Background: In pregnant women with type 1 diabetes tight glycemic control reduces perinatal complications. Intensive observation of glucose profiles is essential in the achievement of tight glycemic control. The recent availability of the Continuous Glucose Monitoring System (CGMS, Minimed, Sylmar, CA) creates the opportunity to obtain more complete glucose profiles. This study was aimed at evaluating the accuracy of the CGMS in pregnant women with type 1 diabetes.
Methods: Five pregnant women with type 1 diabetes were asked to use two CGMS devices simultaneously. The simultaneously measured glucose levels were analyzed using the Pearson correlation, the mean absolute difference, and Bland-Altman analysis. Second, the percentage of concordance of paired data in the hypoglycemic, normoglycemic, or hyperglycemic range was calculated.
Results: The correlation coefficient between simultaneously measured data was 0.94 (P < 0.001). The mean absolute difference was 1.1 +/- 0.8 mmol/L. Bland-Altman analysis shows that 95% of the data pairs have a difference < or = 1.74 mmol/L. Almost 80% of the data pairs could be classified in the same glucose range. In 81% of the non-concordant pairs, one glucose value was classified in the hypoglycemic range and one in the normoglycemic range.
Conclusions: This study shows that the reproducibility of the CGMS in pregnant women with type 1 diabetes is adequate. This indicates that the CGMS is a useful tool in the management of type 1 diabetes in pregnant women. However, the CGMS should only be used as a supplementary method of daily glucose level measurement as a small degree of error, mainly in the hypoglycemic range, is present.