!
  • Treatment Options for Diabetic Neuropathy
  • Lecture Transcript
  • Male Speaker: Next speaker is Dr. Larry Lavery. He probably needs no introduction. He is a professor in the Department of Surgery at Scott and White Hospital in Texas A&M Health Science Center College of Medicine. He’s the past chair of the American Diabetes Association Foot Care Council. Serves on the Editorial Board for Diabetic Care. He has received extramural grant funding and has over 140 peer-reviewed publications. At this time, Dr. Lavery is going to share Treatment Options, Peripheral Sensory Neuropathy. Please welcome Dr. Lawrence Lavery.

    [Applause]

    Lawrence Lavery: Okay. Thank you very much. I really like the topic because these are kind of the patients that I see most of the time. About three years ago, I moved from Scott and White to the University of Texas Southwestern Medical Center in Dallas where I really have a large urban population with really kind of virulent diabetes. Today, I want to discuss evidence for alternative treatments for diabetic neuropathy. I really want to talk about four areas. I wish I could talk more in depth about several of these because they are really interesting. I’m going to talk about METANX, electrical stimulation, monochromatic light therapy, and surgical decompression for neuropathy. I think a lot of these have some evidence. They’re controversial in our profession and other people’s clinical practices. I think the way that the body of the literature is growing is very interesting, either interesting because it’s good data or interesting because it’s not. When you’re either looking at sensory neuropathy or painful neuropathy, small fiber neuropathy, all of the studies suffered from some of the same barriers. Almost all of the studies are small studies, especially when there’s no intellectual property or no pharma behind them. They’re usually small. They’re underpowered. You’ll see a lot of these studies. The biggest study in a group would be 25 people in each treatment arm. They’re usually of a short duration. Especially when you’re looking at large fiber neuropathy, it probably takes years to turn that around. Many of these studies are very short in comparison. In both studies with large fiber neuropathy and pain studies, people with small fiber neuropathy, there’s a very interesting placebo effect that you think something that’s really looking great, then you compare it to sham and they have nearly the same effect. Many of these studies is due to wrong analysis. I mean, they are classic to discuss in general club because they classically select the wrong statistical analysis to do. It looks good on their data, but when you analyze it the right way, it’s not. It’s no longer significant. Some of these papers have been, except through some really good journals, with the wrong analysis. Then, if you kind of dig for a lot of these studies, their target population is unclear, if it’s large fiber neuropathy or small fiber neuropathy, how they get in, their testing criteria is sometimes nontraditional or doesn’t have a high standard I think would be another way to look at it. There’s some real issues in this area that make it challenging. If you go and look at the pharma studies for Lyrica and Cymbalta, I mean those are the classic sit-down with the feds. We’re going to spend a couple of million dollars. We’re going to get this right the first time studies and they don’t make those mistakes. I mean, those are probably good comparisons. If you look at the current treatment, they are the for mostly for small fiber neuropathy. It’s for pain. It’s Lyrica and Cymbalta products in our toolbox. There are no successful studies. There are no successful drug therapy for large fiber neuropathy. All of those studies had been miserable failures. I don’t know of any currently that are going on. But for a while, there are always four or five neuropathy studies going on every year and none of them had been successful. In 1988, there was a consensus panel that was held in San Antonio. One of the things that came away with this is consensus definition of what diabetic sensory neuropathy is. I think it’s like a lot of things in the consensus panel. I don’t think it means a damn thing. It says their consensus definition was the presence of symptoms and/or signs of peripheral nerve dysfunction in people with diabetes after you exclude other causes.

    [05:06]

