| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Psychology and Harlow Center for Biological Psychology, University of Wisconsin, Madison, Wisconsin.
Address reprint requests to: Kimberly C. Price, PhD, Harlow Primate Laboratory, 22 North Charter Street, Madison, WI 53715, E-mail: kcprice{at}facstaff.wisc.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: A multigenerational data set was used to evaluate the relationship between familial and contemporaneous pregnancy factors and infant birth weight across several generations. These records provided 25 years of information on the maternal and paternal ancestries and reproductive histories, gestation lengths, and birth weights for 1321 infants.
Results: Pregnancy weight gain, gestation length, and maternal familial factors were the most important predictors of infant birth weight, followed by infant sex, paternity, and maternal pregravid weight (P < .001 for each variable). Furthermore, the trend in fetal growth across generations followed a matrilineal pattern of transmission that was much more pronounced for female than male offspring (P < .001). Although secular increases in maternal pregravid weight and pregnancy weight gain were detected, the upward shift in female birth weight was not explained solely by these changes in maternal weight parameters.
Conclusion: With the delivery of ample nutrition and health care in a laboratory setting, there was a dramatic increase in the birth weight of daughters within certain matrilines, providing evidence that an intrauterine mechanism transmitted through female progeny can regulate fetal development. Further, the upward trend in female birth weight had a beneficial influence on the reproductive performance of female descendants in those lineages.
Although both maternal and paternal genetic factors clearly have an important influence in determining infant birth weight,1–4 familial patterns in birth weight appear to operate primarily through the mother’s lineage. Because the mother also provides the gestational environment, she has a greater capacity than the father to modulate the expression of the fetus’s inherited growth potential.5–12 Pregnancy variables, such as pre-gravid weight and maternal weight gain, are strongly associated with infant birth weight and presumably reflect the mother’s health and nutrition status at the time of conception and during gestation.13–19 Demographic factors can also have a general effect on birth weights across populations, as economic forces and emigration patterns have been associated with both positive and negative trends in birth weight over several generations.20–23 Changes in maternal fitness and stature resulting from circumstantial improvements or declines may be partially responsible for these population-level shifts. To better understand how changes in the environment, mediated through certain maternal reproductive features, influence birth weight, the following study was designed to analyze familial trends across several generations. Further, we tested the hypothesis that female offspring were more important than male offspring for perpetuating this matrilineal predominance of birth weight patterns.
Previous attempts to establish the relative importance of generational, familial, and contemporaneous influences on birth weight have been limited by formidable methodologic challenges. Specifically, data collected on human populations typically extend only one to three generations and often do not include thorough reproductive histories or birth weights of distant relatives. Large-scale breeding programs with animals provide unique opportunities for investigating the interrelationships among these maternal attributes and birth weight. By using an extensive data set acquired from our long-established colony of rhesus monkeys, we conducted a comprehensive analysis of factors known to influence birth weight in humans.
The rhesus monkey (Macaca mulatta) is used widely to model human reproductive processes because of its long gestation (6 months), predominantly singleton births, and many similarities with human placental and endocrine physiology.24 The genealogies, the reproductive histories, and accurate gestational ages of our study population span more than 40 years, and were available for most animals. Because our records included information from five generations of monkeys derived from a relatively small feral population, we had the novel opportunity to explore how generational increases in the birth weights of laboratory-born monkeys, due to the provision of ample nutrition and health care, operated through features characterizing the matriline. Specifically, we sought to develop a descriptive model that quantified the contributions of the environmental, familial, and pregnancy-specific variables to infant birth weight, and to assess how changes in maternal condition (eg, maternal pregravid weight, weight gain during pregnancy) contributed to the intergenerational transmission of birth weight in each matriline.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Typically the neonate was weighed on the day of birth; weights measured after the first day were adjusted according to a curve-fitting formula that provided the best-fit estimate of the likely birth weight. The data from nearly 200 infants weighed after the first week of life were excluded. As the objective of this analysis was to examine the normative transmission patterns of maternal and infant weight variables across generations, and not to estimate risk for unfavorable outcomes, we measured these associations using only offspring born appropriate for gestational age (AGA). Thus, monkeys born with weights in the upper (n = 70) and lower (n = 78) fifth percentile for their sex and gestational age were eliminated from the sample, thereby excluding a priori potential genetic or disease factors associated with fetal growth restriction (FGR) or acceleration.
