Abstract Objective Accurate detection of cervical lymph node metastasis in patients with head and neck cancer (HNC) is critical for determining the need for neck dissection. The lymph node-to-primary tumor SUVmax ratio (NTR), derived from ¹⁸Ffluorodeoxyglucose positron emission tomography/computed tomography (¹⁸FFDG PET/CT), is a robust and reproducible imaging biomarker. This study investigated the diagnostic utility of NTR in predicting cervical lymph node metastasis in patients with HNC. Methods This retrospective study included 167 patients with HNC who underwent surgical treatment with neck dissection. All patients underwent 18F-FDG PET/CT and contrast-enhanced head and neck CT/MRI before definitive treatment. The diagnostic performance of these imaging modalities in detecting cervical lymph node metastases was assessed using the pathological results as the reference standard. Clinical and imaging factors associated with nodal metastases were analyzed using univariate and multivariate analyses, and a scoring model was developed based on the independent predictors. Results 18FFDG PET/CT demonstrated superior diagnostic performance over CT/MRI with higher sensitivity (87.5% vs. 77.4%) and specificity (75.7% vs. 68.6%) but showed a suboptimal positive predictive value (PPV) (69.1%) in detecting cervical nodal metastasis. Multivariate analysis identified the maximum SUV of the neck lymph nodes (LNSUVmax, p = 0.049) and NTR (p = 0.038) as independent predictors of cervical nodal metastasis. A two-parameter scoring model was developed by incorporating these predictors. The risk stratification model categorized visually positive cases into high- (PPV 86.8%), intermediate- (PPV 41.1%), and low-risk (PPV 36.3%) groups, indicating its ability to further differentiate the metastatic risk. Conclusions When used as an adjunct to visual interpretation, the NTR-based scoring model aids in better classifying neck nodal metastasis risk in patients with HNC. This approach has the potential to improve diagnostic confidence and reduce unnecessary surgical intervention.
Chi et al. (Tue,) studied this question.