Nutrition During Pregnancy: Part I: Weight Gain, Part II: Nutrient Supplements (1990)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

8 Effects of Gestational Weight Gain on Outcome in Singleton Pregnancies The subcommittee reviewed the evidence concerning the effects of gestational weight gain on short-term fetal, infant, and maternal health outcomes, as well as maternal factors that could modify those effects. The following outcomes were considered: fetal and neonatal mortality, fetal growth, gestational duration, spontaneous abortion (miscarriage), congen- ital anomalies, maternal mortality, complications of pregnancy, lactation performance, and postpartum obesity. The concepts and terms illustrated in Figure 2-1 were used to analyze and review published studies of human populations bearing on these potential consequences of maternal weight gain. Particular attention was given to controlling for other maternal factors that could confound the relationship between weight gain and preg- nancy outcome. Animal studies were considered only when the clinical and epidemiologic literature was too sparse or contradictory to permit rea- sonable inferences (e.g., for lactation performance). The discussion here is restricted to singleton pregnancies; twin pregnancies are considered in Chapter 9. The subcommittee focused on the links between gestational weight gain and short-term pregnancy outcomes because data relating weight gain to long-term outcomes are relatively scanty, and there is no strong evidence indicating that weight gain affects long-term outcomes directly, i.e., without first affecting shorter-term outcomes. For example, several reports (Naeye and Chez, 1981; Singer et al., 1968; Tavris and Read, 1982) have linked maternal weight gain to subsequent cognitive development in the offspring, but none has shown that such effects occur independently of the effects on 176

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 25 20 CO a) a: .^ 15 o 10 a) CL 5 o - \ \ >135% (Overweight) _ _ _ _ ~ 90-13 Curves Are % of Mean Prepregnancy Weight-for-Height Values (Metropolitan) . - < 90% (Underweight) 10 177 15 Pregnancy Weight Gain, kg FIGURE ~1 Pennatal mortality as a function of maternal weight gain. From Naeye (1979) with permission. fetal (including brain) growth. The long-term child health consequences of preterm birth and intrauterine growth retardation (IUGR) are reviewed briefly later in this chapter. Links between other short-term and longer- term outcomes for both mothers and children (see Figure 2-3) are discussed in Chapter 10, along with other general issues regarding the entire causal pathway. FETAL AND INFANT OUTCOMES Mortality Fetal and infant mortality rates have been used extensively to track progress in improving infant health and, indeed, to reflect the overall health status of the nation. Because these rates are quite low (approximately 1%), however, very large numbers of births are required to study the relationship between weight gain and fetal and infant deaths. Thus, few such studies have been conducted. No exceptions are the Collaborative Perinatal Project (Naeye, 1979) and a National Center for Health Statistics (NCHS) study linking data from the 1980 National Fetal Mortality Survey and the 1980 National Natality SuIvey (duffel, 1986). Data from the Collaborative Perinatal Project (Figure 8-1) indicate that the relationship between gestational weight gain and perinatal mortality is strongly influenced by maternal prepregnancy nutritional status; i.e., there is evidence for important effect modification (see Chapter 2). For women

178 NUTRITIONAL STATUS AND WEIGHT GAIN who were underweight prior to pregnancy, the greater the gestational weight gain, the lower the perinatal mortality. However, for women with desirable prepregnancy weight for height (based on the 1959 Metropolitan Life Insurance Company's tables), perinatal mortality began to rise with gestational weight gains in excess of 11.4 kg (25 lb), which might be partially explained by a rise in the rate of high birth weight and a corresponding increased risk for shoulder dystocia and other complications of labor and delivery (see below). For weight gains above 6.8 to 7.3 kg, (15 to 16 lb) the highest perinatal mortality rates occurred among overweight women (i.e., those with prepregnancy weights greater than 135% of standard weight for height). Data shown in Figure 8-1 may be biased (because women who deliver preterm infants will have had less time to gain weight, and preterm infants are at increased risk for perinatal death). Nevertheless, other data in the same report indicate similar trends even when gestational weight gain was considered as a percentage of a gestational age-adjusted "optimum" gain. According to Naeye, the effects shown were not confounded by maternal age, parity, race, family income, number of prenatal care visits, cigarette smoking, or prior pregnancy history. The NCHS study (duffel, 1986) focused on late fetal deaths (>28 weeks of gestational age). The results, stratified by gestational age (Figure 8-2), are consistent with those from the Collaborative Perinatal Project. The trend toward higher fetal deaths per 1,000 live births for women with lower weight gains (below 11.8 kg, or 26 lb) was most marked among women with low prepregnancy weights and persisted after stratification (one variable at a time) for maternal age, education, and cigarette smoking. Beyond this direct evidence, there is a fairly strong link between fetal growth and mortality (see discussion below). Because of this strong rela- tionship, it is also reasonable to assume, even in the absence of abundant direct evidence, that any effects of gestational weight gain on intrauter- ine growth will be reflected by corresponding, albeit smaller, effects on mortality. Fetal Growth Importance of Birth Weight as a Pregnancy Outcome Infant size at birth is a key determinant of child health, especially in early infancy, but even beyond (see the review by McCormick, 1985~. As shown in Figure 8-3A, for example, neonatal mortality decreases sharply with increasing birth weight up to 2,700 or 2,800 g, declines more slowly up to 3,500 g, is relatively flat from 3,500 to 4,250 g, and then begins to rise slightly (Hogue et al., 1987~. A similar but less pronounced trend is seen for postneonatal mortality (Figure 8-3B).

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 60 50 40 30 a) CO o 20 In ._ m a) ._ J lo lo lo - a) Q In - Ct a) CO a) a) - _ 10 9 8 7 6 4 3 2 1 . ., i, Under 32 weeks 32-35 weeks 36 weeks  AJI gestational periods 37-39 weeks 40 weeks and over . 1 1 1 1 413 ~9.3 1 ~9A' ss6 Maternal Weight Gain, kg `~5 sib\ ~67 7 179 FIGURE S-2 Fetal death ratios by maternal weight gain and period of gestation in the United States, based on data from the 1980 National Natality and National Fetal Mortality Surveys. From duffel (1986~. In an attempt to identify those infants at highest risk, many researchers and policymakers have compared infants with low birth weights (LBWs), i.e., <2,500 g, with infants who weigh more. This dichotomy is crude but provides striking contrasts in outcomes: compared with infants who weigh >2,500 g, LBW babies are nearly 40 times as likely to die during the neonatal period, and those that survive are five times as likely to die during the postneonatal period. Of those who survive infancy, LBW babies are

18() NUTRITIONAL STATUS AND WEIGHT GAIN 1 ,000 cn ._ m ~ 100 to to to a) Q 10 in - ct o In 1,000 o 2 in ~ 100 o it g ~10 a, Q in Cat ~O -a A _ \` I'm V-N All Races Blacks Whites V'N N.x -;~ 500 1,500 ~ B  1 1 500 1,500 2,500 3,500 4,500 Birth Weight, 9 HI Races - Blacks VVhites 2,500 3,500 4,500 Birth Weight, 9 FIGURE 8-3 (A) Neonatal mortality risks By race and birth weight, United States, 1980 live birth cohort. (B) Postneonatal mortality risks by race and birth weight, United States, neonatal survivors of 1980 live-birth cohorts. From Hogue et al. (1987). about 50% more likely to have serious developmental problems or other illnesses (Shapiro et al., 1980). Further subdivisions based on birth weight have been used to refine risk categories. For example, the LBW group is often subdivided into very low birth weight (VLBW), i.e., <1,500 g, and moderately LBW, i.e., 1,500 to 2,499 g. The VLBW infants are at much greater risk of death and

GESTATIONAL WEIGH GAIN IN SINGLETON PREGNANCIES 181 disability than are infants in the moderately LBW group (Kleinman and Kessel, 1987~. At the other end of the scale, high-birth-weight (>4,000 g) infants, especially those weighing >4,500 g, are also at higher risk than normal-weight infants (2,500 to 4,000 g) for adverse outcomes, including mortality (but less so than for the moderately LBW group; see Figure 8-3A), meconium aspiration, clavicular fracture, brachial plexus injury, and birth asphyxia (Boyd et al., 1983; Koff and Potter, 1939; Modanlou et al., 1980~. Birth weight is a composite of two outcomes: the rate of fetal growth and gestational duration. Thus, the use of birth weight often hides more than it reveals. For example, survival among VLBW infants with the same birth weights is considerably higher among those who are small for gestational age (SGA) than it is among those who have a lower gestational age but are larger for their age (Arnold et al., 1988~. A combined classification based on both birth weight and gestational age provides a more discriminating basis for etiologic and prognostic distinc- tions. It is possible to distinguish those LBW infants who are small because they are born preterm (gestational age <37 weeks) from those with IUGR (also referred to as SGA), which is usually defined as a birth weight below the 10th percentile for gestational age. This definition obviously depends on the choice of reference population. Birth weight and gestational age have independent effects on fetal and neonatal mortality (Erhardt et al., 1964; Hoffman et al., 1977; Koops et al., 1982; Lubchenco et al., 1972; Yerushalmy et al., 1965~. Both IUGR and, to a greater extent, preterm infants have an increased risk of developing cerebral palsy (Ellenberg and Nelson, 1979~. Preterm infants (especially those born extremely early) have a far greater risk of developing respira- tory distress syndrome, apnea, ~ntracranial hemorrhage, seps~s, retrolental fibroplasia, and other conditions related to physiologic immaturity. IUGR infants appear to have increased risks of hypoglycemia, hypocal- cemia, polycythemia, and birth asphyxia (Arora et al., 1987; Kramer et al., 1989; Ounsted et al., 1988; Usher, 1970~. The extent to which these neonatal complications are responsible for the increased risk of mortality or later neurocognitive deficits (see below) is not clear. Some degree of deficit in both stature and head circumference may persist (Babson, 1970; Babson and Phillips, 1973; Fancourt et al., 1976; Fitzhardinge and Inwood, 1989; Fitzhardinge and Steven, 1972; Hill et al., 1984; Low et al., 1982; Neligan et al., 1976; Ounsted and lkylor, 1971; Villa r et al., 1984; Walther, 1988; Walther and Ramaekers, 1982; Westwood et al., 1983~. Long-term deficits in neurocognitive performance have been observed in IUGR infants (Fitzhardinge and Steven, 1972; Neligan et al., 1976; Ounsted et al., 1984; Rubin et al., 1973; Westwood et al., 1983; Ylitalo et al., 1988~. However, since asphyxia is a frequent concomitant of growth retardation and studies

