Gestational Hyperglycemia, Excessive Pregnancy Weight Gain and Risk of Fetal Overgrowth

Between 1989 and 2004 the prevalence of gestational diabetes mellitus (GDM) in the US increased from 1.9% to 4.2% in parallel with the well documented obesity epidemic (Getahun et al., 2008; Mokdad et al., 2001). However, an additional 9-20% of pregnant women, with a milder form of glucose intolerance which does not meet the diagnostic criteria for GDM, may also be at risk of type 2 diabetes, cardiovascular disease and problems while pregnant (Stamilio et al., 2004; Mello et al., 1997; Bo et al., 2004; Nordin et al., 2006). Fetal overgrowth, defined as macrosomia (birth weight >4000g) or large-for-gestational-age birth (LGA, birth weight >90th percentile for a given gestational age) increases maternal morbidity from operative delivery and also causes serious consequences to the offspring including birth trauma, obesity during childhood and type 2 diabetes and metabolic syndrome in adult life (Boney et al., 2005; Zhang & Bowes, 1995; Langer, 2000). GDM and excessive pregnancy weight gain, especially in obese women are known risk factors for fetal overgrowth (Ray et al., 2001; Hillier et al., 2008). Although previous research has suggested that metabolic abnormalities are also present in pregnant women with less severe hyperglycemia, the clinical implications of milder maternal hyperglycemia are poorly described (Chen et al., 2010; Cheng et al., 2006; Nordin et al., 2006). The primary objective of this chapter is to use prospective data from a population of low income and minority women to examine (i) the influence of maternal hyperglycemia including GDM and less severe maternal hyperglycemia on risk of fetal overgrowth; (ii) the association of longitudinally measured excessive pregnancy weight gain with risk of fetal overgrowth (IOM, 2009); (iii) the independent contribution of gestational hyperglycemia and excessive pregnancy weight gain to fetal overgrowth. Because gestational hyperglycemia and excess weight gain during pregnancy are preventable risk factors, early detection and treatment of mild hyperglycemia and monitoring pregnancy weight gain may be important for reducing the risk of LGA, for reducing childhood and later obesity and for improving the long term risk for metabolic disease.


Introduction
Between 1989 and 2004 the prevalence of gestational diabetes mellitus (GDM) in the US increased from 1.9% to 4.2% in parallel with the well documented obesity epidemic (Getahun et al., 2008;Mokdad et al., 2001).However, an additional 9-20% of pregnant women, with a milder form of glucose intolerance which does not meet the diagnostic criteria for GDM, may also be at risk of type 2 diabetes, cardiovascular disease and problems while pregnant (Stamilio et al., 2004;Mello et al., 1997;Bo et al., 2004;Nordin et al., 2006).Fetal overgrowth, defined as macrosomia (birth weight >4000g) or large-for-gestational-age birth (LGA, birth weight >90th percentile for a given gestational age) increases maternal morbidity from operative delivery and also causes serious consequences to the offspring including birth trauma, obesity during childhood and type 2 diabetes and metabolic syndrome in adult life (Boney et al., 2005;Zhang & Bowes, 1995;Langer, 2000).GDM and excessive pregnancy weight gain, especially in obese women are known risk factors for fetal overgrowth (Ray et al., 2001;Hillier et al., 2008).Although previous research has suggested that metabolic abnormalities are also present in pregnant women with less severe hyperglycemia, the clinical implications of milder maternal hyperglycemia are poorly described (Chen et al., 2010;Cheng et al., 2006;Nordin et al., 2006).The primary objective of this chapter is to use prospective data from a population of low income and minority women to examine (i) the influence of maternal hyperglycemia including GDM and less severe maternal hyperglycemia on risk of fetal overgrowth; (ii) the association of longitudinally measured excessive pregnancy weight gain with risk of fetal overgrowth (IOM, 2009); (iii) the independent contribution of gestational hyperglycemia and excessive pregnancy weight gain to fetal overgrowth.Because gestational hyperglycemia and excess weight gain during pregnancy are preventable risk factors, early detection and treatment of mild hyperglycemia and monitoring pregnancy weight gain may be important for reducing the risk of LGA, for reducing childhood and later obesity and for improving the long term risk for metabolic disease.
Conflicting data on mild hyperglycemia and fetal overgrowth can be, at least in part, due to the different population studied and the criteria used for mild hyperglycemia.Although the HAPO study included multiple countries and populations and had a large sample size, it was focused on developing an international consensus for the diagnosis and treatment of carbohydrate intolerance during pregnancy, and was not designed to compare the difference in LGA between GDM and less severe hyperglycemic pregnancies (HAPO, 2008).Thus, to establish the risk of adverse fetal outcomes in relation to milder degrees of maternal hyperglycemia is clinically important, particularly in high risk US populations.

