OBJECTIVE: We sought to determine the absolute risk of having a congenital anomaly in relation to periconceptional GHb concentration among women with prepregnancy diabetes. RESEARCH DESIGN AND METHODS: Two reviewers independently retrieved all cohort studies through a systematic literature search between January 1985 and May 2006. For each study, the absolute risk of having a pregnancy affected by a major or minor structural anomaly (diagnosed either antenatally or up to 28 days after conception) was calculated according to the number of SDs of GHb above the mean for nondiabetic, nonpregnant control subjects. A multilevel logistic-normal model was used to pool the data, which were expressed in tabular and graphic formats. RESULTS: In seven cohort studies, there were 117 anomalies among 1,977 pregnancies. At a periconceptional GHb concentration 0 SD above normal, the absolute risk of a pregnancy affected by a congenital anomaly was approximately 2% (95% CI 0.0-4.4). At 2 SD above normal, the risk was 3% (0.4-6.1), and at 8 SD it was approximately 10% (2.3-17.8). For each 1-SD unit increase in GHb, the associated risk of a congenital malformation increased by an odds ratio of 1.2 (95% CI 1.1-1.4). The risk in relation to A1C followed the same pattern. CONCLUSIONS: Using data from a limited number of published studies, a practical aid was developed to optimize use of the GHb and A1C concentrations for estimating the absolute risk of a congenital anomaly in the offspring of women with prepregnancy diabetes.
OBJECTIVE: We sought to determine the absolute risk of having a congenital anomaly in relation to periconceptional GHb concentration among women with prepregnancy diabetes. RESEARCH DESIGN AND METHODS: Two reviewers independently retrieved all cohort studies through a systematic literature search between January 1985 and May 2006. For each study, the absolute risk of having a pregnancy affected by a major or minor structural anomaly (diagnosed either antenatally or up to 28 days after conception) was calculated according to the number of SDs of GHb above the mean for nondiabetic, nonpregnant control subjects. A multilevel logistic-normal model was used to pool the data, which were expressed in tabular and graphic formats. RESULTS: In seven cohort studies, there were 117 anomalies among 1,977 pregnancies. At a periconceptional GHb concentration 0 SD above normal, the absolute risk of a pregnancy affected by a congenital anomaly was approximately 2% (95% CI 0.0-4.4). At 2 SD above normal, the risk was 3% (0.4-6.1), and at 8 SD it was approximately 10% (2.3-17.8). For each 1-SD unit increase in GHb, the associated risk of a congenital malformation increased by an odds ratio of 1.2 (95% CI 1.1-1.4). The risk in relation to A1C followed the same pattern. CONCLUSIONS: Using data from a limited number of published studies, a practical aid was developed to optimize use of the GHb and A1C concentrations for estimating the absolute risk of a congenital anomaly in the offspring of women with prepregnancy diabetes.