Central tumour location should be considered when comparing N1 upstaging between thoracoscopic and open surgery for clinical stage I non-small-cell lung cancer

Abstract

OBJECTIVES: Nodal upstaging is a quality indicator for oncological thoracic surgery and is found in up to 25% of patients with clinical stage I (cStage-I) non-small-cell lung cancer (NSCLC). In large retrospective series, lower N1 upstaging was reported after video-assisted thoracic surgery (VATS) resections. We studied the impact of central primary tumour location on nodal upstaging in cStage-I NSCLC. METHODS: Consecutive patients operated for cStage-I NSCLC were selected from a prospectively managed surgical database. Tumour location was classified as central if the lesion was visible during standard video bronchoscopy. A nodal station mapping was drawn for each patient based on final pathological examination. Univariable and additive multivariable binary logistic regression analyses were performed. RESULTS: Between 2007-2014, 334 patients underwent anatomical resection for cStage-I NSCLC, either by open thoracotomy (n = 158) or by VATS (n = 176; conversion rate 1.7%). All patients underwent imaging with [18F]-fluorodeoxyglucose positron emission tomography and computer tomography. Invasive mediastinal staging was performed in 24.6% of patients. There were more central tumours in the open group (24.1%, n = 38) compared with the VATS group (4.5%, n = 8). There was no significant difference between the number (mean ± standard deviation) of nodal stations examined (open 5 ± 1.9 vs VATS 5 ± 1.7, P = 0.99). Pathological nodal upstaging was found in 15.9% (n = 53) of cStage-I patients. Nodal pN1 and pN2 upstaging were 13.3 and 8.2%, respectively, for the open group, and 6.3 and 4.5%, respectively, for the VATS group. In 32.6% (n = 15/46) of patients with a central cStage-I tumour pN1, upstaging was found. A binary logistic regression model (including tumour location, technique, tumour size, gender and histology) showed that only tumour location had a significant impact on pN1 upstaging [peripheral versus central; odds ratio (OR) 5.07 (confidence interval, CI: 1.89-13.60), P = 0.001], while surgical technique had no significant impact [VATS versus open; OR 0.74 (CI: 0.31-1.78), P = 0.50]. CONCLUSIONS: The number of lymph node stations examined during VATS resections is similar to open resections for cStage-I NSCLC. Almost one-third of the patients with a central cStage-I NSCLC were upstaged to pN1. Tumour location was the only independent variable for pN1 upstaging in logistic regression analysis. It is a potential bias in retrospective studies and should therefore be accounted for when comparing different surgical resection techniques for cStage-I NSCLC.