Diagnostic performance of the H2FPEF score and cardiovascular burden in patients with chronic kidney disease: a cross-sectional study

Background

Patients with chronic kidney disease (CKD) carry a high risk of cardiovascular complications. Among these, heart failure with preserved ejection fraction (HFpEF) is prevalent but often remains unrecognised in routine outpatient care. In addition to traditional cardiovascular risk factors, CKD is characterised by non-traditional contributors such as chronic inflammation, vascular stiffness, anemia, and volume overload, which may promote a distinct cardio-renal HFpEF phenotype.

Purpose

The present study aimed to identify occult HFpEF in CKD and to evaluate the diagnostic performance of the H2FPEF score.

Methods

Adults with CKD stages 3–4 (eGFR 15–59 mL/min/1.73 m², CKD-EPI 2021) were prospectively recruited from outpatient visits. Patients with preserved ejection fraction (LEVF) >45% were included, while exclusion criteria included dialysis, congenital heart disease, pregnancy, active malignancy, and established coronary artery disease. The study was approved by the local ethics committee (no. 1797/2024). All participants provided written informed consent. Clinical evaluation, laboratory testing, and transthoracic echocardiography were performed at enrolment according to American Society of Echocardiography / European Association of Cardiovascular Imaging (ASE/EACVI) recommendations. HFpEF probability was estimated using the H2FPEF score, while HFpEF diagnosis was established using the HFA-PEFF algorithm recommended by the European Society of Cardiology (ESC) heart failure guidelines. Statistical analyses were performed using JASP version 0.95.4.

Results

The study included 161 patients with CKD stages 3–4 (baseline characteristics shown in Table 1).

According to the HFA-PEFF diagnostic algorithm, HFpEF was identified in 50.9% of patients. The median H2FPEF score was 3 (IQR 2–4), with 10.6% classified as high probability. The H2FPEF score showed modest discrimination for HFpEF (AUC 0.62). A high score (≥6) demonstrated high specificity (91.1%) but low sensitivity (12.2%) for definite HFpEF and was significantly associated with the diagnosis (p = 0.008), supporting its role as a rule-in rather than rule-out tool in CKD (Figure 2). Most patients with HFpEF were classified in the intermediate probability category, whereas low probability scores were uncommon among confirmed HFpEF cases. Conversely, high probability scores were more frequent in patients with HFpEF than in those without, consistent with preserved specificity but limited screening capability.

Conclusions

In patients with moderate-to-advanced CKD, previously unrecognised HFpEF was identified in approximately half of participants using guideline-based diagnostic criteria. The H2FPEF score showed modest discrimination but high specificity at elevated values, supporting its use primarily as a rule-in rather than a screening tool in this population.Figure 1