“Failed back surgery” syndrome – who or what failed?

It’s a term I thought had died a natural death. Failed back surgery syndrome. I used to hear it in the late 1980’s when people with low back pain often had five or six surgeries to try to “fix” whatever wasn’t working in the back. I hadn’t heard it for a while and I really thought it had gone the way it should have – into that great rubbish bin diagnosis pile in the sky, along with “somatisation” and “functional pain” and trigger points. But, to my astonishment, there it is, now a separate MeSH title when you search through Ovid or Medline.

The question I always ask is “Who or what has failed?”

Is it the patient who dares to carry on saying that they have pain, sometimes even more pain than before surgery?

Is it the surgeon for choosing the wrong surgery, or perhaps for doing surgery at all?

Is it the nervous system, already sensitised, now subject to more sensitisation through the planned trauma of the surgical scalpel?

Is it the system for funding surgery?

In a paper written by Shapiro (2015), the rate of “failed back surgery syndrome” is reported as being between 10 – 40%. It seems to increase when the surgery is complex, and it hasn’t improved despite developing less invasive surgical techniques.  Shapiro goes so far as to define FBSS as  “a multidimensional chronic pain syndrome that has significant myofascial and psychosocial components that are directly related to the high incidence of lumbar surgeries in the United States.” Sorry to say, but it’s not just present in the US.

These are the psychosocial risk factors, as identified by Chan and Peng :

  • significant levels of depression,
  • anxiety,
  • poor coping,
  • somatisation
  • hypochondriasis

To be honest, I think these indications would be risk factors for ANY surgery (and probably ANY invasive procedure, including dry needling). But there’s more than this – preoperative risks such as having had prior surgery (spinal instability seems to become worse with repeated revisions); surgery carried out when imaging results suggest there may be good reason, but there is no corresponding clinical correlation; nonsurgical cases of radiculopathy and neuropathy (eg diabetes, viruses, inflammation, vascular disease). And intra-operative risks such as difficult radiographic imaging, inadequate decompression – this paper uses the term “pain generator” for the potential source of nociception, but you and I know that pain only occurs as an output of the brain within a biopsychosocial system; lots of other surgical issues including bleeding, fragments of tissue, “battered root syndrome” (!) where the nerve root is damaged during discectomy; and finally, there are a load of post-operative issues as well – such as haematoma, pseudoarthritis, fibrosis, irritation of the nerve roots, spinal instability, loss of disc height and the list goes on.

Shapiro suggests that some types of ongoing pain after surgery are misdiagnosed myofascial pain, stirring up or producing trigger points. He does add that central sensitisation in the dorsal horn could also play a part, expanding receptive fields in the spinal cord and amplifying “perception” (once again, think of this as amplifying nociceptive input reaching the brain). He also favours the notion of motor control problems following surgery, again compounding the problems that occur with movement after surgery and feeding back into the ongoing pain experienced by the individual.

Shapiro also goes on to suggest various ways the problems could be ameliorated, primarily through careful imaging and then possible surgical management.  The overall management, however, he argues should be pharmacological, “with the goal of increasing physical activity and community involvement”.  Interestingly, Shapiro says “FBSS patients are often pushed aside and told that their surgeons have little to offer them after their surgeries were unsuccessful. These patients are generally managed by community-based physicians, including primary care physicians, anesthesiologists, physiatrists, and neurologists.”  He goes on to say “Treatment of FBSS demands recognition that it is a chronic pain syndrome. One reason that FBSS is a difficult-to-treat syndrome and a public health problem is that the antecedent back pain did not respond to intervention and was often augmented with additional pain complaints caused by the surgery. The psychosocial burden for individuals is huge, especially if there is job loss or loss of function physically. Many of the immediate postoperative pharmacologic strategies complicate the situation further by altering the patient sensorium, mood, affect, and even libido. Opioid-induced hyperalgesia is probably under-recognized and under-reported.”

What does this mean to me?

It must be terribly distressing to be offered a treatment for low back pain and leg pain that is both invasive and expensive – only to find that it hasn’t helped one bit. And even more distressing to then be advised that “there is nothing more I can do, it’s a risk of having surgery”. Worse, then to be advised that the options for managing this pain are quite limited – and for many people, means repeated surgery, perhaps even to the point of having a spinal cord stimulator implanted (with all the known complications of this kind of intervention). Shapiro describes people being offered facet joint radiofrequency neurotomy, but even he acknowledges that “it does not offer a permanent fix to facet-mediated pain, which should raise concern with regard to overutilisation”.

