OBJECTIVE: To estimate whether the risk of recurrent preeclampsia is affected by interpregnancy change in body mass index (BMI).
METHODS: We conducted a population-based cohort study using Missouri maternally linked birth certificates for 17,773 women whose first pregnancies were complicated by preeclampsia. The women were placed into three groups: those who decreased their BMIs, those who maintained their BMIs, and those who increased their BMIs between their first two pregnancies. The primary outcome was recurrent preeclampsia in the second pregnancy. Adjusted risk ratios and 95% confidence intervals were calculated using Poisson regression analysis.
RESULTS: The overall rate of recurrent preeclampsia in women who decreased their BMIs between pregnancies was 12.8% (risk ratio 0.70, confidence interval 0.60–0.81) compared with 14.8% if BMI was maintained and 18.5% in those who increased their BMIs (risk ratio 1.29, confidence interval 1.20–1.38). Within the normal weight, overweight, and obese weight categories, women who decreased BMI between pregnancies were less likely to experience recurrent preeclampsia. Women in all weight categories who increased their BMIs between pregnancies were more likely to experience recurrent preeclampsia.
CONCLUSION: Interpregnancy weight reduction decreases the risk of recurrent preeclampsia and should be encouraged in women who experience preeclampsia.
LEVEL OF EVIDENCE: II
Women who lose weight between pregnancies reduce their risk of recurrent preeclampsia, and women who gain weight increase their risk of recurrent preeclampsia.
From the Saint Louis University School of Medicine, Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology, and Women's Health, and School of Public Health, Department of Community Health, St. Louis, Missouri.
The authors acknowledge the Missouri Department of Health and Senior Services, Section of Epidemiology for Public Health Practice, as the original source of the data.
The analysis, interpretations, and conclusions in the present study are those of the authors and not the Missouri Department of Health and Senior Services, Section of Epidemiology for Public Health Practice.
Presented at the 27th Annual Meeting of the Society for Maternal-Fetal Medicine, February 5–10, 2007, San Francisco, California.
Corresponding author: Dorothea Mostello, MD, Division of Maternal-Fetal Medicine, 6420 Clayton Road, St. Louis, MO 63117; e-mail: email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
Our previous work has demonstrated that the strongest risk factors for recurrence of preeclampsia are maternal prepregnancy body mass index (BMI) and the gestational age at delivery for the first pregnancy complicated by preeclampsia.1 Other factors identified to increase the risk of preeclampsia in the parous woman, such as race, preexisting medical conditions, interbirth interval, change in paternity, previous fetal loss, and previous growth restriction,2 do not contribute to the risk of recurrence.1 Although maternal smoking seems to be protective,1,3 body weight remains the most clinically important modifiable risk factor for recurrent preeclampsia.1 Because of the potential to avoid the perinatal morbidity and mortality associated with preeclampsia,4 determining whether reducing body weight decreases the incidence of recurrent preeclampsia is paramount.
Many studies demonstrate a relationship between prepregnancy body weight and preeclampsia.5–7 In a systematic review of controlled cohort studies, the risk of preeclampsia doubled with each 5- to 7-point increase in prepregnancy BMI.7 Bodnar et al6 demonstrated a “dose-dependent” relationship between maternal adiposity and preeclampsia risk among nulliparous women without preexisting conditions, showing that the risk of preeclampsia increases even within traditional BMI categories.
Several authors8,9 show alterations in various risks with interpregnancy weight change, but this concept has not been applied to recurrence of preeclampsia. We sought to examine whether interpregnancy weight gain or loss affected the risk of recurrent preeclampsia.
MATERIALS AND METHODS
We conducted a population-based, retrospective cohort study using data from Missouri maternally linked birth and fetal death certificates. Our study population consisted of resident mothers who delivered their first two singleton pregnancies at more than 20 weeks of gestation between January 1, 1989, and December 31, 2005. The variable “clinical estimate of gestational age,” a more accurate representation of gestational age at delivery than length of pregnancy calculated using last menstrual period,10 became a required field on the birth certificate in 1989. The last year of available data from the cohort is 2005. Only singleton births were included to eliminate the confounding effects of multiple gestation on pregnancy duration, birth weight, and likelihood of preeclampsia.
Our study population comprised women with preeclampsia during their first pregnancy; “hypertension, pregnancy-induced (preeclampsia)” or “eclampsia” checked on the birth certificate served to satisfy this condition. Preeclampsia is defined in the Missouri birth certificate registry as the development of hypertension (blood pressure of 140/90 or higher or an increase of 30 mmHg systolic or 15 mmHg diastolic over baseline values on at least two occasions at least 6 hours apart) plus proteinuria or edema that is generalized and overt. Eclampsia is defined as preeclampsia with the occurrence of convulsions, coma, or both of these. Because of its infrequency, eclampsia was combined with preeclampsia. Recurrent preeclampsia was the outcome of interest; women with preeclampsia noted on both the first and second birth certificates met the criteria for recurrent preeclampsia.
