404. Implementation of an Inpatient Urine Culture Algorithm Decreased Catheter-Associated Urinary Tract Infections
Session: Poster Abstract Session: HAI: Preventing Device-Associated Infections
Thursday, October 27, 2016
Room: Poster Hall
Posters
  • UCX Change Poster Final.pdf (686.2 kB)
  • Background: Urine cultures (UC) are often obtained when symptoms of urinary tract infection (UTI) are absent. Indiscriminate UC can lead to misidentification of catheter-associated UTI (CAUTI). Pyuria (>10 WBC/hpf) has an excellent negative predictive value for UTI in immunologically normal patients.

    Methods: In April 2015 the ability to order an inpatient UC was removed at our 650-bed academic medical center. UC was available only via a UTI Evaluation Panel requiring documentation of symptoms and special criteria supporting UC in the absence of pyuria (neutropenia, kidney/pancreas transplant, pregnant, impending urologic surgery, age <3 years, other). A UC was always performed if special criteria were met. Asymptomatic patients not meeting special criteria were not cultured. Symptomatic patients not meeting special criteria had UC reflexively performed based on an algorithm: >100 squamous cell = no UC; <100 squamous cells and ≤10 WBC/hpf = no UC; <100 squamous cells and >10 WBC/hpf = UC. NHSN CAUTI definitions were used which changed in 2015 (excluded candida and cultures with <100,000 CFU/ml). 2014 CAUTI rates were recalculated using 2015 definitions (A-CAUTI). Institutional catheter utilization (CD/PD) and CAUTI rates using catheter days (CD) and patient days (PD) before (1/14-3/15) and after (4/15-3/16) were compared using Poisson regression. UC and contaminated UC (UC with ≥3 isolates; C-UC) were compared one year before (4/14-3/15) and after (4/15-3/16).

    Results: Catheter utilization decreased significantly (0.27 CD/PD vs. 0.20 CD/PD, P<0.0001) as did rates of UC, C-UC, CAUTI and A-CAUTI (Table). CAUTI rates measured per patient day showed a greater decrease those measured by catheter day.

    Measurement

    Before

    After

    P

    Model Estimated Risk (95% CI)

    CAUTI/ 1000 CD

    3.29

    1.8

    0.0004

    0.55 (0.39, 0.76)

    A-CAUTI/ 1000 CD

    2.58

    1.8

    0.039

    0.70 (0.49, 0.98)

    CAUTI/ 1000 PD

    0.85

    0.36

    <0.0001

    0.42 (0.30, 0.59)

    A-CAUTI/ 1000 PD

    0.67

    0.36

    0.0005

    0.54 (0.38, 0.76)

    UC/ 1000 PD

    35.9

    19.3

    <0.0001

    0.54 (0.48, 0.60)

    C-UC/ 1000 PD

    3.3

    1.8

    <0.0001

    0.56 (0.46, 0.68)

    Conclusion: Implementation of a urine culture algorithm using symptoms and pyuria resulted in a significant decrease in UC, contaminated UC, and CAUTI.

    Trevor C. Van Schooneveld, MD1, Nicole Turille, RN, BSN2, Elizabeth Lyden, MS3, Ryan Clevenger, BS4, Michelle Schwedhelm, MSN, RN4 and Mark E. Rupp, MD, FIDSA, FSHEA1, (1)Division of Infectious Diseases, University of Nebraska Medical Center, Omaha, NE, (2)Quality and Patient Safety, Nebraska Medicine, Omaha, NE, (3)Epidemiology, University of Nebraska Medical Center, Omaha, NE, (4)Infection Control & Epidemiology, Nebraska Medicine, Omaha, NE

    Disclosures:

    T. C. Van Schooneveld, None

    N. Turille, None

    E. Lyden, None

    R. Clevenger, None

    M. Schwedhelm, None

    M. E. Rupp, None

    Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.