    If someone comes in and says, “Hey, my foot is a little numb.” They have neuropathy. If they had NCVs and vibration perception threshold that is abnormal, they have neuropathy and then everyone in between. It’s hard to get your arms around these people without a much tighter definition. There’s a bunch, we have different nerves. We have different neuro fibers. They relay different sensation and drugs work differently on all of them. I mean, I’m tickled often when students come to visit me and I say, “What are you testing?” They say, “I’m testing loss of protective sensation.” I said, “Well, that’s not really a test. That’s not defined on here. It’s a different animal. You do some test and that’s your diagnosis but that’s not a neurofiber type.” Most students come and they have like a Semmes-Weinstein monofilament that look like they sleep on it and comb their hair with it and has been around for a long time. That’s their only understanding of how to test for sensory neuropathy. I mean, that’s a whole different discussion in and of itself. A lot of the studies that I’m going to talk about just address small fiber neuropathy that is primarily pain or allodynia. A couple will only address large fiber neuropathy and some address both or they seem to try to address both. It’s unclear in their inclusion and exclusion criteria. The first thing I want to talk about is monochromatic infrared phototherapy for neuropathy. This is the classic journal club article where they got data, they use bad testing instruments, and they analyze their data wrong. They said their data was positive, it was negative and it got published in a really good journal. You don’t see this around very much. Not very long ago, it was probably a $20 million company. There are two positive and two negative studies, all supposedly randomized clinical studies. This is a large fiber neuropathy study, all four of them. They want to see if they can reverse loss of protective sensation or large fiber neuropathy. I’m going to talk about two of the studies. The first was published in diabetes care and it kind of start with this. It was by Leonard [Phonetic]. They looked at 27 subjects. They enrolled people that missed at least two out of five sites for the 10 gram Semmes-Weinstein monofilament. Their endpoint was only if their 10 gram Semmes-Weinstein monofilament changed or their primary endpoint. Their secondary endpoint was changes in the foot exam, visual analog pain scale, which was a minority of their patients because a lot of these people have no pain, and a symptom score. This was a very, very short study. It was I think six weeks in total. They saw no change in people that had severe neuropathy, so they removed them from the analysis, which the FDA doesn’t let you do any place else and those journals won’t let you do. You have to analyze data on intent to treat. When you look at their data, they have their phase, their group one and their group two based on their severity of disease. When you look at there, they said, “Well look, there’s a change from 3.5 for 2.4 sites with the Semmes-Weinstein monofilament, that’s significant.” “Look, there’s a change from 3.6 to 3.0 in sham, that’s significant.” It’s not from the beginning to the end. It’s the change, comparing the change in group one and group two. The other thing is they used Semmes-Weinstein monofilament, it’s like they’re a continuous variable and they’re not. It’s a discreet dichotomous variable. Do you feel this on your big toe, yes or no? It’s not the number out of five. They didn’t really compare the treatment groups. They threw some people out because they were inconvenient to analyze and they said that they had positive data. This is a study that we did when I was at Scott and White in 2007. It was investigated, initiated study funded by the company that makes Anodyne. All the other studies had been done by physical therapists, so they were done a couple of times a week. This was the largest study. We randomized 69 people and they all got Anodyne or sham treatment that they could use everyday at home. All people that had type 2 diabetes, they had to have a vibration perception threshold of 20 to 45 volts, so a standard measure of large fiber neuropathy. They got either active monochromatic therapy or they got sham, deactivated bites that the lights went on but there was no infrared therapy.

    [10:07]

    They used this 40 minutes a day for 90 days and we evaluated with a number of measures. We use four graded 10 grams Semmes-Weinstein monofilaments, 6 grams, 10 grams, 23 grams, and 60-gram monofilaments. We used the vibration perception threshold device. We used the neuropathy quality of life instrument. We used a 10-centimeter visual analog scale, nerve conduction velocities. Yeah, and I think that’s it. There was no difference in the sham and after-treatment groups in any of the parameters we looked at. The first that was kind of funny like that was the Semmes-Weinstein monofilament. In this group, there were 11 patients in the sham group that got better, that increased the grade of monofilament, and there were only six in the active group that improved, essentially identical results. If you look at their vibration perception threshold, the average that we started with was 33 or so, exactly the same difference between groups and across groups. There was no change in the vibration perception threshold that was significant. You can see the P values are 30% and 92%, no difference. When we look at nerve conduction studies, pre and post, exactly the same results, no change. I think this was a really nice part of the exam because everyone was blinded but especially the EMG person didn’t know the patients at all. Kind of the pluses and minuses of the studies are negative and the two studies that are positive are the same. I mean, they made the same errors in the other study. There doesn’t appear to be any effect and with this therapy when it’s compared to sham. No change in pain. No change in peripheral sensation by any measure that we used. We used the therapy about 70 times longer or they’re at about 70 times the dose that they would on the studies where people got four treatments in physical therapy with no benefit at all. Kind of a classic example of a device that analyzed wrong, taken to clinicians and they accepted it, and actually sales were booming for a while. Next, I want to talk about METANX. METANX is an interesting product. It’s defined by FDA as a medical food which is nutritional support that’s specifically to modify disease for the distinct management of nutritional needs that are related to the disease. It has to be dispensed by a physician, so you write a prescription and it addresses this underlying pathology. I don’t want to go into the pathway that is proposed for this to be effective because there’s a lot more to understand. But the nitric oxide or oxidative stress pathway I think has been associated with neuropathy, wound healing, and a number of other disease processes, fracture healing. Data from this as they build this could have relatively inexpensive implications for patients for other things besides neuropathy. This is the active form of these vitamins that METANX contains, and actually, and this is the dose, 32 and 35 milligrams. I think this company is building a data. They had a couple prospective observational studies. Recently a very nice animal study was published and a randomized clinical trial was published. I’ll share that data with you. This is the universe of what their data looks like. These are the references for those of you who have really good eyes. I’ll walk through a couple of the important studies. This is a recent study that evaluated METANX in Zucker fatty rats that have diabetic neuropathy. Zucker fatty rat is a kind of a classic type 2 rat model. I mean, they look like ball with teeth sticking out. They just get really obese when you feed them a high fat diet. Basically, they looked at Zucker fatty rats and part of the population got METANX and the other part was just kind of wild type and fed regular rat chow. Then they were looking at some of the biomarkers that are associated with neuropathy and diabetes.

