FORMALDEHYDE EXPOSURE IN PREGNANT WOMEN AND ITS RELATIONSHIP TO FETAL GROWTH

by AZITA AMIRI

ANNE TURNER-HENSON, CHAIR CHARLES A DOWNS MICHELLE FANUCCHI ERICA PRYOR MARTI RICE LISA SCHWIEBERT

A DISSERTATION Submitted to the graduate faculty of The University of Alabama at Birmingham, in partial fulfillment of the requirements for the degree of Doctor of Philosophy BIRMINGHAM, ALABAMA 2014

Copyright by AZITA AMIRI 2014

FORMALDEHYDE EXPOSURE IN PREGNANT WOMEN AND ITS RELATIONSHIP TO FETAL GROWTH AZITA AMIRI SCHOOL OF NURSING ABSTRACT Formaldehyde exposure during pregnancy has been linked to adverse pregnancy outcomes such as poor fetal growth, although few studies have examined formaldehyde exposure during pregnancy and its relationship to fetal growth. The purpose of this study was to determine the level of formaldehyde exposure during pregnancy and examine the relationship between formaldehyde exposure and fetal growth. Formaldehyde exposure was examined, using vapor monitor badge and urine formic acid, in 140 women in their second trimester of pregnancy. One time urine samples were collected during a routine prenatal visit, and women wore the vapor badges for 24 hours. Fetal growth was measured through fetal ultrasound biometry measurements including, biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), femur length (FL), and estimated fetal weight (EFW). The mean level of formaldehyde exposure using the vapor monitor badges was 0.04 parts per million (ppm) (SD=0.06); 36.4% of participants exceeded Minimum Risk Levels (MRL's) of 0.03 ppm, the Agency for Toxic Substances and Disease Registry (ATSDR) standard for personal exposure for 14-364 days. Formaldehyde levels by vapor monitor badge (< 0.03 ppm & > 0.03 ppm) were correlated with season of data collection (p < .008), indoor temperature of dwellings (p < .014), and house remodeling (p < .037). The linear regression model including covariates showed that ATSDR dichotomized level of formaldehyde exposure was a significant predictor of BPD percentile, after controlling iii

for maternal race. The relationships between isoprostane and fetal growth were nonsignificant. Home and lifestyle behaviors can lead to air pollutant risks due to formaldehyde. Over one third of participants in this study had formaldehyde exposures levels for a 24 hour period exceeding ATSDR recommended levels for 14-364 days. Promoting home and lifestyle behaviors to reduce exposure to environmental toxic chemicals, including, formaldehyde should be included as part of prenatal care.

Keywords: formaldehyde, pregnancy, fetal growth indicators, formic acid, isoprostane, cotinine, tobacco smoke.

iv

This dissertation is lovingly dedicated to:

The memory of my beloved father, Ali Norouzi, his words of encouragement and push for tenacity still rings in my ears

My mother, Zari Firouzi, her support and unconditional love have sustained me throughout my life

My beautiful children, Armita, Parmida, and Arya, for their patience and benevolence

And

My dear husband, Nasser Amiri, for his love, constant source of support, and sacrifices in each step of the way

v

ACKNOWLEDGMENTS

There are people in everyone’s life who make success both possible and rewarding. I owe my deepest gratitude to my sister Azin, who unconditionally and steadfastly supported and encouraged me in this journey. My other sisters, Azar, Faranak, and Giti, who have never left my side and are very special. I cannot find words to express my gratitude to Dean Fay Raines of the University of Alabama in Huntsville, College of Nursing, for her encouraging words and constant support, without her expertise I would not have accomplished as much and many thanks to my mentor, Dr. Darlene Showalter, who prodded me when I needed most. I would like to thank my dissertation committee members Dr. Erica Pryor, for exquisite attention to detail and being a student advocate, Dr. Marti Rice for her demand for excellence, and Dr. Charles Downs, for his superb scientific suggestions and sincere interest in my work. I would also like to thank my other committee members, Dr. Michelle Fanucchi and Dr. Lisa Schwiebert for their scientific assistance, and my advisor Anne Turner-Henson, who had a significant impact on my graduate career. In addition, a thank you to Dr. Becky Christian, who continually and convincingly conveyed a spirit of adventure in regard to research and scholarship, and an excitement in regard to my profession. Special thanks goes to the physicians and staff of the obstetrics and gynecologic clinics who helped me in data collection, Drs Krishna Kakani, James Light, Leon Lewis,

vi

Kennett Pitts, and Ms Xóchitl Dupré for their willingness for help and encouragement during my data collection. I thank the University of Alabama in Huntsville which provided financial support IIDR-260335. Also, I consider it an honor to work with Dr. Debra Moriarity and Dr. Gordon McGregor from the Biology Department of the University of Alabama in Huntsville, who gave me home in their lab and supported me unconditionally. Finally, I would like to thank my supportive friends in the Leadership and Education in Child Health Nursing program, Dr. Thuy Lam, Dr. Luz Huntington-Moskos, Dr. Chrissy Feeley, Susan Williams, Jeannie Rodriguez, Ann Johnson, Heather Soistmann, Allison McQuirter, and Sara Davis, for their advice and support through the years.

Note. The dissertation study was supported by funding from the University of Alabama in Huntsville, IIDR 260335. In addition, the coursework was supported by funding from Leadership Education in Child-Health Nursing (LECHN) - Maternal Child Health Bureau (grant number T80MC09653, PI: A. Turner-Henson), 2010-2012 Gladys Farmer Colvin Memorial Scholarship, and 2011-2013 Alabama Board of Nursing Scholarship.

vii

TABLE OF CONTENTS Page ABSTRACT....................................................................................................................... iii DEDICATIONS ...................................................................................................................v ACKNOWLEDGMENTS ................................................................................................. vi LIST OF TABLES ........................................................................................................... xiii LIST OF FIGURES ...........................................................................................................xv LIST OF ABBREVIATIONS .......................................................................................... xvi CHAPTER 1. INTRODUCTION .....................................................................................................1 Fetal Growth ............................................................................................................1 Formaldehyde Exposure ..........................................................................................6 Tobacco Smoke Exposure......................................................................................11 Oxidative Stress…………… .................................................................................12 Conceptual Framework ..........................................................................................15 Conceptual Definition of Study Variables… .........................................................18 Fetal Growth… ................................................................................................19 Formaldehyde Exposure Level… ....................................................................19 Indoor Residential Sources of Formaldehyde Exposure … .............................19 Tobacco Smoke Exposure…............................................................................20 Oxidative Stress… ...........................................................................................20 Demographic Maternal Characteristics and Pregnancy Characteristics. .................................................................................................20 Statement of the Problem .......................................................................................20 Statement of the Purpose… ...................................................................................22 Research Questions ………… ...............................................................................23 Research Question 1………… ........................................................................23 Research Question 2………… ........................................................................23 Research Question 3………… ........................................................................23 Research Question 4………… ........................................................................24 Significance of Study ………… ............................................................................24 viii

TABLE OF CONTENTS (continued) Page Assumptions …………..........................................................................................25 Summary ………… ...............................................................................................25 2. REVIEW OF LITERATURE .......................................................................................27 Fetal Growth………… ..........................................................................................28 Formaldehyde Exposure and Fetal Growth………… ...........................................34 Tobacco Smoke Exposure and Fetal Growth …....................................................39 Measurement of Level of Formaldehyde Exposure during Pregnancy ………… 43 Formaldehyde Exposure Measures………… ..................................................43 Formaldehyde Exposure: Formic acid versus Vapor Monitor …. ...................45 Standards for Formaldehyde Exposure…. .......................................................47 Formaldehyde Exposure in Pregnant Women… .............................................48 Indoor Residential Sources of Formaldehyde Exposure ………… .......................56 Tobacco Smoke Exposure Measures During Pregnancy ………… ......................63 Tobacco Smoke and Formaldehyde Exposure………………….. ………… ……67 Oxidative Stress as a Mediator of the Relationships between the Level of Formaldehyde Exposure and Fetal Growth …………….. …………….……..70 Oxidative Stress as a Mediator of the Relationships between the Level of Tobacco Smoke Exposure and Fetal Growth ……………………...…………75 Summary ………… ...............................................................................................78 3. METHODOLOGY .......................................................................................................82 Study Design ..........................................................................................................82 Characteristics of the Sample.................................................................................83 Characteristics of the Setting. ................................................................................85 Sample Size and Power. .........................................................................................86 Protection of Vulnerable Subjects. ........................................................................88 Instrumentation . ....................................................................................................90 Questionnaire.....................................................................................................91 Demographic Characteristics ......................................................................91 Pregnancy Characteristics… .......................................................................93 Residential Dwelling Characteristics and Household Practices.......................................................................................................97 Study Biomarkers …............................................................................................100 Formaldehyde…. .............................................................................................101 Vapor Monitor Badge… .............................................................................103 Formic acid … ............................................................................................105 Creatinine Standardization ..........................................................................106 Cotinine … ..................................................................................................107 15-Isoprostane F2t .......................................................................................108 ix

TABLE OF CONTENTS (continued) Page Pilot Study …. ......................................................................................................109 Pilot Sample Characteristics …. ......................................................................111 Evaluation of Methods …. ..............................................................................112 Data Collection Procedure ...................................................................................113 Recruitment ....................................................................................................113 Informed Consent Process .............................................................................114 Data Collection ..............................................................................................114 Medical Record Reviews ...............................................................................116 Biomarkers Analysis Processing....................................................................116 Data Management and Analysis ..........................................................................117 Data Management.........................................................................................117 Data Analysis ...............................................................................................118 Research Question 1 ...............................................................................119 Research Question 2 ...............................................................................120 Research Question 3 ...............................................................................121 Research Question 4 ...............................................................................122 Summary………… ..............................................................................................124 4. FINDINGS ..................................................................................................................126 Description of the Study Population .....................................................................126 Missing Values ......................................................................................................127 Maternal Demographic and Pregnancy Characteristics ........................................128 Fetal Ultrasound Biometry Measurements ............................................................130 Formaldehyde Vapor Monitor and Urinary Biomarker Measurements ................132 Research Questions ...............................................................................................133 Research Question 1 .........................................................................................133 1a. What Is the Level of Formaldehyde Exposure? ....................................133 1b. What Are the Indoor Residential Sources of Formaldehyde Exposure? ..................................................................................................133 1c. Are There Relationships among Level of Formaldehyde Exposure, Tobacco Smoke Exposure, and Indoor Residential Sources of Formaldehyde? ..........................................................................................137 Research Question 2 .........................................................................................140 2a. Are There Relationships between Maternal Demographic Characteristics and Fetal Growth? ..............................................................140 2b. Are There Relationships between Pregnancy Characteristics and Fetal Growth? .......................................................................................141

x

TABLE OF CONTENTS (continued) Page Research Question 3 .........................................................................................142 Do Level of Formaldehyde Exposure, Indoor Residential Sources of Formaldehyde Exposure, and Tobacco Smoke Formaldehyde Exposure Level?..........................................................................................142 Research Question 4 .........................................................................................147 Does Oxidative Stress Mediate the Relationships between (a) Level of Formaldehyde Exposure and Fetal Growth, and (b) Tobacco Smoke Exposure and Fetal Growth?...........................................................147 Additional Analysis ...............................................................................................147 Tobacco Smoke and Formaldehyde Exposure .................................................147 Tobacco Smoke Exposure Prevalence: Cotinine and Self-Report ....................................................................................................149 Summary ..............................................................................................................150 5. DISCUSSION ............................................................................................................152 Findings Related to the Study Sample ..................................................................152 Findings Related to Research Questions ...............................................................155 Research Question 1 .........................................................................................155 1a. What Is the Level of Formaldehyde Exposure? ....................................155 Formic Acid ...........................................................................................155 Formaldehyde Vapor Monitor Badge....................................................156 1b.What Are the Indoor Residential Sources of Formaldehyde Exposure? ....................................................................................................158 1c. Are There Relationships among Levels of Formaldehyde Exposure, Tobacco Smoke Exposure, and Indoor Residential Sources of Formaldehyde Exposure?..........................................................158 Formaldehyde Exposure (Formic Acid vs Vapor Monitor Badge) ..................................................................................................158 Formaldehyde and Tobacco Smoke Exposure.....................................159 Formaldehyde Exposure and Indoor Residential Sources of Formaldehyde ..................................................................................162 Residential Dwelling Characteristics ...................................................162 Formaldehyde and Season of Data Collection .....................................164 Household Practices .............................................................................165 Research Question 2 .........................................................................................167 2a. Are there Relationships between Maternal Demographic Characteristics and Fetal Growth? ...........................................................167 2b. Are There Relationships between Pregnancy Characteristics (Gravida, Fetal Gender, Interval between Pregnancies, and Maternal Smoking Status) and Fetal Growth? ..........................................170

xi

TABLE OF CONTENTS (continued) Page Research Question 3 .........................................................................................172 Do Levels of Formaldehyde Exposure, Indoor Residential Sources of Formaldehyde Exposure, and Tobacco Smoke Exposure Influence Fetal Growth? ..............................................172 Formaldehyde (Formic Acid) ..............................................................172 Formaldehyde (Vapor Monitor Badge) ...............................................172 Indoor Residential Sources of Formaldehyde Exposure ..............................................................................................173 Tobacco smoke exposure .....................................................................174 Research Question 4 .........................................................................................174 Among Women in the Second Trimester of Pregnancy, Does Oxidative Stress Mediate the Relationships between (a) Level of Formaldehyde Exposure and Fetal Growth, and (b) Tobacco Smoke Exposure and Fetal Growth?.....................................................174 Conceptual Framework ........................................................................................178 Tobacco Smoke Exposure: Cotinine and Self-Report .........................................179 Study Limitations .................................................................................................181 Implications for Nursing Practice and Education ................................................184 Implications for Future Research .........................................................................185 Summary ..............................................................................................................188 LIST OF REFERENCES .................................................................................................190 APPENDICES A

IRB APPROVAL LETTERS .......................................................................246

B

INDIVIDUAL INVESTIGATOR DISTINGUISHED RESEARCH GRANT/FINANCIAL SPONSOR .........................................254

C

LETTERS OF SUPPORT.............................................................................256

D

QUESTIONAIRE/FORMS ..........................................................................262

E

INSTRUCTION FOR USING VAPOR MONITOR BADGE ....................273

F G

BIOSAFETY LEVEL 1 & 2 FOR INFECTIOUS AGENTS ......................275 TESTS PROCEDURES ...............................................................................278 Cotinine ELISA ...........................................................................................279 Formaldehyde Vapor Monitor ......................................................................281 15-Isoprostane F2t .........................................................................................301

xii

LIST OF TABLES

Table

Page

1

Variable Classification and Measurement Levels for Maternal Demographic Characteristics .................................................................................91

2

Variable Classification and Measurement Levels for Pregnancy Characteristics .......................................................................................................92

3

Measurement Formulas for Fetal Ultrasound Biometry ........................................95

4

Variable Definitions and Measurement Levels for Indoor Residential Dwelling Characteristics .....................................................................98

5

Measurement Tools and Analytic Methods for Examining the Level of Formaldehyde Exposure, Levels of Cotinine, and 15-isoprostane F2t. ..............101

6

Distribution of Demographic Characteristics ......................................................128

7

Distribution of Pregnancy Characteristics. ..........................................................129

8

Distribution of Fetal Ultrasound Biometry Measurements (percentile) ..............130

9

Means and Standard Deviations for Formaldehyde Vapor Monitor Badge Readings, and Urinary Biomarkers ......................................................................131

10

Distribution of Residential Dwelling Characteristics (Self Report) ....................134

11

Distribution of Household Practices Characteristics (Self-Report) .....................135

12

Correlations between Indoor Residential Sources of Formaldehyde Exposure with Formaldehyde Level Based on Vapor Monitor Badge ................137

13

Correlations between Formaldehyde (Vapor Monitor), Formic Acid, and Cotinine .........................................................................................................138

14

Correlations between Selected Demographic Characteristics and Fetal Ultrasonic Biometry Measurements (Percentiles) ......................................140

xiii

LIST OF TABLES (continued)

Table

Page

15

Correlations between Selected Pregnancy Characteristics and Fetal Ultrasonic Biometry Measurements (Percentiles) .....................................141

16

Correlation between Formaldehyde (Vapor Monitor Badge and Formic Acid), and Cotinine with Fetal Ultrasound Biometry Measurements’ Percentiles ..................................................................................140

17

Correlations between Selected Indoor Residential Dwelling Characteristics and Fetal Ultrasonic Biometry Measurement (Percentiles) ........................................................................................................145

18

Exposure Duration, Formaldehyde Level, and Eight Hour Exposure Based on Number of Cigarettes Smoked Under the Chamber .......................................................................................................147

19

Frequencies and Percentages of Smokers Based on Self-Report, and Classification of Cotinine Based on Cut-off Points Reported by SRNT (2002), and Pickett et al. (2005) ........................................147

xiv

LIST OF FIGURES

Figure

Page

1

Proposed biological framework for exploring possible effect Modification of PM-birth outcomes by maternal nutrition....................................15

2

Conceptual Framework ..........................................................................................18

3

Mediation Model for oxidative stress in the relationship between formaldehyde and fetal growth indicators ...........................................................122

xv

LIST OF ABBREVIATIONS

AC

abdominal circumference

AC/FL

abdominal circumference to Femur Length Ratio

ACOG

The American College of Obstetricians and Gynecologists

ASRM

American Society for Reproductive Medicine

ATSDR

Agency for Toxic Substances and Disease Registry

BPD

biparietal diameter

CDC

Center for Disease Control

CO2

carbon dioxide

COPD

chronic obstructive pulmonary disease

CPEP

calcium for preeclampsia prevention

DNPH

dinitrophenylhydrazine

EFW

estimated fetal weight

EPA

Environmental Protection Agency

ETS

Environmental Tobacco Smoke

FEMA

Federal Emergency Management Agency

FL

femur length

g

gram

GA

gestational age

GC

gas chromatography

xvi

LIST OF ABBREVIATIONS (continued)

GC/MS

Gas chromatography/mass spectrometry

HC

head circumference

HPLC

high-performance liquid chromatography

IARC

International Agency for Research on Cancer

IUGR

intrauterine growth restriction

IRB

Institutional Review Board

LBW

low birth weight

LMP

last menstrual period

MACT

maximum achievable control technology

µg

microgram

mg

milligram

m

meter

ml

milliliter

MDA

malondialdehyde

MRL

minimum risk level

ng

nanogram

NIOSH

National Institute for Occupational Safety and Health

NTP

National Toxicology Program

OHRP

Office for Human Research Protections

OSHA

Occupational Safety and Health Administration

PAH

polycyclic aromatic hydrocarbons

xvii

LIST OF ABBREVIATIONS (continued) PAHO

Pan American Health Organization

PM

particulate matter

ppb

parts per billion

ppm

parts per million

REL

Recommended exposure limit

ROS

reactive oxygen species

RNS

reactive nitrogen species

SD

Standard Deviation

SGA

small for gestational age

SRNT

Society for Research on Nicotine and Tobacco

UAB

University of Alabama at Birmingham

UAH

University of Alabama in Huntsville

VOC

volatile organic compounds

US

United States

WHO

World Health Organization

xviii

CHAPTER 1 INTRODUCTION Fetal Growth Fetal growth begins with the union of egg and sperm, which creates the zygote, and ends at birth (Cunningham et al., 2010). Normal fetal growth can be defined as uneventful cell divisions of the zygote that yields a full-term infant with full expression of its genetic potential (Fescina et al., 2011). The normal nine month fetal growth period is divided into three trimesters. During the first trimester (1-14 weeks of gestational age), development of brain, spinal cord, heart, central nervous system, and other organs begins (Cunningham et al., 2010). Organ development continues in the second trimester (13-28 weeks of gestational age), with a growth peak in the neural system (Fescina et al., 2011). The third trimester (29-42 weeks of gestational age) is accompanied by the development of organs, including long bones and lungs, and rapid weight gain (Cunningham et al., 2010). Any deviation in growth of a fetus or failure to reach growth potential is called intrauterine fetal growth restriction (IUGR) (Bernstein, Gabbe, & Reed, 2002; Wolfe & Gross, 1989). Poor fetal growth in fetuses with IUGR is an important cause of perinatal mortality and morbidity because of its correlation with low birth weight (LBW), defined as less than 2500 grams (Cunningham et al., 2010; Longo et al., 2013; McCormick, 1985; Pollack & Divon, 1992; Singh & Toppo, 2011). LBW is a major worldwide health

1

concern (WHO, 2002), and has been identified as a priority initiative among United States (US) health priorities in the Healthy People 2020 document (2011). The prevalence of LBW in the US was 8.2% in 2010, and Alabama ranked 48th among US states with a prevalence of 10.4% (America's Health Rankings, 2011). Worldwide, the perinatal mortality rate in LBW infants is approximately 20 times that of infants with birth weight more than 2500 grams (WHO, 2002). IUGR is the second leading contributor to perinatal morbidity and mortality (Bernstein et al., 2002; Longo et al., 2013; Wolfe & Gross, 1989). Perinatal mortality is defined as death of the child during the period between the 24th week of gestational age until 4 weeks after birth (WHO, 2006). Perinatal morbidity is defined as a disorder in the neonate, child or adult that occurs as a result of adverse influences of genetic or environmental factors either on the fetus during pregnancy, or on the infant during the first four weeks of life (National Health and Medical Research Council, 1995). Although perinatal mortality is a serious adverse outcome, it occurs relatively infrequently in developed countries because of the high number of neonatal specialists and facilities. However, the fetal period strongly influences perinatal morbidity, including short term and long term health effects. Neonatal consequences of IUGR have been found to be associated with respiratory distress, necrotizing enterocolitis, intraventricular hemorrhage, patent ductus arteriosus, necrotizing enterocolitis, retinopathy, metabolic and hematological disturbances (e.g., hypoglycemia, polycythemia), and hypothermia (Bernstein et al., 2002; Longo et al., 2013; WHO, 2002; Wolfe & Gross, 1989). IUGR has been linked to adverse effects that persist across the life course (Barker & Osmond, 1986). Infants who had IUGR are at greater risk for chronic disabilities such as cerebral

2

palsy, vision and hearing impairments (McMartin, Martin-Amat, Noker, & Tephly, 1979; Zhang, Merialdi, Platt, & Kramer, 2010), growth and cognitive developmental defects (WHO, 2002), autism (Abel et al., 2013), speech and language deficits, internalizing and attention problems, social difficulties (Veen et al., 1991), hyperactivity (Guellec et al., 2011; Pharoah, Stevenson, Cooke, & Stevenson, 1994), and learning disabilities (Johnson & Breslau, 2000). In addition, IUGR has been reported to be associated with chronic diseases in adulthood such as coronary heart disease, type II diabetes, and abnormalities of blood coagulation and lipid metabolism (Barker et al., 1993; Byrne & Phillips, 2000; Calkins & Devaskar, 2011; Frankel et al., 1996; Robillard & Segar, 2006; Zhang et al., 2010). IUGR can result in significant healthcare costs due to the use of advanced medical technologies, complex medical and surgical procedures, and longer hospital stays for infants and mothers (Schmitt, Sneed, & Phibbs, 2006). Approximately 1 in 12 live born infants are admitted to a neonatal facility, with half receiving neonatal intensive care, and many of these infants have IUGR or LBW(Jenkins, McCall, Gardner, Casson, & Dolk, 2009). Furthermore, families may experience increased caregiving burdens and substantial costs due to the complex care and special education these infants may require over their lifetimes (Petrou et al., 2003). The assessment of fetal growth is based on the comparison of the fetal ultrasound biometric measurements with standards obtained from pregnancies with no known abnormalities (Fescina et al., 2011). The American College of Obstetricians and Gynecologists (2013b), reported four common fetal ultrasound biometry measurements that are used to assess fetal growth based on gestational age. These measurements

3

include: biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), and femur length (FL). In addition, fetal weight can be estimated using fetal ultrasound biometry measurements and this is often used as an indicator for fetal growth (Cunningham et al., 2010; Fescina et al., 2011; Owen & Khan, 1998; The American College of Obstetricians and Gynecologists, 2013b). The other fetal ultrasound biometric measurement used to assess fetal growth independent of gestational age is the fetal abdominal circumference to femur length ratio (AC/FL) (Fescina et al., 2011). The fetal ultrasound biometry measurements are calculated using validated mathematical formulas (Hadlock, Deter, Harrist, & Park, 1984) that were developed based on the standards collected from healthy fetuses in a society (Fescina et al., 2011). Previous studies have focused on fetal growth in third trimester of pregnancy, however, centering on fetal growth during third trimester may lead researchers to overlook the importance of early fetal growth that potentially is associated with irreversible damage later in pregnancy (Pedersen, Wojdemann, Scheike, & Tabor, 2008). Recent studies have demonstrated that fetal growth restriction may start as early as the first trimester of pregnancy; therefore, the second trimester ultrasound, when the fetal ultrasound biometric measurements are computable, is valuable in determining the outcome of pregnancy (Bukowski, 2004; Smith, Smith, McNay, & Fleming, 1998; Smith et al., 2002). Results from several published studies showed that smaller than expected fetal ultrasound biometric measurements in the second trimester are associated with LBW (Nakling & Backe, 2002; Nguyen, Larsen, Engholm, & Moller, 2000; Pedersen, Wojdemann, et al., 2008). Ultrasonic measurement of BPD in the second trimester is the widespread method of choice in predicting the gestational age (Tunon, Eik-Nes, &

4

Grottum, 1999). In addition, Pedersen, Wojdemann, et al. (2008) and Vasudeva, Abraham, and Kamath (2013) found a significant relationship between the growth rate measured by BPD and fetal growth in the second trimester. Fetal growth is influenced by a multitude of fetal and maternal factors. There are several maternal factors that have been identified, including, but not limited to: age, race, socioeconomic status, parity, smoking during pregnancy, and maternal education (Cunningham et al., 2010; Fescina et al., 2011). In addition, a short interval between two consequent pregnancies (less than 18 months) is another risk factor associated with fetal growth restriction (Fuentes-Afflick & Hessol, 2000; Gulmezoglu, de Onis, & Villar, 1997; Zhu, Rolfs, Nangle, & Horan, 1999). One fetal factor influencing growth is gender. The relationship between fetal gender and fetal growth has been investigated previously, although results are conflicting (Melamed et al., 2013; Parker, Davies, Mayho, & Newton, 1984; Smulian et al., 1995). Researchers have reported that the fetus also is vulnerable to environmental pollutants (Love, David, Rankin, & Collins, 2010; Makri, Goveia, Balbus, & Parkin, 2004; Makris, Thompson, Euling, Selevan, & Sonawane, 2008; Vrijheid et al., 2011). Exposure to indoor and outdoor air pollutants, such as formaldehyde and tobacco smoke, during pregnancy was associated with adverse pregnancy outcomes in several studies (Grazulevicience, Dulskiene, & Vencloviene, 1998; Maroziene & Grazuleviciene, 2002). A majority of the studies focused on the effect of formaldehyde on abortion and congenital malformations in occupational settings; however, a few studies examined the relationship between formaldehyde exposure in non-occupational setting on LBW and preterm labor (birth before 37 weeks of gestational age) (Grazulevicience et al., 1998;

5

Maroziene & Grazuleviciene, 2002). Tobacco smoke is another air pollutant that has an effect on fetal growth. Several studies have reported the effect of tobacco smoke exposure during pregnancy on fetal ultrasonic biomarker measurements (smaller head circumference, femur length, and estimated fetal weight) in the third trimester (Bergsjo, Bakketeig, & Lindmark, 2007; Jaddoe et al., 2007; Prabhu et al., 2010). However, a few researchers have reported that fetal growth can be affected by tobacco smoke as early as the second trimester (Iniguez et al., 2012; Iniguez et al., 2013; Roza et al., 2007).

Formaldehyde Exposure Formaldehyde is an indoor air pollutant with the chemical formula of HCHO. It is an aldehyde, which means it is an organic compound that contains a formyl group which may modify gene expression (IARC, 2006; National Toxicology Program, 2010). The International Agency for Cancer Research (IARC) reclassified formaldehyde from “probable human carcinogen” to “known human carcinogen” in 2006 (IARC, 2006); and formaldehyde is defined by the US Clean Air Act Maximum Achievable Control Technology (MACT) standard as a hazardous air pollutant (EPA, 1990). Synonyms of formaldehyde are formalin, formic aldehyde, paraform, formol, formalin (methanol-free), fyde, formalith, methyl aldehyde, methylene glycol; methylene oxide, tetraoxymethalene, oxomethane, and oxymethylene (OSHA, 2013). The US Environmental Protection Agency (EPA) classifies formaldehyde as a volatile organic compound (VOC) (EPA, 2011). More than 46 billion pounds of formaldehyde are produced worldwide annually and used in building materials and household products (WHO, 2010). Studies have reported that formaldehyde is a toxic

6

compound for the fetus (Cogliano et al., 2005; IARC, 2006; WHO, 2002), and formaldehyde that presents in residential dwellings may serve as a potential pollutant during pregnancy. Formaldehyde is produced in a small amount endogenously in all living organisms. However, the Agency for Toxic Substances and Disease Registry (ATSDR) reported that indoor air is the dominating contributor to formaldehyde exposure through inhalation (2010). The indoor formaldehyde is two to ten times more than the outdoor concentration (ATSDR, 2010; Dassonville et al., 2009). An important factor that can influence exposure is the season of the year, which affects the amount of time spent indoors (Dannemiller et al., 2013; Dassonville et al., 2009). Indoor residential sources of formaldehyde exposure can be classified into two broad categories: residential dwelling characteristics and household practices (WHO, 2010). Residential dwelling characteristics include type of dwelling, age of dwelling, and indoor temperature (ATSDR, 2010; Dassonville et al., 2009). Formaldehyde is detectable to some degree in almost all residential dwellings. Indoor formaldehyde in residential dwellings is mainly emitted from building and household materials such as paints, adhesives, wall boards, ceiling tiles, carpets, furniture, fiberglass, fabrics, household cleaners, molding, insulation foams, and tobacco smoke (IARC, 2006). The level of indoor formaldehyde is higher in single dwelling homes than in apartments, and in new homes compared to old homes (Dassonville et al., 2009). Mobile homes and trailers have the highest levels of formaldehyde in residential buildings due to the type of building materials (EPA, 2011). The EPA estimated that the average indoor air formaldehyde level

7

in homes, with significant amounts of new pressed wood products, can be greater than 0.3 ppm (EPA, 2011). Other residential characteristics associated with higher levels of formaldehyde are home remodeling (Dassonville et al., 2009), installing new carpet (Cracowski, Carpentier, et al., 2002; Dassonville et al., 2009; Hodgson, Wooley, & Daisey, 1993; Rogers et al., 2007), and adding new furniture (ATSDR, 2010; Dassonville et al., 2009). In addition, presence of an attached garage (related to formaldehyde produced by automobile engine emissions), tobacco smoke that may come from neighboring dwellings in apartment complexes, and non-electric home cooking and heating systems are also potential sources of indoor residential formaldehyde exposure (ATSDR, 2010; Hisamitsu et al., 2011; IARC, 2006). The other major category of formaldehyde exposure sources is household practices. Formaldehyde can be produced by the blending of chemicals, such as household cleaning products, air fresheners or perfume with ozone (Liu, Mason, Krebs, & Sparks, 2004; Uhde & Salthammer, 2007). In addition, burning candles emit several chemicals including formaldehyde (Petry, Cazelle, Lloyd, Mascarenhas, & Stijntjes, 2013). Further, indoor use of nail polish is associated with the increased levels of formaldehyde is residential dwellings (Alaves, Sleeth, Thiese, & Larson, 2013; Sainio, Engstrom, Henriks-Eckerman, & Kanerva, 1997). A potential source of indoor residential formaldehyde exposure is tobacco smoke. Formaldehyde is generated from the combustion of saccharide materials in tobacco, such as sugars and cellulose (Baker, 2006). These materials are naturally present in tobacco, and are often added as an ingredient by some companies to improve taste (Baker, 2006;

8

Rustemeier, Stabbert, Haussmann, Roemer, & Carmines, 2002). Therefore, tobacco smoke also can be a source of formaldehyde in indoor air (IARC, 2006). Although several studies have reported that the indoor formaldehyde level is not influenced by tobacco smoking (Dassonville et al., 2009; Gustafson, Barregard, Lindahl, & Sallsten, 2005; Heroux et al., 2010; Zhang & Cai, 2003); tobacco smoke remains a potential source of formaldehyde for pregnant women. Formaldehyde is absorbed through the dermis, by inhalation, or by ingestion (IARC, 2006), and transmitted to the fetus through the placenta(Katakura, Kishi, Okui, Ikeda, & Miyake, 1993). Some biological changes in respiratory system during pregnancy may increase the likelihood of inhalation of formaldehyde (Cunningham et al., 2010). Overton, Kimbell, and Miller (2001), in a study of dosimetry modeling of inhaled formaldehyde in the human respiratory tract, reported that over 95% of inhaled formaldehyde was retained from the respiratory system. Further, Thrasher and Kilburn (2001) reported that elimination of formaldehyde is slower from fetal tissues than maternal tissues. Formaldehyde exposure may have acute (short-term) or chronic (long-term) health effects on humans (ATSDR, 2010). Acute health effects include irritation of the eye, nose, and throat and allergic reactions (International Agency for Research on Cancer, 2006). The followings have been identified as adverse effects of chronic formaldehyde exposure: asthma, dysplasia, hyperplasia, metaplasia, epithelial cell proliferation, lung cancer, childhood leukemia, genotoxicity, premature birth, LBW, congenital anomalies, abortion, and immunological effects (Duong, Steinmaus, McHale, Vaughan, & Zhang, 2011; IARC, 2006).

9

The biological mechanism by which formaldehyde may interfere with the processes of fetal growth is not clear (IARC, 2006). However, oxidative stress was found to be a determining factor in the cytotoxic effects of formaldehyde (Gulec et al., 2006; Gurel, Coskun, Armutcu, Kanter, & Ozen, 2005; Im et al., 2006; Kum, Kiral, Sekkin, Seyrek, & Boyacioglu, 2007; Saito, Nishio, Yoshida, & Niki, 2005; Sögüt, Songur, Özen, Özyurt, & Sarsilmaz, 2004). In addition, oxidative stress has been linked to IUGR (Ashok, Sajal, & Rakesh, 2005; Negi, Pande, Kumar, Khanna, & Khanna, 2012). Oxidative stress occurs when the production of reactive oxidant molecules go beyond the capacity of the cell’s antioxidant defense mechanisms (Thompson & Al-Hasan, 2012). The Agency for Toxic Substances and Disease Registry (ATSDR) defines Minimal Risk Level (MRL) as “an estimate of daily human exposure to a substance that is likely to be without an appreciable risk of adverse effects (noncarcinogenic) over a specified duration of exposure” (ATSDR, 2010, p. 10). Due to the carcinogenic effect of formaldehyde, no level is known to be risk free (California Air Resources Board, 2005). MRLs have been estimated for noncancerous health effects of formaldehyde exposure through inhalation (ATSDR, 2010). The MRLs of 0.04 ppm, 0.03 ppm, and 0.008 ppm for formaldehyde exposure have been derived for acute-duration inhalation exposure (14 days or less), intermediate-duration inhalation exposure (15–364 days), and chronicduration inhalation exposure (365 days or more), respectively (ATSDR, 2010).There is no standard for formaldehyde exposure level in residential dwellings (EPA, 2011). However, ATSDR recommends less than 0.03 ppm for sensitive populations including children, the elderly, and the infirm (ATSDR, 2010).

10

Formaldehyde as an air pollutant can be measured by indoor (stationary) and personal exposure measures (Lazenby, Hinwood, Callan, & Franklin, 2012) using vapor monitor badges (Advanced Chemical Sensors, 2012). Formic acid, as a metabolite of formaldehyde exposure, can be measured in urine (Bhatt, Lober, & Combes, 1988; Billings, 1984; Heck, Chin, & Schmitz, 1983; Johansson & Tjalve, 1978; Keefer et al., 1987; Upreti, Farooqui, Ahmed, & Ansari, 1987). However, based on animal studies only 17% of inhaled formaldehyde is excreted in urine of rats (Heck et al., 1983). Therefore, with a low exposure to formaldehyde, the level of detected formaldehyde in urine could be minimal.

Tobacco Smoke Exposure Tobacco smoke during pregnancy influences fetal growth (Fescina et al., 2011; Grazuleviciene, Danileviciute, Nadisauskiene, & Vencloviene, 2009; Iniguez et al., 2012; Iniguez et al., 2013). Maternal exposure to tobacco smoke has been identified as the single major modifiable risk factor for IUGR in developed countries (Bada et al., 2005). A significant association between tobacco smoke and fetal growth in third trimester has been reported previously (Bergsjo et al., 2007; Jaddoe et al., 2007; Prabhu et al., 2010; Pringle et al., 2005). However, Iniguez et al. (2013) reported that tobacco smoke can affect fetal growth as early as mid-pregnancy. Further, tobacco smoke has been reported to be associated with oxidative stress (Cai & Harrison, 2000), the mechanism that may interfere with placenta insufficiency and fetal growth (Hempstock, Jauniaux, Greenwold, & Burton, 2003).

11

Nicotine is the principal toxic alkaloid of tobacco (Hukkanen, Jacob, & Benowitz, 2005) and cotinine is its primary metabolite. Cotinine has a long half-life and is considered the “gold standard” measure of tobacco smoke (SRNT, 2002). Cotinine may be measured in blood, hair, saliva, and urine specimens (Benowitz et al., 2009). The cotinine level in urine is 4-6 times more than blood and saliva, therefore, urine cotinine is a more sensitive biomarker to detect low level exposure to tobacco smoke, such as secondhand smoke or environmental tobacco smoke (ETS) (Benowitz et al., 2009). However, Dempsey, Jacob, and Benowitz (2002) reported that the clearance of nicotine is substantially increased during pregnancy. In addition, they found that the metabolic clearance of cotinine is considerably elevated during pregnancy, which can result in a substantial decrease in the half-life of cotinine and impact urine levels in pregnant women.

Oxidative Stress Jones (2008) has addressed oxidative stress “as a consequence of disruption of thiol redox circuits, which normally function in cell signaling and physiological regulation” (p. C849). In other words, oxidative stress can result in from both radical and non-radical mechanisms (Jones, 2008). Free radical oxidative stress, which is associated with macromolecular damage, occurs when the rate of free radical production exceeds the rate of removal or buffering of free radicals by the cellular defense mechanisms (Burton & Jauniaux, 2011; Jones, 2008). Free radicals that contain oxygen molecules are called reactive oxygen species (ROS) (Halliwell & Gutteridge, 2007). ROS have important roles in cell signaling and homeostasis (Devasagayam et al., 2004). ROS are produced as a

12

normal byproduct of aerobic metabolism. Some situations, for example, aging or exposure to chemicals, can result in production of higher levels of ROS (Sies, 1997). In these situations, ROS damage or kill the healthy cells by inducing cell death through oxidation. Additionally, oxidation causes damage to the cell proteins, and DNA (Ashok et al., 2005; Burton & Jauniaux, 2011). Non-radical mechanism of oxidative stress is related to redox elements that exit in all biological systems. In fact, oxidative stress is a result of disruptions in the function of the redox due to specific reaction with the redox-sensitive thiol elements. Some environmental sources, such as air pollutants, including formaldehyde and tobacco smoke, are associated with oxidative stress (Halliwell & Gutteridge, 2007). One of the by-products of oxidative stress is isoprostane (Roberts & Morrow, 2002). Through the biological process, cell membrane phospholipids are hydrolyzed by the phospholipase enzyme to produce non-esterified arachidonic acid. Non-esterified arachidonic acid can undergo peroxidation through two different pathways (a) the enzymatic pathway, which involves cyclooxygenases and lipoxygenases; and (b) the nonenzymatic pathway that is through the participation of ROS, reactive nitrogen species (RNS), transition metals and other free radicals (Halliwell, 1994). The final products of the lipid peroxidation through non-enzymatic pathways include isoprostane, malondialdehyde, and hydroxynonenal (Montuschi, Barnes, & Roberts, 2004). Therefore, isoprostane is produced by ROS-catalyzed peroxidation of arachidonic acid and is considered a specific marker of lipid peroxidation (Morrow & Roberts, 1997; Viviano & Vanderwielen, 2013).

