If you live with pain that’s lasting longer than three months, it’s likely you’ll be feeling frustrated, worried and possibly a little desperate. People in this situation are inclined to look at possible treatments with a bit of fatalism – like ‘As long as it doesn’t kill me, I’ll give it a shot’ – and as a result (and because of effective marketing and the occasional success) go ahead with treatments that can be both expensive AND not all that great. Worse – if there’s a relatively ‘good’ result in the short-term, people are inclined to think doing it again is a good thing.
And so it is with lots of treatments for chronic back and neck pain. If you don’t believe me, just do a quick Google search (or Duck Duck Go if you don’t want to feed that machine).
Now I get it, we all want relief from pain because it is tiring, makes life hard, and none of us like to feel helpless. Those juicy treatment carrots dangling in front of us are appealing.
But what if they aren’t quite as effective as promised? What if they do harm? What if they’re seductive and offer short-term appeal but hook us in to long-term not quite as lovely consequences?
The BMJ has published a systematic review and network meta-analysis of ‘common interventional procedures’ for chronic non-cancer (and non-inflammatory) neck and back pain (Wang et al., 2025). This is a study examining a heap of interventional procedures (see box below from Busse, et al., 2025, from the Clinical Practice Guideline published in the same issue as Wang et al.), evaluating them against sham procedures and across outcomes including pain intensity and physical function. The authors attempted to conduct network meta-analysis for mental, role and social functioning, sleep quality, opioid use and return to work but there were insufficient studies for them to complete these – however no intervention compared with a sham procedure showed important improvement in these areas (low certainty evidence).
Well, what did they find?
Ummm… little to no difference in pain relief from epidural injections of steroids and local anaesthetic and the same for radiofrequency of the dorsal root ganglion. Epidural injection of local anaesthetic or steroids, and effects for dorsal root ganglion radiofrequency with an epidural anaesthetic with/without steroid – very low certainty. Joint radiofrequency may be more harmful than sham procedures (but with low certainty). Overall the certainty of evidence was low or very low for adverse events, and from moderate to very low for physical functioning, and for pain intensity.
OK, so these procedures don’t offer a lot of pain relief (or improved function), but they’re very popular – Busse et al., (2025) point out that between 1994 – 2001 in the US there was a 271% increase in lumbar epidural steroid injections, and a 231% increase in face injections – and for facet injections under US Medicare, there was an increase from 425 000 in 2000 to 2.2 million interventions in 2013.
If these procedures aren’t very effective (in a similar vein to exercise for pain, to be fair), the risks seem relatively low, but the procedures get done often, what’s a person with pain to do?
The Busse et al (2025) practice guideline included people with pain on the panel – four people living with chronic spine pain, 10 clinicians with experience managing chronic spine pain, and eight methodologists, and followed the GRADE approach. The paper is careful to point out that evidence may change recommendations, and that it’s especially important to note that the evidence is low or very low certainty and outcomes especially important to people with pain such as opioid use, return to work, and sleep quality, simply have not been done.
They make strong recommendations against some procedures that are frequently used – for axial spine (not down the arms or legs) epidural injections with/without local anaesthetic and/or steroids; joint radiofrequency ablation without/without joint targeted injection or anaesthetic and/or steroids; joint-targeted injection of local anaesthetic and/or steroids; and intramuscular injection of anaesthetic and/or steroids. For radicular spine pain – dorsal root ganglion radiofrequency with/without epidural injection of anaesthetic and/or steroids, and epidural injection of anaesthetic and/or steroids.
The rationale? Lack of effectiveness on important patient outcomes, cost (to the patient), repeated travel to clinics to get the treatments ‘every 2 weeks to 3 months for injections and approx every 6 months for nerve ablation procedures.’
Now this does seem a little often to me, given some of the claims made for especially radiofrequency ablations (up to 12 months with some claiming even longer; Engel et al., 2016) but in any event, each time a person has to return to a clinic for treatment is time out of life and more people for a health system to have on its books. In other words, even the best of these procedures doesn’t ‘fix’ the problem permanently. Along with these practical issues, there are risks (small, but moderate to severe harms) and some even with ‘catastrophic’ harms following epidural steroid injection.
The pain reduction offered by these procedures was not enough for people to want to go through with them, according to the outcomes measured – people wanted a 1.5cm reduction in pain on a 10cm visual analogue scale, and this wasn’t found for epidurals or dorsal root ganglion radiofrequency ablations. Most importantly, studies just haven’t been conducted on the outcomes people with pain most value – impact on functioning, mood or returning to work.
Intriguingly, Busse et al., (2025) point out ‘the substantial reimbursement associated with these procedures may act as a perverse incentive for their delivery as opposed to less well paying, and more time-consuming, interventions that have evidence of effectiveness’ – and suggest cognitive functional therapy, exercise and even pain reprocessing therapy. I’d posit that ACT-based approaches might also be suitable for this group, given the explicit focus of therapy on doing things that people most value in their life.
