7
Determining Optimal Weight Gain

INTRODUCTION

In this chapter, the approach used by the committee for arriving at its recommendations for revision of the current guidelines for weight gain during pregnancy is discussed. First, a brief discussion of the principles used by the committee to develop a strategy for making its recommendations is presented. Next, previous approaches for developing gestational weight gain (GWG) guidelines, including those detailed in the Institute of Medicine (IOM) (1990) report, but also others, are discussed. The strategy used by this committee is then described in some detail, along with the results of applying this approach. Finally, the committee’s recommendations are detailed and discussed.

PRINCIPLES USED TO DEVELOP A STRATEGY

As was the case for the report, Nutrition During Pregnancy (IOM, 1990), the committee used a conceptual framework to organize the evidence and identify a set of consequences for the short- or long-term health of both the mother and the child that are potentially causally related to GWG. These consequences included those evaluated in a systematic review of outcomes of maternal weight gain prepared for the Agency for Healthcare Research and Quality (AHRQ) (Viswanathan et al., 2008) as well as others based on data from the literature outside the time window considered in that report. The committee considered both the severity of these



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 241
7 Determining Optimal Weight Gain INTRODUCTION In this chapter, the approach used by the committee for arriving at its recommendations for revision of the current guidelines for weight gain dur- ing pregnancy is discussed. First, a brief discussion of the principles used by the committee to develop a strategy for making its recommendations is presented. Next, previous approaches for developing gestational weight gain (GWG) guidelines, including those detailed in the Institute of Medi- cine (IOM) (1990) report, but also others, are discussed. The strategy used by this committee is then described in some detail, along with the results of applying this approach. Finally, the committee’s recommendations are detailed and discussed. PRINCIPLES USED TO DEVELOP A STRATEGY As was the case for the report, Nutrition During Pregnancy (IOM, 1990), the committee used a conceptual framework to organize the evi- dence and identify a set of consequences for the short- or long-term health of both the mother and the child that are potentially causally related to GWG. These consequences included those evaluated in a systematic review of outcomes of maternal weight gain prepared for the Agency for Health- care Research and Quality (AHRQ) (Viswanathan et al., 2008) as well as others based on data from the literature outside the time window consid- ered in that report. The committee considered both the severity of these 

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY outcomes and their frequency in the population. To develop estimates of risk and frequency, the committee used data from the published literature and from additional, commissioned analyses (see below). The committee considered the incidences, long-term sequelae, and base- line risks of several potential outcomes associated with GWG (additional information about these outcomes appears in Appendix G). Postpartum weight retention, cesarean delivery, gestational diabetes mellitus (GDM), and pregnancy-induced hypertension or preeclampsia emerged from this process as being the most important maternal health outcomes. The com- mittee removed preeclampsia from consideration because of the lack of sufficient evidence that GWG was a cause of preeclampsia and not just a reflection of the disease process. The committee also removed GDM from consideration because of the lack of sufficient evidence that GWG was a cause of this condition. Postpartum weight retention and, in par- ticular, unscheduled primary cesarean delivery were retained for further consideration. Measures of size at birth (e.g., small-for-gestational age [SGA] and large-for-gestational age [LGA]), preterm birth and childhood obesity emerged from this process as being the most important infant health out- comes. The committee recognized that both SGA and LGA, when defined as 90th percentile of weight for gestational age, respectively, represent a mix of individuals who are appropriately or inap- propriately small or large. In addition, the committee recognized that being SGA was likely to be associated with deleterious outcomes for the infant but not the mother, while being LGA was likely to be associated with con- sequences for both the infant and the mother (e.g., cesarean delivery). The committee addressed this mix of outcomes in the approach used to develop its recommendations. Importantly, although the Institute of Medicine report (IOM, 1990) recognized a trade-off between maternal and child health was recognized as a possible consequence of changing the weight-gain guidelines, evalua- tion of that trade-off was not possible with the data then available. This committee made evaluating this trade-off a central element of its process to develop new guidelines while recognizing that, although the available data have increased, they are still less than fully adequate for this purpose. In making its recommendations, the committee also sought to recognize unintended consequences and to develop guidelines that are both feasible and potentially achievable. It is important to note that these guidelines are intended for use among women in the United States. They may be ap- plicable to women in other developed countries; however, they are not in- tended for use in areas of the world where women are substantially shorter or thinner than American women or where adequate obstetric services are unavailable.

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN PREVIOUS APPROACHES FOR DEVELOPING WEIGHT GAIN RECOMMENDATIONS Many approaches have been and are currently being used for making recommendations for how much weight women should gain during preg- nancy. At one extreme is the advice from the National Center for Clinical Excellence in the United Kingdom that women should not be weighed at all during pregnancy, “as it may produce unnecessary anxiety with no added benefit” with the exception being “pregnant women in whom nutrition is of concern” (National Collaborating Centre for Women’s and Children’s Health, 2008). At the other extreme is the single target approach. For ex- ample, in the United States, the 1970 report Maternal Nutrition and the Course of Pregnancy (NRC, 1970) recommended a single target: an average gain of 10.9 kg (24 pounds), with a range of 9.1-11.3 kg (20-25 pounds). This target was based on the amount of weight that healthy women gain when meeting the physiologic needs of pregnancy (e.g., the products of conception, expansion of plasma volume, red cell mass, and maternal fat stores). Still another approach has been used in Chile. Since 1987, maternal weight gain recommendations have been based on a single target, although instead of an absolute amount of weight, a proportion (120 percent) of the woman’s “standard weight” for her height is used (Rosso, 1985; Mardones and Rosso, 2005). Consequently, the recommendation is for a higher gain in underweight women and a lower gain in heavier women, with an upper limit of 7 kg for women with prepregnant weights over 120 percent of the standard (Figure 7-1). The objective of this recommendation is to increase birth weight among underweight women, and it is considered successful in having done so (Mardones and Rosso, 2005). Similar to the Chilean recommendations, the IOM (1990) report also recommended higher gains for underweight women and lower gains (but at least 6.8 kg) for heavier women. The desired outcome was expressed as specific target ranges for each of three prepregnant body mass index (BMI) groups. The rationale for this approach was to achieve the birth weight (i.e., 3-4 kg) associated with “a favorable pregnancy outcome” in all pre- pregnant BMI groups while avoiding the birth of infants with weight > 4 kg because of “the possible risks to the mother and infant of complicated labor and delivery” (IOM, 1990). In constructing their recommendations, both the Chilean investigators (Mardones and Rosso, 2005) and the IOM (1990) committee explicitly recognized the trade-off between raising the birth weight of infants born to underweight women and increasing the risk of high birth weight in some infants as well as obesity and other undesirable outcomes in their mothers. In fact, the IOM (1990) committee recommended that a formal decision

