Heart Rate Recovery Index and Pulse Wave Velocity in HFpEF: Exploratory Cutoff Values from a Cross-Sectional Study

Background/Objectives: Heart failure with preserved ejection fraction (HFpEF) remains challenging to diagnose, and established algorithms such as HFA-PEFF and H2FPEF frequently classify patients within intermediate probability categories requiring additional testing. Arterial stiffness and cardiac autonomic dysfunction are increasingly recognized as relevant components of HFpEF pathophysiology, yet the combined discriminatory value of pulse wave velocity (PWV) and heart rate recovery index (HRRI) has not been well established. This study aimed to evaluate the individual and combined discriminatory performance of PWV and HRRI for HFpEF and to assess their incremental contribution beyond established clinical risk factors. Methods: In this prospective, single-center, cross-sectional study, 180 participants were included, comprising 86 patients with HFpEF and 94 participants without heart failure. Group allocation was clinically assigned by the treating cardiologist with access to PWV and HRRI results. PWV was assessed non-invasively using an oscillometric device, and HRRI was derived from standardized cycle ergometer exercise testing. Univariable and multivariable logistic regression, receiver operating characteristic (ROC) curve analysis, and likelihood-ratio testing were performed to assess independent associations and incremental model contribution. Sensitivity analyses evaluated the influence of age modelling and beta-blocker use. Results: Both PWV and HRRI were independently associated with clinically assigned HFpEF status after adjustment for age, sex, body mass index, hypertension, diabetes mellitus, dyslipidemia, and atrial fibrillation (adjusted OR 1.330 per 1 m/s increase in PWV and 0.857 per 0.1-unit increase in HRRI; both p ≤ 0.023). PWV and HRRI demonstrated AUCs of 0.770 and 0.792, respectively, with Youden index-derived cutoff values of ≥9.60 m/s and ≤1.97. Their combination significantly improved discrimination compared with either marker alone (AUC 0.833). Adding both markers to the clinical model improved model fit, but the AUC increase from 0.865 to 0.896 was not statistically significant after Holm adjustment (adjusted p = 0.132). The associations of both markers with HFpEF remained essentially unchanged after adjustment for beta-blocker therapy. Conclusions: PWV and HRRI were associated with clinically assigned HFpEF status; the observed discrimination does not establish independently validated diagnostic accuracy. Incremental discrimination beyond clinical characteristics and clinical usefulness remain to be established. The cutoff values are exploratory and should not be used as clinical decision thresholds without formal measurement reliability assessment and independent prospective validation.