BTK MIS Episode 3 Journal Review Robotic Cholecystectomy and Bile Duct Injury

Hello, everyone. Welcome back to another episode of Behind the Knife. I'm here with the rest of our minimally invasive team at the University of Washington. Ben Vieira, a fellow resident, and our three faculty members, Dr. White, Dr. Wright, and Dr. Citrullo. Today, for our journal review, we're going to be discussing a recent paper on outcomes of robotic versus laparoscopic cholecystectomy, which created quite a stir.

Ben, can you give us a bit of background? Absolutely. Thank you, Paul. So as you all know, the robotic surgery revolution is well underway in the United States and abroad. As discussed in our prior episodes, the robotic platform is being applied to many surgical diseases, including emergency general surgery and biliary disease.

There's been a seven fold increase in the use of the robot for colostectomies in the past few years. But even so, the data is not really clear that the robot actually improves outcomes in such surgeries. Plenty of data has been published about this topic, but most of the data has been limited to single series or small institution case studies.

So today we're going to talk

about a larger paper recently published last year in 2023 that shed some new light into this topic. So before we dive in to this paper, we thought it would be best to review where we have been and all the work that has gone into optimizing the lap coli. Yes. About 30 years ago, laparoscopic colistectomy quickly became the standard of care.

Due to overall low complication rates, lower pain, early return to work. Originally, it had concern for higher bile duct injury rates, but over time those rates have started to disappear. Due to numerous advances in technique, instrumentation and overall just a drop in the devastating complications of the common bile duct injury.

This bile duct injury rate with laparoscopic vasectomy has been realized due to many individuals and teams following their outcomes and seeking to improve the procedure since 2014, the sage of Safe Sesame Task Force has spearheaded this effort, quote, with the mission of creating a universal culture of safety

around this operation and reducing barrier injuries.

We don't have time to review the many foundational studies in s Colosectomy. But we'll link several in the show notes. Be sure to check out Sage's safe cholecystectomy and multi site practice guidelines as well. Thanks, Dr. Citrullo. Dr. White, I wondered if you could weigh in. Despite all these efforts that have gone into reducing bioelect injury, and all trainees having the critical view of safety and other safe chole principles drilled into their memory countless times, bioelect injuries are still happening too frequently.

Thanks, Paul. And you're right. Despite all that we have done to minimize complications of biliary surgery, bile duct injuries can still happen even to the most experienced surgeons. There's a paper by Way et al. titled Causes and Prevention of Laparoscopic Bile Duct Injuries. Analysis of 252 cases from a human factors

and cognitive psychology perspective.

This delves into the causes behind CBD injuries. The authors analyzed 250 laparoscopic major bowel duct injuries that had occurred and grouped them into four classes based on the mechanism and anatomy of injury. They reviewed the operative reports and found that 97 percent of the injuries were due to a visual perceptual illusion and that only 25 percent of the injuries were recognized at the time of injury.

In only 6 percent of the cases, was the problem identified early enough to limit injury. The most common type of injury was a Class III injury, in which the common bile duct was mistaken for the cystic duct and deliberately cut. Thus, it's now

widely accepted that common bile duct injuries occur from misperception.

Not errors of skill, knowledge, or judgment. Thank you for the background, Dr. White. So let's jump into the major topic of discussion today. This paper that came out in 2023. Paul, can you tell us a little bit about it? Yeah, so the title of this paper is comparative safety of robotic assisted versus laparoscopic cholecystectomy.

It was published in JAMA Surgery in September 2023 by Dr. Collada, Dr. Sheets, and their group from the University of Michigan. It was a retrospective cohort study. Data was extracted from the Medicare claims database between January 2010 and March 2020. and December 2019, selecting patients by using ICD codes who underwent 23 hour observation or inpatient admissions for cholecystectomy.

Patients were notably between the ages of 66 and 99 and were excluded if they had any hepatobiliary malignancies. The paper compared

outcomes in laparoscopic and robotic cholecystectomy, but also had numbers on the number of open cholecystectomies completed and injury rates of open cholecystectomy during this time period as well.

