Nocturnal hypoxemia secondary to sleep apnea has long been implicated as a cardiovascular risk factor in renal failure, but to date there is no study that links nocturnal hypoxemia to cardiovascular outcomes in end-stage renal disease. Fifty uremic patients on regular dialysis treatment without primary sleep apnea, pulmonary diseases, or other illnesses that may cause sleep apnea underwent pulse oximetry studies during night and were followed up for 32 mo. Average nocturnal SaO(2), minimal SaO(2), and the number of episodes of hypoxemia were similar in patients who died during the follow-up and in patients who survived, and none of these parameters predicted all-cause mortality. Average nocturnal SaO(2) was significantly lower (P = 0.006) in patients who had cardiovascular events during the follow-up (94.7 +/- 2.9%) than in event-free patients (97.1 +/- 1.3%). In a Cox model, average nocturnal SaO(2) was the second factor in rank explaining these outcomes. In this model a 1% decrease in average nocturnal SaO(2) was associated with a 33% increase in the incident risk of fatal and nonfatal cardiovascular events. Furthermore the risk of cardiovascular events was 5.05 times higher in patients with average nocturnal SaO(2) <95% (95% CI 1.61 to 15.86) than in those above this threshold (P = 0.005). This study adds weight to the hypothesis that nocturnal hypoxemia in dialysis patients represents an important cardiovascular risk factor.