182 NUTRITIONAL STATUS AND WEIGHT GAIN have not been limited to nonasphyxiated infants (Westwood et al., 1983), the magnitude of neurocognitive deficits due to growth retardation may be somewhat less than is generally reported. Heterogeneity of IUGR Several methodologic issues should be kept in mind before considering the relationship between gestational weight gain and fetal growth. Problems include measurement of gestational age, as discussed in Chapter 4, and the definition of retarded fetal growth (IUGR). Growth-retarded infants represent a highly heterogeneous group in terms of etiology, severity, and body proportionality. A number of chromosomal and other congenital anomalies associated with growth retardation may lead to prognoses much worse than those for infants without those anomalies. Major congenital anomalies affect only a small percentage of IUGR infants but account for a disproportionate number of deaths. For example, Ounsted et al. (1981) reported that 6.9% of the IUGR infants in their study had such anomalies but represented 62% of the total deaths. It would be quite surprising if two full-term infants, one weighing 2,000 g and the other weighing 2,800 g, had the same prognosis for subsequent morbidity and mortality. Yet, follow-up studies have not subdivided their IUGR cohorts by severity of growth retardation. Thus, little is known about the magnitude of such prognostic distinctions. In recent studies, IUGR infants have been subdivided according to their body proportions, especially as defined by Rohrer's ponderal index (birth weight divided by the length cubed). Those with low ponderal indices are said to be d~sproporaonal (also referred to as asymmetric or wasted). Several investigators have reported higher neonatal mortality rates among disproportional IUGR infants (Guaschino et al., 1986; Haas et al., 1987; Hoffman and Bakketeig, 1984), but better early catch-up growth and better prognoses for long-term growth and development than for those among proportional IUGR infants (Fancourt et al., 1976; Harvey et al., 1982; Hill et al., 1984; Villar et al., 1984~. Unfortunately, most studies in this area have not controlled for the severity of IUGR, with which disproportionality appears to be associated (Kramer et al., 1989), nor have they ensured accurate measurements of gestational age or controlled for confounding by short maternal stature or the postnatal nutritional and other environmental influences listed in Figure 2-2. Other Methodologic Caveats Interpretation of the literature relating gestational weight gain to fetal growth requires adequate consideration of several other factors: problems in measurement of length of gestation and gestational weight gain (Chapter

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 183 4), differences in components of the gain (Chapter 6), and maternal factors that might either confound or modify the relationship (Chapter 5~. The subcommittee emphasizes that use of total weight gain leads to overstatements of the association between gestational weight gain and fetal growth. That is, if the baby's weight is not subtracted from the mother's weight gain, the association is biased by a part-whole correlation problem (i.e., y is being correlated with x + y). Net gain avoids this problem by subtracting the baby's weight. Overstatements of the association of gestational weight gain and fetal growth are also expected unless birth weight is adjusted for gestational age, either by dividing net weight gain by the number of weeks of gestation or by using analytic methods to adjust for the expected gain at each week of gestation. The most appropriate measure of weight gain would be based on serial measurements of weight gain (i.e., the pattern of weight gain) during the course of normal pregnancies. Effects on Birth Weight (for Gestational Age) Despite the methodologic caveats discussed in the preceding section, the published data concerning the effect of gestational weight gain on fetal growth are quite convincing. Methodologically acceptable studies have been virtually unanimous in reporting a positive relationship of gestational weight gain with gestational age-adjusted birth weight and with the risk for IUGR. Based on a meta-analysis (Kramer, 1987) of 61 English- and French-language studies published between 1970 and 1984, the average magnitude of the effect on mean birth weight in women with adequate prepregnancy weight for height is approximately 20 g/kg of total weight gain. The relative risk for IUGR in women with low (<7 kg, or 15 lb) total gestational weight gain is approximately 2.0. Given the prevalence of low weight gain, the etiologic fraction (population attributable risk) in women with average prepregnancy weight for height in developed countries is approximately 14%. In other words, low weight gain can account for about one in seven cases of IUGR. All these quantitative estimates are likely to be inflated, because they are based on total weight gain and thus reflect some degree of part-whole correlation. The effect on mean birth weight, for example, appears to be reduced by about one-third (from 20 to 13 g/kg) when based on net gain rather than total gain (Kramer et al., 1989~. Investigators who have examined the effect of a given gestational weight gain in women with different prepregnancy weight-for-height status have been virtually unanimous in concluding that the two factors strongly interact (i.e., that prepregnancy weight for height is an effect modipery (see Chapter 2~. Miller and Merritt (1979), for example, showed a clear trend for increasing rates of IUGR with decreasing prepregnant weight

184 NUTRITIONAL STATUS AND WEIGHT GAIN 3,600 3,500 , 3 400 ~, ._ a) s m a) ._ an 3,300 3,200 3,100 3,000 2,900 Very Overweight - Moderately Overweight ~~ Ideal weight/' ~ Underweight I 1 1 1 1 1 1 1 1 0 2 4 6 8 10 12 14 16 18 20 Maternal Weight Gain, kg FIGURE 8-4 Birth weight as a [unction of maternal weight and prepregnangy weight for height. Adapted from Abrams and Laros (1986) with permission. for height among women with low gestational weight gain. Similar results were reported in several studies investigating mean birth weight (Abrams and Laros, 1986; Frentzen et al., 1988; Mitchell and Lerner, 1989; Naeye, 1981b,d; Seidman et al., 1989; Winikoff and Debrovner, 1981~. Illustrative data from Abrams and Laros (1986), as adapted by B. Abrams (University of California at Berkeley, personal communication, 1989), are shown in Figure 8-4. Thus, underweight women appear to derive a greater benefit from a given gestational weight gain than do those with adequate or excessive weights. Nonetheless, prepregnancy weight for height is itself a determinant of fetal growth above and beyond the effect of gestational weight gain (Kramer, 1987~. Women who are thinner before pregnancy tend to have smaller babies than do heavier women with the same weight gain. Thus, desirable weight gains in thin women are higher than those in normal- weight women, despite the effect modification, and desirable weight gains for overweight and obese women are lower. The effect of gestational weight gain on fetal growth is weak, or perhaps

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 185 even absent, in obese women (~35% of standard prepregnancy weight for height) (Abrams and Laros, 1986; Brown et al., 1986; Frentzen et al., 1988; Harrison et al., 1980; Luke et al., 1981; Mitchell and Lerner, 1987; Naeye, 1981b; Rosso, 1985; Winikoff and Debrovner, 1981~. Nonetheless, obese women clearly have infants that are larger than those of nonobese women for the same weight gain (Kramer, 1987~. It seems prudent to recommend that obese women gain a minimum equivalent to the weight of the products of conception (6.8 kg, or 15 lb), although lower weight gains in such women are often compatible with optimal birth weights. The subcommittee has not identified an upper limit for this group. The evidence for other effect modifiers is not nearly as strong as that for prepregnancy weight for height. Recent data indicate that the relationship of gestational weight gain to fetal growth is similar in adolescents and older women after controlling for prepregnancy weight for height and other potentially confounding differences (Scholl et al., 1988), although one recent Israeli study reported a substantially (but nonsignificantly) reduced relationship in women under age 20 (Seidman et al., 1989~. These data are concordant with the results of several earlier studies indicating no significant differences in fetal growth in adolescents (even those within 1 or 2 years of menarche), once differences in gestational weight gain, prepregnancy weight, and other confounders have been controlled (Duenhoelter et al., 1975; Horon et al., 1983; Scholl et al., 1984), thus undermining the notion of a competition between the adolescent's own requirements for growth and those of the fetus. Research findings have not been unanimous on this point, however, especially for younger adolescents (<16 years). In an analysis of young, black adolescent mothers in the Collaborative Perinatal Project, Naeye (1981d) found significantly lower mean birth weights among infants born at 38 to 44 weeks of gestation to nonsmokers who were not obese prior to pregnancy. This was particularly true in those aged 10 to 14, in whom deficits averaged approximately 150 to 200 g. However, potential differences in alcohol or drug use were not controlled. In a study of poor, young, urban Peruvian mothers, Frisancho et al. (1985) reported a birth weight deficit of approximately 200 g in young adolescents (<15 years) compared with that in older women (17 to 25 years), even after controlling for gestational weight gain. But these results were not controlled for potentially confounding differences in parity or socioeconomic status. A recent study from New York City (Haiek and Lederman, 1989) showed large (200 to 400 g) deficits in birth weight among full-term infants born to young adolescents (<15 years) compared with those born to 19- to 30-year-old women, even after stratification by weight for height at full-term, unless the adolescents had achieved 140~o of their standard (nonpregnant) weight for height.

186 NUTRITIONAL STATUS AND WEIGHT GAIN Potentially confounding differences in cigarette, alcohol, and drug use were not controlled. Even those studies showing reduced fetal growth in young adolescents do not necessarily demonstrate a true effect modification of weight gain by age. Even if the infants of young teenage mothers have lower birth weights for gestational age after controlling for gestational weight gain (and a variety of potential confounders), this may not indicate a smaller effect of a given weight gain on fetal growth. The lower birth weights might reflect true biologic differences in potential for fetal growth (perhaps related to the young adolescents' own nutritional requirements for growth (Scholl et al., 1989) or to other, unknown mechanisms) in fetal growth or unmeasured or inadequately controlled confounding factors. Of the three studies cited above (Frisancho et al., 1985; Haiek and Lederman, 1989; Naeye, 1981d), only Frisancho et al. present data that directly bear on effect modification. Although the regression coefficients (adjusted slopes) for gestational weight gain in that study decrease with lower maternal age (13 to 15 years), the absolute magnitude of the slopes for 16 year aids (44.4-g birth weight per kilogram of total gestational weight gain) and 17 to 25 year olds (52.2 g/kg) is far higher than the usual effect size of approximately 20 g/kg cited above and, therefore, is difficult to accept at face value, even considering the poor, potentially undernourished population under study. These extremely large effect sizes strongly suggest the existence of residual confounding by socioeconomic or other differences. But lower birth weights seen in infants of young adolescents compared with those seen in infants of older women with the same weight gain, even in the absence of effect modification, argue for promotion of weight gains toward the upper end of the range recommended for older women with a similar weight for height. The subcommittee was able to locate only a single study (Seidman et al., 1989) bearing on possible effect modification by older age (i.e., >35 years). That study reported a slightly but significantly increased effect of gestational weight gain in Israeli women over age 30 as compared with those between the ages of 20 and 30, after controlling for prepregnancy weight for height and other potential confounding variables. (The reported effect was 16.6- compared with 14.0-g birth weight per kilogram of gestational weight gain, respectively.) In addition, since weight does increase with age, older women might be protected to some degree against the adverse effects of low weight gains. Few data are available concerning differences in the effect of weight gain on fetal growth among women of different racial or ethnic back- grounds. Analysis of the 1980 National Natality Survey Duffel, 1986), however, indicates similar effects of gestational weight gain on mean birth weight among white as well as black women. But as with the case of young

