Objective: To compare the diagnostic yield of the transcervical extended mediastinal lymphadenectomy (TEMLA) and the cervical mediastinoscopy (CM) in detecting metastatic mediastinal lymph nodes in NSCLC patients.
Methods: Prospective, randomized, single-blind clinical study.
Results: There were 41 NSCLC patients enrolled in the study; 21 were randomized to the TEMLA group and 20 to the cervical mediastinoscopy group. The TEMLA revealed mediastinal metastases in 7 patients, and mediastinoscopy in 3. In the TEMLA group one patient out of the 14 with negative nodes was finally found unfit for surgery, and in the remaining 13 lung resections with mediastinal dissection were performed. In the mediastinoscopy group one patient out of the 17 with negative nodes was finally found unfit for surgery and another one refused surgery, so in 15 of them lung resections with mediastinal dissection were performed. In no patient in the TEMLA group did the pathological examination of the operative specimen reveal metastatic lymph nodes, whereas in the mediastinoscopy group metastatic nodes were found in 5 patients. The number of false negative results was significantly greater in the mediastinoscopy group (5 vs 0, p=0.019), and the difference was the reason for terminating the randomization before reaching the initially planned number of 100 patients. The sensitivity of mediastinoscopy was 37.5% and its negative predictive value was 66.7%, compared to 100% and 100% in the TEMLA group. The comparison of the time of the operation, blood loss, complications, postoperative pain and the use of analgetics has shown significant differences between groups only regarding the operative time and the pain intensity, being greater in the TEMLA group.
Conclusions: 1. The sensitivity and the NPV of the TEMLA in detecting mediastinal metastases in NSCLC are significantly greater than those of cervical mediastinoscopy. 2. The invasiveness of TEMLA and mediastinoscopy does not significantly differ, except for the postoperative pain.