Written by: Kishan Ughreja , MD (NUEM ‘23) Edited by: Ade Akhentuamhen, MD (NUEM ‘21)
Expert Commentary by: Tim Loftus, MD, MBA
Journal Club: Ultrasonography versus Computed Tomography for Suspected Nephrolithiasis
A 70-year-old man with BPH s/p TURP, hypertension, hyperlipidemia and stroke presents to the ED with acute onset of intermittent sharp left flank pain radiating into the groin that awoke him from sleep. He endorses nausea without vomiting and denies fever. He also endorses slightly decreased urination with “dribbling.” His urinalysis shows >100 RBC and no signs of infection. Nephrolithiasis is likely high on your differential diagnosis. How do you proceed?
What is your initial imaging test of choice, ultrasound (US) or non-contrast CT, and why?
Would you be satisfied with only US and no follow-up CT?
Would you be confident in a point-of-care-ultrasound evaluation or a formal ultrasound?
Do outcomes for patients with suspected nephrolithiasis differ based on the initial imaging?
Should your medical decision-making change if the patient has a history of nephrolithiasis?
What would you do if the same patient presented again with persistent pain from a previously diagnosed stone?
Pain from suspected nephrolithiasis is a very common complaint in the ED and the incidence of the disease continues to increase. The estimated incidence over the past two decades is up to 340 visits per 100,000 individuals.1 Low-dose non-contrast abdominal CT has become the gold standard for diagnosis as it has become readily available in emergency departments nationwide, with some studies touting sensitivity and specificity of 97% and 95%, respectively.2 However, low dose CT still exposes the patient to radiation and may increase their risk of cancer, as many nephrolithiasis patients often undergo repeat imaging because of recurring pain or urological intervention. Additionally, CT scans prolong average ED lengths of stay. However, with ultrasonography becoming more prevalent in EDs, it may be possible that initial imaging may avoid this radiation risk and still have similar outcomes for patients. Let’s analyze this NEJM article comparing US to CT for the assessment of nephrolithiasis.
Study design: a multicenter, pragmatic, randomized comparative effectiveness trial
Population
N = 2759
ages 18- 76 yo
reported flank or abdominal pain that the treating physician wished to order imaging to establish or rule out a primary diagnosis of nephrolithiasis
not considered at high risk for serious alternative diagnoses e.g. cholecystitis, appendicitis, aortic aneurysm, or bowel disorders
no pregnant patients
no men >129 kg, no women >113 kg
no history of single kidney, renal transplantation, undergoing dialysis
Patient selection
Intervention protocol
patients randomized to 3 groups each using a different initial imaging modality (POCUS vs. Radiology US vs. CT)
patients contacted at 3, 7, 30, 90, and 180 days after randomization to assess study outcomes
Outcome measures
Primary Outcomes
high-risk diagnoses with complications that could be related to missed or delayed diagnoses — within 30 days of ED visit, including:
AAA w/rupture, PNA w/sepsis, appendicitis w/rupture, diverticulitis w/abscess or sepsis, bowel ischemia or perforation, renal infarction, renal stone w/abscess, pyelonephritis w/urosepsis or bacteremia, ovarian torsion w/necrosis, aortic dissection w/ischemia
cumulative radiation exposure from all imaging within 6 months after randomization
total cost (not reported in this study, ongoing analysis)
Secondary Outcomes
serious adverse events (FDA definition)
serious adverse events related to study participation
delayed diagnosis, like acute cholecystitis, appendicitis, bowel obstruction
return ED visits
hospitalizations after being discharged from ED
self-reported pain scores
diagnostic accuracy for nephrolithiasis
by comparing ED diagnosis at discharge to reference standard of confirmed stone by patient’s observation of passage or report of surgical removal
Results
no significant differences among groups in terms of pain scores, medical history, physical exam findings, and ED physician’s assessment of the likelihood of other diagnoses (Table 2)
POCUS and US groups had significantly lower cumulative radiation exposure over 6 months than the CT group (difference attributed to initial ED visit’s imaging choice)
11 patients (0.4%) had high risk diagnoses with complications during first 30 days after randomization, with no significant difference among the 3 groups
no difference when stratified by patients with a history of nephrolithiasis
no significant difference among groups in the number of patients with serious adverse events; total of 466 SAE in 316 patients (91.4% were hospitalizations during f/u period; 26.4% involved surgical treatment of complications of nephrolithiasis)
5 reported deaths (occurred between 38 and 174 days after randomization) — none thought to be related to study participation
the proportion of patients with a confirmed stone diagnosis within 6 months was similar in all 3 groups (POCUS 34.5% vs. US 31.2% vs. 32.7% CT)
diagnostic accuracy based on result of initial imaging modality
POCUS sensitivity 54% [48 - 60]; specificity 71% [67 - 75]
US sensitivity 57% [51 - 64]; specificity 73% [69 - 77]
CT sensitivity 88% [84 - 92]; specificity 58% [55 - 62]
Interpretation
The US group was exposed to less radiation than the CT group and had no significant differences in the incidence of high-risk diagnoses with complications, total serious adverse events, or related serious adverse events.