13

Isoprostane is a prostaglandin-like compound and can be measured in urine and plasma, and its level correlates with disease severity in humans, oxidative injury, and the level of consumed antioxidants (Morrow & Roberts, 1997; Viviano & Vanderwielen, 2013). There are several types of isoprostane. The most abundant isoprostane is 15isoprostane F2t, which has a vasoconstriction effect (García-Estañ, Ortiz, & Lee, 2002). 15-isoprostane F2t is a biologically active isoprostane known to be a reliable biomarker of lipid peroxidation (Cracowski, Durand, & Bessard, 2002; Gill et al., 2009). Levels of 15isoprostane F2t in biological fluids have been shown to be useful for assessment of oxidative stress in vivo (Oxford Biomedial Research, 2012). Oxidative stress has an important role in the pathogenesis of many systemic diseases in humans (Nathens et al., 2002; Roth, Manhart, & Wessner, 2004) through involvement in lipid oxidation, inflammation, endothelial dysfunction, platelet aggregation, collagen degradation, and up-regulation of adhesion molecules and macrophages recruitment (Rodella et al., 2013). Damage induced by excessive production of ROS have been shown in atherosclerosis (Rodella et al., 2013) as well as asthma, Chronic Obstructive Pulmonary Disease (COPD), and cancers (Halliwell & Gutteridge, 2007). Oxidative stress during pregnancy, which determined through measurement of glutathione peroxidase, gluthatione reductase, superoxide dismutase, or malaondialdehyde, has been associated with abortion, loss of placental function, preeclampsia (García-Estañ et al., 2002), preterm labor, and IUGR (Buhimschi, Buhimschi, Pupkin, & Weiner, 2003; O'Donovan & Fernandes, 2004). Oxidative stress may contribute to placental insufficiency that may result in poor fetal growth (Adcock,

14

Brown, Kwon, & Barnes, 1994; Crowley, 2013). Karowicz-Bilinska, Suzin, and Sieroszewski (2002) reported that biomarkers of oxidative stress including malondialdehyde and lipid peroxides were higher in IUGR. Kim et al. (2005), found that levels of maternal urinary 8-hydroxydeoxyguanosine and malondialdehyde were inversely associated with birth weight of full-term deliveries. Kamath, Rao, Kamath, and Rai (2006), reported that malondialdehyde levels were significantly elevated in mothers of growth restricted babies when compared to controls. Published studies to date have not examined isoprostane as a biomarker for oxidative stress and formaldehyde exposure in pregnant women.

Conceptual Framework This study was based on a biological framework proposed by Kannan, Misra, Dvonch, and Krishnakumar (2006) (Figure 1). In this framework, particulate matter and maternal nutrition act through the five exclusive biologic pathways (oxidative stress, pulmonary and placental inflammation, blood coagulation, endothelial function, and hemodynamic response) to result in placental insufficiency. Placental insufficiency is, in turn, associated with IUGR and LBW. Underlying the Kannan et al. (2006) biological framework is the Barker hypothesis. This hypothesis suggested that adverse influences during intrauterine development can increase the risk of diseases in adulthood (Barker & Fall, 1998). David Barker, a physician and professor, was the first scientist who reported that individuals born with LBW were at greater risk of developing coronary heart disease (Paneth & Susser, 1995). According to this hypothesis, intrauterine circumstances may have positive

15

or negative effects on fetal physiology, a phenomenon referred to as fetal programming (Fraser & Cresswell, 1997).

Figure 1. Proposed biological framework for exploring possible effect modification of particulate matter-birth outcomes by maternal nutrition. Reproduced with permission from Environmental Health Perspectives (Kannan et al., 2006)

Particulate matter, an air pollutant and also part of cigarette smoke, is composed of a complex mixture of particles, primarily organic compounds, including VOCs . (Arhami et al., 2010; Liden, Ek, Palmberg, Okret, & Larsson, 2003; Osornio-Vargas et al., 2003; Valavanidis, Fiotakis, & Vlachogianni, 2008). The adverse health outcomes associated to particulate matter can be related to VOCs as well. The EPA has classified formaldehyde as a VOC (EPA, 2011). Although other VOCs, such as benzene, toluene, and styrene have been identified in particular matter, no published study was found that identified formaldehyde as an element of particulate matter. This could be due to the fact 16

that particulate matter is mostly an outdoor air pollutant, while formaldehyde is predominantly an indoor air pollutant. Although there are differences in the chemical structures of the VOCs found in particulate matter and formaldehyde, the framework proposed by Kannan et al. (2006) provides a potential explanation of the physiological pathway for the effect of formaldehyde exposures on fetal growth. To the best of principal investigator’s knowledge, no published study has examined the physiological pathway of formaldehyde on fetal growth. However, the IARC reported the following pathways related to carcinogenic effects of formaldehyde: oxidative stress, tissue inflammation, endothelial function, and hemodynamic responses (IARC, 2006). In addition, previous studies reported a possible relationship between tobacco smoke and oxidative stress (IARC, 2006; Lowe, Luettich, & Gregg, 2013; National Toxicology Program, 2010; Wozniak et al., 2012) and Negi et al. (2012) reported that oxidative stress can be associated with fetal growth restriction due to its effect on placental insufficiency. Therefore, based on the conceptual model of Kannan et al. (2006), oxidative stress, was proposed as a mediator in the relationship between formaldehyde exposure and fetal growth outcomes for this study’s conceptual framework for the following reasons: oxidative stress is associated with fetal growth; oxidative stress is associated with formaldehyde exposure, oxidative stress is associated with tobacco smoke, and a biomarker of oxidative stress and lipid peroxidation (15-isoprostane F2t) is measurable in urine samples. For the current study, using the Kannan et al (2006) biological framework as a guide, the following conceptual model (Figure 2) was developed. The conceptual model integrates the study variables and illustrates the hypothesized relationship between

17

variables that were examined in this study. Maternal demographic and pregnancy characteristics have been added to the model as factors that are known to be associated with fetal growth (Cunningham et al., 2010; Fescina et al., 2011).

Formaldehyde exposure (ppm and formic acid)

Tobacco smoke exposure (cotinine)

Fetal growth* (BPD, HC, AC, FL, EFW, AC/FL)

Oxidative stress (15-isoprostane F2t)

Indoor residential sources of formaldehyde exposure

Maternal demographic and pregnancy characteristics

Figure 2. Conceptual Framework * BPD: Biparietal diameter; HC: Head circumference; AC: Abdominal circumference; FL: Femur length; EFW: Estimated Fetal weight; Abdominal Circumference to Femur Length (AC/FL) Ratio

Conceptual Definition of Study Variables The dependent variable in this study is fetal growth. The independent variables are formaldehyde exposure and tobacco smoke exposure. Formaldehyde exposure was measured in three different ways (formaldehyde monitor badge, formic acid in urine, and indoor residential sources of formaldehyde exposure). Oxidative stress was considered as a mediator in this study. The conceptual definitions of study variables are as follow:

18

Fetal Growth Fescina et al. (2011), in a report published by WHO, defined normal embryo-fetal growth as “the growth that results from uneventful cell division and growth, yielding a full-term infant with full expression of its genetic potential as its end product” (p. 4). Fetal growth was measured using fetal ultrasound biometry measurement percentiles based on gestational age (Cunningham et al., 2010; Fescina et al., 2011; Owen & Khan, 1998; The American College of Obstetricians and Gynecologists, 2013). Specifically biparietal diameter in millimeters, abdominal circumference in millimeters, head circumference in millimeters, femur length in millimeters, estimated fetal weight in grams; and abdominal circumference to femur length ratio were obtained from the ultrasound measurements.

Formaldehyde Exposure Level Formaldehyde exposure level is defined as the level to which an organism is exposed during a specified period (NIOSH, 2007). The exposure level to formaldehyde was measured using objective measurements, including vapor monitor badge (Advanced Chemical Sensors, 2012) and formic acid in urine (NMS LABS, 2013).

Indoor Residential Sources of Formaldehyde Exposure The level of exposure to formaldehyde was measured through self-reported measurements using a questionnaire. Indoor residential sources of formaldehyde exposure included two groups of questions: (1) dwelling characteristics and (2) household practices (WHO, 2010).

19

Tobacco Smoke Exposure Tobacco smoke exposure level is defined as the level to which an organism is exposed during a specified period (NIOSH, 2007). Urine cotinine was measured as a biomarker of tobacco smoke exposure (Benowitz et al., 2009) and self-report.

Oxidative Stress Oxidative stress is defined as “an imbalance of pro-oxidants and antioxidants” (Jones, 2006, p. 9). 15-isoprostane F2t, as a biomarker of oxidative stress, was measured in urine (Halliwell & Gutteridge, 2007).

Demographic Maternal Characteristics and Pregnancy Characteristics Maternal demographic characteristics included age, education, marital status, race/ethnicity, employment status, and yearly family income. Pregnancy characteristics that are associated with fetal growth included gender, gravida, maternal smoking status, and interval between two pregnancies (Cunningham et al., 2010; Fescina et al., 2011).

Statement of the Problem Poor fetal growth is associated with higher prenatal mortality and morbidity rates (Fescina et al., 2011; WHO, 2002). It also has long term adverse effects, such as cognitive and developmental disabilities, metabolic disorders, which contribute to higher economic burdens to family and society (WHO, 2002). Among the environmental risk factors known to be associated with fetal growth are formaldehyde and tobacco smoke exposure. However, there is limited empirical evidence regarding level of formaldehyde

20

exposure in pregnant women and its relationship to fetal growth. Furthermore, there is limited evidence examining the relationship between tobacco smoke and fetal ultrasound measurements in the second trimester. Additionally, the role of oxidative stress as a mediator, in the relationship between formaldehyde and tobacco smoke exposure is not known. Occupational exposures to formaldehyde have been found to be associated with adverse pregnancy outcomes (e.g., abortion, congenital malformation, IUGR, and preterm labor [less than 37 weeks of pregnancy]) (Buhimschi et al., 2003; Pearson et al., 2003). The adverse effects of formaldehyde exposure on pregnancy outcomes have primarily been reported in occupational indoor air, although a few researchers have examined the association between outdoor formaldehyde exposures with birth weight and have reported an increase in LBW and preterm labor in groups of women who were exposed to higher rates of formaldehyde (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002). No published studies were found that examined the role of indoor non-occupational formaldehyde exposure on fetal growth. Therefore, this study will help address an important gap in the research literature. From a behavioral aspect, home preparations for the birth of a new baby may lead to increased environmental risks. Parents typically get ready for babies by making repairs to the home environment and buying new baby furniture and bedding sets (Dassonville et al., 2009). Thus, environmental changes in homes may increase formaldehyde in indoor air. Home remodeling, new furniture, and new carpets may increase indoor formaldehyde levels (Dassonville et al., 2009). Since exposure to formaldehyde is mostly

21

through inhalation (WHO, 2010), both physiologic and behavioral changes during pregnancy may alter formaldehyde exposure at this nine-month period. Examining the following will address current gaps in knowledge: level of formaldehyde exposure during pregnancy, indoor residential sources of formaldehyde exposure (including tobacco smoke), the relationship of formaldehyde exposure levels with fetal growth in the second trimester, the relationship of indoor residential sources of formaldehyde exposure and fetal growth in the second trimester, the relationship of tobacco smoke exposure with fetal growth in the second trimester, and the role of oxidative stress as a mediator in the relationship between formaldehyde and tobacco exposures with fetal growth.

Statement of the Purpose The overall purpose of this study was to determine the level of formaldehyde exposure during pregnancy and examine the relationship between formaldehyde exposure and fetal growth (fetal ultrasound biometry measurements) in pregnant women. This study was also designed to determine the following: (1) the indoor residential sources of formaldehyde exposure (including tobacco smoke exposure) and their relationships with formaldehyde level and fetal growth outcomes, and (2) the potential mediating role of oxidative stress in the relationship between formaldehyde exposure and fetal growth in pregnant women.

22

Research Questions Research Question 1 Among women in the second trimester of pregnancy: 1a. What is the level of formaldehyde exposure? 1b. What are the indoor residential sources of formaldehyde exposure? 1c. Are there relationships among level of formaldehyde exposure, tobacco smoke exposure, and indoor residential sources of formaldehyde exposure?

Research Question 2 Among women in the second trimester of pregnancy: 2a. Are there relationships between maternal demographic characteristics (age, race, education, marital status, income, employment status) and fetal growth (HC, BPD, AC, FL, EFW, and AC/FL)? 2b. Are there relationships between pregnancy characteristics (gravida, fetal gender, interval between pregnancies, and maternal smoking status) and fetal growth (HC, BPD, AC, FL, EFW, and AC/FL)?

Research Question 3 Among women in second trimester of pregnancy: Do level of formaldehyde exposure, indoor residential sources of formaldehyde exposure, and tobacco smoke exposure influence fetal growth (HC, BPD, AC, FL, EFW, and AC/FL)?

23

Research Question 4 Among women in the second trimester of pregnancy: Does oxidative stress mediate the relationships between (a) level of formaldehyde exposure and fetal growth (HC, BPD, AC, FL, EFW, and AC/FL), (b) tobacco smoke exposure and fetal growth (HC, BPD, AC, FL, EFW, and AC/FL)?

Significance of Study Poor fetal growth is strongly associated with IUGR and LBW (Barker & Osmond, 1986; The American College of Obstetricians and Gynecologists, 2013b; WHO, 2002). Approximately 8.2% of US births are LBW (America's Health Rankings, 2011) and the US Healthy People 2020 has identified LBW as a priority initiative (HealthyPeople.gov, 2011). Formaldehyde and tobacco smoke exposures have been purported to affect fetal growth, resulting in LBW. However, limited evidence exits about formaldehyde exposure in pregnant women, indoor residential sources of formaldehyde exposure, the effect of formaldehyde and tobacco smoke on fetal growth in the second trimester, and the potential mediating role of oxidative stress on the effect of formaldehyde and tobacco smoke on fetal growth. The knowledge gained from this study will be helpful in establishing the risk exposure categories for formaldehyde during pregnancy. Understanding the formaldehyde exposure level during pregnancy, indoor residential sources of formaldehyde exposure, and the effect of formaldehyde and tobacco smoke on fetal growth can be used to develop techniques to reduce the potential for exposure. In

24

addition, this new knowledge can be added to prenatal care education and maternity care programs in health centers and clinics by informing pregnant women regarding the indoor residential sources of formaldehyde exposure and accessible preventive strategies.

Assumptions The following assumptions were made for the purpose of this study: 1. The level of formaldehyde remains constant in the residential dwellings of pregnant women. 2. The study population was exposed to the same levels of outdoor pollutants during the study period.

Summary Poor fetal growth is accompanied with higher infant mortality and morbidity rate. Formaldehyde is produced worldwide in vast amounts annually and is widely used in household items (e.g., furniture, textiles), and home construction materials. A few studies have shown that formaldehyde exposure during pregnancy is associated with an increased risk of LBW (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002). Evidence indicates that formaldehyde exposure may interfere with pregnancy outcomes through oxidative stress (Thrasher & Kilburn, 2001). The purpose of this study was to determine formaldehyde exposure level during the second trimester of pregnancy, indoor residential sources of formaldehyde exposure, and the relationship between formaldehyde and tobacco smoke exposures and fetal growth. The selected conceptual framework is based on the proposed biological framework for exploring possible effect modification of

25

particulate matter and birth outcomes (Kannan et al., 2006). Based on the selected conceptual model, another purpose of this study was to evaluate the effect of oxidative stress, as a mediator in the biological pathways of the effect of level of formaldehyde exposure on fetal growth.

26

CHAPTER 2 REVIEW OF LITERATURE In this chapter, published research literature was reviewed that addressed the level of formaldehyde exposure during pregnancy and its relationship with fetal growth in the second trimester. Literature related to indoor residential sources of formaldehyde and tobacco smoke exposure and their relationships with fetal growth in the second trimester was also evaluated. In addition, the literature that supported the mediating role of oxidative stress in the relationship between formaldehyde and fetal growth and tobacco smoke and fetal growth in the second trimester was also considered. To review the above literature, internet and journal databases, specifically CINAHL Plus with full text, Pub Med, GreenFILE, ERIC, EPSCO, the Centers for Disease Control (CDC), U.S. Department of Health and Human Services (National Toxicology Program and Agency for Toxic Substances and Disease Registry [ATSDR]), the U.S. Environmental Protection Agency (EPA), and the World Health Organization (WHO) were reviewed. The search terms used included formaldehyde, formol, formalin, tobacco smoke, smoking, pregnancy outcomes, reproductive toxicity, intrauterine growth restriction (IUGR), growth restriction, fetal growth, small for gestational age (SGA), low birth weight (LBW), pregnancy, second trimester, third trimester, cotinine, oxidative stress, 15-isoprostane F2t, indoor formaldehyde, and sources of formaldehyde. The published articles in languages other than English were excluded from this review. For

27

the purpose of the current study, literature from nursing, medicine, psychology, environmental health, and occupational health were reviewed for relevance. This chapter begins with a review of the literature on fetal growth. This is followed by a review of literature on the following topics: (1) the relationship between formaldehyde exposure and fetal growth, (2) the relationship between tobacco smoke exposure and fetal growth, (3) measurement of the level of formaldehyde exposure during pregnancy, (4) indoor residential sources of formaldehyde exposure, (5) tobacco smoke exposure measures during pregnancy, (6) tobacco smoke and formaldehyde exposure, and (7) oxidative stress as a mediator of the relationships between the level of formaldehyde exposure and fetal growth and tobacco exposure and fetal growth. . Fetal Growth Fetal growth is defined as “the growth that results from uneventful cell division and growth, yielding a full-term infant with full expression of its genetic potential as its end product” (Fescina et al., 2011, p. 4). Poor fetal growth, which can result in IUGR, LBW, and SGA, is associated with perinatal mortality and morbidity (WHO, 2002) and contributes to diseases and disabilities during the life course (Barker & Fall, 1998; Barker et al., 1993; Zhang et al., 2010). IUGR, LBW, and SGA are terms that are often used interchangeably; however, their definitions are different. IUGR is defined as a fetus whose estimated weight is below the 10th percentile for its gestational age, LBW is classified as an infant whose birth weight is less than 2500 grams and SGA is an infant whose birth weight is below the 10th percentile (Cunningham et al., 2010). IUGR can be a predictor for LBW or SGA (Fescina et al., 2011).

28

The duration of a normal pregnancy is divided into three trimesters (Cunningham et al., 2010) and fetal growth is characterized by sequential patterns of tissue and organ growth, differentiation, and maturation during these three trimesters (Fescina et al., 2011). Lin and Santolaya-Forgas (1998) divided fetal growth into three sequential phases: phase 1includes the first 16 weeks and is characterized by a rapid increase in cell number; phase 2 extends up to 32 weeks and includes both cellular hyperplasia and hypertrophy; phase 3 involves the time after 32 weeks and is characterized by cellular hypertrophy where the majority of fetal fat and glycogen synthesis take place. Gestational age, which is measured by weeks, is a term that is commonly used during pregnancy to describe the progression of the pregnancy (Cunningham et al., 2010). Gestational age can be calculated based on last menstrual period (LMP) or through early ultrasound measurements (Cunningham et al., 2010). LMP is less reliable than gestational age estimated by an early ultrasound (Hoffman et al., 2008; Kramer, McLean, Boyd, & Usher, 1988; Lynch & Zhang, 2007). The most accurate parameter for determining gestational age is the crown-rump length (length of fetus from the top of the head to the buttock) in the first trimester, which has a linear relationship with gestational age (Fescina et al., 2011), and biparietal diameter (BPD) in the second trimester (Cunningham et al., 2010). However, several studies showed that BPD is not accurate in the third trimester and as pregnancy progresses due to the considerable organ enlargement and weight gain of the fetus (Altman & Chitty, 1997; Egley, Seeds, & Cefalo, 1986; Kurmanavicius et al., 1999; Pedersen, Wojdemann, et al., 2008). Hadlock et al. (1984) used a combination of head circumference (HC), abdominal circumference (AC), and femur length (FL) to determine gestational age in the second and third

29

trimesters. In the current study, gestational age was measured based on the Hadlock formulation (Hadlock et al., 1984; Hadlock, Harrist, & Martinez-Poyer, 1991). Ultrasound is the most accurate method to assess fetal growth within the first and second trimesters (Fescina et al., 2011; The American College of Obstetricians and Gynecologists, 2013b). The most commonly used ultrasound fetal biometric measurements are BPD, HC, AC, FL, estimated fetal weight (EFW), and abdominal circumference to femur length ratio (AC/FL) (Fescina et al., 2011). Fetal ultrasound biometric measurements are important when they are adjusted for gestational age and compared to ultrasound norms (Fescina et al., 2011; The American College of Obstetricians and Gynecologists, 2013b). Ultrasound norms were determined based on serial estimates of fetal ultrasound biometric measurements from a group of uncomplicated pregnancies in a specific population (Bukowski et al., 2008). Fetal ultrasound biometric measurements during the third trimester are used commonly as predictors of fetal growth and pregnancy outcomes by clinicians and researchers (David, Tagliavini, Pilu, Rudenholz, & Bovicelli, 1996; De Reu, Smits, Oosterbaan, & Nijhuis, 2008; Skovron, Berkowitz, Lapinski, Kim, & Chitkara, 1991; Souka, Papastefanou, Pilalis, Michalitsi, & Kassanos, 2012). However, several studies have reported that the second trimester fetal ultrasound biometric measurements are valuable in determining fetal growth and the outcome of pregnancy as well (Bukowski, 2004; Pedersen, Wojdemann, et al., 2008; Smith et al., 1998; Smith et al., 2002; Vasudeva et al., 2013). During the third trimester, AC and EFW are the best predictors of fetal growth problems. Although all ultrasound biometric masurements are valuable in the second trimester, several studies have established BPD as the most valid indicator of

30

gestational age in this trimester (Bukowski, 2004; Pedersen, Wojdemann, et al., 2008; Smith et al., 1998; Smith et al., 2002; Tunon et al., 1999; Vasudeva et al., 2013). Known risk factors of fetal growth can be classified as preconception factors (maternal demographic factors such as age, race, ethnicity, income, and occupational status), factors apparent during pregnancy (gravida, fetal gender, interval between two pregnancies, maternal chronic disease, and fetal abnormalities), environmental factors (indoor or outdoor air pollutants) and maternal behaviors (smoking and alcohol) (Cunningham et al., 2010; Fescina et al., 2011). Several studies showed that maternal demographic characteristics were associated with LBW and IUGR (Choudhary, Choudhary, Tiwari, & Dwivedi, 2013; Cunningham et al., 2010; Olsen, Groveman, Lawson, Clark, & Zemel, 2010). In addition, pregnancy characteristics have been linked to the pregnancy outcomes, such as LBW and IUGR (Abrams & Selvin, 1995; Cunningham et al., 2010; Fuentes-Afflick & Hessol, 2000; Gulmezoglu et al., 1997; Luke, 1994; Papageorghiou, Bakoulas, Sebire, & Nicolaides, 2008; Rode et al., 2007; Shah & Ohlsson, 2002; Zhu et al., 1999). Several studies have provided evidence that maternal race is associated with variability in ultrasound biometric measures. In a cross-sectional study, Lai and Yeo (1995) studied fetal ultrasound biometry measurements (BPD, HC, AC, FL, humerus length [HL], and mandible length [ML]) among 6,374 women in Singapore whose delivery date was within 2 weeks of their estimated date of delivery as calculated from the last menstrual period. They collected data on ultrasound results from medical records and found slightly smaller BPD in Asian compared with white fetuses. The investigators included the earliest ultrasound record for each participant in their data analysis, which is

31

a limitation due to the differences in fetal growth patterns in different trimesters. The other limitation is the use of different ultrasound devices and different technicians, which may affect the reliability of the ultrasound measurements. Ogasawara (2009) examined whether fetal ultrasound biometry is affected by variation in fetal ethnicity in a retrospective study in Hawaii. They studied ultrasound reports from 105 White, 370 Asian, 895 part Hawaiian, 76 Pacific Islander, and 311 white Asian fetuses. They found that at 18 weeks gestation, in comparison to White controls, FL was significantly shorter in Asian and white Asian and humerus length was significantly shorter in Asian, part Hawaiian, and white Asian. However, they did not find any significant differences between second trimester’s BPD in different ethnic groups. African Americans were not part of their sample, so results cannot be generalized to this group. The main limitation of this study was the retrospective design with inclusion of ultrasound reports from different ultrasound devices and different technicians. Parikh, Nolan, Tefera, and Driggers (2013), determined the effect of race on fetal biometry (BPD, HC, AC, FL, and humerus length), by conducting a retrospective chart review of prenatal ultrasounds from January 2009 to December 2010 in a perinatal center in Washington DC. Uncomplicated singleton pregnancies at 17 to 22.9 weeks gestation were included. The sample included 1,327 African American, 147 White, and 86 Hispanic subjects. The investigators reported that AC was significantly smaller in African American than Caucasian fetuses, whereas they found no significant relationship between African Americans and Caucasians in BPD, HC, FL, or HL. They also found no differences between Hispanics and either Caucasians or African American in any of the

32

fetal biometric measurements. The main limitation of this study was inclusion of a limited number of White and Hispanic subjects. In addition, investigators did not consider the effect of maternal smoking status on fetal growth biometric measurements. Among fetal risk factors, after excluding fetal anomalies or metabolic diseases, fetal gender can be considered an important factor (Cunningham et al., 2010). Pang, Leung, Sahota, Lau, and Chang (2003) studied the effects of maternal and pregnancy characteristics on fetal biometric measurements using longitudinal ultrasound measurements. They recruited 533 healthy pregnant women with normal singleton pregnancies and conducted regular ultrasound examination between 24 and 40 weeks of gestational age. They found significant relationship between fetal gender and BPD, HC, and FL. In addition, they found that HC and AC were significantly associated with extremes of maternal age. Maternal height was significantly associated with BPD. Maternal weight gain during pregnancy had an influence on AC and FL. In addition, parity had an influence on fetal head circumference and abdominal circumference. The main limitation of this study was that in the investigators did not consider the effect of maternal weight gain or tobacco smoke on ultrasound biometric measurements. Melamed et al. (2013) analyzed the effect of fetal gender on fetal growth during the second and third trimesters in a cross-sectional study. They recruited 12,132 women with uncomplicated singleton pregnancies who underwent ultrasonic fetal weight estimation during the second and third trimesters in a single tertiary center. They studied the effect of fetal gender on each fetal growth biomarkers (BPD, HC, AC, and FL) and their ratios. They found significant relationships between fetal gender and biparietal diameter (male/female ratio, 1.021) and the head circumference/femur length and

33

biparietal diameter/femur length ratios. BPD and HC were significantly related with gender as early as second trimester; however, abdominal circumference was correlated with late second and late third trimester measurements. Due to the retrospective design of this study, the investigators were not able to collect confounding variables such as maternal BMI, height, maternal weight gain during pregnancy, and maternal smoking status. In a retrospective study in the second and third trimesters of pregnancy, Schwenzer (2008) compared the fetal ultrasound biometric measurements (BPD, HC, AC, and FL) in female and male fetuses who had a normal fetal ultrasound report at 1014 weeks. They recruited 4,234 women from different race/ethnicities. The ultrasounds were performed by four trained sonographers, according to a standardized protocol. They found significant differences in fetal BPD, HC, AC and estimated fetal weight, but not FL, between male and female fetuses. The limitations of this study are similar to others, including use of different ultrasound devices and different technicians and not considering the effect of other confounding variables.

Formaldehyde Exposure and Fetal Growth Formaldehyde can be rapidly absorbed through respiratory and gastrointestinal tracts due to its high solubility in water (Salthammer, 2013). Overton et al. (2001) showed that the respiratory tract retained over 95% of inhaled formaldehyde during routine daily activities including sleeping, sitting, light exercises, and heavy exercise. Cunningham et al. (2010) reported that there are some biological changes in the respiratory system during pregnancy. Biologically, pregnant women have a progressive

34

increase in minute ventilation soon after conception, which peaks at 50% above normal levels around the second trimester (Cunningham et al., 2010). Physiologic changes in the respiratory tract are evidenced by a 40% rise in tidal volume and a 15% rise in respiratory rate (2-3 breaths per minute) (Jeffcoat, Chasalow, Feldman, & Marr, 1983). Although, the dead space in the lungs remains unchanged, alveolar ventilation is about 70% higher at the end of pregnancy compared to pre-pregnancy (Bartnik, Gloxhuber, & Zimmermann, 1985). Therefore, it could be speculated that the above mentioned changes in the respiratory system may be associated with an increase in the inhalation of formaldehyde and consequently in formaldehyde exposure level. Formaldehyde can reach the fetus through the placenta (IARC, 2006). Further, Thrasher and Kilburn (2001) reported that elimination of formaldehyde is slower from fetal tissues than maternal tissues. Formaldehyde has been linked to reproductive and developmental toxicities based on human and animal studies (Duong et al., 2011; IARC, 2006; WHO, 2010). Reproductive toxicity includes menstrual irregularities, delayed conception, and endometriosis (Duong et al., 2011; Olsen & Dossing, 1982; Shumilina, 1975) while developmental toxicity involves spontaneous abortion (Duong et al., 2011; Hemminki, Kyyronen, & Lindbohm, 1985), congenital anomalies or birth defects (Hemminki et al., 1985; Saurel-Cubizolles, Hays, & Estryn-Behar, 1994), low birth weight, and premature birth (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002). There is no published study that was found that examined the effect of formaldehyde exposure on fetal growth during the second trimester of pregnancy. In addition, little is known about the relationship between formaldehyde exposure and birth outcomes, such as LBW and preterm birth. However, given that fetal growth, birth

35

weight, and LBW are highly associated (Cunningham et al., 2010; Fiscella, 2005), the limited animal and human studies that examined the effect of formaldehyde exposure on fetal growth during pregnancy will be analyzed in the following paragraphs. Saillenfait, Bonnet, and de Ceaurriz (1989) assessed the effects of inhaled formaldehyde on the embryonic and fetal development in Sprague-Dawley rats by exposing the rats to 0, 5, 10, 20, and 40 ppm formaldehyde for six hours per day from day 6 to 20 of gestation. They reported significant decreases in fetal body weight at 20 ppm and 40 ppm. In another study, Martin (1990) exposed groups of mated Sprague-Dawley rats by inhalation and whole body exposure technique for 6 hours per day, with formaldehyde at dosages of 2, 5, or 10 ppm from day 6-15 of gestation. The findings reported a slight but not significant decrease in fetal weights at 5 and 10 ppm. In a study by Overman (1985), pregnant hamsters were exposed to 0.5 ml formaldehyde solution (case group) or water (control group) directly through the skin (dermal exposure) by means of syringes on day 8, 9, 10 and 11 of gestation. The findings indicated that the mean fetal weight was slightly increased, but the difference was not significant. The animal studies vary in study design, route of exposure, duration of exposure, and length of follow-up time, all of which could influence the outcomes and therefore these findings cannot be generalized to human outcomes. The studies that have examined the effect of formaldehyde during pregnancy in occupational settings were focused on the effect of formaldehyde on pregnancy outcomes within the first trimester, which included abortion and congenital malformations. There is only one study that examined whether exposure to formaldehyde can result in an increased risk of LBW(Zhu, 2006). Zhu (2006) compared

36

pregnant women in Denmark who worked as laboratory technicians (case group) with pregnant teachers (control group) and found no significant difference in LBW between laboratory technicians and teachers. However , the results of the study by Zhu (2006) should be interpreted with caution because the investigators did not measure the personal exposure to formaldehyde and relied on self-reported data. In their study, Grazulevicience et al. (1998) conducted a population based study in Kaunas City, Lithuania on 4,343 women who delivered live infants in 1994. Based on the data collected from ecological monitoring posts regarding formaldehyde concentration in the districts, three exposure level areas were determined including: low exposure 1.94 µg/m3 (0.002 ppm) , moderate exposure 3.48 µg/m3 (0.003 ppm), and high exposure more than 4.67 µg/m3 (0.004 ppm). The infants’ birth date and sex were obtained from hospital records and the other information was obtained from the Kaunas municipal newborns registry through a personal interview with the mothers during perinatal care. Grazulevicience et al. (1998) found that LBW among women residing in high formaldehyde concentration areas was significantly higher than women residing in low exposure areas. Moreover, they reported that increasing levels of formaldehyde exposure resulted in increased incidence of LBW, with 48.3, 49.5, and 81.1 per 1000 in low, moderate, and high concentration areas, respectively. In a subsequent study in Kaunas, Lithuania, Maroziene and Grazuleviciene (2002) measured the formaldehyde level in residential districts through municipal monitoring sites and used it as a biomarker of formaldehyde exposure during pregnancy. The investigators grouped the formaldehyde concentrations into three categories as described above based on Grazulevicience et al. (1998). Results of the study showed that the

37

prevalence of LBW increased with increasing formaldehyde level from 2.9% in low formaldehyde exposure category to 4.0% in high exposure. When considering formaldehyde exposure by trimester, the risk of LBW was higher when exposure to formaldehyde happens in the first trimester. Recently, there has been an increased interest in the relationship between the level of indoor formaldehyde in residential dwellings and the causative health effects (ATSDR, 2010). This emphasis began after Hurricane Katrina, a category four storm affecting the US Gulf Coast between New Orleans, Louisiana and Mobile, Alabama in late August 2005, when The Federal Emergency Management Agency (FEMA) provided temporary houses with higher formaldehyde level to the displaced families (Centers for Disease Control, 2008). Before Hurricane Katrina, the focus of formaldehyde related research was the health effects of formaldehyde in occupational settings. Although the body of knowledge related to indoor formaldehyde levels in residential areas and health effects has been growing, research involving pregnant women and fetuses remains scarce. Some of the above reviewed articles are animal studies (Martin, 1990; Overman, 1985; Saillenfait, Bonnet, & de Ceaurriz, 1989), which cannot be generalized to humans. The study by Zhu (2006) considered laboratory technicians and teachers; however, it did not measure the level of formaldehyde exposure. The other two studies that examined the relationship between the formaldehyde exposure and LBW in general population (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002) measured outdoor formaldehyde level retrospectively thorough the municipality vapor monitors in outdoor, whereas formaldehyde is predominantly an indoor air pollutant (ATSDR, 2010). In addition,

38

these studies used birth weight as an outcome variable, which can be affected by many factors such as gestational age. The current study is one of the first to examine the effect of indoor residential sources of formaldehyde on formaldehyde exposure levels and the relationship between formaldehyde exposure levels and second trimester fetal growth biometric measures.

Tobacco Smoke Exposure and Fetal Growth Tobacco smoke is known to be associated with adverse pregnancy outcomes (Patrick et al., 1994; Wen et al., 2005). The most common adverse outcomes include abortion, infertility, LBW, IUGR, neonatal and prenatal mortality, and sudden infant death (Fescina et al., 2011). Braun et al. (2010) measured the level of tobacco smoke metabolites such as nicotine and cotinine in the infant meconium and showed that nicotine and cotinine pass through the placenta and can be measured in the meconium. Tobacco smoke exposure can be measured by self-report, however, different studies showed that it may underestimate the tobacco smoke exposure status (Lindqvist, Lendahls, Tollbom, Aberg, & Hakansson, 2002; Parna et al., 2005; Post, Gilljam, Bremberg, & Galanti, 2008). Exposure to tobacco smoke can also be detected by measuring nicotine, the active ingredient in tobacco, and its metabolite, cotinine. Cotinine has a longer half-life (20 hours), which allows detection for a few days after smoking (SRNT, 2002). Many studies reported the effect of tobacco smoke during pregnancy on birth outcomes such as LBW with making measured through self-report. Dejmek, Solansk y, Podrazilova, and Sram (2002) studied the impact of maternal tobacco smoke on birth

39

weight according to self-reported tobacco smoke status in a sample of 6,866 singleton births. They classified tobacco smoke status as nonsmokers and those who smoked 1-10 or >10 cigarettes per day. Dejmek et al. (2002) observed a significant reduction in mean birth weight in infants of mothers who smoked during pregnancy. Horta, Victora, Menezes, Halpern, and Barros (1997) studied the association between tobacco smoke during pregnancy and the frequency of LBW and IUGR. Soon after delivery, they interviewed mothers of 5,166 live births occurring in the city of Pelotas, Brazil, during 1993, and found that children whose mothers smoked during pregnancy had a significantly lower birthweight those of non-smoking mothers. They also found that tobacco smoke was associated with an increase in the prevalence of IUGR. Sclowitz, Santos, Domingues, Matijasevich, and Barros (2013) evaluated tobacco smoke during pregnancy and its association with repetition of LBW in another Brazilian sample. They included 565 mothers with previous histories of LBW in their study. They found that the prevalence of LBW in mothers who smoked was higher than among nonsmokers. Aagaard-Tillery, Porter, Lane, Varner, and Lacoursiere (2008) conducted a population-based retrospective analysis of term (37 weeks or longer) singleton pregnancies delivered in Utah from 1991 to 2001 and found that mean birthweight was significantly less and the prevalence of IUGR infants was significantly greater in tobacco-exposed infants. The significant limitation of the studies described above is that they relied on self-report on classification of tobacco smoke exposure, which is not as reliable as cotinine level in body fluids (serum, saliva, or urine) (Aurrekoetxea et al.,

40

2013; Pickett, Rathouz, Kasza, Wakschlag, & Wright, 2005). Some other studies used both cotinine and self-report to determine tobacco smoke exposure and its relationship with fetal growth. In their study, Iniguez et al. (2012) used the self-report method to collect information regarding smoking from 780 participants in a cohort of Spanish pregnant women in their second and third trimesters. Iniguez et al. (2012) classified maternal smoking as “non-smokers during pregnancy” without considering if they smoked before pregnancy, “smokers who give up smoking before week 12”, and “smokers still at week 12.” The last category included women who continued smoking at least until week 32 of pregnancy. Iniguez et al. (2012) reported that maternal smoking in all gestational weeks (12, 20, and 32) was inversely associated with BPD, AC, FL, and EFW at weeks 32 and 38. In 32-38 weeks, the effect was significant for AC and EFW, whereas BPD and FL were less affected at this stage. They also found that among smokers at week 12, fetal size was reduced around 2% at week 32 and around 10% at week 38 for all parameters. In this study the fetal biometric measurements were not adjusted based on confounding factors such as maternal race, whereas some studies reported that fetal growth is associated with race (Choudhary et al., 2013; Olsen et al., 2010). The other limitation is the ultrasonic results were not adjusted by maternal race and fetal gender that are known to be associated with fetal growth (Fescina et al., 2011). Furthermore, the investigators did not use the creatinine standardized cotinine level, which is important due to cotinine fluctuations in urine caused by changing specific gravity of urine (Benowitz et al., 2009). In a follow-up study, Iniguez et al. (2013) examined maternal tobacco smoke exposure through self-report and urinary cotinine level at week 32 of gestation. They

41

used the cutoff point of 50 ng/ml as an agreement point between self-report and urine cotinine level. Iniguez et al. (2013) found that FL and BPD were significantly smaller in mothers who were exposed to tobacco smoke (cotinine >50ng/ml) during early pregnancy (week 12 of gestation). The most affected parameter at week 34 was FL; which was shorter in the fetuses of smokers than in those of nonsmokers. In contrast, the least affected fetal growth parameter was AC. The limitation of this study is that the investigators used the third trimester’s ultrasound biometric measurements, which is not as accurate as first and second trimester (Cunningham et al., 2010). Jaddoe et al. (2007) studied the associations between tobacco smoke exposure during pregnancy with various fetal growth characteristics (HC, AC, and FL) among 7,098 pregnant women in Rotterdam, the Netherlands. Maternal smoking was assessed by self-report through questionnaires. Fetal ultrasound examinations and administration of questionnaires were planned for the first, second, and third trimesters. In their study, Jaddoe et al. (2007) found that tobacco exposure during pregnancy was associated with reduced growth in HC, AC, and FL, nevertheless, FL was the only parameter that was inversely associated with tobacco smoke exposure in the second trimester. They did not consider BPD as an ultrasound measurement. One of the limitations of this study is that they did not measure cotinine level as a biomarker of tobacco smoke and relied on selfreport. In addition, the ultrasonic results were not adjusted by maternal race and fetal gender that are associated with fetal growth (Fescina et al., 2011). In the current study, tobacco smoke was studied as a possible source of formaldehyde exposure in residential dwellings (ATSDR, 2010) and a risk factor for poor fetal growth as discussed above. The significant limitation of some of the above studies

42

was that they relied on self-report regarding tobacco smoke status. Furthermore, the other limitation of previous studies was that the level of urine cotinine was not standardized based on urine creatinine, which is important due to cotinine fluctuations in urine (Benowitz et al., 2009). In addition, in some of the studies, fetal biometric measurements were not adjusted based on race and gender that are associated with fetal growth (Fescina et al., 2011). In the current study, tobacco smoke exposure was measured using selfreport and creatinine-standardized urinary cotinine.

