Validation of a clinical score for assessing the risk of resistant pathogens in patients with pneumonia presenting to the emergency department

(See the Editorial Commentary by Murri and De Pascale, on pages 199-201.)Background. Resistant organisms (ROs) are increasingly implicated in pneumonia in patients presenting to the emergency department (ED). The concept of healthcare-associated pneumonia (HCAP) exists to help identify patients infected with ROs but may be overly broad. We sought to validate a previously developed score for determining the risk for an RO and to compare it with the HCAP definition.Methods.We evaluated adult patients admitted via the ED with bacterial pneumonia (January-December 2010). We defined methicillin-resistant Staphylococcus aureus (MRSA), Pseudomonas aeruginosa, and extended-spectrum β-lactamases as ROs. The risk score was as follows: 4, recent hospitalization; 3, nursing home; 2, chronic hemodialysis; 1, critically ill. We evaluated the screening value of the score and of HCAP by determining their areas under the receiver-operating characteristic (AUROC) curves for predicting ROs. Results. The cohort included 977 patients, and ROs were isolated in 46.7%. The most common organisms included MRSA (22.7%), P. aeruginosa (19.1%), and Streptococcus pneumoniae (19.1%). The risk score was higher in those with an RO (median score, 4 vs 1; P <. 001). The AUROC for HCAP equaled 0.62 (95% confidence interval [CI],. 58-.65) versus 0.71 (95% CI,. 66-.73) for the risk score. As a screening test for ROs, a score > 0 had a high negative predictive value (84.5%) and could lead to fewer patients unnecessarily receiving broad-spectrum antibiotics.Conclusions.ROs are common in patients presenting to the ED with pneumonia. A simple clinical risk score performs moderately well at classifying patients regarding their risk for an RO.