The Rational Approach to Diagnostic Imaging for Ureterolithiasis

Kevin Klauer, DO

With greater scrutiny of CT utilization due to cost and its associated ionizing radiation exposure, many have challenged the routine use of CT for the diagnosis of ureterolithiasis. So, if not CT, then what modality should be employed? Perhaps the more important question is, “Are any radiologic diagnostics necessary at all?”

As we challenge the reflexive ordering of CT in this essay, this discussion may also prompt other critical questions that may further challenge the diagnostic and therapeutic approach to patients with suspect kidney stones. Even our terminology has evolved. “Kidney stone” will likely be the standard lay term, which is, perhaps, synonymous with nephrolithiasis. However, all “kidney stones” are not symptomatic, and those that reside in the renal parenchyma are not active. However, those traversing the ureter are active and symptomatic. Thus, ureterolithiasis is a more appropriate descriptive term.

Dating back to the mid to late 1990s, the intravenous pyelogram (IVP) was the gold standard for evaluation of suspected ureterolithiasis. However, CT quickly gained favor as it exceeded the diagnostic performance of IVP1. Just as today, the cost of CT was a concern and was a barrier to the adoption of CT over IVP. As hospitals began to reduce the price of CT to approximate that of IVP, CT quickly became the gold standard. However, IVP has retained its dominance as the study of choice for kidney function.

Ultrasound has been a viable alternative to CT for quite some time, but has recently been more seriously considered due to the concerns relating to ionizing radiation exposure and the costs associated with CT. With the widespread acceptance and availability of point-of-care ultrasound (POCUS) for many purposes, investigators have challenged the premise that CT is better. Most will agree that ultrasound cannot compare to the CT scan with respect to its ability to identify alternative pathology and its level of detail in which pathology can be defined. However, for those suspected of ureterolithiasis, ultrasound shouldn’t have to meet the standard that CT has set. Alternatively, ultrasound just has to be good enough, good enough to strongly suggest the presence of a kidney stone and strong enough to rule out life threats (e.g. abdominal aortic aneurysm).

In 2014, a comparative effectiveness trial was published in the New England Journal of Medicine. The authors randomized 2,759 patients (ages 18–76 years) to either CT, formal ultrasound or point-of-care ultrasound for the diagnosis of ureterolithiasis. The incidence of a high-risk diagnosis (abdominal aortic aneurysm with rupture, pneumonia with sepsis, appendicitis with rupture, diverticulitis with abscess or sepsis, bowel ischemia or perforation, renal infarction, renal stone with abscess, pyelonephritis with urosepsis or bacteremia, ovarian torsion with necrosis, or aortic dissection with ischemia) with complications in the first 30 days was 0.4% and was the same for all three groups2. Regardless of diagnostic approach, there were very few high-risk diagnoses with complications and all three approaches performed equally as well in this regard.

It seems clear that ultrasound is good enough, when compared to CT, particularly when etiologies such as abdominal aortic aneurysm (AAA) can be adequately evaluated via ultrasound. However, this question begs the next, and we shouldn’t have to wait a decade to address it. “In properly selected patients, is any imaging necessary?” Although data may be forthcoming to definitely and academically answer this question, it seems that common sense can guide us until that time. If we have selected a population that is at high risk for ureterolithiasis and low risk, or no risk, for serious or life-threatening pathology, it stands to reason that foregoing imaging of any kind will not pose any undue risk to these patients and should not result in serious rates of misdiagnosis. Further, when a test won’t change your management, perhaps you don’t need the test. According to Singh, who published a population-based study of 6,386 men and women aged 25–84 years, no one under the age of 48 had an AAA, men were more likely to experience AAA then women (8.9% versus 2.2%), ever use for antihypertensives was associated with an odds ratio of 1.61 and 1.14, respectively for men and women, and finally, an odds ratio of 8.0 was associated with tobacco use for forty or more years3. Forsdahl, et al. found similar findings in that traditional risk factors for atherosclerotic disease were also risk factors for AAA4.

