30 day mortality and eosinopenia in patients with pneumonia

Introduction Blood eosinophil count is an independent predictor of mortality in patients: with bacteraemia; requiring intensive care; and with acute exacerbations of COPD (AECOPD). In AECOPD eosinopenia may be related to steroid use prior to admission. We describe eosinopenia and mortality in patients with pneumonia and no chronic lung disease or maintenance steroid use. Methods 198 patients, mean (SD) age 74 (16) years, were identified and data for mortality, eosinopenia (<0.05x10^9/l) and CURB-65 score (Confusion, Urea >7 mmol/l, Respiratory rate >30/min, low systolic (<90 mm Hg) or diastolic (<60 mm Hg) Blood pressure), age >65 years) were collected. Odds ratios (OR) for death at 30 days were calculated. Results 30 day mortality Dead Total % dead OR (95% CI) Confusion score 1 38 87 43.7 4.3 (2.2-8.4) Confusion score 0 17 111 15.3 Urea score 1 47 125 37.6 4.9 (2.2-11.1) Urea score 0 8 73 11.0 Resp. Rate score 1 15 38 39.5 2.0 (0.9-4.1) Resp. Rate score 0 40 160 25.0 BP score 1 17 48 35.4 1.6 (0.8-3.3) BP score 0 38 151 25.2 Age score 1 53 150 35.3 12.6 (2.9-53.9) Age score 0 2 48 4.2 Eosinopenia 41 123 33.3 2.2 (1.1-4.3) No eosinopenia 14 75 18.7 A higher proportion of patients with eosinopenia had a raised CRP (>50mg/l) compared with those without eosinopenia (81.3% vs. 62.7% p<0.05). Discussion Blood eosinopenia <0.05x10^9/l was a strong predictor of 30 day mortality in this cohort with pneumonia, and no chronic respiratory disease. This association cannot be ascribed to steroid use. Eosinopenia may be a marker of severe sepsis and consequent mortality risk (Shaaban et al 2010; Abidi et al 2008), however the association between eosinopenia and CRP >50mg/l was weak.