Predictors and outcomes of pneumonia in patients with spontaneous intracerebral hemorrhage

• Published in: Journal of intensive care medicine Vol. 28; no. 2; p. 118
• Main Authors: Alsumrain, Mohammad, Melillo, Nicholas, Debari, Vincent A, Kirmani, Jawad, Moussavi, Mohammad, Doraiswamy, Vikram, Katapally, Ram, Korya, Daniel, Adelman, Marc, Miller, Richard
• Format: Journal Article
• Language: English
• Published: United States 01.03.2013
• Subjects: Blood Glucose - physiology; Cerebral Hemorrhage - complications; Cerebral Hemorrhage - therapy; Confounding Factors (Epidemiology); Glasgow Coma Scale; Humans; Pneumonia - complications; Pneumonia - therapy; Predictive Value of Tests; Reproducibility of Results; Treatment Outcome
• ISSN: 1525-1489
• DOI: 10.1177/0885066612437512

Pneumonia is an important complication of spontaneous intracerebral hemorrhage (sICH). The purpose of this study was to determine the predictors and outcomes of the development of pneumonia in patients with sICH. In total, 290 consecutive patients with sICH admitted within 24 hours of stroke onset were investigated in a single center, retrospective study from January 2006 to July 2009. Clinical, biochemical, and imaging variables were registered. Stroke severity and functional outcomes were evaluated with the Glasgow Coma Scale (GCS) and modified Rankin Scale (mRS). Baseline variables that predicted pneumonia were investigated and outcomes were assessed. The association of our primary exposure variables, such as mechanical ventilation, tube feeding, dysphagia, and tracheostomy, with pneumonia was highly significant (P < .0001, for each variable). For mechanical ventilation, we observed an odds ratio (OR; 95% confidence interval [CI]) of 9.42 (4.24-20.9); for tube feeding, OR = 22.3 (8.91-55.8); for dysphagia, OR = 13.1 (4.66-36.7); and for tracheostomy, OR = 26.8 (8.02-89.3). After adjustment of potential confounders including GCS and mRS on admission, the use of angiotensin-converting enzyme inhibitors, proton pump inhibitors, and H2 blockers, all the adjusted OR (ORa) remained significant. For mechanical ventilation, the minimum ORa was 3.72 (95% CI: 1.68-8.26) when adjusted for GCS. For both dysphagia and tracheostomy, mRS reduced OR to 7.46 (95% CI: 3.34-10.6) in the case of dysphagia with an ORa of 16.2 (95% CI: 4.98 to 52.8) for tracheostomy. For tube feeding, both GCS and mRS reduced ORa; the former to 14.7 (95% CI: 6.16-35.0) and the latter to 15.7 (95% CI: 6.63-37.0). Pneumonia shows a significant effect on the morbidity (P = .003), length of stay (P < .0001), and mortality (P = .041) rate of the patients. Mechanical ventilation, tube feeding, dysphagia, and tracheostomy are exposures associated with increased risk of the development of pneumonia in patients with sICH. Pneumonia is associated with an increase in morbidity, length of stay, and mortality among patients with sICH.