The role of magnetic resonance imaging in assessing the extent of tongue squamous cell carcinoma: A prospective cohort study

Abstract

PurposeTo assess the false-positive and false-negative MRI results in evaluating the extent of tongue squamous cell carcinoma. MethodsA prospective cohort series of 165 patients was enrolled to assess the false-positive and false-negative MRI results in evaluating the extent of tongue squamous cell carcinoma by comparing intraoperative tumor profile images and postoperative pathological sections. The differences between two-dimensional tumor margins were analyzed using Mimics 15.0 and Geomagic Control 16.0. A paired-samples t-test was used to analyze the agreement among MRI, intraoperative and pathological findings regarding the extent of tongue tumors. Multiple linear regression analysis was used to analyze associated factors. ResultsThe mean and maximum false-positive values of pathological specimens was 1.95±1.39 mm (95% limit of agreement (LoA) 1.70–2.14) and 3.21 mm, respectively; the false-negative value was 0.44±0.49 mm. The false-positive value of intraoperative specimens was 1.52±0.87 mm (95% LoA 1.36–1.64); the false-negative value was 0.35±0.20 mm. Tumor morphology (ulcer type) (p<0.01) and depth of invasion (DOI) (≤5 mm) (p<0.01) were significantly correlated with the false-positive values of intraoperative and pathology specimens. ConclusionThe false-positive values are important when judging the invasion margin of tongue cancer and forming MRI-based operative plans; the false-negative value was almost negligible.