Clinician-Reported NYHA Class vs Patient-Reported Outcomes in Hospitalized HF

Patient-reported outcomes differ from clinician-reported NYHA functional class in patients hospitalized for heart failure.

In a large global cohort of patients hospitalized for heart failure (HF), the majority of individuals exhibited discordance between clinician-reported New York Heart Association (NYHA) functional class and patient reported outcomes (PROs), according to the results a study published in the journal Circulation: Heart Failure.

The prospective, double-blind, randomized, controlled ASCEND-HF (Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure; ClinicalTrials.gov Identifier: NCT00475852) study sought to assess the level of agreement between NYHA functional class and EuroQoL 5-dimensions (EQ-5D) questionnaire scores, identify unique patient characteristics associated with discordance between these metrics, and evaluate the clinical implications of this discordance. The study compared treatment with nesiritide vs placebo in patients hospitalized for HF, regardless of their ejection fraction.

In patients with complete baseline data available for NYHA class and the EQ-5D (both utility index [UI] and visual analog scale [VAS]), levels of each scale were mapped across 4 prespecified categories of “best” to “worst.” Minor discordance was defined as NYHA class and EQ-5D differing by 1 level, whereas moderate to severe discordance was defined as NYHA class and EQ-5D differing by 2 levels or more.

Among a total of 5741 patients, concordance, minor discordance, and moderate to severe discordance between NYHA class and EQ-5D UI were reported in 22%, 40%, and 38% of participants, respectively. Regarding NYHA class and EQ-5D VAS, concordance, minor discordance, and moderate to severe discordance were reported in 29%, 48%, and 23% of participants, respectively.

Multiple patient factors were independently associated with higher likelihood of moderate-severe discordance, including race and geographic region.

Among participants with any discordance between NYHA class and EQ-5D, better EQ-5D UI and EQ-5D VAS relative to NYHA class were reported in 78.1% and 70.2% of participants, respectively. Both EQ-5D UI and EQ-5D VAS demonstrated moderate correlation (Pearson correlation coefficient, 0.43; P <.001).

Among patients in whom discordance of EQ-5D UI was worse than NYHA class, a statistically significant association with increased 180-day all-cause mortality was reported (hazard ratio [HR], 1.27; 95% CI, 1.01-1.60; P =.042). This association continued within the sensitivity analysis, in which NYHA class I participants were imputed as NYHA Class II (HR, 1.29; 95% CI, 1.02-1.62; P =.036).

Per Kaplan-Meier survival analysis, patients with EQ-5D UI that was worse than NYHA class had the highest 180-day all-cause mortality, whereas those with concordance had intermediate mortality and those with EQ-5D UI better than NYHA class had the lowest mortality (P <.001).

Several limitations of the study include NYHA class and EQ-5D data both being recorded at the baseline visit of the trial and evaluation of NYHA class and EQ-5D assessment not occurring simultaneously. Additionally, no clear EQ-5D score cutoffs are available that match the 4 NYHA classes. Further, although full concordance between NYHA class and EQ-5D were not expected, the remarkable extent of discordance implies that clinician and patient perspectives on functional status generally may differ.

“Multiple patient factors were independently associated with higher likelihood of moderate-severe discordance, including race and geographic region,” the study authors wrote. “Patients who perceive their health status as worse than the status perceived by clinicians may have particularly high risk of poor outcomes, but further studies are required to confirm this finding.”

Disclosure: Some of the study authors have declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures. 

References:

Cosiano MF, Vista A, Sun JL, et al. Comparing New York Heart Association class and patient-reported outcomes among patients hospitalized for heart failure. Circ Heart Fail. Published online October 31, 2022. doi:10.1161/CIRCHEARTFAILURE.122.010107