Fig 1.
The flow chart of sample selection.
The discovery cohort included 4,097 outpatients with suspected UTIs prospectively enrolled from January to October 2023. 1,335 (32.6%) outpatients were defined as UTI cases, with a urine culture ≥10⁵ CFU/μL. Cases were matched 1:1 by age and sex with 1,282 non-UTI controls, resulting in a final discovery cohort. The validation cohort was similarly constructed from retrospectively collected data from 7/2022 to 12/2022.
Table 1.
Demographic characteristics and urinalysis results of urinary tract infection cases and controls in discovery cohort.
Table 2.
Demographic characteristics and urinalysis results of urinary tract infection cases and controls in validation cohort.
Table 3.
Flow cytometry features of urinary tract infection cases and controls in discovery cohort.
Fig 2.
UTIRisk score of UTI cases and non-UTI controls in discovery and validation cohorts.
In the discovery cohort, UTI cases had a significantly higher median UTIRisk score than non-UTI controls (11.2 vs. 5.6, p < 0.001, 2A). Similar results were observed in the validation cohort (9.8 vs. 5.6, p < 0.001, 2B).
Table 4.
The components of our novel UTIRisk score.
Fig 3.
The UTI discrimination performance of our novel UTIRisk score and other parameters of flow cytometry and urinalysis in discovery and validation cohorts.
The UTIRisk score demonstrated strong discriminatory ability, comparable to bacterial counts from flow cytometry, and outperformed commonly used parameters, including white blood cell WBC count from flow cytometry, nitrite, and leukocyte positivity from urinalysis, in both the discovery (3A) and validation cohorts (3B).
Table 5.
UTIRisk score of urinary tract infection cases and controls in discovery and validation cohorts.
Fig 4.
Discrimination performance of UTIRisk score across age and sex subgroups.
The UTIRisk score showed consistently strong diagnostic ability across age and sex subgroups (AUC ≥ 0.80), with excellent performance in patients aged ≥65 (4A) and males (4B).