Devin Peck, MD

#PainManagement #SpinalCordStimulation
🧵

A recent JAMA article about spinal cord stimulators used for management of pain is getting some press, so I thought I would do a modestly deep dive into this article from the perspective of an interventional pain specialist.

TLDR: Some reporting is overstating the implications of this small study of a narrow application of this technology. 1/

https://jamanetwork.com/journals/jama/fullarticle/2797419

Effect of Spinal Cord Burst Stimulation vs Placebo Stimulation on Disability in Patients After Lumbar Spine Surgery

This placebo-controlled, crossover randomized clinical trial compares the effects of spinal cord burst stimulation vs placebo stimulation in patients with chronic radiculopathy after surgery for degenerative lumbar spine disorders.

Dec 3, 2022 at 3:39amWeb
Show threadDevin Peck, MDDec 3, 2022

I’m an anesthesiologist and interventional pain specialist, and so the below might get in the weeds a bit. For that, I apologize deeply.

A group of neurosurgeons in Norway performed a crossover study on spinal cord stimulation (SCS) and its effects on disability. SCS’s are implantable electronic devices that have been in use since as early as the 1960s, but in the last couple of decades, and especially the last few years have become increasingly utilized for the treatment of chronic pain. 2/

Show threadDevin Peck, MDDec 3, 2022
SCS’s can often provide meaningful improvement in patients who have failed other treatments (i.e. medications, injections, rehab, surgery). They consist of electronic leads (small wires implanted in the epidural space outside the spinal cord) and an implanted, programmable battery which sends electronic impulses to the spinal cord to “block” pain signals. There are several theoretical mechanisms, and I won’t get into that here. 3/
Show threadDevin Peck, MDDec 3, 2022
This study utilized a double-blinded, crossover model. After implantation, patients were randomly assigned to a treatment arm or a placebo. After three months, the were assessed and then randomized again to either a treatment arm or placebo for another 3-month period. 4/
Show threadDevin Peck, MDDec 3, 2022
In all, patients were followed for a total of 12 months (four 3-month evaluation periods). All patients had 2 periods each of placebo or treatment, in a random order but with an equal number of patients with each possible sequence. During treatment, stimulation was provided through the device using settings that generally make the stimulation undetectable by the patient. That is, they can’t feel it (more on that in a bit). This is what allowed for blinding of the patients. 5/
Show threadDevin Peck, MDDec 3, 2022
What about blinding of those performing the study? Well, the physicians implanting the SCS used the same technique on all patients, and were not involved in the q3-month assessments. So, they were well-blinded. 6/
Show threadDevin Peck, MDDec 3, 2022
A study nurse performed both the ostensible reprogramming as well as the assessments. So, the nurse was not blinded, and that certainly could affect their assessments. So, overall, this is a poorly-blinded study. It’s unfortunate, because it seems like it would have been easy to use a different investigator to assess than the one doing the reprogramming. Not sure why they didn’t. 7/
Show threadDevin Peck, MDDec 3, 2022
The primary outcome for which this study was powered was the Oswestry Disability Index (ODI), a well-validated measure of disability, but not of pain or pain relief. I say this not to minimize the importance of improving disability in patients with pain: in fact, for the majority of my patients it is the physical limitations their pain imposes, more than the pain itself, which brings them to see me. 8/
Show threadDevin Peck, MDDec 3, 2022
That said, meaningful changes in disability and function take time and consistency, and follow but rarely precede pain relief. This crossover study design meant that, over 12 months, patients spent half that time with a placebo (no stimulation), and potentially the entire time believing they had a placebo. 9/
Show threadDevin Peck, MDDec 3, 2022
While this study included the secondary outcomes of leg and back pain, quality of life, and physical activity levels, it was powered for the primary outcome of ODI, not for these secondary outcomes. 10/
Show threadDevin Peck, MDDec 3, 2022
Back to the stimulation, and why subjects were able to receive stimulation without knowing/feeling it, allowing them to be blinded. The types of stimulation that SCS can deliver can be divided into two broad categories: traditional stimulation that causes a paresthesia (tingling/vibrating sensation) over the painful area, and paresthesia-free stimulation which was pioneered in the mid-2010’s by Nevro, in that case by using very high frequencies. 11/
Show threadDevin Peck, MDDec 3, 2022
From that time, innovation in SCS accelerated and other SCS manufacturers created their own proprietary stimulation profiles, including paresthesia-free modalities. Programming options are now incredibly diverse, adaptable and individualized. 12/
Show threadDevin Peck, MDDec 3, 2022
Those who perform (SCS) know that programming, which occurs after implantation, is key. Patients often require programming adjustment several times during the first few months as the buildup of scar tissue around the leads may change the effects of stimulation over about 12 weeks. And, as I mentioned, ideal programming is very patient-specific. But for this study, only very specific programming parameters were used for the entire duration of the study, and no adjustment was allowed. 13/
Show threadDevin Peck, MDDec 3, 2022
The SCS “trial” (meaning the “test” period in which a temporary lead was placed and kept for 2 weeks, and which needed to show significant pain reduction to proceed to randomization) was done with tonic stimulation. This means the subjects could feel paresthesias in the affected dermatome throughout the “test” period. This was then used as an inclusion criterion for a study on paresthesia-free stimulation. 14/
Show threadDevin Peck, MDDec 3, 2022
These are very different treatment programs and likely work via different physiologic mechanisms and so it makes no sense to use one as inclusion criteria for study of the other. 15/
Show threadDevin Peck, MDDec 3, 2022
A paresthesia-free setting was used in the treatment arm only after having “mapped” coverage using a paresthesia setting. I’m not sure what affect having felt paresthesia and then having it taken away, all while knowing you may have a placebo, would affect outcome; but I do know it’s not the typical way this is done, at least not in the US (the authors are in Norway). 16/
Show threadDevin Peck, MDDec 3, 2022
In terms of the equipment used, there are a range of electronic leads which can be placed. Per the authors: “A 16-contact lead…was implanted for unilateral leg pain or two 8-contact leads…were implanted for bilateral leg pain.” 17/
Show threadDevin Peck, MDDec 3, 2022
At least in the US, we rarely use 8-contact leads from this manufacturer as 16-contact leads allow for more programming capability. Also the 8-contact lead spans 31mm, less than half the 67mm spanned by the 16-contact lead. Less vertebral levels will be covered as a result. We also rarely will place a single lead, even for unilateral leg pain, for similar reasons. 18/
Show threadDevin Peck, MDDec 3, 2022
Technically, I am confused by the authors’ description of implant technique: I’m not understanding how the leads were placed. They seem to be using percutaneous leads, which are typically placed, well, percutaneously (through a needle). Their description is confusing and makes no mention of any introducer needle or loss-of-resistance syringe (the nearly universal way the epidural space is identified). 19/
Show threadDevin Peck, MDDec 3, 2022
If they did not use this technique, how did they identify the epidural space? If it was exposed surgically, a laminotomy would be required for access, but this is not mentioned (a pretty significant omission). 20/
Show threadDevin Peck, MDDec 3, 2022
I think this is an important study in that it demonstrates that a true placebo arm can work in a study on SCS. Prior to the advent of paresthesia-free stimulation, this was not possible. That said, I do not find this study to be damning for SCS technology writ large. Rather, I hope that like all small, well-designed studies, it can be built on to generate more robust data. 21/fin