Maternal medicine

DOI: 10.1111/j.1471-0528.2011.03158.x www.bjog.org

Changes in pulmonary function during pregnancy: a longitudinal cohort study G Grindheim,a K Toska,b M-E Estensen,c,d LA Rosselanda a Department of Anaesthesiology, Division of Critical Care, Oslo University Hospital, Rikshospitalet, Oslo, Norway b Department of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway c National Resource Centre for Women’s Health and d Department of Cardiology, Oslo University Hospital, Rikshospitalet, Oslo, Norway Correspondence: Dr G Grindheim, Department of Anaesthesiology, Division of Critical Care, Oslo University Hospital HF, Rikshospitalet, Postboks 4950, Nydalen, N-0424 Oslo, Norway. Email [email protected]

Accepted 21 August 2011. Published Online 18 October 2011.

Objective To record any physiological changes in lung function

during healthy pregnancies, and evaluate the influence of parity, pregestational overweight, and excessive weight gain. Design Longitudinal cohort study. Setting Antenatal clinic at Oslo University Hospital. Population One hundred healthy white women with singleton

pregnancies. Methods The women were studied with repeated measures

of lung function using spirometry at a gestational age of 14–16, 22–24, 30–32, and 36 weeks, and at 6 months postpartum. Main outcome measures Forced vital capacity (FVC),

forced expiratory volume in 1 second (FEV1), and peak expiratory flow (PEF), also expressed as a percentage of predicted values according to age and height: i.e. FVC%, FEV1%, and PEF%.

Results Both FVC and FVC% increased significantly after 14– 16 weeks of gestation (P = 0.001), as was the case for both PEF and PEF% (P < 0.001). FVC, FVC%, PEF, and PEF% in early and mid-pregnancy were significantly lower compared with the postpartum value (all P < 0.05). Nulliparous women had an overall 4.4% lower value of FVC% than parous women (P = 0.039). There were no differences in FVC, FEV1, or PEF dependent upon pregestational overweight or excessive weight gain. Conclusions Forced vital capacity (FVC) increases significantly

after 14–16 weeks of gestation. The FVC% is significantly higher in parous compared with primigravida women, suggesting that the changes in FVC occurring during pregnancy persist postpartum. PEF increases significantly during healthy pregnancies, and should be interpreted cautiously in pregnant women with impaired lung function. Keywords Longitudinal study, lung function during pregnancy,

spirometry.

Please cite this paper as: Grindheim G, Toska K, Estensen M, Rosseland L. Changes in pulmonary function during pregnancy: a longitudinal cohort study. BJOG 2012;119:94–101.

Introduction In pregnancy, hormonal changes and the progressive increase in abdominal volume may have mechanical and functional impact on respiratory function. However, an increased transverse diameter of the chest, resulting from a widened subcostal angle, opposes the effect of the enlarging pregnant uterus and elevated diaphragm, leaving pulmonary function altered, but not compromised, during pregnancy.1 Previous studies evaluating the effect of pregnancy on pulmonary function have shown that both minute ventilation (VE) and tidal volume (VT) are increased, whereas the functional residual capacity (FRC) and expiratory reserve volume (ERV) are decreased.2–4 The values obtained by

94

forced spirometry, including forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), and peak expiratory flow (PEF) have largely been found to remain unchanged during pregnancy.2,3,5–9 In other studies, PEF is found to decrease with advancing gestational age and to be affected by maternal positioning,10,11 and by living at high altitude.12 Earlier studies addressing changes in pulmonary function during pregnancy have various methodological weaknesses, such as small sample size,8 cross-sectional study design,3,7,12 or insufficient statistical methods,4,5,8 which may have limited the validity of their conclusions. When seeking to describe the natural course of physiological changes in lung function during pregnancy, the study design and choice of statistical methods and tests are of importance. A cross-sectional study design cannot provide

ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG

Longitudinal changes in pulmonary function during pregnancy

data on individual changes over time, which limits the ability to elucidate the influence of advancing gestational age on the variables of interest. A longitudinal study design with repeated measures of the variables of interest throughout pregnancy, although time-consuming and prone to withdrawal, allows the observation of changes during pregnancy. An analysis of longitudinal data should be performed using statistical methods that take into account the dependency between repeated measures from the same subject. Suboptimal pulmonary function in pregnancy has been associated with adverse pregnancy outcome. Pulmonary disease can affect pregnancy outcome and pregnancy can affect the course of pulmonary disease. The pregnancies of women with asthma are more likely to be complicated by pre-eclampsia, preterm birth, and lower birthweight than pregnancies in non-asthmatic women.13 Studies have reported a direct relationship between maternal FEV1 during pregnancy and infant birthweight,14,15 and an inverse relationship with intrauterine growth retardation,14 gestational hypertension, and preterm birth in asthmatic women.15 In pregnant women with cystic fibrosis, low FEV1 is associated with preterm delivery,16–18 and with a greater loss of lung function during pregnancy.17,18 As a consequence, pregnant women with pulmonary disease need regular monitoring of symptoms and measures of lung function by spirometry in order to optimise their lung function throughout pregnancy. Hence, understanding pulmonary changes in pregnancy through the evaluation of spirometry in normal pregnancy is of major clinical importance when facing pregnant women with pulmonary disease. Pregestational overweight and excessive weight gain during pregnancy are both recognised as being associated with an increased risk of maternal complications during pregnancy.19,20 It is well established that overweight individuals have a decrease in FVC and FEV1.21 The effect of pregestational overweight and excessive weight gain on the values obtained by forced spirometry during pregnancy is to our knowledge not yet described. Based on these considerations, we endeavoured to perform a more extensive study with repeated measures of healthy pregnant women in order to provide pertinent data on the physiological changes in lung function during pregnancy. Furthermore, we sought to evaluate the influence of parity, pregestational overweight, and excessive weight gain on lung function during pregnancy.

Methods A prospective longitudinal study of respiratory function was performed in 100 healthy women with singleton pregnancies. Exclusion criteria were asthma or other self-

reported pulmonary disease, current tobacco use, hypertension (i.e. systolic/diastolic blood pressure > 140/90 mmHg), or other cardiovascular disease. The women were without current respiratory infection at the time of each measurement. The women were recruited by an open invitation to all patients at the antenatal clinic at Oslo University Hospital – Rikshospitalet from August 2007 to June 2008. The follow-up period lasted until May 2009. The study was approved by The Regional Medical Research Ethics Committee for Southern Norway, and the women gave their written informed consent to participation. Respiratory function was measured repeatedly: four times during pregnancy (14–16, 22–24, 30–32, and 36 weeks of gestation), and at 6 months postpartum, thus allowing the women to serve as their own controls. Gestational age was estimated from an ultrasonography test, performed up to 18 weeks of gestation. Spirometry data, including FVC, FEV1, and PEF were recorded using a Vmax 22 (Sensor Medics; SensorMedics Corp., Yorba Linda, CA, USA). The volume signal of the equipment was calibrated once daily with a 3-L syringe. Tests were performed under calm conditions with the subject in a sitting position according to American Thoracic Society (ATS) guidelines.22 After oral instruction the subjects exhaled forcefully until three acceptable curves were obtained. The highest values achieved were selected for analysis. The obtained values were also expressed as a percentage of predicted value according to age and height, using the European Community of Coal and Steel (ECCS) reference equation.23 These values are referred to as FVC%, FEV1%, and PEF%, respectively. Pregestational body mass index (BMI) and gestational weight gain were recorded and classified according to the Institute of Medicine (IOM) guidelines.24 A pregestational BMI > 25 and weight gain higher than recommended were registered as dichotomous variables. Parity was also recorded as a dichotomous variable: nulliparous or parous (having one or more children). Data analysis was carried out using spss 16 (SPSS Inc., Chicago, IL, USA). Data normality was investigated using the spss exploration option with summary statistics and graphical display. A linear mixed model with repeated measures was used in order to describe the changes occurring during pregnancy. Model selection was performed for each variable (FVC, FVC%, FEV1, FEV1%, PEF, and PEF%) by choosing the model achieving the lowest information criteria: )2 restricted log likelihood and Akaike’s information criterion. Time was introduced as the repeated effect, and dependencies in the data were handled by an unstructured covariance matrix. Time, parity, pregestational BMI over 25, and excessive weight gain during pregnancy were explanatory factors, and were treated as fixed effects. The significance of each partial effect was evaluated by F-tests. When relevant, the pairwise comparisons between

ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG

95

Grindheim et al.

the estimated means were adjusted for multiple comparisons using the Bonferroni method. We increased the sample size considerably compared with previous longitudinal studies, as we aimed at having sufficient statistical power to analyse the variables of interest (time, parity, pregestational overweight, and excessive weight gain). A post hoc power calculation based on the observed effects of time on FVC% shows that the study has a power of approximately 90% for detecting an effect such as the one observed.

