Diagnostic thresholds for gestational diabetes and their impact on pregnancy outcomes: a systematic review.

AIMS: To assess different diagnostic thresholds for gestational diabetes on outcomes for mothers and their offspring in the absence of treatment for gestational diabetes. This information was used to inform a National Institutes of Health consensus conference on diagnosing gestational diabetes. METHODS: We searched 15 electronic databases from 1995 to May 2012. Study selection was conducted independently by two reviewers. Randomized controlled trials or cohort studies were eligible if they involved women without known pre-existing diabetes mellitus and who did not undergo treatment for gestational diabetes. One reviewer extracted, and a second reviewer verified, data for accuracy. Two reviewers independently assessed methodological quality. RESULTS: Thirty-eight studies were included. Three large, methodologically strong studies showed a continuous positive relationship between increasing glucose levels and the incidence of Caesarean section and macrosomia. When data were examined categorically (i.e. women meeting or not meeting specific diagnostic thresholds), women with gestational diabetes across all glucose criteria had significantly more Caesarean sections, shoulder dystocia, macrosomia (except for International Association of Diabetes in Pregnancy Study Groups' criteria) and large for gestational age. Higher glucose thresholds did not consistently demonstrate greater risk for all outcomes. CONCLUSIONS: Higher glucose thresholds did not consistently demonstrate greater risk, possibly because studies did not compare mutually exclusive groups of women. A pragmatic approach for diagnosis of gestational diabetes using Hyperglycemia and Adverse Pregnancy Outcome Study odds ratio 2.0 thresholds warrants further consideration until additional analysis of the data comparing mutually exclusive groups of women is provided and large randomized controlled trials investigating different diagnostic and treatment thresholds are completed.

Voluntary regionalization and associated trends in perinatal care: the Nova Scotia Reproductive Care Program.

The Nova Scotia Reproductive Care Program is a system of voluntary regionalization that involves the 37 hospitals in the province that provide obstetric care to a population of 850,000. Between 1971 and 1980, the perinatal mortality rate in the central tertiary care unit for nonreferred patients fell progressively from 12.5 per 1,000 total births to 5.16. For all cases, including high-risk referrals, this rate has fallen from 12.7 to 7.2. During the same interval, the perinatal mortality rate for the province's seven regional hospitals fell from 18.7 to 12.2, and that for the 28 community hospitals fell from 18.4 to 7.0. Analysis of these reductions by fitted trend lines demonstrates statistical significance. Further analysis demonstrates that, with regionalization of perinatal services, it is possible to reduce the perinatal mortality rate in small community hospitals to levels that approximate those of a sophisticated tertiary care hospital.