Use of STONE Score to Predict Urolithiasis in an Asian Emergency Department.

Abstract

The STONE score was developed to predict uncomplicated ureteral stones in patients so that they can be managed without imaging. Validation studies had been conducted previously but the results were varied. This study aims to investigate the utility of the STONE score in an emergency department in Singapore. We retrospectively analyzed the records of adult patients presenting with ureteric colic in the emergency department in 2015. STONE score as well as the proportion of urolithiasis diagnosed on advanced imaging in each STONE score group were calculated. Logistic regression was used to calculate the odds ratios (ORs) for the STONE score components in our study and compared with the ORs obtained in the original study. Measures of diagnostic accuracy for a high STONE score were also calculated. 753 cases were included in the final analysis. Among patients with a high STONE score, 66.7% had urolithiasis and 2.6% had significant alternative diagnoses. Compared to original studies, ORs for the STONE score components obtained for our study were different. From our study, the sensitivity of a high STONE score was 47.0%, specificity was 68.7%, positive predictive value was 66.7%, negative predictive value was 49.3%, positive likelihood ratio was 1.50, and negative likelihood ratio was 0.77. The STONE score is not expected to perform well in Singapore based on our study. It should be used with caution in similar Asian populations.