Alternative methods of diagnosing gestational diabetes mellitus

OBJECTIVE: In an attempt to find more efficacious alternatives far the diag nosis of gestational diabetes mellitus, we evaluated whether (1) there is a glucose loading test value above which all glucose tolerance test results are positive, (2) omission of the third-hour plasma glucose measurement of the glucose tolerance test alters the sensitivity of the test, and (3) the presence of a fasting plasma glucose concentration greater than or equal to 105 mg/dL suffices as a diagnostic standard after an abnormal glucose load ing test result. STUDY DESIGN: The charts of 512 patients who underwent 3-hour glucose toler ance tests at our institution between January 1995 and December 1996 were r eviewed. Only subjects for whom the glucose loading test yielded plasma glu cose levels greater than or equal to 140 mg/dL were selected. The positive predictive value of a glucose loading test result greater than or equal to 185 mg/dL was calculated. Results of glucose tolerance tests of subjects wi th elevated fasting plasma glucose concentrations were then evaluated to de termine the positive predictive value for gestational diabetes mellitus of an elevated fasting plasma glucose concentration. RESULTS: Among the subjects who underwent glucose tolerance tests, 22% (114 /512) met positive test criteria for gestational diabetes mellitus. The pos itive predictive, value for a glucose loading test result greater than or e qual to 185 mg/dL was 57% (25/44), whereas a glucose loading test result >1 99 mg/dL showed a positive predictive value of 69% (4/13). Omission of the third-hour glucose tolerance test value yielded a sensitivity of 87% (99/11 4). Among the 24 women with fasting plasma glucose concentrations greater t han or equal to 105 mg/dL, 96% had positive glucose tolerance test results. An elevated fasting plasma glucose concentration was highly associated wit h gestational diabetes mellitus necessitating insulin therapy (65%). CONCLUSION: An elevated glucose loading test result was associated with but not highly predictive of gestational diabetes mellitus. Omission of the a- hour glucose tolerance test measurement resulted in failure to diagnose 13% of gestational diabetes mellitus cases. A fasting plasma glucose concentra tion greater than or equal to 105 mg/dL was highly predictive of an abnorma l glucose tolerance test result among patients with an elevated glucose loa ding test value.