Our exposure variable, the difference in prepregnancy BMI, calculated as weight (kg)/[height (m)]2, between the first two births, was computed for all participants. The participants were placed into three groups based on a change in BMI: women who decreased their BMIs by more than two units, women who increased their BMIs by more than two units, and women who maintained their BMIs (ie, BMI did not change by more than 2 units). Those who maintained their BMIs between births were selected as the reference group. Self-reported maternal prepregnancy weight and height values were used to calculate prepregnancy BMI.
Factors previously associated with recurrent preeclampsia were evaluated from information available from the birth certificate. Maternal demographic covariates included maternal age, prepregnancy BMI category, cigarette use during pregnancy, and socioeconomic status. Medicaid enrollment status served as a proxy for socioeconomic status. BMI was categorized as underweight (BMI less than 18.5), normal weight (BMI 18.5–24.9), overweight (BMI 25–29.9), and obese class I (BMI 30–34.9), obese class II (BMI 35–39.9), and obese class III (BMI 40 or higher). Medical covariates included were history of chronic hypertension, diabetes, and renal disease. Obstetric factors included were gestational age of newborn at the first delivery (20–28, 29–32, 33–36, and 37 weeks or more), interval between pregnancies, and average weekly gestational weight gain (in pounds) in the second pregnancy. The interval between pregnancies was calculated as the time (in years) from the birth date of the first pregnancy to the conception of the second pregnancy (estimated from clinical gestational age) and was categorized as less than 1 year, 1 to up to 2 years, 2 to 4 years, and more than 4 years. The average weekly gestational weight gains were calculated as the difference between maximum weight during pregnancy minus prepregnancy weight (self-reported) divided by gestational age.
Differences in sample characteristics among the three groups (decreased BMI, maintained BMI, and increased BMI) were assessed using the Pearson χ2 test for categorical variables and analysis of variance for continuous variables. A Poisson regression model with robust error variance was constructed to estimate relative risk and 95% confidence intervals for recurrent preeclampsia to avoid the pitfall of using the odds ratio to estimate the risk ratio when the outcome is common (more than 10%).11 To reduce the bias in the estimation of risk, potential confounders, including maternal demographic, medical, and obstetric factors, were included in the multivariable analysis. We compared the crude rates of recurrent preeclampsia by interpregnancy BMI unit change and tested for trend using χ2 test for trend.12 All tests were two-tailed, and P less than .05 was considered significant. All statistical analyses were performed with STATA 10.0.
This research did not involve interaction with individuals, nor did the dataset contain identifiable private information. Therefore, according to federal regulation 45 CFR 46.102, the Saint Louis University Institutional Review Board deemed that this study did not require formal review.
From 1989 through 2005 in the Missouri maternally linked birth registry, 19,044 women had preeclampsia diagnosed during their first pregnancy. Of these, information required to calculate the prepregnancy BMI was present for both first and second pregnancies in 17,913 women. Information for all variables considered in Tables 1, 2, and 3 were available for 17,773 women. This cohort comprised the final study population.
Tables 1 and 2 describe the demographic, lifestyle, medical, and obstetric characteristics of women who decreased, maintained, and increased their BMIs. Although some observed differences are statistically significant, a few clinically meaningful patterns emerge. Women who decreased their BMIs between pregnancies were more likely to be obese, smoke, have had diabetes, and delivered at less than 28 weeks of gestation during the first pregnancy and gained more weight in the second pregnancy. Women who increased their BMIs were more likely to be younger and receive Medicaid during the first pregnancy, have a pregnancy interval more than 4 years, and gained less weight during the second pregnancy.
The rate of recurrent preeclampsia overall was 16.3%. The rate in women who decreased their BMIs between pregnancies was 12.8% compared with 14.8% if BMI was maintained, and 18.5% in those who increased their BMIs. Figure 1 shows the risk (crude rates) of recurrent preeclampsia based on weight change between pregnancies, stratified by BMI before the first pregnancy. In normal weight, overweight, and obese women, a decrease in BMI was associated with a decreased risk of recurrent preeclampsia. In each weight category, an increase in BMI was associated with an increased risk of recurrent preeclampsia. Figure 2 shows the crude risk for recurrent preeclampsia by interpregnancy BMI unit change. Specifically, there was a “dose-response” with a linear increase in risk of recurrent preeclampsia with increasing interpregnancy weight gain (P for χ2 test for trend <.001).
