2233. Clinical Risk Scores (CRSs) to Predict Resistance to Trimethoprim-Sulfamethoxazole (TMP-SMX), Fluoroquinolone (FQ), Nitrofurantoin (NIT), or Third Generation Cephalosporins (3GC) among Adult Outpatients (OPs) with Complicated Urinary Tract Infections (cUTIs). (15th December 2022)

• Record Type: Journal Article
• Authors: Lodise, Thomas P; Chen, Lie Hong H; Bruxvoort, Katia J; Wei, Rong; Im, Theresa M; Contreras, Richard; Rodriguez, Mauricio; Friedrich, Larry; Reese, Jennifer; Tartof, Sara Y
• Is Part Of: Open forum infectious diseases. Volume 9:(2022)Supplement 2
• Publication Date: 2022-12-15
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofac492.1851
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Increased resistance rates to available oral antibiotics (ABs) contribute to delays in receipt of appropriate treatment and adverse outcomes among patients with cUTI in the OP setting. To optimize empiric AB selection in adult OPs with cUTIs, we developed CRSs using information available at presentation to estimate the risk of having a cUTI that was resistant to TMP-SMX, FQ, NIT, or 3GC. Methods: Retrospective cohort study among Kaiser Permanente Southern California members from 2017-2020. Inclusion criteria: age ≥18 yrs; cUTI diagnosis during an OP visit; positive urine culture with antibiotic susceptibility results; receipt of antibiotic ±3 days of index urine culture, and not hospitalized on day of OP visit. Resistance to TMP-SMX, FQ, NIT, and 3GC on index urine culture was quantified. OPs were randomly split (60:40) into training and validation datasets. Covariates present on clinical presentation were collected. Least absolute shrinkage and selection operator logistic regression (LR) were used to develop separate models to estimate the likelihood of resistance to TMP-SMX, FQ, NIT, and 3GC. The prediction models were developed using training and validation datasets. For all 4 LR models, CRSs were calculated as the weighted sums of regression coefficients. Variables in each of the final LR models were assigned point(s), and an OP's CRS for cUTI resistant to TMP-SMX, FQ, NIT, or 3GC was based on total points in each of the respective models. Results: A total of 30, 450 cUTIs among 26, 326 OPs met study criteria. Mean age was 61, 54% were female, and E. coli (66%) was the most common pathogen. Resistance to TMP-SMX, FQ, NIT, and 3GC was 37%, 27%, 24%, and 19%, respectively. Baseline covariates and associated points for the 4 LR models are shown in Table 1 . A CRS of 0, 5, 8, 12 corresponded to a >20% risk of resistance to TMP-SMX, FQ, NIT, or 3GC, respectively (Figure 1 ). Table 1. Baseline Clinical Covariates and Associated Point Values in Each of the 4 Clinical Risk Scores Covariates in each of the Final Resistance Figure 1. Clinical Risk Scores for Resistance to TMP-SMX, FQ, NIT, and 3GC Conclusion: We developed a high-performing parsimonious CRS to aid clinicians in appropriate treatment selection of adult OPs with cUTI. This tool can be used to facilitate empiric antibiotic selection and ensure adult OPs with cUTI have a greater probability of receiving early appropriate therapy. Given its high baseline resistance, TMP/SMX should not be considered for empiric therapy. Disclosures: Thomas P. Lodise, PharmD, PhD, Spero Therapeutics: Advisor/Consultant Lie Hong H. Chen, DrPH, MSPH, Spero Therapeutics: Grant/Research Support Katia J. Bruxvoort, PhD, MPH, Dynavax: Grant/Research Support|Gilead: Grant/Research Support|Glaxosmithkline: Grant/Research Support|Moderna: Grant/Research Support|Pfizer: Grant/Research Support|Seqirus: Grant/Research Support Rong Wei, MA, Spero Therapeutics: Grant/Research Support|Spero Therapeutics: Grant/Research Support Theresa M. Im, MPH, Spero Therapeutics: Grant/Research Support Richard Contreras, MS, Spero Therapeutics: Grant/Research Support Mauricio Rodriguez, PharmD, MS-HEOR, BCPS, BCCCP, BCIDP, Spero Therapeutics: Employee Larry Friedrich, PharmD, Spero Therapeutics: Employee Jennifer Reese, PharmD, Spero Therapeutics: Employee Sara Y. Tartof, PhD MPH, Pfizer: Grant/Research Support|Spero: Grant/Research Support.