Reduced-Dose Computed Tomography vs Multiview Radiographs in Pediatric Foreign Body Aspiration.

Abstract

To compare the overall safety, efficacy, and resource utilization of ultra-low-dose computed tomography (ULDCT) with multiview chest x-ray (CXR) in children presenting with suspected foreign body aspiration (FBA). A single-center, retrospective chart review was conducted for patients ≤17 years of age with concern for FBA and presenting between January 1, 2018, and July 31, 2024. Computed tomography (CT) studies were categorized as standard, low, and ultra-low dose, based on reported CT dose index volume. Measures of diagnostic performance, estimated effective doses, and total variable costs per patient encounter were compared between imaging modalities. Three hundred thirty patients were evaluated for FBA with CXR only (n = 171), CT only (n = 47), or both modalities (n = 112). One hundred thirty-two patients underwent bronchoscopy with 107 foreign bodies discovered. Negative bronchoscopy rates ranged from 11% (n = 1) in patients evaluated with CT only to 21% (n = 18) in patients evaluated with CXR only. Seven patients evaluated with CXR only were later found to have missed airway foreign bodies. The sensitivity and specificity of ULDCT for tracheobronchial foreign bodies were 100%, respectively. Sensitivity and specificity were 74% and 83% for two-view CXR and 82% and 67% for four-view CXR. Mean effective dose of ULDCT was 0.18 mSv higher than four-view CXR (P < .001). Median total variable costs were similar among patients initially evaluated with CXR and CT (P = .21). Reduced-dose CT showed greater sensitivity and specificity than CXR to diagnose FBA and can be delivered without a clinically significant increase in radiation exposure or hospital expenditure.