Glucose telerance in pregnant African women: a study of outcome of pregnancy in diabetic patients and a study of serial glucose tolerance tests in normal non diabetic women

In Caucasians, pregnancy is considered to be associated with deterioration in glucose tolerance or even to be diabetogenic. In Africans, the effect of pregnancy on gluco tolerance has been reported in only two studies both of which reported no evidence of worsening of glucose tolerance. In one, however, the subjects were seen in or after the 30th week of gestation and in the other only 20 subjects were studied. The present study was undertaken to determine the outcome of diabetic pregnancies in African women and to establish the effect of pregnancy an glucose tolerance in non diabetic African women. Forty seven African women (with a total of SO pregnancies) with diabetes during pregnancy were seen at Muhimbili Medical Centre, Dar es Salaam between June 1986 and August 1989 and followed up until delivery. In the second part of the study 89 pregnant women seen in Dar es Salaam antenatal clinics during their first trimester had an oral glucose tolerance test in each trimester and postpartum. Of the diabetic women, 10 (40%) of the 25 insulin requiring patients had an obvious decrease in their insulin requirements, 6 (24%;) required more insulin and 5 (20%)had no change in their insulin requirements with advancing pregnancy. The outcome of the present pregnancies {44(88%) live births, 5 (l2%) perinatal deaths and 1 (2%) abortions) did not differ from the outcome of the 165 previous deliveries by these women: 136 (82%) live births, 17 (10%) perinatal deaths and 11 (7%) abortions. Neither did their pregnancy outcome differ significantly from the deliveries by non diabetic subjects at Muhimbili Medical Centre between July 1987 and June 1988: 13622 (93.5%) live births, 939 (6.4%) perinatal deaths. The distribution of birth weights were the same for the three groups of deliveries. In the study of serial glucose tolerance tests in non diabetic subjects, the mean fasting blood glucose level in the first trimester (3.9mmo1/1) was significantly higher than in the second (3.4mmol/1) and third trimesters (3.5mmol/1) and the postpartum period (3.6mmol/l). There were no significant changes throughout pregnancy and postpartum in the mean blood glucose level one hour after the glucose load. The mean blood glucose two hours after the glucose load for the first trimester (4.6mmol/l) was higher than the postpartum mean level of (4.2mmol /!). A11 other differences in the two hour blood glucose level between trimesters were not significant. Fourteen (24.1%) of 58 women who completed four glucose tolerance tests showed a progressive decrease or no change in fasting blood glucose throughout pregnancy and 15 (25.9%) showed a progressive decrease or no change in blood glucose 2 hour after the glucose load. Only one woman (1.7%) and five (8.6%) had a progressive increase in fasting and 2 hour blood glucose respectively. The mean body mass index and birth weight for women who showed improved glucose tolerance did not differ significantly when compared with subjects whose glucose tolerance did not improve. Also the proportion of subjects who showed improvement in glucose tolerance among the primigravidae and the housewives were not significantly different from the rest of the women. The results show that with good integrated cart of diabetic pregnant women, the fetal outcome is comparable to that of non diabetic women. Pregnancy does not impair glucose tolerance in most African women. International criteria for identification of glucose intolerance in pregnancy may not be applicable in Africans.