Imaging for renal colic

I have always thought imaging in renal colic is pretty easy. It is required if you are searching for an alternative diagnosis. It is required if the patient is septic and needs the OR urgently. It is required if the patient’s pain can’t be controlled and a surgical intervention might be needed. Otherwise, it generally isn’t necessary. However, the issue of imaging in renal colic still seems to create a lot of controversy. There was the excellent RCT from 2014 that demonstrated no differences in clinical outcomes (aside from increased radiation with CT) between bedside ultrasound, formal ultrasound, and CT as the first line imaging choice in emergency department patients with renal colic. (Smith-Bindman 2014) (I was sure that I had covered this paper on First10EM in the past, but I can’t find it.) However, the use of CT has grown significantly over the last few decades, and despite the increased use of CT, patient oriented outcomes (including admission and surgical intervention) have not changed at all, indicating that many of these CTs are unnecessary. (Gottlieb 2002; Westphalen 2017) Therefore, it sounds like we need more guidance on imaging…

The paper

Moore CL, Carpenter CR, Heilbrun ME, et al. Imaging in Suspected Renal Colic: Systematic Review of the Literature and Multispecialty Consensus. Ann Emerg Med. 2019;74(3):391–399. PMID: 31402153 DOI: 10.1016/j.annemergmed.2019.04.021

The Methods

I love the methodology of this paper. They start with a systematic review looking at the evidence for imaging in renal colic. Then , after reviewing the evidence, they created a series of 29 clinical vignettes to determine what imaging, if any, a group of experts (made up of emergency physicians, urologists, and radiologists) would recommend. They used a modified Delphi process to determine the imaging strategy recommended by this group of 9 experts. 

The Results

The systematic review

They found 232 articles that were deemed relevant and had appropriate methodology. 

Importantly, CT is assumed to be the reference standard in all these studies, so we can’t actually say how sensitive or specific the test is (and it is definitely not perfect). The best we can say is that CT finds an alternative diagnosis is between 0 and 33% of patients (the range is wide because the definition of alternative diagnosis is unclear). They say that a clinically important diagnosis is found on CT in less than 5% of patients (and we can presumably get better by selectively scanning higher risk patients). Furthermore, despite increasing CT use over the years, there has not been a change in interventions, so we are presumably ordering a lot of CTs that don’t change management. 

Of course, ultrasound is not as accurate as CT, but it may be good enough for many (if not most) clinical scenarios. The sensitivity of a formal ultrasound is reported as between 3% and 98%, so obviously that evidence is pretty useless. (The range is so high because it depends on whether you actually needed to see the stone, or would accept indirect evidence such as hydronephrosis.) For POCUS, they report a pooled sensitivity of 70% and specificity of 75%. However, these poor numbers are for all stones, and we don’t really care about the small ones, which are likely to pass on their own without complication. Multiple studies show that ultrasound is very unlikely to miss stones that require surgical intervention.

The Delphi Process

They had perfect consensus on 15 of the 29 (52%) of the vignettes. There was excellent consensus (8 of 9 clinicians agreed) on another (28%) of vignettes. So, overall, the agreement was really good, but not perfect. 

These experts thought that no imaging was required in 45% of the cases presented, ultrasound was the best choice in 31%, and CT was the best imaging choice in only 24%.

I can’t go through all 29 vignettes. I think there is value in reading them for yourself, and thinking about whether you agree with the authors. Specifically, I think there is a lot of value in looking at the cases in which you disagree with the authors, and thinking carefully about your reasoning. I will highlight a few important cases:

• A 35 year old man with a history of stones and a classic presentation whose pain improves with analgesia: NO IMAGING
• A 35 year man with no history of stones with a classic presentation whose pain improves with analgesia: ULTRASOUND (This one is interesting. Although they suggest ultrasound, 2 of the later vignettes ask what we should do with this same patient if the ultrasound is positive and if the ultrasound in negative. In both cases there was perfect agreement: NO FURTHER IMAGING. Thus, it isn’t clear why you need the ultrasound in the first place.)
• A 35 year man with no history of stones with a classic presentation whose pain does not improve with analgesia: LOW DOSE CT
• A 55 year old man with a history of stones and a classic presentation whose pain improves with analgesia: ULTRASOUND
• A 55 year old man with no history of stones and a classic presentation whose pain improves with analgesia: LOW DOSE CT
• A 75 year old man with a history of stones and a classic presentation whose pain improves with analgesia: LOW DOSE CT

I think you can see the general trend here. Young patients with clear stories get no imaging. If their pain doesn’t resolve, or you add a complicating factor like abdominal tenderness or no past history of stone, move on to ultrasound. If the patient is older, or you can’t adequately control the pain, use CT scan. Pregnant women and children get ultrasound (or no imaging). 