The Camden study
Camden Study is a prospective cohort study of pregnancy outcome and complications in young, generally healthy women residing in one of the poorest cities in the continental United States (Webster & Bishaw, 2006).Women with serious non-obstetric problems (e.g., Lupus, type 1 or 2 diabetes, seizure disorders, malignancies, acute or chronic liver disease, drug or alcohol abuse and psychiatric problems) were excluded from participation.The sample for this analysis totaled 2,373 pregnant women with live births who enrolled between 1996-2006.

Definition of GDM and mild hyperglycemia non-GDM in Camden study
The diagnosis of GDM was made using a two-step approach.All participants were initially screened by measuring the plasma glucose concentration 1h after a 50-g oral glucose challenge test (GCT) at 28±0.1 (mean±SE) weeks' gestation.A diagnostic OGTT was performed on that subset of women exceeding the glucose threshold value (>140 mg/dl).The diagnostic criteria for GDM was the Carpenter/Coustan conversion as recommended by the American Diabetes Association (ADA) which defines GDM by two or more glucose values over the cut points of 95,180,155,140 mg/dl at fasting, 1, 2, and 3 hours during a 100g OGTT (American Diabetes Association, 2000).Women with a positive GCT and fewer than two abnormal glucose values were identified as having an impaired GCT without GDM (impaired GCT non-GDM).All of patients diagnosed with GDM obtained dietary counseling and/or insulin treatment for their glycemic control; patients with impaired CGT non-GDM were untreated.We identified 2,092 women (88.1%) with normal GCT, 182 women (7.7%) with impaired GCT non-GDM and 99 women (4.2%) with GDM (table 1).Ethnic composition of the cohort was 46% Hispanic, 37.6% African American and 16.6% Caucasian plus others.When the maternal characteristics were rank ordered with respect to glucose tolerance status, older maternal age, higher pre-pregnant body mass index (BMI), obesity and prior history of GDM, all were found to be positively associated with maternal glycemic status (normal GCT, impaired GCT non-GDM or GDM) and represented a linear increasing trend in terms of severity (p for trend <0.001 for each).

Maternal hyperglycemia and LGA birth
LGA is defined as a neonatal birth weight greater than the 90 th percentile for gestational age that has been adjusted for ethnicity, parity and infant sex (Zhang & Bowes, 1995).Women with GDM or impaired GCT without GDM had a ~2-fold increased risk for bearing an LGA infant as compared to women with normal GCT (table 2, model 1) after controlling for maternal confounding variables and using a standard which adjusted LGA for maternal ethnicity, parity and fetal gender.This increased risk persisted after additional adjustments LGA, large-for-gestational age infant; AOR, adjusted odds ratio; 95% CI, 95% confidence interval (same as in tables 4-5). 1 p for linear trend <0.01.
2 Model 1 was adjusted for age and cigarette smoking in addition to using a standard which adjusted LGA for maternal ethnicity, parity and fetal gender.
3 Model 2 was adjusted for all variables in model 1 with additional adjustment for pre-pregnant BMI and net of weight gain (kg) during pregnancy.
Table 2. Maternal Hyperglycemia and LGA birth

What do we learn from the Camden study?
Complicating the interpretation of previous studies has been the inability to adjust the observations for important factors related to fetal growth such as pre-pregnant obesity or gestational weight gain, often because the data were not available (Hillier et al., 2008;Rosenberg et al., 2005;Aberg et al., 2001).Our study in Camden also differs because of the inclusion of several ethnic groups (Vambergue et al., 2000;Bo et al., 2004;Mello et al., 1997).
With our large cohort, we had the power to control for known potential confounders that influence the fetal overgrowth, including pre-pregnant BMI and net gestational weight gain.We confirmed that untreated maternal mild hyperglycemia (impaired GCT with one abnormal or no abnormal glucose value at OGTT) in otherwise young and healthy women in the US was associated with a 2-3 fold increased risk of delivering an LGA infant (table 2).These observations are consistent with previous findings from populations studied worldwide by the HAPO group and other studies (HAPO, 2008;Stamilio et al 2004).In addition, increased risk of LGA birth in GDM women (all of whom were treated by diet or with insulin) was associated with high maternal pre-pregnant BMI.

Significance of optimal pregnancy weight gain and fetal growth
To optimize gestational weight gain for both mother and fetus is critical and remains controversial.A large number of studies have found that excess gestational weight gain is associated with decreased risk of small-for-gestational age birth and with increased risk for LGA birth (Oken et al., 2009;Jensen et al., 2005;Hinkle et a., 2010), regardless the definition or the scales used for excess pregnancy weight gain (Kiel et al., 2007;Jain et al., 2007).

