I invite Dr. Lucy Gee to discuss Spinal Cord Stimulation Reduces Opioid Use in Patients with Chronic Pain. The statistical discussion will be by Ab Gulkarni, and then the discussant will be Kim Berchiel. Thank you for your kind introduction. My name is Lucy Gee. I'm a fourth year medical student at Albany Medical College, and I work closely with Dr. Julie Paletsis, and we're interested in opioid use during spinal cord stimulation. These are our disclosures. So chronic pain, as many of you know, affects approximately 100 million Americans. That's more than cancer, diabetes, and heart disease combined. It's difficult to treat, and it's a significant contributor to increased opioid use in the United States. The National Institute on Drug Abuse has estimated approximately 2 million Americans suffer from substance use disorders related to prescribed opioid medications. In their recent report, they suggest that 21% to 29% of these patients that we prescribe these medications to will misuse them, 8% to 12% will develop an opioid use disorder, and 4% to 6% will transition to heroin use. From 1992 to 2008, we've been looking at substance abuse treatment numbers and have found that for patients greater than 50 years old, the substance use treatment increased from 6.6% to 12.2%. Similarly, primary admissions for substance abuse in older patients increased from 0.7% to 30.5%, and it follows that opioid overdose rates have increased. In 2016, they estimated that greater than 40% of these overdoses were from prescribed medications. So how do we treat chronic pain currently? So we can use medications such as opioids and non-opioid analgesics. We have invasive approaches such as neuroablation, neuromodulation, and injections, and we can use physical and alternative therapies including PT, acupuncture, chiropractic manipulation, and mind-body intervention. So we wanted to focus on spinal cord stimulation, which is an FDA-approved treatment of chronic pain. And we hypothesized for our study that pain and disability outcomes following spinal cord stimulation would improve when simultaneous opioid use was limited or stopped. We included patients at Albany Medical Center who were already destined for spinal cord stimulation surgery for chronic pain between September 2012 and August 2015. All of these patients were greater than 18 years old, and they had a 50% or more pain relief with a three- to seven-day spinal cord stimulation trial and provided informed consent. Before the surgery, we evaluated all of our patients using several questionnaires. They included the Visual Analog Scale, the McGill Pain Questionnaire, the Pain Catastrophizing Scale, the Oswestry Disability Index, and Beck's Depression Inventory. We also looked at their opioid use based on the New York State ISTAP database, which is our tracking system for prescribed opioids, and we used a chart review. Patients underwent their scheduled surgery, and then one year later, we followed up with the same questionnaires and opioid use look. So 86 patients met our inclusion criteria and had primary outcomes at the one-year time point, and of these 86 patients, we found that 33 patients were not taking any opioids at the beginning of the study, and at the one-year time point, they also were not taking any opioids. So we put them in one group together. 53 patients were taking opioids at the beginning of the study, and of those 53, 31 of them decreased or stopped their opioid use during the one-year study. 26 patients stopped completely, and two were reduced by 50 and 75 percent. The other 22 patients who were in this group, we put them in a separate category called Similar and Increased, where 21 of them used a similar dose at the one-year time point, and one patient actually started using opioids during our study. When we looked at demographics across our three groups, we saw the mean disease duration, the gender, the average age at time of surgery, the average BMI at the time of surgery, and diagnoses did not vary between groups. So when we looked at our visual analog scores for all of our patients, we can see here in the black line is the preoperative score, and we didn't see any significant difference across the three different groups for visual analog scale. But at one-year time point, all three of our groups seemed to have a reduced visual analog scale score. The interesting point about this graph here is that the no-opioid group, patients who never used opioids, did significantly better at one year than patients who decreased or stopped their opioids and patients who had the same or increased opioid dose. This suggests that patients who don't use opioids are doing a little bit better at the one-year time point. When we looked at disability scores, we saw again the no-opioid group had a significant reduction in disability score as well as the decrease stopped at one year. However, again, we see the no-opioid group doing significantly better than the patients who decreased or stopped their opioids and the patients who used the same amount. We wanted to look at the emotional and cognitive measures of pain as it's a three-dimensional theory. So we looked at the McGill Pain Questionnaire and we saw that both no-opioid and decreased stop opioid groups did better at one year. However, patients who had the same amount of opioids only trended