Objective: The CHOKAI and STONE scores are clinical prediction rules to predict ureteral stones in patients presenting with renal colic. Both systems contribute to reducing diagnostic radiation exposure; however, few studies have compared the two scoring systems. Therefore, we aimed to compare these systems and assess their diagnostic accuracy for ureteral stones.
Methods: This was a multicenter prospective observational study performed between 2017 and 2018, including patients aged >15 years with renal colic and suspected with ureteral stones. We calculated the CHOKAI and STONE scores of each patient based on their medical interviews and physical and laboratory findings. Primary outcome was differences in the area under the receiver operating characteristic curve in each model, and secondary outcome was diagnostic accuracy at the optimal cut-off point.
Results: Of the 124 patients included, 84 were diagnosed with ureteral stones. The area under the curve of the CHOKAI score was 0.95, showing a sensitivity of 0.93, specificity of 0.90, positive likelihood ratio of 9.3, and negative likelihood ratio of 0.079, at an optimal cut-off point of 6. The area under the curve of the STONE score was 0.88, showing a sensitivity of 0.68, specificity of 0.90, positive likelihood ratio of 6.8, and negative likelihood ratio of 0.36, at an optimal cut-off point of 9. Thus, the area under the curve was significantly higher for the CHOKAI score than for the STONE score (p = 0.0028).
Conclusions: The CHOKAI score has a diagnostic performance superior to that of the STONE score in this population.
Keywords: Clinical prediction rule; Emergency; Renal colic; Ureterolithiasis.
Copyright © 2019 Elsevier Inc. All rights reserved.