Robotic Assisted Thoracic Surgery Lobectomy versus Video Assisted Thoracic Surgery Lobectomy: Is a Randomized Trial Really Necessary?

Abstract

Minimally invasive pulmonary resection (MIPR) for lung cancer has become standard in many centers, and is slowly being adopted throughout the United States. An abundance of literature supports MIPR over traditional thoracotomy in terms of short-term surgical outcomes with no oncologic inferiority, and perhaps benefit. 1 What has become the latest point of contention, however, is not whether or not MIPR should be performed, but what instrumentation should be used to perform it. In this regard, Agzarian et al 2 have performed a systematic review of published literature (no randomized, controlled trials [RCT]) concerning robotic-assisted thoracic surgery (RATS), ultimately comparing it to nonrobotic MIPR (video assisted thoracic surgery (VATS)) as well as traditional thoracotomy. As one would anticipate, RATS appeared to provide superior outcomes compared to traditional thoracotomy. However, no significant differences were appreciated in surgical and oncologic outcomes between RATS and VATS, with the exception that RATS appeared to be more costlier than VATS. The authors concluded (as have other investigators that have evaluated this question) that a prospective randomized trial is needed to investigate differences in outcomes between RATS and VATS. Is this a question that the thoracic surgical community really needs to attempt to answer in a RCT? We would argue “no,” for several reasons. First, any such trial cannot be performed in a blindedfashion.Different surgeons will need to performRATS vs VATS, which will introduce significant surgeon bias. Surgeons have different technical abilities and philosophies regarding how to conduct the operation. Some surgeons have adopted RATS because they are not comfortable in their ability to perform VATS. As a result, even if a significant difference in an outcome variable was detected in a RCT, it may be contested on this basis. Second, what is the hypothesis that would be tested in a RCT? There is no evidence to date that one approach is better than the other (with the possible exception