Cedergren, Marie I. MD, PhD
From the Division of Obstetrics and Gynecology, Department of Molecular and Clinical Medicine, Faculty of Health Sciences, Linköping University, Linköping, Sweden.
See related editorial on page 743.
Supported by the Östergötland County Council.
The author thanks the National Board of Health and Social Welfare, Stockholm, for access to the health registers.
Corresponding author: Marie I. Cedergren, Division of Obstetrics and Gynecology, University Hospital, SE-581 85 Linköping, Sweden; e-mail: firstname.lastname@example.org.
Financial Disclosure The author has no potential conflicts of interest to disclose.
OBJECTIVE: To establish optimal gestational weight gain for each maternal body mass index (BMI) category based on significant risk estimates of adverse maternal and fetal outcome.
METHODS: The study population consisted of 298,648 singleton pregnancies delivered in Sweden between January 1, 1994, and December 31, 2004. The number of individuals in each weight gain class was compared with the number of individuals in all other weight gain classes in the same BMI group with regard to adverse maternal and fetal outcome. Odds ratios were calculated after suitable adjustments.
RESULTS: The optimal gestational weight gain in women by prepregnancy BMI was 9–22 lb (4–10 kg) for BMI less than 20; 5–22 lb (2–10 kg) for BMI 20–24.9; less than 20 lb (less than 9 kg) for BMI 25–29.9; and less than 13 lb (less than 6 kg) for BMI of 30 or more.
CONCLUSION: The gestational weight gain limits for BMI categories determined in this large population-based cohort study from Swedish Medical Registers showed that a decreased risk of adverse obstetric and neonatal outcomes was associated with lower gestational weight gain limits than was earlier recommended, especially among obese women.
LEVEL OF EVIDENCE: II
There is as yet no general agreement on the recommended weight gain for pregnant women in different prepregnancy body mass index (BMI) classes. The available guidelines are based on pregnancies with “good outcomes.” A good outcome is often defined as a vaginal delivery at term of a live infant with a normal birth weight to a mother without prenatal complications. The most widely adopted recommendations concerning pregnancy weight gains are those of the Institute of Medicine (IOM), published in 1990.1 The subcommittee concluded that the target range for desirable maternal weight gain should be based on prepregnancy weight for height and should include the mean weight gain for women delivering full-term infants weighing between 3 and 4 kg. Because the observed range for such mothers was too broad to be used clinically, the subcommittee used its judgment in setting narrower target ranges by weight for height categories.
The broad ranges of weight gain among women with good outcomes were illustrated in a study from California where only 40% of women with good pregnancy outcomes had total gestational weight gains within the IOM guidelines.2 After establishing the IOM weight gain recommendations with different cutoff limits for each maternal BMI class, further research has to a large extent consisted of efforts to test the meaningful relationship of these guidelines to important clinical outcomes. The majority of these studies found that weight gains within the IOM's ranges were associated with better pregnancy outcomes (birth weight, preterm delivery, Apgar score, meconium aspiration syndrome, hypoglycemia, and cesarean delivery) than weight gains outside these ranges.3–7
Relatively little is known about maternal and fetal outcomes along the entire spectrum of weight change during pregnancy. Improved knowledge about this might result in other BMI-specific weight gain ranges for pregnant women. The aim of this study was to evaluate the risk of adverse maternal and fetal outcomes along the whole range of maternal weight gain and to establish optimal gestational weight gain for each maternal BMI category with cutoff levels based on significant risk estimates less than 1.
MATERIALS AND METHODS
The study population consisted of 298,648 singleton pregnancies delivered in Sweden between January 1, 1994, and December 31, 2004, where information on maternal height, maternal prepregnancy weight, and gestational weight gain was available. The women were identified using the Swedish Medical Birth Registry. Medical data on almost all (98–99%) deliveries in Sweden are listed in this register. The register contains a large number of items concerning pregnancy, delivery, and pediatric neonatal examination. It is based on copies of the standardized medical record forms completed at the maternity health care centers at the start of prenatal care, usually in gestational weeks 10–12, as well as records from the delivery units and on the pediatric examination of the newborn. The system is identical throughout the country. A description and validation of the register content is available.8,9
The midwife records maternal prepregnancy weight and height on a standardized form when the mother-to-be first visits the maternity health care center. Ninety percent of the women present themselves to the antenatal clinic during the first trimester of their pregnancy.
