Lymphnode dissection

Since the advent of surgical resection for lung cancer by Graham's successful pneumonectomy and the subsequent demonstration that an anatomical lobectomy was sufficient for a curative lung cancer procedure the removal of regional and mediastinal lymph nodes has been thought to be an integral part of surgical therapy of lung cancer. The initial impetus for this was from extrapolation of resections of other solid organ tumors, such as a lymphadenectomy with a gastrectomy for stomach cancer and a axillary lymph node dissection as part of a mastectomy for breast cancer. The practice of including routing lymphadenectomy as a part of lung cancer resection was solidified by two reports from Memorial Sloan Kettering Cancer Center stating that mediastinal lymph node dissection (MLND) leads to a more favorable long-term survival. Unfortunately, a routine lymph node dissection is not performed in the majority of lung cancer resections. Little and colleagues found that only 48.1% of patients undergoing a lung cancer resection had even one mediastinal lymph node sampled at the time of surgery. MLND is the gold standard for staging the mediastinum. Other techniques such as computed tomography, proton emission tomography, endoscopic bronchial ultrasound or even mediastinoscopy all have a significant false negative rate and can not be considered as the 'best “ method to accurately stage the mediastinum. Whether or not MLND increases survival rate is a separate question. Subgroup analysis of Intergroup trial 0115 by Keller et al did show improved long-term survival in a group of patients with right upper lobe tumors that underwent MLND compared to a group that just underwent sampling; however, this was not a randomized trial and all the patients had positive N1 or N2 lymph nodes. In a randomized trial reported by Izbicki and colleagues in 1998, they were not able to demonstrate a difference in survival between lung cancer patients undergoing resection that had a MLND and those that just had sampling of suspicious lymph nodes. After a follow- up of 47.5 months they found the disease free and overall survival rates were the same in the two groups. In another study by Sugi et al studying just patients with small (<2cm) T1 non-small cell lung cancers they also found no difference in survival between patients that had a complete MLND and those that had sampling alone. The only positive study has been reported by Wu et all where they examined 532 patients and included stages I, II and III in the study population. There was an improvement of 11 months in the median survival in patients that had a MLND compared to those that just had sampling of the mediastinal lymph nodes. The recently reported Z0030 study from the American College of Surgery Oncology group reported on 1,111 patients with T1 or T2 tumors that were clinically N0 or non-hilar N1 non-small cell carcinomas of the lung and were randomized to either mediastinal node sampling (removal of a single node from stations 2R, 4R,7 and 10R for right sided tumors and 5,6,7 and 10L for left sided tumors) or to MLND. The randomization yielded a very similar group of patients in the two groups, the only demographic difference was that patients in the MLND arm had a median age 1 year older than the sampling arm. The patients were followed for a median of 6.5 years and there was no difference in the survival between the two groups. of note, there were 21 (4%) patients that were discovered to have at least one positive mediastinal lymph node when they were randomized to MLND after having undergone initial mediastinal lymph node sampling. These patients were all carefully staged preoperatively as well with CT, PET and careful history and physical examinations. It is also important to note that the performance of a complete MLND did not increase the incidence of lymph leaks, bleeding, bronchial fistulae or other complications, but did extend the length of the operation by about 15 minutes. The conclusion from the Z0030 trial is that MLND does not improve long term survival or decrease local recurrence in patients with an early stage lung cancer. When clinical staging, intraoperative sampling and visual inspection do not reveal any mediastinal or hilar involvement, a complete MLND is not beneficial for patients with an early stage lung cancer. This should not be extrapolated to patients with more advanced T3 or hilar N1 cancers. Since the trial specifically excluded these patients, no conclusions can be made concerning this group of patients. In this instance, a complete MLND would seem prudent until further randomized trials clarify the situation.