Measurement of Level of Formaldehyde Exposure during Pregnancy Formaldehyde Exposure Measures Formaldehyde can be measured in air and in biological fluids or tissues (National Toxicology Program, 2010). The major route of exposure to formaldehyde in human is respiratory tract (ATSDR, 2010). Personal exposure in high risk occupational areas, such as shoe companies, is measured by monitoring the breathing air (NIOSH, 2007) through use of a vapor monitor. A vapor monitor collects formaldehyde continuously when is exposed to air and levels can be analyzed by an instrumental technique of analytical chemistry, high-performance liquid chromatography (HPLC) (Advanced Chemical Sensors Inc, 2011). Formic acid in urine is another method to measure the exposure level to formaldehyde in occupational settings (ATSDR, 2010; NMS LABS, 2013). Formic acid is a metabolite of formaldehyde exposure and can be measured in urine and blood using gas chromatography (GC), another analytical chemistry method (NMS LABS, 2013). In addition, formaldehyde can be routinely monitored in ambient outdoor air through outdoor monitors, the most convenient method to measure formaldehyde in the

43

outdoor environment, with analysis using spectroscopic techniques (Salthammer, Mentese, & Marutzky, 2010). Formaldehyde can be measured in urine and blood (Heck et al., 1983; Shin, Ahn, & Lee, 2007). Direct measures of exposure to formaldehyde involves determination of formaldehyde or its major metabolite, formic acid (formate), in blood or urine (National Toxicology Program, 2010). Formaldehyde has been measured in urine and blood using gas chromatography-mass spectrometry and spectrophotometry (Baumann & Angerer, 1979; Kato, Burke, Koch, & Bierbaum, 2001; Spanel, Smith, Holland, Al Singary, & Elder, 1999). Recently, a novel fluorescence-based method had been developed to detect formaldehyde in urine (Cayman Chemical Company, 2011), using fluoroscopes (Pellegrini et al., 2007). However, in the pilot study, conducted by the principal investigator, and described in detail in Chapter 3, the method failed to yeild resutls. The Cayman Chemical company tested the method on different urine samples and agreed on the failure of the kit for detection of formaldehyde in urine. This resulted in the removal of thiscommercial formaldehyde detection kit from the market. Formic acid, as a metabolite of formaldehyde, has been measured in the urine in previous studies (Schmid, Schaller, Angerer, & Lehnert, 1994) using gas chromatography (GC) method. However, there is some evidence that biological monitoring of exposure to formaldehyde by measurement of formic acid in urine is not an appropriate method with low level exposure (Gottschling, Beaulieu, & Melvin, 1984; Schmid et al., 1994; Triebig, Schaller, Beyer, Muller, & Valentin, 1989; Yasugi et al., 1992). This could be due the difference in levels measured at inhalation versus by urinary excretion. In an early animal

44

study, only 17% of inhaled formaldehyde was excreted in the urine of rats (Heck et al., 1983).

Formaldehyde Exposure: Formic acid versus Vapor Monitor There were no published studies found regarding the level of formic acid and use of vapor monitors in non-occupational settings or in pregnant women. Literature in this section is related to the relationship between formic acid and vapor monitor badges in determining the level of formaldehyde exposure in occupational settings.Gottschling et al. (1984), measured formaldehyde exposure in 15 veterinary medicine students through air monitors. They also measured formic acid in urine, as a metabolite of formaldehyde. First they established normal baseline levels of urinary formic acid for each subject before exposure to formaldehyde. Then, after exposure to 0.5 ppm formaldehyde, three sets of pre- and post-exposure urine samples were taken. They found no differences between levels of formic acid before and after exposure to formaldehyde. The main limitation of this study is the ethical issue of exposing participants to a level of formaldehyde which is higher than the limit for occupational exposure of 0.016 ppm based on NIOSH (2007). Another limitation is the small sample size, which also threatens the validity of the study. Schmid et al. (1994), studied the suitability of the formic acid excretion in the urine as a parameter for the biological monitoring of inhalational exposure to formaldehyde. They investigated the level of formaldehyde using personal air sampling in 70 individuals who were not occupationally exposed to formaldehyde. They also measured the level of formaldehyde in 30 medical students, who were exposed to a short

45

but intensive inhalational of formaldehyde during an anatomical dissection course and eight employees of a pathological-anatomical laboratory. They examined fluctuations of formic acid excretion in the urine during a work week with a continuous exposure to formaldehyde below the value of 0.5 ppm. Schmid et al. (1994), reported the value of 23 mg/g creatinine standardized formic acid as the upper norm level of formaldehyde in non-pregnant adults. In the group of pathologists who were exposed to higher level of formaldehyde (more than 0.5 ppm), no significant increase in the formic acid concentration in urine was detected. In addition, in the group of medical students who had a short but intensive exposure to formaldehyde (0.32-3.48 ppm) the formic acid concentration in the urine did not change significantly before and after exposure. The main limitations of this study are the small sample size and time of urine sampling. Time of urine collection is important because, as with cotinine, the concentration of formic acid can be altered based on the specific gravity of the urine (Benowitz et al., 2009). Only a few studies have examined formic acid as a biomarker of formaldehyde exposure. This could be due to the fact that formic acid has not been found to be a specific test to measure formaldehyde exposure in adults (Gottschling, Beaulieu, & Melvin, 1984 Schmid et al., 1994); however, no published studies to date have examined formic acid as a biomarker of formaldehye in pregnant women. The current study is the first study regarding measurement of the level of formic acid in pregnant women and determination of the relationship between formic acid and formaldehyde (vapor monitor) in pregnant women.

46

Standards for Formaldehyde Exposure Due to the carcinogenic effect of formaldehyde, no level is known to be risk free (California Air Resources Board, 2005). There is no standard for formaldehyde in nonoccupational setting. A standard for formaldehyde exposure has been set for occupational settings (less than 0.016); however, it is not specified for pregnant women (NIOSH, 2007). The Agency for Toxic Substances and Disease Registry (ATSDR) classified formaldehyde exposure as acute (14 days or less), intermediate (15-365 days), and chronic (635 days or more) (ATSDR, 2010). However, the California Environmental Protection Agency recommends exposure limits of less than 0.027 ppm for residential areas (California Environmental Protection Agency, 2004). The World Health Organization (WHO) recommends a concentration level (air quality guideline value) of 0.1 mg/m3 (0.08 ppm) formaldehyde, for a 30-minute period, for the general population (WHO, 2010); each ppm formaldehyde equals 1.23 mg/m3 (NIOSH, 2007). Sweden proposed the guideline value of 0.009-0.048 ppm (12-60 µg/m3) for indoor settings (Gustafson et al., 2005) . The European Union has established a 30-minute average limit value of 0.024 ppm (30 µg/m3) (Bruinen De Bruin et al., 2008). The Canada Department of National Health and Welfare set an hour exposure limit of 0.1 ppm and an 8 hour exposure limit of 0.04 ppm (Health Canada, 2012). In Japan, there are two limit values for environmental exposure to formaldehyde, 0.08 ppm in general workplaces and 0.25 ppm for specific workplaces such as formaldehyde production factories (Ohmichi et al., 2006). The highest acceptable concentration of 0.1 ppm has been designated as an action or ceiling level by the U.S. EPA (2008).

47

The limits that were discussed above are set for prevention of allergic reactions such as eye irritation and asthma hospitalization in general and are not specific for children or pregnant women or for chronic outcomes, such as carcinogenic effects. However, Minimal Risk Levels (MRLs) have been estimated for formaldehyde exposure through inhalation to protect sensitive populations, including children and seniors (ATSDR, 2010). The MRL derives from reliable and sufficient evidence and is “an estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse non-cancer health effects over a specified duration of exposure” (ATSDR, 2010, p. 10). The MRLs of 0.04 ppm, 0.03 ppm, and 0.008 ppm have been derived for acute-duration inhalation exposure (14 days or less), intermediateduration inhalation exposure (15–364 days), and chronic-duration inhalation exposure (365 days or more), respectively (ATSDR, 2010). Despite the fact that almost no governmental or private organizations consider the level of formaldehyde in designing and building residential dwellings, in this study, the MRL of 0.03 ppm was considered a cut-off point for classifying high and low formaldehyde exposure level. This MRL has been considered for 15-364 days (ATSDR, 2010), which covers the duration of pregnancy.

Formaldehyde Exposure in Pregnant Women Studies which measured personal formaldehyde exposure level using vapor monitors have been conducted in children and in non-pregnant adults. In addition, there are several other studies that examined formaldehyde exposure level through measuring the formaldehyde level in residential dwellings using vapor monitors. Given that indoor

48

formaldehyde levels can be used as a surrogate for personal formaldehyde exposure level (Dannemiller et al., 2013; Dassonville et al., 2009), in this section the evidence related to personal formaldehyde exposure in children and non-pregnant adults, and the formaldehyde levels in the residential dwellings will be discussed. Lazenby et al. (2012), studied the relationship between personal exposure formaldehyde concentration in children in Australia. They included 41 elementary school children (9-12 years old, both male and female) and measured the formaldehyde exposure level, using vapor monitors, for two 24 hours periods in two seasons (summer and winter). Simultaneously, they measured participants’ residential dwellings’ indoor and outdoor formaldehyde levels. Participants completed 24-hour daily activity diaries and a questionnaire about lifestyle and behavior. The results of this study showed the mean personal exposure formaldehyde concentration in children was 0.009 ppm. The researchers reported a positive correlation between personal exposure concentrations and indoor formaldehyde level. They did not find any relationship between season (winter or summer) and formaldehyde levels, which could be due to either mild meteorological conditions experienced during both seasons, or the small sample size and mostly older residential dwellings. Lioy et al. (2011), measured the level of 24 hour personal exposure to formaldehyde and other pollutants in the Waterfront South (54 participants) and Copewood-Davis (53 participants) neighborhoods in New Jersey. Waterfront South was considered a hot spot for air pollutants, having elevated concentrations of air toxics compared with those of other more distant areas in New Jersey and across the United States. However, Copewood-Davis was considered a less polluted neighborhood and was

49

chosen as a control to compare concentrations of air pollutants. The researchers found personal formaldehyde exposure levels of 0.014 ± 0.012 ppm (16.8 ± 15.5 µg/m3) in Waterfront South and 0.013 ± 0.14 (16.0 ± 16.7 µg/m3) in Copewood-Davis with no significant difference between communities. One of the limitations of this study was the small sample size. In addition, the investigators did not consider the role of occupational exposure. Jurvelin, Vartiainen, and Jantunen (2001), measured the personal 48-hr exposures to formaldehyde of 15 randomly selected participants during the summer and fall in 1997 using Sep-Pak DNPH-Silica cartridges in Helsinki, Finland. This cartridge is coated with acidified 2, 4-dinitrophenylhydrazine (DNPH), which is a chemical compound that is often used to qualitatively test for VOCs and reacts with formaldehyde. The DNPH derivatives are eluted from the sampling cartridges using HPLC (California Environmental Protection Agency, 2001). Therefore, the cartridge’s function is similar to a vapor monitor (California Environmental Protection Agency, 2001). In addition to personal exposures, simultaneously the investigators measured each participant's residence (indoor and outdoor) formaldehyde level. They reported a mean personal formaldehyde exposure level of 0.021 ppm. Personal exposure was lower than indoor residential concentrations, and ambient air concentrations were lower than both indoor residential concentrations and personal exposure levels. Personal 48 hour exposure to formaldehyde was positively correlated with indoor residential dwelling formaldehyde level, but not outdoor formaldehyde level. One of the limitations of this study was the small sample size. In addition, the investigators did not consider the type and age of the residential dwellings.

50

Gustafson et al. (2005) conducted a study to measure formaldehyde exposure level in the general population (20-50 years old male and females), using vapor monitors, in two Swedish cities, Borås (campaign A) and Göteborg (campaign B) , among a total of 65 randomly selected subjects. In Campaign A, the sampling was performed for 24 hours in October-November 1999. In campaign B, the sampling was performed for 6 days in 2000 (the season of data collection for campaign B was not mentioned in the report). Simultaneously, they measured individual indoor (participants’ bedroom) for campaign A and individual indoor and outdoor (outside the residential dwellings) for campaign B. Repeated measurements were also conducted in order to study the variability between and within individuals. Characteristics of the residential dwellings were evaluated using a questionnaire, which included time spent at home and household activities such as tobacco smoke at home. These investigators reported a median personal formaldehyde exposure of 0.02 ppm for both campaigns (22 – 23 µg/m3). Bedroom concentrations were slightly higher than personal exposure, while an outdoor concentration, which was measured only in campaign B was lower than personal or indoor exposure. Similar to the previous studies, this study is limited by a small sample size. Unlike other studies, Gustafson et al. (2005) did consider the square footage of the residential dwelling in determining the formaldehyde exposure level; however, they did not consider the indoor sources of formaldehyde exposure. Roda et al. (2011), measured formaldehyde in a random sample of 196 residential dwellings of infants (full-term with a birth weight of >2500 grams) in Paris, France. They measured formaldehyde using vapor monitors that were placed in the infant’s bedroom for 7 days. The aim of this study was to assess the influence of formaldehyde exposure on

51

lower respiratory tract infection, because there is a link between formaldehyde exposure and asthma in children (McGwin, Lienert, & Kennedy, 2010; Raaschou-Nielsen et al., 2010). However, in this section only the formaldehyde exposure level is considered. Roda et al. (2011) collected the health data from parents by regular self-administered questionnaires and reported that infants were exposed to an average of 0.02 ppm (SD = 0.004) of formaldehyde. However, the main limitation with this study was measurement of formaldehyde level for 7 days that raised concerns regarding saturation of the formaldehyde level in the monitors, which can result in over estimation of the exposure level (Advanced Chemical Sensors Inc, 2011). Dannemiller et al. (2013) used a different sampling method to measure the formaldehyde level in residential dwellings. They used Kitagawa 710 formaldehyde detector tubes. They measured formaldehyde in dwellings of 70 asthmatic children and adult patients in the Boston, Massachusetts area. The goals of their study were to evaluate the new formaldehyde measurement method and to determine what housing factors affect formaldehyde concentration. In this section, only the results related to the formaldehyde exposure level are discussed; the effect of housing factors on formaldehyde exposure level will be discussed later in this chapter. Dannemiller et al. (2013) conducted sampling in the kitchen on top of the refrigerator using their newly developed sampler for 30 minutes and found a mean of 0.0351 ppm, with a range of 0.005 to 0.132 ppm. The main limitation of this study is the use of a new sampler that had not been tested before, which makes the comparison of the results difficult. In addition, they only exposed the sampler to the air for 30 minutes, without considering what type of activities had been done within the selected 30 minutes. This is a limitation because, based on the recent studies

52

(Dassonville et al., 2009),the formaldehyde exposure level is associated with household activities such as cooking and cleaning. Sexton, Petreas, and Liu (1989), measured the one-week, average indoor formaldehyde levels in summer (July-August 1984) and winter (February-March 1985) in mobile homes within the State of California . As formaldehyde emissions from building materials decrease over time (ATSDR, 2010; Dassonville et al., 2009; WHO, 2010), Sexton et al. (1989) used an age-stratified random sample technique and selected 470 mobile homes out of 50,000 mobile home in California and measured formaldehyde using vapor monitors. Results indicated a mean one-week formaldehyde exposure level of 0.07 - 0.09 ppm. In this study formaldehyde levels appeared to be decreasing inside mobile homes manufactured since about 1980, which is probably a result of the increased use of low-formaldehyde-emitting building materials. The main limitation of this study is the sampling time of 7 days, which, as discussed before, may increase saturation of formaldehyde in the monitor and result in overestimating of the formaldehyde level (Advanced Chemical Sensors Inc, 2011). Murphy et al. (2013) identified formaldehyde exposure levels in occupied trailers, including travel trailers, park models, and mobile homes, in relationship to the characteristics of trailers, including model and manufacturer. A sample of 519 FEMA supplied trailers, out of 46,970 trailers, was identified in Louisiana and Mississippi, in December of 2007 to February of 2008, using stratified random sampling. The investigators collected a one hour indoor formaldehyde level, using standard industrial hygiene pumps and Supelco S10 LpDNPH cartridges (Supelco, St. Louis, MO, USA). In addition, they measured indoor temperature and relative humidity concurrently. They also

53

administered a short questionnaire to obtain demographic information about trailer residents and their typical daily activities. Residents were asked to configure doors and windows as they would have them while they slept, and they were not allowed to cook or smoke in the trailer during the one hour sampling period. Murphy et al. (2013) reported that formaldehyde levels among all trailers ranged from 0.003 to 0.590 ppm, with a mean of 0.077 ppm. The results regarding the relationship between trailer type and season with the formaldehyde level will be discussed later in this chapter. However, the low level formaldehyde level in this study can be related to the timing of data collection during winter. Therefore, levels measured in this study may underestimate peak summer levels when trailers would be hotter and more humid. Residential dwelling temperature and humidity are known to be positively correlated with the level of indoor formaldehyde (ATSDR, 2010; Dannemiller et al., 2013; Dassonville et al., 2009). Stock and Mendez (1985), conducted a survey of indoor air quality under warm weather conditions, in a variety of Houston area residential dwellings. They surveyed a total of 78 residential dwelling from 13 separate geographic clusters. Seven hour air sampling was performed in a room frequently used by the occupants (usually den or living room). They reported a distribution of indoor formaldehyde concentrations ranging from less than 0.008 ppm to 0.29 ppm, with a mean of 0.07 ppm. Approximately 15% of the monitored residences had concentrations greater than 0.1 ppm. The formaldehyde level for non-conventional homes was significantly higher than conventional homes. They also found inverse relationship between the age of the residential dwellings and the level of formaldehyde. Considering that formaldehyde exposure level is related to the indoor activities, one of the limitations of this study is that the investigators used vapor

54

monitors for only seven hours, and it is not clear what part of the day or night they included in their study and what types of household activities were happening during the sampling period. This study is the first in regards to the formaldehyde exposure level in pregnant women; therefore, there is lack of evidence to review related to this target population. The reviewed studies have measured personal or residential formaldehyde exposure level in non-pregnant adults. As mentioned earlier, these studies cannot represent formaldehyde exposure level during pregnancy due to some physiological changes and some behavioral practices during pregnancy. For example, some common practices during pregnancy, such as buying new furniture and remodeling a nursery, are potential sources of formaldehyde exposure in residential dwellings, but have not been reflected in those studies. Therefore, caution should be exercised when making comparisons among the findings of different studies, because differences in personal exposures might result from variances in study populations and daily practices. Moreover, the differences between the personal and indoor formaldehyde concentrations in various studies might be attributable, at least in part, to the various sampling monitors and devices and different exposure times, which may affect the saturation of formaldehyde in the selected monitors. The current study also is the first study regarding measurement of the level of urinary formic acid in pregnant women and determination of the relationship between formic acid and formaldehyde (vapor monitor) levels in pregnant women. There are a few previous studies that measured formic acid in urine specimens of non-pregnant males and females (Gottschling et al., 1984; Schmid et al., 1994) that were reviewed earlier.

55

Indoor Residential Sources of Formaldehyde Exposure The major sources of indoor formaldehyde exposure are mainly building and household materials such as paints, adhesives, wall boards, ceiling tiles, carpets, furniture, fiberglass, fabrics, molding, and insulation foams (IARC, 2006). Dassonville et al. (2009) revealed that almost all houses have formaldehyde, however, the level of formaldehyde is related to the type and age of the residential dwelling, as well as the type and manufacturer of the building materials. One of the main indoor sources of formaldehyde in residential dwellings is wood products. The EPA indicates that in homes with significant amounts of new pressed wood products, the level of formaldehyde can be greater than 0.3 ppm (EPA, 2011). This is significantly above the MRLs of 0.03 and 0.008 ppm, which are considered by ATSDR for indoor intermediate-duration inhalation exposure (15–364 days) and chronic-duration inhalation exposure (365 days or more), respectively (ATSDR, 2010) and the indoor formaldehyde exposure limit of 0.027 ppm set by the California Environmental Protection Agency (2001). Several studies have reported the association between building materials and the level of formaldehyde. As discussed and critiqued earlier, Roda et al. (2011), measured formaldehyde in a random sample of 196 residential dwellings of infants (full-term with a birth weight of >2500 grams) in Paris, France, using vapor monitors that were placed in the infant’s bedroom for 7 days. Roda et al. (2011) observed higher levels of formaldehyde in houses that were built with wall coating paint and particle boards. However, as discussed earlier the major limit of this study is using 7 days sampling that may increase the saturation of formaldehyde in the monitors (Advanced Chemical Sensors Inc, 2011).

56

Maddalena, Russell, Sullivan, and Apte (2009) measured the level of formaldehyde in four unoccupied temporary housing units, including travel trailers, park models, and mobile homes from stock at the FEMA staging yard in Purvis, Mississippi using vapor monitors. The sampling was conducted in November. The higher level of formaldehyde concentration was attributed to the use of composite wood products, sealants, and vinyl covering in combination with low air exchange rates relative to material surface area. The limitations of this study are small sample size and the season of data collection. As discussed earlier, the level of formaldehyde emission is higher in warmer seasons; therefore the level of formaldehyde may be underreported due to the selected season. Dassonville et al. (2009), examined indoor airborne aldehyde levels, including formaldehyde within 1 year (1, 6, 9 and 12 months) in the bedroom of 196, randomly selected Paris infants. Formaldehyde was collected using a vapor monitor (Radiello, Fondazione Salvatore Maugeri – IRCCS, Italy), which was placed in the infant’s bedroom for seven days. Predictors for formaldehyde concentration were measured using a questionnaire, which included questions about home characteristics and living conditions, such as, type of residential dwelling, date of building construction, number of occupants, home surface area, heating and cooking system; type and age of wall and floor coverings in the infant’s bedroom, living room and kitchen; quantity and age of particleboard furniture in the infant’s bedroom and living room, signs of dampness, and ventilation system, frequency and length of windows opening; and frequency of use of air fresheners, cleaning products, and decorating or crafting. They also measured indoor temperature and relative humidity concurrent with measurement of formaldehyde.

57

Dassonville et al. (2009), reported the formaldehyde levels were higher in houses than in apartments, particularly when the building was newer. The amount of particle board furniture in the infant’s bedroom, and the presence of pressed wood products for flooring in the residential dwelling, significantly increased formaldehyde levels, especially when these materials were new (less than a year old). These investigators did not find any relationship between interior covering such as wall paper or wall to wall carpet with formaldehyde levels. The use of a wood burning fire in the dwelling tended to increase formaldehyde levels; however, a gas heating system was not associated with formaldehyde level. Furthermore, formaldehyde level was negatively associated with the length of window opening. This study has many strengths that include serial measurement of formaldehyde level, large sample size, and measurement of humidity concurrent with formaldehyde, which is associated with the level of formaldehyde emission (ATSDR, 2010). However, one of the limitations of this study is not considering the square footage of the studied residential dwellings, which is correlated with the formaldehyde level in dwellings (Hodgson, Beal, & McIlvaine, 2002). As discussed earlier, Dannemiller et al. (2013), who measured formaldehyde level in residential dwellings in 70 asthmatic children and adult patients in the Boston, Massachusetts area, found higher levels of formaldehyde in the presence of decorative laminates, fiberglass, and/or permanent press fabrics in residential dwellings. They also found significant relationship between formaldehyde concentration and presence of decorative laminates, fiberglass, and/or permanent press fabrics. This study has been critiqued earlier. The age of the residential dwelling is valuable from the stand point of the level of

58

formaldehyde given that as the residential dwelling ages, the concentration of formaldehyde in the building materials decreases (IARC, 2006). Maruo, Yamada, Nakamura, Izumi, and Uchiyama (2010), determined the formaldehyde level in 5 Japanese houses using sensor elements. These sensors contains 1-phenyl-1,3-butandione that reacts between formaldehyde and b-diketone with ammonium ion, and this mechanism was used for detecting formaldehyde in this study. In their study, Maruo et al. (2010) reported the highest formaldehyde concentration of 0.07 ± 0.01 ppm (92 ± 15 µg/m3) for apartments 0–2 years old after their renovation and a simple linear relationship was found between formaldehyde concentration and the age of the apartment. The limitation of this study is very small sample size. In addition, the method that they used to detect formaldehyde is different than for other studies, which mostly used vapor monitors, and this makes the comparison difficult. As previously mentioned, Lazenby et al. (2012), examined personal exposure formaldehyde concentrations in children in Australia. They did not find any relationship between the formaldehyde level and age of the residential dwelling. This could be due to a low level of formaldehyde used in the building materials based on Australian standard (Ecos, 2006). Formaldehyde can also be produced by the blending of indoor air pollutants, such as household cleaning materials, perfume, nail polish, candles, or air fresheners and ozone, (Liu et al., 2004; Uhde & Salthammer, 2007). Petry et al. (2013), conducted a pilot study to examnine VOCs’ emission on the basis of existing standards, including formaldehyde. They measured formaldehyde emission from a candle under a chamber using vapor monitors and repeated measurements twice. They found that formaldehyde is

59

emitted as a product of burning candles. The limitation of this study is that they used only one type of candle and type of candle is associated with the level of formaldehyde emitted (Liu et al., 2003). Lamorena and Lee (2008) examined the emissions of VOCs, including formaldehyde, from an aliquot amount of liquid car air freshener (0.5 ml) and identified the formation of ultra-fine particles and secondary gaseous compounds during the ozoneinitiated oxidations with emitted VOCs. They placed air freshener in the bottom of 50 mL beaker and allowed it to emit the biogenic VOCs for one hour inside the chamber bag. The temperature was controlled at 30 and 40 C. The chamber was flushed with ozone and purified air for several hours before and after each experimental run. First, they identified primary constituents emitted from the car air freshener, which were alpha-pinene, betapinene, p-cymene, and limonene. Later, they injected ozone into the chamber and tested the formation of ultra-fine particles (4.4-160 nm) and measured them. The total reaction time was approximately four hours. They found that the irritating secondary gaseous products formed during the ozone-initiated reactions include formaldehyde, acetaldehyde, acrolein, acetone, and propionaldehyde. They also reported that ozone concentration (50 and 100 ppb) and temperature (30° and 40° degrees of Celsius) significantly correlated with the formation of formaldehyde and other VOCs during the ozone-initiated reactions. The main limitation of this study is the small sample size. Alaves et al. (2013), examined VOCs in 12 randomly selected commercial nail salons in Salt Lake County, Utah. They also considered physical/chemical parameters (room volume, carbon dioxide [CO2] levels) of nail salons. VOCs levels were collected using summa air canisters and sorbent media tubes for an 8-hour period. The results of

60

this study showed a relationship between salon physical/chemical characteristics and the level of formaldehyde and other VOCs. In addition, they found formaldehyde above the NIOSH recommendation of 0.016 ppm in 58% of the selected salons. This study is also limited by a small sample size. Cooking and heating systems also have been found to be related to the formaldehyde level (IARC, 2006). As discussed earlier, Dassonville et al. (2009) examined formaldehyde level and its variability within one year (1, 6, 9 and 12 months) in the bedrooms of 196 randomly selected Paris infants and found that the use of a wood burning fire in the dwelling tended to increase formaldehyde levels. In their study, Dassonville et al. (2009) reported that a gas heating system was not associated with the formaldehyde level. Also as noted earlier, Lazenby et al. (2012) did not find any relationship between the formaldehyde level and the type of heating or cooling system. Lindstrom, Proffitt, and Fortune (1995), studied the number of occupants and their activities residential dwellings in relation to formaldehyde exposure level. They assessed the level of formaldehyde in residential dwellings in an experimental community where houses were made of low pollutant materials and compared them to conventional houses with the same size and price range. They measured formaldehyde using vapor monitors three different times in six dwellings: (1) shortly after construction, (2) before the houses were occupied, and (3) five months after occupancy. They found higher formaldehyde levels in the conventional homes than in the experimental homes. In addition, Lindstrom et al. (1995) found that the level of formaldehyde was 0.007-0.054 ppm, with a mean of 0.021ppm before occupancy, and 0.027-0.066 ppm, with a mean of 0.040 ppm five months after occupancy. The major limitation of this study is its small

61

sample size. As discussed earlier, Roda et al. (2011) established higher levels of formaldehyde in houses that had more than 3 occupants. The emission of formaldehyde from the building materials has been shown to be higher in warm seasons of the year, high humidity, and high indoor temperature (ATSDR, 2010). Several studies that were reviewed earlier in this section found the highest formaldehyde levels during summer (Dannemiller et al., 2013; Dassonville et al., 2009; Gustafson et al., 2005; Lioy et al., 2011). However, Lazenby et al. (2012), did not find any relationship between season (winter or summer). Dassonville et al. (2009) and Murphy et al. (2013) found that formaldehyde level was positively associated with temperature and humidity, and negatively associated with the length of window openings. Dannemiller et al. (2013) found a significant relationship between formaldehyde concentration and humidity and warmer seasons. Maruo et al. (2010) reported that formaldehyde level in a room containing furniture increased by 10% when the temperature increased by 1Celisius degree. In summary, there is an agreement among researchers that the main sources of formaldehyde in residential dwellings are building materials (Dannemiller et al., 2013; Dassonville et al., 2009; Maddalena et al., 2009; Roda et al., 2011; Wieslander, Norback, Bjornsson, Janson, & Boman, 1997) and that formaldehyde levels vary based on the age and type of residential dwelling (ATSDR, 2010; EPA, 2008). In addition, certain household practices such as use of candles and air fresheners have been identified as potential sources of formaldehyde (ATSDR, 2010; EPA, 2008; Petry et al., 2013; Salthammer, 2013). The relationship between season, indoor temperature, and humidity is well established as well (Dannemiller et al., 2013; Dassonville et al., 2009; Gustafson

62

et al., 2005; Lioy et al., 2011). The current study is the first to examine indoor residential sources of formaldehyde during pregnancy, including residential characteristics and household practices, and their relationships with fetal growth.

Tobacco Smoke Exposure Measures During Pregnancy Metabolic clearance of nicotine is increased during pregnancy, which results in marked acceleration of cotinine production (Dempsey et al., 2002) . In addition, due to the physiological changes during pregnancy the half-life of cotinine is substantially decreased in this period (Dempsey et al., 2002). In non-pregnant adults, the average halflife of cotinine is approximately 12 hours; however, during pregnancy the cotinine halflife is a little less than 9 hours (Benowitz & Jacob, 1994). These metabolic changes are associated with a considerable decrease in the excretion of cotinine in urine based on a study by Dempsey et al. (2002). In addition, Wu et al. (2008) found that the clearance of cotinine in urine differs in different trimesters, with a mean of creatinine standardized cotinine levels of 6.10 mg/g in first trimester, an increase up to 24.18 mg/g in the second trimester, and a decrease to 5.78 mg/g in the third trimester. The mechanism for these metabolic changes during pregnancy is not clear. However, it could be related to the increase of cardiac output and blood volume during pregnancy (Cunningham et al., 2010), which can be associated with the acceleration of liver blood flow. Increase in liver blood flow can increase nicotine metabolic clearance (Dempsey et al., 2002). Tobacco smoke can be measured through cotinine or self-report. However, evidence shows that self-report is not as reliable as cotinine. England et al. (2007), examined tobacco exposure misclassification among 4,289 eligible pregnant women.

63

Healthy, nulliparous women at 13-21 weeks’ gestation were selected from the Calcium for Preeclampsia Prevention (CPEP) trial, a randomized, double-blind clinical trial of the effects of daily calcium supplementation on the incidence of preeclampsia and gestational hypertension. The participants were asked regarding their smoking status before urine collection. England et al. (2007) collected 24 hour urine sample and first morning void specimen at the screening prior to study enrollment and at two subsequent visits at 26-29 and 36 weeks of gestational age. Urinary cotinine concentration was used to validate quit status. Of 4,289 women enrolled, 508 (11.8%) were self-reported smokers and 737were self-reported quitters. Of 737 self-reported quitters with a valid cotinine measurement (cut of point of 50ng/ml), 21.6% had evidence of active smoking and were reclassified as smokers. With the addition of these cotinine-validated smokers to self-reported smokers the proportion of the study population considered to be considered smokers during pregnancy increased from 11.9% to 15.5%. One of the strengths of the study by England et al. (2007) is the collection of 24 hour urine sample, which is more reliable than spot urine sample in determining the level of cotinine (Benowitz et al., 2009). The limitations of this study are regarding the potential misclassification of smoking status. For instance, participants’ smoking history was obtained before their urine specimens were collected; therefore, the investigators were not able to directly validate the accuracy of maternal self-report in this study. In addition, although cotinine is the gold standard for validation of nonsmokers (SRNT, 2002), environmental tobacco smoke (ETS) exposure may have occurred in some instances and resulted in misclassification of pregnant women as smokers. The other limitation was regarding the physiological changes of cotinine clearance within various

64

trimesters (Dempsey et al., 2002). The classification of tobacco smoke status might be affected based on trimester at the time of urine sample collection, as discussed earlier in this section. Aurrekoetxea et al. (2013), interviewed 2,263 singleton pregnant women (10-13 weeks of gestational age) in Spain in their third trimester and collected morning urine samples. They considered women who reported smoking occasionally or daily as smokers regardless of their urine cotinine levels. Women who had urine cotinine levels above 50ng/ml were considered smokers. In their study, Aurrekoetxea et al. (2013) found that 77.6% of the women reported that they did not smoke and had urine cotinine levels of less than 50 ng/ml. However, 0.8% had urine cotinine levels less than 50 ng/ml despite claiming to smoke and 3.9% reported that they did not smoke but were found to have urine cotinine levels above 50 ng/ml. The proportion of true positive smokers in this study was 17.7%, and 32.4%, were occasional or regular smokers when they became pregnant. Based on the finding of their study, Aurrekoetxea et al. (2013) reported an optimal cut-off point of 82 ng/ml (95% CI 42 to 133) for smokers and 27 ng/ml to discriminate occasional smokers from non-smokers. Due to the differences in physiological mechanism and household activities in various ethnicities, such as opening windows or smoking outdoors instead of indoors, it acceptable for each study to set a new cut-off point for classifying smoking status of its participants (Center for Disease Control, 2006). The limitations of this study are similar to the study by England et al. (2007), with an exception that Aurrekoetxea et al. (2013) asked the status of smoking concurrent with urine sampling. However, the race and culture of the participants in the

65

study byAurrekoetxea et al. (2013) are different than England et al. (2007) (Spain versus the US), which makes the comparison difficult. The cut-off point for tobacco smoke status differentiation is different between the two studies as well. This also may interfere with the interpretation of results from different studies. Pickett et al. (2005), examined self-reported smoking patterns and compared them to patterns of creatinine standardized urinary cotinine levels in a prospective study of 998 pregnant women who were less than 20 weeks of gestational age in Boston, Massachusetts. They examined urinary cotinine cut-off values of 100, 200, and 500ng/mg and found that a cut-off value of 200 ng/mg had the best balance of true positive and true negative reports. This study has the same limitations of the study by England et al. (2007), with an exception that the time of data collection for tobacco smoke self-report and urine collection was concurrent. Jhun et al. (2010), studied urinary cotinine levels and self-reported smoking among a total of 1,090 pregnant women in Korea and the factors associated with smoking during pregnancy. The subjects were selected from pregnant women who visited 30 randomly sampled obstetric clinics and prenatal care hospitals in Korea in 2006. Selfreporting and urinary cotinine measurement used to determine the tobacco smoke status. In this study only 0.55% of participants self-reported smoking. The prevalence of smoking with urinary cotinine measurement, with a cut-off point of >100 ng/ml for smokers, was 3.03%. The agreement between self-reported smoking status and urinary cotinine measurement was 0.20. These investigators found urinary cotinine levels were lowest in the second trimester of pregnancy. They reported that early gestational period, low educational level, and being married to a smoker were significant risk factors for

66

smoking during pregnancy. The limitations of this study are similar to the study by Aurrekoetxea et al. (2013), which was done in Spain. Based on the above studies, self-report may under estimate the rate of tobacco smoke exposure during pregnancy and creatinine standardized urine cotinine level seems to be more accurate than self-report (Aurrekoetxea et al., 2013; England et al., 2007; Pickett et al., 2005). However, there are some common limitations with these studies, including not using a unique cut-off point for urine cotinine, selecting participants from different trimesters, and enrolling participants from different ethnicities. These limitations make comparison of the results of different studies difficult.

Tobacco Smoke and Formaldehyde Exposure Formaldehyde is a major by-product of the oxidation of tobacco, which has been reported in both mainstream and side stream, and in substantial levels in ETS (Godish, 1989; Parrish & Harward, 2000). Main stream concentration of formaldehyde ranged from 0.008 to 0.08 ppm in different studies (Godish, 1989; Hoffmann, Hoffmann, & ElBayoumy, 2001; Kensler & Battista, 1963) and in the reports of the Surgeon General (Center for Disease Control, 2006). This difference can be related to the type and brand of tobacco (Center for Disease Control, 2006; USDHEW, 1979). Formaldehyde levels associated with ETS indicate that the level of formaldehyde in rooms where there is a smoker is higher than nonsmoker’s rooms (Center for Disease Control, 2006; USDHEW, 1979). ETS, however, is only a relatively minor contributor to indoor air concentrations of formaldehyde (Jenkins, Guerin, & Tiomkins, 2000). Results in two studies presented earlier, Gustafson et al. (2005). and Jurvelin et al. (2001), found no relationship between

67

the level of formaldehyde exposure in residential dwellings and tobacco smoking. One of the limitations of the studies by Gustafson et al. (2005). and Jurvelin et al. (2001) is that they relied on self-report of tobacco smoke exposure, and, as discussed before, this method can underestimate the tobacco smoke exposure during pregnancy (Aurrekoetxea et al., 2013; Pickett et al., 2005). Dassonville et al. (2009), examined indoor airborne aldehyde levels, including formaldehyde, using vapor monitors and their variability within 1 year (1, 6, 9 and 12 months) in the bedroom of 196, randomly selected Paris infants. Moreover, they identified tobacco smoking in the residential area through questionnaire and examined the level of nicotine for seven days using indoor monitor. In this study, parental smoking and nicotine levels did not have significant effects on formaldehyde levels. Bono et al. (2006) did not specifically study the relationship between formaldehyde and tobacco smoke, however, they conducted a pilot study to examine the correlation of human exposure to formaldehyde with N-methylenvaline, a molecular adduct formed by addition of formaldehyde to the N-terminal valine in hemoglobin. They studied 21 subjects who were employed in a plywood factory and a laminate factory (occupationally exposed to formaldehyde). They also selected 30 individuals (no occupational exposure to formaldehyde) as controls from volunteers who consented to be tested for biomarkers. Formaldehyde exposure level was measured through a questionnaire (self-report) and vapor monitors. They measured cotinine in urine spot specimens as a contributor of formaldehyde exposure and reported significant correlation between level of urinary cotinine and formaldehyde exposure. They reported direct

68

positive relationship between formaldehyde and N-Methylenvaline in blood. The main limitation of this study was its small sample size. Risner and Martin (1994), conducted experimental study and used a method named “fishtail” chimney procedure to collect and quantify formaldehyde in side stream cigarette smoke. In the fishtail method, the side stream smoke was drawn by vacuum through the chimney (a Cambridge filter pad) and into an impinger that contained 2,4dinitrophenylhydrazine (DNPH). As discussed earlier DNPH is a chemical compound that is often used to qualitatively test for VOCs and quickly reacts with formaldehyde (California Environmental Protection Agency, 2001). In their study, Risner and Martin (1994) found a linear relationship between formaldehyde and the number of cigarettes smoked. They also reported the formaldehyde level of 0.004 to 0.013 ppm per cigarette. This study measured formaldehyde level in a small environmental chamber, which makes it difficult to compare the results with other studies that were conducted in residential dwellings. The current study is the first study regarding comparison of formaldehyde exposure and tobacco smoke exposure in pregnant women. Previous studies are few and were conducted in non-pregnant smokers or as an experimental laboratory study.. Given the changes in the metabolism of nicotine during pregnancy that result in increase of nicotine and cotinine clearance and decrease in half-life of cotinine (Dempsey et al., 2002), the results of the reviewed studies cannot be generalized to pregnant women.