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY Baseline 150 150 140 140 130 130 Weight/Height 120 120 110 110 100 100 90 90 80 80 10 15 20 25 30 35 40 Gestational Age (Weeks) FIGURE 7-1 Graphic showing weight increase for pregnant women. NOTE: A = underweight; B = normal weight; C = overweight; D = obese. Figure 7-1.eps SOURCE: A weight gain chart for pregnant women designed in Chile, Mardones F. bitmap image with vector type elements and P. Rosso. Copyright © 2005, Maternal and Child Nutrition. Reproduced with permission of Blackwell Publishing Ltd.

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN analysis be undertaken “in which probabilities and utilities (values) are assigned to each potential outcome” to assist in balancing the risks and benefits of any recommendation. Since the publication of the IOM (1990) report, several groups of inves- tigators have offered other unique approaches for determining the optimal GWG. It is noteworthy that, with one exception (Nohr et al., 2008), these investigators did not consider maternal and infant outcomes beyond the im- mediate neonatal period. Also, in all of these investigations, the researchers studied GWG as a categorical, not a continuous, variable, and each group of investigators defined the GWG categories differently. All of the studies of this type identified by the committee are discussed below. First, with data from 20,971 pregnant women and their singleton in- fants who were delivered at a single hospital in New York City (1987-1993), Bracero and Byrne (1998) identified the range of weight gains at which the proportion of women who had infants with any 1 of 11 adverse perinatal outcomes was minimal. The list included outcomes not generally associated with GWG, and adverse maternal outcomes were not considered. In gen- eral, they found that this range of GWG was higher than recommended in the IOM (1990) report. They recommended gains of 16.3-18.1 kg (36-40 pounds), 14.1-18.2 kg (31-40 pounds), and 11.8-13.6 kg (26-30 pounds) for underweight, normal weight, and overweight or obese women catego- rized by the cutoff points in the IOM (1990) report, respectively. More recently, with data from the Swedish Medical Birth Registry Cedergren (2007) conducted a population-based cohort study (1994-2004) of 298,648 women and calculated the risk of a variety of pregnancy out- comes by maternal prepregnant BMI category. She did this “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” (Cedergren, 2007). It is important to note that her selection of adverse out- comes was, as she stated, “not based primarily on possible correlations with weight gain or maternal BMI.” In addition to SGA and LGA, her analysis included six maternal and seven fetal outcomes that were unweighted for either their frequency or severity. Preeclampsia was included, but not GDM. With this approach, Cedergren (2007) found that the optimal GWG was lower than that recommended in the IOM (1990) report in all categories, especially for overweight or obese women. In three other recent studies, all of which used population-based cohort data from Missouri, DeVader et al. (2007) studied 94,696 normal weight women (1999-2001), Langford et al. (2008) studied 34,143 overweight women (1990-2004), and Kiel et al. (2007) studied 120,251 obese women (1990-2001) who delivered full-term, singleton infants. All three groups of investigators calculated the risks of pregnancy outcomes routinely col- lected on all birth certificates according to reported GWG. As was the case

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY for Cedergren’s analysis (2007), none of these investigators considered the frequency or severity of these events, and the outcomes of pregnancy were restricted to those at delivery. DeVader et al. (2007) and Langford et al. (2008) both assessed the risk of these outcomes according to whether the women had gained < 11.4 kg (25 pounds), 11.4-15.9 kg (25-35 pounds), or > 15.9 kg (> 35 pounds) dur- ing pregnancy. They found that the primary hazard of gaining less than the IOM (1990) report recommendation was delivering an SGA or low birth weight (< 2,500 g) infant (only Langford et al., 2008); gaining in excess of the recommendation was associated with an increased risk of several ad- verse outcomes, including preeclampsia, cesarean delivery, and delivery of an LGA or macrosomic infant (only Langford et al., 2008). After balancing these risks, DeVader et al. (2007) concluded that the “ideal” gestational weight gain for their population of normal weight women was 11.4-15.5 kg (25-34 pounds). Langford et al. (2008) found that overweight women “should gain within the current recommendations (15-25 lbs)” and that “there may be additional benefit of gaining below the recommendations, specifically in the 6-14 lbs range.” In addition to considering the major categories of prepregnancy BMI, the number of obese women in their sample was large enough so that Kiel et al. (2007) were able to distinguish among obesity classes I, II, and III as well. They found that the risk of delivering an SGA infant continued to de- crease with increasing degrees of maternal obesity and was minimal among women who gained < 6.8 kg (15 pounds) during pregnancy. In addition, although the pattern of increasing risk of preeclampsia, cesarean delivery, and LGA birth with increasing GWG was the same across the obesity classes, Kiel et al. (2007) found that the point at which the risk of these outcomes considered as a group was minimal differed for each obesity class. This minimal risk corresponded to GWG of 4.5-15.5 kg (10-25 pounds) and 0-4.1 kg (0-9 pounds) for obesity class I and obesity classes II and III, respectively. In all of these studies of women from Missouri, the authors chose to consider outcomes that have been related to GWG (although the validity of using preeclampsia is open to question, see Chapter 5). As was the case for Cedergren’s analysis (2007), these investigators did not consider the frequency or severity of these events, and the outcomes of pregnancy were restricted to those at delivery. Finally, in the most recent of the research reports in which authors have tried to identify optimal GWG, Nohr et al. (2008) analyzed Danish National Birth Cohort (1996-2002) data on 60,892 women with term pregnancies. This study included data on weight before pregnancy, weight gain during pregnancy, and postpartum weight obtained during telephone interviews of the mother; outcome data were obtained from birth and hospital discharge registries. Nohr et al. (2008) calculated the risks of a