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 187 teenagers, black infants tend to be smaller than white infants for the same weight gain of the mother (Kramer, 1987; Taffel, 1986~. Black women should therefore strive for weight gains toward the upper end of the ranges recommended for white women with similar prepregnancy weights for height. The evidence also indicates that women with large gestational weight gains are at increased risk for highbirth-weight infants (Ounsted and Scott, 1981; Scholl et al., 1988; Udall et al., 1978), which can secondarily lead to dysfunctional labor, midforceps delivery, cesarean delivery, shoulder dystocia, meconium aspiration, clavicular fracture, brachial plexus injury, and asphyxia (Acker et al., 1985; Boyd et al., 1983; Koff and Potter, 1939; Modanlou et al., 1980; Sandmire and O'Halloin, 1988~. Most studies cited previously are based on nonrepresentative samples. Thus, their results are of uncertain generalizability. Moreover, few have based their measurements of gestational weight gain on net gain or rate of net gain, so that many of the reported effect sizes may be inflated. A recent analysis of the 1980 National Natality Survey offers an im- portant advance regarding both of these methodologic issues (Kleinman, 1990~. This survey, described in Chapter 5, oversampled LBW infants and included both a physician's and a mother's questionnaire (Taffel and Kep- pel, 1986~. The analysis focused on the risk for full-term LBW (<2,500 g and >37 weeks of gestational age) among married, white, non-Hispanic women who responded to the mother's questionnaire. (The term LBW is a reasonable proxy for one class of IUGR, since infants born at full- term who weigh <2,500 g are clearly growth-retarded. Infants weighing >2,500 g at birth who also fall below the 10th percentile are not included by this definition, however.) The analysis uses multiple logistic regression techniques to control for the following potentially confounding maternal variables: age (<20, 20 to 29, and >30 years), total birth order (1 versus >2), the interaction of age with birth order, high birth order (>3 if age <20, >4 if age >20), education (<12, 12, and >13 years), height (<160, 160 to 168, and >168 cm, or <63, 63 to 66, >67 inches), cigarette smoking (yes or no), and alcohol consumption (none, light two drinks or less once a month or less, moderate more than once a month or more than 2 drinks when they drink). Maternal weight for height was measured by using the body mass index (BMI; i.e., weight in kilograms divided by height in meters squared) in three groups: low (BMI <19.8), moderate (BMI 19.8 to 26.0), and high (BMI >26.0~. The cutoff at 19.8 corresponds to the lowest quar- tile of BMI among the survey subjects, whereas the cutoff at 26.0 closely approximates 120% of the 1959 Metropolitan Life Insurance Company's standards for women with a medium frame (Metropolitan Life Insurance Company, 1959~. Because most other studies of weight gain have found

188 NUTRITIONAL STATUS AND WEIGHT GAIN 20 a) 10 s oh .= a) 0 ~ G 5 CD ._ m o.> 2 ~ j - , to 1 0.5 Low BMI (< 19.8) Moderate BMI (19.8-26.0) High BMI (>26.0) in, _. -_ _, - 0 5 10 15 20 25 Total Weight Gain, kg FIGURE 8-5 Full-te~m low birth weight of live-born singleton infants by total maternal weight gain and prepregnangy BMI of white, non-Hispanic mamed mothers in the United States in 1980. From Kleinman (1990~. different effects, depending upon the mother's prepregnancy weight (see above), the effects of all weight gain measures were assessed separately for each BMI group. Three measures of maternal weight gain were used: total gain, net gain, and the rate of net gain. The effect of weight gain on LBW was assumed to have a quadratic rather than a linear form. That is, weight gain and its square were entered in the logistic regression models. Figures 8-5 to 8-7 show the percentage of full-term LBW infants (on a logarithmic scale) as a function of each weight gain measure, with separate curves for each of the three prepregnancy BMI strata. In each figure, the effect of gestational weight gain is largest (i.e., the slope is steepest) in the low BMI group and least in the high BMI group. For all three groups, however, the effect is greatly attenuated in moving from total weight gain to net weight gain to net weight gain per week. Table 8-1 compares the full-term LBW odds ratios for women with low weight gain by prepregnancy BMI according to each weight gain measure (e.g., total and net weight gain). The cutoff for low weight gain was the 25th percentile for each measure among all women. The effects were similar among women with low and moderate prepregnancy BMIs; those with total weight gains of <10 kg (22 lb) were three to four times as likely to have a full-term LBW baby as those with a gain of >10 kg. However, if either the net weight gain or the net weight gain per week measure is used instead (adjusting the cutoff for low weight gain), the relative odds are reduced to

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 20 ,`5 1 0 c: cn a G cn .= m > J 0 - o 0- 0.5 0 5 10 15 Net Weight Gain, kg 189 - - ~ _ Low BMI (< 19.8) -- Moderate BMI (19.8-26.0) - High BMI (>26.0) - - - 20 25 FIGURE 8-6 Full-term low birth weight of live-born singleton infants by net weight gain and prepregnangy BMI of white, non-Hispanic married mothers in the United States in 1980. From Kleinman (1990~. 20 a) ,<5 1 0 is c: ~ in 'a) 0 G an m m 3 a) 0 > '= co a) =) ~ at, = 0 11 ~ 0.5 0 0.1 0.2 - - 1 1 1 Low BMI (c 19.8) Moderate BMI (19.8-26.0)  1 1 - ~' High BMI (>26.0) .' 0.3 0.4 0.5 Net Weight Gain per Week, kg FIGURE 8-7 Full-term low birth weight of live-born singleton infants by net weight gain per week and prepregnancy BMI of white, non-Hispanic married mothers in the United States in 1980. From Kleinman (1990~.

190 NUTRITIONAL STATUS AND WEIGHT GAIN TABLE 8-1 Full-Term Low-Birth-Weight Odds Ratios for Low Maternal Weight Gain,a by Prepregnancy Body Mass Index (BMI)b C Odds Ratios (95% Confidence Intervals), by Prepregnancy BMI Weight Gain Measure Low (<19.8) Moderate High (>26.0) Total weight gain, 2.4 3.1 1.3 =10 vs >10 kg (1.5, 4.0) (2.2, 4.5) (0.6, 2.8) (c22 vs >22 lb) Net weight gain, 1.4 2.0 1.0 '6.8 vs >6.8 kg (0.8, 2.4) (1.4, 2.9) (0.4, 2.3) ('15 vs >15 lb) Net weight gain per week 1.4 1.7 0.9 '0.17 vs >0.17 kg/wk (0.8, 2.3) (1.1, 2.5) (0.4, 2.1) (c0.375 vs >0.375 lb/wk) a Odds of full-term low birth weight (<2,500 g and-37 weeks of gestation) for mothers with low weight gain (below 25th percentile) compared to other mothers. Based on live births in single deliveries to white non-Hispanic married mothers. Adjusted for maternal age, parity, height, cigarette smoking, and education. b BMI expressed in metric units. c From Kleinman (1990~. 1.5 or 2 (i.e., an estimated increased risk of 50 to 100%~. Regardless of which measure of weight gain is used, the association is weakest (relative odds near 1.0) and becomes nonsignificant (i.e., the 95% confidence interval around the odds ratio includes 1.0) among women with high prepregnancy BMIs. In fact, among those with the highest prepregnangy BMIs, there is an increased risk of a full-term LBW infant among those who gain more than 15.0 kg (35 lb) total (Figure 8-5), 11.4 kg (25 lb) net (Figure 8-6), or 0.27 net kg (0.6 net lb) per week (Figure 8-7~. Finally, the survey data also permit an analysis of weight gains asso- ciated with optimal fetal growth, defined here as a birth weight of 3,000 to 4,000 g and a gestational age of 39 to 41 weeks. The range for optimal fetal growth is based on a balance between lower infant mortality and higher birth weight at full-term and high rates of meconium aspiration, birth trauma, and asphyxia for infants with weights above 4,000 g. In this analysis, a fourth very high BMI (>29) group, which approximates 135% of the 1959 Metropolitan Life Insurance Company standards for women of medium frame, has been split off from the high (BMI >26 to 29) group. As shown in Figure 8-8, the distribution of total weight gains among women giving birth to optimally-grown infants is extremely wide for all four BMI groups. In the middle 70% (i.e., those between the 15th and 85th percentiles), mothers with low or moderate prepregnancy BMIs had total gains of between 7.3 and 18.2 kg (16 and 40 lb). For mothers in the high and very high BMI groups, the corresponding ranges were somewhat wider:

191 o, CM m ... .,..' o CM m o ~o C~ 0 m 0 a v o CO 1= _3 _~ -__ ~,.~ \\\\\N r\\\\\\- \\\\\\~^ r\\\\\N ~ \\\\\\\\\\~ ~\\\\\\\\\\\\~\\\\\\\\\\ r\\\\\~ E Z-£Z< L-ZZ-6 0Z S-OZ~-8L Z-8l-b-91 6 Sl-l-~L 9-Cl-8 ll t-Ll-S-6 l-6-~-] 8 9-O S S-~-Z C-Z-S-0  O > Z-EZ< 2-ZZ-6-0Z S OZ~-8 ~ Z-8l-b-9l 6-Sl-l-bl 9-£l-8 ll t-Ll-S-6 l-6-£-l 8-9-O S S-~-L-Z C-C-S-0 O > S0Z-98l Z8l-b9l 6-SI-l-~L 9-£l-8-ll t-Ll-S-6 ~ 6-£-' 8 9-O S S-b-~-Z £-Z-S 0 O > 4 S OZ-9 8l t\\\\ Z 81-b 9l - ~\\\\\\\\\~N' [\\\\\' ~\\\\\\\\\~; ~\~\\\\\\\\\\\\\\\\\\\\\\\~\\ . N\\\-\-\\\\\\\\\\\\~N C~\\\\\\\\\\\\\\- F\\\' o C~ % ~s~uejul alOu!s '6~!1 6U!J9A!laC slaq~oW pe!lleW o!ueds!H-uoN '01!UM o o ._ Ct U) C) oc o Ct ._ ~ ~_ ou Q 3 ~ _' Ct  _ o _ Ct oo ~ -o [,L, Q O d. O .= ~ ~ ~C) C) ~3.- ._ ~ ~: o _ `: ._ ._ 3 o ~ :> c a Ct 6 S L-I ~ ~ 9-£l-8 LL t-ll-S-6 l 6-£-L 8 9-O S S-~-L-Z £-Z S-0 O > CQ - Ct ._ o~ _ ;;> .,= ·- os ._ 3 s: _ ct au ~ .~ _ ~ a ~ C_4 O o a D ~:5 - E . oo ~ ~ 'Z CY C: _4 o

192 NUTRITIONAL STATUS AND WEIGHT GAIN 5.0 to 18.2 kg (11 to 40 lb) and 0.5 to 15.9 kg (1 to 35 lb), respectively. Mean it standard deviation gains in the low and moderate BMI groups were about 13.6 ~ 5 kg (30 ~ 11 lb). For the high-BMI group, the mean was somewhat lower (12.3 kg, or 27 lb), but the standard deviation was considerably higher (6.8 kg, or 15 lb). Very high BMI mothers had an even lower mean (9.5 kg, or 21 lb) and higher standard deviation (8.2 kg, or 18 lb). Effect of Weight Gain Pattern Few investigators have examined the relationship between the pattern of weight gain and fetal growth. Because Comparatively little weight is gained in the first trimester, one would expect second- and third-trimester weight gains to have the largest impact on fetal growth. This theoretical argument is supported by the results of the Dutch famine study (Stein et al., 1975) and those of supplementation trials, in which caloric supplements were usually begun in the second or third trimester (Lechtig et al., 1975; Mora et al., 1979; Prentice et al., 1983; Viegas et al., 1982~. Several investigations of weight gain patterns are consistent with this evidence. For example, Thomson and Billewicz (1957) reported that low weight gains during weeks 20 to 30 or weeks 30 to 36 of gestation were associated with an increased risk of LBW. Picone et al. (1982) found that birth weight was significantly correlated with weight gain during the second and third trimesters, but not the first. Hediger et al. (1989) reported anal adolescents warn low weight gains either before or after 24 weeks of gestation were at increased risk of delivering an IUGR infant. In a recent case-control study (Lawson et al., 1988), women giving birth to IUGR infants were found to have only half the average weight gain between 28 and 32 weeks of gestation as that of women giving birth to non-IUGR infants (other specific gestational periods were not examined). The first trimester of weight gain may also be important, however. Tompkins and colleagues Tompkins and Wiehl, 1951; Tompkins et al., 1955) found elevated rates of LBW associated with low first- or second- trimester weight gains, even after stratification by prepregnancy weight. In France, Lazar (1981) reported that correlations between maternal weight gain and sex- and gestational age-adjusted birth weights were low and did not increase with advancing length of gestation. In fact, the ultimate average difference in gestational weight gain of approximately 2 kg (4.4 lb) between mothers giving birth to IUGR and large-for-gestational-age infants appeared to be established by midgestation. Finally, a recent study (Gross et al., 1989) of pregnancy outcome in women with hyperemesis gravidarum (generally associated with improved outcomes) reported impaired fetal growth in those who lost more than 3.6 kg (8 lb) of their prepregnant . , . . , . ~. . . . . . ~

