Key result
Haemoglobin shows a U-shaped association with HF outcomes, with optimal risk ~2 g/dL above WHO thresholds.
Why the study?
World Health Organization anaemia thresholds (haemoglobin <12 g/dL in women and <13 g/dL in men) are derived from healthy population distributions rather than clinical outcomes, leaving it uncertain whether they appropriately stratify prognostic risk in patients with heart failure.
What sex-specific haemoglobin concentrations are associated with the lowest risk of adverse clinical outcomes in patients with HFrEF and HFpEF compared to current WHO anaemia thresholds?
Observational (n=45,230)
Yes
What sex-specific haemoglobin concentrations are associated with the lowest risk of adverse clinical outcomes in patients with HFrEF and HFpEF compared to current WHO anaemia thresholds?
In patients with HFrEF and HFpEF, the haemoglobin concentrations associated with the lowest risk of adverse clinical outcomes (~14 g/dL in women and ~15 g/dL in men) are approximately 2 g/dL higher than current WHO anaemia thresholds, suggesting standard anaemia definitions may underestimate prognostic risk in heart failure.
Challenges WHO anemia thresholds in HF; extends pooled trial data supporting higher sex-specific Hb nadirs for lowest risk across HFrEF/HFpEF.
Background The World Health Organization (WHO) defines anaemia as haemoglobin (Hb) <12 g/dL in women and <13 g/dL in men. These thresholds are widely used in clinical practice and research; however, they are based on the lower fifth percentile of Hb distribution in young healthy populations and were not derived from associations with clinical outcomes. Accordingly, whether these definitions appropriately stratify prognostic risk in heart failure (HF) - a condition defined by reduced tissue oxygen delivery - remains uncertain. We have re-examined the association between Hb concentration and risk of adverse outcomes in men and women with HFrEF and HFpEF (across the full range of Hb). Purpose To examine the association between Hb concentration and clinical outcomes in patients with HF and to identify sex-specific Hb values associated with the nadir in risk across HF phenotypes (HFrEF and HFpEF). Methods Patient-level data from 5 HFrEF trials and 5 HFpEF trials (EMPHASIS-HF, PARADIGM-HF, DAPA-HF, GALACTIC-HF, CHARM, I-Preserve, PARAGON-HF, TOPCAT-America, FINEARTS-HF) were pooled. Associations between Hb concentration and all-cause death, cardiovascular (CV) death, first HF hospitalization, and the composite of HF hospitalization or CV death were evaluated using Cox proportional hazards models. Non-linear relationships were assessed using restricted cubic spline analyses. Models were stratified by trial and adjusted for age, heart rate, NYHA class III/IV, body mass index, left ventricular ejection fraction (LVEF), serum creatinine, log-transformed NT-proBNP, history of atrial fibrillation, prior myocardial infarction, diabetes, prior stroke, previous HF hospitalization, race, diuretic use, and treatment assignment. Nadir Hb values were identified from spline curves for each outcome. Results Among 31826 patients with HFrEF, 7070 (22.2%) were women (age 65.9±11.5 years, LVEF 29.0±6.4%, Hb 12.9±1.5 g/dL) and 24,756 (77.8%) were men (age 64.3±11.2 years, LVEF 28.2±6.4%, Hb 14.0±1.6 g/dL). Among 13,404 patients with HFpEF, 7,453 (55.6%) were women (age 73.0±8.4 years, LVEF 60.3±7.4%, Hb 13.0±1.5 g/dL) and 5,951 (44.4%) were men (age 71.5±9.0 years, LVEF 58.1±6.7%, Hb 13.8±1.8 g/dL). Hb concentration demonstrated a reverse J-shaped or U-shaped association with all outcomes (Figure). In the adjusted model for the overall HF population, the nadir Hb concentration across all outcomes was approximately 14 g/dL in women and 15 g/dL in men. Similar patterns were observed in both HFrEF and HFpEF (Table). Conclusions In patients with HF, the Hb concentration associated with the lowest risk of adverse outcomes is approximately 2 g/dL higher than the current WHO anaemia thresholds. Importantly, higher risk was observed even among patients with Hb concentrations up to 2 g/dL above WHO-defined anaemia. These findings suggest that optimal Hb thresholds may warrant reconsideration in patients with HF.
No takes yet. Share an insight, caveat, or question.
Chimura et al. (2026) conducted an observational in Heart failure (HFrEF and HFpEF) (n=45,230). Haemoglobin (Hb) concentration vs. WHO anaemia thresholds (<12 g/dL in women, <13 g/dL in men) was evaluated on All-cause death, cardiovascular (CV) death, first HF hospitalization, and the composite of HF hospitalization or CV death. Haemoglobin concentration showed a U-shaped association with adverse HF outcomes, with lowest risk at ~14 g/dL in women and ~15 g/dL in men (~2 g/dL above WHO anaemia thresholds).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: