Alexei Adan, PGY-1
The Article: Smith-Bindman, et al. “Ultrasonagraphy versus Computed Tomography for Suspected Nephrolithiasis.” NEJM 2014; 371:1100-10.
The Idea: Over the last decade, patients with suspected nephrolithiasis have been receiving CT scans for initial diagnostic imaging given its high sensitivity. However, no studies have shown whether imaging patients with suspected renal stones actually changes outcomes. Additionally, the increased use of CT scans means greater radiation exposure for patients. This study compares the use of ultrasound versus CT scan as the initial diagnostic imaging study in patients with suspected renal stones and comparing clinical outcomes.
The Study: This study was a randomized multi-center comparative effectiveness trial from October 2011-February 2013.
Inclusion Criteria: patients aged 18-76 years old presenting to an ED with reported abdominal or flank pain on the track board. Patient was deemed eligible if physician decided to order imaging to establish or rule out a diagnosis of renal stones.
Exclusion Criteria: patients deemed high risk for serious diagnoses (ie cholecystitis, appendicitis, AAA rupture, bowel disorders); pregnant patients; males weighing more than 285 lbs and females weighing more than 250 lbs (given decreased diagnostic accuracy of U/S in obese patients); patients with underlying renal issues including solitary kidney, renal transplant, or on dialysis
There were 2759 patients randomized to 1 of 3 initial imaging modality groups – ultrasound by an EM physician, formal ultrasound by a radiologist, or CT scan. Any subsequent imaging was at the discretion of the ordering physician. Patients were then followed for 6 months and surveyed at intervals regarding subsequent healthcare encounters, which included a follow up of their subsequent medical records and imaging.
Primary Endpoints:
1) 30-day incidence of high-risk diagnoses with complications that could be related to a missed or delayed diagnosis. Diagnoses included: AAA rupture, pneumonia with sepsis, appendicitis with rupture, diverticulitis with abscess or sepsis, bowel ischemia or perforation, renal infarct, renal stones with abscess, pyelonephritis with urosepsis or bacteremia, ovarian torsion with necrosis, or aortic dissection with ischemia.
2) 6-month cumulative radiation exposure
Secondary Endpoints:
1) Serious adverse events related to study
2) Pain score
3) Return ED visits
4) Hospitalizations
5) Diagnostic accuracy
Results: There were a total of 11 patients (0.4%) with high risk diagnoses with complication within the first 30 days among the three groups – 6 in the EM U/S group, 3 in the Radiologist U/S group, and 2 in the CT group, with no statistically significant difference. The total 6-month radiation exposure was lower in both U/S groups – 10.1 mSv in the EM U/S group, 9.3 mSv in the Radiologist U/S group, and 17.2 mSv in the CT group. Looking at secondary endpoints – there was no significant difference among study groups in numbers of patients with serious adverse events, no difference in return to ED rate within 7 to 30 days, no difference in admission rate within 7, 30, and 180 days, no difference in pain score, and no difference in total ED stay time (unless the patient received ONLY an EM U/S for imaging, in which case total ED time was reduced by about one hour). On the other hand, diagnostic accuracy study revealed the CT scan to be 88% sensitive for nephrolithiasis compared to just 54% and 57% in the EM and Radiologist U/S groups respectively.
The Takeaway: The paper reveals a couple important points. First and foremost, although CT scan proves to be more sensitive for the diagnosis of nephrolithiasis, the initial imaging modality did not affect clinical outcomes. The caveat being that this should be applied to the relatively healthy and uncomplicated patient with whom there is a low suspicion for more serious pathology. Given the relatively benign course of uncomplicated nephrolithiasis and its predominantly outpatient management, this raises a more controversial next question – does any sort of initial imaging need to be done for healthy patients presenting with first-time flank pain? In the meantime, this article suggests that using ultrasound is a great place to start, subjecting our patients to less radiation by decreasing the number of unnecessary CT scans.