Hello, everyone. My name is Pavlos Texikoulidis, and I am Dr. Boulos' postdoctoral fellow. Today, I will present to you our 12-year trigeminal ganglion stimulation experience from Emory University Hospital. Briefly, as a background, facial pain has several etiologies and can be refractory to medical therapy with multiple agents. These are the cases that will require consideration for a neurosurgical intervention. There are some ablative procedures available, like nucleus caudalis dress lesioning and trigeminal tractotomy, but those are distracted and irreversible. On the other hand, there are some neuromodulatory approaches that have been around for the past few decades. Some examples are DBS for facial pain and motor cortex stimulation, which has variable results in different hands. And then we also have trigeminal ganglion and or trigeminal branch stimulation. The advantages of the neuromodulatory approaches over the ablative procedures mainly lie in their reversible nature and the low risk for major neurological morbidity. Now, just for the historical perspective, the idea of trigeminal neuromodulation was first introduced by Shelton et al. in 1967. Then over a decade later, the implantation of a bipolar plate electrode was reported through an open surgical approach. Now the first reports of the percutaneous approach with trigeminal ganglion were in 1978 and 1984. Since then, several case series have been published reporting on the outcomes of this approach. As you probably already know, this is not a very commonly performed procedure. So with this retrospective case series, which is actually one of the largest around, we try to identify potential predictors of outcome and report the MRI experience with a percutaneous trigeminal ganglion stimulation. A few words about the surgical technique. A stab incision is made 2.5 centimeters lateral to the oral commissure. Then by using the stealth neuronavigation system, we cannulate the foramen ovale and advance the lead to the Meikle's cave. We usually place the zero contact at the trigeminal nerve root and the seventh contact at the level of the foramen. Of course, we use interoperative fluoro to confirm lead positioning. Now an important point here. When patients had targeted pain in the supraorbital and infraorbital region, leads were also placed percutaneously at the supraorbital and infraorbital branches as seen in these interoperative pictures. In this case, the patient, for example, received all three electrodes to cover concurrent pain at these regions. Typically, leads were left in place for a two-week period of trial simulation and trials were considered successful when more than 50% pain relief was endorsed by the patient. During these years, we adapted the procedure and to be more specific, before 2012, for stage one, the distal end of the electrode was connected through the temple and then preauricularly to a temporary extension wire. If the patient reported adequate pain relief, they would come back for the permanent implantation of the system. During this stage, the temporary extension wire was disconnected from the electrode and then connected through a new wire to the IPG. Now after 2012, during trial simulation, the distal end of the electrode was tunneled through the neck skin and if the patient reported more than 50% pain relief, leads were removed in clinic and then during stage two, a new stimulation system was implanted by retargeting the trigeminal ganglion. After retrospective review of the medical records, we identified 59 patients overall, which were included in this analysis. A very important point here is the selection of patients that underwent this procedure and this had to do with the pain characteristics. Of course, an MRI had to exclude neurovascular conflict first, but then the pain had to be constant and background. If for example, a patient had the typical lancinating trigeminal neuralgia pain, they would not be eligible for this procedure. As you can see in the table, the mean age was 60 years old, 78% of the patients were female and the average duration of pain was 5.5 years. The International Headache Society IHS3 classification system was used to provide a diagnosis for each of these patients. Briefly the most common cause of facial pain in our cohort was painful trigeminal neuropathy in 30 patients, followed by atypical facial pain, now called persistent idiopathic facial pain in 12 patients, then idiopathic trigeminal neuralgia in 8 patients, and secondary classical trigeminal neuralgia in 2 patients each. This is an overview of all the outcomes in our cohort. As you can see, 42 patients endorsed the successful trial simulation, which is not bad, it corresponds to 71% of the patients. However, due to unforeseen reasons, for example, denied coverage from the insurance or the patients deferred the permanent implantation and some other medical reasons, a total of 35 patients received permanent stimulation. During this stage, there was one technical failure, which leaves us with 34 patients being followed up after the stage 2 procedure. I will come back to the complications that occurred during the follow-up later in detail. Moving on to our analysis for the trial stimulation period, we found some very interesting results. Traumatic pain etiology was the sole predictor of a failed trial stimulation with an odds ratio of 0.15. Unfortunately, with the present data, we cannot explain this finding, but we have one potential theory. It's possible that peripheral nerve trauma may induce severe de-afferentiation enough to mitigate the central propagation of electro-stimulation-generated impulses to the trigeminal nuclei in the brainstem, midbrain nuclei, thalamus, and cortex, which in turn can eliminate the therapeutic effect of trigeminal ganglion stimulation. However, like I said, this is just a theory and future studies would be needed to dig in further. All the other variables used in our analysis did not show any statistically significant results, and these variables were duration of pain, number of implanted electrodes, all the different permutations of stimulation targets, and type of the facial pain syndrome. Okay, moving on to the results of the permanent stimulation now, 35 patients received this operation and we had one technical failure. So overall, 34 patients received permanent stimulation. During an average follow-up of 20 months, the numerical rating scale for pain decreased from 6.5 to 4, which corresponds to 38% pain reduction. It's important to bear in mind here that these patients suffered from intractable facial pain for years without any relief from other measures, so they were really happy about that. Now, it seems that this procedure does not have the best safety profile with a relatively high complication rate. 13 patients suffered an erosion or infection, and 6 of them had both. The most common site of lead erosion was over the temple. Here you can see two Kaplan-Meier survival curves for erosion and infection. In both cases, you can see that there's a trend for more erosions and infections when one lead, which is the blue line, versus two or three leads, which is the red line, were implanted. Also, a very important point here is that the adaptation of the technique that we did after 2012 was absolutely worth it, as the infection rate before 2012 was 50% compared to 20% after 2012. Just to remind you, a procedure after 2012 means that an entirely new stimulation system was implanted after the trial leads were removed in clinic. In addition, we found six lead migrations in these patients. Four were in the trigeminal ganglion, and two were in the supra- and infra-orbital nerves. Of the trigeminal ganglion lead migrations, three had migrated outwards from the foramen ovale. In the fourth case, the trigeminal ganglion lead was found traversing the right petrous bone into the posterior fossa, which presented with diplopia and new-onset facial pain. The lead migrations were managed with an additional operation that involved retargeting of the trigeminal ganglion under live fluoroscopy. To summarize, this is a visual abstract of our work, and as you can see, trigeminal neuromodulation was utilized for a variety of facial pain syndromes. 42 out of 59 patients reported a successful trial stimulation, which corresponds to 71% of the patients. Hysteriofacial trauma was the only predictor of a failed trial stimulation. During the follow-up and after the permanent implantation of the system in 34 patients, approximately a third had a complication related to the leads. The most common complications were lead erosions and infections. In closing, this brings me to the future perspectives I'd like to share with you. Now, like I said, this procedure appears to work in a subset of patients with intractable facial pain, however, the high complication rate is concerning. So we believe that this approach has a great potential, but we first need to optimize the system. For example, by designing a single system that will cover the whole face, we could be able to reduce the number of lead migrations, erosions, infections, and their associated increase in revision surgeries and healthcare costs. Another important step in the future would be to see how this procedure does compared to other available neuromodulation procedures. And with that, I'd like to thank you for attending this virtual presentation.

trigeminal ganglion stimulation

facial pain

neurosurgical intervention

ablative procedures

neuromodulatory approaches