OBJECTIVES Intraoperative conversion may be required during video-assisted thoracoscopic surgery (VATS) for lung cancer. We evaluated the morbidity and mortality rates associated with VATS for anatomical pulmonary resection with conversion to thoracotomy and compared this technique with full VATS and an open thoracotomic approach. METHODS We performed a retrospective, single-centre study between January 2011 and January 2017 and included 610 consecutive patients having undergone either VATS (with or without intraoperative conversion) or open thoracotomy for anatomical pulmonary resection. Pneumonectomies and angioplastic/bronchoplastic/chest wall resections were excluded. After propensity score adjustment, we assessed the 90-day mortality and determined whether the surgical approach was a risk factor for mortality. RESULTS Of the 610 patients, 253 patients underwent full VATS, 56 patients underwent VATS+conversion and 301 patients underwent up-front open thoracotomy. Relative to the open thoracotomy group, the VATS+conversion group had a higher incidence of cardiac or respiratory comorbidities and was more likely to have an early-stage tumour. Following adjustment, the 90-day postoperative mortality rate was 5.4% (n=3/56) in the VATS+conversion group and 3.7% (n=11/301) in the open thoracotomy group (P=0.58). Likewise, the morbidity rate was similar in these 2 groups. In a multivariable analysis, the surgical approach was not a risk factor for postoperative mortality. CONCLUSIONS Following anatomical resection for lung cancer, VATS with conversion and open thoracotomy were associated with similar early postoperative morbidity and mortality rates. When in doubt, VATS should be preferred to thoracotomy; it potentially provides the patient with benefits of a fully VATS-based resection but is not disadvantageous when intraoperative conversion is required.

Identification and section of pulmonary veins are an essential part of anatomical pulmonary resections. Intraoperative misunderstandings of pulmonary venous anatomy can lead to serious complications such as bleeding and delayed lung infarction or necrosis. We evaluated principally the rate of pulmonary veno-us anatomical variations, and secondarily the reliability and clinical outcomes of a preoperative morphological analysis. Between November 2012 and October 2013, we studied 100 consecutive patients with highly suspected or diagnosed stage I-II primitive lung cancer lesion. The surgical procedure initially retained was video-assisted thoracoscopic surgery (VATS) pulmonary resections and we studied preoperatively the proximal pulmonary venous anatomy using 64 channels multi- -detector computed tomography (CT)-scan angiography to describe the venous anatomical variations. There were 65 men and 35 women with a mean age of 63 years. A pulmonary venous anatomical variation was present in 36 (36%) patients, and right-sided anatomical variations were more frequent than on left-sided ones (25% vs. 11%). The most frequent variation encountered on the right side was the existence of three separate pulmonary veins (16%), and on the left side a single pulmonary vein (8%). Surgical conversion occurred in 21% and we didn't experience a pulmonary venous lesion (0%) or a post-operative lung infarction (0%). We described pulmonary venous anatomical variations and their frequency. Anatomical variations exist and preoperative assessment of pulmo-nary venous anatomy using CT scan is a useful tool in VATS lobectomy to avoid unnecessary extension of pulmonary resections or iatrogenic complications in lung cancer surgery.

Background: Predicted postoperative FEV1 (ppoFEV1) must be estimated preoperatively prior to surgery for non-small cell lung cancer (NSCLC). We evaluated a lung volumetry approach based on chest computed tomography (CT). Methods: A prospective study was conducted over a period of one year in eligible lung cancer patients to evaluate the difference between ppoFEV1 and the 3-month postoperative FEV1 (poFEV1). Patients in whom CT was performed in another hospital and those with factors influencing poFEV1, such as atelectasis, pleural effusion, pneumothorax, or pneumonia, were excluded. A total of 23 patients were included and ppoFEV1 was calculated according to 4 usual Methods: Nakahara formula, Juhl and Frost formula, ventilation scintigraphy, perfusion scintigraphy, and a fifth method based on quantitative CT. Lung volume was calculated twice and separately by 2 radiologists. Tumor volume, and emphysema defined by a −950 HU limit were subtracted from the total lung volume in order to estimate ppoFEV1. Results: We compared 5 methods of ppoFEV1 estimation and calculated the mean volume difference between ppoFEV1 and poFEV1. A better correlation was observed for quantitative CT than for Nakahara formula, Juhl and Frost formula, perfusion scintigraphy and ventilation scintigraphy with respectively: R²=0.79 vs. 0.75, 0.75, 0.67 and 0.64 with a mean volume difference of 266±229 mL (P<0.01) vs. 320±262 mL (P<0.01), 332±251 mL (P<0.01), 304±295 mL (P<0.01) and 312±303 mL (P<0.01). Conclusions: Quantitative CT appears to be a satisfactory method to evaluate ppoFEV1 evaluation method, and appears to be more reliable than other approaches. Estimation of ppoFEV1, as part of the preoperative assessment, does not involve additional morphologic examinations, particularly scintigraphy. This method may become the reference method for ppoFEV1 evaluation.

OBJECTIVES: Intraoperative conversion from video-assisted thoracic surgery (VATS) to thoracotomy may occur during anatomical lung resection. The objectives of the present study were to identify risk factors for intraoperative conversion and to develop a predictive score. METHODS: We performed a multicentre retrospective analysis of French thoracic surgery departments that contributed data on anatomical lung resections to the Epithor database over a 10-year period (from January-2010 to December-2019). Using univariate and multivariate logistic regression analyses, we determined risk factors for intraoperative conversion and elaborated the Epithor conversion score (ECS). The ECS was then validated in a cohort of patients operated on between January- and June-2020. RESULTS: From January-2010 to December-2019, 210,037 patients had been registered in the Epithor database. Of these, 55,030 had undergone anatomical lung resection. We excluded patients who had upfront a thoracotomy or robotic-assisted thoracoscopic surgery (n = 40,293) and those with missing data (6,794). Hence, 7943 patients with intent-to-treat VATS were assessed: 7100 with a full VATS procedure and 843 patients with intraoperative conversion to thoracotomy (conversion rate: 10.6%). Thirteen potential risk factors were identified among patients’ preoperative characteristics and planned surgical procedures and were weighted accordingly to give the ECS. The score showed acceptable discriminatory power (area under the curve: 0.62 in the development cohort and 0.64 in the validation cohort) and good calibration (P = 0.23 in the development cohort and 0.30 in the validation cohort). CONCLUSIONS: Thirteen potential preoperative risk factors were identified, enabling us to develop and validate the ECS—an easy-to-use, reproducible tool for estimating the risk of intraoperative conversion during VATS.