Relative glucose tolerance and subsequent development of hypertension in pregnancy

Objective: To test the hypothesis that relative carbohydrate tolerance, an indicator of insulin resistance, predicts subsequent risk for hypertension of pregnancy among previously normoglycemic, normotensive women. Methods: We conducted a nested case-control study in women enrolled at a large Colorado urban health maintenance organization. Subjects were previously healthy pregnant women who tested abnormal on their initial 50-g glucose screens and subsequently completed 3-hour, 100-g oral glucose tolerance tests. Cases were 54 previously normotensive women who subsequently developed hypertension and controls were 51 subjects with normotensive pregnancies, matched to cases on parity. Subjects diagnosed with gestational diabetes (17 cases, six controls) were excluded from the main analyses. Results: Among the 82 normoglycemic women (45 controls, 37 cases, 13 preeclampsia, 24 gestational hypertension), mean post-load glucose levels and total glucose area under the curve were significantly higher in cases than in controls (P less than or equal to .04) and were positively correlated with peak mean arterial pressure. After adjustment for potential confounders, P-hour post-load glucose levels remained strongly related to risk for hypertension (adjusted odds ratios = 1.48; 95% confidence interval 1.13, 1.92 per 10 mg/dL increase) and to peak mean arterial blood pressure (r = .23, P = .04), as did total glucose area under the curve (P less than or equal to .04). Cases were also more likely to have had one abnormal glucose tolerance test (28% versus 5%, P = .004). Stratifying analyses by case severity (preeclampsia and gestational hypertension) yielded similar results. Among all subjects, more cases than controls were also diagnosed with gestational diabetes (31% versus 12%, P = .008). Conclusion: These findings are consistent with the hypothesis that insulin resistance precedes the clinical onset of hypertension in pregnancy, and may be important in the etiology of hypertension. (Obstet Gynecol 2001;97:905-10. (C) 2001 by The American College of Obstetricians and Gynecologists.).