Methods: Between January 2006 - September 2009, 302 patients with histopathologically confirmed NSCLC were prospectively analyzed and 264 patients (30 females, 234 males; mean age 58±9.3 years; range 27 to 84 years) who underwent PET-CT for treatment or staging were included. Lymph nodes with standardized uptake value (SUVmax) ≥2.5 were considered malignant. Cervical mediastinoscopy was performed for suspicious lymph node involvement of 2R, 4R and 7 nodal stations. Video-assisted thoracoscopy was performed for suspicious lymph node involvement of 5, 6 and contralateral 10, 11 nodal stations. Through right-sided thoracotomy 2,4,7,8,9,10,11 nodal stations were samples, while 5,6,7,8,9,10,11 nodal stations were sampled by the left-sided thoracotomy. 1997 6th version of staging system were used.
Results: Squamous cell carcinoma was reported in 57.8%, adenocarcinoma in 38%, large cell carcinoma in 3%, and carcinoid tumor in 1.1%. The false positive rate of PET-CT was 26.2%, whereas the false negative rate was 13.1%. The overall sensitivity, specificity, negative predictive value, positive predictive value and accuracy of PET-CT for detecting mediastinal lymph node involvement were 76.8%, 86.6%, 73.7%, 86.7% and 81.8%, respectively. The diameter of lymph nodes and the presence of N1 disease were statistically significant in terms of the accurate staging of mediastinal lymph nodes through PET-CT (p<0.001, p=0.002).
Conclusion: Positron emission tomography-CT is required for the clinical staging of NSCLC on a regular basis and in the presence of lymph nodes of >1 cm, we believe that invasive staging is beneficial for the patients with N1 positive, even if SUVmax value at N2 stations was below the malignancy threshold, as indicated in the PET-CT scan.