So the primary outcome was a bile duct injury requiring surgical repair with hepatico jejunostomy or choledoco jejunostomy within one year. Secondary outcomes included the overall incidence of post operative complications and incidence of post operative biliary interventions, including surgical or endoscopic biliary interventions.

Multiple variables were accounted for. And beyond that, the authors did do some additional complicated analyses to try to further limit confounding. for listening. Can you tell us a little bit about this, Paul? A large possible confounder is that there may be an inherent difference among patients selected for a robot versus laparoscopic cholecystectomy.

Some surgeons may reserve the robot for their healthiest patients with the best anatomy when they're early in their robotic surgery career, giving themselves essentially the best cases to learn on. On the other

hand, owing to improved dexterity and visualization, many surgeons use the robot only for the hardest cases.

So the authors tried to account for this by performing an instrumental variable analysis. In this analysis, they basically modeled a patient's chances of undergoing a robotic cholecystectomy given the rates of robotic surgery in that hospital system in the year prior, and then used various patient level factors to estimate the relative risk of complications, including bile duct injury.

Finally, sensitivity analyses were performed, accounting for total hospital volume with the robot, clinical urgency and patient demographics. Dr. Wright, before we dive into the results, this is a pretty complex methodology. Yeah. So first of all, the group at the university of Michigan are really leaders in health sciences research and they're experts in this area, but this is really difficult and I would say notoriously fraught with potential errors, including some confounders.

I think a couple things here, one is that the Medicare provider analysis and review data

that they used apparently included the 23 hour observation in inpatient admissions. It's not clear that they included outpatient cholecystectomy, so that may have selected for a slightly different patient population than usual.

The other thing is that there aren't actually CPT codes that separate between robotic and laparoscopic cholecystectomy. So they had to use some complex modeling to try to identify robotic cases. And it's not really clear whether that's a valid approach. So when you look at this data, I think it's really intriguing and a warning signal.

But it's also should really be thought of as a hypothesis generating paper. In other words it's a signal that there may be a problem here. that needs further work. I don't know that because of the issues with patient identification, we can clearly see that this represents the ground truth.

It's a warning signal. Thanks, Dr. Wright. So, Ben, what did they find?

Yeah, so between 2010 and 2019, the rate of robotic cholecystectomy increased 37 fold, making up about 5 percent of cholecystectomies in 2019. The rate of laparoscopic surgery increased from 75. 4 to 77. 8 percent, while the rate of open cholecystectomy decreased from 24.

4 to 17 percent. As far as patient demographics, those who underwent robotic surgery were younger, had more comorbidities, and were more likely to be Black or Hispanic. They were also more likely to have an elective cholecystectomy and less likely to undergo the surgery for acute cholecystitis. In other words, more often for biliary gallstone pancreatitis.

One thing I think is really interesting about that point is, Dr. Wright and Dr. White, I'm sure you can mention it too. Sometimes you think you're taking a nice biliary colic patient to the OR and then it turns out to be a little closer to acute cholecystitis. And I've certainly had some hard gallstone pancreatitis gallbladders in my life.

So calling those

not acute cholecystitis by relying only on Medicare data without being able to go back and review operative reports to see the actual description of the case may be another area where some of the data may have gotten muddied. Absolutely. Thanks, Dr. Strula. So, as far as some of the more interesting points in terms of their outcomes, the overall rates of bile duct injury decreased during the study time frame for both robotic and lap coles.

Overall, the rate of bile duct injury requiring repair within one year was 0. 7 for robotic colostectomy and 0. 2 percent for laparoscopic colostectomy. Robotic surgery was also associated with a higher rate of biliary interventions, 7. 4 compared to 6%, and serious complications, 9. 3 versus 8. 6%. Although overall there was no difference in overall 30 day complication rate.

When performing the instrumental and sensitivity analyses, robotic surgery was still associated with higher complication rates in multiple areas.