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN variety of maternal and neonatal outcomes associated with prepregnant BMI and GWG and their interaction. For those outcomes with a strong independent association with GWG and little possibility of reverse causal- ity (unscheduled primary cesarean delivery, SGA, LGA, and postpartum weight retention ≥ 5 kg), the researchers calculated the absolute risk for women in each of the four major categories of prepregnant BMI. Although the trade-off between reducing the risk of SGA and increasing the risk of cesarean delivery was evident in these data, as it was in those from Sweden (Cedergren, 2007) and Missouri (Devader et al., 2007; Kiel et al., 2007; Langford et al., 2008), what is unique in this presentation is the inclusion of postpartum weight retention. Nohr et al. (2008) detected a dramatic increase in postpartum weight retention ≥ 5 kg with increasing GWG in all categories of prepregnant BMI. In addition, they calculated the propor- tion of women who had changed from one BMI category to another at 6 months postpartum according to GWG. They found that only 0.4 percent of underweight women became overweight at the highest GWG (≥ 20 kg) studied. Thus, they concluded that high GWG was “probably not disadvan- tageous for either underweight women or their infants.” For normal weight, overweight, and obese women, however, the trade-off between SGA and these other outcomes, particularly postpartum weight retention, occurred at lower GWG values: 16-19 kg, 10-15 kg, and < 10 kg, respectively. As was the case for the other studies, Nohr et al. (2008) did not weight their outcomes by their frequency or severity; however, it is clear that the authors sought the point of minimum risk of SGA and postpartum weight retention ≥ 5 kg in their decision making. Although the analytic approaches used by these research groups have many similarities, their conclusions about optimal weight gain vary widely (Table 7-1). This is particularly striking for underweight and normal weight women but is also the case for overweight women. The differences in con- clusions may have resulted from the different mix of outcomes considered in each study. For example, the Nohr et al. (2008) study was the only one that excluded preeclampsia and included postpartum weight retention. Cedergren (2007) included a number of pregnancy outcomes that lack a clear association with GWG. None of these reports included the develop- ment of obesity during childhood as an outcome or provided information about the consequences of variation in GWG among women in the racial and ethnic subgroups common in the American population or among women who are young or short—groups that were explicitly considered in the IOM (1990) report. As noted above, none of these analyses was weighted (at least explicitly) by the severity or frequency of the adverse out- comes considered, and the categories of GWG were constructed separately by each group of investigators.

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY TABLE 7-1 Summary of Research Published Since the IOM (1990) Report in Which Recommendations for Optimal Weight Gain During Pregnancy Are Developed Proposed Optimal Weight Gain During Pregnancy (kg) Bracero Maternal 1990 IOM and DeVader Kiel Langford Nohr Prepregnant Guidelines Byrne, Cedergren, et al., et al., et al., et al., BMI (kg/m2) (kg) 1998 2007 2007 2007 2008 2008 IOM BMI Categories Underweight 12.5-18 16.4-18.2 — — — — — (< 19.8) Normal 11.5-16 14.1-18.2 — 11.4-15.5 — — — weight (19.8-26.0) Overweight 7-11.5 11.8-13.6 — — — 6.8-10.9 — (26.0-29.0) (or 2.7-6.4) Obese (> 29) ≥ 6 11.8-13.6 — — — — — WHO BMI Categories > 20 Underweight — — 4-10* — — — (< 18.5) Normal — — 2-10* — — — 16-19 weight (18.5-24.9) <9 Overweight — — — — — 10-15 (25-29.9) <6 < 10 Obese — — — — — (≥ 30) Obese — — — — 4.5-11.4 — Class I (30-34.9) Obese — — — — 0-4.1 — — Class II (35-39.9) Obese — — — — loss of — — Class III 0-4.1 (≥ 40) *BMI cutoff of 20 kg/m2.

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN STRATEGIC APPROACH USED BY THE COMMITTEE IN DEVELOPING ITS RECOMMENDATIONS To address these conflicts and gaps within the available literature, the committee commissioned several additional analyses that informed its deci- sion making (Table 7-2) (see Appendix G). First, Dr. Ellen Nohr expanded TABLE 7-2 Research Commissioned by the Committee: Characteristics of the Datasets Used Consultant Characteristic Nohr Herring Stein Population • Danish National • National Maternal • New York City births Studied Birth Cohort and Infant Health • 1995-2003 • National sample Survey (NMIHS) • Subset of 34,307 • 1996-2002 • Nationally births (those with • n = 60,892 representative maternal height • Singleton term sample among 913,320 births • 1988 linked to singleton births) 1991 follow-up Subgroups • Primiparous vs. • White vs. black • White vs. black Available multiparous • < 20 years old vs. older • Smokers vs. nonsmokers • Short vs. non-short stature • Obesity classes II and III • GWG < 5 kg and = 25 kg Outcomes • SGA/LGA • Primary • Spontaneous preterm Included • Emergency cesarean delivery birth cesarean delivery (n = 5,433) • Primary cesarean • PPWR (≥ 5 kg at • Preterm birth delivery 6 months) (n = 7,728) • SGA/LGA • SGA and LGA (n = 7,748) • PPWR, 6-12 months (n = 1,089) • Breastfeeding initiation and duration • Infant mortality NOTE: GWG = gestational weight gain; LGA = large-for-gestational age; PPWR = postpartum weight retention; SGA = small-for-gestational age.

