Clinical Predictors of Heart Failure and Short-Term Outcomes in Frail Adults: A Single-Center Retrospective Cohort Study

Abstract

Introduction Heart failure (HF) remains a leading cause of morbidity and mortality in frail adults. Traditional prognostication relies on ejection fraction and cardiac biomarkers, yet this approach may not adequately capture multifactorial vulnerability in frail populations. Although N-terminal pro-B-type natriuretic peptide (NT-proBNP) along with ejection fraction phenotype has emerged as an important risk stratification marker, its prognostic utility in specific populations remains incompletely characterized. Frailty is increasingly recognized as a critical determinant of prognosis in HF patients. The interrelationship between biomarker elevation, frailty, and short-term outcomes in hospitalized older adults requires investigation. Methods This retrospective observational cohort study was conducted at a secondary care hospital in the United Kingdom between September and December 2024, utilizing data from 240 true HF admissions from a local audit and service improvement initiative. The study enrolled 98 consecutive eligible admissions with confirmed Rockwood Clinical Frailty Score ≥ 5, acute HF diagnosis, and complete 12-week follow up data. Frailty was measured using the Rockwood Clinical Frailty Scale, and NT-proBNP was measured using standard laboratory assays. Ejection fraction parameters were obtained from admission echocardiographic assessment. Pre-existing cardiovascular risk factors-atrial fibrillation (AF), chronic kidney disease (CKD), and coronary artery disease (CAD)-were gathered retrospectively on a case-by-case basis. The composite primary outcome was death or hospitalization for acute HF within 12 weeks. Statistical analyses included Mann-Whitney U tests, chi-squared analyses, binary logistic regression, and receiver operating characteristic (ROC) analysis. Results The cohort comprised 98 adults (mean age 76.4 years; Frailty Scale 5.38) with a mean admission NT-proBNP of 5,060 pg/mL. CKD, AF, and CAD were present in 20.4%, 32.7%, and 33.7%, respectively. HF phenotypes included 41.5% in HF with preserved ejection fraction (HFpEF), 35.4% in HF with mildly reduced ejection fraction (HFmrEF), and 23.1% in HF with reduced ejection fraction (HFrEF). Adverse outcomes occurred in 33.7%. Age was significantly associated with poor prognosis. Despite limited multivariable analysis due to small sample size, NT-proBNP and frailty showed directional associations. Multivariable logistic regression identified no independent predictors after adjustment. ROC analysis showed acceptable discrimination by admission NT-proBNP (area under the curve (AUC) 0.703). Conclusion Short-term prognosis in frail HF populations reflects complex multifactorial vulnerability rather than isolated biomarker elevation. Although NT-proBNP demonstrated acceptable discriminatory ability, the absence of independent multivariable predictors indicates that prognostication cannot rely on isolated cardiac biomarkers alone. Rather, outcomes are fundamentally determined by age-related physiological decline, comorbidity burden, neurohormonal activation, and reduced physiological reserve. Furthermore, ejection fraction phenotype was not significantly associated with adverse outcome and showed that the vulnerability to acute stressors often eclipses isolated ventricular metrics. These findings emphasize incorporating comprehensive assessment principles into HF prognostication rather than adhering to HF guidelines.