Summary
Predictors of drug-resistant pathogens in community-onset pneumonia: Are factors considered in health-care–associated pneumonia useful in the emergency department?
Affiliation of the authors
DOI
Quote
Torres Bonafonte OH, Gil Olivas E, Pérez Macho E, Pacho Pacho C, Mateo Roca M, Casademont Pou J, et al. Predictors of drug-resistant pathogens in community-onset pneumonia: Are factors considered in health-care–associated pneumonia useful in the emergency department?. Emergencias. 2017;29:306-12
Summary
Objectives.
To analyze factors related to drug-resistant pathogens (DRPs) in community-onset pneumonia (COP) and whether previously suggested criteria are useful in our emergency-department.
Methods.
Prospective 1-year study of adults coming to the emergency department for COP. We assessed the usefulness of criteria used in health-care–associated pneumonia (HCAP), as well the Shorr index, the Barthel index, and clinical suspicion of resistant pathogens. Data were analyzed by multiple logistic regression and the area under the receiver
operating characteristic curve (AUC).
Results.
We included 139 patients with a mean (SD) age of 75.9 (15.3) years; 63.3% were men. Forty-nine COP patients (35.2%) were at risk for DRP-caused pneumonia according to HCAP criteria; 43 (30.9%) according to the Shorr index, and 56 (40.3%) according to the Aliberti index. A score of less than 60 derived from the Barthel index was recorded for
25 patients (18%). Clinical suspicion of a DRP was recorded for 11 (7.9%). A DRP was isolated in 5 patients (3.6%) (3, Pseudomonas aeruginosa; 2, methicillin-resistant Staphylococcus aureus). Multiple logistic regression analysis identified 2 predictors of DRP-caused COP: hospital admission within the last 90 days (odds ratio [OR], 8.92; 95% CI, 1.92–41.45) and initial arterial blood oxygen saturation (OR, 0.85; 95% CI, 0.74–0.98). The AUC was 0.91 (95% CI, 0.85–0.98). The model identified 22 patients (16.8%) at risk for DRP-caused pneumonia. The positive and negative predictive values were 20% and 99.1%, respectively, for the model 90–day period (vs 8.7% and 98.9%, respectively, for criteria used in HCAP).
Conclusions.
Hospitalization within the 90–day period before a COP emergency and arterial blood oxygen saturation were good predictors of DRP in our setting. Criteria of DRP in HCAP, on the other hand, had lower ability to identify patients at risk in COP.
