Review of Trimester-Specific Gestational Weight Gain and Childhood Adiposity

Childhood obesity is a global epidemic and a major public health challenge. There has been increasing evidence that intrauterine exposures, such as alcohol, smoking, and maternal nutritional status, may affect both the long and short term health consequences of the mother and offspring. Childhood adiposity may be affected by the mother’s pre-pregnancy weight and her weight gain during pregnancy. Consequently, interventions may need to start before conception of the child to prevent childhood obesity. In 2009, the Institute of Medicine updated its gestational weight gain recommendations by incorporating rates of gestational weight gain in the second and third trimesters based on the mother’s pre-pregnancy Body Mass Index. There is extensive research on the association between total gestational weight gain and short-term offspring adiposity. However, this review focuses on the association between trimester-specific gestational weight gain and childhood adiposity for singleton pregnancies with respect to the Institute of Medicine’s newly defined weight gain recommendations as very few studies have examined the association between the gestational weight gain during each trimester and childhood adiposity. Identifying the trimester that is most associated with childhood adiposity may help in the development of targeted interventions, guide physician’s nutritional and weight-gain recommendations for child-bearing mothers, and direct future research. DOI: http://dx.doi.org/10.3126/jnps.v34i1.8429 J Nepal Paediatr Soc 2014;34(1):48-53


Introduction
O besity is a worldwide epidemic across all age groups with li le evidence of decline 1 .Childhood overweight and obesity are leading causes of early type two diabetes and cardiovascular disease 1 .In 2011, more than 40 million children under the age of fi ve were overweight, 30 million of who are living in developing countries 2 .The lack of evidence on the long term effi cacy of tradi onal obesity management emphasizes that preven on should begin early in human development.Researchers have iden fi ed numerous modifi able factors in the prenatal and early postnatal periods that may increase the risk of childhood overweight, such as maternal smoking during pregnancy, reduced breas eeding dura on, and rapid weight gain during the fi rst few months of life 3 .
As excessive gesta onal weight gain (GWG) has been correlated with childhood and later in life overweight and obesity, the importance of appropriate GWG has gained increasing a en on in recent mes 3- 20 .Mo vated by recent research on the changing characteris cs of the child-bearing popula on, the Ins tute of Medicine (IOM) revised GWG guidelines in 2009 to include a specifi c recommenda on for obese women, and second and third trimester GWG rates based on pre-pregnancy Body Mass Index (Table 1) 4 .
In light of the current IOM recommenda ons, the primary objec ve of this review was to examine current knowledge regarding the associa on between trimester-specifi c GWG and childhood adiposity.This literature review examines current evidence inves ga ng GWG and childhood adiposity with a focus on trimester-specifi c GWG and childhood adiposity for singleton pregnancies.As such, the review hopes to provide direc on for future research on early predictors of childhood adiposity and for the development of early interven ons to prevent obesity.

Methods
Web of Science and Pubmed were used to obtain peer-reviewed studies for the period of January 1990 and March 2013.The search was also limited to journal ar cles, meta-analyses, randomized controlled trials, reviews, and systema c reviews in the English language.The string of terms that were considered include: gesta onal weight gain and childhood BMI; gesta onal weight gain and childhood obesity or off spring adiposity; maternal weight gain during pregnancy and off spring obesity or off spring adiposity; gesta onal weight gain and adolescent obesity; and trimesterspecifi c gesta onal weight gain.Both observa onal studies and review ar cles were included.The ini al search yielded a total of 341 studies.Studies were subsequently screened by abstrac ng informa on on study characteris cs, study par cipants, eligibility criteria, interven ons, outcome measures, the method of ascertainment, and the outcomes.Opinion and editorial pieces were excluded.Studies that exclusively discussed the biological aspects of gesta onal weight gain e.g.physiologic mechanisms and those that predominantly focused on exposures other than GWG, such as gesta onal diabetes, were excluded.25 studies met the inclusion criteria and were ul mately selected for review.

