Repair Failed Back Surgeries

Failed back syndrome or post-laminectomy syndrome is a condition characterized by persistent pain following back surgeries.

Failed back syndrome (FBS), also called “failed back surgery syndrome” (FBSS), refers to chronic back and/or leg pain that occurs after back (spinal) surgery, usually after laminectomy. It is characterized as a chronic pain syndrome. Multiple factors can contribute to the onset or development of FBS. Contributing factors include but are not limited to residual or recurrent disc herniation, persistent post-operative pressure on a spinal nerve, altered joint mobility, joint hypermobility with instability, scar tissue (fibrosis), depression, anxiety, sleeplessness and spinal muscular deconditioning. An individual may be predisposed to the development of FBS due to systemic disorders such as diabetes, autoimmune disease and peripheral blood vessels (vascular) disease.

Common symptoms associated with FBS include diffuse, dull and aching pain involving the back and/or legs. Abnormal sensibility may include sharp, pricking, and stabbing pain in the extremities. The term “post-laminectomy syndrome” is used by some doctors to indicate the same condition as failed back syndrome.

The treatments of post-laminectomy syndrome include physical therapy, Microcurrent electrical neuromuscular stimulator, minor nerve blocks, transcutaneous electrical nerve stimulation (TENS), behavioral medicine, non-steroidal anti-inflammatory (NSAID) medications, membrane stabilizers, antidepressants, spinal cord stimulation, and intrathecal morphine pump. Use of epidural steroid injections may be minimally helpful in some cases. The targeted anatomic use of a potent anti-inflammatory anti-TNF therapeutics is being investigated.

The number of spinal surgeries varies around the world. The United States and the Netherlands report the highest number of spinal surgeries, while the United Kingdom and Sweden report the fewest. Recently, there have been calls for more aggressive surgical treatment in Europe (see infra). Success rates of spinal surgery vary for many reasons.


Spinal surgeons operating on a back.

Patients who have undergone one or more operations on the lumbar spine, and continue to experience and report pain afterward can be divided into two groups. The first group are those in whom surgery was never indicated, or the surgery performed was never likely to achieve the desired result; and those in whom the surgery was indicated, but which technically did not achieve the intended result. It has been observed that patients who have a predominant painful presentation in a radicular pattern will have a better result than those who have predominant complaints of back pain.

The second group includes patients who had incomplete or inadequate operations. Lumbar spinal stenosis may be overlooked, especially when it is associated with disc protrusion or herniation. Removal of a disc, while not addressing the underlying presence of stenosis can lead to disappointing results. Occasionally operating on the wrong level occurs, as does failure to recognize an extruded or sequestered disc fragment. Inadequate or inappropriate surgical exposure can lead to other problems in not getting to the underlying pathology. Hakelius reported a 3% incidence of serious nerve root damage.

In 1992, Turner et al. published a survey of 74 journal articles which reported the results after decompression for spinal stenosis. Good to excellent results were on average reported by 64% of the patients. There was, however, a wide variation in outcomes reported. There was a better result in patients who had a degenerative spondylolisthesis. A similarly designed study by Mardjekto et al. found that a concomitant spinal arthrodesis (fusion) had a greater success rate. Herron and Trippi evaluated 24 patients, all with degenerative spondylolisthesis treated with laminectomy alone. At follow-up varying between 18 to 71 months after surgery, 20 out of the 24 patients reported a good result. Epstein reported on 290 patients treated over a 25-year period. Excellent results were obtained in 69% and good results in 13%. However, these optimistic reports do not correlate with “return to competitive employment” rates, which for the most part are dismal in most spinal surgery series.