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 193 weight. (Most of this loss presumably occurred during the first trimester, although details of the timing were not provided.) Given the small amount of new maternal tissue (lean or fat) that accumulates during the first trimester, any relationship between early weight gain and fetal growth may reflect expansion of plasma volume or even noncausal mechanisms. Effect of Weight Gain Composition The importance of the composition of tissues gained is far from clear. Studies in Peru (Frisancho et al., 1977) and Sweden (Langhoff-Roos et al., 1987) suggest that increases in lean body mass may be more important than fat accretion for intrauterine growth. Similarly, data collected by Pipe et al. (1979) at 10 to 14 weeks of gestation and later analyzed by Campbell-Brown and McFadyen (1985) showed correlation coefficients (R) between birth weight and maternal fat-free mass, height, weight, and skinfold thickness of .44, .24, .22, and .07, respectively. In Senegal, Briend (1985) found a negative correlation between maternal triceps skinfold thickness at full- term and birth weight in full-term infants after controlling for maternal weight (also at full-term), parity, and sex of the infant. Although Briend interpreted these data to indicate that maternal energy reserves do not limit fetal growth, they are equally compatible with the inference that mobilization of fat stores in late pregnancy results in improved fetal growth. Data from two recent studies suggest that an increase in maternal fat stores (as approximated by skinfold thicknesses) may enhance fetal growth. Viegas et al. (1987) found that infants born to mothers with a <0.02-mm weekly increase in triceps skinfold thickness between 18 and 28 weeks of gestation not only had a lower mean birth weight but also a smaller head circumference and mid-upper arm circumference. Similarly, Maso et al. (1988) reported anthropometric changes between 22 and 32 weeks of gestation in 100 black teenage mothers. The 10 women who gave birth to LBW infants (8 of the 10 were preterm) had significantly smaller increases in their mid-upper arm circumferences, and actual decreases in their triceps skinfold thicknesses. However, the results are difficult to interpret because of the small number of LBW infants and the failure to control for other potentially confounding differences between the two groups. Differences in gestational weight gain may also reHect changes in nonnutritional body components, especially those due to body water content and plasma volume. Several studies indicate that changes in body water may affect intrauterine growth independently of the nutritional aspects of maternal weight gain (Campbell and MacGillivray, 1975; Duffus et al., 1971~. It has long been recognized that edema alone (i.e., not accompanied by proteinuria or hypertension) is associated with improved fetal growth (Billewicz and Thomson, 1970; Naeye, 1981a,c; Thomson et al., 1967~.

194 NUTRITIONAL STATUS AND WEIGHT GAIN The beneficial effect is seen even in women with dependent (leg) edema only, despite the absence of any increase in total body water (Hytten and Thomson, 1976~. Effects on Birth Length and Head Circumference The term fetal growth should be interpreted as indicating more than merely birth weight or birth weight for gestational age. In particular, other dimensions of fetal growth, especially body length and head circumference, have often been used as indices of pregnancy outcome. Although few data directly link gestational weight gain to infant length, head circumference, or body proportionality, IUGR infants tend to be shorter and have smaller heads than those of infants with a normal weight for gestational age (Kramer et al., 1989~. There does appear to be some degree of head and length sparing, however, such that the relative shapes of infants change with successive degrees of growth retardation. The well-controlled observational study by Miller and Merritt (1979) of women in the Kansas City area showed that rates of short-for-date infants and, to a far lesser extent, infants with small head circumferences both increased with low gestational weight gain. This trend was especially marked in women with low prepregnancy weights for height. Studies of nutritional deprivation and energy supplementation support these overall findings. For example, the Dutch famine study (Stein et al., 1975) demonstrated a significant reduction of 1.3 cm in length and 1.0 cm in head circumference among infants born to women exposed to famine conditions in the third trimester. Conversely, a supplementation trial of poorly nourished women in Colombia (Herrera et al., 1980) suggested positive effects of supplementation on birth length and head circumference. Larger head circumferences were also reported among infants born to supplemented women in The Gambia (Prentice et al., 1983~; no data on birth length were provided in that study. Gestational Duration Compared with the extensive amount of literature on fetal growth, there are relatively few published reports on the relationship between maternal weight gain and gestational duration. Papiernik and Kaminski (1974) found a significantly increased risk of subsequent preterm birth to mothers with either low (<5 kg, or 11 lb) or high (~9 kg, or 20 lb) weight gains up to 32 weeks of gestation; however, these results were based on bivariate (i.e., potentially confounded) associations. A companion report (Kaminski and Papiernik, 1974) showed that low weight gain by 32 weeks of gestation significantly discriminated between subsequent low (<2,500 g)

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 195 and normal (>2,500 g)-birth-weight infants, but a specific relationship with preterm birth (as opposed to IUGR) was not examined. Miller and Merritt (1979) reported a higher rate of preterm birth among 108 white women who gained less than 227 g/week during the last two trimesters, but no such effect was seen among the 70 black women with low weight gains. Berkowitz (1981) reported a fourfold increased risk of preterm delivery in women with inadequate weight gains compared with that in women with adequate gains. No details were provided, however, on the criteria used for assessing adequacy other than a general statement that the assessment was based on "a schedule . . . which adjusted for pregnancy duration," p. 87. Hingson et al. (1982) found a significant positive partial correlation (i.e., simultaneously adjusted for multiple potential confounding variables) between total weight gain and mean gestational age, but this result was based on total gain, rather than rate of gain, and may therefore reflect a reverse causal influence of gestational duration on total weight gain. Mitchell and Lerner (1989) found an increased risk of preterm delivery in women with both normal and subnormal prepregnancy relative weights and total weight gains of <9 kg (20 lb). Here, too, the increase in total weight gain does not permit firm inferences to be made about causal direction. Picone et al. (1982) compared women with predicted low (<6.8 kg, or 15 lb) and adequate (>6.8 kg) total weight gains. They based their predictions on weight gains of <3.6 kg, or >3.6 kg (<8 or >8 lb) at 20 weeks of gestation. There was no difference in mean gestational age between the two groups when based on the mother's last menstrual period (LMP), but there was a slightly shorter (38.5 compared with 39.2 weeks of gestation; p <.01) gestational age when based on neonatal (Dubowitz) examination. These results are even more difficult to interpret in light of the fact that the investigators reclassified two of the women whose total weight gains did not agree with their predictions at 20 weeks of gestation. van den Berg and Oechsli (1984) reported a highly significant increased risk of preterm birth (based on LMP) in women with weight gains averaging <0.23 kg/week (<0.5 lb/week) after 20 weeks of gestation in the large Child Health and Development Studies conducted in the San Francisco Bay area. These results are closely paralleled by those in a recent study by Abrams et al. (1989), who observed an increased risk of preterm delivery in women with a low (<0.27 kg/week, or 0.6 lb/week) rate of weight gain (a 60% increase in risk over women gaining 0.27 to 0.52 kg/week, or 0.6 to ~1 lb/week). The magnitude of elevated risk reported by Abrams and colleagues was not materially altered when the analysis was restricted to preterm births of infants whose gestational ages had been confirmed by ultrasound examination before 28 weeks of gestation. Two other reports of a recent study among adolescents (Hediger et al., 1989; Scholl et al., 1989) also indicated an association between low rate of weight gain during

196 NUTRITIONAL STATUS AND WEIGHT GAIN gestation and preterm delivery. The magnitude of the increased risk varied from 50 to 75%, depending on whether the gestational age was based on the LMP or an "obstetric" (undefined) estimate. The risk appeared to be even higher if the low gain occurred both before and after 24 weeks of gestation. Data from the 1980 National Natalie Survey (Kleinman, 1990) were also reviewed by the subcommittee. Preterm LBW (<2,500 g at <37 weeks of gestation) was examined as a function of maternal weight gain. Stratification of prepregnancy BMI and the three different weight gain measures used were the same as those in the analysis for full-term LBW infants. As shown in Figures 8-9 to 8-11, women in the lowest BMI group generally had the highest rates of preterm LBW infants. The apparent effect of weight gain in women in all three BMI groups diminished markedly, however, in moving from total weight gain to net weight gain to net weight gain per weelc In fact, Figure 8-11 indicates little or no effect of net weight gain per week, except perhaps for a slight reduction in LBW with higher gains in the low BMI group. That the results differ so markedly according to which weight gain measure is used strongly suggests that the apparent effect of total (or even net) gain is due to reverse causality; i.e., women who deliver their infants preterm will have had less time to gain weight. Table 8-2 shows the preterm LBW odds ratios for women with low 20 a) ~ 10 .= 0 O) G ~ A: 5 ~ ._ ·= m m a, ~ .> 0 ~ ~2 an a 1 0.5 - - LowBMI(< 19.8) Moderate BMI (19.~26.0) High BMI (>26.0) - 1 1 1 1 0 5 10 15 20 25 Total Weight Gain, kg FIGURE 8-9 Preterm low birth weight of live-born singleton infants by total maternal weight gain and prepregnancy BMI of white, non-Hispanic married mothers in the United States in 1980. From Kleinman (1990~.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 20 a) ct a) _ 5 .= ~ m m a, in.= 2 en E a) a) ° 1 0.5 197 Low BMI (< 19.8) Moderate BMI (19.8-26.0) High BMI (>26.0) - _ 0 5 Net Weight Gain, kg FIGURE 8-10 Pretend low birth weight of live-born singleton infants by net maternal weight gain and prepregnangy BMI of white, non-Hispanic married mothers in the United States in 1980. From Kleinman (1990~. 20 a) 10 ,= 0 a) G ._ ._ m a, in-> 2 c a) a) an j o 1 - o 0- 0.5 Low BMI (< 19.8) Moderate BMI (19.8-26.0) L _ High BMI (>26.0j 1 1 1 1 1 1 0.1 0.2 0.3 0.4 0.5 Net Weight Gain per Week, kg FIGURE 8-11 Preterm low birth weight of live-born singleton infants by net maternal weight gain per week and prepregnancy BMI of white non-Hispanic mamed mothers in the United States in 1980. From Kleinman (1990~.

