Discordance in prognostic ability between physician assessed NYHA classification and self-reported health status in patients with acute heart failure

Abstract

Abstract Background Especially in patients with acute heart failure (AHF) the NYHA classification remains of uncertain representation of patients' actual health state. Alternatively, patient's subjective well-being, in terms of health-related quality of life (HRQL), showed to have an excellent prognostic ability in out clinic patients with chronic heart failure. Objectives It is unknown whether HRQL instruments can assess a more reliable prognostication in patients hospitalized due to AHF than the NYHA classification. Methods Goal Directed Afterload Reduction in Acute Congestive Cardiac Decompensation Study (GALACTIC) was a multicenter, randomized, open-label blinded-end-point trial that emphasized early intensive and sustained vasodilation in adult patients hospitalized due to AHF with NYHA functional class III/IV, however provided neutral findings. HRQL was assessed by the generic EQ-5D-3L which is a 3-leveled 5-item instrument and the disease-specific Kansas City Cardiomyopathy Questionnaire (KCCQ). Unadjusted and adjusted Cox regression models were performed after patients were grouped into low (EQ-5D −0.074<0.25; KCCQ 0<25), moderately low (0.25<0.5; 25<50), moderately high (0.5<0.75; 50<75) and high HRQL (0.75–1.0; 75–100). Results 781 patients were enrolled in 10 centres in 5 countries over 2 continents among which 536 (69%) patientshad completed theEQ-5D and 419 (54%) the KCCQ shortly after admission. Within 180 days of follow-up69 (13%) and 54 (13%) patients died and 151 (28%) and 122 (29%) died or were rehospitalized due to AHF, respectively. Cumulative incidence as well as HRs in patients grouped according to NYHA (n=536) indicated a comparable or significantly lower risk in patients with NYHA IV: e.g. for the combined outcome HR 1.07 (95% CI 0.777–1.473) and aHR 0.463 (95% CI 0.245–0.875). Whereas HRs in patients grouped according to both, EQ-5D (n=536) and KCCQ (n=419), increased from the group with highest to the group with the lowest HRQL: e.g. aHR for moderately high 1.11 (95% CI 0.718–1.715), for moderately low 1.721 (95% CI 1.102–2.688) and for low EQ-5D index 1.891 (95% CI 1.136–3.149) referenced to high HRQL (EQ-5D index 0.75–1.0). Conclusions These findings corroborate and extend previous work suggesting that NYHA classification poorly discriminates AHF patients' prognosis and challenge its' extensive application. HRQL might be a possible alternative to easily assess these patients' heath state. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Swiss National Science Foundation, the Swiss Heart Foundation A. 180-day mortality; B. composite outcome