Although formal validation is always useful and often necessary to sway the masses, it is reasonable to consider that young patients (under 48 years of age), without any risk factors for atherosclerosis (particularly females) with a classic presentation for ureterolithiasis are very unlikely to have an AAA masquerading as a stone. Many would agree that a young healthy person with a previous history of ureterolithiasis and a classic presentation that is similar to their previous kidney stones would be enough to base your diagnosis on. I would agree, but would take this line of reasoning one step further. Why do we need the extra validation of a presentation similar to their previous? Low risk is low risk. Thus, we may be able to avoid imaging even in first time presentations, provided that such patients are carefully selected and have reasonable access to follow up and also understand that their diagnosis is presumptive and has not been definitively confirmed. Such cases are excellent opportunities to employ shared decision-making, in which the patient would agree to accept some degree of uncertainty to avoid the diagnostic test. As this debate and data have been noted in the lay press, patients are becoming more aware of their options5.

For those who may need additional data to help verify diagnostic accuracy, the urine may be the solution. Evaluating urine for infection is critical in the context of ureterolithiasis. Thus, this should be ordered in any case, and the presence of hematuria may help confirm the presumptive diagnosis of ureterolithiasis. Argyropoulos A, et al. reported that in 609 patients with imaging-confirmed stones of 3mm or greater, the dipstick urinalysis was negative in 7.1% and follow-up microscopic urinalysis yielded an additional 2%6. Thus, based on this data, 95% of patients with a stone greater than 3mm will have evidence of hematuria5. So, even if those with false-negative urinalyses are imaged for diagnostic verification, the vast majority may avoid imaging altogether.

Some may be concerned about the presence of hematuria being falsely positive for ureterolithiasis and an important sign of an AAA. Hematuria is not common with AAA but may be seen in approximately 17% of patients with AAA and an aortocaval fistula, which is a rare complication occurring in 2–4% of patients with AAA7,8. Therefore, if the presence of hematuria is due to an AAA, as opposed to a kidney stone, the patient’s presentation should be evidence of such advanced disease.

Despite challenging the need for imaging, there still remain some patients who should be imaged and for whom CT is the most appropriate modality. Certainly, imaging is paramount in patients who either are at risk for an AAA or other more serious pathology or whose presentations are suggestive, but not classic, for uncomplicated ureterolithiasis (i.e., fever or abdominal tenderness). In the former, ultrasound, either point-of-care or formal, may be sufficient. However, in the latter group, considering appendicitis, perinephric abscess, retroperitoneal hematoma, ovarian torsion, etc., CT or ultrasound may be more appropriate depending on the etiologies of primary concern.

Although a minority of patients with suspected ureterolithiasis may still need imaging, the data, combined with a large dose of common sense, can help providers develop a rational diagnostic strategy to avoid unnecessary exposure to CT-associated radiation and the cost and time associated with imaging altogether.

Kevin Klauer, DO, EJD
Chief Medical Officer, Emergency Medicine
TeamHealth, Knoxville, TN
Assistant Clinical Professor
Michigan State University College of Osteopathic Medicine
Co-Host, ED Leadership Monthly

Citations

1. Worster A, et al. The accuracy of noncontrast helical computed tomography versus intravenous pyelography in the diagnosis of suspected acute urolithiasis: a meta-analysis.. Ann Emerg Med. September 2002;40:280–6.
2. Smith-Bindman, R et al. Ultrasonography versus computed tomography for suspected nephrolithiasis. N Engl J Med. 2014 Sept 18;371(12):1100–10.
3. Singh K., et al. Prevalence of and Risk Factors for Abdominal Aortic Aneurysms in a Population-based Study: The Tromso Study Am. J. Epidemiol. (2001) 154(3): 236–244.
4. Forsdahl SH, et al. Risk Factors for Abdominal Aortic Aneurysms
   A 7-Year Prospective Study: The Tromso Study, 1994–2001. 2009; 119: 2202–2208.
5. Nicholas Bakalar. The New York Times online. Ultrasound vs. CT for Kidney Stones.
   September 24, 2014; accessed October 7, 2015.
6. Argyropoulos A, et al. The presence of microscopic hematuria detected by urine dipstick test in the evaluation of patients with renal colic. Urol Res. 2004 Aug;32(4):294–7.
7. Brewster DC, Ottinger LW, Darling RC. Hematuria as a sign of aorto-caval fistula. Annals of Surgery. 1977;186(6):766–771.
8. Woolley, Daniel S. et al. Aortocaval fistula treated by aortic exclusion.
   Journal of Vascular Surgery, Volume 22, Issue 5, 639–642.