# Diagnostic stewardship cutoffs for urinalysis results prior to performing a urine culture: analysis of data from a healthcare system

**Authors:** Deborah Kahler Kupferwasser, Amy Y. Kang, Michael Bolaris, Holly Huse, Liz Chen, Loren Miller

PMC · DOI: 10.1017/ice.2025.10265 · Infection Control and Hospital Epidemiology · 2025-09-16

## TL;DR

This study finds that white blood cell count in urinalysis is the best predictor of true urinary tract infections, helping reduce unnecessary urine cultures.

## Contribution

The study identifies optimal urinalysis cutoffs for diagnostic stewardship to reduce overdiagnosis of urinary tract infections.

## Key findings

- White blood cells (WBCs) had the highest diagnostic accuracy (AUC = 0.722) for predicting uropathogens.
- WBCs performed consistently well across inpatient, emergency department, and outpatient settings.
- Combining WBCs with bacteria in urinalysis did not improve diagnostic accuracy over WBCs alone.

## Abstract

Urinary tract infections are commonly overdiagnosed. To minimize overdiagnosis, some laboratories utilize reflex algorithms that use urinalyses as preliminary screening before potentially proceeding to urine culture. However, the optimal urinalysis cutoffs for this diagnostic stewardship intervention remain poorly defined.

We performed a retrospective, cross-sectional analysis from 2/1/21–1/31/23 in the Los Angeles County Department of Health Services healthcare system. We examined patient encounters in which urinalysis was ordered synchronously with urine culture. We categorized urine culture isolates as uropathogens or non-uropathogens. We calculated receiver operating characteristic curves of urinalysis parameters’ ability, singularly or in combination, to identify uropathogens.

Among 80,949 paired urinalysis and urine cultures (17,488 inpatient, 20,716 emergency department, 42,745 outpatient), cultures yielded 35% (n = 28,993) uropathogens, 4% (n = 2960) non-uropathogens, 37% (n = 29,951) contaminants, and 24% (n = 19,045) no growth. Among urinalysis parameters, white blood cells (WBCs) had the highest diagnostic accuracy (area under the curve (AUC)=0.722, [95% CI 0.718–0.725]), followed by leukocyte esterase (AUC = 0.700, [95% CI 0.690–0.701]), bacteria (AUC = 0.673, [95% CI 0.670–0.677]), nitrite (AUC = 0.627, [95% CI 0.625–0.630]), and squamous epithelial cells (AUC = 0.530, [95% CI 0.526–0.534]). WBC AUC values were consistent across healthcare settings (outpatient, emergency department, and inpatient). The urinalysis parameter combination with the highest AUC, WBC plus bacteria, performed worse than WBCs alone (AUC = 0.711 vs. AUC = 0.722, p = 0.001).

WBC on microscopic urinalysis demonstrated the highest diagnostic accuracy for predicting uropathogens in urine cultures. Stewardship programs should consider prioritizing urinalysis WBC count as the screening tool to optimize urine culture utilization.

## Full-text entities

- **Diseases:** Urinary tract infections (MESH:D014552)
- **Chemicals:** nitrite (MESH:D009573)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12620065/full.md

## Figures

16 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12620065/full.md

## References

38 references — full list in the complete paper: https://tomesphere.com/paper/PMC12620065/full.md

---
Source: https://tomesphere.com/paper/PMC12620065