69

Oxidative Stress as a Mediator of the Relationships between the Level of Formaldehyde Exposure and Fetal Growth There were no published studies identified regarding the mechanism linking formaldehyde exposure during pregnancy and fetal growth. Therefore, the physiological pathway of the effect of formaldehyde on fetal growth is not clear. However, there is evidence regarding the potential effect of formaldehyde during pregnancy and on pregnancy outcomes and also related to the potential role of oxidative stress is this relationship. Findings from animal studies report that formaldehyde crosses through the placenta (Katakura et al., 1993; Thrasher & Kilburn, 2001), which means that similar to the other chemicals, such as drugs in category B, C, and D (Cunningham et al., 2010), formaldehyde might not be safe for the fetus. Animal and human studies have reported that formaldehyde is a toxic compound for the fetus and may cause adverse pregnancy outcomes (Al-Saraj, 2009; Cogliano et al., 2005; Grazulevicience et al., 1998; Hansen, Contreras, & Harris, 2005; Magras, 1996; Maroziene & Grazuleviciene, 2002; Saillenfait et al., 1989; Thrasher & Kilburn, 2001; Zhu, Haines, Le, McGrath-Miller, & Boulton, 2001). Furthermore, investigators have reported relationships between formaldehyde exposure and oxidative stress in the organs and tissues of non-pregnant animals (Gulec et al., 2006; Gurel et al., 2005; Im et al., 2006; Kum et al., 2007; Sögüt et al., 2004), which is important because this process can happen in the human fetus as well. Finally, evidence shows a relationship between oxidative stress and fetal growth (Kamath et al., 2006; Kim et al., 2005; Tabacova, Baird, & Balabaeva, 1998). Therefore, in this section, the process of oxidative stress and its metabolite, 15-isoprostaneF2t, are reviewed.

70

Oxidative stress occurs when the rate of free radical production exceeds the rate of removal or buffering by the cellular defense mechanisms (Burton & Jauniaux, 2011). An increase in free radicals increases the rate of oxidative stress in the absence of adequate antioxidants and results in tissue damage (Halliwell, 1994). Lipid peroxidation, as part the of oxidative stress progress, has been associated with cellular damage, in vivo, in both animals and plants during aging and in certain diseases (Siciliano et al., 2012). Lipid peroxidation is a process that happens naturally in small amounts in the body due to the effect of several reactive oxygen species (Halliwell & Gutteridge, 2007). Reactive oxygen species (ROS) is a term used to describe a number of reactive molecules and free radicals derivative from molecular oxygen (Halliwell & Gutteridge, 2007) . Isoprostanes can be considered reliable markers of lipid peroxidation (Cracowski, Durand, et al., 2002; Siciliano et al., 2012). Isoprostanes are a unique series of prostaglandin-like compounds that are produced from arachidonic acid via a free-radicalcatalyzed mechanism (Cracowski, Durand, et al., 2002). In 1990, Morrow et al. discovered the first class of isoprostanes, the F2-isoprostane, so named because it contains F-type prostane rings similar to prostaglandin F2α (Roberts & Milne, 2009). There are four series of F2-isoprostanes, including 5-, 12-, 8-, or 15, with names based on the carbon atom to which the side chain is attached (Roberts & Milne, 2009). Correlations between high level of 15-isoprostane F2t and diseases associated with ischemia-reperfusion, atherosclerosis and inflammation, have been reported (Cracowski, Durand, et al., 2002). There are no published studies regarding the effect of 15isoprostane F2t on fetal growth or placental function, however, given that 15-isoprostane F2t mediates vasoconstriction (Halliwell & Gutteridge, 2007) and poor fetal growth is

71

associated with feto-placental vasoconstriction (Cunningham et al., 2010; S. Zhang et al., 2013), it can be speculated that 15-isoprostane F2t can be associated with placental insufficiency or poor fetal growth. The flow of formaldehyde from mother to fetus has been addressed by Katakura et 14

al. (1993), who administered C-formaldehyde (radioactive marker) intravenously to 14

pregnant mice. C-formaldehyde was found in maternal tissues and organs (maternal liver, intestinal mucosa, bone marrow, kidneys, and salivary glands) and fetal tissues and blood immediately after injection. In their study , Katakura et al. (1993), reported considerable accumulation and retention of formaldehyde three hours after injection in maternal organs and tissues and six hours after injection in fetal organs and tissues. 14

Thrasher and Kilburn (2001), investigated the distribution of C-formaldehyde in 14

pregnant mice and observed a rapid uptake of C-formaldehyde in maternal liver, lungs, heart, salivary glands, gall bladder, spleen, kidneys, bone marrow, nasal mucosa, uterus, placenta, and fetal tissues. Further, Thrasher and Kilburn (2001) found higher 14

concentrations of C-formaldehyde in fetal tissues, including the fetal brain, compared to the maternal brain. Katakura et al. (1993) and Thrasher and Kilburn (2001) both reported that elimination of formaldehyde was slower from fetal tissues than maternal tissues and that the fetus is the main site of accumulation of maternal formaldehyde. Therefore, maternal formaldehyde can accumulate in the fetal organs and eliminate at a slower pace. In the human studies discussed earlier, a relationship between LBW and formaldehyde exposure level was reported (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002). Furthermore, developmental toxicity in animal studies has been linked to formaldehyde exposure. Thrasher and Kilburn (2001) found an increased rate of 72

embryo degeneration, chromosomal changes, involution of lymphoid tissues, and hypertrophy in Kupffer’s cells in the fetus after injection of formaldehyde. Saillenfait et al. (1989), exposed Sprague-Dawley rats to 0, 5, 10, 20 or 40 ppm formaldehyde for 6 hour/day from day 6 to 20 of gestation. They reported significant concentration-related reduction of fetal body weight (approximately 20%) in groups that exposed to 20 and 40 ppm. Likewise, Al-Saraj (2009) reported congenital abnormalities, preterm birth, and LBW in rabbits who were exposed to 12 ppm formaldehyde during the entire pregnancy. The major limitations of these animal studies is the physiological differences between different animals and between animals and humans, In addition, different studies used various levels of formaldheyde exposure in diverse exposure duration. These limitations, make the interpretation of the animal studies challenging. Formaldehyde may result in oxidative stress in cells. Several in vivo studies have examined oxidative stress in rats exposed to formaldehyde. Formaldehyde exposure in these in vivo studies was found to be associated with oxidative stress in the rat liver, plasma, lymphocytes, heart, and brain (Gulec et al., 2006; Gurel et al., 2005; Im et al., 2006; Kum et al., 2007; Sögüt et al., 2004). The presence and absence of ROS was found to be a determining factor in the cytotoxic effects of formaldehyde (Saito et al., 2005). Therefore, the presence of oxidative stress at the cellular level can occur in placental cells as well (Crowley, 2013; New et al., 2013; Soylemez et al., 2013). Oxidative stress may contribute to placental insufficiency (Adcock et al., 1994; Crowley, 2013; New et al., 2013; Soylemez et al., 2013), which is defined as insifficient blood flow to the placnetal during pregnancy (Cunningham et al., 2010). Placental

73

insufficiency may result in placental dysfunction and IUGR and is considered one of the underlying risk factors for poor fetal growth (Cunningham et al., 2010). Tabacova et al. (1998) measured blood lipid peroxides as an oxidative stress biomarker in cord blood and found that poor birth outcome, including LBW, was associated with high oxidative stress. Kim et al. (2005) studied the role of maternal oxidative stress in LBW in Korea and found that concentrations of biomarkers of oxidative stress, including urinary 8-hydroxydeoxyguanosine (8-OH-dG) and malondialdehyde (MDA), were inversely associated with birth weight. Kamath et al. (2006) reported higher MDA in mothers of IUGR babies when compared to mothers who did not have IUGR. Kamath et al. (2006) examined the possibility of increased lipid peroxidation and protein oxidation in both maternal (maternal blood) and fetal erythrocytes (cord blood) as markers of oxidative stress during intrauterine growth restriction. They reported that erythrocyte MDA levels were significantly elevated in mothers of IUGR babies when compared to mothers who did not have IUGR (control). In addition, the endogenous protein damage due to oxidative stress was significantly higher in IUGR mothers when compared to controls. Moreover, they reported that in fetuses born with IUGR, lipid peroxidation was significantly higher than normal newborns. The above studies support the possibility of the mediating role of oxidative stress. However, there are several limitations. For instance, with animal experiments there is the issue of physiological differences between humans and animals, specifically in immunological and neurological systems. The other issue is that, pregnancy by itself is associated with oxidative stress on the maternal side (Chelchowska, Ambroszkiewicz,

74

Gajewska, Laskowska-Klita, & Leibschang, 2011). Whereas, the levels of oxidative stress in maternal biological samples or cord blood and/or placenta have been used as an alternate for oxidative stress in the fetus in reviewed articles. There is a barrier to further research because examining the level of oxidative stress in the fetus is not possible due to the safety of the fetus. Therfore, more animal studies might be helpful cosidering the metioned limitations.

Oxidative Stress as a Mediator of the Relationships between the Level of Tobacco Smoke and Fetal Growth There is more evidence regarding the physiological pathway of the effect of tobacco smoke exposure on fetal growth than formaldehyde exposure. The physiological pathway of the relationship between tobacco smoke exposure and fetal growth is related to placental insufficiency (Cunningham et al., 2010). In addition evidence shows that tobacco smoke can result in vascular damages due to mechanisms involving the production of ROS (Nishizawa, Kohno, Nishimura, Kitagawa, & Niwano, 2005). Aycicek, Varma, Ahmet, Abdurrahim, and Erel (2011), studied the influence of active and passive maternal smoking on placenta oxidative stress status in term infants. They measured levels of cord blood total oxidant status (TOS), and oxidative stress index (OSI) in samples of fetal placental tissue, cord blood, and the maternal peripheral blood serum in active smokers (n = 19), passive smokers (n = 19), and nonsmokers (n = 22). TOC is a method that measures total oxidants that present in body fluids or plant extracts. OSI is a combined measurement that is calculated by a ratio between pro-oxidants and antioxidants that could be measured through specific assays(Aycicek et al., 2011). The

75

selected participants were women with uncomplicated pregnancies, who delivered vaginally between 37-40 weeks of gestational age. These investigators reported that birth weight and head circumference in the active smokers were significantly lower than passive or non-smokers. Placenta, cord blood, and the maternal peripheral TOS and OSI levels were significantly higher in the active and passive smokers than in the controls. In addition, they found positive significant correlation between active maternal smoking and placenta oxidative stress index. The major limitation of this study is the small sample size. Rossner et al. (2009), studied the relationship between tobacco smoke exposure on oxidative stress, and assessed correlations between oxidative stress. They measured 15-isoprostane F2t and cotinine, as biomarkers of oxidative stress and tobacco smoke. They found 15-isoprostane F2t in mother’s plasma was significantly higher than cord blood plasma. In addition, no relationship was found between tobacco smoke exposure and oxidative stress biomarkers. The main limitation of this study is that investigator measured the oxidative stress at birth. However, the oxidative stress could be related to the medications used at the time of the delivery, for example analgesics or antibiotics. Chelchowska et al. (2011) studied the effect of tobacco smoking during pregnancy on oxidative damage and compared antioxidant defense in samples of maternal blood and cord blood. They selected 140 healthy, pregnant women and divided into non-smoking and smoking groups according to the concentration of cotinine in serum and urine. They measured oxidative damage through levels of MDA. They determined the plasma antioxidant status by measuring concentrations of total radical trapping parameters (TRAP) and selected antioxidants (β-carotene, vitamin A, vitamin E,

76

uric acid). They found that during pregnancy the concentration of MDA increased. In addition, they found a positive correlation between higher values of MDA in smoking women than in non-smoking ones. They also found significantly lower TRAP in the smoking group than in the controls (p < 0.05). Plasma concentration of uric acid and antioxidant vitamins E, A and β-carotene were lower in smokers as compared with nonsmokers. Furthermore, they found that the level of MDA in plasma of cord blood of newborns of smoking mothers was significantly higher than in non-smokers. One limitation is similar to Rossner et al. (2009) regarding the interaction between the medications that are used at the time of delivery and oxidative stress. The other limitation is the time of data collection for tobacco smoke exposure status and plasma antioxidant levels, which was at the first visit of prenatal care for tobacco smoke exposure and at the time of delivery for antioxidants. As discussed earlier, the physiological changes for clearance of cotinine in different trimesters could lower the reliability of the results of this study. As discussed earlier, tobacco smoke can be a source of formaldehyde exposure (ATSDR, 2010). Cotinine crosses through the placenta (Cunningham et al., 2010; Wu et al., 2008) similar to formaldehyde (Thrasher & Kilburn, 2001). Evidence shows that formaldehyde is associated with oxidative stress (Gulec et al., 2006; Gurel et al., 2005; Im et al., 2006; Kum et al., 2007; Sögüt et al., 2004) . Furthermore, several studies reported that cotinine are associated with the oxidative stress as well (Aycicek et al., 2011; Chelchowska et al., 2011; Rossner et al., 2009). Adding to the above evidence, studies that support the association between oxidative stress and fetal growth (Kamath et al., 2006; Kim et al., 2005; Tabacova et al., 1998), it can be speculated that oxidative

77

stress can play a mediating role in pathological pathway of both formaldehyde and tobacco smoke exposure on fetal growth. One purpose of the current study was to explore these hypothesized relationships.

Summary Poor fetal growth is associated with increased perinatal, infant, and adulthood morbidity and mortality (Barker, Osmond, Forsen, Kajantie, & Eriksson, 2005; Kajantie et al., 2005). Formaldehyde and tobacco smoke exposure have been linked to fetal growth, such as IUGR and LBW (Erichsen et al., 2006; International Agency for Research on Cancer, 2006; Wang et al., 2000). No previous study has examined the level of formaldehyde exposure during pregnancy and its relationship with fetal growth. However, studies that measured the level of formaldehyde using vapor monitors in residential areas and personal exposure in non-pregnant women reported a level of 0.009 ppm to 0.44 ppm (Lazenby et al., 2012; Maddalena et al., 2009). Formic acid is another biomarker of exposure to formaldehyde, however several studies have shown that formic acid did not reflect of the level of formaldehyde exposure in low level formaldehyde exposure (less than 0.5 ppm) (Schmid et al., 1994). Fetal growth may be affected as early as first trimester of pregnancy. Therefore, it is important to examine the fetal growth within the first and second trimester trough ultrasound for preventing the adverse outcomes of poor fetal growth (Bukowski, 2004; Smith et al., 1998; Smith et al., 2002). However, most of studies have examined fetal growth in the third trimester or at birth and few studies have reported second trimester fetal ultrasound biometric measurements and resulting pregnancy outcomes. In addition, the second trimester fetal ultrasound biometric measurement are more accurate than the 78

third trimester due to the fetal weight gain during the third trimester of pregnancy (Cunningham et al., 2010). During the second trimester, BPD is the most valuable biometric measurement (Fescina et al., 2011). Formaldehyde is a by-product of oxidation of tobacco, which has been reported in both mainstream and side stream smoke and in substantial levels in Environmental Tobacco Smoke (ETS) (Godish, 1989; Parrish & Harward, 2000). However, several studies showed no relationship between tobacco smoke and formaldehyde inside residential dwellings (Dannemiller et al., 2013; Dassonville et al., 2009). Tobacco smoke is associated with adverse pregnancy outcomes, such as IUGR and LBW (Dejmek et al., 2002; Horta et al., 1997; Sclowitz et al., 2013). Moreover, only a few previous studies have examined the role of tobacco smoke exposure on fetal ultrasound biomarkers in the second trimester. The major sources of formaldehyde exposure in residential dwellings are building and household materials (IARC, 2006). Season (Dannemiller et al., 2013; Dassonville et al., 2009), and exposure levels are influenced by the age and type of residential dwelling (Lazenby et al., 2012; Maruo et al., 2010), Indoor household practices such as use of cleaning products, candles or air fresheners (Lindstrom et al., 1995; Liu et al., 2004; Uhde & Salthammer, 2007) also have important roles in the level of formaldehyde exposure. Although tobacco smoke is a source of formaldehyde exposure, evidence shows that smoking in the residential dwellings is not a leading source of formaldehyde (Gustafson et al., 2005; Jurvelin et al., 2001; Rode et al., 2007); however, this relationship has not been examined in pregnant women. Self-report may miscalculate the rate of tobacco smoke exposure during pregnancy; however, creatinine

79

standardized urine cotinine level could be an alternative for self-report (Aurrekoetxea et al., 2013; England et al., 2007). The flow of formaldehyde though placenta has been confirmed by several studies (Katakura et al., 1993; Thrasher & Kilburn, 2001). Oxidative stress has been shown to affect fetal growth through its contribution to placental insufficiency (Adcock et al., 1994; Crowley, 2013; New et al., 2013; Soylemez et al., 2013). In addition, the evidence shows that formaldehyde is associated with oxidative stress in animal’s fetal cells and tissues (Gulec et al., 2006; Gurel et al., 2005), oxidative stress with tobacco smoke (Aycicek et al., 2011; Nishizawa et al., 2005). Therefore, considering that formaldehyde is associated with oxidative stress and oxidative stress is associated with fetal growth, oxidative stress can be considered as a potential mediator in the relationship between formaldehyde exposure and fetal growth. Currently, there is a gap in the literature addressing the mediating role of oxidative stress in the relationship between formaldehyde and fetal growth. One of the challenges in the preceding review of literature was the lack of evidence in many areas. This study was the first regarding: (1) formaldehyde exposure level (formic acid and vapor monitor badge) during pregnancy, (2) formaldehyde exposure level and its relationship with fetal growth(ultrasound biomarkers) in the second trimester, (3) the relationship between indoor residential sources of formaldehyde and fetal growth, (4) the role of oxidative stress as a mediator of the relationship between formaldehyde exposure and fetal growth, (5) the role of oxidative stress as a mediator of the relationship between tobacco smoke and fetal growth, (6) the relationship between formaldehyde detected by vapor monitor badge and urine formic

80

acid during pregnancy, and (7) and the relationship between tobacco smoke (cotinine and self-report) and the formaldehyde exposure level (vapor monitor badge and formic acid) during pregnancy.

81

CHAPTER 3 METHODOLOGY The overall purpose of this study was to determine the level of formaldehyde exposure during pregnancy and examine the relationship between formaldehyde exposure and fetal growth (fetal ultrasound biometry measurements) in pregnant women. This study was also designed to determine the following: (1) the indoor residential sources of formaldehyde exposure (including tobacco smoke exposure) and their relationships with formaldehyde level and fetal growth outcomes, and (2) the potential mediating role of oxidative stress in the relationship between formaldehyde exposure and fetal growth in pregnant women. The sample for this study was pregnant women recruited during their prenatal care visits at selected OG & GYN clinics, who had an ultrasound report within 17-32 weeks of gestational age, and who were in their second trimester of pregnancy. A pilot study was conducted to address feasibility and to inform the larger study. In this chapter, the research design, sample and setting, and protection of vulnerable subjects are discussed. Each study instrument and its corresponding protocol are reviewed in the context of this study. Finally, data management and data analysis plans are presented.

Study Design The study was based on a descriptive, cross-sectional, correlational design. This type of study design does not determine causation, but is used to make inferences about possible relationships, or to collect initial data to support further research and 82

experimentation for new research areas (Levin, 2013). This research design was selected because little is known about the level of formaldehyde exposure during pregnancy. According to Polit and Beck (2008), descriptive studies describe the relationships among the variables by directly examining samples from the target population. The study variables (independent and dependent) in cross-sectional studies are captured at a fixed point in time (Polit & Beck, 2008). Therefore, a descriptive cross-sectional study was appropriate for this study as the independent variables were measured at one time point with no intervention. In addition, this study examined the associations among formaldehyde levels and fetal ultrasound biometry measurements, and the potential mediating role of oxidative stress in these relationships, for which a descriptive correlational design is appropriate. In descriptive study designs, a mediator variable is used to identify the mechanism or process of the relationship between the independent and dependent variables (MacKinnon, 2008).

Characteristics of the Sample The nonprobability, convenience sample for this study consisted of pregnant women from the selected obstetrics and gynecology (OB & GYN) clinics located in an urban area in the southeastern United States (US). This type of sampling method is useful in cross-sectional studies, particularly in new areas of research (Polit & Beck, 2008). The sample consisted of pregnant women who (1) were capable of understanding, speaking, and responding in English, (2) were between the ages 19 to 40, (3) were pregnant with one fetus (singleton), (4) were in their second trimester of pregnancy, and (4) had at least one fetal ultrasound biometry report within 17-28 weeks of gestational age. Ultrasound

83

reports were reviewed in the potential participants’ medical records to determine eligibility. Exclusion criteria for this study included pregnant women who were (1) nonEnglish speaking, (2) were ages less than 19 or more than 40, (3) were pregnant with more than one fetus, (4) were in their first or third trimester of pregnancy, and (5) did not have a fetal ultrasound biometry report within 17-28 weeks of gestational age . The extremes of maternal age (less than 16 and above 40 years) (Roth, Hendrickson, Schilling, & Stowell, 1998; Scholl, Hediger, & Schall, 1996; Shah & Ohlsson, 2002) and pregnancies with more than one fetus (Cunningham et al., 2010) have been linked to poor fetal growth. In addition, several known risk factors that have been identified in the literature as influencing fetal growth were also considered as exclusion criteria. These risk factors included: (1) history of substance abuse during current pregnancy (Cunningham et al., 2010; Mastrobattista & Gomez-Lobo, 2008), (2) no prenatal care during the current pregnancy (or late initiation of prenatal care) (Cunningham et al., 2010), (3) history of chronic diseases, such as hypertension (Odegard, Vatten, Nilsen, Salvesen, & Austgulen, 2000), anemia (Cunningham et al., 2010; Yildiz, Ozgu, Unlu, Salman, & Eyi, 2013), asthma (Bakhireva, Schatz, & Chambers, 2007), sickle cell diseases (Chakravarty, Khanna, & Chung, 2008), gestational diabetes (Bowers et al., 2013; Cunningham et al., 2010), diabetes (Lampl & Jeanty, 2004), vascular disease (Cunningham et al., 2010), and chronic renal diseases (Cunningham, Cox, Harstad, Mason, & Pritchard, 1990; Vidaeff, Yeomans, & Ramin, 2008), (4) fetal abnormality, fetal congenital malformations, and genetic disorders such as trisomy 13, 18, and 21, triploidy, and Turner’s syndrome

84

(Khoury, Erickson, Cordero, & McCarthy, 1988; Rochelson et al., 1990), (Khoury et al., 1988; Rochelson et al., 1990), (5) pregnancy resulting from infertility interventions (Cunningham et al., 2010), (6) placental or amniotic fluid abnormalities (Cunningham et al., 2010), and (7) history of infection during the current pregnancy, such as rubella, cytomegalovirus, and varicella, that have been found to be associated with IUGR (Droste, 1992; Hejtmancik et al., 1992; Stagno et al., 1977; Waterson, 1979; Zhu, Obel, Hammer Bech, Olsen, & Basso, 2007). While some of the above mentioned risk factors were identified as exclusion criteria, other risk factors related to IUGR were considered as independent variables or potential covariates. Tobacco smoke exposure was considered as an independent variable. Selected maternal demographic and pregnancy characteristics were considered as covariates. These included maternal age, race/ethnicity, marital status, and socioeconomic status (employment, education, and income), parity, intervals between two pregnancies, and fetal gender.

Characteristics of the Setting The settings for this study were one university associated and three private OB & GYN clinics located in a US southeastern community. The selected clinics had high number of clients. The clients that were seen in the selected clinics represented various racial, ethnic, cultural, and socioeconomic characteristics. A pilot study was conducted in the university-associated and in one private clinic in February and March of 2013. A total of 94% of participants were recruited from the university-associated clinic and 6% were recruited from the private clinic. The distribution of race for the entire practice was not disclosed by the clinic; however, in the

85

pilot study, 74% of participants were identified as Caucasian. Based on the results of pilot study, two other private clinics that served higher rates of African American clients were added to the study.

Sample Size and Power No effect sizes were available in the published literature, since, to the knowledge of principal investigator, no previously published study has addressed the specific set of variables of this study. Therefore, recommendations by Cohen (1984) were used to calculate sample size and power. According to Cohen (1988), small, medium, and large effect sizes can be estimated using R2 = 0.02, 0.13, and 0.30, respectively. A medium effect size of R2 = 0.13 was selected as a balance between achieving the desired power level of .80 at α = 0.05, as well as taking into account the feasibility of subject recruitment and data collection for the necessary sample size (n). For this study, the covariate variables included in the analysis were maternal demographic and pregnancy characteristics. The independent variables were indoor residential sources of formaldehyde, the levels of formaldehyde exposure (urine or vapor monitor), tobacco smoke exposure (urine), and the mediating variable was oxidative stress. The formula used for calculating effect size was drawn from Cohen’s effect sizes (Cohen, 1988), where N represents the estimated sample size, γ represents the estimated effect size, L represents the tabled value for the desired alpha and power, and k represents the number of independent or predictor variables.

86

Estimated effect size: γ = R2/ (1 – R2) Estimated number of potential participants: N = [L / γ] + k + 1 Considering a medium effect size of (.13), γ= 0.13/1-0.13=0.149. Considering k=8, for power = .80, a value of L = 15.02 is obtained (Polit & Beck, 2008). Therefore for this study, a sample size of n = 109 based on a medium effect size of ES = .13, power of .80, and α = .05 for 8 predictor variables was calculated. To achieve the total number of potential participants who were required for the final study analysis, practical issues such as missing data were considered (Suresh & Chandrashekara, 2012). The following formula was used to calculate the optimal number of potential participants for this study: N1= N/ 1-q, where N1 was the total number that would have to be recruited to make sure that the final sample size (N=109) was achieved, and q is the proportion of attrition (Suresh & Chandrashekara, 2012). The proportion of attrition is generally considered to be 10%, however, in this study q was considered as 20%. The reason for selecting 20%, instead of 10%, was based on the results of pilot study that showed 36% of participants failed to return their vapor monitor badge back to the clinic. Considering q=34%, the calculated sample size would increase the total number of potential participants to 170. Considering the feasibility, and due to budget limitations, q was considered at 20% and a total of 140 participants (N1=140) were calculated as final sample size. An increase in the amount of the incentive, which is discussed below in protection of vulnerable subjects, was taken into account in determining the final sample size. It was anticipated that this increase would encourage participants to return their badges, thereby decreasing attrition.

87

Protection of Vulnerable Subjects Pregnant women are considered vulnerable study subjects because of the fetus, who does not have the ability to protect himself or herself (Schwenzer, 2008). When conducting research with pregnant women, investigators must provide additional safeguards to protect the rights and welfare of these subjects (Office for Human Research Protections (OHRP), 1993). Based on the US federal guidelines research associated risks to the fetus must not be more than minimal risk (OHRP, 1993). Minimal risk is defined as “the probability and magnitude of physical or psychological harm that is normally encountered in the daily lives, or in the routine medical, dental, or psychological examination of healthy persons” (OHRP, 2009, p. 11). In this study, there were no known risks to the fetus but minimal risks for the mother were considered, which included potential breach of confidentiality and discomfort in sharing personal information. Approval to conduct this study was obtained from the Institutional Review Board (IRB) at the University of Alabama at Birmingham (UAB), prior to the initiation of this study (see Appendix A). The UAB IRB approved the study protocols, informed consent documents, and the questionnaire. Based on the pilot study, minor amendments (e.g, incentive increased [pilot study, only one $5.00 gift card was given]) were made in the study procedures and protocol in the larger dissertation study. In this study, the incentive was two gift cards ($5.00 each). One gift card was given at the time of interview and a second gift card was given when they returned the badge. The eligibility of the participants was determined by the principal investigator through accessing the patient’s health record in the university-related clinic, and by the physicians in the private clinics. A random number code was assigned to each participant

88

to reduce the risk of breach of confidentiality. The same random number code was used to label the questionnaire, urine specimen, and vapor monitor badge instead of using personal identifiers. Personal identifying information linking the study participant to the unique participant identification code, including the patient’s name and phone number, were recorded in a log book. The participant’s phone number was used for vapor monitor badge follow-up. All documents, including questionnaires and consent forms, were kept in a locked file cabinet at the principal investigator’s office. The log book was kept in a locked file cabinet separate from the consent form and questionnaires. All data were stored on an IronKeyTM encrypted flash drive and on the principal investigator’s encrypted office computer (hard drive). Only the principal investigator had access to the encrypted flash drive and office computer. The urine samples were stored in a -80oC freezer in a laboratory and the vapor monitor badges were stored in room temperature at the principal investigator’s office. Efforts were made to ensure privacy during data collection by using an exam room with a closed door at the clinics. Women who were eligible were informed about the study objectives and the research process in a private room in the clinic. A copy of consent form was given to each woman for review. If the woman agreed to participate she signed two copies of the informed consent and kept one. Women were given an opportunity to ask questions and told that they could withdraw from the study at any time without repercussions. The participants were informed that no identifying information would appear on any data collection sheet or within any database. They also were informed that a random

89

code number was used on all study forms and specimens (questionnaire, vapor monitor badge package, and urine samples). Also, participants were informed if they experienced discomfort about answering any questions at any time, they did not have to answer the questions and could stop their participation in the study without penalty. In addition, participants were informed that wearing the vapor monitor badge during the night might be uncomfortable. Therefore, the participants were asked to take off their badge and put it uncovered close to their head on a night stand, chair, or table while asleep. Levels of concern in exposure to formaldehyde and formic acid were identified based on standard levels reported by the laboratories, as there are no standard levels for pregnant women. When the amount of formaldehyde indicated by the vapor monitor badge was greater than 0.1 ppm (Advanced Chemical Sensors, 2012) or the level of formic acid was higher than 36 µg/ml (NMS LABS, 2013), the results were discussed with the participant’s physician, who then discussed them with the participant. The principal investigator also provided the physicians with information on how to reduce the level of exposure to formaldehyde if there was high exposure.

Instrumentation The research design for this study incorporated a biobehavioral approach, using subjective (questionnaire) and objective (assays, ultrasound report) measures. The dependent variable of fetal growth was measured using fetal growth ultrasound biometry, including measurements of head circumference (HC), abdominal circumference (AC), femur length (FL), biparietal diameter (BPD), estimated fetal weight (EFW), and abdominal circumference to femur length ratio (AC/FL). The objective measures of

90

formaldehyde vapor monitor badge and urinary formic acid were used to measure the independent variable of the levels of formaldehyde exposure. A questionnaire that included items about indoor residential sources of formaldehyde, including residential characteristics and household practices was administered. This questionnaire also included questions about maternal demographic and pregnancy characteristics. The level of tobacco smoke exposure was measured through the collection of urine for cotinine. Oxidative stress was measured through collection of urine for 15-isoprostane F2t. Study measures were selected based on the conceptual framework, and the accuracy, reliability, and feasibility of available biomarker assays.

Questionnaire The researcher-developed questionnaire included the following sections: maternal demographic characteristics, pregnancy characteristics, residential dwelling characteristics, and household practices. The pilot study testing of the questionnaire revealed no problems with administration and pilot study participants were able to answer all of the items.

Demographic Characteristics This section of the questionnaire was developed by the principal investigator and was designed to collect demographic data. An extensive review of the literature was used to identify demographic factors that influence fetal growth. This questionnaire consisted of seven questions including the participants’ race/ethnicity, education, marital status, employment status, occupation, and the level of family income. Demographic variables

91

were measured through self-report. Maternal demographic variables and the level of measurement for each variable are presented in Table 1.

Table 1 Variable Classification and Measurement Levels for Maternal Demographic Characteristics Level of Measurement Ratio

Variables Age Ethnicity Black or African-American White Others (Hispanic, Asian, Native Hawaiian, AmericanIndian, Native-American)

Nominal

Highest level of education High school or less/GED Some college and higher

Dichotomous

Marital status Married Single (Divorced, never-married)

Dichotomous

Current Employment status Fulltime Part-time Not employed

Nominal

Level of family income $40,000

Ordinal

Indicators for US socioeconomic status in research are income and education (Daly, Duncan, McDonough, & Williams, 2002). In this study a description of income and education were used as indicators of socioeconomic status. Socioeconomic status was 92

operationally defined as self-report of the highest level of education and yearly family income.

Pregnancy Characteristics The pregnancy characteristics section of the questionnaire was designed to collect pregnancy related data and fetal ultrasound biometry measurements from existing clinical sources. Pregnancy characteristics data were collected by reviewing the participants’ medical record. This was done by the principal investigator at the university clinic. At the private offices, the data collection procedure was the same, however, the physicians accessed the participants’ electronic data and reported the information to the principal investigator. Pregnancy characteristics included gravida (number of pregnancies including current pregnancy), fetal gender (male/female), interval between pregnancies (the months calculated between the end of the previous pregnancy to the beginning of the current pregnancy), and maternal smoking status (self-report). Pregnancy characteristics variables and the level of measurement for each variable are presented in Table 2. In addition, gestational age and fetal ultrasound biometry measurements were included in this questionnaire. Gestational age can be calculated based on both ultrasound and last menstrual period (LMP) (Cunningham et al., 2010). Gestational age assessed by ultrasound biometry in early pregnancy has been considered to be more precise in predicting the day of delivery than counting from the first day of the last menstrual period (LMP) (Gardosi & Geirsson, 1998; Geirsson, 1991; Nakling & Backe, 2002).

93

Table 2 Variable Classification and Measurement Levels for Pregnancy Characteristics Level of Measurement Dichotomous

Variables Fetal gender Male Female Gravida

Ratio

Maternal Smoking Status (Self-report) Smoker Non-Smoker Interval between Two Pregnancies Primigravida >18 months 5

51 66 20 3

36.0 47.5 14.4 2.1

Maternal Smoking Status (Self-report) (N = 138) Smoker Non-Smoker

19 119

15.4 84.6

Interval between Two Pregnancies (N = 129) Primigravida >18 months 0.05), therefore, it was not considered in subsequent regression models (Table 16). A significant relationship was found between formaldehyde vapor monitor badge measured as a dichotomous variable (< .03 ppm and > .03 ppm) and BPD percentile (rpb = -.303, p < .006). Independent-samples t-test revealed that the mean BPD percentile was higher in participants with formaldehyde levels (determined by vapor monitor badge) < .03 ppm (M = 57.02 versus M = 45.00, p < .005). In addition, a significant relationship was found between FL percentile and urinary cotinine level (r = -0.175, p < .045). Linear regression models were examined only for the above mentioned significant bivariate relationships. A linear regression model considering level of formaldehyde exposure (determined by vapor monitor badge) as a dichotomous variable of < .03 and > .03 ppm was tested. In this model, formaldehyde and BPD were considered as the independent and dependent variables respectively. For this model maternal age (continuous variable), fetal gender, and race (reference-cell coded with White as the reference category) were included as control variables. Based on the results presented in Tables 11-13, race and fetal gender had significant relationships with several fetal ultrasound biometry measurements. Although, there were no relationships between maternal age and fetal ultrasound biometry measurements in this study, it was included in the linear models as an important risk factor in fetal growth (Cunningham et al., 2010; Fescina et al., 2011). The overall model to predict BPD was significant (F = 3.596, p < .006) with an R2 of .20. Fetal gender and maternal age were not significant and were removed from the

143

model. Controlling for race, ATSDR dichotomized formaldehyde levels was a significant predictor of BPD percentile (β = -.271, p < .013). One of the 2 dummy coded race variables was also significant (Other compared to White; p < .001). Examination of the normal p-p plot did not indicate a violation of the normality assumption and the test of homoscedasticity also indicated that this assumption was not violated (χ2 = 14.61, p = .48). Post hoc power analysis showed a power of 0.65 for formaldehyde and 0.87 for race.

Indoor residential sources of formaldehyde exposure. Table 17 summarizes the relationships between indoor residential dwelling characteristics with fetal ultrasound biometry measurements. There were no significant relationships between indoor residential dwelling characteristics and fetal ultrasound biometric measurements. Using one way ANOVA, no significant difference was found for the means fetal ultrasound biometry measurements and among the categories of “type of dwelling” or “age of dwelling.”

144

Table 16 Correlation between Formaldehyde (Vapor Monitor Badge and Formic Acid), and Cotinine with Fetal Ultrasound Biometry Measurements’ Percentiles AC

FL

HC

Variables

.049 .661

.135 .228

-.059 .598

Formaldehyde Exposureb

.081 .468

.073 .513

-.091 .414

-.081 .358

.002 .984 -.175 .045

Urine Cotinine (N = 131)a

EFW

AC/FL

Correlation coefficient p value

Formaldehyde£ (N = 82)a

Formic Acid (N = 131)a

BPD

.109 .217

£

a

Pearson (r) Point-biserial correlation coefficient (rpb) § p < .01 level; £ p < .05 level £ vapor momitor Classification Based on ATSDR Recommendation b

145

-.168 .132

.089 .428

-.054 .629

-.303 .006

.111 .323

.018 .876

-.003 .972

.081 .360

-.064 .471

-.059 .501

-.049 .575

.013 .882

-.042 .635

.160 .068

§

Table 17 Correlations between Selected Indoor Residential Dwelling Characteristics and Fetal Ultrasonic Biometry Measurement (Percentiles) Variablesb

AC

FL

HC

BPD

EFW

AC/FL

Correlation Coefficient p value New Furniture (N = 128)a

.079 .374

-.160 .071

.106 .233

-.033 .707

-.039 .663

.151 .089

New Carpet (N = 125)a

-.025 .781

.052 .568

-.093 .305

-.077 .393

.072 .426

-.055 .539

Remodeled House (N = 126)a

-.117 .190

.081 .368

-.058 .520

.095 .290

-.103 .253

-.124 .165

Use of Perfume (N = 125)a

-.012 .891

-.104 .249

.135 .134

.133 .138

-.075 .404

.041 .650

Use of Candle (N = 129)a

-.004 .963

-.146 .098

.002 .985

-.124 .162

-.109 .219

.058 .517

Use of Nail Polish (N = 125)a

-.146 .104

.017 .850

.010 .916

-.041 .649

-.134 .136

-.127 .157

.028 .753

-.070 .431

.014 .877

.000 .998

.076 .396

.039 .662

Use of Air Freshener (N = 128)a a

Point-biserial correlation coefficient (rpb) Tobacco smoke exposure. A linear regression model was examined for FL

percentile (as the dependent variable) and creatinine standardized urinary cotinine (as the independent variable). An extreme outlier of urine cotinine, 3893.0 (µg/g), was removed from the data analysis. The overall model was significant (F = 3.07, p = 200 µg/g

116 22

84.1 15.9

Non Creatinine standardized Creatinine standardized

Summary The mean formaldehyde exposure level in this study, as measured by vapor monitor badges, was 0.04 ppm (SD = 0.06). The mean of the creatinine-standardized formic acid level in urine was 18.00 mg/g (SD = 9.9). About 36% of the participants were found to be exposed to ≥ .03 ppm of formaldehyde based on ATSDR classification (< .03 or > .03 ppm). The mean level creatinine-standardized cotinine in urine was 153.0 µg/g. The mean level of creatinine-standardized 15-isoprostaneF2t in this study was 5.15 µg/g (SD = 3.0). No significant correlation was found between urinary formic acid and vapor monitor badge readings. In addition, no relationships were found between the level of formaldehyde using vapor monitor badge and urinary formic acid with the level of urinary cotinine as the biomarker of tobacco smoke. There were no correlations 150

between the urinary 15-isoprostane F2t and urinary levels of formic acid and cotinine, and the vapor monitor badge readings. Significant relationships were found between the level of formaldehyde using vapor monitor badge (continuous measure) and season of data collection, indoor temperature, use of nail polish, and new carpet. Significant correlations were also found between the ATSDR classification of formaldehyde (dichotomous) and season of data collection, indoor temperature, and remodeling. Formic acid in urine did not show any relationship with fetal ultrasound biometry measurements. However, a significant relationship was found between level of formaldehyde exposure (dichotomous , < 0.03 ppm and > 0.03 ppm) as measured by the vapor monitor badge and BPD percentile. In addition, a significant relationship was found between FL percentile and urinary cotinine level. The linear regression model including covariates showed that ATSDR dichotomized level of formaldehyde exposure was a significant predictor of BPD percentile, after controlling for maternal race. However, the linear regression model for cotinine level and FL with covariates showed that cotinine was not a significant predictor of FL percentile after controlling for fetal gender (β = -.123, p < .168). The relationships between isoprostane and fetal growth as measured by ultrasound biometry measurements were non-significant. In addition, there were no significant associations between isoprostane and level of formaldehyde exposure and tobacco smoke exposure in this study (vapor monitor badge readings for formaldehyde, urinary formic acid levels, or urinary cotinine levels). As a result, no analyses were performed related to examining isoprostane as a potential mediator of the relationships between the key study exposure variables and the fetal ultrasound biometry measurement outcomes.

