The 2005 ATS/IDSA guidelines proposed the term healthcare-associated pneumonia (HCAP) and suggested broad-spectrum antibiotic therapy for these patients. However, no single model including the HCAP criteria has yet demonstrated sufficient predictive value for the presence of drug-resistant pathogens (DRPs). Recognition of patients with DRPs is crucial for treatment of pneumonia. The aim of this study was to validate a previously developed scoring model for determining the risk for DRPs and to compare it with currently published 8 clinical prediction models including HCAP criteria.
We conducted a retrospective study of adult patients with culture-positive pneumonia at a tertiary academic medical center from January 1, 2013 to December 31, 2014. The scoring model consists of: nonambulatory status; previous episode of DRPs; and immunosuppression. Each risk factor had 1 point, resulting in a maximum of 3 possible points. We evaluated the screening value of the scores by determining their areas under the receiver-operating characteristic (AUROC) curves for predicting DPRs.
In total, 312 patients were analyzed and DRPs were isolated in 33.0%. The most common organisms included Pseudomonas aeruginosa (20.2%), Hemophilus influenza (15.7%), and Streptococcus pneumoniae (13.5%). The risk score was higher in those with DRPs than those without DRPs (mean±SD, 0.92±0.68 vs. 0.27±0.51; P<0.001). The AUROC curves were 0.78 (95% confidence interval [CI], 0.71–0.84) for the risk score and 0.66 (95% CI, 0.60–0.72) for HCAP criteria. A score of 0 had a high negative predictive value (85.9%).
The new scoring model had better AUROC curve than the current HCAP criteria. This model had a high negative predictive value and could prevent unnecessary use of broad-spectrum antibiotics.
G. Devendra, None
See more of: Poster Abstract Session