OCR for page 241
0 WEIGHT GAIN DURING PREGNANCY her published analyses from the Danish National Birth Cohort (Nohr et al., 2008). She provided two sets of analyses and information on an additional lower and an additional higher category of GWG and replicated her pub- lished analyses for obese class I women separately from obese class II and III women. She conducted analogous new analyses for several important subgroups of the population of pregnant women, namely primiparous, short, and young women, as well as smokers (information contributed to the committee in consultation with Nohr [see Appendix G, Part I]). Second, with data from the 1988 National Maternal and Infant Health Survey (NMIHS), Dr. Amy Herring analyzed the association between GWG and outcomes important to the committee separately for white and black women. She also linked the 1988 survey data to its 1991 follow-up and examined the association between GWG and postpartum weight retention. She was unable to examine the long-term weight status of infants born LGA because access to the data could not be obtained in a timely manner (information contributed to the committee in consultation with Herring [see Appendix G, Part II]). Third, Dr. Cheryl Stein analyzed adverse outcomes associated with GWG stratified by racial/ethnic group in the subsample of births during 1995-2003 in New York City for which prepregnant BMI was available (information contributed to the committee in consultation with Stein [see Appendix G, Part III]). The fourth commissioned analysis, described in more detail below, was a quantitative analysis of risk trade-offs between maternal and child health outcomes associated with GWG by Dr. James Hammitt (information contributed to the committee in consultation with Hammitt [see Appendix G, Part IV]). The committee relied on both standard criteria for evaluating the qual- ity of research studies (such as those provided by the American Academy of Pediatrics, 2004) as well as its expert judgment when evaluating the evidence. It used evidence from the published scientific literature as well as the analyses it commissioned. In the development of its recommendations, the committee evaluated the overall quality of the evidence as well as the balance between benefits and risks. Although the committee relied on the highest level of evidence (randomized controlled trials, and experimental studies in women and animal models), few such experimental studies were available in the literature relevant to the committee’s task. In addition, the committee used data from the general population in those instances in which data on minority populations were unavailable. Prepregnant BMI Category After the publication of the IOM (1990) report, the World Health Or- ganization (WHO) held a consultation that developed a categorization of BMI values for adults based on different cutoff points (WHO, 1995). The WHO cutoff points were subsequently endorsed by the National Institutes

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN of Health (NHLBI, 1998). These categories have been widely adopted in the United States and internationally and, if used in formulating recommen- dations for GWG, would provide opportunities for a consistent message to women and health care providers about weight status for all groups of adults, including women of childbearing age. For these reasons, the com- mittee adopted the WHO BMI categories for its recommendations. Evidence from the scientific literature is remarkably clear that pre- pregnant BMI is an independent predictor of many adverse outcomes of pregnancy (see Chapter 5). These data provide ample justification for the choice made in the IOM (1990) report to construct weight-gain guidelines according to prepregnant BMI. That approach has been retained in the current document. Special Populations The following discussion summarizes the committee’s decision-making regarding whether any special populations warrant separate guidelines. The committee considered women of short stature, adolescents, women with multiple fetuses, racial or ethnic group, obesity classes II and III, parity, and smokers. Of these, evidence suggests that only women with multiple fetuses warrant modified guidelines. Women of Short Stature The IOM (1990) report guidelines recommended that women of short stature (< 157 cm) gain at the lower end of the range for their prepregnant BMI. The committee was unable to identify evidence sufficient to continue to support a modification of GWG guidelines for women of short stature (Vishwanathan et al., 2008). The limited data available to the committee indicated that women of short stature had an increased risk of emergency cesarean delivery but that this risk was not modified by GWG; they did not have an increased risk of having an SGA or LGA infant or of excessive postpartum weight retention compared to taller women (Appendix G). No information was available with which to evaluate whether a modification of guidelines might be necessary for very short (< 150 cm) women. Adolescents As discussed in Chapter 4, the committee was unable to identify suf- ficient evidence to continue to support a modification of the GWG guide- lines for adolescents (females < 20 years old) (Vishwanathan et al., 2008) (see Chapter 4). The committee also had to resolve the difference in cutoff values for BMI categories between the growth charts commonly used for adolescents and those used for adults. This is because, for adolescents

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY < 18 years old, the WHO BMI cutoff points for overweight and obesity of- ten do not correspond to the 85th and 95th percentiles, respectively, of the Centers for Disease Control and Prevention (CDC) pediatric growth charts that used to assess growth in these girls (available online at http://www.cdc. gov/nchs/data/nhanes/growthcharts/set2/chart%2016.pdf [accessed Decem- ber 3, 2008]). The younger the girl, the more likely it is that she will reach the 85th or 95th percentile of the growth charts at a lower BMI value than the corresponding WHO cutoff points. Thus, if adult cutoff points are used to determine the prepregnant BMI category of younger adolescents, some girls will be categorized as being in a lighter group, leading to higher GWG recommendations than would be the case if the pediatric growth charts were used to categorize them. The committee determined that this was a tolerable risk for two reasons. First, research has shown that young teens often need to gain more weight than adult women to have an infant of the same size (Scholl, 2008). Second, it would be difficult to implement a recommendation in obstetric practices to use pediatric growth charts to categorize the prepregnant BMI of these girls. Women with Multiple Fetuses The evidence base for women carrying multiple fetuses remains, as it was in 1990, limited. In that report (IOM, 1990), women carrying twins were encouraged to gain 16-20.5 kg (35-45 pounds) without respect to their prepregnancy BMI category. However, as discussed in Chapter 3, recent data suggest that the weight gain of women with twins who have good outcomes varies with prepregnancy BMI (see Chapter 3) as is clearly the case for women with singleton fetuses. Unfortunately, the committee was unable to conduct the same kind of analysis for women with twins as it did for women with singletons because the necessary data are unavailable. Therefore, the committee offers the following provisional guidelines, which are based specifically on the work of Luke and Hediger (Appendix C) and are corroborated by the work of others (Chapter 4): • Normal weight women should gain 17-25 kg (37-54 pounds) at term. • Overweight women should gain 14-23 kg (31-50 pounds) at term. • Obese women should gain 11-19 kg (25-42 pounds) at term. Unfortunately, these data sources do not provide sufficient informa- tion to develop provisional guidelines for underweight women. These pro- visional guidelines reflect the interquartile (25th-75th percentiles) range among women who delivered their twins, who weighed ≥ 2,500 g on aver- age, at 37-42 weeks of gestation.

