Risk factor analysis of mediastinal lymph node metastasis in non-small cell lung cancer patients and the strategy of mediastinoscopy prior to surgery / 中华肿瘤杂志

ABSTRACT

<p><b>OBJECTIVE</b>To discuss the strategy of mediastinoscopy for the evaluation of mediastinal lymph node status (metastasis or not) of non-small cell lung cancer (NSCLC) prior to surgery.</p><p><b>METHODS</b>From October 2000 to June 2007, 152 consecutive NSCLC cases pathologically proven and clinically staged I-III were enrolled in the study. Of the 152 cases, there were 118 males and 34 females. Age ranged 24-79 years old and the median age was 58. All cases underwent CT and mediastinoscopy for the evaluation of mediastinal lymph node status prior to surgery. Compared with the results of final pathology, the positive rate of mediastinoscopy and the prevalence of mediastinal lymph node metastasis were calculated in the NSCLC patients with negative mediastinal or hilar lymph nodes on CT scan (the shortest axis of mediastinal or hilar lymph nodes <1 cm). Clinical characteristics used as predictive factor including sex, age, cancer location, type of pathology, T status, cancer type (central or peripheral), size of mediastinal lymph nodes (the shortest axis <1 cm or >1 cm) on CT scan and serum CEA level were analyzed by univariate and multivariate analysis with Binary logistic regression model to identify risk factors of mediastinal metastasis.</p><p><b>RESULTS</b>The positive rate of mediastinoscopy was 11.6% (8/69) and the prevalence of mediastinal metastasis was 20.1% (14/69) in NSCLC with negative mediastinal or hilar lymph nodes on CT scan respectively. In clinical stage I (cT1-2N0M0) NSCLC the positive rate of mediastinoscopy was 11.3% (7/62), N2 accounting for 6.5% (4/62) and N3 4.8% (3/62), respectively; and the prevalence of mediastinal lymph node metastasis was 19.4% (12/62), N2 ccounting for 14.6% (9/62) and N3 4.8% (3/62), respectively. In the whole group both univariate and multivariate analysis showed that adenocarcinoma or mediastinal lymph nodes > or =1 cm in the shortest axis on CT scan was an independent risk factor to predict mediastinal lymph node metastasis. In NSCLC with negative mediastinal or hilar lymph nodes on CT scan both univariate and multivariate analysis showed that adenocarcinoma was a predictor of mediastinal lymph node metastasis. Conclusion We recommend the policy of routine mediastinoscopy in NSCLC prior to surgery if the mediastinal staging was only based on CT scan. Mediastinal lymph nodes > or =1 cm in the shortest axis on CT scan mandates preoperative mediastinoscopy. Adenocarcinoma also indicates mandatory mediastinoscopy even with negative mediastinal or hilar lymph nodes on CT scan.</p>