Reproducibility of the oral glucose tolerance test in pregnant women.

Objective: The purpose of this study was to evaluate the reproducibility of the 3-hour oral glucose tolerance test during pregnancy and the potential factors associated with nonreproducible results.
Study Design: Thirty-eight women with a 1-hour glucose level > or 135 mg/dl had a 100 gm oral glucose tolerance test. During the test samples were obtained for glucose, insulin, cortisol, human placental lactogen, and norepinephrine levels. The oral glucose tolerance test was repeated 1 week later under similar metabolic conditions.
Results: The intraassay coefficient of variation in glucose from week 1 to week 2 was < 2%. There were no significant differences in the paired fasting 1-, 2-, or 3-hour glucose concentrations (p = 0.51 to 0.96) or the area under the glucose curve (p = 0.43) from week 1 to week 2, although the mean absolute difference in glucose values ranged from 4 (fasting) to 18 (3 hours) mg/dl. Oral glucose tolerance test results were classified as either normal or abnormal from week 1 to week 2; 16 normal/normal, 13 abnormal/abnormal, seven abnormal/normal, and two normal/abnormal. Norepinephrine (p = 0.03) and insulin (p = 0.05) were significantly greater in week 1 but not in week 2 in the abnormal/normal versus normal/normal and abnormal/abnormal groups. There were no significant differences in cortisol or human placental lactogen levels among groups at any time.
Conclusions: The oral glucose tolerance test was not reproducible for diagnosis in 24% (nine of 38) of pregnant women. We speculate that maternal stress (increased norepinephrine) may have been a factor for the abnormal results in week 1 in the abnormal/normal group.