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN Racial/Ethnic Group The descriptive observational data cited in Chapter 4 suggested that in- adequate GWG was more common in some racial/ethnic groups. However, only Dr. Stein’s analysis of data from New York City in 1995-2003 and Dr. Herring’s analysis of the nationally representative data from the NMIHS in 1988-1991 provided insight into whether a woman’s racial or ethnic group modified the relationship between GWG and the various outcomes of in- terest. The predominant finding from these analyses was that racial/ethnic group did not modify the association between GWG and these outcomes. As a result, the committee concluded that, although confirmatory research is needed, its recommendations should be generally applicable to the vari- ous racial or ethnic groups that make up the U.S. population. Obesity Classes II and III Although a record-high number of American women of childbearing age have BMI values in obesity classes II and III, the evidence identified and reviewed by the committee was insufficient to develop more specific recom- mendations for GWG among these women. Parity It has long been known that primiparous women have smaller infants than multiparous women (as reviewed in Chapter 4) and that they gain more weight during pregnancy. The analyses by Nohr (information contrib- uted to the committee in consultation with Nohr [see Appendix G, Part I]) show that primiparous women must gain more weight during pregnancy than multiparous women do to have an equally low risk of an SGA birth but that primiparous women are similar to multiparous women in their likelihood of retaining ≥ 5 kg at 6 months postpartum in every category of prepregnant BMI. This means that the trade-off between lowering the risk having an SGA infant and increasing the risk of retaining an exces- sive amount of weight postpartum occurs at a different GWG value for primiparous and multiparous women. This is a novel finding that warrants additional study. Smokers It has also long been known that smokers have smaller infants than nonsmokers. Analyses prepared by Nohr (information contributed to the committee in consultation with Nohr [see Appendix G]) show that smokers who gain more weight, as expected, have larger infants, but they also retain more weight postpartum. For example, among normal weight multiparous

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY women, smokers would have to gain at least 16-19 kg instead of 5-9 kg to have a 10 percent risk of having an SGA infant. If they were to gain in this higher range, their risk of retaining ≥ 5 kg at 6 months postpartum would be over 20 percent instead of being about 5 percent. Thus, the weight gain trade-off to prevent an SGA birth is particularly unfavorable for smokers, which is perhaps because at least some of the effect of smoking on birth weight is independent of GWG (as reviewed in Chapter 4). As a result, additional GWG may fail to increase birth weight but, nonetheless, still increase postpartum weight retention. This unfavorable trade-off is best resolved by smoking cessation. DEVELOPMENT OF RECOMMENDED WEIGHT-GAIN RANGES Guidelines for Gestational Weight Gain As was the case for the current guidelines for GWG, the committee chose to formulate the new guidelines with a range for each category of prepregnant BMI. This range reflects the imprecision of the estimates on which these recommendations are based, the reality that good outcomes are achieved with a range of weight gains, and the many additional factors that may need to be considered when making a recommendation for an individual woman. To develop these ranges (listed in Table 7-3), the committee proceeded as follows. Based on the available published literature (Appendixes E and F) as well as the reports of its consultants (Appendix G), the committee ascertained the GWG value or range of GWG values associated with low- est prevalence of the outcomes of greatest interest (i.e., the five outcomes identified earlier: (1) cesarean delivery, (2) postpartum weight retention, (3) preterm birth, (4) small- or large-for-gestational age birth, and (5) child- TABLE 7-3 New Recommendations for Total and Rate of Weight Gain during Pregnancy, by Prepregnancy BMI Total Weight Rates of Weight Gain* Gain 2nd and 3rd Trimester Range Range Mean (range) Mean (range) Prepregnancy BMI in kg in lbs in kg/week in lbs/week Underweight (< 18.5 kg/m2) 12.5-18 28-40 0.51 (0.44-0.58) 1 (1-1.3) Normal weight (18.5-24.9 kg/m2) 11.5-16 25-35 0.42 (0.35-0.50) 1 (0.8-1) Overweight (25.0-29.9 kg/m2) 7-11.5 15-25 0.28 (0.23-0.33) 0.6 (0.5-0.7) Obese (≥ 30.0 kg/m2) 5-9 11-20 0.22 (0.17-0.27) 0.5 (0.4-0.6) *Calculations assume a 0.5-2 kg (1.1-4.4 lbs) weight gain in the first trimester (based on Siega-Riz et al., 1994; Abrams et al., 1995; Carmichael et al., 1997).

