Next, we have Dr. Paff, and she'll be speaking to us on clinical outcomes associated with robot-assisted DBS placement into the STN in patients with PD. Hi, I'm Michelle Paff, I'm from UC Irvine, and I'm going to be talking about our experience using the ROSA robot for DBS surgery in patients with Parkinson's disease. So I'm going to touch on technique and workflow, clinical outcomes, accuracy, and operative time. And just a brief history, so even though robotics and neurosurgery hasn't caught on until recently, they've actually been around since the mid-1980s, starting in 1985. And with the first neurosurgical robot gaining FDA approval in 1991, that was the Neuromate by Renishaw. And then we have the SpineAssist by Maser in 2004, and then finally, our ROSA robot, which I'll be talking about today, which was FDA approved in 2012. So this is just the basic setup. In contrast to the Maser robot, we actually use a Lexel frame just to immobilize the patient's head during the procedure, and that attaches to the ROSA during the length of the procedure to keep it immobilized and at a certain length from the robotic arm. And we have the option of using laser facial recognition or bony fiducials for registration. We use bony fiducials to improve accuracy. And this is the setup of what it looks like intraoperatively with the patient in the CT center and the robot at the patient's head with the microdriver in position. So our workflow consists of preoperative planning on the preoperative MRI on the ROSA software itself. And then in the OR, we place the bony fiducials and the patient in the head frame. We get an intraoperative CT, and we merge that to the preoperative planning MRI on the ROSA software. At that point, we perform registration. And we generally look for an error of less than 0.75 millimeters. And then we move on to the procedure. And then after that's completed, we perform intraoperative head CT and merge the postoperative CT with the planning MRI so we can see how accurate we are. Oh, this is just a kind of a typical day in the OR doing DBS. With the ROSA, there's a lot of people involved. All of our patients are asleep under general anesthesia during this procedure. That's just how we've always done it. And so we do do microelectrode recordings with the patient asleep, which do not give us a spike frequency as great as you would see in an awake patient. But it is useful to us to help give us a little bit more information so that we can ensure optimal placement of the final lead. So at the end of the procedure, we get a postoperative head CT with the bony fiducials in place and the head frame in place in case we need to go back and change something. And then we superimpose the postoperative CT with the planning MRI. If it's within about 2 millimeters of what we wanted, then we consider that successful. So the purpose of this project was kind of to determine, once we started using the system, how are we doing compared to how we were doing using the CRW frame? When we first started in 2012, we were using the CRW frame with the Framelink software exclusively. And then when we switched over in June of 2015, we started using the ROSA exclusively to do DBS surgery. So we wanted to assess our results. And we had 33 patients that underwent DBS placement. This is specifically with Parkinson's patients undergoing DBS insertion into the STN. We had 33 patients. Ten of those were from outside referrals, so we didn't have clinical data available for those. And some of them did not reach the six-month follow-up. So a final number of 18, for which we had clinical data available. And in the ROSA group, we had 27, three were outside referrals, four did not reach six months of follow-up. And so we have clinical data for 20 patients in the ROSA group. And this is just a table summarizing the patient characteristics. There wasn't a significant difference in the age at the time of surgery or the duration of symptoms. And the vast majority of our patients were placed bilaterally into the STN. Only a few had unilateral placement. So this is a bar graph just illustrating the improvement in the Movement Disorder Society Unified Parkinson's Disease Rating Scale score, Part 3 motor score, six months, 12 months, and 24 months after surgery. All of these assessments, just to make it clear, were done with the patient on levodopa and in their on state. So what we found is that at each follow-up point, both groups demonstrated a significant reduction in their motor score compared to their baseline. However, when we compared between groups, we found no significant difference between the ROSA and the CRW frame. Same thing with the levodopa-equivalence doses. We found that all the patients at each follow-up point demonstrated a significant reduction in their levodopa-equivalence dose. And that was significant compared to their preoperative levodopa-equivalence doses. However, we did not find a significant change between or a difference between the frame and the ROSA. We did look at accuracy in our ROSA group. We measured the distance from the end of the lead on the postoperative CT compared to the planned target in cases where the lead was actually placed at target. And we found that we were able to achieve a vector error of 1.77 plus or minus 0.8 millimeters. I had some data from some historical patients at our institution that were implanted with the CRW frame, which was found to have a vector error of 2.53 millimeters. And in the literature, some reports are around a little bit less than two. So basically, in conclusion, our ROSA is able to achieve an accuracy on par with the frame-based method. One thing is with the OR time, we did notice that overall, using the ROSA for this procedure can add about an hour to 70 minutes more of total operative time, which I attribute to the registration process, especially with surgeons or staff who are not used to using the ROSA. They may need to register more than one time to get an acceptable amount of error in order to proceed. So we did notice that over time, we did improve, and when the same surgeon uses the system over and over again, they do get faster. As far as complications, we had two infections or wound dehiscences in the CRW frame-based group and three in the ROSA group. This is a relatively small number of patients, so we didn't really notice that much of a difference in complications between the two groups. So in conclusion, I feel like there's advantages and disadvantages to both systems. We really felt that the ROSA eliminates a lot of the potential for error during the transfer of coordinates and placement of the arc. It eliminates all that. One of the benefits is that we're able to make changes to our trajectory during the surgery and then delete those changes if we want. So it very much streamlines the procedure. I would say the biggest disadvantage is with the learning curve associated with using it and the potential to have increased OR time as a result. So that's it. Thank you. I have a question. Michelle, you showed the results in terms of outcomes for Parkinson's disease, the ePDRS results. Yeah. And I noted that there was a little bit of a trend for worse outcomes in the ROSA group, but there was no difference between the two groups. Yes. And this sample may be small so far, right? Yes. I agree. And I actually expected at first, when I was first doing this analysis, I almost thought that the ROSA was going to be inferior to the frame and that that would be real. But there was a lot of variation, and so that turned out not to be significant. Do you have a sense if you would do a power analysis to figure out how many cases you would need to do to show the difference, what that number would be, what's the magnitude of it? That's the next step. Once we get more patients, I think that will be the next step. But there's a lot of things involved with the outcomes, and so I think that that's kind of like where I'm going in the future with that. In terms of doing the MER under anesthesia, that's a little bit different, and in terms of the cocktail that you use, is it like Ativa, as we would do for neuromonitoring? It's remifentanil and propofol. So when we use biz monitoring, we keep the biz within a certain range. And we do not get as good of signals as you would with an awake patient, but we do find it useful, and we do change our target maybe 20% of the time. So based on that, and along with the macrostimulation, it is useful to us. And are they intubated? Yes. Thank you very much. Thank you.

robot-assisted deep brain stimulation

Parkinson's disease

ROSA robot

DBS surgery

clinical outcomes