Diagnostic Performance of CT/MRI-Based Lymph Node Morphology With Three-Plane Size Measurements in Oral Cavity Cancer
Mohd Hafiz Johari, Eric Bartlett, Jie Su, Brian O’Sullivan, Ilan Weinreb, Stephen Smith, John de Almeida, Christopher M. K. L. Yao, John Kim, Andrew Bayley, Eugene Yu, Shao Hui HuangPurpose:
This study evaluates the diagnostic value of CT/MRI-based nodal morphology and three-plane size criteria for identifying pathologically positive (pN+) cervical lymph nodes (LNs) in oral squamous cell carcinoma (OSCC).
Methods:
OSCC patients who underwent neck dissection (2020-2021) were included. Preoperative CT/MRI were reviewed by an observer blinded to clinicopathological information. Visible LNs were assessed for morphologic abnormality and size in short-axial (SAD), maximum-axial (MAD), and craniocaudal (CC) diameters. Sensitivity and specificity of each imaging parameter were calculated using pathology as the reference standard.
Results:
CT/MRI scans from 148 patients were analyzed. Sensitivity/specificity for detecting pN+ LNs were: 56%/76% (abnormal morphology-alone), 57%/72% (morphology ± SAD >1.0 cm), 83%/28% (morphology ± MAD >1.0 cm), 86%/29% (morphology ± CC >1.0 cm), and 88%/24% (morphology ± any-plane >1.0 cm). Among 64 morphologically abnormal LNs, 25, 6, and 5 measured ≤1.0 cm in SAD, MAD, and CC, respectively, with pN+ rates of 16/25 (64%), 4/6 (66.7%), and 3/5 (60%). In morphologically “normal” (homogenous) LNs, sensitivity/specificity for pN+ detection by >1.0 cm size threshold were: 3%/95% (SAD), 62%/36% (MAD), 68%/39% (CC), and 72%/32% (any plane).
Conclusions:
Abnormal LN morphology is the most reliable CT/MRI indicator of pN+ disease in OSCC, irrespective of nodal size, although sensitivity remains modest. In morphologically normal LNs, three-plane size assessment, particularly MAD and CC, improves detection of pN+ beyond SAD. These findings support a morphology-first approach to CT/MRI nodal evaluation and inform an exploratory OSCC-specific decision framework incorporating morphology, multi-plane size assessment, and lymphatic drainage patterns.