The comparison of wall thickness of esophagus and gastroesophagial junction measured in computed tomography with endoscopy and / OR biopsy results

Aim: In this retrospective study, we aimed to detect a possible cut-off value in esophagus wall thickening which is measured on CT scans for the discrimination of benign and malign processes in the patients who underwent to endoscopy and/or biopsy procedures. Material and methods: Between January 2015 and June 2018, a total of 144 patients (61 male, 83 female) aged between 24-86 years with a thickness of esophagus/gastroesophagial junction on computerized tomography were included in the study. All of these patients underwent to endoscopy but biopsies were taken in 69 of them. According to the provisional diagnosis of the patient, tomography images taken in various protocols were examined by two radiologists and wall morphology, thickness, anatomical localization and accompanying findings were examined. Benign and malignant patients were determined according to endoscopy and / or biopsy results. We used a ROC analysis for the detection of a cut-off wall-thickening value for discrimination of benign and malign processes. ROC analysis was performed to determine the cut-off value of ratio of lesion level thickness to normal segment thickness for discrimination of benign and malign processes. Chi-square test was also performed in order to detect asymmetry in benign and malign cases. Statistical analysis was performed using SPSS 22 statistical software program.. Results: There was a statistically significant difference in the wall thickness between the cases with esophageal cancer and benign conditions. When the cut-off value was 13.5mm for the wall thickness, the sensitivity was determined as 94.3% and the specificity was 100%. The ratio of the thickness of the lesion to the normal segment thickness was statistically significant in the malignant-benign differentiation. A significant relationship was found between asymmetric thickening and malignancy. The most common histological type of esophageal cancer was SCC (79.2%) and adenocarcinoma (13.3%). The most common location was the distal segment of the esophagus. Most of the SCCs were seen in the middle and distal segments of esophagus and all of the adenocarcinomas were seen in distal segments of esophagus. Most of the benign lesions were seen in distal segments of esophagus. The most common diagnosis was esophagitis. Conclusion: We think that CT will contribute to early diagnosis of esophageal cancer because of its highly prevalence in our region. Key words: Esophageal Cancer, Esophageal Wall Thickness, Computerized Tomography (CT)