151

CHAPTER 5 DISCUSSION The purpose of this study was to determine the level of formaldehyde exposure during pregnancy and to explore the relationships between level of formaldehyde and tobacco smoke exposure and fetal growth in the second trimester of pregnancy. In addition, the principal investigator sought to explore (1) the indoor residential sources of formaldehyde exposure (including tobacco smoke) and their relationship with formaldehyde exposure level, and (2) the potential mediating role of oxidative stress in the relationships between formaldehyde and tobacco smoke with fetal growth. This chapter begins with a discussion of the characteristics of the study sample, followed by findings related to the research questions, then additional analyses. The final section of this chapter provides study implications and recommendations for future research. The discussion of the study limitations and implications for nursing education and future research are included in this chapter as well.

Findings Related to the Study Sample Participant recruitment took place in one university-associated and three private Obstetrics and Gynecology (OB& GYN) clinics. The sample for this study was enrolled during winter and spring of 2013 (February to June). The final study sample size was 140, including 31 participants from the pilot study. The sample size met the

152

requirements as determined by an a priori power analysis, which yielded a desired N = 109, for an effect size of .13, power = .80, and α = .05. To account for expected attrition and missing data, 20% was added to the calculated sample size, making the optimal number of subjects N = 140. For all dependent and independent variables, with the exception of formaldehyde exposure as measured by the vapor monitor badge, the sample size exceeded the required number of participants (N = 109). Only eighty-eight participants (63%) returned the vapor monitor badge. However, in regression models the calculated post hoc power was 65 for formaldehyde level as measured by vapor monitor badge. Participants in the study sample ranged in age from 19-40 years (M = 25.9, SD = 5.09). The majority of the study participants were between age 21-25 (38.1%) and 2630 (27.2%) years. These results are consistent with the Alabama demographic for pregnancy rates; that is Alabama pregnancy rates are highest in the 20-24 and 25-29 age groups (Woolbright, Zheng, & Stones, 2013). In this study, 46% of the sample was white, 37.1% was African American, and 16.4% was classified as other (Asian, Hispanic, American-Indian, and others). Based on the 2011Alabama Health Profile, 67% of births in Alabama occurred in whites and 33% occurred in blacks or others (Woolbright, 2013). With respect to the county demographic characteristics, the distribution of females in the age group of 15-44, was 63% white and 37% black and others in the county (Madison County, Alabama) (Woolbright, 2013). The proportion of whites in the sample was lower than for Alabama as a whole; however, the ethnic group proportions were consistent with the patient population seen in the clinics used for recruitment sites. The study sample was more reflective of ethnic groups at the

153

county level. The study sample (55%) had a higher percent of unmarried women as compared to the state of Alabama (42.2%) and Madison county (37.2%) rates (Woolbright et al., 2013). The insurance provider’s breakdown showed that in the clinic that provided the majority of study participants in this study, 45% of the participants were Medicaid, and only 30% had insurance. The distribution of self-reported yearly family income revealed approximately 54% of participants had an income of less than $20,000. The per capita income for Alabama was $23,483 in 2011 (United States Census Bureau, 2013). These differences could be due in part to the selected clinics that served low socioeconomic populations. In this study, 72.8% of the selected participants held a high school diploma or GED equivalent. The Alabama Department of Health reported education defined “educated” as “12 years or more education,” and “undereducated” as “less than 12 years of education.” When the educational characteristics of this study were re-categorized based on the Alabama Department of Health (educated and undereducated), 34.3% of this study’s participants were classified as uneducated. This number is higher than the statistic of 15.7% reported by the Alabama Department of Health division (Woolbright et al., 2013). This finding could possibly be due to the inclusion criteria that limited the study samples to the ages of 19-40 years old as well as the clinics selected for recruitment, which mostly served low socioeconomic groups. Education and socioeconomic levels are known to be associated with fetal growth (Madden, 2013; Neel & Alvarez, 1991), therefore, overrepresentation of participants from lower socioeconomic and educational groups is a limitation to this study that needs to be

154

considered when interpreting and generalizing the findings and when designing future studies.

Findings Related to Research Questions Research Question 1 1a. What Is the Level of Formaldehyde Exposure? The mean formaldehyde exposure level (vapor monitor badge), with 64% response rate, was 0.04 ppm (SD = .06) with a range of 0.003-0.54 ppm. Based on ATSDR classification (2010), 63.6% of the participants were exposed to > 0.03 ppm. As another biomarker of formaldehyde, the mean of the creatinine-standardized formic acid level in urine was 18.00 mg/g (SD = 9.9) with a range of 0-47 mg/g, where 4.4% were above normal (36 mg/g).

Formic acid. No published study has reported the standard level of urinary formic acid in pregnant women. The evidence with regard to urinary formic acid is limited to the personnel exposed in occupational settings where there is a greater risk for exposure to higher levels of formaldehyde (> 0.5 ppm). There are few studies that have included non-pregnant samples from non-occupational settings in control groups (Triebig et al., 1989; Yasugi et al., 1992). Yasugi et al. (1992) reported the level of urine creatinine-standardize formic acid in women and men at 28.46 (SD = 11.23) and 17.41 (SD = 12.08), respectively. In addition, Triebig et al. (1989) reported a level of 2 to 30 mg/l of formic acid in the urine of 20 European men and women (combined). Heinrich and Angerer (1982), reported a range of 3-50 mg/l of formic acid in the urine of 26

155

European males. The reports by Triebig et al. (1989) and Heinrich and Angerer (1982) did not standardize the level of formic acid based on creatinine. However, Ogata and Iwamoto (1990) found that the creatinine-standardized formic acid level in the urine of 31 Japanese males was 7.82 mg/g. The results of the present study’s observations on 140 pregnant women are in agreement with the range of findings reported by others. However, none of the published studies examined pregnant women, and most did not differentiate the level of formic acid for men and women and/or did not standardize the level of formic acid based on creatinine. Due to the possible physiological changes during pregnancy in the respiratory system (Cunningham et al., 2010), which may influence the absorption and elimination of formaldehyde, the comparison between the finding of this study and others may not be valid. These possible changes in urinary excretion of formic acid during pregnancy are an area for further study.

Formaldehyde vapor monitor badge. This was the first known study to examine the formaldehyde exposure level in pregnant women using vapor monitor badges. Given that indoor residential dwelling exposure to formaldehyde can be used as an alternative for personal exposure (Dannemiller et al., 2013; Dassonville et al., 2009), the literature related to formaldehyde level in residential dwellings and non-pregnant personal exposure was selected for discussion in this chapter. The mean formaldehyde exposure level (0.04 ppm) in the current sample was higher than the California Environmental Protection Agency’s recommended limit (0.027 ppm) for residential dwellings (2004). Additionally, this study sample had higher exposure levels than European standards (0.024 ppm) for indoor settings (Bruinen De Bruin et al., 2008). In Finland, 48 hour

156

mean personal exposure levels for formaldehyde were 0.017 and 0.006 ppm in summer and fall, respectively (Jurvelin et al., 2001). In Sweden, the median exposure level was 0.017-0.019 ppm (22-23 µg/m3) for 24 hours (Gustafson et al., 2005). In a European urban environment, the level of formaldehyde exposure has been estimated to be as low as 0.004 ppm (Bruinen De Bruin et al., 2008). Therefore, the level of formaldehyde exposure found in this sample of pregnant women was higher than the exposure level recommended by U.S. environmental agencies and several European countries. There are numerous potential explanations for this finding of elevated formaldehyde exposures among pregnant women. One factor may be the high levels of formaldehyde in building materials used in the US residential structures, which would contribute to exposure to indoor residential sources of formaldehyde. In addition, all of the study participants were female, and women may spend more time in an indoor setting, which has a 2-10 times higher level of formaldehyde when compared to outdoor settings (ATSDR, 2010; California Air Resources Board, 2005). In addition, women may be more exposed to some indoor residential formaldehyde sources, such as nail polish and cleaning materials. Pregnant women’s household activities may also influence the level of formaldehyde exposure during pregnancy. Many pregnant women ready their homes for babies by making small or large modifications in their residential dwellings, such as remodeling, buying new baby furniture and bedding sets, clothes, or adding new carpets or furniture that may expose them to higher levels of formaldehyde. Unlike formic acid, however, the effect of possible physiological changes in absorption and elimination of formaldehyde due to pregnancy does not affect the level of formaldehyde measured by

157

vapor monitors, mainly because vapor monitors measure the level of formaldehyde in the air. Given that higher humidity and temperature increase the level of indoor formaldehyde (ATSDR, 2010; California Air Resources Board, 2005), temperature and humidity may explain the higher exposure levels of formaldehyde. This study was conducted in the Southeastern part of the US, where the temperature and humidity are higher than in European countries or California where other studies were conducted (Bruinen De Bruin et al., 2008; Gustafson et al., 2005; Jurvelin et al., 2001).

1b. What Are the Indoor Residential Sources of Formaldehyde Exposure? In this study, and in agreement with other studies, recent house remodeling (Dannemiller et al., 2013; Dassonville et al., 2009; Kim et al., 2013) and the installment of new carpet (Cracowski, Carpentier, et al., 2002; Dassonville et al., 2009; Hodgson et al., 1993; Rogers et al., 2007) were commonly reported sources of indoor residential formaldehyde exposure. Additionally, 46% of participants reported the use of nail polish in the current study, which was found to be a potential source of formaldehyde. This finding is consistent with other studies (Alaves et al., 2013; Sainio et al., 1997).

1c. Are There Relationships among Levels of Formaldehyde Exposure, Tobacco Smoke Exposure, and Indoor Residential Sources of Formaldehyde Exposure? Formaldehyde exposure (formic acid vs vapor monitor badge). Two different biomarkers were selected to measure the formaldehyde exposure level, the vapor monitor badge and creatinine standardized urinary formic acid. In the current study, no significant correlation was found between measurements from the two methods. These findings are

158

consistent with other studies (Schmid et al., 1994; Triebig et al., 1989; Yasugi et al., 1992), and provide evidence that the urine formic acid levels are not a specific and sensitive biological biomarker for monitoring low-dose formaldehyde exposure. In addition, urinary formic acid can be affected by the physiological changes that occur during pregnancy, as discussed earlier. In contrast, the sensitivity of vapor monitor badges was high, with the lowest level reported at 0.008 ppm (Advanced Chemical Sensors, 2012), and these levels were not affected by pregnancy-associated physiologic changes in metabolism of formaldehyde. For these reasons, vapor monitor badges may be a better formaldehyde exposure measurement method for non-occupational exposure.

Formaldehyde and tobacco smoke exposure. Tobacco smoke may contribute to formaldehyde exposure (ATSDR, 2010; WHO, 2010). In this study, no correlation was found between urinary formic acid and formaldehyde measured by vapor monitor badge or tobacco smoke exposure (self-report and urinary cotinine). There are no published studies examining these relationships in pregnant women. Findings from several other studies, conducted were conducted in non-pregnant adults, are consistent with the finding of the current study that showed no relationship between formaldehyde exposure (vapor monitor badge) and urinary cotinine level (Dassonville et al., 2009; Gustafson et al., 2005; Heroux et al., 2010; Jenkins et al., 2000; Nazaroff & Singer, 2004; Zhang et al., 2003). Conversely, Godish (1989) reported that formaldehyde levels were associated with Environmental Tobacco Smoke (ETS) and indicated that the level of formaldehyde in rooms where a smoker was present was higher. Moreover, Bono et al. (2006) found a positive correlation between formaldehyde exposure (measured by passive air samplers)

159

and urine cotinine level, which also is not in agreement with the findings of the current study. There are several possible explanations for the inconsistency of findings across published studies. The evidence shows that the level of formaldehyde produced by tobacco smoke is related to the type of tobacco and brand of cigarettes; for example, formaldehyde is higher in non-filter cigarettes (Center for Disease Control, 2006; Godish, 1989; Hoffmann et al., 2001; Kensler & Battista, 1963; USDHEW, 1979). In this study, similar to other studies, the participants were not questioned about the type and brand of cigarettes. Therefore, the inconsistency in findings of different studies could be related to limitations of data collection. Dempsey et al. (2002), reported that the clearance of nicotine is substantially increased during pregnancy. Furthermore, Dempsey et al. (2002) found that the metabolic clearance of cotinine is considerably elevated during pregnancy, which results in a marked decrease in the half-life of cotinine. As a result, the inconsistency between the findings of the current study and the study by Bono et al. (2006) could be related to the difference between the selected populations (pregnant versus non-pregnant). Another reason for conflicting results between different studies is the time of urine collection. Benowitz et al. (2009), reported that cotinine level is associated with the time of urine collection and that the first void has the highest level of cotinine. In addition, Benowitz et al. (2009) reported that cotinine corrected for creatinine should be considered as the best predictor of plasma cotinine level in urine if frequent sampling is not possible. In the current study, collection of frequent samples and obtaining the first void were not feasible. However, cotinine was standardized based on the level of

160

creatinine in the same urine sample to reduce the effect of sampling time and urine specific gravity on the level of detected cotinine (Benowitz et al., 2009). In the future studies, first void urine samples may need to be collected for accurate cotinine measurement. Another limitation is the capability of formaldehyde monitors in detecting formaldehyde (ATSDR, 2011). Different studies have used various types of monitors. After finding non-significant relationships between tobacco smoke and formaldehyde exposure level in the current study, an additional laboratory experiment was conducted by the principal investigator to determine the ability of selected vapor monitor badges (Advanced Chemical Sensors, 2012) in absorbing formaldehyde that is produced by tobacco smoke. Although the experiment showed a linear trend between the level of formaldehyde and number of cigarettes smoked under the chamber, it also revealed that long term exposure to higher doses of cigarette smoke is needed to be able to detect formaldehyde in vapor badges. Consequently, another explanation for not finding significant relationships between the level of cotinine and formaldehyde exposure by vapor monitor badges in the current study is that the participants smoked one cigarette at a time in a larger environment (i.e., larger than the laboratory chamber) and in a short period of time. In addition, although the principal investigator asked the participants to wear the badge for 24 hours, there was no assurance that they wore the badge as instructed. These limitations and the physiological changes in the clearance of cotinine during pregnancy need to be reflected while comparing the current findings with other available literature.

161

Formaldehyde exposure and indoor residential sources of formaldehyde. In the current study, no relationship was found between urinary formic acid and indoor residential sources of formaldehyde exposure. As noted previously, this could be due to the lack of sensitivity of urinary formic acid for use in detecting low level formaldehyde exposure (Schmid et al., 1994; Triebig et al., 1989; Yasugi et al., 1992). This is an area for further research.

Residential dwelling characteristics. In this study, significant relationships were found between some, but not all, residential dwelling characteristics and the level of formaldehyde as measured by vapor monitor badge. There was a relationship between residential remodeling and the level of formaldehyde, which is in agreement with other studies (Dannemiller et al., 2013; Dassonville et al., 2009; Kim et al., 2013). Several studies also support the findings of this study that carpet is a source of formaldehyde (Cracowski, Carpentier, et al., 2002; Dassonville et al., 2009; Hodgson et al., 1993; Rogers et al., 2007). No relationship was found between the age of the residential dwellings and the level of formaldehyde in the present study, which is consistent with the findings of Hun, Corsi, Morandi, and Siegel (2010). However, Maruo et al. (2010) and Dassonville et al. (2009) reported a linear correlation between the age of apartments and the level of formaldehyde in new apartments. One explanation for the conflict between the findings of this study and others is that Maruo et al. (2010) studied only new apartments and Dassonville et al. (2009) studied mostly new apartments. In the current study, the majority of residential dwellings (61.1%) were more than 10 years old and only 10%

162

were new, although 12% of the participants could not recall the age of the dwelling. In addition, in this study, the majority of dwellings were houses (54.5%) and only 35.8% were apartments. The low number of new residences among participants, as well as missing data, may have limited the ability to detect a relationship between age of dwelling and formaldehyde exposure. Another possible explanation for the conflicting results is the season of data collection. Maruo et al. (2010) studied the exposure level in fall season, although this study was conducted in winter and spring season. The relationship between season and the emission of formaldehyde in the air is discussed in more detail later in the chapter. The other difference between this study and the study conducted by Maruo et al. (2010) is that they measured the indoor formaldehyde level by placement of monitors in a room for 24 hours, while this study measured personal exposure level where participants were asked to wear the badge consistently for 24 hours and to wear it wherever they went. Therefore, in this study exposure to formaldehyde could be due to occupational exposure. The differences between data collection methods make the comparison between the results of various studies difficult. In the current study, no relationships were found between the formaldehyde level and the type of residential dwellings which is in agreement with Hun et al. (2010). Previous investigators, however, have reported that apartments and mobile homes have higher levels of formaldehyde (Brown, 2002; Dassonville et al., 2009; Hun et al., 2010). Furthermore, the higher rate of exposure in apartments in some of the studies (Dassonville et al., 2009), can be due to the smaller square footage of the residential dwellings. In the current study, no information regarding square footage of dwelling has been collected, which is a limitation.

163

Another important factor, besides age and type of residential dwelling which influences the indoor level of formaldehyde, is the indoor temperature of residential dwellings (Dassonville et al., 2009; Hun et al., 2010). In the current observation, formaldehyde exposure level was higher in residential dwellings where the indoor temperature was kept high. These findings are in agreement with those of other researchers who found that the increase of temperature causes the level of formaldehyde emission in the air to rise (Centers for Disease Control, 2008; Dannemiller et al., 2013; Dassonville et al., 2009; Heroux et al., 2010; Hodgson, Rudd, Beal, & Chandra, 2000; Lioy et al., 2011; Maruo et al., 2010). Another limitation of the current study and other studies is that the level of humidity in dwellings has not been measured. Experiments showed that increases in humidity, as well as temperature, have been reported to increase the emission of formaldehyde (Parthasarathy, Maddalena, Russell, & Apte, 2011). Formaldehyde can be emitted from gas cooking and heating systems (Dannemiller et al., 2013; Dassonville et al., 2009; WHO, 2010). However, the majority of the current study participants used electric cooking (91.3%) or heating systems (84%), and no relationship was found between the type of cooking and heating system and the level of formaldehyde. Therefore, for this sample, cooking and heating systems could not be considered as indoor residential sources of formaldehyde exposure.

Formaldehyde and season of data collection. No relationship was found between the season of data collection and the level of formic acid in the current study. As discussed before, this could be due to the failure of the urinary formic acid measure in detecting low level formaldehyde exposure (Schmid et al., 1994; Triebig et al., 1989;

164

Yasugi et al., 1992). Formaldehyde levels measured by vapor monitor badge, as both continuous and dichotomous variables, were significantly correlated with the season of data collection (winter, spring) and showed exposure level to be higher in the spring. These findings are in agreement with those of other researchers who found that the increase of temperature causes the level of formaldehyde emission in the air to rise (Centers for Disease Control, 2008; Dannemiller et al., 2013; Dassonville et al., 2009; Heroux et al., 2010; Hodgson et al., 2000; Lioy et al., 2011; Maruo et al., 2010).

Household practices. The second group of characteristics potentially contributing to indoor residential formaldehyde exposure was household practices. The use of nail polish was found to be associated with an increase in formaldehyde exposure level in the current study. Sainio et al. (1997) and Alaves et al. (2013) found formaldehyde levels to be above the National Institute for Occupational Safety and Health (NIOSH) recommended exposure limit (REL) level (0.016 ppm) in nail salons in Utah. However, in the current study, no information was collected regarding whether the participants used nail polish in residential dwellings or in nail salons. This was a limitation in the current study and is an area that should be explored in future studies. The use of air freshener and perfume are known to increase the level of formaldehyde as a result of the reaction of some of their components with ozone (Fan et al., 2003; Singer, Destaillats, Hodgson, & Nazaroff, 2006). Although some studies reported a significant relationship between the use of air freshener or perfume with the level of indoor formaldehyde exposure (Dassonville et al., 2009; Fan et al., 2003; Lamorena & Lee, 2008; Singer et al., 2006), the present study did not detect such

165

relationships. One reason for the inconsistences between studies’ findings is that, in the current study, it was not known if participants used air freshener or perfume while wearing the badge. Moreover, considering that the percentage of formaldehyde production differs based on the type of household product, such as brand of perfume or air freshener, the other limitation is that researchers in none of the studies, including the current study, determined the type and brand of perfume or air freshener that were used in the residential dwellings. Burning candles release a number of chemicals including formaldehyde (Petry et al., 2013). In this study, no relationship was found between formaldehyde exposure levels and use of candles. As noted previously, one explanation is that the participants were not asked if they used candles while wearing the badge. In addition, the level of formaldehyde differs in different brands and is higher in candles that are scented or contain paraffin wax (Lau, Fiedler, Hutzinger, Schwind, & Hosseinpour, 1997). In the current study, the type and brand of candles were not determined, which is a limitation. Another source of formaldehyde in residential dwellings is the use of household cleaning products (Dassonville et al., 2009; Singer et al., 2006). Those products increase the level of formaldehyde through reaction with ozone (ATSDR, 2010; WHO, 2010). However, in the current study, participants were not asked regarding the use of household cleaning products while wearing the badge. No relationship was found between the level of formaldehyde and tobacco smoke exposure (self-report). No published study noted a relationship between self-reported tobacco smoke exposure and the formaldehyde exposure level. However, this finding is consistent with the other finding of this study of no significant relationship between

166

urinary cotinine level and formaldehyde exposure level. In addition, self-report of tobacco smoke has some limitations, such as underestimating the true value of tobacco smoke exposure (Jhun et al., 2010; West, Zatonski, Przewozniak, & Jarvis, 2007), so this finding should be interpreted with caution.

Research Question 2 2a. Are there Relationships between Maternal Demographic Characteristics and Fetal Growth? The current study sample was selected based on inclusion and exclusion criteria. The inclusion criteria for the study included women who were pregnant with one fetus, between the ages of 19 to 40, and had at least one fetal ultrasound biometry report within the first and second trimester. The exclusion criteria were non-English speaking pregnant women, history of substance abuse, no prenatal care (or late initiation of prenatal care), history of chronic diseases, fetal abnormality, pregnancy resulting from infertility interventions, placental or amniotic fluid abnormalities, and a history of infection during the current pregnancy. The above criteria restricted the target population to a specific group who did not have risk factors known to influence fetal growth. Therefore, generalization of the results related to the relationships between maternal demographic and pregnancy characteristics with fetal growth should be avoided and comparison of the results with other studies should be interpreted carefully. In this study, no significant relationship was found between maternal age and fetal growth. The extremes of maternal age (less than 16 and above 40 years) are associated with impaired fetal growth (Roth et al., 1998; Scholl et al., 1996; Shah & Ohlsson, 2002).

167

Therefore, the reason for the inconsistency between the findings of this study and others could be related to the study design which included only pregnant women between the ages of 19-40. BPD in Whites was significantly lower than in other races (African-American and others) in the current study. Parker, Davies, and Newton (1982), found no significant difference between BPD in White and Asian fetuses up to 20 weeks of gestational age. However, Lai and Yeo (1995) demonstrated slightly smaller BPD in the Asian fetuses compared with White fetuses. Ogasawara (2009), did not find any significant differences between second trimester BPD in different ethnic groups of Hawaii (White, Asian, part Hawaiian, Pacific Islander, and Whites Asian); however, African Americans were not part of their study population. The inconsistencies between the current findings and those of other studies could be related to inclusion and exclusion criteria in the current study. No significant relationship was found between race and HC, AC, FL, and AC/FL in the current study. However, Parikh et al. (2013) found that African Americans have smaller AC in singleton pregnancies between 17-23 weeks of gestational age when compared to Whites. However, they found no differences in African Americans in BPD, HC, and FL. Yeo, Chan, Lun, and Lai (1994) reported that the average length of the fetal femur may even differ among various Asian subpopulations. Ogasawara (2009), reported shorter FL in the Asians than in the White population. Shipp, Bromley, Mascola, and Benacerraf (2001) noted that the fetuses of Asian women had shorter FL and fetuses of African American women had longer FL during 15 to 20 weeks of gestation age. As mentioned earlier, the inconsistencies between the findings of the current study with others could be due to the selected target population in the current study.

168

No relationship between maternal education and fetal growth was found in the current study, which is inconsistent with results in several other studies (Cunningham et al., 2010; Fescina et al., 2011; Lekea-Karanika, Tzoumaka-Bakoula, & Matsaniotis, 1999). In addition, Berngard et al. (2013) found that education is a predictor of fetal growth. Maternal education level has been associated with fetal growth (Cunningham et al., 2010; Fescina et al., 2011; Lekea-Karanika et al., 1999) due to a larger rate of some high risk behaviors in less educated mothers, such as smoking, alcohol, and drug use (Eisner, Brazie, Pratt, & Hexter, 1979). In this study, pregnant women who were identified as alcoholics or drug users in their medical records were excluded, and among sample participants, the frequency of self-reported smoking was low. Given that the selected sample in this study was mostly from low socioeconomic groups, the finding regarding education and fetal growth is consistent with Bobadilla, Ceron, and Suarez (1988) and Obel (1979) who reported no association between maternal education and fetal growth in low socioeconomic populations. In the current study, no relationship was found between employment and marital status and fetal growth. This finding is inconsistent with others’ findings that showed significant relationship between marital status and LBW (Lekea-Karanika et al., 1999; Vettore, Gama, Lamarca Gde, Schilithz, & Leal Mdo, 2010). Furthermore, in the present study, no relationship was found between income and employment status and fetal growth. Income and employment status are predictors of socioeconomic status. Significant relationships between socioeconomic status and fetal growth have been reported (Gaudineau, 2013; Kramer, 2013). However, as mentioned above, the disagreement between the findings of the current study and others could be a result of

169

recruiting from clinics that had the highest census of low socioeconomic populations and limiting the study target population based on inclusion and exclusion criteria.

2b. Are There Relationships between Pregnancy Characteristics (Gravida, Fetal Gender, Interval between Pregnancies, and Maternal Smoking Status) and Fetal Growth? While the relationship between gravida and fetal growth is well known (Cunningham et al., 2010; Fescina et al., 2011; Lekea-Karanika et al., 1999), in this study no significant relationship was found between gravida and fetal growth. However, the reason for disagreement between the findings of this study and others can be due to the inclusion and exclusion criteria of this study that was discussed previously. In addition, over one-third of the samples (36%) were primiparas, and comparatively few women had experienced more than two prior pregnancies (16.5%). In this study, there was a significant relationship between FL percentiles with fetal gender, where females were more likely to have a higher FL percentile. This result is inconsistent with Pang et al. (2003), Melamed et al. (2013), and Schwarzler, Bland, Holden, Campbell, and Ville (2004) who found no relationship between fetal gender and FL. One reason for the incongruity between the result of this study and others could be the gestational age of the selected samples, which included women in their second trimester in the current study and in their third trimester in the other studies. In addition, fetal growth is highly related to maternal body mass index before pregnancy and weight gain during pregnancy. This additional information was not collected in the current study (Cunningham et al., 2010; Lampl et al., 2010), therefore creating another limitation.

170

In this study, no significant relationship was found among the interval between two pregnancies and fetal growth. A short interval between two pregnancies has been reported to be associated with poor fetal growth (Fuentes-Afflick & Hessol, 2000; Gulmezoglu et al., 1997; Zhu et al., 2001). Given that women with shorter intervals between pregnancies are more likely to have other obstetrical risk factors such as young age (Fuentes-Afflick & Hessol, 2000; Zhu et al., 1999), one reason for the disagreement between studies, as mentioned previously, is the design of the current study that limited the participants to 19-40 years old and those with non-complicated pregnancies. Moreover, lack of data collection regarding maternal body mass index, height, and weight gain during pregnancy is another limitation. Tobacco smoke exposure, as measured by participant self-report, did not significantly relate to fetal growth ultrasound biomarkers in the current study. Tobacco smoke exposure during pregnancy is significantly associated with fetal growth in the third trimester (Bergsjo et al., 2007; Jaddoe et al., 2007; Prabhu et al., 2010; Pringle et al., 2005). Few studies have reported the effect of tobacco smoke exposure on fetal growth during the second trimester. However, as discussed before, self-report has some limitations and cannot be considered a reliable measurement method in determining tobacco smoke exposure in pregnant women. Thus, in this study, the level of cotinine as a biomarker of tobacco smoke was measured in urine. The details regarding the findings of cotinine and self-report are discussed later in this chapter.

171

Research Question 3 Do Levels of Formaldehyde Exposure, Indoor Residential Sources of Formaldehyde Exposure, and Tobacco Smoke Exposure Influence Fetal Growth? Formaldehyde (formic acid). No relationship was found between urinary formic acid and the second trimester fetal growth biomarkers. No published study had examined this association. However, due to the lack of sensitivity of urinary formic acid in determining low level formaldehyde exposure (Schmid et al., 1994; Triebig et al., 1989; Yasugi et al., 1992), these findings should be interpreted carefully.

Formaldehyde (vapor monitor badge). This is the only reported study that has examined the effect of formaldehyde exposure and second trimester fetal growth. In this study, a significant inverse relationship was found between formaldehyde exposure level and BPD. BPD is the most accurate measurement during the second trimester in determining gestational age (Tunon et al., 1999). Pedersen, Figueras, Wojdemann, Tabor, and Gardosi (2008) and Vasudeva et al. (2013) reported that slow growth of BPD between the first and second trimesters of pregnancy is a strong predictor of perinatal death before 34 weeks of gestational age. Additional studies need to examine this relationship to validate the evidence that formaldehyde exposure during pregnancy can adversely affect fetal growth. In the current study, no correlation was found between the level of formaldehyde exposure and the other fetal growth biomarkers of AC, FL, HC, EFW, and AC/FL. This could be due to the lack of sensitivity and specificity of those biomarkers in the second trimester (Fescina et al., 2011). The other reason is that, in the current study, the

172

calculations of EFW, AC, FL, BPD, and HC were conducted using Hadlock et al. (1984) formulas. These formulas were based on a study group that consisted of only one race (Caucasian women), and in this study approximately 54% of the population were minorities. The Hadlock et al. formulas are commonly used in the US, however, racial diversity and fetal gender, which are important determinants in fetal growth (Aibar, Puertas, Valverde, Carrillo, & Montoya, 2012; Ioannou et al., 2012; Krampl et al., 2000; Radulescu, Ferechide, & Popa, 2013), have not been considered in the formulas. No previous study has examined the relationship between formaldehyde exposure and fetal growth biomarkers. Additionally, there are only two other studies which have examined the relationship between outdoor formaldehyde exposure level and pregnancy outcomes and in both a relationship between the level of formaldehyde exposure and LBW was found (Grazulevicience et al., 1998; Maroziene & Grazuleviciene, 2002). Given that LBW can be a biomarker of poor fetal growth (Cunningham et al., 2010), the finding of the current study is consistent with these other studies. Nonetheless, due to lack of related studies, it is not possible to draw any firm conclusion regarding the effect of formaldehyde exposure during pregnancy on fetal growth. The current study was an exploratory study, which needs to be repeated.

Indoor residential sources of formaldehyde exposure. The results of this study found no significant relationships between residential dwelling characteristics and fetal ultrasound biometric measurements. No published study has examined the relationships between indoor residential sources of formaldehyde exposure and fetal growth. Limitations related to data collection on indoor residential sources of formaldehyde were

173

discussed in the previous section, and these limitations should be taken into account in the design of future studies.

Tobacco smoke exposure. A significant relationship was found between FL and the level of cotinine in this study, which is in agreement with other studies (Iniguez et al., 2013; Jaddoe et al., 2007; Pringle et al., 2005). However, this relationship did not remain significant after controlling for fetal gender. Jaddoe et al. (2007) and Pringle et al. (2005) did not consider fetal gender as a confounder in their studies. Conversely, Iniguez et al. (2013) found a significant difference between FL and cotinine level even after accounting for fetal gender in their statistical model. Furthermore, Bergsjo et al. (2007) found no significant relationship between fetal ultrasound biomarkers and maternal smoking in the second trimester. The inconsistency between the studies could be related to gestational age at the time of data collection and the time of urine sample collection. In addition, the limitation of the current study regarding lack of information about the mother’s weight before pregnancy and weight gain during pregnancy should be considered in interpretation of this correlation.

Research Question 4 Does Oxidative Stress Mediate the Relationships between (a) Level of Formaldehyde Exposure and Fetal Growth, (b) Indoor Residential Sources of Formaldehyde Exposure and Fetal Growth, and (c) Tobacco Smoke Exposure and Fetal Growth? This was the first known study to examine the potential mediating role of oxidative stress in the relationship between formaldehyde and tobacco smoke exposure

174

and fetal growth during pregnancy. To examine the mediating role of oxidative stress, the first step was to determine the relationships between formaldehyde (urinary formic acid and vapor monitor) and cotinine exposure with fetal ultrasound biometric measurements. A relationship was found between formaldehyde (vapor monitor) exposure and BPD percentile, but not with cotinine. The second step was to determine the relationship between formaldehyde and cotinine level with15- isoprostane F2t, which was the biomarker used as an indicator of oxidative stress. No relationship was found between 15- isoprostane F2t and the formaldehyde (vapor monitor), urinary formic acid, or cotinine. In addition, as a third step of the mediation model, no relationship was found between fetal ultrasound biometric measurements and the level of 15- isoprostane F2t. Therefore, the mediation model was not supported in this study. A relationship has been found between formaldehyde and oxidative stress in pregnant rats’ maternal and fetal tissues (Gulec et al., 2006; Gurel et al., 2005; Im et al., 2006; Kum et al., 2007; Sögüt et al., 2004). However, in this study, no relationship was found between formaldehyde (vapor monitor and urinary formic acid) 15-isoprostane F2t. The only human study is the study by Romanazzi et al. (2013) who investigated the role of occupational exposure to formaldehyde in the stimulation of oxidative stress status among male workers and found a relationship between formaldehyde exposure level and the level of 15-isoprostane F2t. In that study, the participants were employed in the manufacture of plastic laminates, and thus exposed to higher levels of formaldehyde, and all were male, ages 36-44. There are no published studies that have investigated the role of formaldehyde exposure on oxidative stress during pregnancy. In addition, there are no published studies that have described the physiological changes of oxidative

175

stress during pregnancy. Therefore, this study was an exploratory study and needs to be repeated with a larger sample size or through cellular or animal studies. In the current study, no correlation was found between cotinine and 15isoprostane F2t. There is no published study that has examined the relationship between tobacco smoke exposure and oxidative stress during pregnancy. However, Bono et al. (2013) found a relationship between passive tobacco smoke exposure and 15-isoprostane F2t in adolescents (12-14 years old). The results of the Bono et al. (2013) study are questionable because they found a significant effect of age of participants on oxidative stress. The finding of Bono et al. (2013) regarding the effect of age on oxidative stress is noteworthy for future studies to consider age as a confounder variable, considering that the range of age distribution in their study was excessively narrow (12-14 years old). In addition, Romanazzi et al. (2013) found a relationship between cotinine and 15isoprostane F2t in male workers in the plastic laminates industry. In the latter research, the participants were male aged 36-44. In addition, some other studies have found a correlation between tobacco smoke and other biomarkers of oxidative stress such as urinary 8-hydroxydeoxyguanosine (8-OH-dG), malondialdehyde (MDA), and placenta oxidative stress index (Aycicek et al., 2011; Nishizawa et al., 2005; Tabacova et al., 1998). The main reason for the disagreement with the findings in the current study could be due to the differences in the oxidative stress biomarkers that were studied. In the current study no relationship was found between 15-isoprostane F2t and fetal ultrasound biomarker measurements. No other study examined this relationship in the second trimester. However, several studies did show an association between oxidative stress and birth outcomes, such as LBW (Biri et al., 2007; Burton, Yung,

176

Cindrova-Davies, & Charnock-Jones, 2009; Guvendag Guven, Karcaaltincaba, Kandemir, Kiykac, & Mentese, 2013; Mert et al., 2012). The inconsistency between the results of the current study and others could be due to the gestational age at the time of data collection. Although there is no evidence regarding the mediating role of oxidative stress in the relationship between formaldehyde exposure and fetal growth, there could be several explanations for non-significant findings between the formaldehyde level and 15isoprostane F2t in the current study. First, the physiological changes in oxidative stress and 15-isoprostane F2t metabolism have not been studied during pregnancy. The sample in the study in which a relationship was found (Romanazzi et al., 2013), was male workers who had a higher level of formaldehyde exposure. Second, pregnancy by itself is a state of oxidative stress, which is related to increased metabolic activity in the placental mitochondria (Wisdom, Wilson, McKillop, & Walker, 1991) and production of oxidative stress due to the hypoxia-reoxygenation injury following the establishment of intervillous space blood flow (Jauniaux et al., 2000; Myatt & Cui, 2004), Third, urinary isoprostane is a non-specific oxidative stress biomarker (Romanazzi et al., 2013). Although 15-isoprostaneF2t has been the most studied isomer and is a good indicator of oxidative stress, it relies on the amount of antioxidants that have been taken before collection of urine (Halliwell & Gutteridge, 2007; Larose, Julien, & Bilodeau, 2013). Fourth, evidence shows that exposure to risk factors such as air pollutants that interfere with placental efficiency through oxidative stress, are accompanied with higher risk for pregnancy induced hypertension (pre-eclampsia and eclampsia) (Huang et al., 2013). However, in this study any women with complicated pregnancies, including those with

177

pregnancy-induced hypertension, were excluded from the study. Lastly, this is an exploratory study with a limited sample size; therefore, larger studies are needed to validate the current findings.

Conceptual Framework In this study, the biological pathway suggested by Kannan et al. (2006) was the basis for the conceptual framework that was used. Kannan et al. (2006) proposed metabolic pathways of the particulate matter (PM), small air particles, on pregnancy outcomes. PMs are complex mixtures of chemical and biological elements, and/or liquid droplets, such as metals, salts, volatile organic compounds (VOC), and polycyclic aromatic hydrocarbons (PAH) (Billet et al., 2007; EPA, 2012; Schlesinger, Kunzli, Hidy, Gotschi, & Jerrett, 2006). The PM composition is primarily related to the organic compounds, such as VOCs (Arhami et al., 2010; Liden et al., 2003; Osornio-Vargas et al., 2003; Valavanidis et al., 2008). The EPA classified formaldehyde as a VOC (EPA, 2011). Although, based on the principal investigator’s knowledge, no study has identified formaldehyde as an element of PM; though many studies have reported other VOCs, such as benzene, toluene, and styrene, as components of PM. This could be due to the fact that PM is mostly an outdoor air pollutant and formaldehyde is an indoor air pollutant. In their framework, Kannan et al. (2006) suggested five exclusive biologic pathways (oxidative stress, pulmonary and placental inflammation, blood coagulation, endothelial function, and hemodynamic response) as mediators in the relationship between PM and fetal growth outcomes. However, in the conceptual model of the current study only oxidative stress was proposed as a mediator in the relationship between formaldehyde exposure and fetal growth. Oxidative stress is the commonality between 178

the biological pathway of PM (Arhami et al., 2010; Kannan et al., 2006; Valavanidis et al., 2008) and formaldehyde (Siu, Shapiro, Wiley, & Wells, 2013; Tulpule & Dringen, 2013; Wang, Li, Liu, & Jin, 2013; Y. Zhang et al., 2013). Oxidative stress is also associated with placental insufficiency and pregnancy outcomes including LBW (Biri et al., 2007; Burton et al., 2009; Guvendag Guven et al., 2013; Mert et al., 2012). In this study, due to lack of correlations between 15-isoprostane F2t and outcome variables, the mediating role of oxidative stress was not supported. In addition, no correlation was found between the level of formaldehyde exposure and oxidative stress and no association was reported between oxidative stress and fetal growth biomarkers. However, based on the evidence, the selected biological framework still appears to be appropriate. The same framework should be used for future studies with a larger sample size and with other measures of oxidative stress.

