Why sacral evoked responses could help in evaluating SNM response

Colin Goudelocke, MD

In this article, Colin Goudelocke, MD, discusses the 2024 Society of Urodynamics, Female Pelvic Medicine & Urogenital Reconstruction Winter Meeting abstract #21: Investigating a new threshold for SNM: sensing sacral evoked responses from the lead. Goudelocke is a urologist with Ochsner Health Center in New Orleans, Louisiana.

We presented earlier data on something similar to this about a year ago. I don't know that I got the best reception, because I think to a lot of people, it was sort of foreign and they didn't quite understand what we were talking about. I found that reception from our moderators and from people in the audience seemed to be much better this year, I think because they're now hearing it for the second time. For me, it seemed odd when I first heard about it a few years ago. This is a trial sponsored by Medtronic called the PEER 2 trial. There was a PEER 1 trial that we were the site for. Essentially, this is the multicenter version of that trial.

In traditional sacral neuromodulation, we have this battery, often called a generator, that's attached to a wire that sits next to the sacral nerve root. And we send a tiny electrical signal through little pairs of electrodes. One electrode would be negative; 1 electrode would be positive. Most of our leads today have 4 electrodes, so we'll pick 1 to be positive and 1 to be negative, and we're just sending signal at a frequency and we're doing it continuously. That's kind of the way sacral neuromodulation has worked for 28 years. Initially, they had done some work in sheep, and then about maybe 3 or 4 years ago, they came to me and said, "We'd like to try this out in humans," and that's where the PEER 1 trial came in. Essentially, what they're doing is saying, "Okay, I understand we have 2 electrodes that we can use to stimulate, but that also leads to 3 electrodes." So if we have 4 electrodes on a lead and one is positive and one is negative, the other 2 are just not doing anything. Could we do something with those 2 electrodes and what they actually use them is as essentially a sensor. There are electrical sensors and they send an electrical pulse across those 2 electrodes.

What electrical pulses are we talking about? When you stimulate any nerve, you get what's called an evoked potential, "evoked" because it's a product of us stimulating. So I wouldn't get it if I didn't stimulate the nerve. It's almost like an echo, and it's a response to the nerve itself. But also, the nerve is innervating all these little pelvic muscles, and they create an electric potential. That gets sort of reflected back in the nerve itself. It's the kind of thing that I would stimulate the nerve with an electrical pulse, and then 5 milliseconds later, I'll actually get an electric response, so we call it an evoked response. And because we're doing it with the sacral nerve, it's called a sacral evoked response. There's been previous work that's been done asking, can we do better? Right now, when I stimulate the nerve, I'm looking for a motor response, like a muscle response. So I get bellows response, that contraction of the pelvic floor, or the toe might contract. And so we have these motor responses. Or I might ask a patient, "When I'm stimulating the nerve, what are you feeling? Are you feeling a sensation? Where are you feeling the sensation?" That's what we traditionally use. Other people have looked at EMGs. Steve Siegel [, MD], did a lot of that work, looking at EMG responses. So I put an EMG on your foot, and can I use the EMG? But this is self-contained. They're using the commercial lead. So the lead that I put in someone yesterday, not as part of this clinical trial, that lead could actually be used to do this. And in fact, when we're doing patients in this clinical trial, we're actually using the off-the-shelf lead. Now, we have to use some proprietary systems to actually send the response and listen to the response. The current Medtronic generator is not capable of doing this. But the Medtronic lead is, which means if this is useful to us, we wouldn't need a new lead; we wouldn't need to change our technique. In fact, if I went to change a battery in someone that I put in 5 years ago, I could actually put in a new battery capable of doing all of these things. So using the standard factory lead, I think, makes a really big difference here, because it makes it not only easier to do moving forward, but it often makes it sort of backwards compatible with new generators. We wouldn't have to change out the lead; it's a much bigger deal to change out a lead than it is to change out a generator.

What are the potential uses for this? I think that's really interesting. The first thing that we've talked about doing is asking, could we use this system to place our leads better? Right now, there's always this internal debate over, do we use motor response? Do we use sensory response? That argument goes back and forth and is interesting, but what if we could use something that's truly objective and independent of the patient in some way, and certainly independent of the person putting in to actually tell us whether or not we're putting in a good lead, so maybe we could use it at the time of lead placement? The other thing is, the whole motor sensory response becomes sort of moot once we actually put this in, because we don't program patients to motor response; that's uncomfortable. We actually program patients to sensory response, and there's reason to believe that the sensory response may not be ideal for how we program patients. So could they create a generator that uses this sacral evoked response, to actually say, this would be the best program to use, this would be the second-best program, this would be the third-best program; again, all based on objective measures, not the subjective measures of sensation that the patient is telling us about.

And then even more wild and even more futuristic, if you want to think about the third iteration of this, anytime we talk about neuromodulation in other areas, or even other therapies, we're always talking about closing the loop. It all comes back to, if I'm creating a insulin pump, the insulin is going to be delivered to the patient, but I need to get some feedback to control what that insulin pump is doing. I don't want to just keep giving more and more insulin; I need to know, what's the glucose here? Obviously, that's a feedback mechanism. It gives them insulin. It checks to see what the resulting blood glucose is and it adjusts the insulin according to that; it's the same thing with pacemakers. I don't want to just keep stimulating the heart again and again; I actually want to know, what is the heart doing, and I only want to pace if I need to pace. These are all closed loop. They actually do that in other areas of neuromodulation. I'm by no means an expert on deep brain stimulation; I know just enough to be dangerous. But even deep brain stimulation is a closed loop. They stimulate and then they use those leads to record EEGs, and then can adjust that deep brain stimulation according to that. For a company that makes pacemakers, that makes insulin pumps, and makes deep brain stimulators, I guess, it made sense to these engineers to ask, "why aren't we doing the same thing with neuromodulation?" And so for them, what's exciting about this is the possibility...that, could we close the loop? Could I stimulate the nerve, and then get a certain waveform back, and then adjust my stimulation if that waveform was not what I wanted?

Now, again, that's still really far in the future, and I don't want to give anybody the impression that that's going to happen tomorrow. We're still very preliminary. PEER 1 basically asked, can we collect a signal? And in fact, we found after implanting about 20 patients that we could collect a usable signal in human patients. PEER 2 has been about, can we collect a signal? And then how does that signal change over time, so now we're checking across multiple time points. And at 3 months, now we're going to collect data on patient response, so that we can now start to say, all right, if there's a signal A and a signal B, can signal A predict that a patient is going to get better? That's not what was presented in this trial, because those data are still collecting. This is still more about the feasibility and where we're collecting and how these signals actually correlate to sensory and motor responses. But we are collecting efficacy data, so that in future publications, we can actually hopefully demonstrate that getting a certain signal predicts a certain better or worse response. And then that will let us know, how do we want to place these leads? How do we want to program these leads? And then could we potentially use this as a closed-loop mechanism? It's still really early; I can't stress that enough. You always want to be cautious about getting too excited about something early on like this. But for me, what's exciting about it is...the idea that we could fundamentally change the way that we implant leads, the way that we program patients, and really, even, the way this device works, to me, even if that's a possibility, you can see why I was so excited to get involved with this research. Even the possibility of something really big and revolutionary can still be exciting even if we're in the early stages.