Maternal Gestational weight gain and Childhood Adiposity
Oken et al.'s study on the Growing Up Today study to iden fy developmental origins of childhood overweight was one of the fi rst and pioneering studies that examined the rela on of gesta onal weight gain and childhood obesity 5 .GWG was found to be associated with higher child body mass index at 3 years (0.13 units per 5 kg, 95% CI: 0.08, 0.19).A characteris c methodology used in this study was calcula ng BMI along with, measuring skin fold thickness as a measure of adiposity.Total gesta onal weight gain was posi vely associated with both child BMI z-score 0.13 units, 95% CI: 0.08, 0.19 per 5 kg) as well as the sum of subscapular and triceps skinfold thicknesses (0.26 mm, 95% CI: 0.02, 0.51).In addi on, compared with inadequate weight gain, women with adequate or excessive weight gain had children with increased rela ve odds of having children that were overweight (odds ra os=OR=3.77,95% CI: 1.38, 10.27) and (OR=4.35,95% CI: 1.69, 11.24).Appreciable change of results was not found upon adjustment to covariates, which were: maternal age and smoking, household income and paternal educa on, and child race/ethnicity, gesta onal age, sex, age in 1996, and Tanner stage.
Subsequent cohort studies in the US have resulted in fi ndings consistent to those of Oken et al.Specifi cally, these studies showed that independent associa ons exist between maternal GWG and off spring BMI when the child is 3 years 6 , 5 years 7 , 7 years 8 , and in their adolescence 5 .Consistent fi ndings have been found in interna onal studies.A cross-sec onal study in Germany by von Kries et al. found, a er adjus ng for confounders, children of mothers with high GWG were s ll found to be signifi cantly more likely to be overweight than children of mothers with an average GWG (OR=1.16,95% CI: 1.02-1.32),which remained unchanged a er adjus ng for confounders 9 .Studies have also been conducted to examine the associa on between gesta onal weight gain and off spring's BMI in adulthood.Mamun et al.'s study followed off spring to adulthood (age 21 years) and found off spring BMI was on average 0.3 kg/m 2 (95% confi dence interval, 0.1 to 0.4 kg/m 2 ) higher for each 0.1-kg/wk greater GWG a er adjustment for poten al confounding factors 10 .Similar to Oken et al.'s study, off spring of mothers with excessive weight gain during pregnancy, as defi ned by the IOM, were more likely to be overweight (OR=1.3,95% CI: 1.0 to 1.8) and obese (OR=1.5, 95% CI: 1.0 to 2.1) at 21 years of age compared with those with inadequate weight gain.Although there was a posi ve associa on with adequate GWG and risk of overweight or obesity at age 21 years, results were insignifi cant 11 .
A Danish report on the Copenhagen perinatal cohort that followed-up 1540 off spring to age 42 found that GWG was associated with off spring BMI at all ages: a 2.36 (1.08-5.15)-foldincrease in the risk of obesity and a 1.28 (0.89-1.85)-fold increase in the risk of overweight in the highest (>=16 kg) versus the lowest (<6 kg) GWG category 12 .Furthermore, there was an increasing risk of obesity at the age of 42 years (OR=1.08,95% CI: 1.03-1.14per kg GWG) 12 .
Two studies conducted in the United Kingdom have also found similar results.Reynolds et al found GWG, as well as pre-pregnancy BMI and parity, to be independent predictors of percentage body fat of off spring, even a er adjustment for confounders 13 .Crozier et al. found that children who were born to mothers with excess GWG had a greater fat mass in the neonatal period (SD: 0.17; 95% CI: 0.02, 0.32), at 4 yr (SD: 0.17; 95% CI: 0.00, 0.34), and at 6 yr (SD: 0.30; 95% CI: 0.11, 0.49) compared to those children who were born to mothers with normal GWG 14 .

Trimester of Gestational Weight gain
The associa on between GWG and off spring BMI has been substan ated with increasing amounts of evidence.Only a few studies are beginning to draw a en on to the possible role that "adequate" GWG per trimester, as proposed by the IOM, may have on later-life adiposity [17][18][19][20][21][22] .Diffi cul es in establishing recommenda ons arise from the importance to strike a balance between a weight gain that is not so reduced as to cause low birth weight, restricted intrauterine growth and prematurity, yet which is not so high as to increase the chances of macrosomia, preeclampsia, caesarean sec on and gesta onal diabetes.
Dutch Famine studies and animal models have shown maternal exposure to famine in early pregnancy is associated with increased adult BMI in women 15 .Furthermore, numerous epidemiologic studies have found GWG in the second trimester to be more strongly associated with lower birth weight than the GWG in the fi rst or third trimester [16][17][18][19][20] .Among these, Hickey et al found that low gain corresponding to IOM recommenda ons in the fi rst and second, or in the second and third trimesters, were associated with signifi cant decreases in mean birth weight, ranging from 206 to 265g 18 .However, low gain in fi rst or third trimester alone was not associated with a signifi cant decrease in mean in birth weight.
Drehmer et al. also found no associa on between insuffi cient weight gain in the third trimester and birth weight 21 .In the second trimester, insuffi cient weight gain was associated with small for gesta onal age (rela ve risk= RR 1.72, 95% CI: 1.26-2.33),and excessive weight gain with large for gesta onal weight gain (RR 1.64, 95% CI: 1.16-2.31) 21.
Only three studies were found in the literature search to have examined whether trimester-specifi c GWG associates with off spring adiposity later in life as their primary or secondary objec ves, two of them being European cohort studies 7,22,23 .Fraser et al. examined the importance of IOM guidelines and off spring adiposity directly with Avon Longitudinal Study of Parents and Children (ALSPAC) prospec ve pregnancy cohort in UK 22 .The study found that compared with off spring of women gaining recommended levels, those who gained less than recommended levels had lower rela ve odds of overweight/obesity BMI (OR=0.80,95% CI: 0.67, 0.96) and of waist-based central obesity (based on waist) of (OR=0.79,95% CI: 0.69, 0.90), at age 9. Compared with those who met recommended GWG, off spring of mothers who gained more than recommended levels had greater rela ve odds of being overweight/obesity and central obesity OR=1.73, 95% CI: 1.45, 2.05 and OR=1.36, 95% CI:1.19, 1.57, respec vely, at age 9.In addi on, when examining these associa ons more closely, Fraser et al. found that any weight gain in the fi rst 14 weeks of gesta on was incrementally associated with increased off spring adiposity, but between 14 and 36 weeks gesta on only GWG above 500g/week was associated with increased off spring adiposity 22 .
Margerison-Zilko et al. examined 3,015 singleton births to women without pregnancy complica ons in a more heterogeneous popula on 7 .Based on the US-based Child Health and Development Studies, the study found associa ons between total as well as trimester-specifi c GWG and birth weight for gesta onal age as well as child BMI at age 5, adjus ng for maternal age, race/ethnicity, educa on, marital status, parity, pre-pregnancy body mass index (BMI), and smoking; paternal overweight, gesta onal age, and infant sex.The study concluded that although all trimesters were associated with birth weight, only fi rst trimester GWG was associated with high child BMI (OR for child overweight = 1.05, 95% CI: 1.02, 1.09) at age 5.A limita on of the study was that it did not control for certain poten al confounders: gesta onal diabetes mellitus, gene c characteris cs, nausea, diet, or physical ac vity.In addi on, unlike the Fraser et al., Margerison-Zilko found there to be eff ect modifi ca on by pre-pregnancy BMI 7 .