    [15:02]

    They looked at some measurements of nerve fiber function as well. The findings of the study were that the Zucker fatty rats developed sensory and motor nerve conduction velocity deficits. They had 26% loss of intraepidermal nerve fibers in the wild type animals. They had abnormal pathway markers for high levels of advanced glycation endproducts and nitric oxide markers. The first thing they did that I want to talk about is the difference in the nerve conduction velocity study. I mean, we’ve done this in our lab and it’s really technically difficult to do in NCV on a rat that’s this big and that fat. This slide does a lot of that work. What they showed was a significant improvement in nerve conduction velocities in the treatment group compared to the control group. When they looked at nerve fiber density, the control group had a 20% loss of small fibers and after animals or the group that was treated with METANX had a 50% higher level when they looked at their epidermal nerve fiber density. Two nice studies for autonomic and small fiber neuropathy that shows significant change in animal model in a relatively short period of time. Then, they looked at advanced glycation endproducts and showed about 38% reduction in animals that received METANX compared to the control groups that got regular rat chow with no supplements. A nice study with a couple nice endpoints that I see as a building block, help us understand perhaps with the mechanism and some of what may affect our patients. Then, we crossed over to some of the human studies. This is kind of the classic, this is how you start. We’re going to look at from 30,000 feet and see what this does. This is 16 consecutive patients. These guys suffer from big placebo effects. That’s the first problem. Sixteen consecutive patients that were treated, they used PSSD as the only tool to measure change in sensation. They looked at the medial heel and the great toe as the area that they did their testing. They evaluated them at baseline, six months and one year, and showed a significant reduction with patients as their own control with two-point discrimination at the great toe and heel. Interesting, maybe gives you a little flavor that there’s something going on here, we need to evaluate it at a higher level because maybe the sham is going to do just as well. Yeah, we’ll talk about this. This is one focus. Yes. This is the Allen Jacobs’ paper, so similar. METANX administered to pregabalin and shows this interesting change. Okay, so that leads us to a randomized clinical study. This is a randomized clinical study that I was part of the scientific advisory board and part of the group that conducted the study, so a very nice double blind randomized clinical study. People are getting this pill to take. They don’t know what’s active. We don’t know what’s active. The biggest problem was there’s only a six-month followup. It’s a 24-week study, a very short period of time to see a change in large fiber neuropathy. The primary endpoint was vibration perception threshold, change in VPT. There was no significant difference in the six-month time period. The primary endpoint, we failed. Then, we looked at their symptom score. You would perhaps think that symptoms may improve faster than objective measures of large fiber function. That’s exactly what we saw, at the last two time points in the study, there was a significant separation and subjective. Let me see if I got the tools. There you go. This is the criteria that you ask patients about for the symptom score. Significant improvement in their symptoms, and then finally, there was a significant change in self-reported quality of life in patients that were in the treatment group. There was a positive improvement and there’s actually deterioration in self-reported quality of life in people that received sham therapy. Interesting.

    [19:58]