Weight gain and maternal obesity
The independent contribution of excess gestational weight gain and maternal obesity on the risk of LGA or macrosomia has been extensively investigated (Jensen et al., 2005;Jain et al., 2007), because obesity is a common problem during pregnancy (Chu et al., 2007;Sebire et al., 2001;Rosenberg et al., 2005).Even in women with a normal pre-pregnant BMI, a higher pregnancy weight gain was associated with an increased risk of LGA, while a normal weight gain by the 1990 The Institute of Medicine (IOM) guidelines was associated with a decreased risk of LGA (DeVader et al., 2007).The definition of an optimal gestational weight gain in obese pregnancy remains controversial (Nohr et al., 2008).Jain et al observed that overweight/obese women who gain weight within the IOM recommendation achieve better fetal outcomes (Jain et al., 2007), while others found weight gain within IOM ranges did not reduce the risk of LGA among the obese (Dietz et al., 2009).It was suggested that limited or no weight gain in more severely obese pregnant women may increase favorable neonatal outcomes (Kiel et al., 2007;Jensen et al., 2005).Thus, an optimal gestational weight gain may need to be further defined by obesity severity (Hinkle et al., 2010;Oken et al., 2009).

What are problems in the research of weight gain during pregnancy?
There are limited data that link excess pregnancy weight gain measured prior to screening for GDM and /or longitudinal gestational weight gain measures through out the pregnancy to adverse pregnancy outcomes.In 2009, the IOM published revised gestational weight gain guidelines for how much weight a woman should gain during pregnancy to optimize both maternal and child outcomes.The report highlighted the importance of intervention in pregnancy to prevent obesity in the post-partum for mother and child (IOM, 2009).

The US Institute of medicine new guidelines for gestational weight gain
The

Assessment of pregnancy weight gain
We explored the associations between gestational weight gain assessed throughout pregnancy with LGA using the 2009 IOM guidelines for weight gain during pregnancy.Inadequate, adequate and excessive pregnancy weight gain at weeks 24, 28, 32 and at delivery was categorized according to IOM recommendations (table 3).Maternal obesity was defined as BMI ≥30 (WHO, 1998); height was measured with a stadiometer at entry.Gestational duration was based upon gestation from participants' last normal menstrual period confirmed or modified by ultrasound.Pregnancy weight was measured at week 12, 24, 28, 32 and at delivery.Total gestational weight gain was computed as the difference between weight at delivery and recalled pre-pregnant weight.Gestational weight gain during the 2 nd and 3 rd trimester was computed as the difference between weights measured at each visit minuses the weight measured at week 12.The rate of weight gain was the weight gain (kg) divided by gestational age (weeks).

Excess pregnancy weight gain measured mid to late gestations and LGA birth
The proportion of women with excess pregnancy weight gain at weeks 24, 28, and 32 was similar to delivery (48.5%, 52%, 54% and 50% respectively).Depending on the gestation, 21-27% had an adequate weight gain and 22-30% had an inadequate gain throughout the four time points.Compared to women with adequate weight gain, excessive weight gain was associated with a 1.58-2.66fold increased risk of LGA birth (table 4) after controlling for all the confounding variables with the exception of pre-pregnant BMI (p for trend <0.0001 for each model).Similar results were obtained when additional adjustment for pre-pregnant BMI.In addition, women with inadequate weight gain had a reduced risk of LGA birth at week 28 and 32 (p<0.05).

