Cardiovascular mortality and chronotropic incompetence in systolic heart failure: the importance of a reappraisal of current cut-off criteria.

Aims An independent role for the exercise-induced heart rate (HR) response—and specifically the chronotropic incompetence (CI)—in the prognosis of heart failure (HF) is still debated. The multicentre study reported here sought to investigate the prognostic values of HR and CI variables on cardiovascular mortality in a large cohort of systolic HF patients. Methods and results A total of 1045 HF patients were recruited and prospectively followed in three Italian HF centres. The study endpoint was cardiovascular mortality. Besides a full clinical examination, each patient underwent a maximal cardiopulmonary exercise test at study enrolment. The age-predicted peak HR (%pHR) and the peak HR reserve (%pHRR) according to different cut-off values (60–80% of the maximum predicted) were adopted to identify the presence of CI. The median follow-up was 876 days (interquartile range 386–1590 days). Cardiovascular death occurred in 145 cases (13.8%). Besides LVEF, peak oxygen uptake, ventilation vs. carbon dioxide production slope, and beta-blocker therapy, the multivariate analysis showed that both %pHR and %pHRR were able to predict prognosis when considered as continuous variables. Conversely, the presence of CI was associated with the study endpoint only when the 70% (%pHR <70%, hazard ratio 1.80, confidence interval 1.24–2.61, P = 0.002; %pHRR <70%, hazard ratio 1.77, confidence interval 1.09–2.86, P = 0.020) or the 65% cut-off values (%pHR <65%, hazard ratio 2.04, confidence interval 1.34–3.10, P = 0.001; %pHRR <65%, hazard ratio 1.54, confidence interval 1.03–2.32, P = 0.038) were adopted. Conclusions Our findings demonstrated an additive role of CI in stratifying cardiovascular mortality. Both the 65% and the 70% cut-off values, regardless of the method (%pHR and %pHRR), allow identification of HF patients with the worst prognosis, thus supporting such definitions of CI in HF.