Endocrine Abstracts

Background: Macrosomia (BW >4.5 kg) is a well-recognised complication of maternal hyperglycaemia and is associated with poor maternal and foetal outcomes. Traditional risk factors of macrosomia do not fully explain all variation as it can happen in normoglycaemia and in well-controlled diabetes. The relative contributions of these factors in different ethnicities are unknown. We investigated the predictors of BW in South Asian (SA) and White Caucasian (WC) pregnancies, and the risk of macrosomia.

Method: Data from 88 606 singleton births across Leicester were collected (1994–2006). Multivariate linear and logistic regression models are used to analyse the predictors of BW in full-term births. Ethnicity effects were determined using the Mann–Whitney- U (categorical) and χ² tests (continuous).

Results: 53 128 births had complete dataset (WC: 44 657 (84.1%); SA: 8471 (15.9%)). The mean BWs were: WC: 3445±488 g; SA: 3102±452 g; P<0.001. Macrosomia rates were: WC: 12.6%; SA: 2.9% (odds ratio (OR): 0.21, 95% CI: 0.18, 0.24; P<0.001). Maternal diabetes (n=837) prevalence was higher in SA (3.47 vs 1.22%; OR: 2.92, 95%CI: 2.53,3.37; P<0.001). Adjusting for all covariates, higher gravida, non-smoking, low deprivation, maternal age and BMI, gestational age, and male-sex, independently and positively predicted BW and macrosomia. Blood pressure (n=36 697) was not independently associated with BW. Babies born to WC mothers were heavier (male: 291.2 g, 95% CI: 276.7, 305.7; female: 254.2 g, 95% CI: 239.5, 268.9 g; P<0.001) compared to SA babies. Maternal diabetes contributed differentially to BW within different ethnicities (WC: 284.5 g (95% CI: 248.1, 320.9 g) vs SA: 213.7 g (95% CI: 164.0, 263.4 g), P=0.024).

Conclusion: Maternal diabetes is a strong independent risk factor for macrosomia and the magnitude of weight gain is significantly different between WC and SA. After correcting for all variables, the weight gain due to WC pregnancies is greater than that for SA pregnancies. This result may reflect different levels of care between ethnicities or may be the effect of genetic variation. Further research is required to assess the interaction of ethnicity on deprivation, poor engagement and/or poor compliance with treatment.