A low-glycaemic load (GL) diet in overweight or obese women did not affect the birthweight or body fat of their babies when compared with a low-fat diet. However, women in the former group had smaller increases in triglycerides and total cholesterol and a greater decrease in C-reactive protein (CRP) than those in the latter. Infant head circumference and duration of pregnancy were also greater in this group.

A total of 325 pregnant women with BMI 25-44 kg/m² were assessed for eligibility to the study and following screening 46 met the inclusion criteria and were entered in the trial. These women were randomised to either a standard low-fat diet (21) or to a low-GL diet (25) on enrollment during the second or third trimester. Low-GL diets reduce the proportion of carbohydrates with a high glycaemic index that evoke greater blood sugar surges and as a consequence compensatory insulin responses.

Dietary and exercise counselling was given, and to improve ‘dietary fidelity' participants were provided with various complimentary food supplies. Nutritional adherence was assessed by unannounced telephone interview.

No differences in maternal weight gain or skin-fold thickness were noted between the groups from baseline at around 19 weeks gestation till delivery.

However, the natural rise in cholesterol and triglycerides in pregnancy was significantly less in the low-GL diet women and the decrease in CRP was greater. Median increase in triglycerides in the low-GL group was 49 compared with 93 mg/dl, P=0.03, in the low-fat group; median respective increases in total cholesterol were 13 and 33 mg/dl; P=0.04. CRP decreased by a median of 2.5 in the low-GL group compared with 0.4 mg/dl in the low-fat group; P=0.007.

No difference in birthweight was noted but infants born to low-GL diet mothers had a greater head circumference. Among low-fat diet exposed babies preterm birth was more common, and average gestation at delivery was lower 37.9±3.1 compared with 39.3±1.1 weeks.

Rising rates of overweight and obesity within society in general are increasingly being mirrored in women during pregnancy. Current dietary advice does not seem to be achieving significant improvements and not surprisingly alternative dietary interventions are under scrutiny.

Preterm and early-term births contribute significantly to perinatal morbidity and head circumference is directly related to brain volume and possibly later IQ. Drawing firm conclusions from pilot studies is difficult. Although a low-GL diet was associated with a more favourable maternal biochemistry profile and with better birth statistics, the number of subjects in the study was small. Larger studies are needed to evaluate whether a low-GL diet does improve maternal and infant outcomes.

Dr Chris Barclay