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN hood obesity). When weighting the trade-off among these outcomes, the committee considered, within each category of prepregnant BMI: (a) the in- cidence or prevalence of each of these outcomes, (b) whether the outcomes were permanent (e.g., neurocognitive deficits) or potentially modifiable (e.g., postpartum weight retention), and (c) the quality of the available data. The committee compared the resulting ranges with those developed in the quantitative risk analysis conducted by its consultant, Dr. Hammitt. Finally, the committee considered how its recommendations might be accepted and used by clinicians and women. The committee intends these guidelines be used in concert with good clinical judgment as well as a discussion between the woman and her prenatal care provider about diet and exercise. If a woman’s GWG is not within the proposed guidelines, prenatal care pro- viders should consider other relevant clinical evidence, as well as both the adequacy and consistency of fetal growth and any available information on the nature of excess (e.g., fat or edema) or inadequate GWG, before sug- gesting that the woman modify her pattern of weight gain. The safety of intentional weight loss during pregnancy among obese women has not been determined. Thus, priority should be given to addressing weight-loss issues either preconceptionally or between pregnancies, not during pregnancy. In constructing these guidelines, the committee recognized that they fall within the category of personalized medicine. Use of these guidelines will require standardized assessment procedures to inform clinical judgment as well as support of ancillary services (e.g., counseling on nutrition and physical activity) or other interventions that might be deemed necessary to achieve the recommended levels of weight gain. Thus, the committee rec- ognizes that full implementation of these guidelines may entail additional medical expenses. The committee did not attempt to estimate the magnitude of these potential additional medical expenses. Rate of Weight Gain Pregnant women typically gain ~1-2 kg in the first trimester. According to the new recommended GWG values, normal weight women should gain ~0.4 kg per week in the second and third trimesters of pregnancy. Under- weight women should gain slightly more (~0.5 kg per week) and overweight women slightly less (~0.3 kg per week) than this amount (Table 7-3). Obese women should gain about ~0.2 kg per week (Table 7-3). These guidelines were constructed based on the assumption that GWG is linear during the second and third trimesters of pregnancy. The IOM (1990) report made a series of recommendations about how to implement its guidelines in the context of caring for an individual pa- tient. As they remain appropriate, the committee endorses the key elements of these recommendations:

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY 1. Before conception, use consistent and reliable procedures to mea- sure and record in the medical record the woman’s weight and height without shoes. 2. Determine the woman’s prepregnancy BMI. 3. Carefully measure the woman’s height without shoes and weight in light clothing at the first prenatal visit using procedures that have been rigorously standardized at the site of prenatal care. Use con- sistent, reliable procedures to measure weight at each subsequent visit. 4. Estimate the woman’s gestational age from the onset of her last menstruation or from an early ultrasound examination. 5. At the initial comprehensive prenatal examination and together with the pregnant woman, set a weight-gain goal based on pre- pregnant BMI and other relevant considerations and explain to the woman why weight gain is important. 6. Monitor the woman’s prenatal course to identify any abnormal pattern of gain that may indicate a need to intervene, displaying the results graphically for the woman (see Chapter 8, Figures 8-1 through 8-4). When abnormal gain appears to be real rather than a result of an error in measurement or recording, together with the woman try to determine the cause and then develop and implement corrective actions. DISCUSSION OF THE NEW GUIDELINES These new guidelines differ from those issued in 1990 in two important ways. First, they are based on a different set of cutoff points for prepreg- nant BMI. Compared to the cutoff points used in the 1990 guidelines, using the WHO guidelines reduces the proportion of the population in the underweight and obese groups, as these groups are based on more extreme BMI values, and raises the proportion of the population in the normal weight and overweight groups, as these groups are based on wider ranges of BMI values. Second, these new guidelines include a specific, relatively narrow range of recommend gain for obese women. Although this recommendation ap- plies to all women with a prepregnancy BMI value ≥ 30 kg/m2, it reflects the preponderance of data available to the committee that cover women in obesity class I (BMI 30.0-34.9 kg/m2) rather than obesity classes II and III. As noted in Chapter 2, in the past two decades more American women of childbearing age have prepregnant BMI values in obesity classes II and III. Unfortunately, only two studies provide data on women in these obe- sity classes (Kiel et al., 2007; information contributed to the committee in consultation with Nohr [see Appendix G]), and few of the women studied

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN gained < 5 kg. It is possible, based on the data collected in these investiga- tions and compared to higher gains, that weight gains < 5 kg may be as- sociated with a more favorable trade-off among outcomes. However, the committee’s review showed insufficient evidence to recommend gains this low and was concerned about the potential for doing the type of harm that is associated with fetal growth restriction and ketonemia (see Chapters 3 and 6). Ketonemia, which can occur with the accelerated starvation that is characteristic of pregnancy, may be more frequent with low weight gains. The committee recognized that women in obesity classes II and III may, without intervention, gain little during pregnancy and are able to manage their pattern of dietary intake so as to avoid ketonemia and other problems. However, there is no evidence to determine whether a guideline for very low weight gain during pregnancy among women in obesity classes II and III would be managed well enough by these women and their care providers to avoid ketonemia. Although there is ample justification for continuing to structure the new guidelines according to maternal prepregnancy BMI, this approach is not without limitations. Maternal height, for example, has long been known to be a determinant of birth weight among women with a narrower range of prepregnancy weight (40-80 kg) than commonly observed today (Tanner and Thomson, 1970). In addition, height appears to be a stronger predictor than prepregancy BMI of GWG (Straube et al., 2008). However, the research necessary to show that height or another attribute might be a superior alternative to prepregnancy BMI for constructing guidelines for subgroups of pregnant women has not been conducted. The committee based its guidelines, in part, on the presumption that the extensive, consistent observational data that link GWG to fetal growth, as measured by SGA and LGA, as well as those that link GWG to post- partum weight retention are causal. The limited results from randomized trials among undernourished women provide indications of this pathway in some cases (Susser, 1991), as do results from more recent but very small randomized trials designed to control excess weight gain (see Chapter 8). The committee recognizes, however, that the simple model in which in- creased caloric intake increases maternal weight and maternal weight, in turn, increases fetal weight, is likely to be more complex—and may even be incorrect. There are possible non-causal explanations linking GWG to fetal growth, including diet composition, affecting both GWG and fetal growth independently, or shared genetic determinants of GWG and fetal growth, although none of these alternatives has been proven valid. Therefore, in developing these guidelines, the committee determined that it would be prudent to consider the evidence linking inadequate GWG, especially in underweight and normal weight women, with increased risk of SGA; and the evidence linking excessive GWG, especially in overweight and obese

