Abstract 16569: Exercise Testing Parameters and All-Cause Mortality in Patients Without Coronary Artery Disease: Results From the FIT Project
BACKGROUND: It remains unclear which exercise testing parameters best predict adverse health outcomes. We investigated routinely collected exercise parameters to determine which most strongly correlate with future MI and mortality.
Methods: This was a retrospective cohort of 69,886 adults who were clinically referred for exercise stress testing from 1991-2009 at a single health system. Patients with known CAD (n=12,454) or CHF (n=839) were excluded, leaving 56,593 patients. Future MI was determined by medical claims data, and death by the National Death Index. Forward and backwards stepwise Cox proportional hazards models were used to identify significant demographic and exercise variables for prediction of future MI or total mortality. Model performance was compared by Harrell’s c-statistic.
RESULTS: Over a median follow-up of 10 years (IQR 8-14), 1,209 patients suffered MI and 6,106 died. Median age was 52 (IQR 45-61) and 49% were female. Median METs achieved was 10 (IQR 7-10) and median % heart rate achieved was 92% (IQR 87-97%) of target for age. Of the variables evaluated; age, gender, METs achieved, and adequate study achievement (>90% target heart rate) were most highly predictive of all-cause mortality. Age per year had HR 1.06 (95% CI 1.06-1.07), female gender HR 0.51 (95% CI 0.49-0.54), METs per unit HR 0.83 (95% CI 0.82-0.84), and adequate study HR 0.68 (95% CI 0.65-0.72) for mortality (p <0.001 for all, Harrell’s c-statistic 0.81 for model). Addition of resting heart rate, resting systolic pressure, peak heart rate, peak systolic pressure, double product, and maximum predicted heart rate achievement did not add significant predictive survival to the initial model (p<0.001 for model, Harrell’s c-statistic now 0.82). Patients who achieved an adequate study and 10 METs had adjusted HR 0.32 (95% CI 0.27-0.38) for suffering an MI, and adjusted HR 0.30 (95% CI 0.27-0.32) for all-cause death compared to patients unable to achieve adequate study and 10 METs.
Conclusion: In 56,593 adults referred for stress testing, age, gender, METs, and adequate study achievement were most highly predictive of mortality. Patients who achieved 10 METs and adequate study had 70% lower risk of death over 10 years compared to patients unable to achieve these exercise parameters.
- © 2013 by American Heart Association, Inc.