Contribution of Camden data
These data confirmed that excess pregnancy weight gain throughout the 2 nd and 3 rd trimesters and at delivery significantly increased risk of LGA, whereas inadequate pregnancy weight gain reduced LGA risk.The current study has several important strengths.Firstly, we used the most recent IOM recommendations ( 2009) for the estimation of excess weight gain by BMI categories and the analysis models were fully adjusted for potential confounding, including pre-pregnant BMI.In contrast, more arbitrary criteria were used to define excess weight gain in previous studies (Herring et al., 2009;Rosenberg et al., 2005), and pre-pregnant BMI either was not available or was not adjusted for in the analysis (Hillier et al., 2008;Rosenberg et al., 2005).Secondly, there are limited data on longitudinal measures of pregnancy weight gain (Herring et al., 2009;Saldana et al., 2006).We used multiple measures of pregnancy weight gain starting prior to the screen for GDM.The relationship between excess pregnancy weight gain assessed relatively early in pregnancy and risk of LGA could be important for preventive strategies.Total weight gain is not an ideal measure to evaluate in relation to fetal growth in GDM patients, because most of patients are treated and their weight gain is monitored after the diagnosis.
Week 1 Models were adjusted for age and cigarette smoking in addition to using a standard which adjusted LGA for parity, ethnicity and infant gender. 2 p for trend <0.0001 for week 24, 28, 32 and delivery respectively.Table 4. Excess pregnancy weight gain and LGA birth 1,2   4. Association of various degree of gestational hyperglycemia, excess pregnancy weight gain with LGA birth 4.1 Does excess gestational weight gain prior to screening GDM predict risk of GDM? High gestational weight gain between early and mid pregnancy was positively associated with risk of GDM or impaired glucose tolerance (Hedderson et al., 2010;Herring et al., 2009).In the Camden study, we found a positive association between excess weight gain prior to or at the time of screening for GDM (weeks 24 and 28) with increased risk for GDM (AOR 1.57, 95% CI 1.01, 2.44 for week 24, AOR 1.94, 95% CI 1.28, 2.95 for week 28).This association was not significant at week 32 or at delivery which may suggest an effect of treatment and weight monitoring after the diagnosis of GDM.We did not observe a significant association between excess weight gain and impaired GCT non-GDM (p>0.05 for all time points).

The association of gestational hyperglycemia and excess weight gain with
LGA Combined association of gestational hyperglycemia and excess weight gain on the risk of LGA has not been examined extensively.Hillier et al observed at GDM screening that macrosomia risk was increased across the spectrum of maternal glucose levels and that this relationship was further modified by excessive maternal weight gain (Hillier et al., 2008).A weight gain of 40lbs or more nearly doubled the risk of fetal macrosomia among glucose intolerant women including those with GDM.However, results were not adjusted for pre-pregnancy BMI which can complicate the relationship between maternal hyperglycemia and pregnancy weight gain.Thus, our next goal was to examine the independent and combined contributions of hyperglycemia and excess weight gain on risk of LGA.An analysis stratified by excessive weight gain was used to index the influence of various degrees of maternal glycemic status on LGA risk, using women with non-excess pregnancy weight gain (adequate and inadequate gain combined) and a normal GCT as the reference group.Our results showed that excess weight gain (at week 24, 28, 32 or at delivery) was positively associated with a 2-6 fold increased risks for delivering an LGA infant (table 5) regardless of whether the women were diagnosed with GDM, impaired GCT non-GDM or normal GCT.Furthermore, women with non-excess weight gain and an impaired GCT non-GDM also had a 2-3 fold increased risk of LGA birth during all of four time points, whereas risk of LGA was not increased in the GDM group with non-excess weight gain.

What do we learn?
These data confirmed that hyperglycemia and excess weight gain are independently associated with risk of fetal overgrowth.Women with excess weight gain at any of four times in gestation and hyperglycemia, even a mild hyperglycemia, had a substantially increased risk of LGA.Risk of LGA was not increased among gestational diabetics without an excessive weight gain but only among the group with impaired GCT non-GDM.Thus, excess pregnancy weight gain and hyperglycemia are independent risk factors for LGA.

Conclusion
By using the most updated IOM guidelines (2009) on pregnancy weight gain, we found that healthy, young women with mild but untreated hyperglycemia were at increased risk for fetal overgrowth (LGA).The risk of LGA birth in women with GDM, who are treated by diet or with insulin after diagnosis, was dependent on maternal pre-pregnant weight.
Excess pregnancy weight gain assessed longitudinally through out mid to late gestation consistently showed strong associations with LGA risk.In addition, excess pregnancy weight gain and hyperglycemia appeared to be independent risk factors for LGA.The risk for LGA was increased still further in women with excess weight gain, even in those with mild hyperglycemia.Because prepregnancy obesity, mild gestational hyperglycemia, and excess weight gain during pregnancy are preventable risk factors, these findings suggest that early detection and treatment of mild hyperglycemia and monitoring pregnancy weight gain during early gestation are both important for reducing the risk of LGA, for reducing childhood and later obesity and for improving the long term risk for metabolic disease in the offspring.

Table 3 .
new 2009 IOM guidelines are based on the WHO BMI (kg/m 2 ) categories(WHO, 1998)and include the recommended total pregnancy weight gain (kg) or rates of weight gain during the 2 nd and 3 rd trimester (kg/week).Weight gain below or above the recommended range is defined as inadequate or excessive gain, respectively.New recommendations for total and rate of weight gain during pregnancy, by prepregnancy BMI (IOM, 2009) 1

Table 5 .
Maternal hyperglycemia and LGA infant: Stratified by excess weight gain 1 www.intechopen.com