Studies by Cohen show that up to 25% of all low back pain is sacroiliac joint in origin and that the diagnosis of sacroiliac joint disease is frequently overlooked by physicians. Studies by Ha, et al., show that the incidence of SI joint degeneration in post-lumbar fusion surgery is 75% at 5 years post-surgery, based on imaging.[15] Studies by DePalma and Liliang, et al., demonstrate that 40-61% of post-lumbar fusion patients were symptomatic for SI joint dysfunction based on diagnostic blocks

In the past two decades there has been a dramatic increase in fusion surgery in the U.S.: in 2001 over 122,000 lumbar fusions were performed, a 22% increase from 1990 in fusions per 100,000 population, increasing to an estimate of 250,000 in 2003, and 500,000 in 2006. In 2003, the national bill for the hardware for fusion alone was estimated to have soared to $2.5 billion a year. For patients with continued pain after surgery which is not due to the above complications or conditions, interventional pain physicians speak of the need to identify the “pain generator” i.e. the anatomical structure responsible for the patient’s pain. To be effective, the surgeon must operate on the correct anatomic structure; however it is often not possible to determine the source of the pain. The reason for this is that many patients with chronic pain often have disc bulges at multiple spinal levels and the physical examination and imaging studies are unable to pinpoint the source of pain. In addition, spinal fusion itself, particularly if more than one spinal level is operated on, may result in “adjacent segment degeneration”. This is thought to occur because the fused segments may result in increased torsional and stress forces being transmitted to the intervertebral discs located above and below the fused vertebrae. This pathology is one reason behind the development of artificial discs as a possible alternative to fusion surgery. But fusion surgeons argue that spinal fusion is more time-tested, and artificial discs contain metal hardware that is unlikely to last as long as biological material without shattering and leaving metal fragments in the spinal canal. These represent different schools of thought. (See discussion on disc replacement infra.)

Another highly relevant consideration is the increasing recognition of the importance of “chemical radiculitis” in the generation of back pain. A primary focus of surgery is to remove “pressure” or reduce mechanical compression on a neural element: either the spinal cord, or a nerve root. But it is increasingly recognized that back pain, rather than being solely due to compression, may instead entirely be due to chemical inflammation of the nerve root. It has been known for several decades that disc herniations result in a massive inflammation of the associated nerve root. In the past five years increasing evidence has pointed to a specific inflammatory mediator of this pain. This inflammatory molecule, called tumor necrosis factor-alpha (TNF), is released not only by the herniated or protruding disc, but also in cases of disc tear (annular tear), by facet joints, and in spinal stenosis. In addition to causing pain and inflammation, TNF may also contribute to disc degeneration. If the cause of the pain is not compression, but rather is inflammation mediated by TNF, then this may well explain why surgery might not relieve the pain, and might even exacerbate it, resulting in FBSS.


Before the advent of CT scanning, the pathology in failed back syndrome was difficult to understand. Computerized tomography in conjunction with metrizamide myelography in the late 1960s and 1970s allowed direct observation of the mechanisms involved in post operative failures. Six distinct pathologic conditions were identified:

  • Recurrent or persistent disc herniation
  • Spinal stenosis
  • Post operative infection
  • Epidural post-operative fibrosis
  • Adhesive arachnoiditis
  • Nerve Injury

Recurrent or persistent disc herniation

CT scan image of large herniated disc in the lumbar spine.

Removal of a disc at one level can lead to disc herniation at the same level or a different level at a later time. Even the most complete surgical excision of the disc still leaves 30-40% of the disc, which cannot be safely removed. This retained disc can re-herniate sometime after surgery. Virtually every major structure in the abdomen and the posterior retroperitoneal space has been injured, at some point, by removing discs using posterior laminectomy/discectomy surgical procedures. The most prominent of these is a laceration of the left internal iliac vein, which lies in close proximity to the anterior portion of the disc. In some studies, recurrent pain in the same radicular pattern or a different pattern can be as high as 50% after disc surgery. Many observers have noted that the most common cause of a failed back syndrome is caused from recurrent disc herniation at the same level originally operated. A rapid removal in a second surgery can be curative. The clinical picture of a recurrent disc herniation usually involves a significant pain-free interval. However, physical findings may be lacking, and a good history is necessary. The time period for the emergence of new symptoms can be short or long. Diagnostic signs such as the straight leg raise test may be negative even if real pathology is present. The presence of a positive myelogram may represent a new disc herniation, but can also be indicative of a post operative scarring situation simply mimicking a new disc. Newer MRI imaging techniques have clarified this dilemma somewhat. Conversely, a recurrent disc can be difficult to detect in the presence of post op scarring. Myelography is inadequate to completely evaluate the patient for recurrent disc disease, and CT or MRI scanning is necessary. Measurement of tissue density can be helpful.