Patients included during study periode N = 100

Reasons for non-participation Lost upon first measurement: Miscarriage (2) Lost after first measurement: Lost to follow-up (6) Intrauterine fetal disease (2) Late miscarriage (2) Pregnancy related depression (1)

Patients included in statistical analyses N = 87

Results A total of 115 women were invited to participate in the study, of which 100 women agreed to participate and were included. The repeated measurements of 87 women were analysed. The statistical analyses were performed based on measurements repeated on five occasions in 87 women. Figure 1 shows a flow diagram illustrating the number of data analysed at each consultation and the various reasons for non-participation. The main characteristics of the study population are shown in Table 1. Five of the pregnant women developed late-onset pre-eclampsia (PE), one woman developed gestational diabetes, one woman experienced a stillbirth at 35 weeks of gestation, and two women had a premature delivery at 35 weeks of gestation. These women are included in the analyses. The repeated measures of the women who developed late-onset PE were analysed separately. There were no statistically significant differences between the pulmonary function of these five women and the women with uncomplicated pregnancies. The mean values of FVC, FEV1, and PEF, and their corresponding values expressed as the percentage of the predicted values, were within normal limits at all times in pregnancy (Table 2). FVC and FVC% increased significantly during pregnancy (P = 0.001), and were significantly lower in early and mid pregnancy compared with the postpartum value. Furthermore, nulliparous women had an overall 4.4% lower FVC% than parous women (P = 0.039; Figure 3). PEF and PEF% also increased during pregnancy (both P < 0.001). The mean difference between the postpartum PEF and that at 14–16 weeks of gestation was 0.44 l/second, with a 25–75 percentile range of )0.03 to 0.87 l/second. Forced expiratory volume in 1 second and FEV1% did not show any alterations during pregnancy compared with the postpartum values. Values for FEV1 and PEF did not show any differences between nulliparous and parous women. Figure 2 demonstrates the observed changes in FVC, FEV1, and PEF during pregnancy. We did not find any differences in FVC, FEV1, or PEF depending on pregestational overweight or excessive weight gain during pregnancy (Table S1).

96

Gestational age 15 weeks * N = 85 [87]

Missing at this examination (n = 2): Late inclusion (1) Unsatisfactory quality of data (1)

Gestational age 23 weeks* N = 87 [87]

Gestational age 31 weeks* N = 82 [87]

Lost (n = 3): Lost to follow-up (3) Missing at this examination (n = 2): Intercurrent illness (2)

Gestational age 36 weeks* N = 80 [87]

Lost (n = 4): Lost to follow-up (+1) Missing at this examination (n = 3): Premature labour (2) Intrauterine fetal death (1)

6 months post partum* N = 80 [87]

Lost (n = 4): Missing at this examination (n = 3): Pregnant (1) Postpartum depression (1) Intercurrent illness (1)

Figure 1. Flowchart of patient participation and follow-up throughout the study period. Subjects lost to follow-up between the first and second measurements were not included in the statistical analyses. The count at each time point represents the number of subjects examined. The counts in square brackets [] represents the number of subjects whose data were included in the statistical analyse. *Time of examination according to the mean value for the study population.