Namely, higher rates of biodeck injury for elective surgery, emergent surgery, and cholecystectomy is performed for acute or chronic cholecystitis. Dr. White, can you give us your thoughts on these results?

The biggest takeaway, of course, is the higher rate of bile duct injury, necessity for biliary interventions, and serious complications amongst the robotic group. By using the instrumental and sensitivity analysis. The authors attempted to show that these results were not due to selection bias or other majority confounders.

It's also really important to note that people may justify these findings by saying that the robot is used in more difficult cases. And that is why there is a higher rate of complications. But that's not exactly what the paper found. In fact, the robot, according to the paper, was more often used in elective cases.

And for biliary colic, engulfed stone

pancreatitis rather than cholecystitis. Now, as Dr. Citrullo pointed out, that may not always be the case. But this does tell us, perhaps, that surgeons are actually using the robot in lower risk cases. On the other hand, like I talked about when we discussed methodology, it's really difficult to fully account for all of the possible confounders.

So, It's possible that if this is a real effect, is it due to the robotic platform and the learning curve associated with the robotic platform? Is it due to the changes in the visual perception when you're moving from a laparoscopic view to a robotic view? Or is it some difference in patient comorbidities?

I think a really fascinating and possible explanation is also a difference in the surgeons themselves. Who's adopting robotics, and are those surgeons more likely to have a common duct injury? If you go back to the early days of

interoperative cholangiography, a fascinating paper by Dave Flum showed that IOC use was associated with a lower risk taking behavior, and there was a reduction in common duct injuries among surgeons who scored for a lower risk taking behaviors on personality surveys.

So, is there something about who's doing robotic cholecystectomy that's driving this? Thanks, Dr. Wright. That's a really interesting point. An additional thing that I wanted to comment on is the difference in population about who is getting robotic versus laparoscopic holostectomies. I think as many surgeons adopt the new robot platform, some of them are using more common surgeries and quote unquote easier type surgeries such as gall bladders and appendectomies to get some some time on the console and get their skills up.

It's important to note that black and Hispanic patients were more likely to undergo robotic surgery than their white counterparts. And I think that we need to be sure that minority populations and populations that been stigmatized in the medical field are

not the ones who we are practicing a new technology on while we, you know, get our reps is something that we are not quite as good at.

You know, that's a really interesting point, Ben, and it runs counter to a lot of other studies which have shown a lower rate of adoption of minimally invasive techniques in minoritized populations. And so there's a I think we're still learning more about disparities in access studies. I think we should continue to be mindful of.

Getting to that point of the learning curve, which we're kind of starting to talk about with robotic, or adoption of robotic cases and procedures, you know, we expect to bear a little bit increased burden of complications as we're developing these skills on the robot. You know, when you're first doing laparoscopic cholestectomies, you don't feel as confident as you do when you've done 200.

Same thing with robotic surgery. It takes a while to build up that comfort and skill level. And if we have this minimally invasive option like laparoscopic cholestectomy that we know is safe and has very good rates of bile duct injuries,

should we really be asking some patients to accept this higher risk for the potential benefit down the line?

Which patients should we really be asking to bear the brunt of that burden where they have to take the high, you know, almost three times increased risk of bile duct injury for robotic surgery? I mean, it's a similar but different discussion that was happening a few decades ago when laparoscopic surgery was coming around, and we're going to talk more about that, but, Early on, laparoscopic surgery in general, especially colostectomy, was criticized for higher rates of bile duct injury compared to open.

But with time, those rates have come down to now we have extremely low rates in laparoscopic colostectomy, but it took 30 years to get here. Are we going to accept a 30 year learning curve for the general rate of our bile duct injuries to go down with the robot to equal laparoscopic? Is it realistic to expect that number to go down that much over that time period without a major technological change?

I think these are, there's a

lot of unanswerable questions in this that you have to, that kind of plague all of us when we're trying to decide about starting new technologies. That's a very good point. I just want to jump in because it's not exactly the same. We're taking a procedure that we do well laparoscopic and we're moving on to robotic surgery.