towards their significance. On the PCS, we saw that the no-opioid group was the only group that had a significant difference at one year. Lastly, we wanted to look at depression scores. So we found again that the patients who had no opioids on board did much better at the one-year time point. However, I do need to point out in this graph, not everyone filled out their questionnaire so we have a smaller sample size in the other two groups which needs to be taken into consideration. And then lastly, we looked at failure and revision rates in our spinal cord stimulation. So revisions were anybody who had a lead migration, people who needed increased coverage and anybody who had an infection. And when we look at the Kaplan-Meier curve, we see the same opioid group, the decreased stopped opioid group and the no-opioid group had no significant difference in the rate of revisions. For failures, we had five patients who had spinal cord failure, meaning the device benefit was outweighed by issues with the device or they didn't have any pain benefit at all. And here we see again there was no significant difference between the groups, although you do see that the no-opioid group did not have any failures. So in conclusion, we find that patients who eliminate opioid use or they don't use opioids prior to opioid surgery have superior clinical outcomes than those who continue use, suggesting assessment and weaning off of opioids prior to surgery may have clinical benefit. And we suggest that spinal cord stimulation treatment is used more often as a pain management strategy to avoid long-term prescribing of opioid therapy and potentially reduce the number of patients requiring opioids in the U.S. I just want to say thank you to my lab and everyone who has helped with this project and to AANS for inviting me to talk today. So this study by Jie et al. has some notable strengths, including prospective collection of most of the pre- and post-treatment outcome measures and the use of multiple related outcome measures, which provides further confidence in the overall results of the study. Among the limitations, however, are the use of multiple statistical comparisons without adjustment and the lack of a true control group, particularly one who did not receive spinal cord stimulation. There was also a variable survey completion rate for the prospective outcome measures used in this study, with some as low as 45 percent. So this does raise the possibility of bias in those who did choose to respond. Ultimately, the most complete results of this study come from the retrospectively collected visual analog scale. So as such, despite promising results from a very good study, we need to exercise caution in our overall final interpretation. Thank you for asking me to comment on this very timely paper, and I think the authors have certainly made a good case for the importance of this work. I don't have any conflicts that are relevant to this particular paper. So, the question of this paper was, does spinal cord stimulation decrease opiate usage? They studied, as you heard, 86 patients over about three years. They used the typical criteria of a 50 percent reduction in pain over a three to seven day trial. They looked at a number of tests before surgery, the numerical rating scale, the McGill, the Oswestry, Beck depression, and this pain catastrophizing score, all very relevant to the problem. The VAS was unfortunately obtained retrospectively, and I think that does taint this a little bit, and that was the initial VAS was obtained retrospectively. They did obtain a VAS at one year postoperatively. They looked at opiate use both pre- and post-op, and they did a logistic stepwise regression to determine which things tended to contribute to a better outcome. I'm grateful that they used Kaplan-Meier statistics, which is a relevant way to look at this kind of information, and ultimately their conclusion was, yes, opiate usage is reduced with spinal cord stimulation. This is another class three evidentiary study. It is non-randomized because, as was mentioned, the VAS score was obtained retrospectively. It's a case series of 86, but reading the paper, I couldn't determine what the denominator was and how these patients were selected, so that leaves some question in my mind about the nature of the group. In addition, each group that was examined for each of these tests that were done prospectively seemed to have different numbers, so there were different numbers of patients that were participating in each subgroup. So again, it makes it hard to interpret. They did find that, though across the board, these scores were better in the folks that had not used opiates preoperatively. The authors, I think, got a little bit outside the rails on suggesting that potentially weaning opiates prior to trial of spinal cord stimulation would improve better outcome, and I don't think there's any data in this particular paper that would support that. So I think that's really an outlier. So again, this study adds another class three evidentiary study that opiate use seems to be reduced with spinal cord stimulation. I think for that reason, it's an important paper, timely, and certainly there should be congratulated for this work. Thank you very much. Thank you.

spinal cord stimulation

opioid use

chronic pain

pain relief

disability