Haematocrit and in-hospital mortality in acute decompensated heart failure at 2,600 m above sea level: a retrospective analysis of a prospective registry from Bogotá, Colombia

Conventional risk stratification for acute decompensated heart failure (ADHF) has largely been developed in populations living at or near sea level. At high altitude, chronic hypobaric hypoxia increases expected haematocrit values, and sea-level thresholds may not adequately identify patients with insufficient oxygen-carrying capacity. We evaluated predictors of in-hospital mortality in patients hospitalised with ADHF at 2,600 m above sea level. Retrospective analysis of a single-centre prospective registry (CARDIOMED) of consecutive adults hospitalised for ADHF at Hospital Universitario Mayor-Méderi, Bogotá (2,600 m a.s.l.). Three pre-specified Firth penalised logistic regression models were fitted: (A) primary: age, haematocrit, and chronic obstructive pulmonary disease (COPD) ( n = 882; events per variable [EPV] = 16.7); (B) secondary: additionally including systolic blood pressure and creatinine ( n = 876; EPV = 10.0); (C) exploratory echocardiographic subset ( n = 276; EPV = 4.2). Altitude-adjusted haematocrit ranges (males 44–54%, females 38–47%) were applied. Of 886 patients (median age 79 years; 44.4% female), 50 died in hospital (5.6%). In Model A, haematocrit (aOR 0.95, 95% CI 0.92–0.99, p = 0.008), age (aOR 1.04, 95% CI 1.01–1.08, p = 0.006), and COPD (aOR 2.25, 95% CI 1.17–4.17, p = 0.017) were independently associated with mortality. Results were consistent in Model B. A descriptive nine-fold difference in mortality was observed between haematocrit quintiles Q1 and Q3 (10.1% vs. 1.1%). The exploratory altitude anaemia gap was observed in 20.0% of female deaths versus 9.4% of female survivors, whereas no mortality excess was observed among men. In this retrospective analysis of a prospective ADHF registry at 2,600 m a.s.l., lower haematocrit, older age, and COPD were independently associated with in-hospital mortality. The non-linear haematocrit–mortality pattern and the exploratory female-specific altitude anaemia gap suggest that sea-level haematological thresholds may not fully capture risk at high altitude. Prospective multicentre validation is required.