ARTICLE |
a Department of Surgery, State University of New York Downstate Medical Center, Brooklyn, New York
b Department of Surgery, Laval University, Québec, Québec, Canada
c Laval Hospital Research Center, Québec, Québec, Canada
| ABSTRACT |
|---|
|
|
|---|
METHODS. In this case series, with >88% follow-up in a tertiary referral center, crosssectional office chart and telephone data on childhood and adolescent weights were transformed to z scores.
RESULTS. After maternal surgery, the prevalence of obesity in the offspring decreased by 52% and severe obesity by 45.1%, with no increase in the prevalence of underweight. The z score reduction in obesity was gender specific, with boys reducing from 1.4 ± 1.3 before to 0.57 ± 1.7 after maternal surgery, corrected for birth order. The difference was not significant in girls (0.8 ± 1.3 vs 0.8 ± 1.2). Among children of both genders who were aged 6 to 18 years of age and born after maternal surgery, the prevalence of overweight was reduced to population levels.
CONCLUSIONS. Contrary to outcomes after intrauterine under- and overnutrition, the prevalence of overweight and obesity in children of mothers with large voluntary postsurgical weight loss was similar to that in the general population, with no increase in underweight. The results demonstrate the importance of potentially modifiable epigenetic factors in the cause of obesity.
Key Words: fetal origins gestational weight gain overnutrition severe obesity biliopancreatic diversion
Abbreviations: BPDbiliopancreatic diversion BMSbefore maternal surgery AMSafter maternal surgery SGAsmall for gestational age
Throughout evolution, the greatest natural threat to survival has been malnutrition. Hunger and starvation remain the largest global source of human suffering, although overnutrition is spreading rapidly worldwide. Gestational and fetal undernutrition have dominated the concerns of mothers, grandmothers, and pediatricians for instinctive as well as empirical reasons. With growing awareness of the hazards that are associated with maternal obesity and weight gain to mother, fetus, infant, and ultimately adult, a conflict has arisen between the need to treat and prevent obesity and the obligation to protect the growing progeny from arrested development, malformations, and other complications of undernutrition.1
Recommendations for gestational weight gain adjust for maternal obesity,2,3 although efforts to treat or prevent excessive weight gain are no more successful than weight control in the general population. Animal experiments and the cruel effects on humans of climate, natural catastrophes, and war have provided ample evidence of the adverse sequelae of starvation during pregnancy and early development. Antiobesity surgery is the most effective treatment of the obesities and their comorbidities by achieving maintenance of substantial weight loss.4,5 Although pregnancy should be avoided during the first years after bariatric surgery, several case series describe favorable pregnancy outcomes after weight-loss operations spanning the evolution of these operations from open intestinal bypass to the currently performed laparoscopic restrictive and diversionary operations.69 There are, however, no published data on long-term effects of maternal bariatric surgery on the offspring.
Among recognized antiobesity operations, biliopancreatic diversion (BPD), entailing partial gastric resection and shunting of bile and pancreatic secretions to distal small bowel,4 arguably is the operation with the greatest potential for causing malnutrition, although metabolizable energy intake after BPD is comparable to that of unoperated lean individuals.10 Previously, we compared pregnancy outcomes in severely obese mothers before and after BPD.9,11 Here we present data on offspring who were followed into adolescence. These findings address some of the concerns regarding fetal undernutrition or intrauterine growth restriction (fetal origins of adult disease12,13) and provide information that is relevant to the importance of epigenetic factors in the development of obesity.
| METHODS |
|---|
|
|
|---|
Our hospital database contains operative reports and information on annual office visits and pregnancies. We analyzed mothers weights at time of surgery and most recent weight before conception (usually within 1 year), presence of antidiabetic or antihypertensive medication, fasting blood sugar level and serum albumin before surgery and during pregnancy, gestational weight gain, and birth weights of children. Apart from current height and weight, psychosocial status and history of dieting of the children were evaluated in a brief structured interview with their mothers. Mothers with pregnancies before and after surgery were questioned about the course of their pregnancies, specifically the presence of diabetes, hypertension, preeclampsia, and preterm hospitalization for these complications.
We used the National Health and Nutrition Examination Survey 2000 chart15 to determine the BMI percentile of each child, defining "underweight" as a BMI percentile of 5 or less. We calculated the obesity z score according to the Centers for Disease Control and Prevention chart.16 We also defined "obesity" as a BMI at or above the 95th percentile for age and gender.17 Therefore, in this article, "obesity" refers to BMI above the adult equivalent of 30 kg/m2, whereas "overweight" represents a pediatric BMI above the adult equivalent of 25 and below 30 kg/m2. We defined "severe obesity" as BMI >2 kg/m2 above the cutoff point for obesity, equivalent to BMI of 35 in adults. The 1996 National Longitudinal Survey of Children and Youth served as our population standard.18
Operations
Twenty-three mothers had undergone conventional BPD with distal gastrectomy4 and 90 had had BPD with pylorus-sparing, "duodenal switch" and sleeve gastrectomy (Fig 1) a mean of 6 ± 3 years before this study.14 Vitamins and minerals were routinely prescribed postoperatively and were monitored via blood tests, with supplementation on an as-needed basis.
|
P
.05 was considered statistically significant, whereas P
.15 was defined as trends. The data were analyzed using the statistical package SAS 9.1.3 (SAS Institute, Cary, NC). The study was approved by the local ethics committee.
| RESULTS |
|---|
|
|
|---|
|
The mean obesity z score decreased from 1.2 ± 1.3 (n = 45) BMS to 0.7 ± 1.5 (n = 172; P = .105) AMS. In 24 boys who were born BMS compared with 79 boys who were born AMS, the z score decreased from 1.40 ± 1.3 to 0.57 ± 1.7 (P = .047) and was unchanged when corrected for birth order (P = .044). There was no statistically significant difference in girls: 0.87 ± 1.3 for 21 girls who were born BMS vs 0.80 ± 1.3 for 93 girls who were born AMS (P = NS); 0.80 ± 1.3 vs 0.80 ± 1.2 corrected for birth order.
Among 17 mothers before and 35 mothers after surgery who were matched for BMI at time of conception (BMS: 42.2 ± 2.6 vs AMS: 41.4 ± 2.8 kg/m2), the offspring prevalence of overweight and obesity decreased from 70% (12 of 17) in children who were born BMS to 35% (13 of 35) in children who were born AMS (P = .037). The original, preoperative mean BMI of 35 mothers who underwent BPD had been 57.1 ± 8.5 kg/m2 before their operation, which resulted in a reduction of >15 BMI units, although they remained severely obese. The rate of cesarean sections among primigravidae decreased from 34% before surgery to 19% after surgery (p = .098), equivalent to the rate in the general population in Quebec.19
Predictors of Overweight and Obesity
Overweight and obesity among offspring, adjusting for age, parity, birth order, and gender, were predicted by maternal presurgical BMI (P = .037), with trends for prepregnancy BMI (P = .086) and gestational weight gain (P = .070). Neither low birth weight (< 2500 g) nor being small for gestational age (SGA) was associated with future overweight or obesity. Among 6 mothers with gestational malnutrition (s-albumin <26 g/L), 4 delivered children who were SGA. At the time of survey, 3 of these children, aged 4, 7, and 10 years, had normal weights, whereas 1 child, 4 years old, was overweight.
Psychosocial Outcomes
Children born AMS seemed to progress normally intellectually and socially according to the questionnaire data, which naturally have strong limitations. Among 123 school-age children, all were performing normally at their appropriate age level except for 2 (1.6%) who had repeated one grade. The prevalence of repeating is reported to be about 5% in primary schools and 17% in secondary schools in Quebec.20 "Learning difficulties" were attributed to 5 of the 172 children (2.9%); the national average for Canada is 2.8%.21 Five boys and 1 girl, or 3.5% of the children, were taking psychotropic medication (Ritalin), less than the reported 5% to 12% in the province of Quebec.22 Eleven children (9 obese and 2 of normal weight) had been on weight loss diets.
| DISCUSSION |
|---|
|
|
|---|
Our data on immediate outcomes of pregnancies after bariatric surgery are in line with the experiences of others.7,8,23 However, it is important to recognize differences between generically different antiobesity operations, with different mechanisms of action.24 Studies of the newer, laparoscopic, purely gastric-restrictive procedures (adjustable banding) have confirmed some of the benefits that we demonstrated previously9,11 with respect to fetal loss, prematurity, and decreases in macrosomia.8,25 However, in the only published series to date with extensive data on gestational diabetes, hypertension, birth weights, and macrosomia, gestational diabetes was present in 6.3% and hypertension was present in 10% of mothers, and reductions in birth weights and macrosomia were smaller than in our series.8,25 These differences have bearing on the mechanisms that we hypothesize might explain our findings.