To evaluate the intergenerational patterns of birth weight, each infant was assigned to a category based on the number of matrilineal (n = 5) and patrilineal (n = 4) generations that separated it from its wild-caught progenitor. In addition, as some changes in nutrition and health care delivery likely occurred during the study period, the infant’s year of birth was included as a continuous variable to control for the secular or nonfamilial component of the temporal trends in birth weight. The incidence of small for gestational age (SGA) births, large for gestational age (LGA) births, and neonatal mortality in this population did not vary by generation in captivity; in contrast, the rate of cesarean delivery decreased markedly with each generation (unpublished observations).
The familial variables were created by tracing the infant’s maternal and paternal lineage to its maternal (n = 128) and paternal (n = 59) progenitor, respectively. A matriline was defined as one of 76 maternal families having intergenerational birth weight data from four or more infants; similarly, a patriline was defined as one of 37 paternal families having intergenerational birth weight data from six or more infants. Each infant was then assigned a rank equivalent to the mean birth weight of the lineage, excluding its own weight. The remaining predictors included maternal pregravid weight and pregnancy weight gain (to the nearest 0.05 kg), maternal age at parturition (to the nearest month), parity (with births 1–5 coded as zero and births 6–15 coded as one), infant sex, and gestational age (to the nearest day). The father’s adult stature, operationally defined as the sire’s weight at 10 years of age, was not significantly associated with either infant birth weight or patrilineal identity for this population, so it was not considered in the present analysis.
We conducted a preliminary descriptive evaluation of the sources of variation in infant birth weight. Analysis of variance and analysis of covariance procedures were used for means comparisons, with Scheffé’s posthoc tests used for follow-up analyses; bivariate relationships were tested using Pearson’s product-moment correlation. Next, multivariate regression techniques were used to establish the contribution of each predictor as a unique source of variation in birth weight. For the purposes of this analysis, maternal and paternal generation, and matriline and patriline were treated as continuous variables. Before the regression procedure, the variables were examined for the accuracy of data entry, missing values, and fit between their distributions and the assumptions of multivariate analysis. Because exact ages were not available for feral mothers, maternal age was not included in the regression model. To improve the linearity of the maternal generation variable, a power transformation was applied before the variable was entered in the regression analysis. Finally, to test whether the intergenerational trends in birth weight varied by infant sex, a sex-by-maternal generation interaction term was created and entered in the final step of the analysis.
Next, to evaluate the associations among the predictors, we used the information from the descriptive and regression analyses to derive a theoretic model in which the variables were ordered according to their temporal relationships with one another. Year of birth, maternal generation, patriline, and infant sex were considered exogenous variables as they were not influenced by any other variables in the model; next, matriline, parity, pregravid weight, gestational age, and weight gain were analyzed in order. Because of the colinearity between maternal age and parity (r = 0.92), parity was entered as a continuous variable to approximate the relationships between maternal age and the infant, maternal, and temporal variables. At each step, the output variable was regressed on preceding predictors, with significant associations retained in the final model. All analyses were conducted using the Statistical Package for the Social Sciences, Version 8.0 (SPSS Inc, Chicago, IL), with significance established at P < .05.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
and 2). Year of birth was positively associated with infant birth weight and with maternal and paternal generation in captivity (r = .71 and .72, respectively, P < .001 for both). However, after entering year of birth as a covariate to control for this temporal component, only the significant relationship between maternal generation and infant birth weight remained (rpartial = .10, P < .001). In addition, the mean age at first birth for laboratory-born mothers decreased 6 months, on average, with each successive maternal generation (r = -.35, P < .001), yet the maternal weight at first conception remained stable (r = -.01, P = .89), suggesting that the females attained adult size at younger ages with each generation. In addition, a mother’s age at first birth was inversely associated with her own birth weight (r = -.30, P < .001). Finally, gestation length also increased modestly with maternal generation (r = .13, P < .001), and this difference remained after controlling for year of birth and maternal size (rpartial = .15, P < .001).
|
|
, P < .001 for both comparisons). Maternal pregravid weights ranged from 3.8 to 10.4 kg, with older mothers of higher parity weighing more than younger mothers (r = .45 for maternal age and r = .39 for parity, P < .001 in both cases). Changes in maternal weight across pregnancy ranged from -1.35 to 3.85 kg. The magnitude of absolute weight gained was unrelated to pregravid weight, maternal age, or parity, although lower conception weights were associated with higher percent body weight gains during pregnancy (r = -0.41, P < .001). The weight increment gain ranged from -18% to 73%, with a mean of 28%.
