By Michael H. Crawford, MD, Editor

SYNOPSIS: An observational study of leisure time physical activity (LTPA) assessed at baseline and two years later in stable coronary artery disease patients, who then were followed for about five more years, showed that LTPA at baseline, at two years, and if it went from zero at baseline to some at two years was associated with lower rates of cardiac death compared to inactive patients.

SOURCE: Lahtinen M, Toukola T, Junttila MJ, et al. Effect of changes in physical activity on risk for cardiac death in patients with coronary artery disease. Am J Cardiol 2018;121:143-148.

Leisure time physical activity (LTPA) is recommended for secondary prevention in patients with stable coronary artery disease (CAD), yet studies of its beneficial effects have produced mixed results. One explanation is that most researchers assess LTPA once and do not know if it changes over the observation period. Investigators from Finland assessed LTPA over two years in CAD patients and associated it with cardiac mortality over up to seven years of observation. The study population was derived from patients in the Innovation to Reduce Cardiovascular Complication of Diabetes at the Intersection study database of 1,946 patients with stable angiographically verified CAD and type 2 diabetes. Patients with serious, life-threating comorbidities such as Class IV heart failure and end-stage renal disease were excluded. Also excluded were those with less than two years follow-up. During the initial follow-up, information about LTPA was collected at baseline and two years. Based on these data, four groups were identified: no LPTA, irregular light LPTA, regular moderate intensity LTPA, and moderate- to high-intensity LTPA for at least 30 minutes three times a week. Also, subjects were divided by their activity at baseline compared to two years into four other groups: Active-Active or greater than or equal to irregular at both time periods, Inactive-Active, Active-Inactive, and Inactive-Inactive. The main follow-up started at the two-year point, and the primary endpoint was cardiac death.

During a median follow-up of 54 months, there were 68 cardiac deaths. Baseline LTPA was associated with cardiac death in multivariate analyses: irregularly active hazard ratio (HR) = 1.4 (95% confidence interval {CI], 0.6-3.2) and inactive HR = 4.7 (95% CI, 2.1-10.6; P < 0.05) compared to the active groups. The LTPA at two years was associated with cardiac death: inactive HR = 8.0 (95% CI, 2.8-22.4; P < 0.05), irregularly active HR = 1.7 (95% CI, 0.6-5.3), and moderate regular HR = 1.9 (95% CI, 0.6-5.7) compared to the highly active group. In addition, the change in LTPA over the two years was associated with cardiac death: Inactive-Inactive HR = 7.6 (95% CI, 4.2-13.6; P < 0.05), Active-Inactive HR = 3.7 (95% CI, 2.1-6.7; P < 0.05), and Inactive-Active HR = 1.6 (95% CI, 0.4-6.7). The authors concluded that LTPA is of prognostic value in stable CAD patients, and even small changes in activity over two years were related to the risk of cardiac death over a median of 4.5 years.


Current acute coronary syndrome guidelines recommend referral to cardiac rehabilitation. One of the goals of cardiac rehab is to convert patients to a long-term exercise habit. This study clearly demonstrates why this is important: Regular physical activity is associated with longevity in CAD patients. The study also reinforces that it doesn’t take much activity to demonstrate this difference: The highly active group performed moderate- to high-intensity LTPA for a minimum of 30 minutes three times a week. Even the irregular light activity and regular moderate activity groups were not statistically different from the highly active groups. Only the completely inactive group demonstrated a statistically significant increase in cardiac death compared to all three active groups. Thus, we can tell our patients that any degree of LTPA carries some survival value.

The unique contribution of this study is the two-year survey of the patient’s activity history before the follow-up period began. The two-year data largely mirrored the baseline activity data in the prediction of cardiac death. It is interesting that more patients became inactive at two years than became active. This pattern was associated with an increase in cardiac death compared to those who remained active. Also, those who did go from inactive at baseline to active at two years reduced their risk of cardiac death. Thus, efforts to get CAD patients active and keep them active are worthwhile.

There are limitations to this study. Since it was not a randomized trial, there may have been a patient self-selection bias where the healthiest were more likely to be active. This potential issue was observed when the baseline exercise capacity was factored into the multivariate model and the association between LTPA and cardiac death was attenuated. Unfortunately, exercise capacity was not remeasured at two years. There are minimal details on LTPA, and it was a subjective assessment. Also, very few patients went from inactive to active, so this limits conclusions about its effects on mortality.

On the positive side, we know that strenuous physical activity in CAD patients can precipitate acute coronary events, but this was not observed with LTPA in this study. Thus, it appears generally safe to recommend regular LTPA for stable CAD patients.