Back

Incorporating Angina into the H2FPEF Score Improves Diagnostic Performance for HFpEF in Women

Qu, K.; Onland-Moret, N. C.; Vos, A.; van Ommen, A.-M.; Mourik, Y.; Riet, E.; Boonman-de Winter, L. J.; Handoko, M. L.; Cramer, M. J.; Teske, A.; Menken, R.; Rutten, F. H.; Den Ruijter, H. M.; Dal Canto, E.

2025-11-02 cardiovascular medicine
10.1101/2025.10.30.25339191 medRxiv
Show abstract

Background and AimsThe H2FPEF score is a widely used prediction tool used during the diagnostic work-up of heart failure with preserved ejection fraction (HFpEF). However, having angina symptoms is not included in the score, despite being common in patients with HFpEF. We hypothesize that incorporating angina in the H2FPEF score may improve its performance. Given the known sex differences in HFpEF, sex-specific analyses are warranted. MethodsWe included 1,266 individuals suspected HFpEF, with 515 from the UHFO-DM cohort and 751 from a combination cohort of STRETCH, TREE and UHFO-COPD. Participants underwent standardized symptom collection, including angina, using WHO questionnaires and expert-panel adjudication of HFpEF. Following evaluation of H2FPEF, we assessed the association of angina with HFpEF independent of H2FPEF using logistic regression. By adding angina to H2FPEF, we developed a modified algorithm and evaluated it by AUC, calibration, reclassification, and decision curve analysis. All analyses were stratified by sex. ResultsIn the UHFO-DM cohort, HFpEF prevalence was 24%. Overall H2FPEF discrimination (AUC) was 0.72, with 0.69 in women and 0.74 in men. Angina was independently associated with HFpEF in women (OR 3.96, 95% CI 1.72-9.11, P=0.001) but not in men (1.90, 0.88-4.10, 0.102). This was also found in the combination cohort (women: 2.13, 1.14-3.97, 0.018; men: 0.85, 0.44-1.66, 0.638). In the UHFO-DM cohort, adding one point for angina in a modified H2FPEF score in women improved AUC from 0.69 to 0.71 (DeLong P=0.030), increased sensitivity (0.53 to 0.60) and negative predictive value (0.80 to 0.82), and yielded a continuous net reclassification improvement of 0.449, with preserved calibration and higher net clinical benefit on decision curves. No performance gain was observed with the same modification in men. ConclusionsIn women with suspected HFpEF, presence of angina provides diagnostic information independent of H2FPEF to uncover HFpEF. A simple sex-specific modification of H2FPEF, adding one point for angina in women, may slightly improve discrimination and rule-out performance in women. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=129 SRC="FIGDIR/small/25339191v1_ufig1.gif" ALT="Figure 1"> View larger version (55K): org.highwire.dtl.DTLVardef@b9eee9org.highwire.dtl.DTLVardef@f41af6org.highwire.dtl.DTLVardef@162595org.highwire.dtl.DTLVardef@1fd8f9c_HPS_FORMAT_FIGEXP M_FIG C_FIG

Published in Open Heart (predicted rank #23) · training set

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.