Even though the complications of laminectomy for disc herniation can be significant, a recent series of studies involving thousands of patients published under auspices of Dartmouth Medical School concluded at four-year follow-up that those who underwent surgery for a lumbar disc herniation achieved greater improvement than nonoperatively treated patients in all primary and secondary outcomes except work status.

Spinal stenosis

CT scan of laminectomy showing scar formation (highlighted in red)causing new stenosis.

Spinal stenosis can be a late complication after laminectomy for disc herniation or when surgery was performed for the primary pathologic condition of spinal stenosis. In the Maine Study, among patients with lumbar spinal stenosis completing 8- to 10-year follow-up, low back pain relief, predominant symptom improvement, and satisfaction with the current state were similar in patients initially treated surgically or nonsurgically. However, leg pain relief and greater back-related functional status continued to favor those initially receiving surgical treatment.

A large study of spinal stenosis from Finland found the prognostic factors for ability to work after surgery were ability to work before surgery, age under 50 years, and no prior back surgery. The very long-term outcome (mean follow-up time of 12.4 years) was excellent-to-good in 68% of patients (59% women and 73% men). Furthermore, in the longitudinal follow-up, the result improved between 1985 and 1991. No special complications were manifested during this very long-term follow-up time. The patients with total or subtotal block in preoperative myelography achieved the best result. Furthermore, patients with block stenosis improved their result significantly in the longitudinal follow-up. The postoperative stenosis seen in computed tomography (CT) scans was observed in 65% of 90 patients, and it was severe in 23 patients (25%). However, this successful or unsuccessful surgical decompression did not correlate with patients’ subjective disability, walking capacity or severity of pain. Previous back surgery had a strong worsening effect on surgical results. This effect was very clear in patients with total block in the preoperative myelography. The surgical result of a patient with previous back surgery was similar to that of a patient without previous back surgery when the time interval between the last two operations was more than 18 months.

Post-operative MRI findings of stenosis are probably of limited value compared to symptoms experienced by patients. Patients’ perception of improvement had a much stronger correlation with long-term surgical outcome than structural findings seen on postoperation magnetic resonance imaging. Degenerative findings had a greater effect on patients’ walking capacity than stenotic findings.

Postoperative radiologic stenosis was very common in patients operated on for lumbar spinal stenosis, but this did not correlate with clinical outcome. The clinician must be cautious when reconciling clinical symptoms and signs with postoperative computed tomography findings in patients operated on for lumbar spinal stenosis.

A study from Georgetown University reported on one-hundred patients who had undergone decompressive surgery for lumbar stenosis between 1980 and 1985. Four patients with postfusion stenosis were included. A 5-year follow-up period was achieved in 88 patients. The mean age was 67 years, and 80% were over 60 years of age. There was a high incidence of coexisting medical diseases, but the principal disability was lumbar stenosis with neurological involvement. Initially there was a high incidence of success, but recurrence of neurological involvement and persistence of low-back pain led to an increasing number of failures. By 5 years this number had reached 27% of the available population pool, suggesting that the failure rate could reach 50% within the projected life expectancies of most patients. Of the 26 failures, 16 were secondary to renewed neurological involvement, which occurred at new levels of stenosis in eight and recurrence of stenosis at operative levels in eight. Reoperation was successful in 12 of these 16 patients, but two required a third operation. The incidence of spondylolisthesis at 5 years was higher in the surgical failures (12 of 26 patients) than in the surgical successes (16 of 64). Spondylolisthetic stenosis tended to recur within a few years following decompression. Because of age and associated illnesses, fusion may be difficult to achieve in this group.

