Both pregnancy-induced hypertension and high-altitude residence are associated with fetal growth retardation, thereby leading us to hypothesize that pregnancy-induced hypertension would be more common at high than at low altitude. Retrospectively collected data in Colorado revealed that pregnancy-induced hypertension was more common at 3,100 m (12%) than at 2,410 m (4%) or 1,600 m (3%) (P less than 0.001). Proteinuria and edema in the upper extremities were also more frequent at 3,100 m than at 1,600 m (proteinuria = 28% versus 9%, P less than 0.001; edema of upper extremities = 22% versus 13%, P less than 0.01). Blood pressure during pregnancy increased with altitude among all women and those without pregnancy-induced hypertension (analysis of variance, P less than 0.05). In a prospective study at 3,100 m, women with pregnancy-induced hypertension had no change in blood pressure during pregnancy prior to the onset of hypertension (analysis of variance, P = NS), whereas blood pressure decreased in the normal women (analysis of variance, P less than 0.05). Arterial oxygen saturation during the third trimester was inversely related to the degree of hypertension in women with pregnancy-induced hypertension at 3,100 m (r = 0.8, P less than 0.05), thus suggesting that maternal hypoxia may play a previously unsuspected role in the etiology of pregnancy-induced hypertension.