What astonishes me, but probably shouldn’t, is that interdisciplinary pain management approaches continue to be seen as costly, yet there is ample research showing these programmes achieve clinical and cost-effective improvements in patient function.  I’ll let you draw your own conclusions from the following paragraph snipped from Shapiro’s paper:

The cost for this individually tailored 20-day outpatient program, including physical therapy, occupational therapy, and psychological support, is approximately $15,000 (Pain SolutionsNetwork, personal communication, 2013). This is similar to the cost for a single-level laminectomy not requiring a hospital stay and is substantially less costly than the direct costs for lumbar fusion when hospital, hardware, surgeon, and anesthesia costs are added up. [and I’d go even further to add that the 30% risk of a poor outcome is even more expensive…]

I think the health system has failed the people who undergo surgery for low back pain. With a failure rate quoted by Shapiro as approximately 30%, and an increase in epidural injections and facet joint injections of over 250% without any similar improvement in health status, something is going wrong. Very wrong.

Allied health professionals can and do provide cost-effective interdisciplinary pain management for people living with so-called “failed” back surgery syndrome. I think it’s time for allied health professionals to begin agitating, and perhaps even advertising direct to people who have pain – we can help. Our treatments don’t have the risks of invasive procedures. We are focused on helping you DO despite your pain. And what we do can hardly make your pain worse – unlike some of the very sad stories I’ve read.



Chan C, Peng P. (2011). Failed back surgery syndrome. Pain Medicine 12:577–606.

Shapiro, C.M. (2014). The failed back surgery syndrome: Pitfalls surrounding evaluation and treatment. Physical Medicine & Rehabilitation Clinics of North America, 25(2), 319-340.

Rhodin, A. (2014). A case of severe low back pain after surgery. Journal of Pain & Palliative Care Pharmacotherapy, 28(2), 167-168; discussion 168-169.

…but I thought a new knee would fix my pain!
Working in an orthopaedic surgery department is quite enlightening. Along with discussions about ceramic vs metal implants and cartilage regeneration (I work with a team of researchers looking at how to create replacement cartilage), the topic of what counts as a surgical success in knee surgery also comes up from time to time. Knee joint replacement isn’t as successful as hip joint replacement for a number of reasons including the complex nature of the joint, the way the joint capsule is disrupted during surgery, and the knee can apparently feel quite different from the original knee.

What this means is there can be quite a mismatch of expectations for people who believe very strongly that there should be no pain and they should have a completely normal knee joint after surgery. In fact, in the NZ National Joint Registry, the main reason recorded for knee joint revision is pain.

Naturally, the problem of pain after surgery is something surgeons and researchers are also interested in. Many studies have shown psychological factors such as catastrophising (thinking the worst) and low mood are associated with poorer outcomes (Riddle, Wade. Jiranek, & Kong, 2010; Roth, Tripp, Harrison, Sullivan & Carson, 2007; Shelby, Somers, Keefe, Pells, Dixon & Blumenthal, 2008).  The outcomes looked at so far include length of stay in hospital at the time of surgery, analgesia use during surgery, increased disability after surgery and revision rates.

Now, one solution to this problem could be simply not performing joint replacement surgery in people who are depressed and tend to think the worst. The problem is that two of the strongest predictors for looking for treatment are – you guessed it: low mood and catastrophising (and disability). Another solution is to provide intervention after surgery to target the people who tend to become fearful – perhaps an enhanced level of physiotherapy for this group, while the less vulnerable may even require less. Some studies have also looked at giving better education about what the surgery entails and the expected outcomes, and doing this before the person undergoes surgery, while another study has looked at providing a pain coping skills training course prior to surgery (Riddle, Wade, Jiranek & Kong, 2011). The outcomes from this last approach were promising, although it was a very small sample size, and the follow-up was for only two months.