Male infants were significantly heavier than females at birth (F [1,1318] = 92.39, P < .001). However, this pattern was not observed for fourth- and fifth-generation offspring because of the dramatic increase in female birth weights (F [4,1309] = 3.50, P < .01 for the interaction between sex and maternal generation; Figure 1). Viable pregnancies in our sample varied in length from 146 to 184 days, with 95% of all pregnancies within the range of 160 to 177 days. The mean gestational age of males was nearly a day longer than that of females (F [1,1130] = 6.86, P < .01). Finally, later-born offspring weighed significantly more than early-born infants did (F [1,1318] = 7.78, P = .005), although this pattern was not observed for first-generation animals (F [4, 1309] = 2.41, P < .05 for the interaction between parity and maternal generation).
|
. The R2adj indicates that together the predictors explained approximately 53% of the variation in birth weight, whereas the semipartial R2 represents the unique variance in birth weight accounted for by each independent variable once the linear effects of the others were removed. According to this measure, maternal weight gain is the strongest predictor of infant birth weight, followed by gestational age, matrilineal identity, infant sex, patrilineal identity, pregravid weight, and maternal generation. Of the temporal variables, only maternal generation in captivity was an important predictor of birth weight, and this association held even after the addition of the sex-by-maternal generation interaction term, which also retained a statistically significant relationship with birth weight in the final model.
The full model derived from the regression procedures is depicted in Figure 2. The numbers represent the significant (P < .05) standardized coefficients (ß) derived from the regression of the output variable on the preceding predictors. Positive and negative values indicate direct and inverse relationships between variables, respectively, with larger numbers (both positive and negative) denoting stronger associations. Although the regression of birth weight on year of birth did not yield a significant association, the relationships of birth weight with pregravid weight and with maternal weight gain suggest that year of birth exerts an indirect influence operating through these other predictors. Similarly, maternal weight at conception and gestational weight gain were positively associated with maternal family of origin, but not with maternal generation, suggesting that they were relatively stable characteristics of the matriline from one generation to the next.
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
There is a tendency for women to bear infants with similar birth weights and gestational ages across successive pregnancies26; further, mothers who deliver an LGA or SGA infant are likely to produce another such birth.27–29 These familial patterns in birth weight across multiple generations appear to operate primarily through the maternal lineage, suggesting a stronger maternal than paternal familial influence on birth weight.5–9 Our data corroborate those findings: the maternal family of origin was one of the most important predictors of birth weight, accounting for 60% more of the variance than the paternal family. This pattern is even more striking given the multigenerational index we developed to represent maternal and paternal pedigrees. Typically, only the parents’ birth weights are used to estimate maternal and paternal familial effects; our measure included all animals related through the same progenitor, thereby allowing us to evaluate this association among members of an entire lineage.
Our finding of a significant influence of the mother’s generation in captivity further illustrates the importance of the maternal line and emphasizes the potential role of the intrauterine environment in producing these birth weight patterns. Comparable trends in birth weight have been reported for humans, with dramatic increases observed in infants of immigrant mothers who relocated to industrialized countries.22,23 However, the increase in birth weight observed in this monkey population was considerably larger for the female offspring and occurred independent of transgenerational increases in the mothers’ weight characteristics, suggesting that maternal physiology might be responding to environmental changes in a manner that regulates fetal growth rate. Thirty years ago, Ounsted and Ounsted proposed that intrauterine processes can govern fetal growth and that the set point for this maternal regulator was determined by the mother’s own gestational experience.10,11 Such a system would favor infant survival as it provides for the maternal-fetal transfer of information regarding resource availability, thereby programming the infant’s developmental trajectory and energy requirements to meet the demands of the external environment. An adjustable uterine mechanism could modify the rate of fetal growth in response to an improvement or deterioration in local conditions and allow for moderately fast birth weight trends in either direction. As female offspring provide the gestational milieu for the subsequent generation, a uterine mechanism that selectively governs the growth of daughters in response to variations in their living conditions could offer the most efficient means for selection toward the optimal birth weight sustainable by the environment. Finally, because sisters share a common intrauterine environment, shifts in constraint would persist in the female descendants of the matriarch, thereby providing for the manifestation of a matrilineal birth weight pattern across generations.
