Multifactorial Risk Stratification in Patients with Heart Failure, Chronic Kidney Disease, and Atrial Fibrillation: A Comprehensive Analysis

Abstract

Background: Heart failure (HF), chronic kidney disease (CKD), and atrial fibrillation (AF) frequently coexist, forming a high-risk triad that amplifies morbidity and mortality through shared pathophysiological mechanisms such as neurohormonal activation, fluid overload, and inflammation. Current risk stratification tools, including CHA₂DS₂-VASc and HAS-BLED, inadequately capture the complexity of these multimorbid patients. This study aims to explore the influence of comorbidities, hypertension severity, anticoagulation strategy, and risk scores on hospitalization outcomes in patients with coexisting HF, CKD, and AF. Materials and Methods: A retrospective case study was conducted on 174 hospitalized patients with HF, CKD, and AF. Clinical data included hypertension grade, HF phenotype (HFpEF vs. HFrEF), NYHA classification, renal function (KDIGO stage), stroke and bleeding risk scores (CHA₂DS₂-VASc: congestive heart failure, hypertension, age ≥ 75, diabetes, and stroke/TIA; HAS-BLED: hypertension, abnormal renal/liver function, stroke, bleeding, labile INR, elderly, and drugs/alcohol), comorbidities (neurological, psychiatric, COPD, and diabetes), anticoagulation type (DOACs vs. VKAs), and length of hospital stay. Statistical analysis included Spearman correlation, independent t-tests, and multivariate regression to evaluate associations between variables and clinical outcomes. Results: Most patients were elderly (mean age 75 ± 12), with advanced CKD (stage 3b) and systolic HF (77% HFrEF). Mean CHA₂DS₂-VASc was 5.67, HAS-BLED was 4.40, and ATRIA was 4.74, indicating high stroke and bleeding risk. Anticoagulation was predominantly via DOACs (69.5%). Hypertension severity did not significantly correlate with NYHA class (ρ = −0.14, p = 0.068). Neurological, psychiatric, and metabolic comorbidities showed no significant associations with HF severity. COPD and diabetes correlated inversely with CHA₂DS₂-VASc scores (ρ = −0.83, p = 0.014). No significant differences were observed in hospital stay between HF phenotypes or prior stroke history. In-hospital mortality was low (2.3%). Conclusions: Traditional risk scores do not fully capture the complexity of multimorbid patients. Metabolic comorbidities showed an inverse correlation with stroke risk scores, and no significant correlation was observed between hypertension severity and HF symptom burden. Hypertension and common comorbidities did not correlate with HF symptom burden, and metabolic diseases may paradoxically associate with lower stroke risk scores. These findings highlight the need for improved multimodal risk assessment strategies that consider the heterogeneity of multimorbid populations. Personalized, integrated approaches are essential to optimize anticoagulation, reduce hospitalization, and improve prognosis.