So open to laparoscopic is a big change, but the skill set is not as varied as that. But the addition of using different instrumentation for robotic cholecystectomy versus laparoscopic cholecystectomy is something to think about. We use many different instruments in laparoscopic surgery because we've already paid for them as the capital.

But when I'm doing a robotic cholecystectomy, I'm using one dissection instrument and I'm avoiding. Using the standard things like endo shears

suction other things that I would standardly use to make it more cost effective. And I wonder if any of that has to do with this change here. I'll throw in there something that's unanswerable from this data set is the push for single Robotic cholecystectomy, which is a I think a difficult and different operation than a multiport robotic cholecystectomy And there's no way to look at those numbers in this data set.

Yeah, I did a lot of those in 2012 to 2014 and They're hard they're hard One thing I think is important to note is that we really also need to talk about that Robotic cholecystectomy has not really shown a true positive benefit over laparoscopic cholecystectomy. When, speaking before, we were talking about the obvious advantages of MIS over open surgery.

To the degree that robotics can replace open cholecystectomy in maybe for a surgeon who

doesn't feel comfortable doing a procedure laparoscopically, that may be a nice net benefit for those patients and that patient population. But right now, if you're a, Yeah, I think that's an interesting point just because laparoscopic cholecystectomy is sort of a core competency of general surgery training, and some of the other, you know, When we talked about robotic surgery benefits in the past, we've talked about it, it allowing surgeons to do something they otherwise might not be able to do laparoscopically, for example, some of some complex hernia repairs that might be completely impossible laparoscopically.

might be made possible to be done in the community by applying the robot and sort of working your way up from basic robotic hernia procedures up to more complex hernia procedures. But it's interesting to think of this, I think a little differently given the

procedure of laparoscopic cholecystectomy, the frequency of cholecystitis, and that this is something all general surgeons need to be able to do at the end of their training.

I want to circle back to Dr. White's point made in the paper about the robot being used more often in the elective setting. In patients with acute, without acute cholecystitis, I wonder if the results may be different if you stratified by surgeon volume with the robot. Anecdotally, particularly with experienced robotic surgeons, I've seen the robot reserved only for the most difficult cases.

I wonder if that's the framework we should be using. If you're doing, say, 90 percent of your gallbladders laparoscopically, maybe switching modalities only for the hardest gallbladders doesn't make sense. What's your take on this, Dr. Wright? You know, one of my favorite papers in general surgery ever is the Lawrence Ray paper that you mentioned earlier.

And that really showed that most common duct injuries are due to issues in perception and interpretation. And the worry that I have with the robot is it might be subtly different in setup and visual angles compared to lab. You have a different angle of the camera, your port

position's a little different, your view's a little bit different.

And that might increase the risk of those visual interpretation errors. And we actually saw this I posted this on X, what formerly was known as Twitter. A video that was presented at a plenary session just a couple weeks ago at the SAGES meeting in Cleveland, where a group presented a video of a robotic common duct injury.

And the operating surgeon had a slightly different view than we would expect in the lab view, and they ended up dissecting much too low, below the sulcus. And that led also to retraction, it was sort of vertical, instead of pulling the infinibulum out and away from the liver. And that aligns the cystic duct with the common duct.

So those two perceptual issues were really due to being on this new platform. Alright, so, if you only pull out the robot for the hard cases, you don't learn those little nuanced differences. And so I think it's a tough one, because if you are Never using the robot except the hard cases, you don't get

good at the robotic platform.

And then you might be leading yourself astray. In some way, I think it's almost like an all or nothing. You decide you're going to use the robot and get good at it, or you decide you're not going to use it. Well, I just had a little lightbulb moment when you talked about the view, because the cameras are different with robotic surgery.

And we actually use a 30 down lens. to do a robotic cholecystectomy. Think about it. We're using a 30 degree scope in general on the laparoscopic, but we're changing the angle ourselves. When you use a 30 down on the robot, you're constantly looking down towards the hilum, right? I wonder if that's one of the main reasons.