Discussion The main finding is that FVC and PEF increase progressively after 14–16 weeks of gestation. As the parous women in our study had an overall significantly higher FVC% than the primigravida women, the increase in FVC occurring during pregnancy may be permanent. Although the magnitude of increase itself may be small, our findings may have

ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG

Longitudinal changes in pulmonary function during pregnancy

Table 1. Basic characteristics of the study population (n = 87) Number (%)

Maternal characteristics Age (years) Height (cm) Pregestational weight (kg) Pregestational BMI (kg/m2) Pregestational BMI < 18.5 kg/m2 Pregestational BMI 18.5–24.9 kg/m2 Pregestational BMI 25.0–29.9 kg/m2 Pregestational BMI ‡ 30 kg/m2 Pregestational BMI > 25 kg/m2 Weight gain during pregnancy (kg) Weight gain over recommended Gravidity Para 0 Para 1+ Para 1 Para 2 GA at first measurement (weeks) GA at second measurement (weeks) GA at third measurement (weeks) GA at fourth measurement (weeks) Postpartum measurement (months) Pregnancy outcome Vaginal delivery Caesarean section (elective/urgent) Fetal birthweight (g) Fetal Apgar score at 1 minute/5 minutes

Mean or median*

SD

Range or 10–90 percentiles**

31.4 167.7 65.1 23.1

4.3 5.2 11.3 3.5

21–40 155–180 50–118 18.3–37.0

14.8

5.5

2–32

1.7

0.9

1–5

4 (4.6) 66 (75.9) 13 (14.9) 4 (4.6) 17 (19.5) 28 (32.2) 51 (58.6) 36 (41.4) 31 (35.6) 5 (5.8) 15* 23* 31* 36* 6.0*

14–17** 22–24** 30–32** 35–37** 5.5–6.5**

74 (85.1) 6/7 (6.9/8.0) 3526 9/9*

603

BMI, body mass index; GA, gestational age.

Table 2. Forced spirometry values during pregnancy and postpartum GA 15 weeks (n = 85) FVC (l) FVC nulliparous FVC parous FVC% FVC% nulliparous FVC% parous FEV1 (l) FEV1% PEF (l/second) PEF%

3.89 3.84 3.96 104.5 102.9 106.8 3.18 98.2 6.71 93.5

± ± ± ± ± ± ± ± ± ±

0.48** 0.43 0.55 10.6** 9.7 11.4 0.44 11.1 1.19* 15.5*

GA 23 weeks (n = 87) 3.92 3.89 3.98 105.4 104.0 107.4 3.16 97.6 6.92 96.5

± ± ± ± ± ± ± ± ± ±

0.48*** 0.45 0.52 11.0*** 10.5 11.5 0.39 10.0 1.13*** 14.9***

GA 31 weeks (n = 82)

GA 36 weeks (n = 80)

3.96 3.92 4.02 106.6 105.1 108.8 3.20 99.1 7.19 100.3

4.00 3.93 4.08 107.6 105.6 110.1 3.21 99.3 7.24 101.0

± ± ± ± ± ± ± ± ± ±

0.51 0.49 0.53 11.6 11.8 11.1 0.43 11.4 1.10 13.8

± ± ± ± ± ± ± ± ± ±

0.53 0.47 0.59 11.9 11.1 12.7 0.43 11.3 1.15 14.6

6 months postpartum (n = 80) 4.00 3.97 4.06 107.7 106.1 110.0 3.20 98.9 7.18 100.2

± ± ± ± ± ± ± ± ± ±

0.51 0.51 0.51 11.9 12.7 10.3 0.41 10.9 1.05 13.1

FEV1, forced expiratory volume in 1 second; FEV1%, forced expiratory volume in 1 second, expressed as a percentage of the predicted value; FVC, forced vital capacity; FVC%, forced vital capacity expressed as a percentage of the predicted value; GA, median gestational age at measurement; PEF, peak expiratory flow; PEF%, peak expiratory flow expressed as a percentage of the predicted value. Values are means ± SDs. *P < 0.001; **P < 0.01; ***P < 0.05 for the comparison with postpartum value.

ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG

97

Grindheim et al.

PEF

7.50

L/s

7.25

¤

7.00

*

6.75 6.50

4.15

FVC

L

4.05

¤

*

3.95 3.85 3.75

3.35

FEV1

L

3.25 3.15

m 6

ee

ks

on

th

s

63

2 –3 W

ks

30

–2 ee W

ks ee W

W

ee

ks

14

22

–1

6

4

3.05

Figure 2. Changes in forced vital capacity (FVC), peak expiratory flow (PEF), and forced expiratory volume in 1 second (FEV1) during pregnancy and postpartum. Data are expressed as means ± SE. Significant changes: *P < 0.01, compared with the postpartum value; ¤P