OBJECTIVES: The purpose of this study was to develop a score to predict the complexity of video-assisted thoracoscopic surgery (VATS) lobectomies preoperatively. METHODS: One hundred and thirty-nine consecutive patients undergoing VATS lobectomy operated on by a single surgeon as the first operator were included. Complex operations were defined as: operation time >180 min (corresponding to the 75th percentile) or a conversion to thoracotomy. Several patient-related baseline and radiological variables were tested for a possible association with surgical complexity by logistic regression analysis. An aggregate score was created by weighing the regression estimates of the significant predictors. Patients were then grouped in classes of risk according to their scores. Finally, the score was validated in an external population of 154 VATS lobectomy patients. RESULTS: Twenty-nine VATS lobectomies (21%) were classified as complex. The following variables were found to be significantly associated with a complex operation and were used to calculate the risk score in each patient (1 point each): male (P = 0.006), presence of thick pleura (P = 0.003), presence of emphysema (P = 0.001), enlarged hilar nodes (P = 0.003). Patients were grouped in 4 classes showing an incremental incidence of complex operations (P < 0.0001): score 0, 7.4%; score 1, 18%; score 2, 27%; score >2, 67%. In the external validation set, the score confirmed its association with the incidence of complex operations (P < 0.001): score 0, 7.3%; score 1, 10%; score 2, 16%; score >2 50%. CONCLUSIONS: The complexity score appeared to be reproducible in an external setting and can be used to preoperatively identify appropriate candidates for VATS lobectomies to improve the efficiency and safety of the training phase.
OBJECTIVES: The purpose of this study was to develop a score to predict the complexity of video-assisted thoracoscopic surgery (VATS) lobectomies preoperatively. METHODS: One hundred and thirty-nine consecutive patients undergoing VATS lobectomy operated on by a single surgeon as the first operator were included. Complex operations were defined as: operation time >180 min (corresponding to the 75th percentile) or a conversion to thoracotomy. Several patient-related baseline and radiological variables were tested for a possible association with surgical complexity by logistic regression analysis. An aggregate score was created by weighing the regression estimates of the significant predictors. Patients were then grouped in classes of risk according to their scores. Finally, the score was validated in an external population of 154 VATS lobectomy patients. RESULTS: Twenty-nine VATS lobectomies (21%) were classified as complex. The following variables were found to be significantly associated with a complex operation and were used to calculate the risk score in each patient (1 point each): male (P = 0.006), presence of thick pleura (P = 0.003), presence of emphysema (P = 0.001), enlarged hilar nodes (P = 0.003). Patients were grouped in 4 classes showing an incremental incidence of complex operations (P < 0.0001): score 0, 7.4%; score 1, 18%; score 2, 27%; score >2, 67%. In the external validation set, the score confirmed its association with the incidence of complex operations (P < 0.001): score 0, 7.3%; score 1, 10%; score 2, 16%; score >2 50%. CONCLUSIONS: The complexity score appeared to be reproducible in an external setting and can be used to preoperatively identify appropriate candidates for VATS lobectomies to improve the efficiency and safety of the training phase.