US vs. CT for Kidney Stones

Dr. Adam Webb


Kidney stones are not uncommon in the ED, and with the explosion of CT scanners we often scan patients looking specifically for stones. However, this exposes a lot of patients to perhaps unnecessary radiation.

I picked this article because I had a patient who was too large for our CT scanner, and I was interested in how accurate and reliable other imaging modalities are in the diagnosis of nephrolithiasis. While this study did not directly answer my question, as obese patients were excluded, it did shed light on the role ultrasonography can play in diagnosing kidney stones in the ED.


Ultrasonography versus Computed Tomography for Suspected Nephrolithiasis

N Engl J Med 2014; 371:1100-1110

DOI: 10.1056/NEJMoa1404446

Study Design

This was a multicenter, pragmatic, comparative effectiveness trial. Patients were randomly assigned to ultrasonography performed by ED physicians, or by radiology, or abdominal CT scan. Primary endpoint was 30-day incidence of high-risk diagnoses with complications that could be related to missed or delayed diagnosis and the 6-month cumulative radiation exposure. Secondary outcomes were serious adverse events, pain, return to the ED, hospitalizations, and diagnostic accuracy.


  • Sensitivity and specificity for CT in this study was 86% and 53%, which are much lower than most reported rates in the rest of the literature. This is likely because the standard for diagnosis (passing a stone, or having surgery) was so high.

  • Sensitivity and specificity between the POCUS and US performed by radiology was very similar

  • There were similar rates of complications, ED length of stay, return to the ED, admissions, and pain between all groups

  • The 6-month cumulative radiation exposure was lower in the ultrasound groups

  • The numbers for the entire US groups approached those of the CT group after people had subsequent imaging (US only was about 54% sensitive and 74% specific)


  • Ultrasound has a place as an imaging modality for kidney stones.

  • For patients in whom you have a high suspicion of a kidney stone, it isn’t unreasonable to start with an ultrasound first.

  • This spares them quite a bit of radiation long term (only 40% went on to get a CT), and there are very similar complications rates between all the groups.

  • This does not necessarily apply to those patients in whom you are looking for alternative pathology.

  • I also learned was what a pragmatic study was. Randomizing a variable and then intentionally not controlling for others as a way of seeing how an intervention/change is measured in real world conditions.


Starting with US for high suspicion of kidney stone as the primary diagnosis is not unreasonable.