886. Pneumococcal urinary antigen testing in US hospitals: underutilized and rarely acted upon.
Session: Oral Abstract Session: Respiratory Infection Diagnosis
Thursday, October 5, 2017: 2:15 PM
Room: 01AB
Background: The IDSA guideline for CAP recommends Pneumococcal urinary antigen testing (UAT) in addition to blood and sputum cultures for patients with severe CAP. In controlled settings, UAT is 50-80% sensitive and > 90% specific; however, its utility and performance on a large-scale in real world use has not been assessed. It is unclear whether UAT is clinically useful or whether the results impact prescribing behavior.

Methods: Retrospective cohort study of adult patients admitted with CAP or HCAP from 2010- 2015 at 170 US hospitals that submit data to Premier. Date and time stamped administrative and microbiologic data was assessed. Patients with a principal diagnosis of pneumonia, or sepsis with a secondary diagnosis of pneumonia  plus a CXR and antibiotics within the first 24 hours were included if they had a UAT plus either a blood or respiratory culture within the first 48 hours of admission.

Results: Of 159,894 eligible pneumonia patients, 24,757 (15.5%) had UAT plus either blood or respiratory cultures performed. Of 1797 (7%) who had a positive UAT, 457 (25%) also grew S. pneumoniae (SP) from blood or respiratory cultures, 1240 (69%) had negative cultures and 100 (6%) an organism other than SP, with S. aureus, Pseudomonas spp, and E.coli being the most common pathogens, predominantly from respiratory cultures. Among 22,960 patients with a negative UAT, 429 (2%) had a positive blood or respiratory culture for SP and 2653(12%) had a culture positive for another organism. UAT was performed among 18.4% of patients admitted to the ICU, and 15.3% of those admitted to wards. Among patients empirically started on broad spectrum antibiotics, 35% who had a positive UAT were de-escalated by day 5, compared to 49% who grew SP in blood cultures and 24% in respiratory cultures.

Conclusion: In a large representative US inpatient database, there was poor concordance between UAT and cultures for SP. A positive UAT decreased the probability of having a non-SP pathogen. Antibiotic de-escalation occurred more often in association with a positive blood culture for SP than for UAT or positive respiratory culture, but occurred in less than half the patients with these markers of pneumococcal pneumonia. Overall, UAT is underutilized and does not appear to have a substantial impact on clinical care.

Sarah Haessler, MD1, Jennifer Schimmel, MD1, Pei-Chun Yu, MS2 and Michael Rothberg, MD, MPH3, (1)Infectious Diseases, Baystate Medical Center, Springfield, MA, (2)Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, (3)Medicine Institute Center for Value-Based Care Research, Cleveland Clinic, Cleveland, OH


S. Haessler, AHRQ: Investigator , Research grant

J. Schimmel, None

P. C. Yu, AHRQ: Investigator , Research grant

M. Rothberg, AHRQ: Investigator , Research grant

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 4th with the exception of research findings presented at the IDWeek press conferences.