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY women, with increased risk of LGA and its consequences. As additional experimental data are generated to confirm or refute a causal interpretation of the evidence linking GWG and fetal growth, this reasoning may need to be revised. In contrast, the likelihood that the link from increased caloric intake to increased GWG and, in turn, from increased GWG to increased postpartum weight retention is causal seems more certain. However, postpartum weight retention reflects not only GWG but also maternal actions postpartum, including but not limited to changes in dietary intake and physical activity associated with new motherhood as well as breastfeeding behavior (Baker et al., 2008). It is noteworthy that these guidelines are structured around GWG ranges associated with good outcomes for both mother and infant. For ex- ample, women who are more concerned with postpartum weight retention than with the birth of a small baby can choose to gain at the lower instead of the higher end of the range for their prepregnancy BMI category. As American women of childbearing age have become heavier, the trade-off between maternal and child health created by variation in GWG has become more difficult to reconcile than it was when prevention of SGA births was paramount and there was relatively low risk of excessive weight retention postpartum and childhood obesity with additional GWG. The effort made by the committee to project the short- and long-term conse- quences of GWG for both mothers and their children so as to reconcile the trade-offs between them is a unique feature of the process used to develop these new guidelines. For this purpose, the committee used data from the NMIHS (information contributed to the committee in consultation with Herring [see Appendix G, Part II]) to provide estimates for the probabil- ity of infant mortality and data from the Danish National Birth Cohort (Nohr et al., 2008) to provide estimates for the probability of postpartum weight retention related to GWG within each category of prepregnant BMI. Dr. Hammitt linked the data on postpartum weight retention to estimates of morbidity and mortality associated with additional maternal weight. Similarly, data from the Growing Up Today Study (Oken et al., 2008) and supporting studies (see Chapter 6) were used to provide estimates of the risk of childhood obesity at ages 9-14 years related to additional GWG. The committee chose these three outcomes because they are quantitatively important and their consequences could be estimated with available data. Dr. Hammitt used the literature currently available to calculate quality adjustments for each outcome, which resulted in quality-adjusted life- years (QALY) for comparison across outcomes (information contributed to the committee in consultation with Hammitt [see Appendix G, Part IV]). Although the results of this quantitative risk analysis by Dr. Hammitt provided general support for the GWG guidelines that the committee de-

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN veloped from published and commissioned research data needed to support a more complete and persuasive analysis were unavailable. In particular, more information is needed on associations between GWG and longer term maternal outcomes, such as postpartum weight retention and later repro- ductive function and health, and child health outcomes such as fetal growth restriction, child neurocognitive outcomes, and obesity. Such data should include not only the frequencies of outcomes but also the utilities associated with each so that appropriate quality adjustments could be calculated. Overall, these guidelines are remarkably similar to those included in the IOM (1990) report. The research that has appeared since that publication as well as the committee’s commissioned analyses support the robustness of the prior recommendations. Specifically, it remains true that, within a given prepregnancy BMI category, healthy women can deliver healthy infants at a relatively wide range of weight gain values. Unfortunately, an already large and increasing proportion of the population is gaining outside of the prior recommendations (see Chapter 2), which is likely to also be the case with these new guidelines. As a result, it is time to focus attention on helping women to adhere to these guidelines. If research on adherence is conducted with experimental designs of adequate statistical power, such studies could finally provide causal evidence that gaining within these new guidelines results in superior outcomes of pregnancy for both mother and infant. FINDINGS AND RECOMMENDATIONS FROM THE COMMITTEE’S ANALYSES Findings The committee found that: 1. The WHO cutoff points have been widely adopted for categorizing BMI among nonpregnant adults and should be used for categoriz- ing prepregnancy BMI as well; the committee found that these cat- egories are also acceptable to use for categorizing the prepregnancy BMI of adolescents. 2. Evidence from the scientific literature is remarkably clear that prepregnant BMI is an independent predictor of many adverse outcomes of pregnancy. As a result, women should enter pregnancy with a BMI in the normal weight category. 3. Although a record-high number of American women of childbear- ing age have BMI values in obesity classes II and III, available evidence is insufficient to develop more specific recommendations for GWG among these women. 4. There are only limited data available to link GWG to health

OCR for page 241
0 WEIGHT GAIN DURING PREGNANCY outcomes of mothers and children that occur after the neonatal period. 5. There is insufficient evidence to continue to support a modification of GWG guidelines for African American women, women of short stature, or adolescents younger than 16 years of age. 6. There is insufficient data with which to establish how much more weight women carrying multiple fetuses should gain beyond that recommended for women carrying singleton fetuses. 7. The committee reaffirms the clinical recommendations in IOM (1990) for implementation of these guidelines. 8. There is insufficient evidence to reject the possibility that racial/eth- nic group modifies the association between GWG and important maternal and child health outcomes. Recommendation for Action Action Recommendation 7-1: The committee recommends that relevant federal agencies, private voluntary organizations, and medical and pub- lic health organizations should adopt these new guidelines for GWG and publicize them to their members and also to women of childbear- ing age. Recommendation for Research Research Recommendation 7-1: To permit the development of improved recommendations for GWG in the future, the committee recommends that the National Institutes of Health and other relevant agencies should provide support to researchers to (a) conduct studies to assess utilities (values) associated with short- and long-term health outcomes associated with GWG for both mother and child and (b) include these values in studies that employ decision analytic frameworks to estimate optimal GWG according to category of maternal prepregnancy BMI and other subgroups. Additional Recommendation for Research Additional Research Recommendation 7-1: The committee recommends that the National Institutes of Health and other relevant agencies should provide support to researchers to conduct studies among women carry- ing multiple fetuses that link GWG to relevant health outcomes among both mothers and their infants.

