The first step in assessing thoracic lesions is chest X-ray, but the optional imaging procedure for the final diagnosis is controversial. We aimed to examine the diagnostic accuracy of imaging modalities in pediatric thoracic masses.
Materials and Methods
This prospective cross-sectional study was conducted from 2017 to 2018 in Dr. Sheikh hospital, Mashhad, Iran. A total of 130 patients with a confirmed pathology report of thoracic masses were recruited in this study. A pediatric radiologist independently evaluated the existing chest X-ray (CXR), ultrasound (US) and CT and reported the probable diagnosis. Imaging reports of CXR, US, and CT were compared with the pathology results.
83 (63.8%) of the patients were boys with the mean age of 72.15 + 46 months. The most prevalent site of the thoracic masses was the lung parenchyma with the frequency of 81 (62.3%), and the most frequent mass was hydatid cyst with the frequency of 57 (43.8%). Thoracic CT had the overall sensitivity of 100% for mass localization and 78.2% for nature determination; while US had the sensitivity of (95.4%) for mass localization and 90.9% for the diagnosis of mass nature. The sensitivity of CXR for thoracic mass localization was 89.4 and for mass nature determination was 35.5%.
Based on the results, CXR and US had a similar appropriate sensitivity in localization of thoracic masses. Although CT had the highest overall sensitivity for mass localization, in comparison with US, it was less diagnostic to define mass nature and US had the highest sensitivity for mass nature determination. Hence, US may potentially obviate further imaging such as CT in most of the cases.