Lung Resection Data Points to Better Outcomes for daVinci than Open Surgery

Intuitive Surgical’s so-called daVinci, Robot-assisted surgery system, has seen its ups and downs so the company will grab any supporting data it can. A new paper, touted by the company, says historical lung resection data shows improved outcomes for daVinci compared with open surgery. It also hints at improved outcomes compared with Video Assisted Thoracic Surgery (VATS), although the difference is not statistically significant.


A new study, published in the October issue of The Annals of Thoracic Surgery, evaluated different techniques for performing the most commonly occurring lung resections in cancer patients.

The study took the form of a restrospective data review using State Inpatient Databases and identified 33,095 patients in eight states who underwent open surgery, VATS, or robotic-assisted surgery.

Of particular note is the growth in robot-assisted surgery, from 0.2% of procedures in 2008 to 3.4% in 2010. Open procedures have steadily declined over the period and VATS procedures, while increasing to 2009, dropped off in 2010, no doubt somewhat influenced by robotic surgery’s increase.

It’s difficult to draw too many statistically significant conclusions from a data set that comprises over 33000 cases, with only a comparative handful of robot-assisted surgeries (420 were performed using the daVinci system and 12500 using VATS). Strikingly though (and frankly unsurprisingly), compared with open surgery, daVinci patients saw a significantly reduced death rate (0.2 percent vs. 2.0 percent, p = 0.016), shorter hospital stays (5.9 days vs. 8.2 days, p < 0.0001) and lower overall complication rate (43.8 percent vs. 54.1 percent, p = 0.003).

Intuitive’s press release hints that their system also out performed VATS, claiming that “the results also indicated that the death rate, length of hospital stays and overall complication rate for robotic-assisted surgery were similar to or better than VATS: “Robotic resection was also associated with reductions in mortality (0.2% vs 1.1%, p = 0.12), length of stay (5.9 days vs 6.3 days, p = 0.45), and overall complication rates (43.8% vs 45.3%, p = 0.68) when compared with VATS.”

The only problem here being that none of these results were statistically significant. So whether you buy indicative data or not, will dictate how much store you set by these observations.


Statistically then, segmentectomy or lobectomy is best done minimally invasively. That’s your conclusion folks.

daVinci is certainly in the game, but claiming superiority over VATS is a bit premature. The data is certainly indicative of an advantage, but do not yet support claims of superiority. Why is that? What’s occurring in these data? It could of course be purely down to numbers. With a small data set from the daVinci group the differences would have to be very striking to yield statistical superiority.

There may be other factors too. It could be that patients undergoing daVinci surgery were preselected based on their suitability. Against that, it could be that VATS cases were artificially favoured because of their more routine application. So it’s hard to read too much into retrospective data, especially when one of the techniques being studies is so relatively new. A randomised single blind study in which the patient and non-OR staff were not aware of the technique employed would do the job much better.

What we’ve learnt from this study then is that minimally invasive surgery yields better outcomes than open surgery when it comes to lung resection. The indicative data suggesting that daVinci beats VATS is encouraging, especially the mortality figure, but surely indicative data shouldn’t be driving the decision.