F-096DOES VIDEO-ASSISTED MEDIASTINOSCOPIC LYMPHADENECTOMY FULFIL CRITERIA TO BECOME A STANDARD PROCEDURE FOR EARLY STAGE LUNG CARCINOMA?

Abstract

Video-assisted mediastinoscopic lymphadenectomy (VAMLA) is a procedure for complete bilateral resection of mediastinal lymph nodes in stations 4, 10 and 7. Here, we present recent data on the use of VAMLA for mediastinal lymph node staging in early stage lung carcinoma. We evaluated the end-points disease free survival (DFS), overall survival (OS) and additionally stratified according to risk factors. Between 2005 and 2011, a total of 76 patients with NSCLC in UICC stage IA-IIB underwent VAMLA followed by thoracoscopic or open lobectomy. Data were compared retrospectively with 54 patients with NSCLC (same stages) who received thoracoscopic lobectomy followed by unilateral lymphadenectomy (control group, no prior mediastinoscopy). There was no significant difference between groups concerning age, gender, histology or involvement of the mediastinal lymph nodes (pN0/1). DFS was significantly lower in the VAMLA group (P = 0.004) than in the contol group, and also significantly lower in VAMLA subgroups with tumours up to 5 cm diameter (P = 0.02), with squamous cell carcinoma (P = 0.007), or with patients over 75 years of age (P = 0.021). OS was not different between the VAMLA and control groups (P = 0.06). In total, involvement of mediastinal lymph nodes was confirmed in 8 patients (pN1). VAMLA detected nodal involvment (pN1) in 3 patients, which had been radiologically classified as cN0, and confirmed lymph node metastases in 2 patients with cN1 classification. Similarily, in the control group, 3 patients were found pN1 that were originally classified cN0. Relapse of mediastinal lymph node metastases was observed in VAMLA group in 2.6% (n = 2) and 5.6% (n = 3) in control group (P = 0.46). The present data indicate that VAMLA patients have lower DFS than patients with unilateral VATS-lymphadenectomy. However, the statistical power of the present study is limited due to small patient numbers. Based on these observations it seems questionable that VAMLA fulfils criteria for a standard procedure. All authors have declared no conflicts of interest.