Background

Low back pain represents a major epidemiological problem, affecting at least 80% of adults during their lifetime. The intervertebral disc is a complex structure that has become the focus of much attention in clinical practice. The concept of internal disc disruption, with and without prolapse, was described by Crock, who introduced the concept and described the pathologic features of this pain entity. Recent literature (Guiot, Fessler) has shed further light into this phenomenon.

Recent studies indicate the existence of a biochemical/biomechanical model of discogenic pain and calls into question, the popular belief that a disc lesion can only cause pain by direct nerve compression from disc prolapse.

Internal disc disruption is a process by which biochemical changes involving phospholipase A₂, substance P and increased fibrinolytic activity ultimately lead to degeneration. The aging disc suffers progressive decline in nutrition, the result of an alteration in diffusion, diminished blood supply at the periphery of the annulus fibrosis and within the vertebral bodies, and decreasing matrix water (see figure A). It has been suggested that a reduction in proteoglycan synthesis, leads to decreased production and maintenance of the matrix, triggering the cascade of degeneration.

Repetitive axial loading of the intervertebral disc results in fatigue failure, disruption of internal collagen fibrils and progressive infiltration of the nucleus pulposes into the annular fibers. Eventually, sufficient degradation of the type I fibers occur, such that the weakened annulus bulges outward, due to pervasive displacement of the herniating nucleus pulposes (see figure B). Pain can occur as a result of compression of adjacent neural structures as well as by release of irritating by-products.

For some time, the present authors did perform intradiscal injections of corticosteroids. However, we did not observe a sustained benefit, on a consistent basis, though there was nearly universal pain relief, immediately post-procedure. We theorized that this pain relief was due to local anesthetic effect from xylocaine or bupivicaine, one of which was routinely mixed with the steroid.

During this time, we had been utilizing 25% Dextrose in water injection, to treat chronic pain arising from fibro-osseus junctions of ligament and tendon attachments. This is based upon research demonstrating that the dextrose mediates a local tissue effect that causes an influx of fibroblasts, which, over time, synthesize collagen at the injection site. The result is an increase in ligament thickness, mass and strength, the associated pain improving as the fibers stabilize.

While results were actually better than what we were experiencing with corticosteroid injections, we made two important observations:

1-Over 3-5 bi-weekly injections, the dextrose injections led to substantial reduction in pain even with performance of previously provocative activity.

2-Many patients experienced sustained reduction in pain after the first injection, quite inconsistent with the literature regarding anticipated outcome.

The authors recognized that the Type I collagen fibrous make-up of the peripheral tissues we were treating, is greatly similar to that of the annulus fibrosis of the intervertebral disc. We theorized that the dextrose must be acting as an osmotic neurolytic to local c-fiber (unmyelinated) nociceptors. Further, we hypothesized whether injection of the dextrose into the chronically deranged fibrous milieu of a degenerative lumbar disc, might not result in mechanical stabilization of the collagen-fibrous matrix as well as chemical pain suppression.

Groen et. al. (The American Journal of Anatomy 188:282-296/1990) created a new staining technique, utilizing an acetylcholinesterase whole-mount method, studied in the human fetus instead of animals, and with more sophisticated microdissection. They were able to identify four to six tiny nerve fibers interwoven within the nerve root canal predominantly located between the disc and nerve root and its ganglion. These nerve structures were found to originate from the sympathetic trunk, rami communicantes, and perivascular nerve plexuses of segmental arteries, ventrally supplying the anterior longitudinal ligament and dorsally supplying the posterior longitudinal ligament. The intervertebral discs are surrounded by a continuous ring of nerve fibers formed by branches of the sympathetic trunks. Even though the origin is solely sympathetic, they were able to differentiate nociceptive, proprioceptive, and vasomotor nerve fibers. These findings are important in the explanation of discogenic pain

Freemont, et al. (Lancet. Jul 19;350(9072):178-81, 1997.), found isolated nerve fibers that expressed substance P deep within diseased intervertebral discs, suggesting an important role for nerve growth in the pathogenesis of chronic low back pain. Forty-six samples of intervertebral discs were collected from thirty-eight patients during spinal fusion. Thirty samples were from pain levels clinically established by discography and sixteen were from adjacent vertebral levels with no pain. Thirty-four control samples were obtained from previously healthy individuals with normal histology within eight hours of recorded death. Standard immunohistochemical techniques were used to test for a general nerve marker, a nociceptive neurotransmitter (substance P), and a protein expressed during axonogenesis (growth-associated protein 43, or GAP 43).

