Changes in health-related quality of life and treatment effects in chronic heart failure: a meta-analysis.

Background: Heart failure (HF) is associated with poor health status, and high morbi-mortality. However, it is not well established how health status changes correlate with treatment effects on clinical outcomes. Our aim was to study the association between treatment-induced changes in health-status, assessed by Kansas City Cardiomyopathy Questionnaire 23 (KCCQ-23), and clinical outcomes in chronic HF.

Methods: Systematic search of phase III-IV pharmacological RCTs in chronic HF that assessed KCCQ-23 changes and clinical outcomes throughout follow-up. We studied the association between treatment induced changes in KCCQ-23 and treatment effects on clinical outcomes (HF hospitalization or cardiovascular death, HF hospitalization, cardiovascular death, and all-cause death) using weighted random-effects meta-regression.

Results: Sixteen trials were included, enrolling a total of 65,608 participants. Treatment induced KCCQ-23 changes were moderately correlated with treatment effects on the combined outcome of HF hospitalization or cardiovascular mortality (regression coefficient (RC) = -0.047, 95%CI: -0.085 to -0.009; R = 49%), a correlation that was mainly driven by HF hospitalization (RC = -0.076, 95%CI: -0.124 to -0.029; R = 56%). Correlations of treatment induced KCCQ-23 changes with cardiovascular death (RC = -0.029, 95%CI: -0.073 to 0.015; R = 10%) and all-cause death (RC = -0.019, 95%CI: -0.057 to 0.019; R = 0%) were weak and non-significant.

Conclusions: Treatment-induced changes in KCCQ-23 were moderately correlated with treatment-effects on HF hospitalizations but were not correlated with the effects on cardiovascular and all-cause mortality. Treatment-induced changes in patient-centered outcomes (i.e., KCCQ-23) may reflect non-fatal symptomatic changes in the clinical course of HF leading to hospitalization.