Trajectories in New York Heart Association functional class in heart failure across the ejection fraction spectrum: data from the Swedish Heart Failure Registry

Abstract

Aims: To investigate incidence, predictors and prognostic implications of longitudinal New York Heart Association (NYHA) class changes (i.e. improving or worsening vs. stable NYHA class) in heart failure (HF) across the ejection fraction (EF) spectrum.

Methods and results: From the Swedish HF Registry, 13 535 patients with EF and ≥2 NYHA class assessments were considered. Multivariable multinomial regressions were fitted to identify the independent predictors of NYHA change. Over a 1-year follow-up, 69% of patients had stable, 17% improved, and 14% worsened NYHA class. Follow-up in specialty care predicted improving NYHA class, whereas an in-hospital patient registration, lower EF, renal disease, lower mean arterial pressure, older age, and longer HF duration predicted worsening. The association between NYHA change and subsequent outcomes was assessed with multivariable Cox models. When adjusting for the NYHA class at baseline, improving NYHA class was independently associated with lower while worsening with higher risk of all-cause and cardiovascular mortality, and first HF hospitalization. After adjustment for the NYHA class at follow-up, NYHA class change did not predict morbidity/mortality. NYHA class assessment at baseline and follow-up predicted morbidity/mortality on top of the changes. Results were consistent across the EF spectrum.

Conclusion: In a large real-world HF population, NYHA class trajectories predicted morbidity/mortality after extensive adjustments. However, the prognostic role was entirely explained by the resulting NYHA class, i.e. the follow-up value. Our results highlight that considering one-time NYHA class assessment, rather than trajectories, might be the preferable approach in clinical practice and for clinical trial design.

Keywords: Ejection fraction; Heart failure; NYHA functional class; Prognosis; Symptoms.

Figure 1

Changes in New York Heart Association (NYHA) class in the full cohort (A), stratified by ejection fraction (EF) (B) and by inpatient/outpatient status (C), and to and from each NYHA class (D).

Figure 2

Predictors of New York Heart Association (NYHA) class improvement. *p < 0.05; **p < 0.01; ***p < 0.001. Multivariable multinomial model included the variables labelled with a dagger (†) in Table  1 , the time between NYHA class assessments, and NYHA class at baseline as covariates. The full model, including non‐significant predictors, is shown in online supplementary Table  S5 . ARNi, angiotensin receptor–neprilysin inhibitor; CI, confidence interval; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; eGFR, estimated glomerular filtration rate (calculated by the Chronic Kidney Disease Epidemiology Collaboration formula); HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; ICD, implantable cardioverter‐defibrillator; MRA, mineralocorticoid receptor antagonist; NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide; NYHA, New York Heart Association; OR, odds ratio; RASi, renin–angiotensin system inhibitor.

Figure 3

Predictors of New York Heart Association (NYHA) class worsening. *p < 0.05; **p < 0.01; ***p < 0.001. Multivariable multinomial model included the variables labelled with a dagger (†) in Table  1 , the time between NYHA class assessments, and NYHA class at baseline as covariates. The full model, including non‐significant predictors, is shown in online supplementary Table  S6 . ARNi, angiotensin receptor–neprilysin inhibitor; CI, confidence interval; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; eGFR, estimated glomerular filtration rate (calculated by the Chronic Kidney Disease Epidemiology Collaboration formula); HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; ICD, implantable cardioverter‐defibrillator; MRA, mineralocorticoid receptor antagonist; NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide; NYHA, New York Heart Association; OR, odds ratio; RASi, renin–angiotensin system inhibitor.

Figure 4

Kaplan–Meier curves for (A) all‐cause mortality, (B) cardiovascular mortality, and (C) first heart failure (HF) hospitalization by New York Heart Association (NYHA) class change trajectory. CI, confidence interval; HR, hazard ratio.

Figure 5

Association between New York Heart Association (NYHA) class change and outcomes. *p < 0.05; **p < 0.01; ***p < 0.001. Adjusted hazard ratios (HR) with 95% confidence intervals (CI) for the association between longitudinal change in NYHA class and outcomes were calculated by multivariable Cox regression. Baseline model: adjusted for variables labelled with a dagger (†) in Table  1 , the time between the NYHA class assessments, and NYHA class at baseline. Follow‐up model: adjusted for variables labelled with a dagger (†) in Table  1 , the time between the NYHA class assessments, and NYHA class at follow‐up. Stratified model: low = NYHA class I–II; high = NYHA class III–IV; adjusted for variables labelled with a dagger (†) in Table  1 , the time between the NYHA class assessments, and NYHA class at follow‐up. CV, cardiovascular; HFH, heart failure hospitalization.