RESUMEN
OBJECTIVE: The ability of emergency physicians (EPs) to identify hydronephrosis using point-of-care ultrasound (POCUS) has been assessed in the past using computed tomography (CT) scans as the reference standard. We aimed to determine the ability of EPs to identify and grade hydronephrosis on POCUS using the consensus interpretation of POCUS by emergency radiologists as the reference standard. METHODS: The study was conducted at an urban academic emergency department (ED) as a secondary analysis of previously collected ultrasound data from the EP-performed POCUS databank. Patients were eligible for inclusion if they had both POCUS and CT scanning performed during the index ED visit. Two board-certified emergency radiologists and six EPs interpreted each POCUS study independently. The interpretations were compared with the consensus interpretation by emergency radiologists. Additionally, the POCUS interpretations were also compared with the corresponding CT findings. Institutional approval was obtained for conducting this study. All the analyses were performed using Stata MP 14.0 (StataCorp). RESULTS: A total of 651 patient image-data sets were eligible for inclusion in this study. Hydronephrosis was reported in 69.6% of POCUS examinations by radiologists and 72.7% of CT scans (p = 0.22). Using the consensus radiology interpretation of POCUS as the reference standard, EPs had an overall sensitivity of 85.7% (95% confidence interval [CI] = 84.3%-87.0%), specificity of 65.9% (95% CI = 63.1%-68.7%), positive likelihood ratio of 2.5 (95% CI = 2.3-2.7), and negative likelihood ratio of 0.22 (95% CI = 0.19-0.24) for hydronephrosis. When using CT scan as the reference standard, the EPs had an overall sensitivity of 81.1% (95% CI = 79.6% to 82.5%), specificity of 59.4% (95% CI = 56.4%-62.5%), positive likelihood ratio of 2.0 (95% CI = 1.8-2.2), and negative likelihood ratio of 0.32 (95% CI = 0.29-0.35) for hydronephrosis. The specificity of EPs was improved to 94.6% (95% CI = 93.7%-95.4%) for categorizing the degree of hydronephrosis as "moderate or severe" versus "none or mild," with positive likelihood ratio of 6.33 (95% CI = 5.3-7.5) and negative likelihood ratio of 0.69 (95% CI = 0.66-0.73). CONCLUSIONS: Emergency physicians were found to have moderate to high sensitivity for identifying hydronephrosis on POCUS when compared with the consensus interpretation of the same studies by emergency radiologists. These POCUS findings by EPs produced more definitive results when at least moderate degree of hydronephrosis was present.