Body mass index (kg/m2) was calculated from maternal prepregnancy weight and height data. Women were grouped into four categories of BMI: underweight (less than 20), average weight (20–24.9), overweight (25–29.9), and obese (30 or more).10 Gestational weight gain was defined as the difference between the maternal weight as recorded when the woman attended the delivery unit and the maternal prepregnancy weight recorded at the first visit to the maternity health care center. The women were grouped into 21 “total weight gain during pregnancy classes,” from 0–44 lb (0–20 kg).
Women with specific medical disorders were not excluded from the analysis. The only exclusion made was multiple pregnancies. Preexisting diabetes occurred in 0.3–0.4% and gestational diabetes in 3% of the study population. Obstetric and neonatal outcome variables included in the study were based on their clinical importance and severity for the mother and the fetus. To select outcome variables directly related to maternal weight gain and BMI was not the purpose of this study. The Swedish Medical Birth Registry contains information on diagnoses given as International Classification of Diseases (ICD) codes. Obstetric and neonatal outcomes evaluated in this study were extracted as ICD codes from the register. The maternal and fetal outcomes and their respective ICD codes are listed in Table 1.
Small for gestational age (SGA) infants were defined as those with birth weight more than two standard deviations below the mean birth weight for gestational age (sex- and parity-specific) according to a Swedish reference curve.11 Large for gestational age (LGA) infants were those with birth weight more than two standard deviations above the mean.
Adjusted odds ratios (ORs) were determined using Mantel-Haenszel technique.12 Estimates of 95% confidence intervals (CIs) were made with a test-based method, based on the Mantel-Haenszel χ2 test.13 The number of individuals in each weight gain class studied was compared with the number of individuals in all other weight gain classes in the same BMI group with regard to the outcome variables studied. Maternal age (six 5-year classes), and parity (1–4+) were regarded as potential confounding factors and were included as covariates in the adjusted analyses. The local ethics committee and the institutional review board approved the study.
The total prevalence of overweight and obesity (BMI 25 or greater) in the study population was 33%. Obesity occurred in 10.7% of all pregnant women. The relationships among maternal age, parity, and maternal BMI groups are presented in Table 2. Overweight and obese women were slightly older and more often multiparous. The prevalence of preterm birth (less than 37 weeks of gestation) was 5%.
Information about maternal prepregnancy weight and height could be retrieved for 84.0% of all births registered during the study period between January 1, 1994, and December 31, 2004. The group of women with missing data on prepregnancy weight and height had been evaluated previously, and no odds ratios concerning the obstetric and neonatal outcome variables studied showed an increase compared with women with known BMI.14 Information on maternal weight gain was available for 37.7% of the registered births with known maternal BMI. The number of individuals with missing data on gestational weight gain in each maternal BMI class was evaluated. Among underweight women, there was information needed for 37.2% to make it possible to calculate gestational weight gain. The corresponding figure among obese women was 38.7%.15
The risks of adverse maternal and fetal outcomes, including SGA and LGA births, along the whole range of maternal weight gain in each maternal BMI class are presented in Figure 1. The optimal gestational weight gain for each maternal BMI category is shown in Table 3. The BMI-specific limits for upper and lower cutoff levels for total gestational weight gain were based on significant risk estimates below 1 for each maternal weight gain determined in Figure 1. The optimal lower and upper limits were extracted from Figure 1 where the odds trend line crosses odds ratio 1. The optimal weight gain limits were set in whole kilograms, and if the line crosses in between two whole kilograms, the first kilogram weight gain with an odds ratio below 1 was set as the lower limit, and the corresponding procedure establishes the upper limit.
This large, population-based cohort study from the Swedish Medical Birth Register made it possible to establish optimal weight gain ranges by maternal BMI based on the relationship between maternal weight gain and undesirable outcomes for both the mother and the infant. The optimal gestational weight gain limits presented in this study were a result of a stepwise evaluation of different possible ways to estimate BMI-specific weight gain limits.
