Exercise and Cardiovascular Outcomes by Hypertensive Status: NHANES I Epidemiological Follow-up Study, 1971–1992

Background A favorable effect of exercise on cardiovascular longevity has been repeatedly demonstrated in the general population. The association of exercise and cardiovascular disease (CVD) outcome among persons with different blood pressure (BP) status is less well known. Methods We examined the epidemiologic follow-up of the First National Health and Nutrition Examination Survey (NHANES I) (1971–1992). Of 14,407 participants, 9791 subjects aged 25 to 74 years met inclusion criteria. All cause, CVD, and non-CVD mortality rates, as well as CVD incidence rates were determined. The associations of levels of exercise and outcomes by BP status were examined. Age- and gender-adjusted rates, as well as Cox proportional hazard models were determined. Results During 17 years of follow-up, there were 3069 deaths, 1465 of which were CVD. In addition, 2808 subjects had incident CVD events. Overall, CVD incidence and mortality rates increased as BP rose. The association of exercise with CVD events differed by BP status (normal, prehypertension, and hypertension). Age- and gender-adjusted CVD mortality rate per 1000 person-years for least, moderate, and most exercise were 5.0, 3.6, and 2.4 among normotensive subjects (P> .05), 6.3, 4.7, and 5.2 among prehypertensive subjects (P> .05), and 11.8, 9.8, and 8.7 among hypertensive subjects (P< .01), respectively. In fact, exercise was a significant independent predictor of reduced CVD event only among hypertensive subjects, after adjusting for other CVD risk factors. Among prehypertensive and normotensive subjects, where events were fewer, those who exercise more vigorously also had lower mortality, but these differences did not reach statistical significance. Conclusions This study, consistent with previous observational data, demonstrates that increased exercise is associated with decreased CVD event. Interestingly, this effect is most robust among hypertensive subjects, whereas for prehypertensive and normotensive subjects, a significant benefit of exercise on CVD outcome, perhaps because of lack of power, was not found.