Algorithm of preoperative N-staging of non-small cell lung cancer

Objectives: Accurate preoperative N-staging is a crucial point for management of non-small cell lung cancer (NSCLC). Aim of our study was to work out an algorithm of N-staging based not only on diagnostic value of different methods but also on long-term survival. Material and Methods: 365 patients with potentially resectable NSCLC were included in a prospective study from 2003 till 2009. Patients were divided into two groups according to preoperative staging: group 1 (n=190) where both videomediastinoscopy (VM) and non-invasive procedures (CT, PET, evaluation of CYFRA 21-1) were used and group 2 (n=175), where only non-invasive methods were included in N-staging plan. Patients without preoperatively proven N2/N3 (n=324) underwent lung resections with systematic lymph nodes dissection. Results: Specificity, sensitivity, accuracy, PPV and NPV of VM (100%, 76%, 92%, 100% 90%) were significantly higher than those of CT (66%, 57%, 63%, 43%, 78%). High NPV and accuracy of PET (89% and 83%) and CYFRA 21-1 (86% and 78%) allow to exclude mediastinal metastasis in patients with peripheral cancer. The rate of postoperative N2 was lower in group 1 than in group 2 (10% vs 27%, p Conclusions: Proposed algorithm improves the accuracy of preoperative N-staging, that results in better long-term survival of operated patients.