One of the limitations I've found with using Indocyanine Green is that although you can see what is filled with bile, it

doesn't actually really help you always identify what structure is what. This robotic video that we've just been discussing, you can see IC Green filling all the structures and it can very easily fool you that it's just the lumen of the gallbladder that's lighting up.

So people think of it that you can use it as a, as almost an adjunct for a cholangiogram or things like that at this point, and I'm not sure we're there yet with it. And it can certainly lead you astray. Yeah, it's interesting. It's harder to do a cholangiogram when you're doing a robotic cholecystectomy.

And so people have said, well, we can use ICG as an, as a replacement. Nothing against ICG. I think it has a role and I have used it and will continue to use it. But you can make mistakes in interpretation of the ICG. And it's another new technology in addition to the new technology of the robotic platform.

That is, has its own learning curve and its own issues. I completely agree. And just, I'm a huge advocate of IC Green. I use

it on every single whole cystectomy, laparoscopic or robotic. And I really use it to train my brain. to have the understanding of the bile duct anatomy and get through that learning curve.

I also firmly believe if you're going to do something laparoscopic you should do it robotically. So, I truly believe you should be doing the robotic cholangiogram, but it does require a different skill set. You have to find different tools to use. You have to work with your Fluoroscopy tech to come in at different angles and it can be painful and it can take more time and that adds to cost.

So at the end of the day, I really think you have to consider the value to your patients. So Dr. White and Dr. Strull, you, you both do a lot of gallbladders that are at our community site, Northwest. What are, how are you guys using the robot now for your Colvirus? Are you ever using it,

are you using it on elective cases sometimes for the most difficult cases?

How often are you guys using it in practice? I use it selectively. I do it when I'm doing a combination procedure, for instance, with a hernia or I have criteria for higher BMI. Okay. So. high, a BMI with a lot of visceral obesity, 40, 45 plus. I choose to use the roll up because I find it's easier for me to have the forearms in there for retraction.

I think that's important. Clearly highlight how systematic you are in choosing your robotic cases. And you're a high volume surgeon in this area. And we were talking about maybe if you're applying the robot just to difficult cases, maybe it wouldn't be a good idea. But I think the fact that you've chosen a population you're consistently applying it to,

you're getting reps in seeing it in those types of cases.

So I think that's important that maybe. Maybe you do choose to apply it to a select part of your patient population rather than saying it's an all or nothing, but that you do it in a way that's not sort of on the fly, oh I'm going to apply the robot in this case, it might be hard. It's a kind of predefined criteria where you're getting reps consistently.

I would also highlight it that it points out that value has a lot of definitions, and we've talked previously about value to the patient, and maybe the robot doesn't provide a different value compared to laparoscopic surgery, but there's a value to the surgeon, and those high DMA, BMI patients are, can be miserable to operate on because of the ergonomic strain, the challenges The robot takes a lot of that ergonomic strain and offloads it.

And it offloads it from the surgeon's body. So when we talk about our own career longevity, there may be value there that we don't capture in a data set like this. Yeah, I

currently do not do any robotic cold cystectomies in my practice. I'm doing everything laparoscopic. But I anticipate in the future adding it to my practice at some point.

I think the teaching that residents can get laparoscopic is very important, but I think robotic cholecystectomy is a worthwhile surgical procedure to pursue, so I think it's going to be a part of my future. And the, you know, one of the things we've been discussing is kind of the conclusions of these paper, this paper in particular is very strong.

And in my personal opinion, I'm not sure I fully buy into the, we need to re evaluate this at an extreme level. I do think at this point that we need to have a pause and decide, are there other techniques, are there other teachings, are there other things we need to do to optimize this procedure to make it more worthwhile, but I still think that we should be doing this procedure in our, in patients.