    I think the next step was we’re going to start a long-term clinical trial based on these findings, but I think these are studies that need to go out for two years or three years to see change in nerve fiber and functional return. Yeah. Okay. Next, I want to talk about electrical stimulation for neuropathy. Most of these studies are designed to study small fiber neuropathy, burning foot pain, rather than to change large fiber neuropathy. There’s actually a decent body of work in this area. The problem is they’re independent studies. There is no intellectual property really around this. The largest study in this area is 25 people in each treatment arm. They’re not big studies. Let me show you some of the data that I think is kind of interesting. This is a study, 15 people in the treatment group, 16 people in the sham arm. What’s nice about these studies is we can give people sham therapy and patients with neuropathy really don’t know. Several of these studies are sub-sensory levels of stimulation, so they don’t feel it. In this study, they looked at a couple of parameters. The pain parameters was daytime and nighttime pain. They looked at a decrease in both the treatment and in the placebo group. Their analysis though was a change from the beginning to the end, not between groups. You might expect there that even with these small groups, there’s a 29% improvement in the people that got active therapy and a 2% increase in their pain in the patients that got sham. That probably, even with 16 people, would be significant although they did not analyze their data that way. They didn’t look at nighttime pain. Overall, they had 25% improvement compared to a 9% improvement in nighttime pain. Again, they didn’t analyze the change, analyze the change between groups. They analyze the change within groups. Then, the last two things they analyzed were vibration perception threshold. This is the only neuropathy study that also looked at some large fiber function. They saw a decrease in the treatment group from 35 to 33. I just can’t believe, they said that’s a significant difference. I can’t believe that is significant with 16 people. Then, they showed a decrease in the placebo group also. I can’t believe you could run those numbers with 15 people and get a significant difference. Well, I don’t think that difference is real. Then they looked at monofilaments and they also said that there’s a decrease in detection in those two groups. Again, I question the way they analyzed their data. I doubt that in this 10-week time period, they’re going to change large fiber neuropathy. I’m going to finish by talking about nerve decompression surgery. This has really been popularized and taught to a lot of people by Dr. Dellon in Boston. One of the plastic surgeons in our group did a fellowship with Dellon, does a lot of nerve release surgery, and is finishing analyzing data of a large randomized clinical study were people actually were randomized to sham therapy. They would have their threshold tunnel opened up or an incision made on their leg and nothing done to their threshold tunnel. I don’t know how they got that to the IRP or talked people into doing it, but he screened about 1600 people to enroll, I think, 225. He’s just analyzing that data. Right now, there’s not high level data. Most of the data is pre/post changes. It’s descriptive studies. There’s a large number but it’s from the registry. Most of it falls into this semi-prospective registry data that’s nonblinded and has all the problems of studies where you have invested interest in thinking things are going to work. Their theory about this is the single crush, double crush theory of compression on the nerve. If you open the nerve up, then your sensation and your pain is going to improve. This is all this table is actually updated, but it’s still busy. I don’t think it gives you any extra data, but there a number of these descriptive studies that show a significant reduction in pain and reduction in ulceration rate. A lot of descriptive stuff that makes you think, well, maybe there is something really here. I just want to talk about two studies, both published by Dr. Dellon.

    [24:58]

    The first is the results where, actually they’re both results of this registry, the neuropathyregistry.com, so 628 patients. They said, “Patients had no ulcer history, 0.2% ulcerated out of 782 patients. Of the people that have previous ulceration history, 3.8 ulcerated or two people out of 57 patients.” I mean, that’s better than anything that anyone has envisioned in the literature or even gets that. This is more recent. If you look at the update or the study that was updated in 2007, I think this kind of either a fun error or a misunderstanding on the writer’s part. Same thing, multicenter registries, 665 people, 2.5 year followup. They said, “A diabetic patient without a previous ulcer history, they went from 15 to 0.6 instance of ulceration.” If 15% had an ulceration, they wouldn’t be without ulcer history. They’d be in this other group down here. Then they said, patients with a previous ulcer history went from 50% to 2%. Well, if they had an ulcer history, it would be 100% of them would have ulcer. Even in their paper, their operational definitions, their primary results are really kind of crazy. In conclusion, I think, I mean, this is a fascinating area that I think a lot of us struggle with. I think we want alternative therapies. A lot of our patients don’t respond well because of all the CNS disturbances with the mainstay treatments today. They are expensive. Our patients are taking a lot of other drugs. There are drug interactions. We’d really like something that was a modality or a medical food that wasn’t going to interfere with other drugs. I think there’s more evidence. I think the challenge is there’s not big pharma backing a lot of these studies or these devices to pay the $2, $3, or $4 million that you need to enroll people and follow them for years. I don’t know if a lot of these things are going to change recently. There’s probably going to be three years or four years before there’s more METANX data from taking this small group and doing a large longer study that will be more robust and help us understand the process better. I think the body of evidence is growing in this area, but it’s not very fast. For a lot of these devices, I think you have to look at the cost and the risk versus benefit. I mean, obviously, a device like Anodyne, light therapy, there’s not a big risk but it’s pretty expensive for a part that shows no benefit, whatsoever. The same thing with surgery, I mean it’s probably the most risky treatment in this group. I think you have to kind of weigh that very diligently and exhaust other options before you pull that trigger. With that, I will end. Thank you for your attention.