Post operative infection

A small minority of lumbar surgical patients will develop a post operative infection. In most cases, this is a bad complication and does not bode well for eventual improvement or future employability. Reports from the surgical literature indicate an infection rate anywhere from 0% to almost 12%. The incidence of infection tends to increase as the complexity of the procedure and operating time increase. Usage of metal implants (instrumentation) tends to increase the risk of infection. Factors associated with an increased infection include diabetes mellitus, obesity, malnutrition, smoking, previous infection, rheumatoid arthritis, and immunodeficiency. Previous wound infection should be considered as a contraindication to any further spinal surgery, since the likelihood of improving such patients with more surgery is small. Antimicrobial prophylaxis (giving antibiotics during or after surgery before an infection begins) reduces the rate of surgical site infection in lumbar spine surgery, but a great deal of variation exists regarding its use. In a Japanese study, utilizing the Centers for Disease Control recommendations for antibiotic prophylaxis, an overall rate of 0.7% infection was noted, with a single dose antibiotic group having 0.4% infection rate and multiple dosage antibiotic infection rate of 0.8%. The authors had previously used prophylactic antibiotics for 5 to 7 postoperative days. Based on the Centers for Disease Control and Prevention guideline, their antibiotic prophylaxis was changed to the day of surgery only. It was concluded there was no statistical difference in the rate of infection between the two different antibiotic protocols. Based on the CDC guideline, a single dose of prophylactic antibiotic was proven to be efficacious for the prevention of infection in lumbar spine surgeries.

Epidural post-operative fibrosis

Epidural scarring following a laminectomy for disc excision is a common feature when re-operating for recurrent sciatica or radiculopathy. When the scarring is associated with a disc herniation and/or recurrent spinal stenosis, it is relatively common, occurring in more than 60% of cases. For a time, it was theorized that placing a fat graft over the dural could prevent post operative scarring. However, initial enthusiasm has waned in recent years. In an extensive laminectomy involving 2 or more vertebra, post operative scarring is the norm. It is most often seen around the L5 and S1 nerve roots.


Under rules promulgated by Titles II and XVI of the United States Social Security Act, chronic radiculopathy, arachnoiditis and spinal stenosis are recognized as disabling conditions under Listing 1.04 A (radiculopathy), 1.04 B (arachnoiditis) and 1.04 C (spinal stenosis).

Return to work

CT scan showing post operative scarring and arachnoiditis.

In a groundbreaking Canadian study, Waddell et al. reported on the value of repeat surgery and the return to work in worker’s compensation cases. They concluded that workers who undergo spinal surgery take longer to return to their jobs. Once two spinal surgeries are performed, few if any ever return to gainful employment of any kind. After two spinal surgeries, most people in the worker’s comp system will not be made better by more surgery. Most will be worse after a third surgery.

Episodes of back pain associated with on the job injuries in the worker’s compensation setting are usually of short duration. About 10% of such episodes will not be simple, and will degenerate into chronic and disabling back pain conditions, even if surgery is not performed.

It has been hypothesized that job dissatisfaction and individual perception of physical demands are associated with an increased time of recovery or an increased risk of no recovery at all.Individual psychological and social work factors, as well as worker-employer relations are also likely to be associated with time and rates of recovery.

A Finnish study of return to work in patients with spinal stenosis treated by surgery found that: (1) none of the patients who had retired before the operation returned to work afterward. (2) The variables that predicted postoperative ability to work for women were: being fit to work at the time of operation, age < 50 years at the time of operation, and duration of lumbar spinal stenosis symptoms < 2 years. (3) For men, these variables were: being fit to work at the time of operation, age < 50 years at the time of operation, no prior surgery, and the extent of the surgical procedure equal to or less than one laminectomy. Women’s and men’s working capacity do not differ after lumbar spinal stenosis operation. If the aim is to maximize working capacity, then, when a lumbar spinal stenosis operation is indicated, it should be performed without delay. In lumbar spinal stenosis patients who are > 50 years old and on sick leave, it is unrealistic to expect that they will return to work. Therefore, after such an extensive surgical procedure, re-education of patients for lighter jobs could improve the chances of these patients returning to work.

