The reproducibility of the 50-g, 1-hour glucose screen for diabetes in pregnancy.

OBJECTIVE

To explore the day-to-day reproducibility of the 50-g, 1-hour glucose screening test performed on 2 consecutive days in the same women.

METHODS

Eight women at 12-23.6 weeks' gestation (early subjects) and 80 women at 24-28 weeks (late subjects) without known diabetes mellitus were studied. The glucose screening test was performed in the morning on 2 consecutive days in the same women, under an identical or opposite sequence of fasting and fed conditions. The women were thus divided into four subgroups: fast-fast, fed-fed, fast-fed, and fed-fast. Duplicate serum glucose concentrations were measured by the glucose oxidase method. Paired Student t test was used to analyze day-1 versus day-2 glucose levels in each woman of each subgroup.

RESULTS

The serum glucose concentrations were higher on day 1 than on day 2 in the subgroups fast-fast and fast-fed (P < .05) in both early and late patients, whereas the opposite was seen in the subgroups fed-fast (P < .05). No significant differences were observed in the subgroups of fed-fed. At three different glucose thresholds (130, 135, and 140 mg/dL), there was more than 90% daily reproducibility for normal results in both groups, and nearly 50 and 83% daily reproducibility for abnormal results in the early and late patients, respectively.

CONCLUSIONS

Up to 28 weeks of pregnancy, the screening test had a high reproducibility for normal results. For abnormal results, daily reproducibility was better after than before 24 weeks' gestation, regardless of prior testing conditions. Depending on the pre-testing conditions, the use of a different serum glucose threshold seems warranted.