The first attempt to establish optimal gestational weight gain for this population was made by using an accepted proportion of SGA infants (2.5%) to give the lower limit of weight gain and an accepted proportion of LGA (2.5%) infants to give the upper limit in each maternal BMI class. This approach was not successful. Fetal size was a usable outcome variable for normal-weight women, but difficulties began to arise as soon as overweight and obese women were included. The number of LGA infants never decreased to less than 2.5%, even if there was no maternal weight gain at all. That is in accordance with findings from Shieve et al16 who concluded that, for overweight and obese women, the benefit of gaining weight within the IOM range was less apparent. Another drawback with that first approach to setting the optimal weight gain based on fetal size was that SGA births and LGA births were looked upon as being equally important. Current data indicate that there is a higher risk if an infant is born SGA rather than LGA,17 although LGA births are also associated with a number of threatening complications.18,19 In a second analysis, the basic goal was to estimate weight gain limits that were associated with a significantly decreased risk of the most clinically dangerous situations for the mother and the infant. The selection of adverse outcomes was not based primarily on possible correlations with weight gain or maternal BMI. The lower weight gain limit was nonexistent for overweight and obese women in the second analysis, and the upper weight gain limit was surprisingly low. There was some concern that these weight gain limits, if implemented, could lead to an increased number of SGA births. Therefore, SGA was included (as well as LGA) in the presented analysis. The inclusion of SGA and LGA births affected neither the lower nor the upper weight gain limit significantly. The optimal gestational weight gain for BMI categories in this study included SGA births as an outcome variable to minimize the likelihood of increasing frequencies of infants with low birth weight, despite low maternal weight gain.
The weight gain ranges based on adverse obstetric and neonatal outcome data presented in this study were lower than the IOM's recommendations, and the differences were most pronounced for overweight or obese women. There have been discussions about whether the IOM guidelines for weight gain are too high and contributing to the ongoing obesity epidemic.20 Studies concerning weight retention after birth in relation to the magnitude of gestational weight gain showed that high-weight gainers during pregnancy retained more of the weight gain at a 1-year and a 15-year follow-up. Fifty-six percent of the high-weight gainers during pregnancy ended up in the high-weight retainers group.21 Weight gain recommendations based on good pregnancy outcomes, which basically means vaginal delivery at term of a live infant with a normal birth weight to a mother without prenatal complications, may be less than satisfactory for certain groups. Such recommendations exclude, for example, cesarean births, which in many countries are looked upon as a normal and even preferable type of delivery for women. Another problem is the broad weight gain ranges that include good pregnancy outcome.1,2
There are limitations in this study. The outcome variables studied are not sufficiently comprehensive in dealing with degree of severity. The outcomes are not equally dangerous for the mother and/or the child. The outcomes studied are short-term variables related to pregnancy, delivery, and the immediate neonatal period. No long-term effects of differing magnitudes of weight gain, whether for the mother or for the infant, were considered. The completeness and accuracy of the diagnoses used may be imperfect, but this hardly has been affected by maternal weight gain. The vast majority of the women in the study population were of European white origin (more than 95%) so the optimal weight gain may not be applicable to other ethnic groups. A study of Chinese women suggested a different classification of prepregnancy BMI and also reported a different weight gain pattern compared with whites.22 The optimal weight gain levels estimated in this study were based on an observational cohort study and were not prospectively tested.
This study was not limited to term infants only. The cohort was so large that the small percentage of preterm deliveries did not markedly affect the risk ratios related to adverse outcomes. Because preterm birth is a major cause of adverse infant outcomes and women delivering preterm have had less time to gain weight, the risk ratios for adverse outcomes in the low weight gain ranges would have been lower if the preterm group had been excluded.
The advantage of register studies is that the large number of individuals available for evaluation gives higher statistical power and makes it possible to demonstrate associations with low prevalence obstetric and neonatal outcome variables. Another strength is the possibility to stratify by prepregnancy BMI. The drawback is the sometimes-low validity of information. Exposure information (maternal prepregnancy weight and height) was recorded in early pregnancy and therefore was prospective as regards the pregnancy outcome variables. Recall bias was thus avoided. The possibility of a selection bias due to missing data of maternal BMI and maternal weight gain was evaluated and found less probable. Two confounding factors associated with maternal obesity and gestational weight gains were adjusted for in this study: maternal age and parity. Smoking habits in early pregnancy and socioeconomic level were not included as potential confounders, and that could have affected the results.