Standing back from this, I think of funding agencies such as New Zealand’s ACC and the current approach to pain management that could very well think of procedures as a ‘quick fix’ to the real problems people with pain experience in daily life. It looks deceptively simple: see one clinician, receive the treatment, and off you go. Come back when it wears off. And off you go.
BUT here’s a problem: if a person develops back pain in their mid-40s (the mean age reported in the Wang et al., 2025 paper), and receives a procedure that reduces pain just a smidge but then must return say, every three to six months – and needs to do this for decades – what are the flow-on effects? The number of clinicians might be lower than for a comprehensive pain management assessment and programme, but the cost per procedure is around $2000 give or take (figures are hard to find online). Do the calculations of three per year, from age 45 – 80 years old, and what do you come up with? Quite an expense. Add to this that it’s not possible to fully discharge people from this kind of service – every three to six months they need to return for their procedure. If you can’t discharge people but need to keep seeing them, you can see how difficult it might be to make room for the new people who develop their pain…
But the more insidious problem is that, in the same way as any approach to pain that only looks to reduce pain intensity, people process what happens to them. My observation from a chronic pain service is that people phoned and asked for a repeat procedure at increasingly short intervals. First it might be every 3 months, then a week before the three months was up, the person would call asking for an early appointment. Then it became two weeks or three weeks before the scheduled appointment. And the time-frame reduced again and again as people monitored their pain – started to feel it return – and called for help. Their confidence to live alongside their pain reduced. Efforts to include self-management strategies alongside procedures failed because why learn them when the answer to pain relief was right there in the form of an injection?
As Jane Ballantyne, in an editorial in the same issue as the papers by Busse and Wang, points out: “What patient struggling with debilitating pain does not want to try an injection that has low risk, even if they know it has little chance of helping? And what patient does not actually improve, at least temporarily, because their pain is acknowledged, and because the white coat and hospital setting are a comfort in themselves?” She goes on to say “On the provider side, pain practitioners are motivated by the gratification of acquiring skills that are in demand, that often produce high patient satisfaction at least in the short term, and that are well reimbursed.” (Ballantyne, 2025).
In a practice setting where quick wins and straightforward positive outcomes are rare, I can see the appeal of these procedures. They’re technically demanding (especially medial branch blocks with RF ablations), patients are grateful for the care and attention they’re given, and the immediate results look great. It’s especially rewarding when there’s a good level of remuneration.
And yet, how useful is a procedure does not address a person’s fear of what the pain means to them; it doesn’t help them develop confidence in their own ability to recover (or even to live alongside their pain) – and they need to repeat it?
Maybe a better pay off comes from investing more time (costly, yes; demanding on clinicians, yes; disappointing for people wanting the quick fix, of course) in good comprehensive assessment AND suitably long-term treatment programmes that give people TIME to change their habits with support? Because pain self management is about changing how you live life, and this is HARD. Bring in some peer support, address the effects on daily life that drive treatment seeking (distress and disability being the main drivers), and emphasise that living with pain is both possible and positive – then we might be heading somewhere useful for the one in five NZers who live with pain that’s lasted more than three months. Not promising pain reduction that’s fleeting and a bit hit and miss, and comes with risks, but attending to what people with pain value, and doing so in ways that respect their lifestyle and the time they need to adjust to this new reality.
Ballantyne, J. C. (2025). Spinal interventions for chronic back pain. BMJ, 388, r179. https://doi.org/10.1136/bmj.r179
Busse, J. W., Genevay, S., Agarwal, A., Standaert, C. J., Carneiro, K., Friedrich, J., Ferreira, M., Verbeke, H., Brox, J. I., Xiao, H., Virdee, J. S., Gunderson, J., Foster, G., Heegsma, C., Samer, C. F., Coen, M., Guyatt, G. H., Wang, X., Sadeghirad, B.,…Agoritsas, T. (2025). Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline. BMJ, 388, e079970. https://doi.org/10.1136/bmj-2024-079970
Engel, A., Rappard, G., King, W., Kennedy, D. J., & Standards Division of the International Spine Intervention, S. (2016). The Effectiveness and Risks of Fluoroscopically-Guided Cervical Medial Branch Thermal Radiofrequency Neurotomy: A Systematic Review with Comprehensive Analysis of the Published Data. Pain Med, 17(4), 658-669. https://doi.org/10.1111/pme.12928
Wang, X., Martin, G., Sadeghirad, B., Chang, Y., Florez, I. D., Couban, R. J., Mehrabi, F., Crandon, H. N., Esfahani, M. A., Sivananthan, L., Sengupta, N., Kum, E., Rathod, P., Yao, L., Morsi, R. Z., Genevay, S., Buckley, N., Guyatt, G. H., Rampersaud, Y. R.,…Busse, J. W. (2025). Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials. BMJ, 388, e079971. https://doi.org/10.1136/bmj-2024-079971