A matched control study of 126 women equally divided into three groups (normal oral glucose tolerance test, one abnormal test value, and gestational diabetes mellitus) was undertaken to examine the relationships among oral glucose tolerance test results, glycemic control in pregnancy, and adverse perinatal outcome. Characterization of metabolic control for the one abnormal oral glucose tolerance test value and the gestational diabetes mellitus groups (before treatment) showed no significant difference. After the start of treatment, however, a significant (p less than 0.01) difference between the groups in level of control was found. While no significant difference in the average birth weight between the three groups was discovered, the incidence of large infants (macrosomia and large for gestational age) was found to be significantly higher in the one abnormal oral glucose tolerance test group when compared with the normal (34% versus 9%; p less than 0.01) and gestational diabetes mellitus group (34% versus 12%; p less than 0.01). No significant difference for the incidence of an infant large for gestational age was found between the normal group and the patients with gestational diabetes mellitus after treatment. Neonatal metabolic disorders were found to be significantly higher for the one abnormal oral glucose tolerance test group (15%) when compared with the control and the gestational diabetes mellitus groups (3%). We conclude that, if left untreated, one abnormal value on an oral glucose tolerance test is strongly associated with adverse perinatal outcome.