We wished to determine whether gestational diabetes was associated with an increased perinatal mortality rate, and to investigate the cause for the observed increase in the incidence of gestational diabetes. We therefore reviewed the results of glucose tolerance tests and pregnancy outcome in 116,303 pregnancies, 1971-1994, at the Mercy Hospital for Women. The main outcome measurements were the presence or absence of gestational diabetes, and perinatal mortality. Over the entire period of the study, gestational diabetes was associated with an increased risk of perinatal mortality (Mantel-Haenszel adjusted odds ratio 1.53, 95% CI 1.13-2.06, p = 0.0069). Women with gestational diabetes that was only diagnosed retrospectively had a higher perinatal mortality rate than their contemporaries with normal glucose tolerance (OR 2.31, 95% CI 1.37-3.91, p = 0.0025). Women in whom a glucose tolerance test was not performed continued to have a higher perinatal mortality rate than women who were tested (adjusted OR 2.21, 95% CI 1.56-3.12, p < 0.00001). There has been an increase in the prevalence of gestational diabetes from 2.9% to 8.8%. Some of this is due to changes in population characteristics (increases in maternal age, obesity and proportion from South-East Asia), but there was still an independent increase over time. We conclude that identification and treatment of women with gestational diabetes can reduce perinatal mortality rates. Similarly to diabetes mellitus in the total population, the prevalence of gestational diabetes has increased over time.