Evaluation of cartilage invasion by laryngeal and hypopharyngeal squamous cell carcinoma with dual-energy CT

Abstract

Purpose: To evaluate the clinical usefulness of dual-energy computed tomography (CT) with weighted-average (WA) images and iodine overlay (IO) images in the evaluation of laryngeal cartilage invasion in patients with laryngeal and hypopharyngeal squamous cell carcinoma (SCC).

Materials and methods: The institutional review board approved this retrospective study, and written comprehensive consent was obtained from all patients. Seventy-two consecutive patients underwent 128-section dual-source dual-energy CT to stage laryngeal (n=27) or hypopharyngeal (n=45) cancer. Three observers who were blinded to the patients' clinical histories and histopathologic findings evaluated cartilage invasion on WA images alone or in combination with IO images (nonossified cartilages were selectively evaluated on IO images) by using a five-point scale. Thirty of the 72 patients (42%) underwent surgery, and findings from histopathologic examination in those patients were used as the standard of reference for the evaluation of diagnostic performance with receiver operating characteristic (ROC) curve analysis and in terms of sensitivity and specificity. Interobserver reproducibility was calculated with κ statistics.

Results: For thyroid cartilage, the area under the ROC curve (AUC) of the WA plus IO images was marginally larger than that for WA images alone (AUC=0.957 vs 0.870, respectively; P=.075). The specificity of WA plus IO images was significantly superior to that of WA images alone (96% vs 70%, respectively; P=.031), with no compromise to the sensitivity (86% for both). For thyroid and cricoid cartilages, the interobserver reproducibility was higher for diagnoses made with WA plus IO images (κ=0.68-0.72 and 0.64-0.79, respectively) than for those made with WA images alone (κ=0.29-0.56 and 0.20-0.64, respectively).

Conclusion: Combined analysis of WA and IO images obtained with dual-energy CT improves the diagnostic performance and interobserver reproducibility of evaluations of laryngeal cartilage invasion by SCC.