Probability scores for heart failure with preserved ejection fraction in obstructive lung disease

Abstract Background Obstructive lung disease and heart failure with preserved ejection fraction (HFpEF) share risk factors and symptoms like dyspnea and reduced exercise capacity, leading to underdiagnosis of HFpEF. While the HFA-PEFF and H2FPEF scores aid in diagnosing HFpEF, their utility in patients with obstructive lung disease remains uncertain. Combined cardiopulmonary exercise testing with echocardiography (CPETecho) enables integrated evaluation of exercise limitations, including cardiac, pulmonary and peripheral causes. Purpose To assess the diagnostic utility of HFpEF risk scores (HFA-PEFF and H2FPEF) in patients with obstructive lung disease. Methods In this retrospective observational study, we evaluated patients with unexplained dyspnea who underwent CPETecho at four third-line hospitals between August 2016 and September 2023. CPETecho was performed on a semi-supine bicycle ergometer following spirometry. Obstructive lung disease was defined by a Tiffeneau index 3 mmHg/L/min with peak mPAP >30 mmHg and either exercise E/e′ >14 or ≥3 pulmonary B-lines. An HFA-PEFF score ≥5 and an H2FPEF score ≥6 indicated a high likelihood of HFpEF; scores ≤1 indicated a low probability. Results A total of 1146 patients were included (mean age 70 ± 10 years, 52% women). Obstructive lung disease was present in 137 patients (12%) and associated with a lower breathing reserve (28 ± 17% vs 41 ± 15%, p < 0.001) and a pulmonary limitation to exercise in 56%. HFpEF was diagnosed in 319 patients, with similar prevalence among those with and without obstructive lung disease (31% vs 27%). Patients with both obstructive lung disease and HFpEF demonstrated lower exercise capacity compared to those with obstructive lung disease alone (peak VO2: 12.4 ± 4.0 vs 17.5 ± 6.2 mL/min/kg, p < 0.001). The H2FPEF and HFA-PEFF scores demonstrated comparable diagnostic performance in patients with and without obstructive lung disease. Low scores (0–1) on either scale ruled out HFpEF with 100% sensitivity and NPV in both groups. High H2FPEF scores (6–9) confirmed HFpEF with high specificity (89% with vs 86% without obstructive lung disease), but modest PPV (52% and 47%) and limited sensitivity (26% and 33%). High HFA-PEFF scores (5–6) showed moderate specificity (72% and 74%) and sensitivity (45% and 61%) and similarly low PPV (41% and 47%). Adding diastolic stress testing to HFA-PEFF reduced specificity (61% and 67%) but increased sensitivity (76% and 80%), resulting in higher AUC (0.710 and 0.752) compared to high resting HFA-PEFF (0.584 and 0.678) or high H2FPEF (0.516 and 0.520). Conclusions In obstructive lung disease, H2FPEF and HFA-PEFF scores maintained diagnostic value. Low scores on either tool reliably excluded HFpEF, while high H2FPEF scores provided greater specificity than resting HFA-PEFF despite limited positive predictive value. Incorporating exercise into the HFA-PEFF score improved overall diagnostic performance.