The Role of PET-CT in Mediastinal Lymph Node Staging in Non-small Cell Lung Cancer

Aim: The aim of this study was to determine the overall accuracy, sensitivity and specificity of both CT and PET-CT in identifying and classifying the presence of metastatic intra-thoracic lymph nodes and their potential role in the treatment of lung cancer. Material and methods: Forty four (mean age 60.7, male/female ratio: 37/7) patients with potentially resectable non-small cell lung cancer have been enrolled. All patients underwent flexibl bronchoscopy, CT scanning and PET-CT imaging prior to mediastinoscopy. Mediastinostomy was performed in two patients. Thirty three patients with negative mediastinoscopy and three patients after neoadjuvant chemotherapy, total of 37 patients went through thoracotomy. Results: The overall sensitivity, specificity and accuracy of CT scan in preoperative staging were found 66.7%, 62.1%, 63.6%, respectively. The negative and positive predictive values of CT scan were 78.3% and 47.6%. CT imaging correctly identified 29 patients (65.9%). Seven patients (15.9%) were under-staged while eight patients (18.1%) were over-staged. Based on PETCT findings, overall sensitivity, specificity and accuracy for staging mediastinal lymph nodes were 86.7%, 65.5% and 72.7%, respectively. The negative and positive predictive values of PET-CT were as follows; 90.5% and 56.5%. Overall, interpretations based on PET-CT staged 65.9% (29 patients) correctly, down-staged six patients (13.6%) and over-staged nine patients. The false-negative rate was 13.3% while the false-positive rate was 34.5%. Conclusion: CT scan and PET-CT perform similarly in the mediastinal staging of non-small cell lung cancer (p= 0.05, Mc Nemar test), both tests are conditionally dependent and provide complementary information, and their diagnostic value mainly resides on the negative results. Although, statistically insignificant, PET-CT has a superior sensitivity and better negative predictive value over CT scan. Based on the relationship between SUV-max value and characteristics of tumor, a significant increase in SUV-max value appeared to be correlated with the tumor size and squamous cell carcinoma sub-type.