Radical surgical resection offers the best chance of survival for up to stage III non-small cell lung cancer (NSCLC). In lung cancer, frozen section analysis of bronchial resection margins is performed with the option of extending the surgical resection to maximize the chance of local control and thereby cure. The outcome of the frozen section analyses has thus far been categorized as: 'resection margins free' (no residual tumour ¼ R0) or 'resection margins tumour positive'. The presence of residual tumour in the final resection margin is further divided into R1 for microscopic and R2 for macroscopic residual tumour. 1 In older studies incomplete resection margins are described in 4.1-14.7% of pulmonary resections. [2][3][4] Incomplete resection was associated with an unfavourable prognosis due to local recurrence. Since intraoperative macroscopic examination does not always detect the presence of residual tumour, frozen section analysis has been advocated for bronchial resection margins. In more recent literature the frequency of incomplete bronchial resection margins lies between 1.2 and 5.4% (mean 3.6%). 5-11 The type of surgery does not seem to correlate to the risk of R1 resection status, except possibly for sleeve resection. [12][13][14][15] Follow-up information reveals that a tumour-positive resection margin is an adverse prognostic factor in many studies. 1-3,5-8,13 A critical methodological note may be that only part of these studies report information on the control arm. 1,6,7,9 In case of an R1 resection and the absence of lymph node metastases (N0 disease), recurrences are more frequently local than at distant sites, whereas in cases with mediastinal lymph node metastases (N2 disease) the opposite is the case. 12,13 However, not all studies report a prognostic difference between R0 and R1 status. 10,16,17 Explanations for publications that do not find differences between R0 and R1 resections may be small numbers where trend is available, and failure to consider morphological subgroups. 17 Frozen section analysis is indicated if, in the preoperative work-up, the tumour is shown to be in close proximity to the bronchial resection margin or if the surgeon considers the resection margin at risk during the operation. In sleeve resections this is practically always the case. For a distance of 15 mm it was estimated that 93% of all non-small cell carcinomas would have a free resection margin. 18 Therefore, it is reasonable to suggest that frozen section analysis should be restricted to those cases in which the preoperative evaluation has shown that the distance of tumour to the supposed bronchial resection margin is 20 mm or less. 18,19 Only an exceptional case will then turn out to have a positive resection margin. 2,18 Remarkably, preoperative data from X-ray, computed tomographic scan analysis or bronchoscopy on this subject are lacking.In several studies a distinction is made in the anatomical localization of tumour tissue in the bronchial resection margins of non-small cell carcinoma. However, the empl...