Intrauterine Environment
Previous laboratory and clinical demonstrations of relationships between the intrauterine metabolic environment and offspring obesity demonstrate the importance of insulin sensitivity in the mother and the fetalplacental unit.26,27 Several investigators have identified maternal insulin resistance, before as well as during gestation, as a robust determinant of fat mass in the newborn.28,29 Furthermore, numerous epidemiologic studies have demonstrated relationships between maternal obesity, birth weight, and adult weight.3032 Although we did not perform studies of insulin sensitivity in our patients or the children, there is a wealth of information demonstrating improved glucose metabolism from weight loss by any method,5,3335 and tight glucose control has dramatic effects on gestational diabetes and its complications.36,37
We postulate that the beneficial effects of BPD are mediated by prevention of the deleterious effects that excess weight gain in gestational obesity causes through "overnutrition" of the fetus38 and by improved tissue and systemic insulin sensitivity in the weight-reduced mother and the fetalplacental unit, preventing glucotoxicity in the fetus and infant. Energy intake that is corrected for fecal losses (mainly of lipid) is equivalent to normal energy intake,10 indicating that the intrauterine energy availability approximates "normal" rather than "obese" levels. Examples of mechanisms for improved insulin sensitivity are reduction of intramyocellular lipid from surgical weight loss39,40 and increased incretin responses through exposure of small bowel to undigested nutrients.24 The latter mechanism likely explains the superior pregnancy outcomes after diversionary operations compared with purely gastric restrictive procedures.
Gestational Weight Gain
Our findings might have implications for understanding the cause of obesity other than supporting the importance of intrauterine overnutrition. The malabsorptive BPD operation sharply decreased gestational weight gain. Indeed, it was associated with weight stability in some mothers who, although advised against it, conceived during the first 18 months, or rapid weight-loss phase, after surgery.11 A recent study indicated that a <5-kg gestational weight gain in moderately obese, glucose-tolerant women (mean BMI: 34) normalizes birth weight.41 The 7-kg gain in our previously severely obese women is at the lower limit of the Institute of Medicines recommended weight gain for obese women.2
In contrast to an expectation of intrauterine growth retardation of the type postulated by Barker and others,12,13 we did not find a tendency for small children at birth to become obese during adolescence. Rather, we found that small children at birth tended to remain small, whereas heavier children remained heavier, in accordance with other reports.32,42 The most important determinant of future insulin resistance seems to be rapid postnatal weight gain,43,44 a phenomenon that we did not detect in the offspring in our study.
Against the background of strong heritability of severe obesity,45,46 it is remarkable to note the normalization of the prevalence of obesity in the offspring AMS in the present study, thus indicating substantial geneenvironment interaction in favor of epigenetic influences on the phenotypic expression of obesity. That siblings BMS and AMS share their mothers postsurgical environment with respect to diet, food purchases, and other activities yet the offspring BMS remain obese indicates that other mechanisms must be sought to explain our findings.
A key distinction between the maternalfetal undernutrition of hunger and famine and BPD surgery is that mothers AMS neither experience hunger nor perceive anything threatening or stressful about their voluntary weight loss. This weight loss, in the absence of stress, is not likely to be associated with exposure of the fetus to excess glucocorticoids, a known mediator of intrauterine growth restriction and postnatal disease.47 It is interesting that elevated levels of corticosteroids have been demonstrated in male compared with female fetuses of stressed ewes. Putatively, the gender-specific expression of obesity in our boys BMS reflects a gender-linked response to stress associated with maternal obesity, alleviated in boys AMS.
Limitations
Limitations of this study are its reliance on telephone questionnaire data that were based on mothers reports of their childrens heights and weights and their recollections of pregnancies, sometimes preceding the interview by many years. Fortunately, we have overlapping as well as supplemental chart data in sufficient numbers to validate the reports with objective data (88% office follow-up in 2004). Although the information is treated as cross-sectional for the purposes of this study, spanning long periods of time in a relatively large clinical series, we continue to maintain close contact with our patients through office visits with us or physicians in the patients communities. Furthermore, our surgical patient population has demonstrated remarkable loyalty in answering questionnaires on various subjects through the years, with response rates of >90%.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to John G. Kral, MD, PhD, Department of Surgery, Box 40, SUNY Downstate Medical Center, 450 Clarkson Ave, Brooklyn, NY 11203. E-mail: jkral{at}downstate.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Effect of Maternal Weight Loss from Obesity Surgery on Obesity in Offspring Journal Watch (General), December 19, 2006; 2006(1219): 4 - 4. [Full Text] |
||||
![]() |
The Effect of Maternal Weight Loss from Obesity Surgery on Obesity in Offspring Journal Watch Pediatrics and Adolescent Medicine, December 13, 2006; 2006(1213): 4 - 4. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||