Thanks, Dr. Trulov. I

think that brings up something that came up is Dr. White was discussing the cases that she has done robotic cholecystectomies for. I'm curious, do you do those in, sort of, we, we discussed some of the changes at the robot in terms of the camera angle, but I'm curious how the steps of your operation, and are they very similar to the way you do it laparoscopically, or are there large differences, and do you apply those kind of strategies from Sage's Safe Cholecystectomy Program?

To the robot, same exact technique. The only thing different is that I'm using hook cautery to do my entire dissection. So I'm, when I do my dichotomy for the cholangiogram, I'm using the hook cautery on cut. But same exact principles. I don't think you can adopt a new technique and do different principles.

Going from doing laparoscopic colectomies to robotic colectomies. I still do the same exact principles. Yeah, and I'll

use that as an opportunity to plug the SAGE's safe cholecystectomy program again. Because it really highlights the strategies and the principles that can protect against bile duct injury, no matter whether you're doing it robotic, lab, or open.

So, things like the true critical view of safety, and I think a lot of residents and attending surgeons actually don't understand what the critical view of safety actually is. Things like understanding what the aberrant and alternate anatomy might be things like taking a pause, a surgical pause before doing an irrevocable step like a ductotomy.

These are all really tools that anyone can use in any operation with any technology and help save our patients from the really horrific complication of a dump. Thanks, Dr. Wright. One other question I'm curious about, Dr. White, you being the one that's doing some of these robotic cholecystectomies, have you noticed any differences with tactile feedback?

I know a big part of my learning so

far is starting to do dissection and cholecystectomies is there's a lot of There's a lot of tension on the infundibulum as you're learning to retract based on how much inflammation there is, learning how much you need to pull to really splay the cystic duct and cystic artery out and bring them away from the common bile duct.

And it's kind of finding the sweet spot in terms of the, both the angle and the amount of tension you're putting on them. And I wonder, do you feel that's pretty easy to reproduce or replicate with the robot? I think it comes down to the learning curve and feeling with your eyes, but I will say having residents who will do teaching on these cases, it's easier to.

tear the infundibulum with the retracting grasp, or the per rasp that we use. And I've seen that more with robotic poleus than in a laparoscopic case. Yeah, it'll be interesting to see what happens with new technology. I mean, we're already seeing the introduction of haptics in the

next generation of the da Vinci.

We haven't used that clinically here yet. It's just rolling out. But I think it'll be really interesting to see what happens when we can start using AI. I mean, my Subaru, which costs 36, 000, will automatically recognize when I'm about to run into somebody and apply the brakes, right? So why can't we have something like that in the robotic platform that says, Hey, this may be a perceptual error, stop, don't cut that duct.

I think that's probably not more than five or six years out. And if you look at some of the work on AI identification of the critical view of safety, I think it's coming soon. Wow. Use of artificial intelligence to assist in identification of the critical view of safety sounds like a really interesting topic.

Sounds like something we might have to explore in a future episode. Thank you all for sharing your thoughts on safe and thoughtful application of the robotic platform to cholecystectomy. Ben, can you summarize with our take home points? Bioduct injuries are

most commonly caused by errors in perception.

It's not specifically technique or knowledge. is by, is because we are misperceiving something while we operate. And more often than not, we don't actually recognize a mistake when we make it. Second point, robotic holostectomy has been shown to have a slightly higher biotic injury rate in this Medicare data set specifically.

But again, there are some limitations to the methodology. And I would encourage everyone to read this paper and come to their own conclusions as well. And I assume that this question is going to continue to be studied and we'll get more data on it. A few final things, we'd like to say that when applying a new approach to a procedure with a well established standard of care, we need to carefully assess any possible increased risks and if significant, they should be disclosed to the patient, so that we always are keeping the patient's safety in mind.

And finally, The safe cholecystectomy principles that we know that Sage has published should not be abandoned. We need to

still ensure that we're maintaining our basic surgical principles. Thanks Ben, and thank you all for tuning in to another episode of Behind the Knife. Until next time, dominate the day.