In a related Finnish study, a total of 439 patients operated on for lumbar spinal stenosis during the period 1974-1987 was re-examined and evaluated for working and functional capacity approximately 4 years after the decompressive surgery. The ability to work before or after the operation and a history of no prior back surgery were variables predictive of a good outcome. Before the operation 86 patients were working, 223 patients were on sick leave, and 130 patients were retired. After the operation 52 of the employed patients and 70 of the unemployed patients returned to work. None of the retired patients returned to work. Ability to work preoperatively, age under 50 years at the time of operation and the absence of prior back surgery predicted a postoperative ability to work.

A report from Belgium noted that patients reportedly return to work an average of 12 to 16 weeks after surgery for lumbar disc herniation. However, there are studies that lend credence to the value of an earlier stimulation for return to work and performance of normal activities after a limited discectomy. At follow-up assessment, it was found that no patient had changed employment because of back or leg pain. The sooner the recommendation is made to return to work and perform normal activities, the more likely the patient is to comply. Patients with ongoing disabling back conditions have a low priority for return to work. The probability of return to work decreases as time off work increases. This is especially true in Belgium, where 20% of individuals did not resume work activities after surgery for a disc herniation of the lumbar spine.

In Belgium, the medical advisers of sickness funds have an important role legally in the assessment of working capacity and medical rehabilitation measures for employees whose fitness for work is jeopardized or diminished for health reasons. The measures are laid down in the sickness and invalidity legislation. They are in accordance with the principle of preventing long-term disability. It is apparent from the authors’ experience that these measures are not adapted consistently in medical practice. Most of the medical advisers are focusing purely on evaluation of corporal damage, leaving little or no time for rehabilitation efforts. In many other countries, the evaluation of work capacity is done by social security doctors with a comparable task.

In a comprehensive set of studies carried out by the University of Washington School of Medicine, it was determined that the outcome of lumbar fusion performed on injured workers was worse than reported in most published case series. They found 68% of lumbar fusion patients still unable to return to work two years after surgery. This was in stark contrast to reports of 68% post-op satisfaction in many series. In a follow-up study it was found that the use of intervertebral fusion devices rose rapidly after their introduction in 1996. This increase in metal usage was associated with a greater risk of complication without improving disability or re-operation rates.


The identification of tumor necrosis factor-alpha (TNF) as a central cause of inflammatory spinal pain now suggests the possibility of an entirely new approach to selected patients with FBSS. Specific and potent inhibitors of TNF became available in the U.S. in 1998, and were demonstrated to be potentially effective for treating sciatica in experimental models beginning in 2001. Targeted anatomic administration of one of these anti-TNF agents, etanercept, a patented treatment method, has been suggested in published pilot studies to be effective for treating selected patients with chronic disc-related pain and FBSS. The scientific basis for pain relief in these patients is supported by the many current review articles. In the future new imaging methods may allow non-invasive identification of sites of neuronal inflammation, thereby enabling more accurate localization of the “pain generators” responsible for symptom production. These treatments are still experimental.

