Background

There are about two million annual emergency department (ED) visits in the United States with suspected renal colic.¹ The lifetime risk of nephrolithiasis (i.e., kidney stones) is 10.6% in men and 7.1% in women in the United States.² Patients frequently seek emergency care for the pain associated with nephrolithiasis because it is typically severe and refractory to over‐the‐counter analgesics. Notably, pain severity does not correlate with the kidney stone size or likelihood of spontaneous passage.^3,4 Clinical features of symptomatic nephrolithiasis include acute flank pain radiating to the groin, nausea, vomiting, and either gross or microscopic hematuria. However, the presence of hydronephrosis on imaging is more specific for obstructing kidney stone than any of these clinical findings.⁵ Standard imaging modalities for the initial diagnosis of nephrolithiasis are either noncontrast spiral computed tomography (CT) scan or ultrasound (US) with no significant difference in cost noted when selecting CT or US.⁶ Healthcare disparities and/or practice variability have been identified in suspected renal colic CT imaging associated with geographic locale and socioeconomic strata.^7,8 Of patients with known nephrolithiasis, almost 50% will develop additional stones within 5–7 years.⁹

Clinical question

How do different diagnostic modalities (unenhanced CT, US, and other diagnostic modalities) compare when it comes to diagnosing nephrolithiasis?

Modern emergency care involves choosing between different options when it comes to the diagnosis of nephrolithiasis in patients with suspected renal colic, in particular the use of unenhanced standard or low‐dose CT, radiology or emergency medicine point‐of‐care in the US, or no imaging. A recent study showed that CT scanning is the predominant test used and is performed in more than 90% of patients diagnosed with kidney stones.¹⁰ Each imaging type gives specific information. In general, unenhanced CT is viewed as the gold standard and can give the most detailed information including the diagnosis of nephrolithiasis, the size of the stone which guides whether intervention is needed (i.e., is ≥10 mm in size), whether hydronephrosis is present, and of overall stone burden.¹¹ In addition, CT can also detect other intraabdominal abnormalities that may be causing symptoms including aortic aneurysm and appendicitis. By contrast, the US is used to detect the presence and degree of hydronephrosis and can also detect anatomical issues or other signs of kidney infection. US is less sensitive in the ability to detect the specific size of the stone.¹² Historically older diagnostic modalities that have been used to diagnose nephrolithiasis have included intravenous (IV) urography, renal tomography, and plain radiography.

An early study on the topic in the late 1980s investigated the sensitivity of the US to detect renal calculi in a three‐phase study in 100 patients.¹³ In the first phase, US was performed after reviewing abdominal radiography and renal tomograms in 30 patients who had undergone extracorporeal shock wave lithotripsy (ESWL), and the sensitivity for detecting stones was 98%. In the second phase, US was performed in 30 post‐ESWL patients without any prior review of radiographs or tomograms. In the second group, the sensitivity of US for stone detection was 95%. In the third phase, 40 patients had the US performed in a blinded way in a random mix of post‐ESWL patients and patients who had undergone urography for other reasons. In this group, the sensitivity of the US was 91%. In combined data, the authors reported an overall sensitivity of the US of 96%.

A 2003 randomized trial compared CT to IV urography for patients with suspected nephrolithiasis enrolled 122 patients with acute flank pain.¹⁴ A total of 59 of them were randomized to CT and 63 to IV urography. The radiographic studies were independently interpreted by four radiologists. Of patients receiving IV urography, mild to moderate adverse reactions from contrast material were seen in three (5%) patients. The mean radiation dose was 3.3 mSv for urography and 6.5 mSv for CT scan. Sensitivity and specificity of CT were 94% and 94%, respectively. For urography, sensitivity and specificity were lower at 85% and 90%, respectively.

Another study investigated the sensitivity of radiology performed diagnostic US compared to CT for the detection of kidney stones in 46 patients with acute flank pain, using a combination of tests and clinical follow‐up as the criterion standard.¹⁵ CT imaging detected 22 of 23 ureteral calculi (sensitivity: 96%), and US detected 14 of 23 ureteral calculi (sensitivity: 61%). The specificity for each technique was 100%. When modalities were compared for the detection of any clinically relevant abnormality, the sensitivities of US and CT were 92% and 100%, respectively.

An additional study was performed in a Spanish hospital and published in 2010.¹⁶ The authors enrolled patients with “persistent renoureteral colic” after standard care. The patients were blindly evaluated by US and CT and, in the 124 patients studied, nephrolithiasis was present in 60%. The specificity and positive predictive value (PPV) of US for nephrolithiasis were 100%, but its sensitivity was only 30%.

Another study determined the test characteristics of US for any ureteral stone and for stones ≥5 mm in size.¹⁷ In 117 patients, the sensitivity for the US was 77% (CI 59–88%) and the specificity was 86% (CI 74–93%) for any stone disease. For stones ≥5 mm, the sensitivity was 90% (CI 54–100%) and the specificity was 64% (CI 53–73%).

A more recent study has examined the sensitivity and specificity of radiology‐performed US for the detection of nephrolithiasis.¹⁸ It also assessed the accuracy to determine the size of calculi and whether this information may help in decisions around counseling patients. In a retrospective study, patients who had undergone an US followed by a noncontrast CT within 60 days were followed at the Cleveland Clinic. Notably, patients were enrolled across a variety of settings, including asymptomatic patients. In 552 patients, the sensitivity and specificity of the US were 54% and 91%. The sensitivity of US increased as stone size increased but was not affected by the location of the stone. When divided by stone size of 0–4 mm (which involve observation) compared to stones ≥5 mm (where intervention is commonly required), there were 54/384 (14%) of cases where US would have led to a recommendation for observation while CT would suggest intervention was needed. In addition, when CT would suggest an intervention, the US would suggest observation in 65/168 (39%) of cases. The authors concluded that by performing US that 119/552 (22%) would be inappropriately counseled with 43% of stones in the 5–10 mm by US having the management recommendation changes when CT was performed.

Clinical question

In ED patients presenting with suspected uncomplicated renal colic which diagnostic strategy should be pursued to optimize patient outcomes?

A large multicenter pragmatic comparative effectiveness clinical trial has demonstrated that there are no differences in serious adverse events when patients were randomized to point‐of‐care ultrasound (POCUS) performed by emergency physicians, US interpreted by radiologists, or unenhanced CT.¹⁹ Specifically, in 2759 patients undergoing randomization to the three interventions, serious adverse events were 12.4%, 10.8%, and 11.2% (p = 0.50), respectively. Pain scores were similar across all three groups and there were no differences in return ED visits, hospitalizations, or diagnostic accuracy. The US groups notably did have less cumulative exposure to radiation at 6 months.

A group of ED physicians, urologists, and radiologists recently convened to answer this question using a systematic review methodology as well as group consensus that was subsequently published in each specialty’s journals.²⁰ The systematic review covered the dates 1995–2018. Following the systematic review of the sensitivity of different imaging studies for renal colic, a series of 29 clinical scenarios was presented with the goal of reaching consensus on the optimal initial imaging. The focus was on scenarios where either standard or reduced‐radiation‐dose CT may not be the clear, initial approach, and other approaches could be taken such as point‐of‐care or radiology‐performed US or no imaging. The authors gained consensus using a modified Delphi process with three rounds of anonymous voting and group discussions with nine members voting.

Related posts:

Adult Septic Arthritis Testicular Torsion Intravascular Volume Status Urinary Tract Infection Rhinosinusitis Cognitive Biases and Mitigation Strategies in Emergency Diagnosis