The

n e w e ng l a n d j o u r na l

of

m e dic i n e

original article

Maternal Vitamin A Supplementation and Lung Function in Offspring William Checkley, M.D., Ph.D., Keith P. West, Jr., Dr.P.H., Robert A. Wise, M.D., Matthew R. Baldwin, M.D., Lee Wu, M.H.S., Steven C. LeClerq, M.H.S., Parul Christian, Dr.P.H., Joanne Katz, Sc.D., James M. Tielsch, Ph.D., Subarna Khatry, M.D., and Alfred Sommer, M.D., M.H.S.

A bs t r ac t Background From the Division of Pulmonary and Critical Care, School of Medicine (W.C., R.A.W., M.R.B.), the Program in Global Disease Epidemiology and Control (W.C., J.K., J.M.T.) and the Center for Human Nutrition (K.P.W., L.W., S.C.L., P.C., J.K., J.M.T.), Department of International Health, and the Department of Epidemiology (A.S.), Bloomberg School of Public Health, Johns Hopkins University, Baltimore; and the Nepal Nutrition Intervention Project Sarlahi, National Society for the Prevention of Blindness, Kathmandu, Nepal (S.C.L., S.K.). Address reprint requests to Dr. Checkley at Johns Hopkins University School of Medicine, Division of Pulmonary and Critical Care, 1830 Monument St., 5th Fl., Baltimore, MD 21205, or at [email protected]. This article (10.1056/NEJMoa0907441) was last updated on December 29, 2010, at NEJM.org. N Engl J Med 2010;362:1784-94. Copyright © 2010 Massachusetts Medical Society.

Vitamin A is important in regulating early lung development and alveolar formation. Maternal vitamin A status may be an important determinant of embryonic alveolar formation, and vitamin A deficiency in a mother during pregnancy could have lasting adverse effects on the lung health of her offspring. We tested this hypothesis by examining the long-term effects of supplementation with vitamin A or beta carotene in women before, during, and after pregnancy on the lung function of their offspring, in a population with chronic vitamin A deficiency. Methods

We examined a cohort of rural Nepali children 9 to 13 years of age whose mothers had participated in a placebo-controlled, double-blind, cluster-randomized trial of vitamin A or beta-carotene supplementation between 1994 and 1997. Results

Of 1894 children who were alive at the end of the original trial, 1658 (88%) were eligible to participate in the follow-up trial. We performed spirometry in 1371 of the children (83% of those eligible) between October 2006 and March 2008. Children whose mothers had received vitamin A had a forced expiratory volume in 1 second (FEV1) and a forced vital capacity (FVC) that were significantly higher than those of children whose mothers had received placebo (FEV1, 46 ml higher with vitamin A; 95% confidence interval [CI], 6 to 86; FVC, 46 ml higher with vitamin A; 95% CI, 8 to 84), after adjustment for height, age, sex, body-mass index, calendar month, caste, and individual spirometer used. Children whose mothers had received beta carotene had adjusted FEV1 and FVC values that were similar to those of children whose mothers had received placebo (FEV1, 14 ml higher with beta carotene; 95% CI, −24 to 54; FVC, 17 ml higher with beta carotene, 95% CI, −21 to 55). Conclusions

In a chronically undernourished population, maternal repletion with vitamin A at recommended dietary levels before, during, and after pregnancy improved lung function in offspring. This public health benefit was apparent in the preadolescent years.

1784

n engl j med 362;19  nejm.org  may 13, 2010

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2017. For personal use only. No other uses without permission. Copyright © 2010 Massachusetts Medical Society. All rights reserved.

Maternal Vitamin A Supplementation and Lung Function in Offspring

V

itamin A deficiency affects 190 million preschool-aged children and 19 million pregnant women worldwide.1 It is the underlying cause of 650,000 early childhood deaths2 and has become recognized as an important problem among women of reproductive age in many developing countries. Chronic vitamin A deficiency may increase the risks of complications and death during pregnancy and in the postpartum period3-9 and, on the basis of evidence from studies in animals, may also adversely affect the embryonic and postnatal development of the offspring.10-14 The importance of vitamin A in regulating growth through cell proliferation and differentiation was recognized early in the 20th century.10-12 Results from animal research have since shown that vitamin A plays a key role in mediating fetal growth, morphogenesis, and maturation of multiple organ systems, including the respiratory system.14-20 Depletion of vitamin A from the diet of female rats before and during pregnancy is associated with agenesis or hypoplasia of the lungs in offspring, conditions that can be prevented with vitamin A supplementation in early, but not late, pregnancy.14 Furthermore, vitamin A depletion in pregnant rats has been associated with dosedependent decreases in DNA content in the lung tissue of their offspring.17,18 Since alveolarization begins in utero at about the 36th week of gestation,21 maternal vitamin A deficiency during pregnancy may have lasting effects on the lung maturation of progeny. Although results from studies in animals have shown that vitamin A is an important determinant of early lung development and size, data are lacking on the long-term consequences of vitamin A deficiency on lung health in human populations. We studied the effect of antenatal vitamin A supplementation on the lung function of preadolescent children in a chronically undernourished population in rural Nepal. The study cohort consisted of children, 9 to 13 years of age, whose mothers had participated in a randomized, placebo-controlled trial of vitamin A or beta-carotene supplementation before, during, and after pregnancy.