There also were no significant differences in pain scores, hospitalizations, ED readmissions among the groups.
Many patients in the ultrasound groups did get additional imaging, but this was not the majority.
Patients with a history of nephrolithiasis were less likely to undergo additional imaging with CT if they already had an ultrasound first (31% vs 36%). They did not have poorer outcomes than patients without a history of nephrolithiasis.
Patients only undergoing POCUS and no other testing had a significantly shorter ED stay (1.3 hours)
It is safe to pursue ultrasound as the initial imaging of choice for suspected nephrolithiasis (with additional imaging ordered as necessary at clinical discretion), though it should not necessarily be the only testing performed.
Strengths
large size, diversity in ED settings, randomized design, assessment of clinically important outcomes, a high follow-up rate
Weaknesses
no blinding of investigators, physicians, or patients as this was a pragmatic trial design
independent review was used to characterize serious adverse events related to study participation
strict reference standard for stone diagnosis which was unbiased, but prone to error based on the patient’s memory of self-reporting of stone passage
Internal/external validity
Given the aforementioned strengths of this study and its pragmatic design, these findings appear both internally and externally valid and may be applied to daily clinical practice
Take-Home Points
What is your initial imaging test of choice, ultrasound (US) or non-contrast CT, and why?
Ultrasound is a good choice for initial imaging as most patients do not end up requiring additional imaging during their visit. This leads to reduced cumulative radiation exposure.
Would you be satisfied with only US and no follow-up CT?
In this study, 40.7% of those in the POCUS group and 27% in the formal ultrasound group underwent subsequent CT. Follow up CT should depend on the patient and ultrasound operator. Keep in mind that this study excluded patients with kidney disease, pregnant patients, and obese patients. They also excluded patients who were high risk for other pelvic and abdominal diseases. Lastly the POCUS operators were ED physicians with training “recommended by ACEP.”
Would you be confident in a point-of-care-ultrasound evaluation compared to a formal ultrasound?
Yes. Sensitivity and specificity between these groups were similar.
Do outcomes for patients with suspected nephrolithiasis differ based on the initial imaging?
No. There was no significant difference in subsequent adverse events, pain, return visits or hospitalizations, or delayed diagnoses of other serious conditions.
Should your medical decision-making change if the patient has a history of nephrolithiasis?
In this study, patients with a history of nephrolithiasis were less likely to undergo additional imaging with CT if they already had an ultrasound first. They did not have poorer outcomes than patients without a history of nephrolithiasis. This suggests that it is safe to avoid ordering a CT in patients with recurrent stones.
What would you do if the same patient presented again with persistent pain from a previously diagnosed stone?
The majority of patients with adverse outcomes were due to infectious causes. Consider alternative diagnoses such as pyelonephritis. Additionally, although rare, renal infarct can present with acute flank pain and is diagnosed with a contrast CT.
References
Fwu, C. W., Eggers, P. W., Kimmel, P. L., Kusek, J. W., & Kirkali, Z. (2013). Emergency department visits, use of imaging, and drugs for urolithiasis have increased in the United States. Kidney international, 83(3), 479-486.
Coursey, C. A., Casalino, D. D., Remer, E. M., Arellano, R. S., Bishoff, J. T., Dighe, M., ... & Leyendecker, J. R. (2012). ACR Appropriateness Criteria® acute onset flank pain–suspicion of stone disease. Ultrasound quarterly, 28(3), 227-233.