There are a number of problems with implementing any of these approaches in clinical practice, at least in the system I’m most familiar with (NZ):

  1. Many orthopaedic surgeons still discount the importance of psychological factors, and consider the information they provide completely adequate for presurgical preparation
  2. It’s difficult to give someone different or special treatment based on psychological factors within a usual clinical setting without also inadvertently giving people a label that may change how they are responded to
  3. Any additional intervention will have an up-front cost (not good for cost accountants!)
  4. These treatments need specially trained clinicians who understand the measures being used, the rationale for treatment, and the type of treatment needed – and because these are psychological factors and surgery is conducted in a “physical” hospital, this is unfamiliar territory for many
  5. Patients may not be very comfortable with a treatment that focuses on thoughts, emotions and pain

It seems to me, and from the results from a recent study looking at analgesic use after surgery, that an argument simply based on the cost of failing to go ahead with effective management of psychosocial factors would make sense. Here’s the detail:

Fuzier, Serres, Bourrel, Palmaro, Montastruc and Lapeyre-Mestre (2014) conducted a large study of patients in the Midi-Pyrenees region of France. They examined the medical records of all 1939 patients who had knee arthroplasty, and compared the drug use (what was dispensed) at four times – 12 months before surgery, 2 months before surgery, and 10 months before and after the knee surgery.  What they found was an increase in analgesic, antineuropathic and opioid drug use in the 12 months after surgery. The actual percentage of patients requiring more medication is astonishing: 47% of people needed more analgesia, 8.6% needed more antineuropathic medications, and 5.6% needed more opioids. That’s a whole lot of drugs!

These researchers also analysed a number of other variables associated with the medication changes, and made some more understandable findings:

  • people having total knee replacement surgery were at a greater risk of using more medication than those having unicompartmental surgery (half a joint)
  • people with high levels of preoperative pain along with “psychiatric vulnerability” were at greater risk of having increased neuropathic medications prescribed
  • but older people were less likely to be prescribed additional medications

Unfortunately, this study didn’t identify the particular “psychiatric vulnerabilities” of the patients who participated – from the paper it looks like data was extracted from clinical records, so psychiatric disorders such as depression, anxiety and so on are the likely culprits. We don’t therefore know whether catastrophising contributed to the medication use – but given the results from at least 20 studies I have in my database, I think it’s probably likely.

Medications are not cheap. While they’re quick to prescribe, dispense and take, they cost not only in fiscal terms (which is a serious consideration, nevertheless), but also in side effects – and more importantly, in loss of personal locus of control. What I mean by this is that these are people who haven’t been given the opportunity to develop pain self management skills. They will continue to catastrophise, and perhaps even more so after surgery because their fears have been confirmed. Maybe it’s cheaper and more humane to bite the bullet and begin giving people appropriate psychosocial treatment before, during and after surgery.

Fuzier, R., Serres, I., Bourrel, R., Palmaro, A., Montastruc, J., & Lapeyre-Mestre, M. (2014). Analgesic drug consumption increases after knee arthroplasty: A pharmacoepidemiological study investigating postoperative pain PAIN®, 155 (7), 1339-1345 DOI: 10.1016/j.pain.2014.04.010

Riddle, D. L., Keefe, F. J., Nay, W. T., McKee, D., Attarian, D. E., & Jensen, M. P. (2011). Pain coping skills training for patients with elevated pain catastrophizing who are scheduled for knee arthroplasty: a quasi-experimental study. Archives of Physical Medicine & Rehabilitation, 92(6), 859-865.

Riddle, D. L., Wade, J. B., Jiranek, W. A., & Kong, X. (2010). Preoperative pain catastrophizing predicts pain outcome after knee arthroplasty. Clinical Orthopaedics & Related Research, 468(3), 798-806.

Roth, M. L., Tripp, D. A., Harrison, M. H., Sullivan, M., & Carson, P. (2007). Demographic and psychosocial predictors of acute perioperative pain for total knee arthroplasty. Pain Research & Management, 12(3), 185-194.

Shelby, Rebecca A., Somers, Tamara J., Keefe, Francis J., Pells, Jennifer J., Dixon, Kim E., & Blumenthal, James A. (2008). Domain specific self-efficacy mediates the impact of pain catastrophizing on pain and disability in overweight and obese osteoarthritis patients. Journal of Pain, 9(10), 912-919.

When chronic pain is there before surgery

People with chronic pain present a special risk for surgeons, anaesthetists, nurses and the rehabilitation team working with them after surgery.  Surgery is, after all, “planned trauma”, and a sensitive nervous system is going to be even more sensitive after a surgical insult – even when that surgery is intended to reduce pain.

One of the most cost-effective, and clinically-effective treatments for osteoarthritic hips and knees is to simply replace the offending joint with a shiny new one.