In this maternal constraint model, these intrauterine processes largely determine the rate of development for growth-restricted infants; with the loosening of constraint with larger offspring, other factors, such as paternal inheritance, will account for more of the variance in birth weight.10,11 The increase in gestation length with successive generations might also be a manifestation of changes in maternal constraint, because the mothers’ ability to sustain longer gestation and thus to bear larger offspring increases with each generation. This phenomenon might partially explain the upward shift in the mean birth weight of the population despite the stability of the SGA and LGA birth rate across maternal generation in captivity. Importantly, the delivery of larger infants has not resulted in greater pregnancy complications in this population, as the incidence of neonatal mortality and cesarean delivery have both decreased with each generation. It should be reiterated, however, that the increase in gestational age was equivalent for male and female infants and therefore does not completely account for the upward trend in birth weight for female monkeys.
Although the term "maternal constraint" might be taken to imply that the mother is physically incapable of carrying a fetus too large for her to bear, data from human studies indicate that pregravid weight does not differ significantly between mothers who deliver SGA and AGA infants.11 Rather, intrauterine constraint possibly reflects altered maternal metabolic processes or uterine-to-placental transport mechanisms that limit the provision of nutrients to the fetus. For example, women, who as fetuses were undernourished during the first trimester but nonetheless had normal birth weight, produced offspring with aberrant birth weight patterns.29 Thus, a mother’s own gestational experience, and not simply her birth weight per se, might influence the uterine environment she provides for her own offspring, thereby creating a link between her fetal development and that of the next generation. Further, women born SGA are not only more likely to repeat the pattern with their own children, but they are also at increased risk for several other adverse obstetrical outcomes, including preterm delivery and neonatal death.4–9 The present study emphasized familial transmission patterns among AGA infants, but subsequent analyses that included infants with deviant birth weight from this population found that both FGR and fetal growth acceleration are traits that follow the maternal, but not paternal, lineage (data not shown). Further, the risk of prematurity, low birth weight, and stillbirth was significantly higher for the offspring of rhesus mothers born SGA, particularly if the infant was female (data not shown).
Gestational weight gain is a strong and consistent predictor of birth weight in humans.14–19 The importance of this factor in the monkey was emphasized in the present analysis, as it alone accounted for nearly 13% of the variation in birth weight and 24% of the variance explained by our statistical model. Both pre-gravid body mass and gestational weight gain have independent, positive effects on infant birth weight and presumably reflect maternal nutritional status before and during pregnancy.13–19 These indices of maternal well-being could have important implications for fetal growth and development.17,19 Additionally, these traits were relatively stable features of the matrilines comprising this population, as both were significantly associated with the maternal family of origin. Pregravid weight and weight gain during pregnancy were not positively associated with generation, suggesting that environmental factors operating throughout the mothers’ lifetimes produced the increases in these maternal measures. Interestingly, maternal weight at first conception was not different with each generation in captivity, despite the earlier onset of reproduction with each successive generation. As maternal birth weight is inversely associated with age at first birth, the earlier attainment of reproductive maturity itself could be an intergenerational phenomenon linked to the increasing trend in female birth weight, which in turn prompts a more rapid rate of early growth and development.
Some investigators raised the concern that access to and utilization of adequate prenatal care has not reduced the risk of repeating a low-birth-weight delivery or mitigated the impact of pregnancy complications on birth weight.28,30 If there is an intergenerational component, then concentrating solely on what transpires during the relatively short period of one pregnancy will be of only modest benefit for correcting such enduring reproductive health disadvantage. The birth weight trends observed in our monkey population were evident only after three generations of laboratory care, leading to the conclusion that important progress depends on the longer-term investment in maternal and infant health care.
|
Footnotes |
|---|
Received October 7, 1998. Received in revised form December 30, 1998. Accepted January 13, 1999.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Little RE. Mother’s and father’s birthweight as predictors of infant birthweight. Paediatr Perinat Epidemiol 1987;1:19–31.[Medline]
3. Alberman E, Emanuel I, Filakti H, Evans SJW. The contrasting effects of parental birthweight and gestational age on the birth-weight of offspring. Paediatr Perinat Epidemiol 1992;6:134–44.[Medline]
4. Coutinho R, David RJ, Collins, JW Jr. Relation of parental birth weights to infant birth weight among African Americans and whites in Illinois: A transgenerational study. Am J Epidemiol 1997;146:804–9.
5. Johnstone F, Inglis L. Familial trends in low birth weight. BMJ 1974;3:659–61.
6. Hackman E, Emanuel I, van Belle G, Daling J. Maternal birth weight and subsequent pregnancy outcome. JAMA 1983;250: 2016–9.[Abstract]
7. Klebanoff MA, Graubard BI, Kessel SS, Berendes HW. Low birth weight across generations. JAMA 1984;252:2423–7.[Abstract]
8. Klebanoff MA, Meirik O, Berendes HW. Second-generation consequences of small-for-dates birth. Pediatrics 1989;84:343–7.
