Predictive accuracy of the H2FPEF score in patients with and without lung disease
C Wegner Wippel, T Wilson, E Deych, N SweitzerAbstract
Background
The H2FPEF score is a tool used to diagnose Heart Failure with Preserved Ejection Fraction (HFpEF). Its accuracy in diagnosing HFpEF in patients with concomitant lung disease is not well-established.
Purpose
This study aimed to evaluate and compare the predictive accuracy of the H2FPEF score for diagnosing HFpEF in patients with and without lung disease undergoing right heart catheterization (RHC).
Methods
We conducted a retrospective chart review of patients who underwent RHC. After excluding patients with ejection fraction <50%, significant valvular disease, or prior heart transplantation, we extracted data from 305 unique patients. HFpEF and lung disease status were defined using clinical and hemodynamic criteria. The H₂FPEF score was calculated for all patients, who were then stratified into two groups based on the presence or absence of lung disease. Lung disease encompassed chronic obstructive pulmonary disease (COPD), asthma, obstructive sleep apnea (OSA), and interstitial lung disease. Patients with primary pulmonary arterial hypertension were excluded. In the primary analysis, missing echocardiographic variables (E/e′ or PASP) were imputed as normal values; these patients were excluded in the sensitivity analysis. The score's diagnostic performance was evaluated using a standard cutoff of 6, with assessment of sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). A logistic regression model was employed to test for interaction between the H₂FPEF score and lung disease status.
Results
Patients with lung disease were older (64.9 vs. 60.2 years, p=0.009), had a higher BMI (32.2 vs. 28.3 kg/m², p<0.001), and a higher prevalence of hypertension, atrial fibrillation, and HFpEF. Prevalence of HFpEF in lung disease group was 44.3%, and 27.4% in non-lung disease group (p=0.007). When excluding patients with OSA only from the lung disease group, prevalence of HFpEF continued to be significantly higher in the lung disease group (39.4% vs 27.4%). The median H2FPEF score was significantly higher in the lung disease group (5.0 vs. 3.0, p<0.001). Using a conventional H2FPEF score cutoff of ≥6, specificity was significantly lower in patients with lung disease compared to those without (0.74 vs. 0.90, p=0.013), while sensitivity did not differ significantly (0.53 vs. 0.38, p=0.268). The interaction term between the H2FPEF score and lung disease status in the logistic regression model was not statistically significant (p=0.09), though it was borderline.
Conclusion
Despite reduced specificity in patients with lung disease, the H2FPEF score retains high sensitivity and good overall diagnostic performance, supporting its use as a screening tool. Given the high prevalence of HFpEF and overlapping symptoms in this population, clinicians should routinely consider the diagnosis of HFpEF in dyspneic patients with lung disease. Use of the H₂FPEF score may facilitate earlier diagnosis and timely management.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.