OCR for page 241
 DETERMINING OPTIMAL WEIGHT GAIN REFERENCES Abrams B., S. Carmichael and S. Selvin. 1995. Factors associated with the pattern of maternal weight gain during pregnancy. Obstetrics and Gynecology 86(2): 170-176. American Academy of Pediatrics. 2004. Classifying recommendations for clinical practice guidelines. Pediatrics 114(3): 874-877. Baker J. L., M. Gamborg, B. L. Heitmann, L. Lissner, T. I. Sorensen and K. M. Rasmussen. 2008. Breastfeeding reduces postpartum weight retention. American Journal of Clinical Nutrition 88(6): 1543-1551. Bracero L. A. and D. W. Byrne. 1998. Optimal maternal weight gain during singleton preg- nancy. Gynecologic and Obstetric Inestigation 46(1): 9-16. Carmichael S., B. Abrams and S. Selvin. 1997. The pattern of maternal weight gain in women with good pregnancy outcomes. American Journal of Public Health 87(12): 1984-1988. Cedergren M. I. 2007. Optimal gestational weight gain for body mass index categories. Ob- stetrics and Gynecology 110(4): 759-764. DeVader S. R., H. L. Neeley, T. D. Myles and T. L. Leet. 2007. Evaluation of gestational weight gain guidelines for women with normal prepregnancy body mass index. Obstetrics and Gynecology 110(4): 745-751. IOM (Institute of Medicine). 1990. Nutrition During Pregnancy. Washington, DC: National Academy Press. Kiel D. W., E. A. Dodson, R. Artal, T. K. Boehmer and T. L. Leet. 2007. Gestational weight gain and pregnancy outcomes in obese women: how much is enough? Obstetrics and Gynecology 110(4): 752-758. Langford A., C. Joshu, J. J. Chang, T. Myles and T. Leet. 2008. Does gestational weight gain affect the risk of adverse maternal and infant outcomes in overweight women? Maternal and Child Health Journal. Epub ahead of print. Mardones F. and P. Rosso. 2005. A weight gain chart for pregnant women designed in Chile. Maternal and Child Nutrition 1(2): 77-90. National Collaborating Centre for Women’s and Children’s Health. 2008. Antenatal care: Routine care for the healthy pregnancy woman. London: Royal College of Obstetricians and Gynaecologists. NHLBI (National Heart, Lung, and Blood Institute). 1998. Clinical Guidelines on the Identifi- cation, Ealuation, and Treatment of Oerweight and Obesity in Adults. National Insti- tutes of Health Publication 98-4083. Washington, DC: National Institutes of Health. Nohr E. A., M. Vaeth, J. L. Baker, T. Sorensen, J. Olsen and K. M. Rasmussen. 2008. Com- bined associations of prepregnancy body mass index and gestational weight gain with the outcome of pregnancy. American Journal of Clinical Nutrition 87(6): 1750-1759. NRC (National Research Council). 1970. Maternal Nutrition and the Course of Pregnancy. Washington, DC: National Academy Press. Oken E., S. L. Rifas-Shiman, A. E. Field, A. L. Frazier and M. W. Gillman. 2008. Maternal gestational weight gain and offspring weight in adolescence. Obstetrics and Gynecology 112(5): 999-1006. Rosso P. 1985. A new chart to monitor weight gain during pregnancy. American Journal of Clinical Nutrition 41(3): 644-652. Scholl T. O. 2008. Biological Determinants of Gestational Weight Gain. Presentation at the Workshop on Implications of Weight Gain for Pregnancy Outcomes: Issues and Evidence, March 10, 2008, Washington, DC. Siega-Riz A. M., L. S. Adair and C. J. Hobel. 1994. Institute of Medicine maternal weight gain recommendations and pregnancy outcome in a predominantly Hispanic population. Obstetrics and Gynecology 84(4): 565-573.

OCR for page 241
 WEIGHT GAIN DURING PREGNANCY Straube S., M. Voigt, V. Briese and K. T. Schneider. 2008. Weight gain in pregnancy according to maternal height and weight. Journal of Perinatal Medicine 36(5): 405-412. Susser M. 1991. Maternal weight gain, infant birth weight, and diet: causal sequences. Ameri- can Journal of Clinical Nutrition 53(6): 1384-1396. Tanner J. M. and A. M. Thomson. 1970. Standards for birthweight as gestation periods from 32 to 42 weeks, allowing for maternal height and weight. Archies of Disease in Child- hood 45(242): 566-569. Viswanathan M., A. M. Siega-Riz, M.-K. Moos, A. Deierlein, S. Mumford, J. Knaack, P. Thieda, L. J. Lux and K. N. Lohr. 2008. Outcomes of Maternal Weight Gain, Eidence Report/Technology Assessment No. . (Prepared by RTI International-University of North Carolina Evidence-based Practice Center under contract No. 290-02-0016.) AHRQ Publication No. 08-E-09. Rockville, MD: Agency for Healthcare Research and Quality. WHO (World Health Organization). 1995. Physical status: the use and interpretation of an- thropometry. Report of a WHO Expert Committee. World Health Organization Techni- cal Report Series 854: 1-452. Website: http://www.cdc.gov/nchs/data/nhanes/growthcharts/set2/chart%2016.pdf