Risk of Minor and Major Fetal Malformations in Diabetics with High Haemoglobin A1c Values in Early Pregnancy

Abstract

The incidence of congenital malformations is about three times higher in infants of insulin-dependent diabetic mothers than in the general population. The exact cause of this increase is not known, but experimental and clinical data show an association between congenital malformations and maternal hyperglycemia in early pregnancy, the critical period with regard to fetal malformations. The present authors report an association between the severity of maternal hyperglycemia in early pregnancy, as measured by blood hemoglobin A1c values, and the occurrence of fetal malformations in mothers with insulin-dependent diabetes. Between April 1978 and December 1982, the maternal hemoglobin A1c values had been determined at least once before the end of the 15th week of gestation in 139 insulin-dependent diabetic patients who gave birth after 24 weeks of gestation. In addition, a fetal malformation was observed in each of three cases of induced abortion in which early hemoglobin A1c determinations had been carried out. Pregnancies were classified as 1) not complicated by malformation, 2) complicated by minor malformation, and 3) complicated by major malformation. A malformation was classified as major if it was fatal or likely to cause serious handicap to the child. Other malformations were classified as minor. A total of 125 pregnancies were not complicated by fetal malformation. A malformed fetus or infant was observed in each of 17 pregnancies. In six cases, the anomalies were classified as minor, and in 11 cases they were classified as major. The mean initial hemoglobin A1c value was significantly higher in the group with minor malformations (9.3 per cent, with an SD of 1.9 per cent) and in the group with major malformations (9.6 per cent; SD, 1.8 per cent) than in the group without malformations (8.0 per cent; SD, 1.4 per cent) (P < 0.05 and P < 0.001, respectively). The difference in hemoglobin A1c values between the groups with minor and major malformations was not significant. In all of the 17 pregnancies complicated by malformation, the mean initial hemoglobin A1c value was 9.5 per cent (SD, 1.8 per cent), which was significantly higher (P < 0.001) than in the group without malformation. There was a significant positive relationship (x2 = 11.9; P = 0.001) between the maternal hemoglobin A1c value in early pregnancy and the occurrence of malformation. An initial hemoglobin A1c value of 10.0 per cent or more was associated with malformations in six of 17 pregnancies (35.3 per cent). Two of these were minor (11.8 per cent) and four were major (23.5 per cent) malformations. An intermediate hemoglobin Alc value (8.0–9.9 per cent) was associated with malformations in eight of 62 pregnancies (12.9 per cent). Relatively low initial hemoglobin A1c values, i.e., below 8.0 per cent, were associated with malformations in only three of 63 pregnancies (4.8 per cent).