Preoperative computed tomography findings predict surgical resectability of thymoma. Academic Article uri icon

Overview

abstract

  • INTRODUCTION: The aim of the study was to identify preoperative computed tomography (CT) imaging characteristics that correlated with surgical resectability. METHODS: We retrospectively reviewed the CT scans of 133 patients who underwent surgical resection for thymoma at our institution between July 21, 1997, and September 22, 2010. Imaging characteristics recorded included tumor size, attenuation, contact of mediastinal vessels, tumor morphology, infiltration of surrounding fat, changes in the adjacent lung parenchyma, lymphadenopathy, and pleural involvement. RESULTS: The study group included 66 men and 67 women, aged 23-88 years (mean 58.8 years). Eighty patients (60.2%) were Masaoka stage I or II and 53 (39.8%) were Masaoka stage III or IV. Twenty-three patients (17.3%) had an incomplete surgical resection. Of these, 15 patients had microscopic residual disease (11.2%) and eight had gross residual disease (6.0%). The preoperative CT characteristics that correlated with an incomplete surgical resection included a lobulated tumor contour (p = 0.016), greater than or equal 50% abutment of the circumference of an adjacent vessel (p < 0.001), thoracic lymphadenopathy (p = 0.029), adjacent lung changes (p = 0.005) and pleural nodularity (p = 0.001). Tumor size was larger in the incompletely versus completely resected groups, with mean values of 9.7 and 6.9 cm (p value 0.013). On multivariate analysis, only degree of abutment of adjacent vessels and pleural nodularity were independent predictors of incomplete resection. CONCLUSIONS: Preoperative CT findings can predict the likelihood of successful surgical resection and could help to identify patients who might benefit from neoadjuvant chemotherapy.

publication date

  • July 1, 2014

Research

keywords

  • Thymoma
  • Thymus Neoplasms
  • Tomography, X-Ray Computed

Identity

Scopus Document Identifier

  • 84902507717

Digital Object Identifier (DOI)

  • 10.1097/JTO.0000000000000204

PubMed ID

  • 24926547

Additional Document Info

volume

  • 9

issue

  • 7