Tobacco Smoke Exposure: Cotinine and Self-Report In this study, the mean level of tobacco smoke based on creatinine-standardized cotinine in urine ELISA was 152.93 µg/g (SD = 463.7 range from 0-3893.0 µg/g). The prevalence of smoking in this study, based on self-report, was 15%, which is similar to the smoking rate of 15.9% published in the report of the Pregnancy Risk Assessment Monitoring System (PRAMS) Surveillance Report for Alabama in 2010 (Zheng, Wouolbright, & Afgan, 2012). However, in the Selected Maternal and Child Health Statistics Alabama (2011) report, the rate of smoking during pregnancy was reported to be 10.6% in Alabama and 8% in Madison County, where this study was conducted. Moreover, in 2003, the prevalence of smoking during pregnancy was reported at 10.7%

179

in the US (Martin et al., 2003). Therefore, the level of tobacco smoke by self-report in this study is higher than the published surveys. This could be due to enrollment of a high number of low socioeconomic populations in this study, which is a group known to have a higher prevalence of smoking (Zheng et al., 2012). The inclination to not report tobacco smoking is higher in pregnant women (Webb, Boyd, Messina, & Windsor, 2003). Therefore, self-reported smoking prevalence can underestimate the true value (Jhun et al., 2010; West et al., 2007). However, West et al. (2007) reported that self-reported tobacco smoking minimally underestimated true tobacco smoking prevalence in the US. Webb et al. (2003) studied the discrepancy between self-reported smoking behavior and actual urine cotinine values among prenatal patients and reported that about three of every four self-reported nonsmokers had cotinine values greater than 80 ng/mL and about half of them had values exceeding 100 ng/ml. The same results have been reported by others (Lindqvist et al., 2002; Parna et al., 2005; Post et al., 2008). The level of urine cotinine was measured in this study to have a better understanding of the true level of tobacco smoke exposure in pregnant women. However, there is not a standard cutoff point for active or ETS. Additionally, the cut off points vary in different ethnic groups (Lindqvist et al., 2002; Parna et al., 2005; Post et al., 2008). Therefore, the comparison of the prevalence rate in different studies may not reflect the real variance. In this study a significant relationship was found between self-report and cotinine level. Out of several cut off points from different countries, two that were conducted in the United States were selected for calculating the rate of tobacco smoke in this study

180

(Pickett et al., 2005; SRNT, 2002). Considering SRNT’s cut off point of 50 µg/g (SRNT, 2002), the prevalence of tobacco smoke exposure increased up to 26.1%. However, when considering Pickett’s cut off point of 200 µg/g, the prevalence based on cotinine (15.9%) is similar to self-report (Pickett et al., 2005). As suggested earlier, the gestational age at the time of urine collection, time of urine sampling, and standardization for creatinine should be considered when discussing the results of different studies. In addition, selfreported data on smoking in pregnant women may underestimate the real smoking prevalence. This underestimation may result in failure by health care providers to initiate tobacco cessation practices and interventions during pregnancy. Therefore, a reliable measurement method, such as urinary cotinine, may be appropriate to be included in the prenatal care panel.

Study Limitations Several limitations must be considered with the design of this study, including instrumentation, one time data collection, and spot urine sampling. In addition, in this study it was not possible to control some covariates on fetal growth, for example, weight before pregnancy, height, and weight gain during pregnancy. Furthermore, this study was not able to evaluate the effect of other air pollutants, such as PMs, on fetal growth. One limitation of this study was the fetal ultrasound biometric measurement method. The gestational age and percentiles in this study were calculated based on the formula by Hadlock et al. (1984) through a software program that was installed on the ultrasound computer in all four clinics. Hadlock et al. performed their research in a group of middle class Caucasian women from the Houston area. Since one single fetal

181

ultrasound biometry measurement is not applicable across all ethnicities (Parikh et al., 2013) or for both genders (Melamed et al., 2013), the use of the Hadlock formulations in this study may have over or underestimated percentiles of fetal ultrasound biometric measurements. The other noteworthy point was the possibility of systemic variations in measurement accuracy between different ultrasound technicians and ultrasound devices (Ioannou et al., 2012). All of the technicians reportedly were experienced and received annual retraining and all of the clinics used devices from the same manufacturer. However, direct observation of the technicians was not feasible; therefore systematic differences among them may have been occurred. Additionally, the accuracy of fetal ultrasound biometric measurements depends on the position of the fetus (Ioannou et al., 2012; Kurmanavicius et al., 1999). These factors represented potential threats to internal validity that could not be controlled for in this study. Another potential limitation was the timing of the fetal growth measurements. Although fetal growth, including fetal weight and abdominal circumference, occurs substantially during the third trimester (after 28 weeks of pregnancy), the routine ultrasound is conducted at the beginning of the second trimester between 17-28 weeks as part of standard prenatal care. In addition, based on health policies, a third trimester ultrasound is conducted in high risk pregnancies, but these were excluded from this study. In this study, the principal investigator collected the fetal ultrasound biometry (from the medical record) during the 17-28 weeks period in uncomplicated pregnancies. Moreover, only one time biometric measurements were selected for each fetus, however, repeated measure designs of fetal ultrasound biometric measures, especially BPD, are better than

182

an isolated one time estimation (Vasudeva et al., 2013). Use of repeat measures should be considered in the design of future studies. In the current study, several risk factors that impact fetal growth were controlled through exclusion criteria in sample selection or through statistical analysis as confounders. However, some risk factors were not measured in this study, including body mass index before pregnancy and weight gain during pregnancy. Those variables also are associated with poor fetal growth (Cunningham et al., 2010; Fescina et al., 2011; Rode et al., 2007). These factors were not included because of the lack of information in the medical records. Limitations of the designed questionnaire for measuring the indoor formaldehyde sources in the residential dwellings should be taken into account as well. The information regarding type, brand, and amount of consumed candles or perfumes had not been included in the questionnaire. In addition, the participants were asked about their household activities at the time of interview, which was before wearing the badge. This is a limitation because their daily activities before wearing the vapor monitor badge may not reflect their activities at the time of sampling. Therefore, there was lack of information regarding the participants’ household activities, such as use of cleaning products or using nail polish while actually wearing the vapor monitor badge. The focus of the current study was on formaldehyde exposure. Other indoor air pollutants (e.g., styrene, nitrogen di-oxide, carbon monoxide, polycyclic aromatic hydrocarbons [PAHs] and other volatile organic compounds), which could be associated with poor fetal growth (Choi, Wang, Lin, Spengler, & Perera, 2012; Cunningham et al., 2010; Venditti, Casselman, & Smith, 2011), were not measured. Future studies should

183

examine the accumulative effect of indoor and outdoor air pollutants such as particulate matter on fetal growth.

Implications for Nursing Practice and Education Barker, the author of the fetal programming hypothesis, stressed the importance of human fetal developmental experiences in determining patterns of diseases in the human life course (Barker, 1995; Barker et al., 1993). The prenatal period is the time that fetal tissues develop in a specific sequence from conception to maturity, therefore, the fetus is vulnerable to both organizing and disorganizing influences on organ development (Barker & Fall, 1998; Bateson et al., 2004; Henrichs et al., 2009). Appropriate fetal growth reduces the long term and short term complications that greatly affect the quality of life of the newborn and the family (WHO & UNICEF, 2004). Although environmental exposures are one of the risk factors for IUGR (Grazulevicience et al., 1998; Kannan et al., 2006; Maroziene & Grazuleviciene, 2002), prenatal care programs typically do not focus on environmental exposures. Based on this study, formaldehyde may be associated with poor fetal growth. Therefore, one recommendation for health care providers is to promote educational interventions to reduce the exposure level to environmental toxic chemicals during pregnancy. Since this is an exploratory research and no prior research has reported on the effect of formaldehyde on pregnancy outcomes, health care providers themselves need to be aware of the risks associated with exposure to environmental toxic chemicals, based on the recommendation of ACOG (The American College of Obstetricians and Gynecologists, 2013a). The American College of Obstetricians and Gynecologists (ACOG) and the American Society for Reproductive Medicine (ASRM) consider environmental chemicals 184

as important risks factors for reproductive health that need to be identified and reduced through policy changes and education (The American College of Obstetricians and Gynecologists, 2013a). Therefore, healthcare providers should provide information during prenatal care about the sources of toxic chemicals such as nail polish (Alaves et al., 2013; Sainio et al., 1997), new carpet (Cracowski, Carpentier, et al., 2002; Dassonville et al., 2009; Hodgson et al., 1993; Rogers et al., 2007), and new furniture (Dannemiller et al., 2013; Dassonville et al., 2009; Kim et al., 2013). They also need to consider educating the pregnant women on cost-effective steps to reduce exposure to formaldehyde such as opening the windows if appropriate, adding non-toxic potted plants in the residential dwellings (Orwell, Wood, Burchett, Tarran, & Torpy, 2006; Wood et al., 2006), or polishing nails while outdoors rather than indoors. However, this study was exploratory, therefore, larger studies are necessary to validate the findings of this study. This study, consistent with other studies, showed that pregnant women’s selfreport regarding smoking is not reliable based on the cutoff point of 50ng/ml (Center for Disease Control, 2006; EPA, 2008; Lindqvist et al., 2002; Parna et al., 2005; Post et al., 2008). Therefore, maternal disinclination to declare smoking during pregnancy through self-report needs to be taken into account in the practice of maternal and child health and tobacco smoke cessation programs. Adding urinary cotinine laboratory tests, along with other prenatal laboratory tests at a pregnant woman’s first visit can be helpful in determining active smokers and ETS.

Implications for Future Research Identification of the short term and long term effects of formaldehyde exposure during pregnancy and its effect on fetal growth should be the focus of future studies. In 185

addition, more research on environmental toxicants that may affect children’s health in the short term or long term is needed. Furthermore, results from this study and ACOG’s statement regarding reduction of exposure to environmental toxic chemicals during pregnancy (The American College of Obstetricians and Gynecologists, 2013a) can be used to support the need for interventional studies regarding educating health professionals and pregnant women about formaldehyde exposure. Barker’s hypothesis stresses long term health effects of poor fetal growth, low birth weight, and preterm newborns. These long term health effects include hypertension, type two diabetes, and coronary artery diseases in adulthood (Almond & Currie, 2011; Barker & Fall, 1998; Barker, 1995; Bateson et al., 2004; Black, Devereux, & Salvanes, 2007; Fall, Vijayakumar, Barker, Osmond, & Duggleby, 1995; Madden, 2013). Therefore, longitudinal studies should be developed to evaluate the long term health effects of exposure to environmental toxicants including formaldehyde during pregnancy in different developmental stages. One finding of this study was the relationship between exposure to higher levels of formaldehyde and smaller BPD. However, this study was considered as an exploratory study; further research with a larger sample size is needed to validate the findings. Replication of this study with more representative samples, and with less exclusion criteria, would allow the investigators to include some at risk pregnancies that are associated with oxidative stress, for example, pre-eclamptic and eclamptic patients (Huang et al., 2013), and would increase the generalizability and applicability of the results (Polit & Beck, 2008). A prospective, longitudinal design that includes the repetitive fetal ultrasound biometric measurements, as well as birth characteristics, such

186

as birth weight, head circumference, abdominal circumference, and height, would validate the current findings about the relationships between formaldehyde exposure and fetal growth (Polit & Beck, 2008). The findings of this study showed no relationship between vapor monitor badges and urine formic acid, therefore, additional studies need to be conducted using blood samples or tissues to determine the level of formaldehyde in serum or DNA-cross-links in tissue as biological markers of exposure to formaldehyde (Polit & Beck, 2008). In addition, further animal or cellular studies that allow the investigators to expose the laboratory animals or cells to different levels of formaldehyde for investigating the metabolism of formaldehyde, the role of oxidative stress, and its effect on pregnancy outcomes would be beneficial. In particular, the biological pathway that may interfere with pregnancy outcomes needs further investigation. In this study the mediating role of oxidative stress was not recognized. However, oxidative stress can be influenced by many confounding factors, for example nutrition and maternal stress (Halliwell & Gutteridge, 2007). Therefore, animal and cellular studies can be designed to eliminate the effect of extraneous variables in the relationship between formaldehyde exposure and oxidative stress. This study found relationships between nail polish, new carpets, and house remodeling with formaldehyde exposure level. Therefore, exploration of the pregnancy outcomes in women who work in nail salons and carpet companies would be valuable. In addition, it would be important to study the pregnancy outcomes in women who lived in a new or remodeled house during pregnancy. In this study, there was a relationship between the level of formaldehyde and remodeled residential dwellings. However, there

187

were not enough participants who lived in new or remodeled residential dwellings to determine fetal growth outcomes in this subsample. Therefore, additional studies with a larger number of new residential dwellings are valuable. Interventional studies can be designed that incorporate environmental hazard risk reduction (including formaldehyde and cotinine) in prenatal care classes, focusing on both the risks for the developing fetus and mother. Health care professionals, such as nurses and physicians, can be considered as study populations for training as well. The curricula of the related courses (child health or obstetric) in health professional schools (e.g., nursing, medical), can be improved by the addition of children’s environmental health sections. Gains in knowledge and risk reduction skills may be examined to determine knowledge/practice changes in health professionals. A larger sample size of pregnant women, who are exposed to first or ETS, would increase the power and generalizability of the study findings regarding the effect of tobacco exposure on fetal growth during the second trimester (Polit & Beck, 2008). Measurement of cotinine in serum or periodical measurement of urine cotinine would increase the accuracy of the findings regarding tobacco exposure and could be used to determine a cotinine cut-off point for first and secondhand smokers or ETS in the US, taking into account the diverse ethnicities.

Summary In this study, the principal investigator examined the relationship among formaldehyde and tobacco exposure and fetal growth in 140 pregnant women from one university-associated and three private clinics in the southeast US. In this group of

188

pregnant women, the mean level of formaldehyde exposure measured 0.04 ppm using a vapor monitor badge. There is no other published study that examined pregnant women; however, the exposure level was higher than other personal exposure levels that were collected from children or non-pregnant adults. No relationship was found between the vapor monitor badges and urinary formic acid, which was supported by other studies. The findings did not support tobacco smoke as a source of formaldehyde in the residential areas, which is inconsistent with several other studies. The relationship between indoor sources of formaldehyde and formaldehyde exposure level was supported by other studies considering the effect of season, indoor temperature, new carpet, nail polish, and home remodeling. A relationship was found between race and BPD; however, due to limiting the study target population to healthy pregnancies, this result cannot be generalized. One of the fetal ultrasound biometry measurements, BPD, was found to have a relationship between the level of exposure to formaldehyde, although there is no other published evidence for comparison. The hypothesized mediating model which speculated the mediating role of oxidative stress in the relationship between formaldehyde exposure and fetal growth, based on Kannan et al. (2006), did not fit the data in the current study. This conceptual model is novel with no other study examining it previously and it should be evaluated in future studies that are designed to address the limitations of the present study.

189

References Aagaard-Tillery, K.M., Porter, T.F., Lane, R.H., Varner, M.W., & Lacoursiere, D.Y. (2008). In utero tobacco exposure is associated with modified effects of maternal factors on fetal growth. American Journal of Obstetrics and Gynecology,, 198(1), 66.e61-66. doi: 10.1016/j.ajog.2007.06.078 Abel, K.M., Dalman, C., Svensson, A.C., Susser, E., Dal, H., Idring, S., . . . Magnusson, C. (2013). Deviance in fetal growth and risk of autism spectrum disorder. Am J Psychiatry, 170(4), 391-398. doi: 10.1176/appi.ajp.2012.12040543 Abrams, B., & Selvin, S. (1995). Maternal weight gain pattern and birth weight. Obstetrics and Gynecology, 86(2), 163-169. Adcock, I.M., Brown, C.R., Kwon, O., & Barnes, P.J. (1994). Oxidative stress induces NF kappa B DNA binding and inducible NOS mRNA in human epithelial cells. Biochemical and Biophysical Research Communications,, 199(3), 1518-1524. doi: 10.1006/bbrc.1994.1403 Advanced Chemical Sensors, I. (2012). Formaldehyde vapor monitor: Catalog No. F-50. Advanced Chemical Sensors, Inc: Technical information sheet. Retrieved July, 17, 2012, from http://acsbadge.com/formaldehyde.shtml Advanced Chemical Sensors Inc. (2011). The ACS formaldehyde vapor monitor: Validation of performace. Retrieved March, 31, 2013, from http://acsbadge.com/formaldehyde.shtml

190

Aibar, L., Puertas, A., Valverde, M., Carrillo, M.P., & Montoya, F. (2012). Fetal sex and perinatal outcomes. Journal of Perinatal Medicine,, 40(3), 271-276. doi: 10.1515/jpm-2011-0137 Al-Saraj, A. (2009). Teratogenic effect of formaldehyde in rabbits. Iraqi Journal of Veterinary Sciences, 23(1), 1-4. Alaves, V.M., Sleeth, D.K., Thiese, M.S., & Larson, R.R. (2013). Characterization of indoor air contaminants in a randomly selected set of commercial nail salons in Salt Lake County, Utah, USA. International Journal of Environmental Health Research,. doi: 10.1080/09603123.2012.755152 Almond, D., & Currie, J. (2011). Killing me softly: The fetal origins hypothesis. The Journal of Economic Perspectives, 25(3), 153-172. doi: 10.1257/jep.25.3.153 Altman, D.G., & Chitty, L.S. (1997). New charts for ultrasound dating of pregnancy. Ultrasound Obstet Gynecol, 10(3), 174-191. doi: 10.1046/j.14690705.1997.10030174.x America's Health Rankings. (2011). E-ranking's by measure. Retrieved March, 2, 2011, from http://www.americashealthrankings.org/Measure.aspx Arany, I., Grifoni, S., Clark, J.S., Csongradi, E., Maric, C., & Juncos, L.A. (2011). Chronic nicotine exposure exacerbates acute renal ischemic injury. American Journal of Physiology. Renal Physiology,, 301(1), F125-133. doi: 10.1152/ajprenal.00041.2011 Arhami, M., Minguillon, M.C., Polidori, A., Schauer, J.J., Delfino, R.J., & Sioutas, C. (2010). Organic compound characterization and source apportionment of indoor

191

and outdoor quasi-ultrafine particulate matter in retirement homes of the Los Angeles Basin. Indoor Air, 20(1), 17-30. doi: 10.1111/j.1600-0668.2009.00620.x Ashok, A., Sajal, G., & Rakesh, K.S. (2005). Role of oxidative stress in female reproduction. Reproductive Biology and Endocrinology, 3(1), 28-28. doi: 10.1186/1477-7827-3-28 ATSDR. (2010). Amendment to the Toxicological Profile for Formaldehyde. Retrieved November, 3, 2013, from http://www.atsdr.cdc.gov/toxprofiles/tp111-p.pdf Aurrekoetxea, J.J., Murcia, M., Rebagliato, M., Lopez, M.J., Castilla, A.M., SantaMarina, L., . . . Ballester, F. (2013). Determinants of self-reported smoking and misclassification during pregnancy, and analysis of optimal cut-off points for urinary cotinine: A cross-sectional study. BMJ Open, 3(1). doi: 10.1136/bmjopen2012-002034 Aycicek, A., Varma, M., Ahmet, K., Abdurrahim, K., & Erel, O. (2011). Maternal active or passive smoking causes oxidative stress in placental tissue. European Journal of Pediatrics,, 170(5), 645-651. doi: 10.1007/s00431-010-1338-9 Bada, H.S., Das, A., Bauer, C.R., Shankaran, S., Lester, B.M., Gard, C.C., . . . Higgins, R. (2005). Low birth weight and preterm births: Etiologic fraction attributable to prenatal drug exposure. Journal of Perinatology,, 25(10), 631-637. doi: 10.1038/sj.jp.7211378 Baker, R.R. (2006). The generation of formaldehyde in cigarettes--Overview and recent experiments. Food and Chemical Toxicology,, 44(11), 1799-1822. doi: 10.1016/j.fct.2006.05.017

192

Bakhireva, L.N., Schatz, M., & Chambers, C.D. (2007). Effect of maternal asthma and gestational asthma therapy on fetal growth. Journal of Asthma,, 44(2), 71-76. doi: 10.1080/02770900601180313 Balhara, Y.P., & Jain, R. (2013). A receiver operated curve-based evaluation of change in sensitivity and specificity of cotinine urinalysis for detecting active tobacco use. Journal of Cancer Research and Therapeutics,, 9(1), 84-89. doi: 10.4103/09731482.110384 Balhara, Y.P., Jain, R., Sundar, A.S., & Sagar, R. (2012). Use of cotinine urinalysis to verify self-reported tobacco use among male psychiatric out-patients. Lung India, 29(3), 217-220. doi: 10.4103/0970-2113.99102 Barker, D., & Fall, C. (1998). The immediate and long-term consequences of low birthweight. Technical Consultation on Low Birthweight. New York: UNICEF. Barker, D.J. (1995). Fetal origins of coronary heart disease. BMJ, 311(6998), 171-174. Barker, D.J., Hales, C.N., Fall, C.H., Osmond, C., Phipps, K., & Clark, P.M. (1993). Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): Relation to reduced fetal growth. Diabetologia, 36(1), 62-67. Barker, D.J., & Osmond, C. (1986). Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet, 1(8489), 1077-1081. Barker, D.J., Osmond, C., Forsen, T.J., Kajantie, E., & Eriksson, J.G. (2005). Trajectories of growth among children who have coronary events as adults. The New England Journal of Medicine,, 353(17), 1802-1809. doi: 10.1056/NEJMoa044160

193

Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol, 51(6), 1173-1182. Bartnik, F.G., Gloxhuber, C., & Zimmermann, V. (1985). Percutaneous absorption of formaldehyde in rats. Toxicology Letters, 25(2), 167-172. Bateson, P., Barker, D., Clutton-Brock, T., Deb, D., D'Udine, B., Foley, R.A., . . . Sultan, S.E. (2004). Developmental plasticity and human health. Nature, 430(6998), 419421. doi: 10.1038/nature02725 Baumann, K., & Angerer, J. (1979). Occupational chronic exposure to organic solvents. VI. Formic acid concentration in blood and urine as an indicator of methanol exposure. International Archives of Occupational and Environmental Health, 42(3-4), 241-249. Benowitz, N.L., Dains, K.M., Dempsey, D., Herrera, B., Yu, L., & Jacob, P. (2009). Urine nicotine metabolite concentrations in relation to plasma cotinine during low-level nicotine exposure. Nicotine & Tobacco Research, , 11(8), 954-960. doi: 10.1093/ntr/ntp092 Benowitz, N.L., & Jacob, P. (1994). Metabolism of nicotine to cotinine studied by a dual stable isotope method. Clinical Pharmacology and Therapeutics,, 56(5), 483-493. doi: 10.1038/clpt.1994.169 Bergman, B.C., Perreault, L., Hunerdosse, D.M., Koehler, M.C., Samek, A.M., & Eckel, R.H. (2009). Intramuscular lipid metabolism in the insulin resistance of smoking. Diabetes, 58(10), 2220-2227. doi: 10.2337/db09-0481

194

Bergsjo, P., Bakketeig, L.S., & Lindmark, G. (2007). Maternal smoking does not affect fetal size as measured in the mid-second trimester. Acta Obstetricia et Gynecologica Scandinavica,, 86(2), 156-160. doi: 10.1080/00016340600984696 Berngard, S.C., Berngard, J.B., Krebs, N.F., Garces, A., Miller, L.V., Westcott, J., . . . Hambidge, K.M. (2013). Newborn length predicts early infant linear growth retardation and disproportionately high weight gain in a low-income population. Early Hum Dev, 89(12), 967-972. doi: 10.1016/j.earlhumdev.2013.09.008 Bernstein, I., Gabbe, S.G., & Reed, K.L. (2002). Intrauterine growth restriction. In S. G. Gabb, J. R. Neibyl & J. L. Simpson (Eds.), Obstetrics: Normal and problem pregnancies (4th ed., pp. 869-891). Philadelphia: Churchill Livingstone. Bhatt, H.S., Lober, S.B., & Combes, B. (1988). Effect of glutathione depletion on aminopyrine and formaldehyde metabolism. Biochemical Pharmacology, 37(8), 1581-1589. Billet, S., Garcon, G., Dagher, Z., Verdin, A., Ledoux, F., Cazier, F., . . . Shirali, P. (2007). Ambient particulate matter (PM2.5): Physicochemical characterization and metabolic activation of the organic fraction in human lung epithelial cells (A549). Environmetal Research, 105(2), 212-223. doi: 10.1016/j.envres.2007.03.001 Billings, R.E. (1984). Interactions between folate metabolism, phenytoin metabolism, and liver microsomal cytochrome P450. Drug Nutr Interact, 3(1), 21-32. Biri, A., Bozkurt, N., Turp, A., Kavutcu, M., Himmetoglu, O., & Durak, I. (2007). Role of oxidative stress in intrauterine growth restriction. Gynecologic and Obstetric Investigation,, 64(4), 187-192. doi: 10.1159/000106488

195

Black, S.E., Devereux, P.J., & Salvanes, K.G. (2007). From the cradle to the labor market? The effect of birth weight on adults outsomes. Quarterly Journal of Economics, 122(1), 409-439. doi: 10.1162/qjec.122.1.409 Bobadilla, J.L., Ceron, S., & Suarez, P. (1988). [Evaluation of the quality of data on perinatal mortality. Vital statistics registries in Mexico]. Salud Publica de Mexico,, 30(1), 101-113. Bono, R., Bellisario, V., Romanazzi, V., Pirro, V., Piccioni, P., Pazzi, M., . . . Vincenti, M. (2013). Oxidative stress in adolescent passive smokers living in urban and rural environments. International Journal of Hygiene and Environmental Health,. doi: 10.1016/j.ijheh.2013.06.008 Bono, R., Vincenti, M., Schiliro, T., Scursatone, E., Pignata, C., & Gilli, G. (2006). NMethylenvaline in a group of subjects occupationally exposed to formaldehyde. Toxicology Letters,, 161(1), 10-17. doi: S0378-4274(05)00238-9 [pii]10.1016/j.toxlet.2005.07.016 Bowers, K., Laughon, S.K., Kiely, M., Brite, J., Chen, Z., & Zhang, C. (2013). Gestational diabetes, pre-pregnancy obesity and pregnancy weight gain in relation to excess fetal growth: Variations by race/ethnicity. Diabetologia, 56(6), 12631271. doi: 10.1007/s00125-013-2881-5 Braun, J.M., Daniels, J.L., Poole, C., Olshan, A.F., Hornung, R., Bernert, J.T., . . . Lanphear, B.P. (2010). A prospective cohort study of biomarkers of prenatal tobacco smoke exposure: The correlation between serum and meconium and their association with infant birth weight. Environmental Health,, 9, 53. doi: 10.1186/1476-069x-9-53

196

Brown, S.K. (2002). Volatile organic pollutants in new and established buildings in Melbourne, Australia. Indoor Air, 12(1), 55-63. Bruinen De Bruin, Y., Koistinen, K., Kephalopoulos, S., Geiss, O., Tirendi, S., & Kotzias, D. (2008). Characterisation of urban inhalation exposures to benzene, formaldehyde and acetaldehyde in the European Union. Environmental Science and Pollution Research International,, 15(5), 417-430. doi: 10.1007/s11356-0080013-4 Buhimschi, I.A., Buhimschi, C.S., Pupkin, M., & Weiner, C.P. (2003). Beneficial impact of term labor: Nonenzymatic antioxidant reserve in the human fetus. American Journal of Obstetrics and Gynecology, 189(1), 181-188. doi: 10.1067/mob.2003.357 Bukowski, R. (2004). Fetal growth potential and pregnancy outcome. Seminars in Perinatology,, 28(1), 51-58. doi: 10.1053/j.semperi.2003.12.003 Bukowski, R., Uchida, T., Smith, G.C., Malone, F.D., Ball, R.H., Nyberg, D.A., . . . D'Alton, M.E. (2008). Individualized norms of optimal fetal growth: Fetal growth potential. Obstetrics & Gynecology, 111(5), 1065-1076. doi: 10.1097/AOG.0b013e3181704e48 Burton, G.J., & Jauniaux, E. (2011). Oxidative stress. Best Practice & Research. Clinical Obstetrics & Gynaecology,, 25(3), 287-299. doi: 10.1016/j.bpobgyn.2010.10.016 Burton, G.J., Yung, H.W., Cindrova-Davies, T., & Charnock-Jones, D.S. (2009). Placental endoplasmic reticulum stress and oxidative stress in the pathophysiology of unexplained intrauterine growth restriction and early onset preeclampsia. Placenta, 30 Suppl A, S43-48. doi: 10.1016/j.placenta.2008.11.003

197

Byrne, C.D., & Phillips, D.I. (2000). Fetal origins of adult disease: Epidemiology and mechanisms. Journal of Clinical Pathology,, 53(11), 822-828. doi: 10.1136/jcp.53.11.822 Calbiotech. (2008). Cotinine ELISA. Retrieved April, 1, 2013, from http://www.calbiotech.com/images/stories/virtuemart/product/co096d-r1-rc2.pdf California Air Resources Board. (2005). Report to the California Legislature: Indoor Air Pollution in California. Retrieved March, 18, 2013, from http://www.arb.ca.gov/research/indoor/ab1173/rpt0705.pdf California Environmental Protection Agency. (2001). Standard operating procedure for teh determination of carbonyl compounds in ambient air. Indoor air quality guideline. Retrieved January, 6, 2014, from http://www.arb.ca.gov/aaqm/sop/sop_22.pdf California Environmental Protection Agency. (2004). Formaldehyde in the home. Indoor air quality guideline. Retrieved March, 18, 2013, from http://www.arb.ca.gov/research/indoor/formaldGL08-04.pdf California Environmental Protection Agency. (2008). Noncancer reference exposure levels. Technical Support Document. Retrieved March, 22, 2013, from http://www.oehha.ca.gov/air/hot_spots/2008/AppendixD1_final.pdf#page=128 Calkins, K., & Devaskar, S.U. (2011). Fetal origins of adult disease. Curr Probl Pediatr Adolesc Health Care, 41(6), 158-176. doi: 10.1016/j.cppeds.2011.01.001 Cayman Chemical Company. (2011). Formaldehdye assay kit. Retrieved March, 31, 2013, from https://www.caymanchem.com/pdfs/700380.pdf

198

Center for Disease Control. (2006). The health consequences of involuntary smoking: A report of General Surgeon. Atlanta, Georgia: Office on Smoking and Health Retrieved from http://www.surgeongeneral.gov/library/reports/secondhandsmoke/reportindex.html. Centers for Disease Control. (2008). Final report on formaldehyde levels in FEMAsupplied travel trailers, park models, and mobile homes: CDC. Chakravarty, E.F., Khanna, D., & Chung, L. (2008). Pregnancy outcomes in systemic sclerosis, primary pulmonary hypertension, and sickle cell disease. Obstetrics and Gynecology, 111(4), 927-934. doi: 10.1097/01.AOG.0000308710.86880.a6 Chelchowska, M., Ambroszkiewicz, J., Gajewska, J., Laskowska-Klita, T., & Leibschang, J. (2011). The effect of tobacco smoking during pregnancy on plasma oxidant and antioxidant status in mother and newborn. European Journal of Obstetrics, Gynecology, and Reproductive Biology,, 155(2), 132-136. doi: 10.1016/j.ejogrb.2010.12.006 Choi, H., Wang, L., Lin, X., Spengler, J.D., & Perera, F.P. (2012). Fetal window of vulnerability to airborne polycyclic aromatic hydrocarbons on proportional intrauterine growth restriction. PLoS One, 7(4), e35464. doi: 10.1371/journal.pone.0035464 Choudhary, A.K., Choudhary, A., Tiwari, S.C., & Dwivedi, R. (2013). Factors associated with low birth weight among newborns in an urban slum community in Bhopal. Indian Journal of Public Health,, 57(1), 20-23. doi: 10.4103/0019-557x.111362

199

Cogliano, V.J., Grosse, Y., Baan, R.A., Straif, K., Secretan, M.B., & El Ghissassi, F. (2005). Meeting report: Summary of IARC monographs on formaldehyde, 2butoxyethanol, and 1-tert-butoxy-2-propanol. Environmental Health Perspectives, 113(9), 1205-1208. doi: 10.1289/ehp.7542 Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Cracowski, J.L., Carpentier, P.H., Imbert, B., Cachot, S., Stanke-Labesque, F., Bessard, J., & Bessard, G. (2002). Increased urinary F2-isoprostanes in systemic sclerosis, but not in primary Raynaud's phenomenon: Effect of cold exposure. Arthritis Rheum, 46(5), 1319-1323. doi: 10.1002/art.10261 Cracowski, J.L., Durand, T., & Bessard, G. (2002). Isoprostanes as a biomarker of lipid peroxidation in humans: Physiology, pharmacology and clinical implications. Trends Pharmacol Sci, 23(8), 360-366. doi: 10.1161/01.CIR.0000077640.74263 Crowley, S.D. (2013). The cooperative roles of inflammation and oxidative stress in the pathogenesis of hypertension. Antioxidant and Redox Signaling. doi: 10.1089/ars.2013.5258 Cunningham, F., Leveno, K., Bloom, S., Hauth, J., Rouse, D., & Spong, C. (2010). Preterm birth Williams Obstetrics (23rd ed., pp. 842-858). New York: McGrawHill Companies Inc. Cunningham, F.G., Cox, S.M., Harstad, T.W., Mason, R.A., & Pritchard, J.A. (1990). Chronic renal disease and pregnancy outcome. American Journal of Obstetrics and Gynecology, 163(2), 453-459.

200

Daly, M.C., Duncan, G.J., McDonough, P., & Williams, D.R. (2002). Optimal indicators of socioeconomic status for health research. American Journal of Public Health,, 92(7), 1151-1157. Dannemiller, K.C., Murphy, J.S., Dixon, S.L., Pennell, K.G., Suuberg, E.M., Jacobs, D.E., & Sandel, M. (2013). Formaldehyde concentrations in household air of asthma patients determined using colorimetric detector tubes. Indoor Air, 23(4), 285-294. doi: 10.1111/ina.12024 Dassonville, C., Demattei, C., Laurent, A.M., Le Moullec, Y., Seta, N., & Momas, I. (2009). Assessment and predictor determination of indoor aldehyde levels in Paris newborn babies' homes. Indoor Air, 19(4), 314-323. doi: INA594 [pii]10.1111/j.1600-0668.2009.00594.x David, C., Tagliavini, G., Pilu, G., Rudenholz, A., & Bovicelli, L. (1996). Receiveroperator characteristic curves for the ultrasonographic prediction of small-forgestational-age fetuses in low-risk pregnancies. American Journal of Obstetrics and Gynecology, 174(3), 1037-1042. doi: http://dx.doi.org/10.1016/S00029378(96)70347-2 Davis, K.A., Crow, J.A., Chambers, H.W., Meek, E.C., & Chambers, J.E. (2009). Racial differences in paraoxonase-1 (PON1): A factor in the health of southerners? Environmental Health Perspectives, 117(8), 1226-1231. doi: 10.1289/ehp.0900569 De Reu, P.A., Smits, L.J., Oosterbaan, H.P., & Nijhuis, J.G. (2008). Value of a single early third trimester fetal biometry for the prediction of birth weight deviations in

201

a low risk population. Journal of Perinatal Medicine,, 36(4), 324-329. doi: 10.1515/JPM.2008.057 Dejmek, J., Solansk y, I., Podrazilova, K., & Sram, R.J. (2002). The exposure of nonsmoking and smoking mothers to environmental tobacco smoke during different gestational phases and fetal growth. Environmental Health Perspectives,, 110(6), 601-606. Dempsey, D., Jacob, P., 3rd, & Benowitz, N.L. (2002). Accelerated metabolism of nicotine and cotinine in pregnant smokers. The Journal of Pharmacology and Experimental Therapeutics,, 301(2), 594-598. doi: 10.1124/jpet.301.2.594 Devasagayam, T.P., Tilak, J.C., Boloor, K.K., Sane, K.S., Ghaskadbi, S.S., & Lele, R.D. (2004). Free radicals and antioxidants in human health: Current status and future prospects. The Journal of the Association of Physicians of India,, 52, 794-804. Drager, L.F., Andrade, L., Barros de Toledo, J.F., Laurindo, F.R., Machado Cesar, L.A., & Seguro, A.C. (2004). Renal effects of N-acetylcysteine in patients at risk for contrast nephropathy: Decrease in oxidant stress-mediated renal tubular injury. Nephrology, Dialysis, Transplantation,, 19(7), 1803-1807. doi: 10.1093/ndt/gfh261 Droste, S. (1992). Fetal growth in aneuploid conditions. Clinical Obstetrics and Gynecology, 35(1), 119-125. Duong, A., Steinmaus, C., McHale, C.M., Vaughan, C.P., & Zhang, L. (2011). Reproductive and developmental toxicity of formaldehyde: A systematic review. Mutation Research, 728(3), 118-138. doi: 10.1016/j.mrrev.2011.07.003 Ecos. (2006). Formaldehyde levels high in wood panel products, Ecos 5.

202

Egley, C.C., Seeds, J.W., & Cefalo, R.C. (1986). Femur length versus biparietal diameter for estimating gestational age in the third trimester. American Journal of Perinatology,, 3(2), 77-79. doi: 10.1055/s-2007-999837 Eisner, V., Brazie, J.V., Pratt, M.W., & Hexter, A.C. (1979). The risk of low birthweight. American Journal of Public Health,, 69(9), 887-893. doi: 10.2105/AJPH.69.9.887 England, L.J., Grauman, A., Qian, C., Wilkins, D.G., Schisterman, E.F., Yu, K.F., & Levine, R.J. (2007). Misclassification of maternal smoking status and its effects on an epidemiologic study of pregnancy outcomes. Nicotine & Tobacco Research,, 9(10), 1005-1013. doi: 10.1080/14622200701491255 EPA. (1990). The Clean Air Act Amendments of 1990 List of Hazardous Air Pollutants. Technology Transfer Network- Air Toxics Web Site. Retrieved March, 20, 2013, from http://www.epa.gov/ttn/atw/orig189.html EPA. (2008). EPA to study formaldehyde. Environmental building news, 17. EPA. (2011). Formaldehyde. An introduction to indoor air quality (IAQ). Retrieved March, 18, 2013, from http://www.epa.gov/iaq/formalde.html EPA. (2012). Particulate matter (PM). Six common pollutants. Retrieved March, 10, 2013, from http://www.epa.gov/pm/ Epi Info. (2013). Epi Info Tm 7. 1. 1. 14. Atlanta, GA: CDC. Retrieved from http://wwwn.cdc.gov/epiinfo/7/ Epplein, M., Franke, A.A., Cooney, R.V., Morris, J.S., Wilkens, L.R., Goodman, M.T., . . . Le Marchand, L. (2009). Association of plasma micronutrient levels and urinary isoprostane with risk of lung cancer: The multiethnic cohort study. Cancer

203

epidemiology, biomarkers & prevention,, 18(7), 1962-1970. doi: 10.1158/10559965.epi-09-0003 Erichsen, H.C., Engel, S.A., Eck, P.K., Welch, R., Yeager, M., Levine, M., . . . Chanock, S.J. (2006). Genetic variation in the sodium-dependent vitamin C transporters, SLC23A1, and SLC23A2 and risk for preterm delivery. American Journal of Epidemiology,, 163(3), 245-254. doi: kwj035 [pii]10.1093/aje/kwj035 Ezimokhai, M., Davison, J.M., Philips, P.R., & Dunlop, W. (1981). Non-postural serial changes in renal function during the third trimester of normal human pregnancy. British Journal of Obstetrics and Gynaecology,, 88(5), 465-471. Fall, C.H., Vijayakumar, M., Barker, D.J., Osmond, C., & Duggleby, S. (1995). Weight in infancy and prevalence of coronary heart disease in adult life. British Medical Journal,, 310(6971), 17-19. Fan, Z., Lioy, P., Weschler, C., Fiedler, N., Kipen, H., & Zhang, J. (2003). Ozoneinitiated reactions with mixtures of volatile organic compounds under simulated indoor conditions. Environmental Science & Technology,, 37(9), 1811-1821. Fescina, R.H., De Mucio, B., Martínez, G., Alemán, A., Sosa, C., Mainero, L., & Rubino, M. (2011). Monitoring fetal growth- Self instruction. Pan American Health Organization Retrieved from http://www.paho.org/clap/index.php?option=com_content&view=article&id=164 &Itemid=234. Fiscella, K. (2005). Race, genes and preterm delivery. Journal of the National Medical Association,, 97(11), 1516-1526.