198 NUTRITIONAL STATUS AND WEIGHT GAIN weight gain according to each of the three weight gain measures. As with full-term LBW, the effect of low weight gain on the relative odds of preterm LBW for women In the high BMI group was weak and insignificant, regardless of the weight gain measure used. For women in the low and moderate BMI groups, there were significant associations between low total or net weight gain and the relative odds of preterm LBW. But the odds ratios were close to, and not significantly different from, 1 when the analysis was based on the more appropriate measure of net gain per weeL One important methodologic caveat should be kept In mind in in- terpreting studies linking gestational weight gain to gestational duration. As discussed earlier In the section on fetal growth, errors in estimation of gestational age (particularly when based on menstrual dates) may well lead to misclassification of some growth-retarded Infants as preterm. Since gestational weight gain has been shown to have an impact on fetal growth and the risk of IUGR, evidence of effects on gestational duration based on menstrual dates should be interpreted with caution. Nevertheless, some data do suggest a possible effect of low weight gain on reducing gestational duration and increasing the risk of preterm delivery. Further research in this area using validated gestational age measurements (e.g., based on early ultrasound examination) should receive high priority, given the well-known importance of preterm delivery on infant mortality and infant and child morbidity and performance. TABLE 8-2 Preterm Low-Birth-Weight Odds Ratios for Low Maternal Weight Gain,a by Prepregnancy Body Mass Index (BMI)b C Odds Ratios (95% Confidence Intervals), by Prepregnancy BMI Weight Gain Measure Low (<19.8) Moderate High (>26.0) Total weight gain, 4.0 2.8 1.6 '10 vs >10 kg (2.7, 6.0) (2.0, 4.0) (0.8, 3.2) (c22 vs >22 lb) Net weight gain, 1.9 1.8 1.1 '6.8 vs >6.8 kg (1.2, 2.8) (1.4, 2.6) (0.6, 2.2) (~15 vs >15 lb) Net weight gain per week 1.2 1.0 1.0 sO.17 vs >0.17 kg/wk (0.8, 1.9) (0.7, 1.5) (0.5, 1.9) (c0.375 vs >0.375 lb/wk) a Odds of preterm low birth weight (<2,500 g and <37 weeks of gestation) for mothers with low weight gain (below overall 25th percentile) compared with those of other mothers. Based on live births in single deliveries to white, non-Hispanic married mothers. Adjusted for maternal age, parity, height, cigarette smoking, and education. b BMI expressed in metric units. c From Kleinman (1990~.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES Spontaneous Abortion (Miscarriage) 199 There is a general paucity of data concerning the effects of gesta- tional weight gain on the risk of spontaneous abortion, i.e., first- and early-second-trimester spontaneous abortions (later-second-trimester spon- taneous abortions merge with preterm deliveries and probably share many of their etiologic determinants). Unfortunately, epidemiologic studies of early miscarriages are quite difficult, since these pregnancy losses may not be recognized or, if recognized, may not come to the attention of a physician or other health care worker. Few data relate early miscarriage to maternal nutrition. A hospital- based study in New York City showed no association between chromoso- mally normal spontaneous abortion and prepregnancy BMI (Stein, 1989~. Risch et al. (1988) found that neither height, weight, nor the presence or absence of obesity (all presumably at the time of the interview) was associated with the woman's risk of a previous spontaneous abortion. In an early study of the effects of the Dutch famine, Smith (1947) noted an increased rate of "abortion and miscarriage" among Rotterdam women who conceived during the famine, but it is not clear whether the reported figures include induced abortions. Smith commented, "There is no reason to assume [the data] are accurate or that conclusions can be drawn from them," p. 603. The only direct data linking gestational weight gain and spontaneous abortion came from a study in Bangladesh (Pebley et al., 1985~. Conception from June to October-a lean period of arduous work and reduced diet was associated with pregnancy loss primarily in the third trimester and, to a lesser extent, before and during the second trimester. Independent effects were reported both for reductions in prepregnancy weight as well as gestational weight gain, both of which were reduced during the lean period. Congenital Anomalies The subcommittee found no studies directly linking gestational weight gain to the risk either of congenital anomalies In general or of specific malformations. Such a link would not be expected, given the negligible weight changes that occur during the very early gestational period of embryonic morphogenesis. MATERNAL OUTCOMES Maternal Mortality In most developed countries, maternal mortality during pregnancy, childbirth, or the puerperium is extremely rare- generally less than 10 per

200 NUTRITIONAL STATUS AND WEIGHT GAIN 100,000. The most common causes are pregnancy-induced hypertension (PIH), pulmonary embolism, ectopic pregnancy, and hemorrhage (ante- and postpartum) (NCHS, 1987~. In the United States, rates are three to four times higher for black women than for white women (Buehler et al., 1986~. This has been attributed to poor prenatal care among black women (Sachs et al., 1987), although the causal relationship between prenatal care and the risk of maternal death remains uncertain. Nutritional differences and many other explanations for the higher rates are possible. Neither gestational weight gain nor other nutritional factors have been investigated directly for this association. Complications of Pregnancy, Labor, and Delivery Large weight gains, especially in the second and third trimesters, have long been associated with an increased risk of PIH and preeclamptic tox- emia (including proteinuria and generalized edema) in primiparous women (Naeye, 1981b; Shepard et al., 1986; Thomson and Billewicz, 1957; ~mp- kins et al., 1955~. Recognition of this association reinforced the earlier obstetric practice of limiting weight gain, which appears to have originated with an observed reduction in eclampsia that was temporally associated with food shortages in Germany during World War I (Anonymous, 1917; NRC, 1970~. But the determination of causality in this association is highly problematic, since the edema and increased body water accompany- ing preeclampsia will, of course, be manifested by an increased maternal weight, irrespective of any change in maternal lean or fat mass. The typical pattern is a sudden increase in weight between visits in the third trimester. The subcommittee was unable to locate any evidence linking increases in maternal lean or muscle mass early in pregnancy to subsequent PIH or preeclampsia. In fact, an expert committee of the World Health Organiza- tion concluded that it is difficult, based on the available evidence, to define the precise role of nutrition in toxemia (WHO, 1965~. The severe energy restriction that occurred during the Dutch famine was associated with a significant reduction in systolic blood pressure near the time of delivery (Ribeiro et al., 1982~. Although this might represent some reduction in the risk of PIH or preeclampsia, it could also be a mechanism for lowering uterine blood flow and thereby for impairing fetal growth. Ibmpkins et al. (1955) did note a higher risk of toxemia in women with low prepregnancy weights for height who had low weight gains during the second trimester. In a review of the published evidence, Chesley (1976) found that, despite the weight gain associated with preeclampsia, most women who develop the condition have total weight gains that are below average. Thus, the causal relationship between gestational weight gain and PIH/preeclampsia remains unclear. Low early gains may be a marker, or even a determinant,

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 201 of subsequent gestational hypertensive disorders, but firmer inferences must await the results of future research. As previously discussed, large gestational weight gains are associated with an increased risk of high birth weight and a corresponding increase in risk for dysfunctional labor, midforceps delivery, and cesarean delivery (Boyd et al., 1983; Koff and Potter, 1939; Modanlou et al., 1980~. Moreover, there is some evidence that these consequences of fetopelvic disproportion are exacerbated in women with short stature or small pelvic size (Frame et al., 1985; Hughes et al., 1987~. Cesarean delivery rates have skyrocketed over the last 15 to 20 years (Placek and Taffel, 1980; Placek et al., 1983), but the remarkable increase has been accompanied by rather modest increases in the rates of high birth weight (see Chapter 3~. Thus, even if larger gestational weight gains are partly responsible for the trend toward slightly larger infants, their contribution to complications of labor and delivery must be quite small. Varma (1984) examined the direct relationship between gestational weight gain and pregnancy complications. Although he reports a signifi- cantly higher rate of forceps and cesarean deliveries among women with fetal weight gains >16 kg (35.2 lb), and especially >21 kg (46.2 lb), these re- sults are unadjusted for potentially confounding differences among women with different weight gains. Using a more sophisticated multivariate ap- proach, however, Shepard et al. (1986) confirmed that women with large weight gains (~35% of their prepregnancy weight) had higher rates of cesarean deliveries and other operative deliveries (forceps and vacuum extraction), as well as a prolonged second stage of labor. Lactation Performance There is a general perception that fat deposition during pregnancy is required for optimal lactation performance. Although several studies have examined the relationship between milk production and maternal nutrition during lactation, few have related lactation performance to gestational weight gain. In one longitudinal study of well-nourished women in the United States, gestational weight gain was not related to milk quantity or quality (Butte et al., 1984~. Fat mobilization was not a prerequisite to adequate milk production, as indicated by the inverse relationship between the amount of energy mobilized from maternal stores and dietary energy intake. Other studies on humans do not support the hypothesis that fat de- posited during pregnancy is necessarily mobilized later during lactation. In one study of Swedish women, the mean gestational weight gain (13.8 kg, or ~30 lb) included substantial quantities of fat (5.8 kg, or ~13 lb) (Sadurskis et al., 1988~. During the first 2 months of lactation, total body fat did not

202 NUTRITIONAL STATUS AND WEIGHT GAIN change; milk production and composition were normal. Energy costs of lactation were met by increased energy intake, not by body fat mobilization. By contrast, investigators from The Gambia have inferred that fat deposition during pregnancy is of crucial importance for lactation perfor- mance. In one Gambian study, milk output at 3 months post partum was negatively correlated with the change in skinfold thicknesses from 6 to 12 weeks post partum (Paul et al., 1979~. In women who were replenish- ing their fat stores during the dry (harvest) season, milk output was low. Although the investigators interpreted this observation to indicate compe- tition between replenishment of maternal body fat and milk production, the data are also consistent with mobilization of maternal fat for milk pro- duction. Subsequent, conflicting results indicated higher milk production rates during the dry season compared with those during the wet (farming) season (Prentice and Whitehead, 1987~. One study conducted in East Java, Indonesia, demonstrated that energy supplementation in the last trimester of pregnancy did not increase milk output among women with habitually low energy intakes (van Steenbergen et al., 1989~. The limited evidence from studies on the relationship between ges- tational weight gain and lactation performance in humans can be supple- mented with findings from animal studies. Extrapolation of data on re- production from nonprimate animal models to humans can be hazardous, since marked differences in the energy costs of gestation and lactation exist between primates and other mammals. Nonetheless, the evidence from animal studies indicates that gestaizonal nutrition is less important than postpartum nutrition for lactation (Jenness, 1986; Kliewer and Rasmussen, 1987; Lodge, 1969; O'Grady et al., 1973; Sadurskis, 1988~. This evidence thus supports the notion that gestational weight gain in humans has little impact on subsequent milk quantity or quality. Postpartum Obesity It is often alleged that women in developing countries become pro- gressively malnourished (experience maternal depleiion) and have corre- spondingly worse outcomes with successive pregnancies (Jelliffe, 1966~. In contrast, many investigators report a net increase in body weight among women in industrialized countries during the interconceptional period that may persist and even increase with successive pregnancies. Studies by Stander and Pastore (1940), Beazley and Swinhoe (1979), and Samra et al. (1988) did not control for the expected weight increase that normally occurs with age. In several population-based cross-sectional studies (Forster et al., 1986; Heliovaara and Aromaa, 1981; McKeown and Record, 1957; Newcombe, 1982; Noppa and Bengtsson, 1980), stratification or multivariate statistical approaches have been used to adjust for the