In this study, odds ratios were evaluated for weight gains of different magnitudes relative to all other weight gains in each BMI class. It would have been possible to have a reference weight gain and compare odds relative to that reference value. The drawback with that approach is that there are no available data on what reference weight gain to choose for each maternal BMI class. The gestational weight gain limits for BMI categories determined in this large population-based cohort study from Swedish Medical Registers showed that a decreased risk of unwanted obstetric outcomes was associated with lower gestational weight gain limits than was earlier recommended, especially among obese women.
1. Institute of Medicine (U.S.), Subcommittee on Nutritional Status and Weight Gain during Pregnancy and Subcommittee on Dietary Intake and Nutrient Supplements during Pregnancy. Nutrition during pregnancy. Washington (DC): National Academies Press; 1990.
2. Carmichael S, Abrams B, Selvin S. The pattern of maternal weight gain in women with good pregnancy outcomes. Am J Public Health 1997;87:1984–8.
3. Abrams B, Altman SL, Pickett KE. Pregnancy weight gain: still controversial. Am J Clin Nutr 2000;71:1233S-41S.
4. Siega-Riz A, Adair L, Hobel C. Institute of Medicine maternal weight gain recommendations and pregnancy outcome in a predominantly Hispanic population. Obstet Gynecol 1994;84:565–73.
5. Cogswell M, Serdula M, Hungerford D, Yip R. Gestational weight gain among average-weight and overweight women–what is excessive? Am J Obstet Gynecol 1995;172:705–12.
6. Stotland NE, Cheng YW, Hopkins LM, Caughey AB. Gestational weight gain and adverse neonatal outcome among term infants. Obstet Gynecol 2006;108:635–43.
7. Hedderson MM, Weiss NS, Sacks DA, Pettitt DJ, Selby JV, Quesenberry CP, et al. Pregnancy weight gain and risk of neonatal complications: macrosomia, hypoglycemia, and hyperbilirubinemia. Obstet Gynecol 2006;108:1153–61.
8. Cnattingius S, Ericson A, Gunnarskog J, Källén B. A quality study of a medical birth registry. Scand J Soc Med 1990;18:143–8.
10. World Health Organization. Obesity: preventing and managing the global epidemic. WHO technical report series 894. Geneva, Switzerland: World Health Organization; 2000.
11. Källén B. A birth weight for gestational age standard based on data in the Swedish Medical Birth Registry, 1985-1989. Eur J Epidemiol 1995;11:601–6.
12. Mantel N, Haenszel W. Statistical aspects of the analyses of data from retrospective studies of disease. J Nat Cancer Inst 1959;22:719–48.
13. Miettinen OS. Simple interval estimation of risk ratio. Am J Epidemiol 1974;100:515–6.
14. Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219–24.
15. Cedergren M. Effects of gestational weight gain and body mass index on obstetric outcome in Sweden. Int J Gynaecol Obstet 2006;93:269–74. Epub 2006 Apr 12.
16. Schieve L, Cogswell M, Scanlon K. An empiric evaluation of the Institute of Medicine's pregnancy weight gain guidelines by race. Obstet Gynecol 1998;91:878–84.
17. Pallotto E, Kilbride H. Perinatal outcome and later implications of intrauterine growth restriction. Clin Obstet Gynecol 2006;49:257–69.
18. Boulet SL, Alexander GR, Salihu HM, Pass M. Macrosomic births in the United States: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol 2003;188:1372–8.
19. Anoon SS, Rizk DE, Ezimokhai M. Obstetric outcome of excessively overgrown fetuses (≥5,000 g): a case-control study. J Perinat Med 2003;31:295–301.
20. Feig D, Naylor C. Eating for two: are guidelines for weight gain during pregnancy too liberal? Lancet 1998;351:1054–5.
21. Linne Y, Dye L, Barkeling B, Rossner S. Long-term weight development in women: a 15-year follow-up of the effects of pregnancy. Obes Res 2004;12:1166–78.
22. Wong W, Tang N, Lau T, Wong T. A new recommendation for maternal weight gain in Chinese women. J Am Diet Assoc 2000;100:791–6.