204

Frankel, S., Elwood, P., Smith, G.D., Frankel, S., Sweetnam, P., & Yarnell, J. (1996). Birthweight, body-mass index in middle age, and incident coronary heart disease. The Lancet, 348(9040), 1478-1480. doi: http://dx.doi.org/10.1016/S01406736(96)03482-4 Fraser, R., Sr., & Cresswell, J., Sr. (1997). What should obstetricians be doing about the Barker hypothesis? British Journal of Obstetrics and Gynaecology,, 104(6), 645647. Fuentes-Afflick, E., & Hessol, N.A. (2000). Interpregnancy interval and the risk of premature infants. Obstetrics and Gynecology, 95(3), 383-390. García-Estañ, J., Ortiz, M.C., & Lee, S.S. (2002). Nitric oxide and renal and cardiac dysfunction in cirrhosis. Clinical Science,, 102(2), 213-222. Gardosi, J., & Geirsson, R.T. (1998). Routine ultrasound is the method of choice for dating pregnancy. British Journal of Obstetrics and Gynaecology,, 105(9), 933936. Gaudineau, A. (2013). [Prevalence, risk factors, maternal and fetal morbidity and mortality of intrauterine growth restriction and small-for-gestational age.]. Journal de Gynecologie, Obstetrique et Biologie de la Reproduction,. doi: 10.1016/j.jgyn.2013.09.013 Geirsson, R.T. (1991). Ultrasound instead of last menstrual period as the basis of gestational age assignment. Ultrasound in Obstetrics & Gynecology,, 1(3), 212219. doi: 10.1046/j.1469-0705.1991.01030212.x Gill, J.K., Franke, A.A., Steven Morris, J., Cooney, R.V., Wilkens, L.R., Le Marchand, L., . . . Kolonel, L.N. (2009). Association of selenium, tocopherols, carotenoids,

205

retinol, and 15-isoprostane F(2t) in serum or urine with prostate cancer risk: The multiethnic cohort. Cancer Causes Control, 20(7), 1161-1171. doi: 10.1007/s10552-009-9304-4 Godish, T. (1989). Formaldehyde exposures from tobacco smoke: A review. American Journal of Public Health,, 79(8), 1044-1045. Gottschling, L.M., Beaulieu, H.J., & Melvin, W.W. (1984). Monitoring of formic acid in urine of humans exposed to low levels of formaldehyde. American Industrial Hygiene Association Journal,, 45(1), 19-23. doi: 10.1080/15298668491399299 Grazulevicience, R., Dulskiene, V., & Vencloviene, J. (1998). Formaldehyde exposure and low birth weight incidence. Journal of Occupational Health, 40, 61-67. Grazuleviciene, R., Danileviciute, A., Nadisauskiene, R., & Vencloviene, J. (2009). Maternal smoking, GSTM1 and GSTT1 polymorphism and susceptibility to adverse pregnancy outcomes. International Journal of Environmental Research and Public Health,, 6(3), 1282-1297. doi: 10.3390/ijerph6031282 Green, R.M., Gally, F., Keeney, J.G., Alper, S., Gao, B., Han, M., . . . Chu, H.W. (2009). Impact of cigarette smoke exposure on innate immunity: A Caenorhabditis elegans model. PLoS One, 4(8), e6860. doi: 10.1371/journal.pone.0006860 Guellec, I., Lapillonne, A., Renolleau, S., Charlaluk, M.L., Roze, J.C., Marret, S., . . . Ancel, P.Y. (2011). Neurologic outcomes at school age in very preterm infants born with severe or mild growth restriction. Pediatrics, 127(4), e883-891. doi: 10.1542/peds.2010-2442 Gulec, M., Songur, A., Sahin, S., Ozen, O.A., Sarsilmaz, M., & Akyol, O. (2006). Antioxidant enzyme activities and lipid peroxidation products in heart tissue of

206

subacute and subchronic formaldehyde-exposed rats: A preliminary study. Toxicology and Industrial Health, 22(3), 117-124. doi: 10.1191/0748233706th248oa. Gulmezoglu, M., de Onis, M., & Villar, J. (1997). Effectiveness of interventions to prevent or treat impaired fetal growth. Obstetrical and Gynecological Survey, 52(2), 139-149. Gurel, A., Coskun, O., Armutcu, F., Kanter, M., & Ozen, O.A. (2005). Vitamin E against oxidative damage caused by formaldehyde in frontal cortex and hippocampus: Biochemical and histological studies. Journal of Chemical Neuroanatomy,, 29(3), 173-178. doi: 10.1016/j.jchemneu.2005.01.001 Gustafson, P., Barregard, L., Lindahl, R., & Sallsten, G. (2005). Formaldehyde levels in Sweden: Personal exposure, indoor, and outdoor concentrations. Journal of Exposure Analysis and Environmental Epidemiology,, 15(3), 252-260. doi: 10.1038/sj.jea.7500399 Guvendag Guven, E.S., Karcaaltincaba, D., Kandemir, O., Kiykac, S., & Mentese, A. (2013). Cord blood oxidative stress markers correlate with umbilical artery pulsatility in fetal growth restriction. The Journal of Maternal-Fetal & Neonatal Medicine,, 26(6), 576-580. doi: 10.3109/14767058.2012.745497 Hadlock, F.P., Deter, R.L., Harrist, R.B., & Park, S.K. (1984). Estimating fetal age: Computer-assisted analysis of multiple fetal growth parameters. Radiology, 152(2), 497-501. Hadlock, F.P., Harrist, R.B., & Martinez-Poyer, J. (1991). In utero analysis of fetal growth: A sonographic weight standard. Radiology, 181(1), 129-133.

207

Halliwell, B. (1994). Free radicals, antioxidants, and human disease: Curiosity, cause, or consequence? Lancet, 344(8924), 721-724. Halliwell, B., & Gutteridge, J.M.C. (2007). Free radicals in biology and medicine, (4nd ed.). New York: Oxford University Press. Hansen, J.M., Contreras, K.M., & Harris, C. (2005). Methanol, formaldehyde, and sodium formate exposure in rat and mouse conceptuses: A potential role of the visceral yolk sac in embryotoxicity. Birth Defects Research. Part A, Clinical and Molecular Teratology,, 73(2), 72-82. doi: 10.1002/bdra.20094 Health Canada. (2012). Formaldehyde. Environmental and work place health. Retrieved October, 12, 2013, from http://www.hc-sc.gc.ca/ewhsemt/air/in/poll/construction/formaldehyde-eng.php HealthyPeople.gov. (2011). Maternal, Infant, and Child Health. Retrieved October, 4, 2011, from http://www.healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topi cId=26 Heck, A., Chin, T.Y., & Schmitz, M.C. (1983). Distribution of [14 C] formaldehyde in rats after inhalation exposure. In J. E. Gibson (Ed.), Formladehyde Toxicity. Washington, D.C.: HemispherePublishign Corporation. Heinrich, R., & Angerer, J. (1982). Occupational chronic exposure to organic solvents. X. Biological monitoring parameters for methanol exposure. International Archives of Occupational and Environmental Health, 50(4), 341-349. Hejtmancik, J.F., Black, S., Harris, S., Ward, P.A., Callaway, C., Ledbetter, D., . . . Pollack, M.S. (1992). Congenital 21-hydroxylase deficiency as a new deletion

208

mutation. Detection in a proband during subsequent prenatal diagnosis by HLA typing and DNA analysis. Human Immunology, 35(4), 246-252. Hemminki, K., Kyyronen, P., & Lindbohm, M.L. (1985). Spontaneous abortions and malformations in the offspring of nurses exposed to anaesthetic gases, cytostatic drugs, and other potential hazards in hospitals, based on registered information of outcome. Journal of Epidemiology and Community Health,, 39(2), 141-147. Hempstock, J., Jauniaux, E., Greenwold, N., & Burton, G.J. (2003). The contribution of placental oxidative stress to early pregnancy failure. Human Pathology,, 34(12), 1265-1275. Henrichs, J., Schenk, J.J., Schmidt, H.G., Arends, L.R., Steegers, E.A., Hofman, A., . . . Tiemeier, H. (2009). Fetal size in mid- and late pregnancy is related to infant alertness: The generation R study. Developmental Psychobiology,, 51(2), 119130. doi: 10.1002/dev.20351 Heroux, M.E., Clark, N., Van Ryswyk, K., Mallick, R., Gilbert, N.L., Harrison, I., . . . Wheeler, A.J. (2010). Predictors of indoor air concentrations in smoking and nonsmoking residences. International Journal of Environmental Research and Public Health,, 7(8), 3080-3099. doi: 10.3390/ijerph7083080 Higgins, S.T., Heil, S.H., Badger, G.J., Mongeon, J.A., Solomon, L.J., McHale, L., & Bernstein, I.M. (2007). Biochemical verification of smoking status in pregnant and recently postpartum women. Experimental and Clinical Psychopharmacology,, 15(1), 58-66. doi: 10.1037/1064-1297.15.1.58 Hisamitsu, M., Okamoto, Y., Chazono, H., Yonekura, S., Sakurai, D., Horiguchi, S., . . . Mori, C. (2011). The influence of environmental exposure to formaldehyde in

209

nasal mucosa of medical students during cadaver dissection. Allergology International,, 60(3), 373-379. doi: 10.2332/allergolint.10-OA-0210 Hodgson, A.T., Beal, D., & McIlvaine, J.E. (2002). Sources of formaldehyde, other aldehydes and terpenes in a new manufactured house. Indoor Air, 12(4), 235-242. Hodgson, A.T., Rudd, A.F., Beal, D., & Chandra, S. (2000). Volatile organic compound concentrations and emission rates in new manufactured and site-built houses. Indoor Air, 10(3), 178-192. Hodgson, A.T., Wooley, J.D., & Daisey, J.M. (1993). Emissions of volatile organic compounds from new carpets measured in a large-scale environmental chamber. Air Waste, 43(3), 316-324. Hoffman, C.S., Messer, L.C., Mendola, P., Savitz, D.A., Herring, A.H., & Hartmann, K.E. (2008). Comparison of gestational age at birth based on last menstrual period and ultrasound during the first trimester. Paediatric and Perinatal Epidemiology,, 22(6), 587-596. doi: 10.1111/j.1365-3016.2008.00965.x Hoffmann, D., Hoffmann, I., & El-Bayoumy, K. (2001). The less harmful cigarette: A controversial issue. a tribute to Ernst L. Wynder. Chemical Research in Toxicology, 14(7), 767-790. doi: 10.1021/tx000260u Horacio Fescina, R., Martell, M., Martinez, G., Lastra, L., & Schwarcz, R. (1987). Small for dates: Evaluation of different diagnostic methods. Acta Obstetricia et Gynecologica Scandinavica,, 66(3), 221-226. doi: 10.3109/00016348709020751 Horta, B.L., Victora, C.G., Menezes, A.M., Halpern, R., & Barros, F.C. (1997). Low birthweight, preterm births and intrauterine growth retardation in relation to maternal smoking. Paediatric Perinatal Epidemiology,, 11(2), 140-151.

210

Huang, Q.T., Zhang, M., Zhong, M., Yu, Y.H., Liang, W.Z., Hang, L.L., . . . Wang, Z.J. (2013). Advanced glycation end products as an upstream molecule triggers ROSinduced sFlt-1 production in extravillous trophoblasts: A novel bridge between oxidative stress and preeclampsia. Placenta. doi: 10.1016/j.placenta.2013.09.017 Hukkanen, J., Jacob, P., 3rd, & Benowitz, N.L. (2005). Metabolism and disposition kinetics of nicotine. Pharmacological Review,, 57(1), 79-115. doi: 10.1124/pr.57.1.3 Hun, D.E., Corsi, R.L., Morandi, M.T., & Siegel, J.A. (2010). Formaldehyde in residences: Long-term indoor concentrations and influencing factors. Indoor Air, 20(3), 196-203. doi: 10.1111/j.1600-0668.2010.00644.x IARC. (2006). Formaldehyde, 2-butoxyethanol and 1-tert-butoxypropan-2-ol. IARC monographs on the evaluation of carcinogenic risks to humans (Vol. 88). Geneva, Switzerland: World HealthOrganization. Im, H., Oh, E., Mun, J., Khim, J.Y., Lee, E., Kang, H.S., . . . Sul, D. (2006). Evaluation of toxicological monitoring markers using proteomic analysis in rats exposed to formaldehyde. The Journal of Proteome Research, , 5(6), 1354-1366. doi: 10.1021/pr050437b Iniguez, C., Ballester, F., Amoros, R., Murcia, M., Plana, A., & Rebagliato, M. (2012). Active and passive smoking during pregnancy and ultrasound measures of fetal growth in a cohort of pregnant women. Journal of Epidemiol and Community Health,, 66(6), 563-570. doi: 10.1136/jech.2010.116756 Iniguez, C., Ballester, F., Costa, O., Murcia, M., Souto, A., Santa-Marina, L., . . . Investigators, I.S. (2013). Maternal smoking during pregnancy and fetal biometry:

211

The INMA Mother and Child Cohort Study. American Journal of Epidemiology,, 178(7), 1067-1075. doi: 10.1093/aje/kwt085 International Agency for Research on Cancer. (2006). Formaldehyde, 2-butoxyethanol and 1-tert-butoxypropan-2-ol IARC monographs on the evaluation of carcinogenic risks to humans (Vol. 18). Geneva, Switzerland: World HealthOrganization. Ioannou, C., Talbot, K., Ohuma, E., Sarris, I., Villar, J., Conde-Agudelo, A., & Papageorghiou, A.T. (2012). Systematic review of methodology used in ultrasound studies aimed at creating charts of fetal size. BJOG, 119(12), 14251439. doi: 10.1111/j.1471-0528.2012.03451.x Jaddoe, V.W., Verburg, B.O., de Ridder, M.A., Hofman, A., Mackenbach, J.P., Moll, H.A., . . . Witteman, J.C. (2007). Maternal smoking and fetal growth characteristics in different periods of pregnancy: The generation R study. American Journal of Epidemiology, 165(10), 1207-1215. doi: 10.1093/aje/kwm014 Jauniaux, E., Watson, A.L., Hempstock, J., Bao, Y.P., Skepper, J.N., & Burton, G.J. (2000). Onset of maternal arterial blood flow and placental oxidative stress. A possible factor in human early pregnancy failure. The American Journal of Pathology,, 157(6), 2111-2122. doi: 10.1016/s0002-9440(10)64849-3 Jeffcoat, A.R., Chasalow, F., Feldman, D.B., & Marr, H. (1983). Dispostition of [14C] formaldehyde after topical exposure to rats, guinea pigs and monkeys. In J. E. Gibson (Ed.), Formaldehyde Toxicity (pp. 38-50). Washington, D. C.: Hemisphese.

212

Jenkins, J., McCall, E., Gardner, E., Casson, K., & Dolk, H. (2009). Socioeconomic inequalities in neonatal intensive care admission rates. Archives of Disease in Childhood. Fetal and Neonatal Edition,, 94(6), F423-428. doi: 10.1136/adc.2008.146464 Jenkins, R.A., Guerin, M.R., & Tiomkins, B.A. (2000). The chemistry of environmental tobacco smoke: Composition and measurement (2nd ed.). Boca Raton: CRC Press LLC. Jhun, H.J., Seo, H.G., Lee, D.H., Sung, M.W., Kang, Y.D., Syn, H.C., & Jun, J.K. (2010). Self-reported smoking and urinary cotinine levels among pregnant women in Korea and factors associated with smoking during pregnancy. Journal of Korean Medical Science,, 25(5), 752-757. doi: 10.3346/jkms.2010.25.5.752 Johansson, E.B., & Tjalve, H. (1978). The distribution of [14C]dimethylnitrosamine in mice. Autoradiographic studies in mice with inhibited and noninhibited dimethylnitrosamine metabolism and a comparison with the distribution of [14C]formaldehyde. Toxicology and Applied Pharmacology,, 45(2), 565-575. Johnson, E.O., & Breslau, N. (2000). Increased risk of learning disabilities in low birth weight boys at age 11 years. Biological Psychiatry,, 47(6), 490-500. Jones, D.P. (2006). Redefining oxidative stress. Antioxid Redox Signal, 8(9-10), 18651879. doi: 10.1089/ars.2006.8.1865 Jones, D.P. (2008). Radical-free biology of oxidative stress. American Journal of Physiology. Cell Physiology,, 295(4), C849-868. doi: 10.1152/ajpcell.00283.2008 Jurvelin, J., Vartiainen, M., & Jantunen, M. (2001). Personal exposure levels and microenvironmental concentrations of formaldehyde and acetaldehyde in the

213

Helsinki metropolitan area, Finland. Journal of the Air & Waste Management Association (1995), 51(1), 17-24. doi: 10.1080/10473289.2001.10464251 Kajantie, E., Osmond, C., Barker, D.J., Forsen, T., Phillips, D.I., & Eriksson, J.G. (2005). Size at birth as a predictor of mortality in adulthood: A follow-up of 350 000 person-years. International Journal of Epidemiology,, 34(3), 655-663. doi: 10.1093/ije/dyi048 Kamath, U., Rao, G., Kamath, S.U., & Rai, L. (2006). Maternal and fetal indicators of oxidative stress during intrauterine growth retardation (IUGR). Indian Journal of Clinical Biochemistry, 21(1), 111-115. Kannan, S., Misra, D.P., Dvonch, J.T., & Krishnakumar, A. (2006). Exposures to airborne particulate matter and adverse perinatal outcomes: A biologically plausible mechanistic framework for exploring potential effect modification by nutrition. Environmental Health Perspectives, 114(11), 1636-1642. doi: 10.1289/ehp.9081 Karowicz-Bilinska, A., Suzin, J., & Sieroszewski, P. (2002). Evaluation of oxidative stress indices during treatment in pregnant women with intrauterine growth retardation. Medical Science Monitor,, 8(3), CR211-216. Katakura, Y., Kishi, R., Okui, T., Ikeda, T., & Miyake, H. (1993). Distribution of radioactivity from 14C-formaldehyde in pregnant mice and their fetuses. British Journal of Industrial Medicine,, 50(2), 176-182. Kato, S., Burke, P.J., Koch, T.H., & Bierbaum, V.M. (2001). Formaldehyde in human cancer cells: Detection by preconcentration-chemical ionization mass spectrometry. Analytical Chemistry,, 73(13), 2992-2997. doi: 10.1021/ac001498q

214

Keefer, L.K., Streeter, A.J., Leung, L.Y., Perry, W.C., Hu, H.S., & Baillie, T.A. (1987). Pharmacokinetic and deuterium isotope effect studies on the metabolism of formaldehyde and formate to carbon dioxide in rats in vivo. Drug Metab Dispos, 15(3), 300-304. Kensler, C.J., & Battista, S.P. (1963). Components of Cigarette Smoke with CiliaryDepressant Activity. Their Selective Removal by Filters Containing Activated Charcoal Granules. The New England Journal of Medicine,, 269, 1161-1166. doi: 10.1056/NEJM196311282692202 Khoury, M.J., Erickson, J.D., Cordero, J.F., & McCarthy, B.J. (1988). Congenital malformations and intrauterine growth retardation: A population study. Pediatrics, 82(1), 83-90. Kim, J., Kim, S., Lee, K., Yoon, D., Lee, J., & Ju, D. (2013). Indoor aldehydes concentration and emission rate of formaldehyde in libraries and private reading rooms. Atmospheric Environment, 71, 1-6. doi: 10.1016/j.atmosenv.2013.01.059 Kim, Y.J., Hong, Y.C., Lee, K.H., Park, H.J., Park, E.A., Moon, H.S., & Ha, E.H. (2005). Oxidative stress in pregnant women and birth weight reduction. Reproductive Toxicology, 19(4), 487-492. doi: 10.1016/j.reprotox.2004.10.007 Kramer, M.S. (2013). The epidemiology of low birthweight. Nestle Nutrition Institute Workshop Series,, 74, 1-10. doi: 10.1159/000348382 Kramer, M.S., McLean, F.H., Boyd, M.E., & Usher, R.H. (1988). The validity of gestational age estimation by menstrual dating in term, preterm, and postterm gestations. The journal of the American Medical Association,, 260(22), 33063308. doi: 10.1001/jama.1988.03410220090034

215

Krampl, E., Lees, C., Bland, J.M., Espinoza Dorado, J., Moscoso, G., & Campbell, S. (2000). Fetal biometry at 4300 m compared to sea level in Peru. Ultrasound in Obstetrics & Gynecology,, 16(1), 9-18. doi: 10.1046/j.1469-0705.2000.00156.x Kum, C., Kiral, F., Sekkin, S., Seyrek, K., & Boyacioglu, M. (2007). Effects of xylene and formaldehyde inhalations on oxidative stress in adult and developing rats livers. Experimental Animals, 56(1), 35-42. Kurmanavicius, J., Wright, E.M., Royston, P., Wisser, J., Huch, R., Huch, A., & Zimmermann, R. (1999). Fetal ultrasound biometry: 1. Head reference values. British Journal of Obstetrics and Gynaecology, 106(2), 126-135. doi: 10.1111/j.1471-0528.1999.tb08212.x Lai, F.M., & Yeo, G.S. (1995). Reference charts of foetal biometry in Asians. Singapore Medical Journal, 36(6), 628-636. Lamorena, R.B., & Lee, W. (2008). Influence of ozone concentration and temperature on ultra-fine particle and gaseous volatile organic compound formations generated during the ozone-initiated reactions with emitted terpenes from a car air freshener. Journal of Hazardous Materials,, 158(2-3), 471-477. doi: 10.1016/j.jhazmat.2008.01.095 Lampl, M., Gotsch, F., Kusanovic, J.P., Gomez, R., Nien, J.K., Frongillo, E.A., & Romero, R. (2010). Sex differences in fetal growth responses to maternal height and weight. American journal of human biology, 22(4), 431-443. doi: 10.1002/ajhb.21014 Lampl, M., & Jeanty, P. (2004). Exposure to maternal diabetes is associated with altered fetal growth patterns: A hypothesis regarding metabolic allocation to growth

216

under hyperglycemic-hypoxemic conditions. American Journal of Human Biology,, 16(3), 237-263. doi: 10.1002/ajhb.20015 Larose, J., Julien, P., & Bilodeau, J.F. (2013). Analysis of F2-isoprostanes in plasma of pregnant women by HPLC-MS/MS using a column packed with core-shell particles. Journal of Lipid Research,, 54(5), 1505-1511. doi: 10.1194/jlr.D034553 Lau, C., Fiedler, H., Hutzinger, O., Schwind, K.H., & Hosseinpour, J. (1997). Levels of selected organic compounds in materials for candle production and human exposure to candle emissions. Chemosphere, 34(5-7), 1623-1630. Lazenby, V., Hinwood, A., Callan, A., & Franklin, P. (2012). Formaldehyde personal exposure measurements and time weighted exposure estimates in children. Chemosphere, 88(8), 966-973. doi: 10.1016/j.chemosphere.2012.03.029 Lekea-Karanika, V., Tzoumaka-Bakoula, C., & Matsaniotis, N.S. (1999). Sociodemographic determinants of low birthweight in Greece: A population study. Paediatric and Perinatal Epidemiology,, 13(1), 65-77. Levin, K.A. (2013). Study design III: Cross-sectional studies. Evidence-Based Dental,, 7(1), 24-25. Liden, J., Ek, A., Palmberg, L., Okret, S., & Larsson, K. (2003). Organic dust activates NF-kappaB in lung epithelial cells. Respiratory Medicine,, 97(8), 882-892. Lin, C.C., & Santolaya-Forgas, J. (1998). Current concepts of fetal growth restriction: Part I. Causes, classification, and pathophysiology. Obstetrics and Gynecology, 92(6), 1044-1055.

217

Lindqvist, R., Lendahls, L., Tollbom, O., Aberg, H., & Hakansson, A. (2002). Smoking during pregnancy: Comparison of self-reports and cotinine levels in 496 women. Acta Obstetricia et Gynecologica Scandinavica,, 81(3), 240-244. Lindstrom, A.B., Proffitt, D., & Fortune, C.R. (1995). Effects of modified residential construction on indoor air quality. Indoor Air, 5(4), 258-269. doi: 10.1111/j.16000668.1995.00005.x Lioy, P.J., Fan, Z., Zhang, J., Georgopoulos, P., Wang, S.W., Ohman-Strickland, P., . . . Neal, J. (2011). Personal and ambient exposures to air toxics in Camden, New Jersey. Research report (Health Effects Institute)(160), 3-127. Liu, X., Mason, M., Krebs, K., & Sparks, L. (2004). Full-scale chamber investigation and simulation of air freshener emissions in the presence of ozone. Environmental Science & Technology,, 38(10), 2802-2812. Liu, Y., Jin, H., Yang, K.Y., Kim, C.Y., Baker, B., & Zhang, S. (2003). Interaction between two mitogen-activated protein kinases during tobacco defense signaling. The Plant Journal for Cell and Molecular Biology,, 34(2), 149-160. Longo, S., Bollani, L., Decembrino, L., Di Comite, A., Angelini, M., & Stronati, M. (2013). Short-term and long-term sequelae in intrauterine growth retardation (IUGR). The Journal of Maternal-Fetal & Neonatal Medicine,, 26(3), 222-225. doi: 10.3109/14767058.2012.715006 Love, C., David, R.J., Rankin, K.M., & Collins, J.W., Jr. (2010). Exploring weathering: Effects of lifelong economic environment and maternal age on low birth weight, small for gestational age, and preterm birth in African-American and white

218

women. American Journal of Epidemiology, 172(2), 127-134. doi: kwq109 [pii]10.1093/aje/kwq109 Lovreglio, P., Carrus, A., Iavicoli, S., Drago, I., Persechino, B., & Soleo, L. (2009). Indoor formaldehyde and acetaldehyde levels in the province of Bari, South Italy, and estimated health risk. Journal of Environmental Monitoring, , 11(5), 955-961. doi: 10.1039/b819843h Lowe, F.J., Luettich, K., & Gregg, E.O. (2013). Lung cancer biomarkers for the assessment of modified risk tobacco products: An oxidative stress perspective. Biomarkers, 18(3), 183-195. doi: 10.3109/1354750x.2013.777116 Luke, B. (1994). Maternal-fetal nutrition. Clinical Obstetrics and Gynecology, 37(1), 93109. Lutterodt, M.C., Sorensen, K.P., Larsen, K.B., Skouby, S.O., Andersen, C.Y., & Byskov, A.G. (2009). The number of oogonia and somatic cells in the human female embryo and fetus in relation to whether or not exposed to maternal cigarette smoking. Human Reproduction,, 24(10), 2558-2566. doi: 10.1093/humrep/dep226 Lynch, C.D., & Zhang, J. (2007). The research implications of the selection of a gestational age estimation method. Paediatric and Perinatal Epidemiology,, 21 Suppl 2, 86-96. doi: 10.1111/j.1365-3016.2007.00865.x MacKinnon, D.P. (1994). Analysis of mediating variables in prevention and intervention research. NIDA Research Monograph,, 139, 127-153. MacKinnon, D.P. (2008). Introduction to statistical mediation analysis. New York: Erlbaum.

219

Maddalena, R., Russell, M., Sullivan, D.P., & Apte, M.G. (2009). Formaldehyde and other volatile organic chemical emissions in four FEMA temporary housing units. Environmental Science & Technology,, 43(15), 5626-5632. Madden, D. (2013). The relationship between low birth weight and socioeconomic status in Ireland. Journal of biosocial science,, 1-18. doi: 10.1017/s0021932013000187 Magras, I.N. (1996). Formaldehyde vapour effects in chicken embryo. Anatomia, Histologia, Embryologia,, 25(3), 197-200. Makri, A., Goveia, M., Balbus, J., & Parkin, R. (2004). Children's susceptibility to chemicals: A review by developmental stage. Journal of Toxicology and Environmental Health. Part B, Critical Reviews, 7(6), 417-435. doi: 10.1080/10937400490512465 Makris, S.L., Thompson, C.M., Euling, S.Y., Selevan, S.G., & Sonawane, B. (2008). A lifestage-specific approach to hazard and dose-response characterization for children's health risk assessment. Birth Defects Research. Part B, Developmental and Reproductive Toxicology Teratology Society,, 83(6), 530-546. doi: 10.1002/bdrb.20176 Mamsen, L.S., Lutterodt, M.C., Andersen, E.W., Skouby, S.O., Sorensen, K.P., Andersen, C.Y., & Byskov, A.G. (2010). Cigarette smoking during early pregnancy reduces the number of embryonic germ and somatic cells. Human Reproduction,, 25(11), 2755-2761. doi: 10.1093/humrep/deq215 Maroziene, L., & Grazuleviciene, R. (2002). Maternal exposure to low-level air pollution and pregnancy outcomes: A population-based study. Environmental Health, 1(1), 6. doi: 10.1186/1476-069X-1-6

220

Martin, J.A., Hamilton, B.E., Sutton, P.D., Ventura, S.J., Menacker, F., & Munson, M.L. (2003). Births: final data for 2002. National vital statistics reports, 52(10), 1-113. Martin, W.J. (1990). A teratology study of inhaled formaldehyde in the rat. Reproductive Toxicology, 4(3), 237-239. Maruo, Y.Y., Yamada, T., Nakamura, J., Izumi, K., & Uchiyama, M. (2010). Formaldehyde measurements in residential indoor air using a developed sensor element in the Kanto area of Japan. Indoor Air, 20(6), 486-493. doi: 10.1111/j.1600-0668.2010.00670.x Mastrobattista, J.M., & Gomez-Lobo, V. (2008). Pregnancy after solid organ transplantation. Obstetrics and Gynecology, 112(4), 919-932. doi: 10.1097/AOG.0b013e318187d00c McCormick, M.C. (1985). The contribution of low birth weight to infant mortality and childhood morbidity. New England Journal of Medicine, 312(2), 82-90. doi: 10.1056/NEJM198501103120204 McGwin, G., Lienert, J., & Kennedy, J.I. (2010). Formaldehyde exposure and asthma in children: A systematic review. Environmental Health Perspectives, 118(3), 313317. doi: 10.1289/ehp.0901143 McMartin, K.E., Martin-Amat, G., Noker, P.E., & Tephly, T.R. (1979). Lack of a role for formaldehyde in methanol poisoning in the monkey. Biochemical Pharmacology, 28(5), 645-649. Melamed, N., Meizner, I., Mashiach, R., Wiznitzer, A., Glezerman, M., & Yogev, Y. (2013). Fetal sex and intrauterine growth patterns. Journal of Ultrasound in Medicine,, 32(1), 35-43.

221

Mert, I., Oruc, A.S., Yuksel, S., Cakar, E.S., Buyukkagnici, U., Karaer, A., & Danisman, N. (2012). Role of oxidative stress in preeclampsia and intrauterine growth restriction. The Journal of Obstetrics and Gynaecology Research,, 38(4), 658664. doi: 10.1111/j.1447-0756.2011.01771.x Migliaccio, C.T., Bergauff, M.A., Palmer, C.P., Jessop, F., Noonan, C.W., & Ward, T.J. (2009). Urinary levoglucosan as a biomarker of wood smoke exposure: Observations in a mouse model and in children. Environmental Health Perspectives, 117(1), 74-79. doi: 10.1289/ehp.11378 Montuschi, P., Barnes, P.J., & Roberts, L.J., 2nd. (2004). Isoprostanes: Markers and mediators of oxidative stress. FASEB Journal, 18(15), 1791-1800. doi: 10.1096/fj.04-2330rev Morrow, J.D., & Roberts, L.J. (1997). The isoprostanes: Unique bioactive products of lipid peroxidation. Progress in Lipid Research,, 36(1), 1-21. Murphy, M.W., Lando, J.F., Kieszak, S.M., Sutter, M.E., Noonan, G.P., Brunkard, J.M., & McGeehin, M.A. (2013). Formaldehyde levels in FEMA-supplied travel trailers, park models, and mobile homes in Louisiana and Mississippi. Indoor Air, 23(2), 134-141. doi: 10.1111/j.1600-0668.2012.00800.x Myatt, L., & Cui, X. (2004). Oxidative stress in the placenta. Histochemistry and Cell Biology,, 122(4), 369-382. doi: 10.1007/s00418-004-0677-x Nakling, J., & Backe, B. (2002). Adverse obstetric outcome in fetuses that are smaller than expected at second trimester routine ultrasound examination. Acta Obstetricia et Gynecologica Scandinavica,, 81(9), 846-851.

222

Nathens, A.B., Neff, M.J., Jurkovich, G.J., Klotz, P., Farver, K., Ruzinski, J.T., . . . Maier, R.V. (2002). Randomized, prospective trial of antioxidant supplementation in critically ill surgical patients. Annals of Surgery,, 236(6), 814-822. National Health and Medical Research Council. (1995). Perinatal morbidity: Report of the health care committee expert panel on perinatal morbidity. Retrieved November 21, 2013, from http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/wh18.pdf National Toxicology Program. (2010). Final report on carcinogens background document for formaldehyde. In National Toxicology Program (Ed.), Report on carcinogens background document for formaldehyde (2010/08/26 ed., pp. i-512). Nazaroff, W.W., & Singer, B.C. (2004). Inhalation of hazardous air pollutants from environmental tobacco smoke in US residences. Journal of Exposure Analysis and Environmental Epidemiology,, 14 Suppl 1, S71-77. doi: 10.1038/sj.jea.7500361 Neel, N.R., & Alvarez, J.O. (1991). Maternal risk factors for low birth weight and intrauterine growth retardation in a Guatemalan population. Bulletin of the Pan American Health Organization,, 25(2), 152-165. Negi, R., Pande, D., Kumar, A., Khanna, R.S., & Khanna, H.D. (2012). Evaluation of biomarkers of oxidative stress and antioxidant capacity in the cord blood of preterm low birth weight neonates. The Journal of Maternal-Fetal & Neonatal Medicine,, 25(8), 1338-1341. doi: 10.3109/14767058.2011.633672 New, K.J., Reilly, M.E., Templeton, K., Ellis, G., James, P.E., McEneny, J., . . . Bailey, D.M. (2013). Free radical-mediated lipid peroxidation and systemic nitric oxide

223

bioavailability: Implications for postexercise hemodynamics. American Journal of Hypertension,, 26(1), 126-134. doi: 10.1093/ajh/hps025 Nguyen, T., Larsen, T., Engholm, G., & Moller, H. (2000). A discrepancy between gestational age estimated by last menstrual period and biparietal diameter may indicate an increased risk of fetal death and adverse pregnancy outcome. An International Journal of Obstetrics and Gynaecology,, 107(9), 1122-1129. NIOSH. (2003). Formaldehyde method: 2016. NIOSH manual of Analytical methods (NMAM). 4th. Retrieved July 17, 2012, from http://www.cdc.gov/niosh/docs/2003-154/pdfs/2016.pdf NIOSH. (2007). NIOSH pocket guide to chemical hazards. Retrieved March, 22, 2013, from http://www.cdc.gov/niosh/docs/2005-149/pdfs/2005-149.pdf Nishizawa, M., Kohno, M., Nishimura, M., Kitagawa, A., & Niwano, Y. (2005). Presence of peroxyradicals in cigarette smoke and the scavenging effect of shikonin, a naphthoquinone pigment. Chemical & Pharmaceutical Bulletin (Tokyo),, 53(7), 796-799. NMS LABS. (2013). Formic acid, urine test (2134U). Retrieved April 17, 2013, from http://www.nmslabs.com/tests/Formic-Acid--Urine/2134U O'Donovan, D.J., & Fernandes, C.J. (2004). Free radicals and diseases in premature infants. Antioxidants & Redox Signaling,, 6(1), 169-176. doi: 10.1089/152308604771978471 Obel, E.B. (1979). Pregnancy complications following legally induced abortion: An analysis of the population with special reference to prematurity. Danish Medical Bulletin,, 26(4), 192-199.

224

Odegard, R.A., Vatten, L.J., Nilsen, S.T., Salvesen, K.A., & Austgulen, R. (2000). Preeclampsia and fetal growth. Obstetrics and Gynecology, 96(6), 950-955. Office for Human Research Protections (OHRP). (1993). Special classes of subjects. IRB Guidebook. Retrieved April 13, 2013, from http://www.hhs.gov/ohrp/archive/irb/irb_chapter6.htm Ogasawara, K.K. (2009). Variation in fetal ultrasound biometry based on differences in fetal ethnicity. American Journal of Obstetrics and Gynecology,, 200(6), 676 e671-674. doi: 10.1016/j.ajog.2009.02.031 Ogata, M., & Iwamoto, T. (1990). Enzymatic assay of formic acid and gas chromatography of methanol for urinary biological monitoring of exposure to methanol. International Archives of Occupational and Environmental Health,, 62(3), 227-232. Ohmichi, K., Komiyama, M., Matsuno, Y., Takanashi, Y., Miyamoto, H., Kadota, T., . . . Mori, C. (2006). Formaldehyde exposure in a gross anatomy laboratory--personal exposure level is higher than indoor concentration. Environmental Science and Pollution Research International,, 13(2), 120-124. OHRP. (1993). Special classes of subjects. IRB Guidebook. Retrieved April 13, 2013, from http://www.hhs.gov/ohrp/archive/irb/irb_chapter6.htm OHRP. (2009). Protection of human subjects. Code of Fedeal regulations. Retrieved April 13, 2013, from http://www.hhs.gov/ohrp/archive/irb/irb_chapter6.htm Olsen, I.E., Groveman, S.A., Lawson, M.L., Clark, R.H., & Zemel, B.S. (2010). New intrauterine growth curves based on United States data. Pediatrics, 125(2), e214224. doi: 10.1542/peds.2009-0913

225

Olsen, J.H., & Dossing, M. (1982). Formaldehyde induced symptoms in day care centers. American Industrial Hygiene Association Journal,, 43(5), 366-370. doi: 10.1080/15298668291409866 Orwell, R.L., Wood, R.A., Burchett, M.D., Tarran, J., & Torpy, F. (2006). The PottedPlant Microcosm Substantially Reduces Indoor Air VOC Pollution: II. Laboratory Study. Water, Air & Soil Pollution, 177(1-4), 59-80. doi: 10.1007/s11270-0069092-3 OSHA. (2013). Regulations (Standards-29CFR). Retrieved March, 24, 2013, from http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_id=10076&p_tabl e=standards Osornio-Vargas, A.R., Bonner, J.C., Alfaro-Moreno, E., Martinez, L., Garcia-Cuellar, C., Ponce-de-Leon Rosales, S., . . . Rosas, I. (2003). Proinflammatory and cytotoxic effects of Mexico City air pollution particulate matter in vitro are dependent on particle size and composition. Environmental Health Perspectives,, 111(10), 1289-1293. Overman, D.O. (1985). Absence of embryotoxic effects of formaldehyde after percutaneous exposure in hamsters. Toxicology Letters, 24(1), 107-110. doi: 0378-4274(85)90147-X [pii] Overton, J.H., Kimbell, J.S., & Miller, F.J. (2001). Dosimetry modeling of inhaled formaldehyde: The human respiratory tract. Toxicological Sciences, 64(1), 122134.

