Treasure Bright, PharmD,
UF Health Jacksonville
Jacksonville, Florida, United States
Disclosure information not submitted.
Michael Erdman, PharmD
University of Florida
Indianapolis, Indiana
Disclosure information not submitted.
Christopher Jankowski, PharmD, BCPS
Clinical Pharmacist – Infectious Diseases
UF Health Jacksonville, United States
Disclosure information not submitted.
Kateline McGory, PharmD, BCPS
Surgery Pharmacist
UF Health Jacksonville, United States
Disclosure information not submitted.
Amanda Pardon, M.D.
Neurocritical Care Intensivist
UF Health Jacksonville, United States
Disclosure information not submitted.
Yvette McCarter, PhD, D(ABMM)
Professor & Director, Microbiology Laboratory Department
UF Health Jacksonville, United States
Disclosure information not submitted.
Title: Impact of Urinalysis Reflex to Culture on Clinical Workup and Management in the Neuroscience ICU
Introduction: Patients admitted to a Neuro ICU (NSICU) with decreased mentation are at high risk for misdiagnosis of a urinary tract infection. In an effort to improve antimicrobial stewardship practices, our institution implemented a reflex to culture (RTC) process in which urine cultures (UCs) will only be performed when preceded by a positive urinalysis (UA). The purpose of this study is to evaluate the rates of urine culture ordering and antimicrobial utilization following the implementation of urine RTC in patients admitted to our NSICU.
Methods: This was a retrospective, observational cohort study conducted at UF Health Jacksonville. Patients admitted to the NSICU with a UA ordered within 48 hours were included and grouped based on whether the UA was ordered before (pre-RTC) or after (post-RTC) the protocol implementation. The primary endpoint was the rate of UCs performed before and after the implementation. Secondary endpoints included the frequency of antibiotic prescribing, hospital length of stay (LOS), NSICU LOS, and total number of UAs ordered. A logistic regression analysis was performed to identify predictors of UC ordering.
Results: A total of 600 patients were screened for inclusion with 273 fulfilling inclusion (pre-RTC = 138 and post-RTC = 135).The number of UCs performed decreased significantly after RTC implementation (23% vs. 67%, p < 0.001). The median number of UAs ordered per admission also decreased (1 [0-1] vs. 0 [0-1], p < 0.001). Median hospital LOS was similar (8 vs. 7 days, p = 0.924), while NSICU LOS decreased (3 vs. 2 days, p = 0.010). Antibiotic utilization was not significantly different between groups (5.8% vs. 8.1%, p = 0.484). A logistic regression analysis found that patients were 10 times less likely to be cultured in the post-RTC group (OR 0.104, p < 0.001), urinary catheterization raised the likelihood of UC by 4 times (OR 4.279, p < 0.001), and increasing serum creatinine raised the likelihood by 1.5 times for every increase by 1 mg/dL (OR 1.550, p = 0.014).
Conclusion: The implementation of a urine RTC process was effective in decreasing the number of UCs performed in patients with non-specific symptoms, and was associated with a decreased LOS in the NSICU. Larger studies may be able to evaluate the impact of reflex to urine culturing on antimicrobial prescribing.