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 203 confounding effect of age and, in some cases, for interpregnancy interval, socioeconomic status, and other potential confounders. But cross-sectional studies are prone to cohort effects; i.e., there has been a trend over time toward higher total body weights at any given age and parity. A recent longitudinal study in The Netherlands (Rookus et al., 1987) reported a slightly (but nonsignificantly) higher increase in BMI at 9 months post partum in 49 pregnant women compared with that in 400 nonpregnant controls followed for the same period. Similarly, in a cross-sectional study (Cederlof and Kaij, 1970) comparing parous monoygotic twins with their childless co-twins, and in longitudinal studies of repeated pregnancies in Scotland (Billewicz and Thomson, 1970) and the United States (Greene et al., 1988), an independent effect of parity on body weight was confirmed. Overall, the evidence suggests an average weight retention of approximately 1 kg (2.2 lb) per birth. The studies in Scotland and the United States are the only ones found by the subcommittee that attempt to relate the magnitude of the parity effect to the amount of weight gained in the preceding pregnancies. Billewicz and Thomson (1970) reported that weight increases (adjusted for age and cohort effects) above 2.5 kg (5.5 lb) between the first and second pregnancies were associated with high weight gains (average 10 to 12 kg, or 22 to 26 lb) after 20 weeks of gestation during the first pregnancy. Greene et al. (1988) reported an analysis of 7,116 women who had at least two singleton births in the 1959-1965 Collaborative Perinatal Project. There was a monotonic trend toward increasing (adjusted) interpregnancy retention of weight with increasing gestational weight gains in the earlier pregnancy. For the minority of women who had very high gestational weight gains, the increases were substantial: 5 kg (10.9 lb) for women gaining 16.4 to 18.2 kg (36 to 40 lb) and 8.0 kg (17.7 lb) for those gaining more than 18.2 kg. (The mean weight gain among the study women was 9.5 kg, or 20.8 lb.) These data should be interpreted with caution, however, since the pregnancies studied occurred nearly four decades ago, when gestational weight gains were considerably lower than those observed more recently (see Chapter 3~. Women gaining >16.4 kg (36 lb) represented only 8% of those studied; that percentage would be far higher today. The fact that the study population was skewed toward black, urban, and poor women and was restricted to the 7,116 women with at least two singleton births during the study period (out of the 58,760 total study population) also limits the generalizabili~ of the findings. In summary, the evidence suggests that women with average gestational weight gains retain about 1 kg (2.2 lb) above and beyond their expected weight increase with age. The 1-kg figure is based largely on data from older studies, however, and may underestimate weight retention associated with the higher gestational weight gains seen in recent years. Women

204 NUTRITIONAL STATUS AND WEIGHT GAIN with very large weight gains appear to be at risk for considerably larger postpartum weight increases. For women who are well- or over-nourished prior to pregnancy, these large increases may contribute to the development of obesity and its adverse health sequelae. Since a given weight increase will have a greater impact on relative weight and, hence, obesity in short women, large weight gains may be particularly undesirable in such women. Further studies are required to document the effects of high gestational weight gain on subsequent maternal obesity. SUMMARY A large body of evidence indicates that gestational weight gain is a determinant of fetal growth, although the magnitude of the causal impact is somewhat less than that usually reported because of the failure of previous studies to adjust total weight gain for fetal weight. Even after such adjustment, however, lower net weight gains are associated with an increased risk of IUGR and increased perinatal mortality (probably mediated by effects on IUGR), whereas higher weight gains are associated with high birth weight and, secondarily, prolonged labor, shoulder dystocia, cesarean delivery, and birth trauma and asphyxia. There is convincing evidence that the effect of maternal weight gain on fetal growth is modified by pregnancy weight for height. Published data do not suggest an effect modification by age or ethnic background. Data concerning the effects of maternal weight gain on gestational duration are suggestive but less conclusive, particularly in light of the difficulties in determining gestational age with accuracy. Further research is clearly indicated in this area, because even small reductions in risk for preterm deliveries, especially those that occur very early in gestation, would have a favorable impact on perinatal and later mortality and on infant and child morbidity and performance. There is little evidence to suggest an important association between gestational weight gain and spontaneous abortion (miscarriage), congeni- tal anomalies, maternal mortality, or lactation performance. There does appear to be a statistical association with PIH and preeclampsia, but it is difficult to interpret this association because of directionality (increased body water leads to increased weight gain) and the absence of data relating PIH to early gestational changes in maternal fat or lean body mass. Pregnancy, in general, and gestational weight gain, in particular, are associated with retained maternal weight post partum. Women with ex- tremely high weight gains during pregnancy may be at increased risk of subsequent obesity.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES CLINICAL IMPLICATIONS 205 · Recommendations for gestational weight gain must be based on an adequate appreciation of potential benefits and risks for fetal growth, perinatal mortality, complications of labor and delivery, and birth trauma and asphyxia. Desirable weight gains are highest in thin women and lowest In obese, overweight, and short women (see Bible 1-1 in Chapter 1~. · Young adolescent and black mothers should be encouraged to strive for weight gains toward the upper range desirable for adult white mothers with similar prepregnancy weights for height and heights. REFERENCES Abrams, B.F., and R.K Laros, Jr. 1986. Prepregnancy weight, weight gain, and birth weight. Am. J. Obstet. Gynecol. 154:503-509. Abrams, B., V. Newman, T. Key, and J. Parker. 1989. Maternal weight gain and preterm delivery. Obstet. Gynecol. 74:577-583. Acker, D.B., B.P. Sachs, and E.^ Friedman. 1985. Risk factors for shoulder dystocia. Obstet. Gynecol. 66:762-768. Anonymous. 1917. Eclampsia rare on war diet in Germany. J. Am. Med. Assoc. 68:732. Arnold, C.C., C.A. Hobbs, R.H. Usher, and M.S. Kramer. 1988. What's wrong with the concept of 'fiery low birth weight" (VLBW)? Pediatr. Res. 23:288A. Arora, N.K., V.K. Paul, and M. Singh. 1987. Morbidity and mortality in term infants with intrauterine growth retardation. J. Stop. Pediatr. 33:186-189. Babson, S.G. 1970. Growth of low-birth-weight infants. J. Pediatr. 77:11-18. Babson, S.G., and D.S. Phillips. 1973. Growth and development of twins dissimilar in size at birth. N. Engl. J. Med. 289:937-940. Beazley, J.M., and R.J. Swinhoe. 1979. Body weight in parous women: is there any alteration between successive pregnancies? Acta Obstet. Gynecol. Scand. 58:45-47. Berkowitz, G.S. 1981. An epidemiologic study of pretend delivery. Am. J. Epidemiol. 113:81-92. Billewicz, W.Z., and A M. Thomson. 1970. Body weight in parous women. Br. J. Prev. Soc. Med. 24:97-104. Boyd, M.E., R.H. Usher, and F.H. McLean. 1983. Fetal macrosomia: prediction, risks, proposed management. Obstet. Gynecol. 61:715-722. Bnend, A. 1985. Do maternal energy reserves limit fetal growth? Lancet 1:38-40. Brown, J.E., K.W. Berdan, P. Splett, M. Robinson, and L^J. Harris. 1986. Prenatal weight gains related to the birth of healthy-sized infants to low-income women. J. Am. Diet. Assoc. 86:1679-1683. Buehler, J.W., AM. Kaunitz, CJ.R. Hogue, J.M. Hughes, J.C. Smith, and R.W. Rochat. 1986. Maternal mortality in women aged 35 years or older United States. J. Am. Med. Assoc. 255:53-57. Butte, N.F., C. Garza, J.E. Stuff, E.O. Smith, and B.L. Nichols. 1984. Effect of maternal diet and body composition on lactational performance. Am. J. Clin. Nutr. 39:296-306. Campbell, D.M., and I. MacGillivray. 1975. The effect of a low calorie diet or a thiazide diuretic on the incidence of pre-eclampsia and on birth-weight. Br. J. Obstet. Gynaecol. 82:572-577. Campbell-Brown, M., and I.R. McFadyen. 1985. Maternal energy reserves and birthweight. Lancet 1:574-575. Cederlof, R., and L. Kaij. 1970. The effect of childbearing on body-weight. Acta Psychiatr. Scand., Suppl. 219:47-49.

