BACKGROUND AND OBJECTIVEApproximately 30% of patients who are diagnosed with non-small cell lung cancer (NSCLC) are classified as N2 on the basis of metastasis to the mediastinal lymph nodes. The effectiveness of surgery for these patients remains controversial. Although surgeries in recent years are proved to be effective to some extent, yet due to many reasons, 5-year survival rate after surgery varies greatly from patient to patient. Thus it is necessary to select patients who have a high probability of being be cured through an operation, who are suitable to receive surgery and the best surgical methods so as to figure out the conditions under which surgical treatment can be chosen and the factors that may influence prognosis.

METHODS165 out of 173 patients with N2 NSCLC were treated with surgery in our department from January 1999 to May 2003, among whom 130 were male, 43 female and the sex ratio was 3:1, average age 53, ranging from 29 to 79. The database covers the patients' complete medical history including the information of their age, sex, location and size of tumor, date of operation, surgical methods, histologic diagnosis, clinical stage, post-operative TNM stage, neoadjuvant treatment and chemoradiotherapy. The methods of clinical stage verification include chest X-ray, chest CT, PET, mediastinoscopy, bronchoscope (+?), brain CT or MRI, abdominal B ultrasound (or CT), and bone ECT. The pathological classification was based on the international standard for lung cancer (UICC 1997). Survival time was analyzed from the operation date to May 2008 with the aid of SPSS (Statistical Package for the Social Sciences) program. Kaplan-Meier survival analysis, Log-rank test and Cox multiplicity were adopted respectively to obtain patients' survival curve, survival rate and the impact possible factors may have on their survival rate.

RESULTSThe median survival time was 22 months, with 3-year survival rate reaching 28.1% and 5-year survival rate reaching 19.0%. Age, sex, different histological classification and postoperative chemoradiotherapy seem to have no correlation with 5-year survival rate. In all N2 subtypes, 5-year survival rate is remarkably higher for unexpected N2 discovered at thoractomy and proven N2 stage before preoperative work-up and receive a mediastinal down-staging after induction therapy (P < 0.01), reaching 30.4% and 27.3% respectively. 5-year survival rate for single station lymph node metastasis were 27.8%, much higher compared with 9.3% for multiple stations (P < 0.001). Induction therapy which downstages proven N2 in 73.3% patients gains them the opportunity of surgery. The 5-year survival rate were 23.6% and 13.0% for patients who had complete resection and those who had incomplete resection (P < 0.001). Patients who underwent lobectomy (23.2%) have higher survival rate, less incidence rate of complication and mortality rate, compared with pneumonectomy (14.8%) (P < 0.01). T4 patients has a 5-year survival rate as low as 11.1%, much less than T1 (31.5%) and T2 (24.3%) patients (P = 0.01). It is noted through Cox analysis that completeness of resection, number of positive lymph node stations and primary T status have significant correlativity with 5-year survival rate.

CONCLUSIONIt is suggested that surgery (lobectomy preferentially) is the best solution for T1 and T2 with primary tumor have not invaded pleura or the distance to carina of trachea no less than 2 cm, unexpected N2 discovered at thoractomy when a complete resection can be applied, and proven N2 discovered during preoperative work-up and is down-staged after induction therapy. Surgical treatment is the best option, lobectomy should be prioritized in operational methods since ise rate of complication and morality are lower than that of pneumonectomy. Patients' survival time will not benefit from surgery if they are with lymph nodes metastasis of multiple stations (Bulky N2 included) and T4 which can be partially removed. Neoadjuvant chemotherapy increases long-term survival rate of those with N2 proven prior to surgery. However, postoperative radiotherapy decreases local recurrence rate but does not contribute to patients' long-term survival rate.