Context There are no clear indicators of which ultrasound findings in an emergency department (ED) renal colic population are predictive of urological surgical intervention.

Objective To determine the sonographic findings of renal colic that predict surgical intervention.

Methods We conducted a retrospective cohort study of 500 consecutive ED patients with a diagnosis of renal colic that had ultrasonography (performed by radiology) during their ED visit. Our main outcome was urological surgical intervention. This was defined as extracorporeal shock wave lithotripsy, percutaneous nephrostomy or ureteroscopy performed within 16 weeks of the initial ED presentation.

Results Of the 500 identified patients, 483 met our eligibility criteria. Of this group, 67 (13.9%) received a surgical intervention. Ultrasound (US) findings were 97% (95% CI 88.7% to 99.5%) sensitive and 28.1% (23.9% to 32.8%) specific in ‘diagnosing’ the requirement for surgery when the ultrasound (US) showed either at least a stone present or showed moderate to severe hydronephrosis. The presence of stone and moderate to severe hydronephrosis had a + likelihood ratio (LR) 3.86 (2.46 to 6.07) and a −LR 0.72 (0.60 to 0.86). Having a stone ≥6 mm had a sensitivity of 77.6% (65.5% to 86.5%), a specificity of 73.6% (69.0% to 77.7%), a +LR of 2.94 (2.39 to 3.6) and a −LR 0.30 (0.19 to 0.48).

Conclusions Radiology performed ultrasonography is a valuable tool for identifying renal colic that will go on to receive a surgical intervention in the 16 weeks following an ED visit. Further studies are needed to confirm the utility of ultrasound findings of stone visualisation, stone size and moderate to severe hydronephrosis in determining which patients need outpatient urology follow-up.