206 NUTRITIONAL STAINS AND WEIGHT GAIN Chesley, L C. 1976. Blood pressure, edema and proteinuria in pregnancy. 1. Historical developments. Prog. Clin. Biol. Res. 7:19 66. Duenhoelter, J.H., J.M. Jimenez, and G. Baumann. 1975. Pregnancy performance of patients under fifteen years of age. Obstet. Gynecol. 46:49-52. Duffus, G.M., I. MacGillivray, and KJ. Dennis. 1971. Ibe relationship between baby weight and changes in maternal weight, total body water, plasma volume, electrolytes and proteins, and urinary oestriol excretion. J. Obstet. Gynaecol. Br. Commonw. 78:97-104. Ellenberg, J.H., and KB. Nelson. 1979. Birth weight and gestational age in children with cerebral palsy or seizure disorders. Am. J. Dis. Child. 133:1044-1048. Erhardt, C.L., G.B. Joshi, F.G. Nelson, B.FI. Kroll, and L. Weiner. 1964. Influence of weight and gestation on perinatal and neonatal mortality by ethnic group. Am. J. Public Health 54:1841-1855. Fancourt, R., S. Campbell, D. Harvey, and ~P. Norman. 1976. Follow-up study of small-for-date babies. Br. Med. J. 1:1435-1437. Fitzhardinge, P.M., and S. Inwood. 1989. Long-term growth in small-for-date children. Acta Paediatr. Scand., Suppl. 349:27-33. Fitzhardinge, P.M., and E.M. Steven. 1972. Il~e small-for-date infant. I. Later growth patterns. Pediatrics 49:671-681. Forster, J.L, E. Bloom, G. Sorensen, R.W. Jeffe~y, and RJ. Prineas. 1986. Reproductive history and body mass index in black and white women. Prev. Med. 15:685-691. Frame, S., J. Moore, A. Peters, and D. Hall. 1985. Maternal height and shoe size as predictors of pelvic disproportion: an assessment. Br. J. Obstet. Gynaecol. 92:1239- 1245. Frentzen, B.H., D.L Dimperio, and AC. Cruz. 1988. Maternal weight gain: effect on infant birth weight among overweight and average-weight low-income women. Am. J. Obstet. Gynecol. 159:1114-1117. Frisancho, AR., J.E. Klayman, and J. Matos. 1977. Influence of maternal nutritional status on prenatal growth in a Peruvian urban population. Am. J. Phys. Anthropol. 46:265-274. Frisancho, A.R., J. Matos, W.R. Leonard, and L.A. Yaroch. 1985. Developmental and nutritional determinants of pregnancy outcome among teenagers. Am. J. Phys. Anthropol. 66:247-261. Greene, G.W., H. Smiciklas-Wright, T.O. Scholl, and RJ. Karp. 1988. Postpartum weight change: how much of the weight gained in pregnancy will be lost after delive~y? Obstet. Gynecol. 71:701-707. Gross, S., C. Librach, and A. Cecutti. 1989. Maternal weight loss associated with hyperemesis gravidarum: a predictor of fetal outcome. Am. J. Obstet. Gynecol. 160:906-909. Guaschino, S., A. Spinillo, E. Stola, P.C Pesando, G.P. Gancia, and G. Rondini. 1986. The significance of ponderal index as a prognostic factor in a low-birth-weight population. Biol. Res. Preg. Perinatol. 7:121-127. Haas, J.D., H. Balcazar, and L. Caulfield. 1987. Variation in early neonatal mortality for different types of fetal growth retardation. Am. J. Phys. Anthropol. 73:467-473. Haiek, L., and S.A. Lederman. 1989. The relationship between maternal weight for height and term birth weight in teens and adult women. J. Adol. Health Care 10:16-22. Harrison, G.G., J.N. Udall, and G. Morrow III. 1980. Maternal obesity, weight gain in pregnancy, and infant birth weight. Am. J. Obstet. Gynecol. 136:411-412. Harvey, D., J. Prince, J. Bunton, C Parkinson, and S. Campbell. 1982. Abilities of children who were small-for-gestational-age babies. Pediatrics 69:296-300. Hediger, M.L., T.O. Scholl, D.H. Belsky, I.G. Ances, and R.W. Salmon. 1989. Patterns of weight gain in adolescent pregnancy: effects on birth weight and preterm delive~y. Obstet. Gynecol. 74:6-12. Heliovaara, M., and A. Aromaa. 1981. Parity and obesiW. J. Epidemiol. Community Health 35:197-199.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 207 Herrera, M.G., J.O. Mora, B. de Paredes, and M. Wagner. 1980. Maternal weight/height and the effect of food supplementation during pregnancy and lactation. Pp. 252-263 in H. Aebi and R. Whitehead, eds. Maternal Nutrition during Pregnancy and Lactation. Hans Huber, Bern. Hill, R.M., SUM. Ve~niaud, R.L. Deter, L M. Tennyson, G.M. Rettig, T.E. Zion, A.L. Vorderman, P.G. Helms, L^B. McCulley, and L L. Hill. 1984. The effect of intrauterine malnutrition on the term infant: a 14-year progressive study. Acta Paediatr. Scand. 73:482-487. Hingson, R., JJ. Alpert, N. Day, E. Dooling, H. Kayne, S. Morelock, E. Oppenheimer, and B. Zuckerman. 1982. Effects of maternal drinking and marijuana use on fetal growth and development. Pediatrics 70:539-546. Hoffman, HJ., and L S. Bakketeig. 1984. Heterogeneity of intrauterine growth retardation and recurrence risks. Semin. Perinatol. 8:15-24. Hoffman, H.J., F.E. Lundin, Jr., US. Bakketeig, and E.E. Harley. 1977. Classification of births by weight and gestational age for future studies of prematurity. Pp. 297-333 in D.M. Reed and F.J. Stanley, eds. The Epidemiology of Prematurity. Urban & Schwa~zenberg, Baltimore. Hague, CJ.R., J.W. Buehler, L.T. Strauss, and J.C Smith. 1987. Overview of the National Infant Mortality Surveillance (NIMS) Project-design, methods, results. Public Health Rep. 102:126-138. Horon, I.L., D.M. Strobino, and H.M. MacDonald. 1983. Birth weights among infants born to adolescent and young adult women. Am. J. Obstet. Gynecol. 146:444449. Hughes, A.B., D.A. Jenkins, R.G. Newcombe, and J.F. Pearson. 1987. Symphysis-fundus height, maternal height, labor pattern, and mode of delivery. Am. J. Obstet. Gynecol. 156:644-648. Hytten, F.E., and A.M. Thomson. 1976. Weight gain in pregnancy. Pp. 179-187 in M.D. Lindheimer, HI. Katz, and F.P. Zuspan, eds. Hypertension in Pregnancy. John Wiley & Sons, New York. Jelliffe, D.B. 1966. The Assessment of the Nutritional Status of the Community. WHO Monograph Series No. 53. World Health Organization, Geneva. 271 pp. Jenness, R. 1986. Lactational performance of various mammalian species. J. Dairy Sci. 69:869-885. Kaminski, M., and E. Papiernik. 1974. Multifactorial study of the risk of prematurity at 32 weeks of gestation. II. A comparison between an empirical prediction and a discriminant analysis. J. Perinat. Med. 2 37-44. Kleinman, J.C. 1990. Maternal Weight Gain During Pregnancy Determinants and Conse- quences. NCHS Working Paper Series No. 33. National Center for Health Statistics, Public Health Service, U.S. Department of Health and Human Services, Hyattsville, Md. 24 pp. Kleinman, J.C., and S.S. Kessel. 1987. Racial differences in low birth weight: trends and risk factors. N. Engl. J. Med. 317:749-753. Kliewer, R.L., and K.M. Rasmussen. 1987. Malnutrition during the reproductive cycle: effects on galactopoietic hormones and lactational performance in the rat. Am. J. Clin. Nutr. 46:926-935. Koff, ~K, and E.L Potter. 1939. The complications associated with excessive development of the human fetus. Am. J. Obstet. Gynecol. 38:412423. Koops, B.L~, LO. Morgan, and F.C. Battaglia. 1982. Neonatal mortality risk in relation to birth weight and gestational age: update. J. Pediatr. 101:969-977. Kramer, M.S. 1987. Determinants of low birth weight: methodological assessment and meta-analysis. Bull. W.H.O. 65:663-737. Kramer, M.S., F.H. McLean, M. Olivier, D.M. Willis, and R.H. Usher. 1989. Body proportionality and head and length 'sparing' in growth-retarded neonates: a critical reappraisal. Pediatrics 84:717-723. Langhoff-Roos, J., G. Lindmark, and M. Gebre-Medhin. 1987. Maternal fat stores and fat accretion during pregnancy in relation to infant birthweight. Br. J. Obstet. Gynaecol. 94:1170-1177.

208 NUTRITIONAL STATUS AND WEIGHT GAIN Lawton, F.G., G.C. Mason, Key Kelly, I.N. Ramsay, and G.A. Morewood. 1988. Poor maternal weight gain between 28 and 32 weeks gestation may predict small-for- gestational-age infants. Br. J. Obstet. Gynaecol. 95:884887. Lazar, R. 1981. General commentary. Pp. 181-186 in J. Dabbing, ed. Maternal Nutrition in Pregnancy: Eating for Two? Academic Press, London. Lechtig, A, J.P. Habicht, H. Delgado, R.E. Klein, C. Yarbrough, and R. Martorell. 1975. Effect of food supplementation during pregnancy on birthweight. Pediatrics 56:508-520. Lodge, G.A. 1969. The effects of pattern of feed distribution during the reproductive cycle on the performance of sows. Anim. Prod. 11:133-143. Low, J.A., R.S. Galbraith, D. Muir, H. Killen, B. Pater, and J. Karchmar. 1982. Intrauterine growth retardation: a study of long-term morbidity. Am. J. Obstet. Gynecol. 142:670- 677. Lubchenco, LO., D.T. Searls, and J.V. Brazie. 1972. Neonatal mortality rate: relationship to birth weight and gestational age. J. Pediatr. 81:814-822. Luke, B., C. Dickinson, and R.H. Petrie. 1981. Intrauterine growth: correlations of maternal nutritional status and rate of gestational weight gain. Eur. J. Obstet., Gynecol. Reprod. Biol. 12:113-121. Maso, M.J., EJ. Gong, M.S. Jacobson, D.S. Bross, and F.P. Heald. 1988. Anthropometric predictors of low birth weight outcome in teenage pregnancy. J. Adol. Health Care 9:188-193. McCormick, M.C. 1985. The contribution of low birth weight to infant mortality and childhood morbidity. N. Engl. J. Med. 31282-90. McKeown, T., and R.G. Record. 1957. The influence of reproduction on body weight in women. J. Endocrinol. 15:393-409. Metropolitan Life Insurance Company. 1959. New weight standards for men and women. Stat. Bull. Metrop. Life Insur. Co. 40:1-4. Miller, H.C., and T.^ Merritt. 1979. Fetal Growth in Humans. Year Book Medical Publishers, Chicago. 180 pp. Mitchell, M.C., and E. Lerner. 1987. Factors that influence the outcome of pregnancy in middle-class women. J. Am. Diet. Assoc. 87:731-735. Mitchell, M.C., and E. Lerner. 1989. Weight gain and pregnancy outcome in underweight and normal weight women. J. Am. Diet. Assoc. 89:634-638. Modanlou, H.D., W.L Dorchester, A. Thorosian, and R.K Freeman. 1980. Macrosomia maternal, fetal, and neonatal implications. Obstet. Gynecol. 55:420-424. Mora, J.O., B. de Paredes, M. Wagner, L~ de Navarro, J. Suescum, N. Christiansen, and M.G. Herrera. 1979. Nutritional supplementation and the outcome of pregnancy. I. Birth weight. Am. J. Clin. Nutr. 32:455-462. Naeye, R.L 1979. Weight gain and the outcome of pregnancy. Am. J. Obstet. Gynecol. 135:3-9. Naeye, R.L~ 1981a. Maternal blood pressure and fetal growth. Am. J. Obstet. Gynecol. 141:780-787. Naeye, R.L~ 1981b. Maternal nutrition and pregnancy outcome. Pp. 89-111 in J. Dobbing, ed. Maternal Nutrition in Pregnancy: Eating for Two? Academic Press, London. Naeye, R.L" 1981c. Nutritional/nonnutritional interactions that affect the outcome of pregnancy. Am. J. Clin. Nutr. 34:727-731. Naeye, R.L~ 1981d. Teenaged and pre-teenaged pregnancies: consequences of the fetal- maternal competition for nutrients. Pediatrics 67:146-150. Naeye, R.L~, and R.A. Chez. 1981. Effects of maternal acetonuria and low pregnan~y weight gain on children's psychomotor development. Am. J. Obstet. Gynecol. 139:189-193. NCHS (National Center for Health Statistics). 1987. Vital Statistics of the United States, 1983. Vol. ~Mortality, Part A. DHHS Publ. No. (PHS) 87-1102. National Center for Health Statistics, Public Health Service, U.S. Department of Health and Human Services, Hyattsville, Md. 713 pp. Neligan, G.A., I. Kolvin, D.McI. Scott, and R.F. Garside. 1976. Born Too Soon or Born Too Small: A Follow-up Study to Seven Years of Age. Clinics in Developmental Medicine, No. 61. Heineman, London. 101 pp.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 209 Newcombe, R.G. 1982. Development of obesity in parous women. J. Epidemiol. Community Health 36:306-309. Noppa, H., and C Bengtsson. 1980. Obesity in relation to socioeconomic status: a population study of women in Goteborg, Sweden. J. Epidemiol. Community Health 34:139-142. NRC (National Research Council). 1970. Maternal Nutrition and the Course of Pregnancy. Report of the Committee on Maternal Nutrition, Food and Nutrition Board. National Academy of Sciences, Washington, D.C. 241 pp. O'Grady, J.F., F.W.H. Elsley, R.M. MacPherson, and I. McDonald. 1973. The response of lactating sows and their litters to different dietary energy allowances. 1. Milk yield and composition, reproductive performance of sows and growth rate of litters. Anim. Prod. 17:65-74. Ounsted, M., and A. Scott. 1981. Associations between maternal weight, height, weight- for-height, weight-gain and birth weight. Pp. 113-129 in J. Dobbing, ed. Maternal Nutrition in Pregnancy: Eating for Two? Academic Press, London. Ounsted, M., and M.E. Taylor. 1971. The postnatal growth of children who were small-for-dates or large-for-dates at birth. Dev. Med. Child Neurol. 13:421-434. Ounsted, M., V. Moar, and SUA. Scott. 1981. Perinatal morbidity and mortality in small- for-date babies: the relative importance of some maternal factom. Early Hum. Dev. 5:367-375. Ounsted, M.K., V.A. Moar, and A. Scott. 1984. Children of deviant birthweight at the age of seven years: health, handicap, size and developmental status. Early Hum. Dev. 9:323-340. Ounsted, M., V.A. Moar, and A. Scott. 1988. Neurological development of small-for- gestational age babies during the first year of life. Early Hum. Dev. 16:163-172. Papiernik, E., and M. Kaminski. 1974. Multifactorial study of the risk of prematurity at 32 weeks of gestation. I. A study of the frequency of 30 predictive characteristics. J. Perinat. Med. 2:30-36. Paul, NA., E.M. Muller, and R.G. Whitehead. 1979. The quantitative effects of maternal dietary energy intake on pregnancy and lactation in rural Gambian women. Trans. R. Soc. Itop. Med. Hyg. 73:686-692. Pebley, A.R., S.L. Huffman, A.K.M.A. Chowdhury, and P.\U Stupp. 1985. Intra-utenne mortality and maternal nutritional status in rural Bangladesh. Popul. Stud. 39:425-440. Picone, T.A., L.H. Allen, P.N. Olsen, and M.E. Ferris. 1982. Pregnancy outcome in North American women. II. Effects of diet, cigarette smoking, stress, and weight gain on placentas, and on neonatal physical and behavioral characteristics. Am. J. Clin. Nutr. 36:1214-1224. Pipe, N.G.J., T. Smith, D. Halliday, CJ. Edmonds, C. W~lliams, and T.M. Coltart. 1979. Changes in fat, fat-free mass and body water in human normal pregnancy. Br. J. Obstet. Gynaecol. 86:929-940. Placek, P.J., and S.M. Taffel. 1980. [lends in cesarean section rates for the United States, 1970-78. Public Health Rep. 95:540-548. Placek, P.J., S. Taffel, and M. Moien. 1983. Cesarean section delivery rates: United States, 1981. Am. J. Public Health 73:861-862. Prentice, AM., and R.G. Whitehead. 1987. The energetics of human reproduction. Symp. Zool. Soc. London 57:275-304. Prentice, AM., R.G. Whitehead, M. Watkinson, W.H. Lamb, and T.J. Cole. 1983. Prenatal dietary supplementation of A~ican women and birth-weight. Lancet 1:489-492. Ribeiro, M.D., Z. Stein, M. Susser, P. Cohen, and R. Neugut. 1982. Prenatal starvation and maternal blood pressure near delive~y. Am. J. Clin. Nutr. 35:535-541. Risch, H.A., N.S. Weiss, E.A. Clarke, and A.B. Miller. 1988. Risk factors for spontaneous abortion and its recurrence. Am. J. Epidemiol. 128:420-430. Rookus, M.A., P. Rokebrand, J. Burema, and P. Deurenberg. 1987. The effect of pregnan~y on the body mass index 9 months postpartum in 49 women. Int. J. Obesity 11:609-618. Rosso, P. 1985. A new chart to monitor weight gain during pregnancy. Am. J. Clin. Nutr. 41:644-652.

