Role of Chest CT Scan to Predict Malignancy on Mediastinal Mass

Saraswati Widyaningrum, Anita Widyoningroem, Anggraini Dwi Sensusiati, Etty Hary Kusumastuti

Abstract


Mediastinal mass is becoming a global health problem due to high mortality. The heterogeneous mediastinal components make the symptoms of mediastinal mass diverse. CT scans are still the imaging modality for examining mediastinal mass before surgery or other therapies. In this study, we evaluate whether a CT scan could predict the malignancy of mediastinal mass, which is expected to help establish a pre-surgical or pre-biopsy diagnosis. Sixty-two samples were taken consecutively from mediastinal mass patients who came to Dr. Soetomo General Academic Hospital to undergo a CT scan of the thorax with contrast and histopathology examination (core biopsy or open biopsy), which was carried out in the period from December 2019 to March 2024. The results of the CT scan imaging used in this study variable include mass location, mass shape, mass size, infiltration with surrounding organs, attenuation values before contrast administration, after contrast administration, and additional attenuation before and after contrast administration. The CT scan and histopathology results were compared, and multivariate analysis was performed to obtain predictor factors. The location of the mediastinal mass (anterior, medius, posterior), the solid heterogeny component, cystic, calcification, mass shape, organ infiltration, and contrast enhancement value could significantly predict the mediastinal mass's malignancy. If obtained simultaneously, the organ infiltration and contrast enhancement value >20 HU can indicate whether a mediastinal mass is malignant with a specificity of up to 100%.

Keywords


Chest CT scan; histopathology; mass; mediastinum



DOI: https://doi.org/10.29313/gmhc.v13i1.14218

pISSN 2301-9123 | eISSN 2460-5441


Visitor since 19 October 2016: 


Free counters!


Global Medical and Health Communication is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.