Table 3 presents the univariable and multivariable analyses in the association among change in BMI, risk of recurrent preeclampsia, and each covariate considered in our analysis. When controlled for other factors, compared with women who maintained their BMIs, those who increased their BMIs increased their risk of preeclampsia recurrence, and those who decreased their BMIs reduced their risk of preeclampsia recurrence. The multivariable analysis confirmed the importance of prepregnancy BMI and gestational age at delivery of the first pregnancy in the likelihood of recurrence. Other important contributors were the presence of diabetes, weight gain during the second pregnancy, and interpregnancy weight change.
Our results demonstrate that relatively small interpregnancy changes in maternal weight, even within BMI categories, significantly alter the risk of preeclampsia recurrence. A change in BMI by two units represents approximately a 10-pound weight change for an overweight woman of average height. Thus, women may be able to lower their risk of preeclampsia recurrence with modest, generally achievable degrees of weight loss before pregnancy.
In general, the effects of weight loss on development of preeclampsia have not been adequately studied. In Villamor's study9 in an unselected population, weight loss of more than one BMI unit was associated with a decreased risk of preeclampsia. Contrary to these findings, previously normotensive obese and overweight women who lost enough weight to change BMI categories remained at increased risk for preeclampsia in a second pregnancy.8 In contrast, our study in a very high-risk population showed clear evidence that weight loss is associated with decreased risk of preeclampsia recurrence.
Weight gain during pregnancy is associated with development of preeclampsia.13,14 Obese women with low gestational weight gain have a lower risk of preeclampsia,14 whereas the risk of preeclampsia increases with increased gestational weight gain.13 Cedergren14 showed that average and overweight women with excessive weight gain doubled their risk of preeclampsia. Our results confirm this association among women with previous preeclampsia. Importantly, the association remains robust in our study when controlled for other factors.
The biologic mechanisms underlying the association of interpregnancy weight change with recurrent preeclampsia are not completely understood. Endothelial dysfunction, which characterizes preeclampsia, is postulated to be part of an exaggerated maternal inflammatory response to pregnancy.15 Markers of inflammation are associated with preeclampsia and are present before clinical symptoms manifest.15,16 Obesity is also associated with elevated markers of inflammation, and these shared features have been used to support the concept of obesity as a risk factor for preeclampsia.5,16 Although some markers of inflammation in preeclampsia are primarily related to maternal BMI and insulin resistance, others are independent of obesity.15,17 The effect of weight change on preeclampsia risk may be mediated through changes in systemic inflammation because weight gain is associated with increased inflammation18 and weight loss has been shown to improve inflammation.19
The strengths of our population-based study include access to a large cohort of women with preeclampsia during their first pregnancy and the availability of information on significant confounding variables that affect risk of preeclampsia recurrence. The use of a cohort design allowed calculations of absolute risks of preeclampsia in the second pregnancy according to weight change. Limitations are inherent in the use of birth certificate data, including the potential for inaccurate reporting, inclusion of self-reported information, and underreporting of medical risk factors, such as chronic hypertension, diabetes mellitus, or renal disease. The birth certificate does not contain information on family history of preeclampsia or severity of preeclampsia. Although these are recognized risk factors for preeclampsia,4,20 we were unable to analyze them as possible confounders of recurrence. Fortunately, the reporting of preeclampsia on birth certificates with a check-box format (such as that used in Missouri) is excellent, ranging from 85% to 97% when compared with risks based on hospital discharge data (Schramm W. Data quality: new certificates. Presentation at AVRHS/VSCP Project Directors Meeting, San Francisco, CA, 1991).21 Similarly, self-reported maternal weights, on which our calculations of BMI and gestational weight gain are based, are highly correlated with and similar to clinically recorded weights.22
For women who face the prospect of recurrent preeclampsia, with its attendant risks of maternal and perinatal morbidity, mortality, and indicated preterm delivery, few options exist that may ameliorate their risk. Certainly, control of underlying medical illnesses, when present, is warranted before conception. Low-dose aspirin therapy4 and prudent weight gain during pregnancy may provide some benefit.13,23 Interpregnancy weight loss, however, is the most reasonable measure to meaningfully lower the risk of preeclampsia recurrence. The loss of at least two units of BMI is achievable with most lifestyle intervention programs.24,25 Prospective studies targeting women with previous preeclampsia can be designed to explore the physiology and effect of interpregnancy weight loss on the risk of recurrent preeclampsia and to substantiate our findings. Meanwhile, based on this evidence, women with preeclampsia during their first pregnancy should be counseled postpartum to actively engage in a weight management program. Because weight control may potentially influence the risk for recurrent preeclampsia, the recommendation to lose weight (for all but underweight women) before conceiving the next pregnancy can be made with assurance.