ern Nepal, in the densely populated, low-lying southern plains (Terai). The weather is usually warm and humid in this region, with temperatures exceeding 40°C in the hot, dry season (April through June), followed by a season of monsoon rains (July through October), and thereafter by a cooler, dry season (November through March). The Terai is an area of chronic undernutrition and vitamin A deficiency.22,23 Rice is the staple of the diet. It is supplemented with small amounts of seasonal fruits, vegetables, lentil soup, and occasionally meat, fish, and eggs. Original Vitamin A Trial

The original study, which was conducted between April 1994 and September 1997, was a doubleblind, placebo-controlled, cluster-randomized trial involving married women of childbearing age. The study was designed to determine the effects of weekly supplementation with a low dose of vitamin A or beta carotene on the rates of maternal death related to pregnancy.7 We invited all eligible women from 30 village development communities (VDCs) to participate in the trial. Each VDC is composed of 9 wards (for a total of 270 wards). The unit of block randomization was the ward, and each ward within a VDC was randomly assigned to one of the three study groups. A total of 44,646 women were enrolled and received weekly supplementation with 7000-μg retinolequivalents of vitamin A, 7000-μg retinol-equivalents of beta carotene, or placebo. Both supplements, as well as the placebo, were given in the form of gelatinous capsules taken orally. A total of 75% of the pregnant women received at least half the allowable dose (i.e., ≥50% of a dietary allowance in the case of those receiving vitamin A or beta carotene).7 We prospectively identified all pregnant women and followed the mothers and their infants for an assessment of vital status and health outcomes. Supplementation with vitamin A or beta carotene resulted in a reduction in the rates of maternal death related to pregnancy, from 704 per 100,000 pregnancies in the placebo group to 395 per 100,000 pregnancies in the combined vitamin A–beta-carotene groups (a 44% relative reduction with the supplements).7 Neither supplement had an effect on infant mortality.24 We enrolled a subgroup of pregnant women and Me thods their live-born infants from three VDCs (27 of the Study Setting 270 wards) in a substudy with a more detailed We conducted the original vitamin A trial and this protocol that involved interviews about illnesses, follow-up study in the Sarlahi District of south- diet, and other exposures; collection of blood n engl j med 362;19  nejm.org  may 13, 2010

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2017. For personal use only. No other uses without permission. Copyright © 2010 Massachusetts Medical Society. All rights reserved.

1785

The

n e w e ng l a n d j o u r na l

samples at mid-pregnancy and at 3 months post partum; clinical examinations; and anthropometric measurements. Serum retinol levels, assessed in the women post partum and in their infants at 3 months of age, were higher in the vitamin A group than in the placebo group and were moderately higher in the beta-carotene group than in the placebo group.7,24 A total of 2055 children were born alive to mothers in the subsample who completed the pregnancy-to-postpartum dosing protocols. Of these children, 1894 (92%) were alive at the end of the trial (September 30, 1997). Enrollment in the Follow-up Study

In 2006, we revisited the households of children whose mothers had participated in the original subsample study. Fieldworkers were unaware of the group assignments of the mothers in the original study. We used a household list derived from the original trial to generate an updated list of children who were eligible to participate in the follow-up study. Households in which the children were absent at the time of the visit were visited up to three times to maximize enrollment. The follow-up study was approved by the institutional review board at the Johns Hopkins University in Baltimore and at the Institute of Medicine, Tribhuvan University, in Kathmandu. We obtained oral or written informed consent from the mothers and assent from the children. Spirometric Assessments

The objective of the follow-up trial was to determine whether there were differences in the forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) among children according to the group assignment of their mothers in the original trial. We used 20 SpiroPro (JAEGER) spirometers during the study period. SpiroPro is a lightweight, battery-operated, portable pneumo­ tachometer with factory-precalibrated pneumo­ tachometer tubes. We used only one pneumotach­ ometer tube per participant. During a 6-month period before the start of the study, we trained 11 technicians and 3 supervisors in the performance of spirometry. We numbered all our spirometers and asked technicians to use a different spirometer each day. Technicians visited the study children at their homes to perform spirometry. Each child underwent spirometry while in a sitting position and wearing a nose clip, until three acceptable and reproducible maneuvers, of a maximum of eight, had been per1786

of

m e dic i n e

formed.25 Technicians reviewed with a supervisor all the flow-volume curves that were obtained each day. Flow-volume curves were transmitted weekly to Johns Hopkins University for additional review. Approximately every 3 months, we performed direct supervision and in-person review of all flowvolume curves with each technician. Statistical Analysis

The values for FEV1 and FVC were adjusted for height, age, sex, body-mass index, calendar month, and caste. Because biases can occur among different spirometers, we also adjusted for the specific spirometer that was used for the measurement. All analyses were performed according to the intention-to-treat principle. Because clustering by ward or VDC may have affected the estimation of standard errors, we used a linear mixed-effects model26 with two levels of random effects — VDC and ward within VDC — to account for the multilevel design of the trial. In subgroup analyses, we examined whether socioeconomic indicators or early exposures confounded the effects of the mother’s original study-group assignment on lung function. In a subgroup of mothers in whom serum retinol levels or serum beta-carotene levels were measured post partum, we examined the relationship between the postpartum levels of these micronutrients in the mothers and the lung function of their preadolescent children. We compared proportions across study groups and according to enrollment status, using standard methods.27 P values of less than 0.05 were considered to indicate statistical significance. We used the R statistical package (www.r-project.org) for analyses.

R e sult s Characteristics of the Study Population

The status of the 2055 live-born children from the original subsample is summarized in Figure 1. Of the 1894 children who were alive at the end of the original trial, 118 (6%) were no longer living in the study area at the time of the follow-up study, and 110 (6%) had died. No information was available for eight children (