Comparison of ultrasound versus computed tomography for the detection of kidney stones in the pediatric population: a clinical effectiveness study

Background The incidence of pediatric nephrolithiasis in the United States is increasing. There is a paucity of literature comparing the diagnostic performance of computed ultrasound (US) to tomography (CT) in the pediatric population. Objective To determine the diagnostic performance of renal US for nephrolithiasis in children using a clinical effectiveness approach. Materials and methods Institutional review board approval with a waiver of informed consent was obtained for this retrospective, HIPAA-complaint investigation. Billing records and imaging reports were used to identify children (<= 18 years old) evaluated for nephrolithiasis by both US and unenhanced CT within 24 h between March 2012 and March 2017. Imaging reports were reviewed for presence, number, size and location of kidney stones. Diagnostic performance of US (reference standard=CT) was calculated per renal unit (left/right kidney) and per renal sector (four sectors per kidney). For sector analysis, US was considered truly positive if a stone was identified at CT in the same or an adjacent sector. Results There were 68 renal stones identified by CT in 30/69 patients (43%). Mean patient age was 14.7 +/- 3.6 years, and 35 were boys. For detecting nephrolithiasis in any kidney, US was 66.7% (48.8-80.8%) sensitive and 97.4% (86.8-99.9%) specific (positive predictive value=95.2% [77.3-99.8%], negative predictive value=79.2% [65.7-88.3%], positive likelihood ratio=26.0). Per renal sector, US was 59.7% (46.7-71.4%) sensitive and 97.4% (95.5-98.5%) specific (positive predictive value=72.3% [58.2-83.1%], negative predictive value=95.4% [93.2-96.9%], positive likelihood ratio=22.5). Of the 30 stones not detected by US, only 3 were > 3 mm at CT. Conclusion In clinical practice, US has high specificity for detecting nephrolithiasis in children but only moderate sensitivity and false negatives are common.