1. Mostello D, Kallogieri D, Tungsiripat R, Leet T. Recurrence of preeclampsia: effects of gestational age at delivery of the first pregnancy, body mass index, paternity, and interval between births. Am J Obstet Gynecol 2008;199:55.e1–7.
2. Mostello D, Catlin TK, Roman L, Holcomb WL Jr, Leet T. Preeclampsia in the parous woman: who is at risk? Am J Obstet Gynecol 2002;187:425–9.
3. Stone CD, Diallo O, Shyken J, Leet T. The combined effect of maternal smoking and obesity on the risk of preeclampsia. J Perinat Med 2007;35:28–31.
4. Barton JR, Sibai BM. Prediction and prevention of recurrent preeclampsia. Obstet Gynecol 2008;112:359–72.
5. Bodnar LM, Catov JM, Klebanoff MA, Ness RB, Roberts JM. Prepregnancy body mass index and the occurrence of severe hypertensive disorders of pregnancy. Epidemiology 2007;18:234–9.
6. Bodnar LM, Ness RB, Markovic N, Roberts JM. The risk of preeclampsia rises with increasing prepregnancy body mass index. Ann Epidemiol 2005;15:475–82.
7. O'Brien TE, Ray JG, Chan WS. Maternal body mass index and the risk of preeclampsia: a systematic overview. Epidemiology 2003;14:368–74.
8. Getahun D, Ananth CV, Oyelese Y, Chavez MR, Kirby RS, Smulian JC. Primary preeclampsia in the second pregnancy: effects of changes in prepregnancy body mass index between pregnancies. Obstet Gynecol 2007;110:1319–25.
9. Villamor E, Cnattingius S. Interpregnancy weight change and risk of adverse pregnancy outcomes: a population-based study. Lancet 2006;368:1164–70.
10. Callaghan WM, Dietz PM. Differences in birth weight for gestational age distributions according to the measures used to assign gestational age. Am J Epidemiol 2010;171:826–36.
11. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6.
12. Armitage P, Berry G, Matthews JNS. Statistical methods in medical research. 4th ed. Oxford (UK): Blackwell Science; 2002.
13. Kiel DW, Dodson EA, Artal R, Boehmer TK, Leet TL. Gestational weight gain and pregnancy outcomes in obese women: How much is enough? Obstet Gynecol 2007;110:752–8.
14. Cedergren M. Effects of gestational weight gain and body mass index on obstetric outcome in Sweden. Int J Gynecol Obstet 2006;93:269–74.
15. Wolf M, Kettyle E, Sandler L, Ecker JL, Roberts J, Thadhani R. Obesity and preeclampsia: the potential role of inflammation. Obstet Gynecol 2001;98:757–62.
16. Walsh SW. Obesity: a risk factor for preeclampsia. Trends Endocrinol Metab 2007;18:365–70.
17. Girouard J, Giguere Y, Moutquin JM, Forest JC. Previous hypertensive disease of pregnancy is associated with alterations of markers of insulin resistance. Hypertension 2007;49:1056–62.
18. Fogarty AW, Glancy C, Jones S, Lewis SA, McKeever TM, Britton JR. A prospective study of weight change and systemic inflammation over 9 y. Am J Clin Nutr 2008;87:30–5.
19. Forsythe LK, Wallace JM, Livingstone MB. Obesity and inflammation: the effects of weight loss. Nutr Res Rev 2008;21:117–33.
20. Duckitt K, Harrington D. Risk factors for preeclampsia at antenatal booking: systematic review of controlled studies. BMJ 2005;330:565–71.
21. Frost F, Starzyk P, George S, McLaughlin JF. Birth complication reporting: the effect of birth certificate design. Am J Public Health 1984;74:505–6.
22. Lederman SA, Paxton A. Maternal reporting of prepregnancy weight and birth outcome: consistency and completeness compared to clinical record. Mat Child Health J 1998;2:123–6.
23. Crane JM, White J, Murphy P, Burrage L, Hutchens D. The effect of gestational weight gain by body mass index on maternal and neonatal outcomes. J Obstet Gynaecol Can 2009;31:28–35.
24. Sacks FM, Bray GA, Carey VJ, Smith SR, Ryan DH, Anton SD, et al. Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N Engl J Med 2009;360:859–73.
© 2010 by The American College of Obstetricians and Gynecologists.
25. O'Toole ML, Sawicki MA, Artal R. Structured diet and physical activity prevent postpartum weight retention. J Wom Health 2003;12:991–8.