If chronic pain in FBSS has a chemical component producing inflammatory pain, then prior to additional surgery it may make sense to use an anti-inflammatory approach. Often this is first attempted with non-steroidal anti-inflammatory medications, but the long-term use of Non-steroidal anti-inflammatory drugs (NSAIDS) for patients with persistent back pain is complicated by their possible cardiovascular and gastrointestinal toxicity; and NSAIDs have limited value to intervene in TNF-mediated processes. An alternative often employed is the injection of cortisone into the spine adjacent to the suspected pain generator, a technique known as “epidural steroid injection”. Although this technique began more than a decade ago for FBSS, the efficacy of epidural steroid injections is now generally thought to be limited to short term pain relief in selected patients only. In addition, epidural steroid injections, in certain settings, may result in serious complications. Fortunately there are now emerging new methods that directly target TNF. These TNF-targeted methods represent a highly promising new approach for patients with chronic severe spinal pain, such as those with FBSS. Ancillary approaches, such as rehabilitation, physical therapy, anti-depressants, and, in particular, graduated exercise programs, may all be useful adjuncts to anti-inflammatory approaches. In addition, more invasive modalities, such as spinal cord stimulation, may offer relief for certain patients with FBSS, but these modalities, although often referred to as “minimally invasive”, require additional surgery, and have complications of their own.

Worldwide perspective

CT scan showing two views of L4-5 disc herniation

A report from Spain noted that the investigation and development of new techniques for instrumented surgery of the spine is not free from conflicts of interest. The influence of financial forces in the development of new technologies and its immediate application to spine surgery, shows the relationship between the published results and the industry support. Authors who have developed and defended fusion techniques have also published new articles praising new spinal technologies. The author calls spinal surgery the “American Stock and Exchange” and “the bubble of spine surgery”. The scientific literature doesn’t show clear evidence in the cost-benefit studies of most instrumented surgical interventions of the spine compared with the conservative treatments. It has not been yet demonstrated that fusion surgery and disc replacement are better options than the conservative treatment. It’s necessary to point out that at present “there are relationships between the industry and back pain, and there is also an industry of the back pain”. Nonetheless, the “market of the spine surgery” is growing because patients are demanding solutions for their back problems. The tide of scientific evidence seems to go against the spinal fusions in the degenerative disc disease, discogenic pain and in specific back pain. After decades of advances in this field, the results of spinal fusions are mediocre. New epidemiological studies show that “spinal fusion must be accepted as a non proved or experimental method for the treatment of back pain”. The surgical literature on spinal fusion published in the last 20 years establishes that instrumentation seems to slightly increase the fusion rate and that instrumentation doesn’t improve the clinical results in general. We still are in need of randomized studies to compare the surgical results with the natural history of the disease, the placebo effect, or conservative treatment. The European Guidelines for lumbar chronic pain management show “strong evidence” indicating that complex and demanding spine surgery where different instrumentation is used, is not more effective than a simple, safer and cheaper posterolateral fusion without instrumentation. Recently, the literature published in this field is sending a message to use “minimally invasive techniques”; – the abandonment of transpedicular fusions. Surgery in general, and usage of metal fixation should be discarded in most cases.

In Sweden, the national registry of lumbar spine surgery reported in the year 2000 that 15% of patients with spinal stenosis surgery underwent a concomitant fusion. Despite the traditionally conservative approach to spinal surgery in Sweden, there have been calls from that country for a more aggressive approach to lumbar procedures in recent years.

Cherkin et al., evaluated worldwide surgical attitudes. There were twice the number of surgeons per capita in the United States compared to the United Kingdom. Numbers were similar to Sweden. Despite having very few spinal surgeons, the Netherlands proved to be quite aggressive in surgery. Sweden, despite having a large number of surgeons was conservative and produced relatively few surgeries. The most surgeries were done in the United States. In the UK, more than a third of non-urgent patients waited over a year to see a spinal surgeon. In Wales, more than half waited over three months for consult. Lower rates of referrals in the United Kingdom was found to discourage surgery in general. Fee for service and easy access to care was thought to encourage spinal surgery in the United States, whereas salaried position and a conservative philosophy led to less surgery in the United Kingdom. There were more spinal surgeons in Sweden than in the United States. However, it was speculated that the Swedish surgeons being limited to compensation of 40–48 hours a week might lead to a conservative philosophy. It should be noted that there have been calls for a more aggressive approach to lumbar surgery in both the United Kingdom and Sweden in recent years.