Smith-Bindman, R., Aubin, C., Bailitz, J., Bengiamin, R. N., Camargo Jr, C. A., Corbo, J., ... & Kang, T. L. (2014). Ultrasonography versus computed tomography for suspected nephrolithiasis. New England Journal of Medicine, 371(12), 1100-1110.
Expert Commentary
Thank you very much to Dr.’s Ughreja and Akhetuamhen for an excellent blog post on a very relevant clinical topic. This is a great summary of the landmark randomized trial published in NEJM in 2014 assessing CT vs two types of US for patients with suspected renal colic in the ED setting. It is worth mentioning that this study was a multicenter study based in the US with representation from ED, Radiology, and Urology. The above study was well summarized and bears repeating that, in this multicenter randomized study assessing CT vs POCUS vs radiology performed US in patients with suspected renal colic in the ED setting, initial US reduced radiation exposure without adversely affecting patient-centered outcomes. It is worth mentioning several additional considerations and placing emphasis on others elucidated from this journal club review.
First, a subsequent systematic review (1) incorporating multispecialty (ED, Radiology, Urology) expert panel consensus recommendations has reiterated that in younger patients without a high suspicion for alternative diagnoses or complicating features of nephroureterolithiasis (such as fever, pyelonephritis, solitary kidney, dialysis, etc), US should be the initial diagnostic imaging modality of choice, if any. It's a great paper, worth reading (and appreciating who the authors are), and worth recalling for bedside teaching to junior learners in the ED.
Additionally, this paper brings to mind my second point, and something that is worth shouting from the hilltops -- a kidney stone is a clinical diagnosis! Now, of course, this is exclusive of those patients with high-risk or complicating features (e.g. pediatrics, pregnancy, solitary kidney, fever, unstable/critically ill, unrelenting pain, atypical features, etc). You don’t need any imaging to tell you the diagnosis in the vast majority of patients. US or CT are helpful in confirming the diagnosis when there is uncertainty or non-trivial pretest probability of alternative diagnoses, excluding alternative diagnoses, and identifying exact stone location and size, which can be used to help counsel patients at the bedside regarding the anticipated clinical course and next steps in management.
Third, for those with proper training, and with some exceptions (see the systematic review paper for case vignettes that highlight these), POCUS is non-inferior to radiology-performed US. And, it's not a “formal” US. I can’t remember the last time I attended a black-tie ultrasonography session, but that's just me.
Fourth, it's worth mentioning that although CT use can lead to the identification of incidental findings more commonly than US, identification of these incidental findings still happens rather often with POCUS (a common example is a renal cyst). Please ensure that you document and discuss with the patient accordingly.
Finally, a burden on us as EM clinicians is training in and awareness of clinical practice guidelines and recommendations from specialties outside of EM. As it relates to the diagnostic evaluation of suspected renal colic in the ED setting, the Choosing Wisely recommendations endorsed by the AUA are worth perusing as are the European/EUA guidelines, both of which suggest US as the initial diagnostic imaging modality of choice, for pediatric (CW) and non-high-risk patients without complicating features (EUA).
The bottom line is that CT is helpful for older patients or those in whom you are less sure about the diagnosis of renal colic. For younger or low-risk patients, suspected renal colic is a clinical diagnosis and often needs no imaging, but ultrasound would be an evidence-based first step. Thanks again toDr.’s Ughreja and Akhetuamhen.
References
1) Moore et al. Imaging in suspected renal colic: a systematic review of the literature and multispecialty consensus. J Urol 2019. 202(3):475-483.
Tim Loftus, MD, MBA
Assistant Professor of Emergency Medicine
Fellowship Director of the Clinical Operations and Administration Fellowship Program, Northwestern Department of Emergency Medicine
Medical Director of Emergency Services Northwestern Lake Forest Hospital and Grayslake Emergency Center
How To Cite This Post:
[Peer-Reviewed, Web Publication] Ughreja, K. Akhentuamhen, A. (2022, May 16). Journal Club: Ultrasonography versus Computed Tomography for Suspected Nephrolithiasis. [NUEM Blog. Expert Commentary by Loftus, T]. Retrieved from http://www.nuemblog.com/blog/nephrolithiasis-ultrasonography-versus-computed-tomography.
