Objective: This study was undertaken to determine the relative contribution of abnormal pregravid maternal body habitus and diabetes on the prevalence of large-for-gestational-age infants.
Study design: Maternal and neonatal records for singleton term (> or =37 weeks' estimated gestational age) deliveries January 1997 through June 2001 were reviewed. Subjects were characterized by pregravid body mass index (BMI), divided into underweight (BMI <19.8 kg/m2), normal (BMI 19.8-25 kg/m2), overweight (BMI 25.1-30 kg/m2), and obese (BMI >30 kg/m2) subgroups. Diabetes was classified as gestational, treated with diet alone (A1GDM), or with insulin (A2GDM), and pregestaional diabetes (PDM). Newborn weight greater than the 90th percentile for gestational age, based on published local birth weight data, defined large for gestational age (LGA). The risk of LGA delivery for underweight, overweight, and obese women were compared with that of women with normal pregravid BMI. Multiple regression models, including parity, newborn sex, BMI, race, and diabetes, were constructed to examine the relative effect of abnormal BMI and diabetes on the risk of the delivery of an LGA infant.
Results: Complete data for 12,950 deliveries were included (1,640 [13.0%] underweight, 2,991 [23.7%] overweight, and 2,928 [23.2%] obese). LGA delivery affected 11.8% of the study sample; 303 (2.3%) of subjects had A1GDM, whereas 94 (0.7%) had A2GDM, and 133 (1.6%) had PDM. Compared with normal BMI subjects, obese women were at elevated risk for LGA delivery (16.8% vs 10.5%; P < .0001) as were overweight women (12.3% vs 10.5%; P = .01). Diabetes was also a risk factor for LGA delivery (A1GDM: [29.4% vs 11.4%]; A2GDM: [29.8% vs 11.7%]; PDM: [38.3% vs 11.6%]; P < .0001 for each). Other risk factors for LGA delivery included parity (13.2% vs 9.5%; P < .0001), and male gender (14.3% vs 9.3%; P < .0001). Black race and low pregravid BMI were associated with a lower risk of LGA delivery (9.0% vs 13.7%; P < .0001) and (6.4% vs 10.5%; P = .006), respectively. Multiple regression revealed the independent influence of pregravid obesity and PDM, increasing the risk of LGA delivery (BMI >30kg/m 2 [Adjusted odds ratio (AOR) = 1.6]), and PDM (AOR = 4.4).
Conclusion: Obesity and pregestational diabetes are independently associated an increased risk of LGA delivery. The impact of abnormal body habitus on birth weight grows as BMI increases. Diabetes has the greatest affect on the normal and underweight population. With the increasing prevalence and relative frequency of overweight and obese women in pregnancy compared with diabetes (46.7% vs 4.1%), abnormal maternal body habitus exhibits the strongest influence on the prevalence of LGA delivery in our population.