There remains controversy on the prognostic value of several common cl
inical factors in NSCLC patients with resected N2-disease. The aim of
this paper is to give a comprehensive overview of the available data o
n this issue. Literature data on surgically treated N2-NSCLC-patients
from 1980-1995, peer reviewed and listed in Index Medicus, were analys
ed. Reported and calculated or estimated survival data were indexed. E
ighteen series were selected: in 12 of them, direct comparisons betwee
n survival curves of subgroups are reported; six contained sufficient
data to make comparisons of survivors at 5 years; three of them also m
ade a multivariate Cox model. The analysis of prognostic factors in a
single study was often hampered by the limited number of patients. Non
etheless, it could be concluded that patients with a clinical N0- or N
1-status (so-called unforeseen N2) do better. There was no clear diffe
rence between patients undergoing lobectomy or pneumonectomy. There wa
s strong evidence that N2-patients with a less advanced primary tumour
(T-stage) have a better prognosis, and this is the case for all opera
ble T-stages (T1 versus T2, T1 versus T3, T2 versus T3). Squamous cell
type was a favourable prognostic factor, as was the presence of only
one metastatic mediastinal lymph node station or absence of metastases
to the subcarinal nodes. There was some evidence that the presence of
extracapsular spread in metastatic MLN is an unfavourable finding. St
ratification for these prognostic factors could help in the planning o
f future trials on combined modality treatment in N2-NSCLC. (C) 1998 E
lsevier Science Ireland Ltd.