210 NUTRITIONAL STATUS AND WEIGHT GAIN Rubin, R.A., C. Rosenblatt, and B. Balow. 1973. Psychological and educational sequelae of prematurity. Pediatrics 52:352-363. Sachs, B.P., D.AJ. Brown, S.G. Driscoll, E. Schulman, D. Acker, BJ. Ransil, and J.F. Jewett. 1987. Maternal mortality in Massachusetts: trends and prevention. N. Engl. J. Med. 316:667~72. Sadurskis, A. 1988. Energy Costs of Pregnant and Lactating Females in Relation to Nutritional Status and Energy Intake. Studies in Humans and Rats. Dissertation, Department of Medical Nutrition, Karolinska Institute, Stockholm, Sweden. 60 pp. Sadurskis, A., N. Kabir, J. Wager, and E. Forsum. 1988. Energy metabolism, body composition, and milk production in healthy Swedish women during lactation. Am. J. Clin. Nutr. 48:4449. Samra, J.S., L.CH. Tang, and M.S. Obhrai. 1988. Changes in body weight between consecutive pregnancies. Lancet 2 1420-1421. Sandmire, H.F., and T.J. O'Halloin. 1988. Shoulder dystocia: its incidence and associated risk factors. Int. J. Gynaecol. Obstet. 26:65-73. Scholl, T.O., E. Decker, RJ. Karp, G. Greene, and M. De Sales. 1984. Early adolescent pregnancy: a comparative study of pregnancy outcome in young adolescents and mature women. J. Adol. Health Care 5:167-171. Scholl, T.O., R.W. Salmon, L.K Miller, P. Vasilenko III, CH. Furey, and M. Christine. 1988. Weight gain during adolescent pregnancy: associated maternal characteristics and effects on birth weight. J. Adol. Health Care 9.286 290. Scholl, T.O., M.L. Hediger, R.W. Salmon, D.H. Belsky, and I.G. Ances. 1989. Influence of prepregnant body mass and weight gain for gestation on spontaneous preterm delive~y and duration of gestation during adolescent pregnancy. Am. J. Hum. Biol. 1:657-664. Seidman, D.S., P. Ever-Hadani, and R. Gale. 1989. The effect of maternal weight gain in pregnancy on birth weight. Obstet. Gynecol. 74:240-246. Shapiro, S., M.C. McCormick, B.H. Starfield, J.P. Krischer, and D. Bross. 1980. Relevance of correlates of infant deaths ~r significant morbidity at 1 year of age. Am. J. Obstet. Gynecol. 136:363-373. Shepard, M.J., K.G. Hellenbrand, and M.B. Bracken. 1986. Proportional weight gain and complications of pregnancy, labor, and deliver~r in healthy women of normal prepregnant stature. Am. J. Obstet. Gynecol. 155:947-954. Singer, J.E., M. Westphal, and K. Niswander. 1968. Relationship of weight gain during pregnan~r to binh weight and infant growth and development in the Srst year of life: a report from the Collaborative Study of Cerebral Palsy. Obstet. Gynecol. 31:417-423. Smith, C.^ 1947. The effect of wartime starvation in Holland upon pregnancy and its product. Am. J. Obstet. Gynecol. 53:599-608. Stander, H.J., and J.B. Pastore. 1940. Weight changes during pregnancy and puerperium. Am. J. Obstet. Gynecol. 39:928-937. Stein, ~ 1989. Maternal pre-pregnant nutritional status and the risk for spontaneous abortion. Master's Essay. Columbia University School of Public Health, New York. Stein, ~, M. Susser, G. Saenger, and F. Marolla. 1975. Famine and Human Development: The Dutch Hunger W~nter of 1944-1945. Oxford University Press, New York. 284 pp. Ta~el, S.M. 1986. Maternal Weight Gain and the Outcome of Pregnancy: United States, 1980. Vital and Health Statistics, Series 21, No. 44. DHHS Publ. No. (PHS) 86-1922. National Center for Health Statistics, Public Health Service, U.S. Department of Health and Human Se~vices, Hyattsville, Md. 25 pp. Taffel, S.M., and K.G. Keppel. 1986. Advice about weight gain during pregnangy and actual weight gain. Am. J. Public Health 76:1396-1399. Tavris, D.R., and J.A. Read. 1982. Effect of maternal weight gain on fetal, infant, and childhood death and on cognitive development. Obstet. Gynecol. 60:689-694. Thomson, A.M., and W.Z. Billewicz. 1957. Clinical significance of weight trends during pregnancy. Br. Med. J. 1:24~247. Thomson, A.M., F.E. Hytten, and W.Z. Billewicz. 1967. The epidemiology of oedema during pregnangy. J. Obstet. Gynaecol. Br. Commonw. 74:1-10. Tompkins, W.T., and D.G. W~ehl. 1951. Nutritional deficiencies as a causal factor in toxemia and premature labor. Am. J. Obstet. Gynecol. 62:898-919.

GESTATIONAL WEIGHT GAIN IN SINGLETON PREGNANCIES 211 Tompkins, W.T., D.G. Stahl, and R.M. Mitchell. 1955. The underweight patient as an increased obstetric hazard. Am. J. Obstet. Gynecol. 69:114-123. Udall, J.N., G.G. Harrison, Y. Vaucher, P.D. Walson, and G. Morrow III. 1978. Interaction of maternal and neonatal obesity. Pediatrics 62:17-21. Usher, R.H. 1970. Clinical and therapeutic aspects of fetal malnutrition. Pediatr. Clin. North Am. 17:169-183. van den Berg, BJ., and F.W. Oechsli. 1984. Prematurity. Pp. 69-85 in M.B. Bracken, ed. Perinatal Epidemiology. Oxford University Press, New York. van Steenbergen, W.~., J.N Kusin, S. Kardjati, and C. de With. 1989. Energy supplemen- tation in the last trimester of pregnancy in East Java, Indonesia: effect on breast-milk output. Am. J. Clin. Nutr. 50:274-279. Varma, OR. 1984. Maternal weight and weight gain in pregnancy and obstetric outcome. Int. J. Gynaecol. Obstet. 22:161-166. Viegas, O.NC., P.H. Scott, TJ. Cole, P. Eaton, P.G. Needham, and B.^ Wharton. 1982. DietaIy protein energy supplementation of pregnant Asian mothem at Sorrento, Birmingham. II. Selective during third trimester only. Br. Med. J. 285:592-595. Viegas, O.AC., T.J. Cole, and B.A. Wharton. 1987. Impaired fat deposition in pregnancy: an indicator for nutritional intervention. Am. J. Clin. Nutr. 45:23-28. Villar, J., V. Smeriglio, R. Martorell, C.H. Brown, and R.E. Klein. 1984. Heterogeneous growth and mental development of intrauterine growth-retarded infants during the first 3 years of life. Pediatrics 74:783-791. Walther, FJ. 19~. Growth and development of term disproportionate small-for-gestational age infants at the age of 7 years. Early Hum. Dev. 18:1-11. Walther, F.J., and L.H.J. Ramaekers. 1982. Growth in early childhood of newborns affected by disproportionate intrauterine growth retardation. Acta Paediatr. Scand. 71:651-656. Westwood, M., M.S. Kramer, D. Munz, J.M. Lovett, and G.V. Watters. 1983. Growth and development of full-tenn nonasphyxiated small-for-gestational-age newborns: follow-up through adolescence. Pediatrics 71:376-382. WHO (World Health Organization). 1965. Nutrition in Pregnancy and I~ctation. Report of a WHO Expert Committee. Technical Report Series No. 302. World Health Organization, Geneva. 54 pp. W~nikoff, B., and C.H. Debrovner. 1981. Anthropometric determinants of birth weight. Obstet. Gynecol. 58:678-684. Yerushalmy, J., B.J. van den Berg, C.L. Erhardt, and H. Jacobziner. 1965. Birth weight and gestation as indices of "immaturity". Am. J. Dis. Child. 109:43-57. Ylitalo, V., P. Kero, and R. Erkkola. 1988. Neurological outcome of twins dissimilar in size at birth. Early Hum. Dev. 17:245-255.