In one major New Zealand hospital, people are referred to the Orthopaedic Department by their GP or family doctor. The referral is screened for suitability, the individual is asked to see a physiotherapist for a physical performance test and to complete a set of questionnaires about function and pain. The results from this testing is returned to the Orthopaedic Surgery triage team who review the findings, and, depending on the level of disability, determine whether the person should then proceed to see a surgeon. Once a surgeon has reviewed the person in a face-to-face clinical consultation, if suitable for surgery, they may be placed on a waiting list and eventually have their surgery.

Surgeons like to be successful, and they try to identify those people who will do well with a joint replacement, but despite this, every year there are a good number of people who don’t get the result they’d hoped for. For some, the result might be a very slow recovery, some may do well then end up with a “squeaky” or “clunky” joint, some have an infection or the prosthesis doesn’t sit correctly or something goes awry and they need to have a revision of the joint.

In the case of knee joint replacements, the number one reason for surgical revision is ongoing pain. Now there are a number of reasons for this: the joint capsule is pulled and pushed around to get the replacement joint into place; there are a good many muscles to manipulate, the joint must be moved consistently after surgery so that range of movement is maintained, and so on.

But one reason could be because a group of patients receiving a joint replacement have additional risk factors for post-operative pain that remains unrecognised.

In a large study in Michigan, people about to undergo hip or knee replacement surgery were asked to complete a range of questionnaires – in this study, one of the items included in the questionnaires was the Michigan Body Map. This is a one-sided body map in which the respondent is asked to tick as many boxes as needed (up to 35) to identify where pain that has lasted longer than 3 months is experienced. The researchers used the responses from this questionnaire to calculate a Widespread Pain Index from the 19 specific body areas identified by the 2011 ACR suvey criteria for fibromyalgia. Additionally, they used the scores from a Symptom Severity Scale, added to the WPI and identified those with a score of 13 or more, which is thought to denote those with fibromyalgia from those without. Purists will note that this is not a typical methodology for diagnosing fibromyalgia, but it seems a pragmatic way to identify those with a greater potential for having this diagnosis than not.

Because people with fibromyalgia paradoxically have more endogenous opioids in their body, therefore fewer opioid receptors available to bind to any additional prescribed opioids, people with fibromyalgia need considerably greater doses of opioid than those without this diagnosis. In this study, therefore, the researchers monitored the use of opioids pre and post-operatively.

After some serious statistical work, the group found that younger people, anaesthetic technique, having a total knee replacement (as opposed to a total hip replacement), and longer stays were more likely to use a greater amount of opioid. And, more importantly, the scores obtained for fibromyalgia corresponded the most – an increased opioid consumption of 9.1mg for every 1-point increase on the 0 – 31 point scale.

What does this mean for nonmedical clinicians working with people in that important post-operative period?

Along with factors we already know about, such as the tendency to catastrophise or “think the worst”, it means this group of people need extra special care. Maybe we need to be very certain that these people have good pain relief on board before we start getting them to move. Maybe we need to carefully grade the amount of movement we ask from them in those early hours and days. Maybe we need to follow them up with very precise and careful prescribed exercises, and review these “at risk” people earlier than we need to for others.

Pre-operatively too, we might have some good things to suggest.

Watching our language – being cautious about emotive or alarmist language. Giving people visualisation and mindfulness training so they can use all that wonderful descending inhibition to reduce the sensitivity of their nervous system. Good relaxation, mindfulness, visualisation and diaphragmatic breathing is something people need to learn and be confident with well in advance of needing it. Letting people know that surgeons put the knee through its paces while they’re anaesthetised, so they know that the range of movement limitation they feel is due to post-operative healing rather than the joint not being properly seated in the joint. Giving people a guideline of the expectations we have for them to be able to get up and about despite pain – and that if they need pain relief, we will ensure they have it, irrespective of whether they seem to need “more” than we would expect.

Most importantly, I think allied health need to assess for the increased risk for having a sensitive nervous system before surgery, so we can take far greater care to reassure them and help them settle their nervous system down themselves.

Brummett CM, Janda AM, Schueller CM, Tsodikov A, Morris M, Williams DA, & Clauw DJ (2013). Survey criteria for fibromyalgia independently predict increased postoperative opioid consumption after lower-extremity joint arthroplasty: a prospective, observational cohort study. Anesthesiology, 119 (6), 1434-43 PMID: 24343289

This is from NOIJam – a great example of how it might work