226

Owen, P., & Khan, K.S. (1998). Fetal growth velocity in the prediction of intrauterine growth retardation in a low risk population. British Journal of Obstetrics and Gynaecology,, 105(5), 536-540. Oxford Biomedial Research. (2012). Enzyme Immunoassay for 15-Isoprostane F2t. Retrieved March, 23, 2013, from http://www.bionovuslifesciences.com.au/images/stories/file/OBR/pdf/EA84.pdf Paneth, N., & Susser, M. (1995). Early origin of coronary heart disease (the "Barker hypothesis"). British Medical Journal,, 310(6977), 411-412. Pang, M.W., Leung, T.N., Sahota, D.S., Lau, T.K., & Chang, A.M. (2003). Customizing fetal biometric charts. Ultrasound in Obstetrics and Gynecology,, 22(3), 271-276. doi: 10.1002/uog.196 Papageorghiou, A.T., Bakoulas, V., Sebire, N.J., & Nicolaides, K.H. (2008). Intrauterine growth in multiple pregnancies in relation to fetal number, chorionicity and gestational age. Ultrasound in Obstetrics and Gynecology,, 32(7), 890-893. doi: 10.1002/uog.6140 Parikh, L.I., Nolan, J., 3rd, Tefera, E., & Driggers, R. (2013). Fetal biometry: Does patient ethnicity matter? The Journal of Maternal-Fetal & Neonatal Medicine,. doi: 10.3109/14767058.2013.820696 Parker, A.J., Davies, P., Mayho, A.M., & Newton, J.R. (1984). The ultrasound estimation of sex-related variations of intrauterine growth. American Journal of Obstetrics and Gynecology,, 149(6), 665-669.

227

Parker, A.J., Davies, P., & Newton, J.R. (1982). Assessment of gestational age of the Asian fetus by the sonar measurement of crown-rump length and biparietal diameter. British Journal of Obstetrics and Gynaecology,, 89(10), 836-838. Parna, K., Rahu, M., Youngman, L.D., Rahu, K., Nygard-Kibur, M., & Koupil, I. (2005). Self-reported and serum cotinine-validated smoking in pregnant women in Estonia. Maternal and Child Health Journal,, 9(4), 385-392. doi: 10.1007/s10995-005-0022-6 Parrish, M.E., & Harward, C.N. (2000). Measurement of formaldehyde in a single puff of cigarette smoke using tunable diode laser infrared spectroscopy Applied Spectroscopy, 54, 1665-1677. Parthasarathy, S., Maddalena, R.L., Russell, M.L., & Apte, M.G. (2011). Effect of temperature and humidity on formaldehyde emissions in temporary housing units. Journal of the Air & Waste Management Association,, 61(6), 689-695. Patrick, D.L., Cheadle, A., Thompson, D.C., Diehr, P., Koepsell, T., & Kinne, S. (1994). The validity of self-reported smoking: A review and meta-analysis. American Journal of Public Health,, 84(7), 1086-1093. Pearson, T.A., Mensah, G.A., Alexander, R.W., Anderson, J.L., Cannon, R.O., 3rd, Criqui, M., . . . American Heart, A. (2003). Markers of inflammation and cardiovascular disease, application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation, 107(3), 499-511.

228

Pedersen, N.G., Figueras, F., Wojdemann, K.R., Tabor, A., & Gardosi, J. (2008). Early fetal size and growth as predictors of adverse outcome. Obstet Gynecol, 112(4), 765-771. doi: 10.1097/AOG.0b013e318187d034 Pedersen, N.G., Wojdemann, K.R., Scheike, T., & Tabor, A. (2008). Fetal growth between the first and second trimesters and the risk of adverse pregnancy outcome. Ultrasound in obstetrics & gynecology,, 32(2), 147-154. doi: 10.1002/uog.6109 Pellegrini, N., Salvatore, S., Valtuena, S., Bedogni, G., Porrini, M., Pala, V., . . . Brighenti, F. (2007). Development and validation of a food frequency questionnaire for the assessment of dietary total antioxidant capacity. Journal of Nutrition, 137(1), 93-98. Petrou, S., Mehta, Z., Hockley, C., Cook-Mozaffari, P., Henderson, J., & Goldacre, M. (2003). The impact of preterm birth on hospital inpatient admissions and costs during the first 5 years of life. Pediatrics, 112(6 Pt 1), 1290-1297. Petry, T., Cazelle, E., Lloyd, P., Mascarenhas, R., & Stijntjes, G. (2013). A standard method for measuring benzene and formaldehyde emissions from candles in emission test chambers for human health risk assessment purposes. Environmental Science. Processes & Impacts,, 15(7), 1369-1382. doi: 10.1039/c3em00011g Pharoah, P.O., Stevenson, C.J., Cooke, R.W., & Stevenson, R.C. (1994). Prevalence of behaviour disorders in low birthweight infants. Archives of Disease in Childhood,, 70(4), 271-274.

229

Pickett, K.E., Rathouz, P.J., Kasza, K., Wakschlag, L.S., & Wright, R. (2005). Selfreported smoking, cotinine levels, and patterns of smoking in pregnancy. Paediatric and Perinatal Epidemiology,, 19(5), 368-376. doi: 10.1111/j.13653016.2005.00660.x Polit, D.F., & Beck, C.T. (2008). Nursing research: Generating and assessing evidence for nursing practice. Philadelphia, PA: Lippincott Williams & Wilkins2. Pollack, R.N., & Divon, M.Y. (1992). Intrauterine growth retardation: Definition, classification, and etiology. Clinical Obstetrics and Gynecology, 35(1), 99-107. Post, A., Gilljam, H., Bremberg, S., & Galanti, M.R. (2008). Maternal smoking during pregnancy: A comparison between concurrent and retrospective self-reports. Paediatric Perinatal Epidemiology,, 22(2), 155-161. doi: 10.1111/j.13653016.2007.00917.x Prabhu, N., Smith, N., Campbell, D., Craig, L.C., Seaton, A., Helms, P.J., . . . Turner, S.W. (2010). First trimester maternal tobacco smoking habits and fetal growth. Thorax, 65(3), 235-240. doi: 10.1136/thx.2009.123232 Pringle, P.J., Geary, M.P., Rodeck, C.H., Kingdom, J.C., Kayamba-Kay's, S., & Hindmarsh, P.C. (2005). The influence of cigarette smoking on antenatal growth, birth size, and the insulin-like growth factor axis. The Journal of Clinical Endocrinology and Metabolism,, 90(5), 2556-2562. doi: 10.1210/jc.2004-1674 Raaschou-Nielsen, O., Hermansen, M.N., Loland, L., Buchvald, F., Pipper, C.B., Sorensen, M., . . . Bisgaard, H. (2010). Long-term exposure to indoor air pollution and wheezing symptoms in infants. Indoor Air, 20(2), 159-167. doi: 10.1111/j.1600-0668.2009.00635.x

230

Radulescu, L., Ferechide, D., & Popa, F. (2013). The importance of fetal gender in intrauterine growth restriction. Journal of Medicine and Life,, 6(1), 38-39. Risner, C.H., & Martin, P. (1994). Quantitation of formaldehyde, acetaldehyde, and acetone in sidestream cigarette smoke by high-performance liquid chromatography. J Chromatogr Sci, 32(3), 76-82. Roberts, L.J., 2nd, & Milne, G.L. (2009). Isoprostanes. Journal of Lipid Research,, 50 Suppl, S219-223. doi: 10.1194/jlr.R800037-JLR200 Roberts, L.J., 2nd, & Morrow, J.D. (2002). Products of the isoprostane pathway: Unique bioactive compounds and markers of lipid peroxidation. Cellular and Molecular Life Sciences,, 59(5), 808-820. Robillard, J.E., & Segar, J.L. (2006). Influence of early life events on health and diseases. Transactions of the American Clinical and Climatological Association,, 117, 313319; discussion 319-320. Rocchi, E., Bursi, F., Ventura, P., Ronzoni, A., Gozzi, C., Casalgrandi, G., . . . Modena, M.G. (2007). Anti- and pro-oxidant factors and endothelial dysfunction in chronic cigarette smokers with coronary heart disease. European Journal of Internal Medicine,, 18(4), 314-320. doi: 10.1016/j.ejim.2006.10.006 Rochelson, B., Kaplan, C., Guzman, E., Arato, M., Hansen, K., & Trunca, C. (1990). A quantitative analysis of placental vasculature in the third-trimester fetus with autosomal trisomy. Obstetrics and Gynecology, 75(1), 59-63. Roda, C., Kousignian, I., Guihenneuc-Jouyaux, C., Dassonville, C., Nicolis, I., Just, J., & Momas, I. (2011). Formaldehyde exposure and lower respiratory infections in

231

infants: Findings from the PARIS cohort study. Environmental Health Perspectives, 119(11), 1653-1658. doi: 10.1289/ehp.1003222 Rode, L., Hegaard, H.K., Kjaergaard, H., Moller, L.F., Tabor, A., & Ottesen, B. (2007). Association between maternal weight gain and birth weight. Obstetrics and Gynecology, 109(6), 1309-1315. doi: 10.1097/01.AOG.0000266556.69952.de Rodella, L.F., Favero, G., Foglio, E., Rossini, C., Castrezzati, S., Lonati, C., & Rezzani, R. (2013). Vascular endothelial cells and dysfunctions: Role of melatonin. Frontiers in bioscience,, 5, 119-129. Rogers, J.V., Choi, Y.W., Richter, W.R., Rudnicki, D.C., Joseph, D.W., Sabourin, C.L., . . . Chang, J.C. (2007). Formaldehyde gas inactivation of Bacillus anthracis, Bacillus subtilis, and Geobacillus stearothermophilus spores on indoor surface materials. Journal of Applied Microbiology,, 103(4), 1104-1112. doi: 10.1111/j.1365-2672.2007.03332.x Romanazzi, V., Pirro, V., Bellisario, V., Mengozzi, G., Peluso, M., Pazzi, M., . . . Bono, R. (2013). 15-F(2)t isoprostane as biomarker of oxidative stress induced by tobacco smoke and occupational exposure to formaldehyde in workers of plastic laminates. The Science of the Total Environment,, 442, 20-25. doi: 10.1016/j.scitotenv.2012.10.057 Rossner, P., Jr., Milcova, A., Libalova, H., Novakova, Z., Topinka, J., Balascak, I., & Sram, R.J. (2009). Biomarkers of exposure to tobacco smoke and environmental pollutants in mothers and their transplacental transfer to the foetus. Part II. Oxidative damage. Mutation Research,, 669(1-2), 20-26. doi: 10.1016/j.mrfmmm.2009.04.010

232

Roth, E., Manhart, N., & Wessner, B. (2004). Assessing the antioxidative status in critically ill patients. Current Opinion in Clinical Nutrition and Metabolic Care,, 7(2), 161-168. Roth, J., Hendrickson, J., Schilling, M., & Stowell, D.W. (1998). The risk of teen mothers having low birth weight babies: Implications of recent medical research for school health personnel. Journal of School Health, 68(7), 271-275. Roza, S.J., Verburg, B.O., Jaddoe, V.W., Hofman, A., Mackenbach, J.P., Steegers, E.A., . . . Tiemeier, H. (2007). Effects of maternal smoking in pregnancy on prenatal brain development. The Generation R Study. The European Journal of Neuroscience,, 25(3), 611-617. doi: 10.1111/j.1460-9568.2007.05393.x Rustemeier, K., Stabbert, R., Haussmann, H.J., Roemer, E., & Carmines, E.L. (2002). Evaluation of the potential effects of ingredients added to cigarettes. Part 2: Chemical composition of mainstream smoke. Food and Chemical Toxicology,, 40(1), 93-104. Saillenfait, A.M., Bonnet, P., & de Ceaurriz, J. (1989). The effects of maternally inhaled formaldehyde on embryonal and foetal development in rats. Food and Chemical Toxicology, 27(8), 545-548. Sainio, E.L., Engstrom, K., Henriks-Eckerman, M.L., & Kanerva, L. (1997). Allergenic ingredients in nail polishes. Contact Dermatitis, 37(4), 155-162. Saito, Y., Nishio, K., Yoshida, Y., & Niki, E. (2005). Cytotoxic effect of formaldehyde with free radicals via increment of cellular reactive oxygen species. Toxicology, 210(2-3), 235-245. doi: 10.1016/j.tox.2005.02.006

233

Sakamoto, W., Isomura, H., Fujie, K., Takahashi, K., Nakao, K., & Izumi, H. (2005). Relationship of coffee consumption with risk factors of atherosclerosis in rats. Annals of Nutrition & Metabolism,, 49(3), 149-154. doi: 10.1159/000086170 Salthammer, T. (2013). Formaldehyde in the ambient atmosphere: From an indoor pollutant to an outdoor pollutant? Angewandte Chemie International Edition in English, 52(12), 3320-3327. doi: 10.1002/anie.201205984 Salthammer, T., Mentese, S., & Marutzky, R. (2010). Formaldehyde in the indoor environment. Chemical Reviews,, 110(4), 2536-2572. doi: 10.1021/cr800399g Saurel-Cubizolles, M.J., Hays, M., & Estryn-Behar, M. (1994). Work in operating rooms and pregnancy outcome among nurses. International Archives of Occupational and Environmental Health,, 66(4), 235-241. Schlesinger, R.B., Kunzli, N., Hidy, G.M., Gotschi, T., & Jerrett, M. (2006). The health relevance of ambient particulate matter characteristics: Coherence of toxicological and epidemiological inferences. Inhalation Toxicology,, 18(2), 95-125. doi: 10.1080/08958370500306016 Schmid, K., Schaller, K.H., Angerer, J., & Lehnert, G. (1994). [The importance of formic acid excretion in the urine for environmental and occupational medicine questions]. Zentralbl Hyg Umweltmed, 196(2), 139-152. Schmitt, S.K., Sneed, L., & Phibbs, C.S. (2006). Costs of newborn care in California: a population-based study. Pediatrics, 117(1), 154-160. doi: 10.1542/peds.20050484 Scholl, T.O., Hediger, M.L., & Schall, J.I. (1996). Re: "Mechanisms for maternal age differences in birth weight". American Journal of Epidemiology, 143(8), 842-843.

234

Schwarzler, P., Bland, J.M., Holden, D., Campbell, S., & Ville, Y. (2004). Sex-specific antenatal reference growth charts for uncomplicated singleton pregnancies at 1540 weeks of gestation. Ultrasound in Obstetrics & Gynecology,, 23(1), 23-29. doi: 10.1002/uog.966 Schwenzer, K.J. (2008). Protecting vulnerable subjects in clinical research: Children, pregnant women, prisoners, and employees. Respiratory Care,, 53(10), 13421349. Sclowitz, I.K., Santos, I.S., Domingues, M.R., Matijasevich, A., & Barros, A.J. (2013). Maternal smoking in successive pregnancies and recurrence of low birthweight: The 2004 Pelotas birth cohort study, Brazil. Cadernos de Saude Publica,, 29(1), 123-130. Sexton, K., Petreas, M.X., & Liu, K.-S. (1989). Formaldehyde exposures inside mobile homes. Environmental Science & Technology,, 23(8), 985. Shah, P., & Ohlsson, A. (2002). Literature review of low birth weight, including small for gestational age and preterm birth. Retrieved July 16, 2012, from http://www.toronto.ca/health/low_birth_weight/pdf/lbw_lit_rewiew.pdf Shin, H.S., Ahn, H.S., & Lee, B.H. (2007). Determination of thiazolidine-4-carboxylates in urine by chloroformate derivatization and gas chromatography-electron impact mass spectrometry. Journal of Mass Spectrometry, 42(9), 1225-1232. doi: 10.1002/jms.1255 Shipp, T.D., Bromley, B., Mascola, M., & Benacerraf, B. (2001). Variation in fetal femur length with respect to maternal race. Journal of Ultrasound in Medicine,, 20(2), 141-144.

235

Shumilina, A.V. (1975). [Menstrual and child-bearing functions of female workers occupationally exposed to the effects of formaldehyde]. Gig Tr Prof Zabol(12), 18-21. Siciliano, G., Simoncini, C., Lo Gerfo, A., Orsucci, D., Ricci, G., & Mancuso, M. (2012). Effects of aerobic training on exercise-related oxidative stress in mitochondrial myopathies. Neuromuscular Disorders,, 22 Suppl 3, S172-177. doi: 10.1016/j.nmd.2012.10.005 Sies, H. (1997). Oxidative stress: Oxidants and antioxidants. Experimental Physiology, 82(2), 291-295. Singer, B.C., Destaillats, H., Hodgson, A.T., & Nazaroff, W.W. (2006). Cleaning products and air fresheners: Emissions and resulting concentrations of glycol ethers and terpenoids. Indoor Air, 16(3), 179-191. doi: 10.1111/j.16000668.2005.00414.x Singh, A., & Toppo, A. (2011). A better classification for determination of still births. Journal of obstetrics and gynaecology of India,, 61(6), 656-658. doi: 10.1007/s13224-011-0115-7 Siu, M.T., Shapiro, A.M., Wiley, M.J., & Wells, P.G. (2013). A role for glutathione, independent of oxidative stress, in the developmental toxicity of methanol. Toxicology and Applied Pharmacology,. doi: 10.1016/j.taap.2013.09.020 Skovron, M.L., Berkowitz, G.S., Lapinski, R.H., Kim, J.M., & Chitkara, U. (1991). Evaluation of early third-trimester ultrasound screening for intrauterine growth retardation. Journal of Ultrasound in Medicine, , 10(3), 153-159.

236

Smith, G.C., Smith, M.F., McNay, M.B., & Fleming, J.E. (1998). First-trimester growth and the risk of low birth weight. The New England Journal of Medicine,, 339(25), 1817-1822. doi: 10.1056/NEJM199812173392504 Smith, G.C., Stenhouse, E.J., Crossley, J.A., Aitken, D.A., Cameron, A.D., & Connor, J.M. (2002). Early-pregnancy origins of low birth weight. Nature, 417(6892), 916. doi: 10.1038/417916a Smulian, J.C., Campbell, W.A., Rodis, J.F., Feeney, L.D., Fabbri, E.L., & Vintzileos, A.M. (1995). Gender-specific second-trimester biometry. American Journal of Obstetrics and Gynecology,, 173(4), 1195-1201. Sögüt, S., Songur, A., Özen, O.A., Özyurt, H., & Sarsilmaz, S. (2004). Does the subacute (4-week) exposure to formaldehyde inhalation lead to oxidant/antioxidant imbalance in rat liver? European Journal of General Medicine, 1(3), 26-32. Souka, A.P., Papastefanou, I., Pilalis, A., Michalitsi, V., & Kassanos, D. (2012). Performance of third-trimester ultrasound for prediction of small-for-gestationalage neonates and evaluation of contingency screening policies. Ultrasound Obstetrics & Gynecology,, 39(5), 535-542. doi: 10.1002/uog.10078 Soylemez, H., Bozkurt, Y., Penbegul, N., Sancaktutar, A.A., Altunoluk, B., Atar, M., . . . Hatipoglu, N.K. (2013). Time-dependent oxidative stress effects of percutaneous nephrolithotomy. Urolithiasis, 41(1), 65-71. doi: 10.1007/s00240-012-0532-9 Spanel, P., Smith, D., Holland, T.A., Al Singary, W., & Elder, J.B. (1999). Analysis of formaldehyde in the headspace of urine from bladder and prostate cancer patients using selected ion flow tube mass spectrometry. Rapid Communications in Mass

237

Spectrometry, , 13(14), 1354-1359. doi: 10.1002/(SICI)10970231(19990730)13:143.0.CO;2-J SRNT. (2002). Biochemical verification of tobacco use and cessation. Nicotine and Tobacco Research,, 4(2), 149-159. doi: 10.1080/14622200210123581 Stagno, S., Reynolds, D.W., Huang, E.S., Thames, S.D., Smith, R.J., & Alford, C.A. (1977). Congenital cytomegalovirus infection. New England Journal of Medicine, 296(22), 1254-1258. doi: 10.1056/NEJM197706022962203 Statistical Package for the Social Sciences. (2013). SPSS 21.0 for Windows. Chicago, IL. Stock, T.H., & Mendez, S.R. (1985). A survey of typical exposures to formaldehyde in Houston area residences. American Industrial Hygiene Association Journal,, 46(6), 313-317. doi: 10.1080/15298668591394888 Suresh, K., & Chandrashekara, S. (2012). Sample size estimation and power analysis for clinical research studies. Journal of Human Reproductive Sciences,, 5(1), 7-13. doi: 10.4103/0974-1208.97779 Tabacova, S., Baird, D.D., & Balabaeva, L. (1998). Exposure to oxidized nitrogen: Lipid peroxidation and neonatal health risk. Archives of Environmental Health, 53(3), 214-221. doi: 10.1080/00039899809605698 The American College of Obstetricians and Gynecologists. (2013a). Environmental chemicals harm reproductive health. Retrieved February 19, 2014, from http://www.acog.org/About_ACOG/News_Room/News_Releases/2013/Environm ental_Chemicals_Harm_Reproductive_Health

238

The American College of Obstetricians and Gynecologists. (2013b). Fetal growth restriction. Practice Bulletin Clinical Management Guidelines for Obstetricians and Gynecologysts(134). Thompson, L.P., & Al-Hasan, Y. (2012). Impact of oxidative stress in fetal programming. J Pregnancy, 2012, 582748. doi: 10.1155/2012/582748 Thrasher, J.D., & Kilburn, K.H. (2001). Embryo toxicity and teratogenicity of formaldehyde. Archives of Environmental Health, 56(4), 300-311. doi: 10.1080/00039890109604460 Triebig, G., Schaller, K.H., Beyer, B., Muller, J., & Valentin, H. (1989). Formaldehyde exposure at various workplaces. The Science of the Total Environment,, 79(2), 191-195. Tulpule, K., & Dringen, R. (2013). Formaldehyde in brain: An overlooked player in neurodegeneration? Journal of Neurochemistry,, 127(1), 7-21. doi: 10.1111/jnc.12356 Tunon, K., Eik-Nes, S.H., & Grottum, P. (1999). Fetal outcome in pregnancies defined as post-term according to the last menstrual period estimate, but not according to the ultrasound estimate. Ultrasound in Obstetrics and Gynecology,, 14(1), 12-16. doi: 10.1046/j.1469-0705.1999.14010012.x Tylicki, L., Puttinger, H., Rutkowski, P., Rutkowski, B., & Horl, W.H. (2006). Smoking as a risk factor for renal injury in essential hypertension. Nephron Clinical Practice,, 103(4), c121-128. doi: 10.1159/000092908 Uhde, E., & Salthammer, T. (2007). Impact of reaction products from building materials and furnishings on indoor air quality—A review of recent advances in indoor

239

chemistry. Atmospheric Environment, 41(15), 3111-3128. doi: 10.1016/j.atmosenv.2006.05.082 United States Census Bureau. (2013). Alabama: People QuickFacts. Retrieved November 6, 2013, from http://quickfacts.census.gov/qfd/states/01000.html Upreti, R.K., Farooqui, M.Y., Ahmed, A.E., & Ansari, G.A. (1987). Toxicokinetics and molecular interaction of [14C]-formaldehyde in rats. Archives of Environmental Contamination and Toxicology,, 16(3), 263-273. USDHEW. (1979). Smoking and health: A report of Surgeon General ((PHS) 79-50069). Washington, DC: Government Printing Office. Valavanidis, A., Fiotakis, K., & Vlachogianni, T. (2008). Airborne particulate matter and human health: Toxicological assessment and importance of size and composition of particles for oxidative damage and carcinogenic mechanisms. Journal of Environmental Science and Health. Part C, Environmental Carcinogenesis & Ecotoxicology Reviews,, 26(4), 339-362. doi: 10.1080/10590500802494538 Vasudeva, A., Abraham, A.A., & Kamath, A. (2013). A prospective observational study of early fetal growth velocity and its association with birth weight, gestational age at delivery, preeclampsia, and perinatal mortality. European Journal of Radiology, 82(8), 1313-1317. doi: http://dx.doi.org/10.1016/j.ejrad.2013.02.016 Veen, S., Ens-Dokkum, M.H., Schreuder, A.M., Verloove-Vanhorick, S.P., Brand, R., & Ruys, J.H. (1991). Impairments, disabilities, and handicaps of very preterm and very-low-birthweight infants at five years of age. The Collaborative Project on Preterm and Small for Gestational Age Infants (POPS) in The Netherlands. Lancet, 338(8758), 33-36.

240

Venditti, C.C., Casselman, R., & Smith, G.N. (2011). Effects of chronic carbon monoxide exposure on fetal growth and development in mice. BMC Pregnancy Childbirth, 11, 101. doi: 10.1186/1471-2393-11-101 Vettore, M.V., Gama, S.G., Lamarca Gde, A., Schilithz, A.O., & Leal Mdo, C. (2010). Housing conditions as a social determinant of low birthweight and preterm low birthweight. Rev Saude Publica, 44(6), 1021-1031. Vidaeff, A.C., Yeomans, E.R., & Ramin, S.M. (2008). Pregnancy in women with renal disease. Part I: General principles. American Journal of Perinatology, 25(7), 385397. doi: 10.1055/s-0028-1083837 Viviano, K.R., & Vanderwielen, B. (2013). Effect of N-Acetylcysteine supplementation on sntracellular glutathione, urine isoprostanes, clinical score, and survival in hospitalized Ill dogs. Journal of Veterinary Internal Medicine,. doi: 10.1111/jvim.12048 Vrijheid, M., Casas, M., Bergstrom, A., Carmichael, A., Cordier, S., Eggesbo, M., . . . Nieuwenhuijsen, M. (2011). European Birth Cohorts for Environmental Health Research. Environmental Health Perspectives. doi: 10.1289/ehp.1103823 Wang, F., Li, C., Liu, W., & Jin, Y. (2013). Oxidative damage and genotoxic effect in mice caused by sub-chronic exposure to low-dose volatile organic compounds. Inhalation Toxicology,, 25(5), 235-242. doi: 10.3109/08958378.2013.779767 Wang, X., Chen, D., Niu, T., Wang, Z., Wang, L., Ryan, L., . . . Xu, X. (2000). Genetic susceptibility to benzene and shortened gestation: Evidence of gene-environment interaction. American Journal of Epidemiology, 152(8), 693-700.

241

Waterson, A.P. (1979). Virus infections (other than rubella) during pregnancy. British Medical Journal, 2(6190), 564-566. Webb, D.A., Boyd, N.R., Messina, D., & Windsor, R.A. (2003). The discrepancy between self-reported smoking status and urine continine levels among women enrolled in prenatal care at four publicly funded clinical sites. Journal of Public Health Management and Practice,, 9(4), 322-325. Wen, C.P., Cheng, T.Y., Lin, C.L., Wu, H.N., Levy, D.T., Chen, L.K., . . . Tsai, S.P. (2005). The health benefits of smoking cessation for adult smokers and for pregnant women in Taiwan. Tobacco Control,, 14 Suppl 1, i56-61. doi: 10.1136/tc.2004.007997 West, R., Zatonski, W., Przewozniak, K., & Jarvis, M.J. (2007). Can we trust national smoking prevalence figures? Discrepancies between biochemically assessed and self-reported smoking rates in three countries. Cancer Epidemiol Biomarkers & Prevention,, 16(4), 820-822. doi: 10.1158/1055-9965.epi-06-0679 WHO. (2002). Using indicators to measure progress on children's environmental health Retrieved August 20, 2013, from http://www.who.int/ceh/publications/cehcallforaction/en/ WHO. (2006). Neonatal and perinatal mortality: Country, regional and global estimates. Retrieved November 21, 2013, from http://whqlibdoc.who.int/publications/2006/9241563206_eng.pdf WHO. (2010). WHO guidelines for indoor air quality: Selected pollutants. Retrieved December 20, 2012, from http://www.euro.who.int/__data/assets/pdf_file/0009/128169/e94535.pdf

242

WHO, & UNICEF. (2004). Low birthweight; Global, regional, and global estimates. Retrieved January 10, 2013, from http://www.unicef.org/publications/files/low_birthweight_from_EY.pdf Wieslander, G., Norback, D., Bjornsson, E., Janson, C., & Boman, G. (1997). Asthma and the indoor environment: The significance of emission of formaldehyde and volatile organic compounds from newly painted indoor surfaces. International Archives of Occupational and Environmental Health,, 69(2), 115-124. Wisdom, S.J., Wilson, R., McKillop, J.H., & Walker, J.J. (1991). Antioxidant systems in normal pregnancy and in pregnancy-induced hypertension. American Journal of Obstetrics and Gynecology,, 165(6 Pt 1), 1701-1704. Wolfe, H.M., & Gross, T.L. (1989). Increased risk of the growth retarded fetus. In T. M. Gross & R. J. Sokol (Eds.), Intrauterine Growth Retardation (pp. 111). Chicago: Year Book Medical Publishres. Wood, R.A., Burchett, M.D., Alquezar, R., Orwell, R.L., Tarran, J., & Torpy, F. (2006). The Potted-Plant Microcosm Substantially Reduces Indoor Air VOC Pollution: I. Office Field-Study. Water, Air & Soil Pollution, 175(1-4), 163-180. doi: 10.1007/s11270-006-9124-z Woolbright, L.A. (2013). County health profiles 2011. Retrieved September 8, 2013, from http://www.adph.org/healthstats/assets/CHP2011.pdf Woolbright, L.A., Zheng, Q., & Stones, A.D. (2013). Selected maternal and child health statistics Alabama 2011. Retrieved September, 9, 2013, from http://www.adph.org/healthstats/assets/MCH2011.pdf

243

Wozniak, A., Kulza, M., Senczuk-Przybylowska, M., Cimino, F., Saija, A., Ignatowicz, E., . . . Florek, E. (2012). [Selected biochemical parameters of oxidative stress as a result of exposure to tobacco smoke in animals addicted to ethyl alcohol]. Przegl Lek, 69(10), 824-832. Wu, F.Y., Chiu, H.T., Wu, H.D., Lin, C.J., Lai, J.S., & Kuo, H.W. (2008). Comparison of urinary and plasma cotinine levels during the three trimesters of pregnancy. Paediatric and Perinatal Epidemiology,, 22(3), 296-301. doi: 10.1111/j.13653016.2008.00927.x Yasugi, T., Kawai, T., Mizunuma, K., Horiguchi, S., Iwami, O., Iguchi, H., & Ikeda, M. (1992). Formic acid excretion in comparison with methanol excretion in urine of workers occupationally exposed to methanol. International Archives of Occupational and Environmental Health,, 64(5), 329-337. Yeo, G.S., Chan, W.B., Lun, K.C., & Lai, F.M. (1994). Racial differences in fetal morphometry in Singapore. Annals of the Academy of Medicine, Singapore,, 23(3), 371-376. Yildiz, Y., Ozgu, E., Unlu, S.B., Salman, B., & Eyi, E.G. (2013). The relationship between third trimester maternal hemoglobin and birth weight/length; Results from the tertiary center in Turkey. The Journal of Maternal-Fetal & Neonatal Medicine,. doi: 10.3109/14767058.2013.837445 Zhang, H., & Cai, B. (2003). The impact of tobacco on lung health in China. Respirology, 8(1), 17-21.

244

Zhang, J., Merialdi, M., Platt, L.D., & Kramer, M.S. (2010). Defining normal and abnormal fetal growth: Promises and challenges. American Journal of Obstetrics and Gynecology, 202(6), 522-528. doi: 10.1016/j.ajog.2009.10.889 Zhang, L., Chung, F.L., Boccia, L., Colosimo, S., Liu, W., & Zhang, J. (2003). Effects of garage employment and tobacco smoking on breathing-zone concentrations of carbonyl compounds. AIHA Journal,, 64(3), 388-393. Zhang, S., Zheng, R., Yang, L., Zhang, X., Zuo, L., Yang, X., . . . Liu, H. (2013). Angiotensin type 1 receptor autoantibody from preeclamptic patients induces human fetoplacental vasoconstriction. Journal of Cellular Physiology,, 228(1), 142-148. doi: 10.1002/jcp.24113 Zhang, Y., Liu, X., McHale, C., Li, R., Zhang, L., Wu, Y., . . . Ding, S. (2013). Bone marrow injury induced via oxidative stress in mice by inhalation exposure to formaldehyde. PLoS One, 8(9), e74974. doi: 10.1371/journal.pone.0074974 Zhao, Z., Xu, Z., Le, K., Azordegan, N., Riediger, N.D., & Moghadasian, M.H. (2009). Lack of evidence for antiatherogenic effects of wheat bran or corn bran in apolipoprotein E-knockout mice. Journal of Agricultural and Food Chemistry,, 57(14), 6455-6460. doi: 10.1021/jf900090q Zheng, Q., Wouolbright, L.A., & Afgan, I. (2012). PRAM surveilance report Alabama 2010. Montgomery, Al: Alabama Department of Public Health. Zhu, B.P., Haines, K.M., Le, T., McGrath-Miller, K., & Boulton, M.L. (2001). Effect of the interval between pregnancies on perinatal outcomes among white and black women. American Journal of Obstetrics and Gynecology, 185(6), 1403-1410. doi: 10.1067/mob.2001.118307

245

Zhu, B.P., Rolfs, R.T., Nangle, B.E., & Horan, J.M. (1999). Effect of the interval between pregnancies on perinatal outcomes. New England Journal of Medicine, 340(8), 589-594. doi: 10.1056/NEJM199902253400801 Zhu, J.L. (2006). Laboratory work and pregnancy outcomes: A study within the National Birth Cohort in Denmark. Occupational and Environmental Medicine, 63(1), 5358. doi: 10.1136/oem.2005.021204 Zhu, J.L., Obel, C., Hammer Bech, B., Olsen, J., & Basso, O. (2007). Infertility, infertility treatment, and fetal growth restriction. Obstetrics and Gynecology, 110(6), 1326-1334. doi: 10.1097/01.AOG.0000290330.80256.97

246

APPENDIX A IRB APPROVAL LETTER

247

248

249

250

251

252

253

254

APPENDIX B INDIVIDUAL INVESTIGATOR DISTINGUISHED RESEARCH GRANT FINANCIAL SPONSOR

255

256

APPENDIX C LETTERS OF SUPPORT

257

258

259

260

261

262

APPENDIX D QUESTIONNAIRE/FORMS

263

Code:----------------------

History of (exclusion criteria): Any Yes means she is not eligible for this study □ Alcohol Abuser

□ Infertility (previous or current)

□ Drug addition

□ In vitro fertilization

□ Hypertension

□ Placental abnormalities (previa or

□ Anemia □ Thyroid diseases □ Kidney diseases □ Diabetes (Type I/II/ or gestational) □ Autoimmune disease □ Fetal abnormality based on last ultrasound □ Chromosomal defects based on amniocentesis or ultrasound

abruption) based on last ultrasound □ History of any viral infection or parasite during pregnancy (rubella, varicella, cytomegalovirus, herpes zoster, toxoplasmosis) □ Doesn’t know how to read and speak English □ First appointment during pregnancy (no prenatal care) □ Oligoamnios or hydramnios based on ultrasound

264

Code: -----------

Date of Completion:-------------------Medical Health Record Data Sheet

Last Menstrual Period (LMP) Date: ----------------- Due date based on Ultrasound: --------------Maternal age: ---------------Gravida: ---------------- Parity: ----------How many visits for prenatal care did she have during this pregnancy? ---------------------Termination date of the past pregnancy (live birth, abortion, or still birth): Cigarette smoke, if yes pack/year: -----------------------------------Gender: Male female

GA based on the report

New calculated GA Percentile (Hadlock) Abdominal circumference Femur length Head circumference Estimated weight Abdominal circumference/femur length

265

Notes

Code: ----------------

Date of interview: ------------------Demographic Questionnaire

1

What is your race/ethnicity or ethnicity? (Select one or more)

□ American-Indian or Alaska native □ Black or African American □ White □ Asian, specify (optional) □ Hispanic □ Native Hawaiian □ Other (Specify-optional)

2

What is your highest level of completed education?

□ High School or less □ GED □ Some College □ Two Year College Graduate □ Four Year College Graduate □ Post Graduate □ Other, please specify

3

What is your current marital status?

□ Married □ Single-divorced □ Single- never married □ Living with partner

5a

What is your current employment status?

□ Employed full time □ Employed part time □ Not-employed □ Not-employed work at home 266

5b

If yes, what is your job?

5c

What kind of facility do you work?

□ Hospital □ Company (please specify) □ School system □ Laboratory (please specify) □ Store (please specify) □ Pharmacy □ Others: --------------------

6

Which one best describes your level of family income?

□ Less than $10,000 □ $10,001 - $20,000 □ $20,001 - $30,000 □ $30,001 - $40,000 □ $40,001 - $50, 000 □ Greater than $50,000

267

Code:--------------Residential Dwelling Environmental Questionnaire Season of data collection: Winter 1

Spring

Summer

What kind of residence do you have?

Fall

□ House □ Apartment □ Mobile home □ Travel trailer

2

When the house/ apartment is built:

□ New home □ Less than 5 year □ Between 5-10 year □ More than 1 0 year

3

What kind of heating system do you have?

□ None □ Electricity □ Gas □ Wood □ Other:---------------------------

4b

If yes, how many hours per day it is on?

Never 2-3 hours a week 2-3 hours a day 4-8 hours a day All day

5

What kind of cooling system do you have?

□ None □ Electricity □ Gas

268

□ Other:--------------------------6

In what temperature do you usually set your heating system?

-----˚ F

7

Does anyone smoke in your house?

□ Yes □ No □ Don’t know

8

If you live in an apartment, do you know of any neighbors who smoke?

□ Yes □ No □ Don’t know

9

What type of cooking appliance?

□ Electricity □ Propane gas □ Natural gas □ Wood □ Charcoal Other :-------------□ Don’t know

10

What is the average amount of time that you spend inside your house each day?

□ Less than one hour □ 1-4 hours □ 5-8 hours □ More than 8 hours □ Don’t know

11

Do you use any of the following regularly? Mark all that apply

□ Candles □ Air fresheners □ Glue, paint, furniture finish

269

□ Mothballs □ Closet fresh (Mildewcide) □ Nail polish □ Perfume 12

If yes how often? (Name of the material)

□ Never □ 1-2 times a months □ 1-2 times a week □ 2-4 times a week □ everyday

13

If yes how often? (Name of the material)

□ Never □ 1-2 times a months □ 1-2 times a week □ 2-4 times a week □ everyday

14

If yes how often? (Name of the material)

□ Never □ 1-2 times a months □ 1-2 times a week □ 2-4 times a week □ everyday

15

Have you added new furniture (including baby’s furniture) to your house within last year?

□ Yes □ No □ Don’t know

16

Have you added any new carpet in your house within the past 2-3 years?

270

□ Yes □ No

□ Don’t know 17

Have you remodeled your home within the past 5 years?

□ Yes □ No □ Don’t know

271

Key Form Code

Name

Phone number

Zipcode

Date of data collection

272

Lab Sample Storage Form Random

Data of

Centrifuge

Number

Number and

Number

Number

Date the

Purpose

number

data

before

of stored

location of

of the

of the

sample

of the

collection

storage

samples

Freezer

freezer

freezer

was

sample

rack

box

removed

removal

(Y/N)

273

APPENDIX E INSTRUCTION FOR USING VAPOR MONITOR BADGE

274

INSTRUCTION FOR USING VAPOR MONITOR BADGE 1. The following instruction was used by the principal investigator for using monitor badge (1) show the badge and the plastic bag to pregnant women, (2) review the purpose and procedure, and safety of using this badge, (3) label the badge and a plastic bag with participant’s random number, (4) assign the start and stop times and dates time on the plastic bag and the instruction page that were given to the participant, (5) clip the badge clothing (shirt, dress) close to the breathing zone on the upper chest area shoulder unless the participant states that she needs to do that later, (6) if they wish to wear the badge later the principal investigator demonstrated to them how to clip it and asked them to do not open the package unless they are ready to use, (7) Provide a copy of the “instruction for using monitor badge”, (8) Explain that they need to keep the badge clipped near the breathing zone for 24 hours, except at the sleep time that they need to take it off and put it uncovered near their head on a table or chair, (9) also they were told to put the sample in a plastic bag that is provided and return it to physicians’ office as soon as possible.

275

APPENDIX F BIOSAFETY LEVEL 1 & 2 FOR INFECTIOUS AGENTS

276

277

278

APPENDIX H TESTS PROCEDURES

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307