Objectives : Despite technical difficulties due to mediastinal fibrosis, remediastinoscopy (reMS) can be a valuable tool in the restaging of non-small cell lung cancer (NSCLC) after neoadjuvant therapy. It may provide essential pathological information on mediastinal downstaging when selecting patients (pts.) for subsequent surgical resection. The aim of our study was to evaluate the feasibility, sensitivity and accuracy of reMS.
Material and methods : From November 1994 to May 2001 we performed a reMS in 27 pts. (24 men, 3 women) after neoadjuvant therapy for NSCLC. Their age ranged from 35 to 80 years (mean 61.9 ± 11.9). Neoadjuvant therapy consisted of cisplatin-based chemotherapy in 24 pts. and chemoradiotherapy in 3 pts.
Results : In all 27 patients it was possible to perform a reMS without major technical difficulties and take biopsies of the lymph nodes that were initially invaded by tumor. ReMS was positive in 11 pts. (40.7 %) and negative in 16 pts. (59.3 %). In the 11 pts. with a positive reMS an unnecessary thoracotomy was avoided.
In 4 pts. reMS turned out to be false negative. So, in our series, sensitivity was 73 %, specificity 100 %, and accuracy 85 %. The positive and negative predictive values were 100 % and 75 %, respectively.
Conclusions : Previous mediastinoscopy is no contra-indication for a repeat one after neoadjuvant therapy. By offering valuable pathological information, reMS is an important restaging tool as persisting mediastinal involvement carries a poor prognosis in stage IIIA and IIIB NSCLC. Although sensitivity and accuracy are lower than that of a first mediastinoscopy, reMS is useful to select patients for surgical resection after induction therapy.