9. Emanuel I, Filakti H, Alberman E, Evans SJW. Intergenerational studies of human birthweight from the 1958 birth cohort. 1. Evidence for a multigenerational effect. Br J Obstet Gynaecol 1992;99:67–74.[Medline]
10. Ounsted M, Ounsted C. Rate of intrauterine growth. Nature 1968;220:599–600.[Medline]
11. Ounsted M, Ounsted C. On fetal growth rate. In: Clinics in developmental medicine. No. 46. London: Heinemann, 1973:72–82.
12. Skjærven R, Wilcox AJ, Øyen N, Magnus P. Mothers’ birth weight and survival of their offspring: Population based study. BMJ 1997;314:1376–80.
13. Peckham CH, Christianson RE. The relationship between prepregnancy weight and certain obstetric factors. Am J Obstet Gynecol 1971;111:1–7.[Medline]
14. Naeye RL. Weight gain and the outcome of pregnancy. Am J Obstet Gynecol 1979;135:3–9.[Medline]
15. Anderson GD, Blidner IN, McClemont S, Sinclair JC. Determinants of size at birth in a Canadian population. Am J Obstet Gynecol 1984;150:236–44.[Medline]
16. Abrams BF, Laros RK. Prepregnancy weight, weight gain, and birth weight. Am J Obstet Gynecol 1986;154:503–9.[Medline]
17. Luke B. Nutritional influences on fetal growth. Clin Obstet Gynecol 1994;37:538–49.[Medline]
18. Cogswell ME, Yip R. The influence of fetal and maternal factors on the distribution of birthweight. Semin Perinatol 1995;19:222–40.[Medline]
19. Li R, Haas JD, Habicht JP. Timing of the influence of maternal nutritional status during pregnancy on fetal growth. Am J Hum Biol 1998;10:529–39.
20. Guendelman F, Gould JB, Hudes M, Eskenazi B. Generational differences in perinatal health among the Mexican American population: Findings from NHANES. Am J Public Health 1990; 80(Suppl):61–5.
21. Alberman E. Are our babies becoming bigger? J R Soc Med 1991;84:257–60.[Abstract]
22. Dhawan S. Birth weights of infants of first generation Asian women in Britain compared with second generation Asian women. BMJ 1995;311:86–8.
23. Chike-Obi U, David RJ, Coutinho R, Wu SY. Birth weight has increased over a generation. Am J Epidemiol 1996;144:563–9.
24. Stolte LAM. Pregnancy in the rhesus monkey. In: Bourne GH, ed. The rhesus monkey: Management, reproduction and pathology. New York: Academic Press, Inc, 1975:171–230.
25. Kemnitz JW. Effects of gender and age on body size and composition of rhesus monkeys (Macaca mulatta). Am J Primatol 1994;33: 220.
26. Bakketeig LS, Hoffman HJ, Harley EE. The tendency to repeat gestational age and birth weight in successive births. Am J Obstet Gynecol 1979;135:1086–103.[Medline]
27. Bakketeig LS, Hoffman HJ, Jacobsen F, Hagen JA, Storvik BE. Intrauterine growth pattern by the tendency to repeat small-for-gestational-age births in successive pregnancies. Acta Obstet Gynecol Scand 1997;165(Suppl):76:3–7.
28. Raine T, Powell S, Krohn MA. The risk of repeating low birth weight and the role of prenatal care. Obstet Gynecol 1994;84:485–9.[Medline]
29. Lumey LH, Stein AD. Offspring birth weights after maternal intrauterine undernutrition: A comparison with sibships. Am J Epidemiol 1997;146:810–9.
30. Alexander GR, Korenbrot CC. The role of prenatal care in preventing low birth weight. Future Child 1995;5:103–20.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S Harding, M Boroujerdi, P Santana, and J. Cruickshank Decline in, and lack of difference between, average birth weights among African and Portuguese babies in Portugal Int. J. Epidemiol., April 1, 2006; 35(2): 270 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hypponen, G. D. Smith, and C. Power Effects of grandmothers' smoking in pregnancy on birth weight: intergenerational cohort study BMJ, October 18, 2003; 327(7420): 898. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.C. Price and C.L. Coe Maternal constraint on fetal growth patterns in the rhesus monkey (Macaca mulatta): the intergenerational link between mothers and daughters Hum. Reprod., February 1, 2000; 15(2): 452 - 457. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
