Reliability of peak exercise testing in patients with heart failure with preserved ejection fraction

Abstract

Exercise intolerance is the primary symptom in patients with heart failure and preserved ejection fraction (HFpEF), a major determinant of their decreased quality of life, and an important outcome in clinical trials. Although cardiopulmonary exercise testing (CPET) provides peak and submaximal diagnostic indexes, the reliability of peak treadmill CPET in patients >55 years of age with HFpEF has not been examined. Two CPETs were performed in 52 patients with HFpEF (70 ± 7 years old). The 2 tests were separated by an average of 23 ± 13 days (median 22) and performed under identical conditions, with no intervention or change in status between visits except for initiation of a placebo run-in. A multistep protocol for patient screening, education, and quality control was used. Mean peak oxygen consumption was similar on tests 1 and 2 (14.4 ± 2.4 vs 14.3 ± 2.3 ml/kg/min). Correlation coefficients and intraclass correlations from the testing days were determined (oxygen consumption, r = 0.85, p <0.001, intraclass correlation 0.855; ventilatory anaerobic threshold, r = 0.79, p <0.001, intraclass correlation 0.790; ventilation per carbon dioxide slope, r = 0.87, p <0.001, intraclass correlation 0.864; heart rate, r = 0.94, p <0.001, intraclass correlation 0.938). These results challenge conventional wisdom that serial baseline testing is required in clinical trials with exercise-capacity outcomes. In conclusion, in women and men with HFpEF and severe physical dysfunction, key submaximal and peak ET variables exhibited good reliability and were not significantly altered by a learning effect or placebo administration.

Figure 1

Bland-Altman plot for change in peak VO₂ versus mean peak VO₂ for test 1 and test 2.