There are 2 special cases worth mentioning: the patients presenting after procedures. If a patient has a stent in place, use ultrasound, because the only question is whether or not the stent is working. Hydronephrosis means that it isn’t. A lack of hydronephrosis means the patient can be sent home after adequate analgesia. Similarly, after lithotripsy, bedside ultrasound was suggested.

My thoughts

I love this combination of evidence based medicine (through a systematic review) with expert opinion (through the Delphi process). People often ask how expertise can be incorporated into evidence based medicine, and I think that this study provides an excellent example. (That being said, if you disagree with their recommendations, their reasoning is opaque, which I think is a significant problem. I think this process could be improved by providing explicit reasoning for the decisions being presented.)

Obviously, there are some weaknesses to the study design. The vignettes are somewhat artificial. In real life practice, patient complaints are frequently more complex or more subtle. The clinicians knew that the study was about renal colic, which could bias the results. (I think this could go either way. They might be less likely to suggest imaging, because they are sure that the diagnosis is renal colic. On the other hand, because they are specifically thinking about imaging, they might be more likely to suggest it.) Furthermore, it isn’t clear how they developed this sample of experts, and their opinions may not be representative of the wider medical community.

In my mind, these recommendations are probably still too liberal in their use of CT. The authors seem to assume that CT is the best imaging modality for alternative diagnoses, but I don’t think that is true. Even if a patient has abdominal tenderness (and therefore an alternative diagnosis is likely), ultrasound may still be the ideal imaging strategy. I start with ultrasound for all appendicitis, and it is also a great test for diverticulitis. Bedside ultrasound is almost perfect for ruling out an abdominal aortic aneurysm. These recommendations seem to ignore ultrasound and just assume that CT is the ideal imaging modality if you are worried about an alternative diagnosis.

Agreement was not perfect in all cases. If the experts can’t agree, no one should be telling you exactly how to practice. However, there was broad agreement here that CT was not necessary in all cases (it was only suggested in 24% of the vignettes), so if you are someone who is currently relying heavily on CT for this diagnosis, you probably want to reassess your practice. 

Bottom line

The older a patient is, and the less sure you are about the diagnosis of renal colic, the more benefit there will be from CT. In younger patients with a clear diagnosis, no imaging is required at all. For intermediate patients, ultrasound is a great starting point.

I will reiterate my initial thought: imaging for renal colic is pretty easy. It is required if you are searching for an alternative diagnosis. It is required if the patient is septic and needs the OR urgently. It is required if the patient’s pain can’t be controlled and a surgical intervention might be required. Otherwise, it generally isn’t necessary.

Other FOAMed

SGEM XTRA: COME TOGETHER, RIGHT NOW – OVER RENAL COLIC

References

Gottlieb RH, La TC, Erturk EN, et al. CT in Detecting Urinary Tract Calculi: Influence on Patient Imaging and Clinical Outcomes Radiology. 2002; 225(2):441-449.

Moore CL, Carpenter CR, Heilbrun ME, et al. Imaging in Suspected Renal Colic: Systematic Review of the Literature and Multispecialty Consensus. Ann Emerg Med. 2019;74(3):391–399. PMID: 31402153 DOI: 10.1016/j.annemergmed.2019.04.021

Smith-Bindman R, Aubin C, Bailitz J, et al. Ultrasonography versus computed tomography for suspected nephrolithiasis. N Engl J Med. 2014;371(12):1100–1110. PMID: 25229916 DOI: 10.1056/NEJMoa1404446

Westphalen AC, Hsia RY, Maselli JH, Wang R, Gonzales R. Radiological Imaging of Patients With Suspected Urinary Tract Stones: National Trends, Diagnoses, and Predictors . 2011; 18(7):699-707.

---

You can find more First10EM critical appraisal here.

---