Complete mediastinal lymph node dissection—does it make a difference?

Abstract

Thoracic surgeons have cataloged and categorized the intrathoracic lymph nodes of patients with nonsmall cell lung cancer (NSCLC) for the past five decades. The anatomic locations in which these lymph nodes were consistently found were labeled with letters or numbers. The currently accepted lymph node level definitions and the N category to which they are assigned [1] are based on the work of Naruke [2]. Since Cahan first described mediastinal lymph node dissection following pulmonary resection numerous techniques for assessing the intrathoracic lymph nodes have been detailed [3]. These range from simple biopsy of abnormal appearing lymph nodes to radical resection of all mediastinal and supraclavicular lymphatic tissue. Disparate claims have been made regarding the accuracy of these techniques in identifying mediastinal lymph node metastases and their effect on patient survival. The lack of uniform terminology has compounded the difficulty in assessing the efficacy of these techniques. The term sampling should be utilized when only those lymph nodes that were obviously abnormal to visual or tactile inspection were removed. Systematic sampling (SS) refers to routine biopsy of lymph nodes at levels specified by the author. Complete mediastinal lymph node dissection (CMLND) indicates that all ipsilateral lymph node-containing tissue was routinely removed at those levels indicated by the investigators. Radical lymph node dissection refers to resection of all ipsilateral mediastinal lymph nodes at those levels indicated by the investigators, as well as, contralateral mediastinal or supraclavicular nodal tissue. Why should CMLND be effective? It is likely that within the spectrum of NSCLC there exists a cohort of patients with metastatic disease that is truly limited to the regional lymph nodes. These are the patients who will benefit from aggressive resection of the intrathoracic lymph nodes. Wu and colleagues have produced an important contribution to the evaluation of CMLND [4]. The intraoperative details of this phase III trial were carefully designed and thoroughly documented. The data demonstrate significantly improved survival following CMLND for stages I and IIIa NSCLC and borderline significance in patients with stage II NSCLC. The authors failed to include information regarding perioperative therapy (neoadjuvant or adjuvant). However, the randomization process should equalize the effect (if any) of these treatments. Prior to the current publication only two randomized prospective trials have compared SS to CMLND. Izbicki randomized 201 with operable NSCLC to either SS or CMLND [5]. All patients found to have metastases to the N2 lymph nodes received adjuvant radiotherapy to the mediastinum. Patients with T3 or T4 tumors received adjuvant radiotherapy to the tumor bed. Thirty-two patients were excluded from the analysis because of residual tumor (n=12), N3 disease (n= 5), M1 disease (n=5), or small cell lung cancer (n=10). Analysis of the 169 remaining patients demonstrated both a disease free survival and overall survival advantage for the 46 patients with N1 and limited N2 disease who underwent CMLND. No survival difference was observed for patients with N0 disease. A retrospective survival analysis of 100 of these patients was conducted after determining the presence or absence of lymph node micrometastases utilizing the Ber-Ep4 antibody [6]. This immuno-histochemical technique detected the presence of micrometastases in an additional 23% of patients. When all 100 patients were analyzed as a group, no survival advantage was demon* Tel.: +1-718-920-7580. E-mail address: skeller@montefiore.org (S.M. Keller). 1 Current address: Department of Cardiothoracic Surgery, Montefiore Medical Center, 3400 Bainbridge Ave., Bronx, New York, NY 10467, USA