Can resident-performed POCUS and clinical predictors effectively risk-stratify emergency department patients with...

Universal reliance on non-contrast computed tomography (NCCT KUB) for suspected renal colic contributes to emergency department (ED) overcrowding and significant radiation exposure. This study evaluates the diagnostic performance of resident-performed point-of-care ultrasound (POCUS) alongside clinical and laboratory markers to establish an early multi-modal risk-stratification framework for managing patients with confirmed urolithiasis.

We conducted a retrospective observational cohort study of 313 consecutive patients with CT-confirmed urolithiasis at a tertiary-care ED over six months. We assessed the diagnostic performance of resident-performed POCUS for hydronephrosis detection against CT as the reference standard, and used multivariable logistic regression to identify independent clinical predictors of CT-confirmed hydronephrosis.

Documented resident-performed POCUS achieved a high sensitivity of 91.1% for identifying hydronephrosis against CT imaging, but displayed a critically low specificity of 21.4%. Microscopic hematuria operated at an extremely high baseline prevalence of 84.4% (n = 255 of 302 patients with urinalysis) across the cohort. Multivariable logistic regression identified three independent predictors of CT-confirmed hydronephrosis: microscopic hematuria (OR = 3.91, 95% CI [1.85, 8.23], p <.001), nausea or vomiting (OR = 3.29, 95% CI [1.27, 8.56], p =.015), and female sex (OR = 0.34, 95% CI [0.17, 0.68], p =.002). Resident-performed POCUS achieved a PPV of 89.3% alongside its high sensitivity, reinforcing its utility as a rule-in adjunct. A secondary bivariate analysis linked elevated serum creatinine (> 1.4 mg/dL) to acute inpatient resource utilization (p <.001).

While resident-performed POCUS is a sensitive but poorly specific tool for detecting hydronephrosis, it cannot safely guide disposition in isolation. Its clinical value lies in integration: when combined with key clinical red flags and secondary biochemical markers, it enables a structured, multi-modal approach to risk stratification that is practical and radiation-sparing in a high-volume emergency setting. Prospective multicenter trials are needed to validate this framework in pre-imaging triage settings.