Sex differences in cardiovascular inflammation defined by high-sensitivity C-reactive protein across heart failure phenotypes: insights from POSEIDON
S Stoerk, W P Abhayaratna, K A Connelly, R Darwesh, E T Kato, S L Kristensen, J Mathisen, P L Myhre, F Roubille, M PiepoliAbstract
Background
Inflammation is common in heart failure (HF), but potential sex differences across HF phenotypes remain uncertain. High-sensitivity C-reactive protein (hsCRP) offers a pragmatic measure of inflammatory burden. Major trials (e.g. CANTOS) have linked a threshold of hsCRP ≥2 mg/L to a persistent inflammatory signal in a cardiovascular (CV) population.
Purpose
To evaluate sex differences in the prevalence of hsCRP-defined CV inflammation (hsCRP ≥2 mg/L) in HF overall and by HF phenotype (HF with preserved or mildly reduced ejection fraction [HFpEF/HFmrEF] vs HF with reduced ejection fraction [HFrEF]), and to describe sex-specific clinical profiles among individuals with hsCRP-defined CV inflammation.
Methods
POSEIDON was a real-world, cross-sectional study (2023–2025) enrolling adults with HF (current or prior New York Heart Association Functional Classification [NYHA] class II–IV) in routine care; those with recent infection, hospitalisation or unplanned visits (≤60 days from date of informed consent) were excluded. HF phenotypes were defined by left ventricular ejection fraction (HFrEF ≤40%; HFmrEF 41–49%; HFpEF ≥50%); HFpEF and HFmrEF were analysed together. We compared inflammation prevalence between women and men overall and within HF phenotypes, and summarised sex differences among participants with inflammation descriptively.
Results
Of 11,885 individuals with HF, recruited in 18 countries on 5 continents, 3519 were women (30%) and 8366 were men (70%). Overall, elevated hsCRP (≥2 mg/L) was present in approximately 42% of women and 37% of men (among those with available hsCRP) and was consistent across HF phenotypes (HFpEF/HFmrEF: 42% vs 37%; HFrEF: 41% vs 37%) (Figure). Use of guideline-directed medical therapy for HF was generally consistent with best clinical practice, with no major differences between sexes or hsCRP categories, with the exception of loop diuretics, which were used more often by women, and by individuals with hsCRP-defined CV inflammation. Among individuals with elevated hsCRP, men more often than women were current/former smokers and more often had type 2 diabetes, in both HF phenotypes. Generally, women were more likely to be in higher NYHA classes and exhibited consistently higher N-terminal pro-B-type natriuretic peptide levels compared with men, both overall and in hsCRP-defined categories, and across HF phenotypes. hsCRP-defined CV inflammation yielded larger subgroup contrasts by sex and HF phenotype than interleukin-6, and women showed greater hsCRP-defined CV inflammation prevalence across HF phenotypes (Table).
Conclusion
In this large, global HF cross-sectional study, hsCRP-defined CV inflammation affected around 4 in 10 patients and was modestly more common in women than men across HF phenotypes. These descriptive findings suggest that the ‘inflammatory HF phenotype’ may not be identical in women and men and may warrant sex-stratified investigation.