Discussion
Rates of excessive GWG have been increasing over me and research has shown that greater GWG is associated with increased weight of baby at birth and infancy [24][25][26][27] .This raises ques ons about the long-term adverse eff ects of higher weight gains in pregnancy.Oken et al.'s inves ga on of the Growing Up Today Study cohort to iden fy developmental origins of childhood overweight was one of the fi rst and pioneering studies that examined the rela on of gesta onal weight gain and childhood obesity 5 .A number of studies therea er have substan ated this associa on.Furthermore, as only part of the associa on of GWG with off spring adult BMI is explained only partly by birthweight and BMI up to 14 years of age, this suggests that excessive GWG induces other media ng processes and factors that in turn result in higher off spring adult BMI 13 .Therefore, explora on of media ng variables in the GWG, such as the trimester of pregnancy, could reveal possible points where interven ons can be applied.
Much controversy surrounds the accuracy of BMI in measuring childhood fat.The literature review demonstrated that Fraser et al.'s and Oken et al.'s studies are two of the only few studies that have examined childhood adiposity, in rela on to GWG, by measures other than BMI 5,22 .Another per nent study is from the Southampton Women's Survey, in which children's body composi on was determined using a dual-energy X-ray adsorp ometry 14 .The Motherwell birth cohort study based in Scotland also examined four-site skinfold thicknesses, waist circumference, and body mass index (BMI) at age 30 yr of off spring 13 .
The literature review iden fi ed only three studies that inves gated the rela onship between trimesterspecifi c GWG and childhood adiposity as a primary or secondary objec ve 7,22,23 .All three found GWG in the fi rst trimester to be associated with childhood and later in life childhood adiposity 6,21,22 .Andersen et al. addi onally found that this associa on to be maintained for the second trimester as well 23 .These fi ndings of early pregnancy being most correlated with later in life adiposity are complimented by results from Dutch Famine Studies and biologic evidence of the most rapid growth-thus, the most sensi ve period-occurring in the fi rst and second trimesters of pregnancy 15,18 .
Limita ons of the studies included in this review need to be considered when interpre ng results.Exposure assessment of trimester-specifi c GWG has been predominantly based on self-report, introducing possible exposure misclassifi ca on due to recall bias or inaccurate recall 6,21,22 .Addi onally, these studies have also been based on European-or USA-based study popula ons with varied methods of exposure and outcome measurement, contribu ng to diffi cul es in valid comparisons between studies.Furthermore, the sample size in Fraser et al.'s study was not large enough to iden fy a signifi cant associa on between GWG from 36 weeks onwards as the dura on of this period varies largely between pregnancies 22 .

Conclusion
The publica ons reviewed in this paper suggest that gesta onal weight gain in the fi rst and second trimesters of singleton pregnancy may be associated with childhood adiposity.However, further research is needed to be er evaluate the possible role of trimester-specifi c gesta onal weight gain on childhood adiposity with regard to the Ins tute of Medicine's recommenda ons.As the majority of past studies have focused on US and European popula ons, there is a need to inves gate the associa on in more diverse popula ons to increase generalizability of fi ndings.More consistent defi ni ons of gesta onal weight gain and childhood adiposity, as well as a more thorough collec on of weight gain data are also needed.

Table 1 :
New Recommenda ons for Total and Rate of Weight Gain during Pregnancy, by Prepregnancy BMI 4Calcula ons assume a 0.5-2.0kg(1.1-4.4 lbs) weight gain in the fi rst trimester Source: Ins tute of Medicine 20094