A new paper is conflicted, fails to meet minimum standards and ‘shouldn’t have been published’, an independent expert says.
A new meta-analysis has found that back and lower-limb pain is helped more by spinal cord stimulation than standard of care – but the author of a previous review says it’s a self-serving exercise by industry and should not have been published.
Chronic back pain is notoriously difficult to treat. Spinal cord stimulation (SCS) is an option of last resort and an alternative to other problematic treatments like spine surgery and opioid use.
The new review, published in JAMA Network Open, was funded by Medtronic, a company that manufactures spinal cord stimulators.
It says the evidence shows the treatment does help patients and has a greater effect than conventional medical management such as pain killers, physical therapy and cognitive behavioural therapy.
This is in stark contrast to prior, independently conducted studies – a Sydney University Cochrane review (Traegar et al., 2023) and another by Brunel University in the UK (O’Connell et al., 2021) – which found there wasn’t evidence to support it.
Dr Adrian Traeger, lead author of the first of these, said the new paper was the latest in a series of pushbacks from industry since his Cochrane review was published.
“The JAMA Open paper is nothing more than a prime example of how to use the scientific literature to benefit corporate interests,” he told RR. “It doesn’t meet the accepted minimum standards of a systematic review and shouldn’t have been published.”
Spinal cord stimulators have been under scrutiny in Australia. An ABC Four Corners investigation in April noted that around 41% of devices implanted required intervention within three years, compared to less than 3% of hip replacements. The ongoing TGA post-market review of the devices, begun in 2022, has so far resulted in cancellation or imposition of conditions on many devices on the market, though it doesn’t recommend removing any that are already implanted.
The Traeger and O’Connell reviews were used by a 2023 Australian government post-listing review which said “the evidence base for the comparative clinical effectiveness of SCS compared to standard care is uncertain”.
The current study refutes the findings of those studies, saying many relevant trials investigating novel stimulation were excluded and the methodology was too restrictive.
The researchers are talking about an implantable device that delivers electrical current to the spinal cord to control pain. The device is first trialled to see if it works for the patient – providing at least a 50% improvement on baseline pain – before being implanted.
The current has traditionally been delivered in tonic stimulation mode, in the 40-80Hz range, but in the last 15 years or so, novel approaches have been used, including high frequency, burst stimulation, closed loop and differential target multiplexed.
This new review included 13 studies of 1561 adults with chronic back and/or leg pain, either related to failed back surgery syndrome (persistent spinal pain syndrome type 1) or nonsurgical refractor back pain (type 2) or painful diabetic peripheral neuropathy.
The studies compared novel forms of spinal cord stimulation with tonic stimulation, compared with placebo and/or conventional management. And the reviewers utilised network meta-analysis (a way to compare more than two treatments at a time by combining direct and indirect evidence from a network of RCTs), rather than simple meta-analysis.
Using this methodology, the researchers found that three times more patients with spinal cord stimulation reported at least a 50% reduction in back pain than patients on conventional management. The number was more than eight-fold for patients on novel stimulation.
Back pain intensity was significantly lower for patients on novel spinal cord stimulation than those on conventional management (a 2.34 point difference on the visual analogue pain scale), and leg pain (a 4.01 point difference). Better quality of life was reported by patients on any type of spinal cord stimulation (a mean difference of 0.15 on the EQ-5D index score between tonic stimulation and conventional management and 0.17 for novel stimulation). And for functional disability, conventional spinal cord stimulation was superior to conventional management (a mean difference of 7.10 on the Oswestry Disability Index).
“Our study provides important insights for clinical decision-making as well as assessment by health technology agencies based on a more representative evidence base and a more inclusive analytical framework than that reported by prior reviews comparing SCS technologies,” the authors write.
Dr Trager told RR there were many examples of industry sponsored studies showing results in their favour. He co-authored a paper earlier this year, published in JAMA Internal Medicine, on corporate tactics to undermine independent research and influence health and science.
The Medtronic-funded review was not registered prior to being undertaken, meaning that the questions could be changed and only the desired results published, he said.
The review was also out of date. Although it was published last month, the data search was completed almost two years ago and was not updated to include a key independent trial (Hara et al) published in June last year in JAMA that found pain relief from spinal cord stimulation was no better than placebo.
“The network meta-analysis doesn’t add anything to the literature,” said Dr Traeger.
“What they’re essentially doing is combining the same group of flawed studies [that] several reviews have looked at and pointed out the methodological limitations.
“They’re trying to use this method to get a different answer, and they obviously got the answer that they wanted.”
