|Topic 2 - Summary
Smith-Bindman R, Aubin C, et al. “Ultrasonography versus Computed Tomography for Suspected
Nephrolithiasis.” N Engl J Med. 2014 Sep 18;371(12):1100-10
|2. Summary/Bottom Lines:
- The current gold standard for diagnosing nephrolithiasis is CT, but patients are exposed
to radiation and the use of CT over other imaging modalities has not been shown to
improve patient outcomes. However, use of ultrasound (U.S.) as an alternative has
been problematic due to low sensitivity.
- This study used an intention-to-treat analysis aimed at identifying differences in
patient outcomes when comparing US and CT as the initial imaging test for suspected nephrolithiasis in the emergency department (ED), with
additional imaging being used as needed.
- Patients initially receiving ultrasound (ED point-of-care U.S. or radiology department
U.S.) were exposed to less radiation, with no differences in patient outcomes (missed/delayed
serious diagnoses, pain, return ED visits, hospital admission after ED discharge),
when compared to patients initially receiving CT.
- Other potential benefits of U.S. included a minor financial benefit ($25 cheaper in
patients randomized to radiology US vs. CT) and shorter length of stay (ED US and
CT vs. radiology US).
- The authors don’t suggest using US as the sole imaging test, but they advocate for US as the initial test and supplementing with additional imaging, if indicated.
- Changes to practice are unlikely to occur on the basis of this article alone, given
some outstanding questions about the study’s protocol and methods. Some of the ideas
and data expressed in the study are encouraging and might instigate changes in practice
after additional studies.
- U.S. groups were exposed to less radiation (about half) during the initial ED visit,
and less cumulative radiation over 6 months.
- No significant difference in patient outcomes (missed/delayed serious diagnoses, pain,
return ED visits, hospital admission after ED discharge)
- Other benefits: minor financial benefit ($25 cheaper if received radiology US vs CT),
shorter ED length of stay (with ED US and CT vs. radiology ultrasound)
- [Reported by group (ED US, radiology U.S., and CT, respectively)]
- No difference in diagnostic accuracy (calculated based on ED discharge diagnosis):
Sensitivity 85%, 84%, 86%, Specificity 50%, 53%, 53%
- Significant difference in diagnostic accuracy (calculated from the results of the
initial imaging method only): Sensitivity 54%, 57%, 88%, Sensitivity 71%, 73%, 58%
- Crossover rates (patients receiving additional imaging tests) 40.7%, 27.0%, 5.1%
- Multicenter, prospective, randomized trial analyzing 2758 patients with a suspected
primary diagnosis of nephrolithiasis
- Randomized to 1 of 3 arms (point-of-care U.S. in ED, radiology department U.S., or
- Patients received the initial imaging test to which they were assigned, but could
receive additional imaging as indicated (determined by the treating physician)
- Data collected from medical records and patient surveys at 3, 7, 30, 90, and 180 days
- Gold standard: patients reporting the passing of a stone, or stone was surgically
- The choice of gold standard was questionable. Only ~33% of the patients in the study
had a confirmed diagnosis according to the gold standard, causing the diagnostic accuracy
results to be inaccurate and unreliable.
- Patients, physicians, and investigators were not blinded to study arm assignment.
- The generalizability limits applicability to large academic EDs, in non-pregnant,
non-obese adults ages 18-76, who have a suspected primary diagnosis of nephrolithiasis. (Other study exclusion criteria: single kidney, post-kidney
transplant, or undergoing dialysis).
- Emergency Physicians involved in the study were trained and certified in point-of-care
U.S.; results may vary based on level of ultrasound training.
- The study protocol did not include specific U.S. criteria used for diagnosis of nephrolithiasis
or reasons cited for ordering additional imaging. (Was more imaging ordered because there were no definitive findings of nephrolithiasis
on U.S. despite high suspicion for a stone, extra characterization of a stone found
on U.S. was needed, or the physicians were trying to assess for an alternate diagnosis?)
|5. Additional Info:
Dr. Nelson, in addition to identifying issues with the methods of the study (namely
the poor choice of gold standard), still advocates for diagnosis by CT
He does not believe the results showed any relevant difference between the groups
(patient outcomes, cumulative radiation over 6 months, length of stay, and costs were
either the same or there was no practical difference). He stated that even though
the US patients were spared radiation at the initial visit, there was still significant
crossover, and these patients likely received a CT within 6 months after discharge.
Additionally, he pointed out that urologists/surgeons would likely want to see CT
results when seeing patients for follow-up after ED discharge. So why not just get
||Dr. Kara Samsel