Empiric antibiotic selection and risk prediction of drug-resistant pathogens in community-onset pneumonia. Issue 2 (April 2016)

• Record Type: Journal Article
• Authors: Webb, Brandon J.; Jones, Barbara; Dean, Nathan C.
• Is Part Of: Current opinion in infectious diseases. Volume 29:Issue 2(2016)
• Publication Date: 2016-04
• Subjects: antibiotic -- antimicrobial stewardship -- community-acquired -- drug resistance -- healthcare-associated -- pneumonia; Communicable diseases -- Periodicals; Communicable Diseases -- Periodicals; Review Literature -- Periodicals; 616.905
• DOI: 10.1097/QCO.0000000000000254
• Languages: English
• ISSNs: 0951-7375

Abstract

Abstract : Purpose of reveiw: Empiric antibiotic selection in community-onset pneumonia is complicated by uncertainty regarding risk of drug-resistant pathogens (DRPs). The healthcare-associated pneumonia (HCAP) criteria have limited predictive value and lead to unnecessary antibiotic use. Better methods of predicting risk of DRP and selecting empiric antibiotics are needed. Here we give an update on risk factors for DRP, available risk prediction models, and treatment strategy in patients with pneumonia. Recent findings: Evidence supporting factors that contribute to risk of DRP has improved since the advent of HCAP. Many of these risk factors have been reproducibly identified in heterogeneous populations. Newer methods of predicting DRP based on these factors demonstrate better performance than HCAP. Recent innovations include the potential to discriminate between risk for methicillin-resistant Staphylococcus aureus and other DRP, and use of severity as a modifier of treatment threshold. However, there is wide variation in included predictor variables, and at proposed thresholds most scores still favor overtreatment. Summary: Until reliable molecular diagnostics are available, additional development and validation of decision support models integrating local resistance rates, estimated DRP risk, severity, and threshold for anti-DRP antibiotics are needed. Once optimized models are identified, implementation studies will be needed to confirm safety and efficacy.