Nerve fibers were identified in the outer third of the annulus fibrosus in forty-eight (60%) of the eighty samples of intervertebral discs. Nerves were restricted to the outer or middle third of the annulus fibrosus in the 34 control samples. In disc material obtained from patients with history of chronic low back pain, nerves extended into the inner third of the annulus fibrosus and into the nucleus pulposus in twenty-one (46%) and ten (22%) samples, respectively. Nerves usually accompanied blood vessels, but in fourteen of the samples from back-pain patients, isolated nerve fibers were seen in the discal matrix. Both types of nerve fibers expressed substance P, but only non-vessel associated fibers expressed GAP 43.

Deep nerve in-growth into the inner third of the annulus fibrosus, the nucleus pulposus, or both was seen in four (25%) of sixteen biopsy samples from non-pain levels and in seventeen (57%) of samples from pain levels. Of the sixteen paired samples from both pain and non-pain levels, five pain-level samples and one non-pain level sample showed deep nerve penetration.

In 1995, the North American Spine Society Diagnostic and Therapeutic Committee published its recommended indications for lumbar discography. They reported that this procedure can be used to evaluate an abnormal disc to assess the extent of abnormality or to investigate the correlation of the abnormality with clinical symptoms. Lumbar discography may be indicated in patients with persistent pain suspected to be discogenic, but whose results from other evaluations have been negative or inconclusive. For patients who are being considered for chemonucleolysis, percutaneous discectomy, or other intradiscal procedures, discography may help determine the degree of disc disruption and indicate the pattern of contrast spread.

Over the years, the present authors have performed many lumbar discograms, demonstrating an incidence of disc-mediated pain in approximately 50% of patients presenting with low back pain, with/without sciatica. Of these patients, some 50% respond to epidural corticosteroid injections combined with appropriate physiotherapy. Of those who failed, it became necessary to identify a more specific pathoanatomic cause. In lieu of neurologic abnormalities, surgical intervention was not indicated.

Materials and Methods

Between January 1998 and January 2001, 85 patients were evaluated for low back pain with/without sciatica in our diagnostic/treatment center. There were 53 males and 32 females. The age range of these patients was 20-86 years with a median of 48 years and an average age of 50.5 years. History and physical examination was combined with weight bearing plain film radiographs and CT or MRI. A number of patients had completed other testing, to include electrodiagnostic evaluation and radionuclide bone scan. All patients demonstrated findings consistent with discogenic (discopathy of mechanical or chemical type) pain suspected at one or more levels, but went on to fail a trial of two fluoroscopically guided epidural corticosteroid injections.

Discography is typically the next step in our diagnostic protocol, so as to correlate a level-specific concordantly provoked pain generator with the level or levels for which we have clinical suspicion. Despite an expected consistent reliability of test sensitivity in patients typically less than 40 years old, we had, in a number of patients with advanced desiccative disc changes, failed to reproduce concordant pain at any tested level. In these patients, the nucleogram and post-nucleogram CT revealed significant circumferential annular tearing and infiltration of the nucleus into the annulus, rendering an essentially impossible differentiation of the two regions. We suspect that the 0.5-1.0ml volume capacity of a normal nucleus is greatly increased as progressive degradation of the Type I collagen fiber annular lamellae, allows the above noted nuclear infiltration. In conjunction with the loss of annular elasticity, the normal disc mechanics are altered and we would propose that one of two possible events occur:

1- A complete annular tear is present and much of the injected contrast escapes into the epidural space or paravertebral soft tissues, disallowing adequate discal pressurization.

2- No tear is evident, however the now expanded nuclear region allows a much greater volume of contrast (up to 6cc or more) at which point the ratio of the volume injected: intradiscal pressure, becomes larger than any reasonably accurate decision of concordant pain response can be made.

As a result, we have largely abandoned routine use of discography in patients with desiccative disc changes of a moderately severe or severe degree. This decision was reached because, as clinicians, our responsibility is to often look beyond that which is not just black and white, but often gray, and treat based upon experience. We recognize that this reasoning is made possible by the knowledge that when spinal injection is performed in a technically proficient manner, a worst-case scenario is one in which we may fail to ameliorate the pain problem, but not typically worsen it. We do not necessarily endorse this thought process with respect to more invasive treatment, as a failed surgery, even when proficiently employed, can complicate the accurate diagnosis and treatment of residual pain.

Based upon this information, the authors created a treatment protocol that would involve intra-discal injection of 50% dextrose in water, mixed 50:50 with bupivicaine 0.25% plain, to a total volume of 3cc. In this manner, we achieved an end concentration of 25% Dextrose. All 85 patients received 3-5 bi-weekly injections, performed under fluoroscopic guidance and typically approached from the side of greatest symptomatic complaint. One half of the injectate was placed, as close to the center of the disc as the approach window would allow, based upon the local anatomy. The remainder was injected into the estimated middle and outer thirds of the disc, as the needle was withdrawn. An average rate of injection was utilized, so as to allow the injectate to reasonably diffuse into the discal environment.

In total, we completed 425 injections in 85 patients, in the 24-month period. The L5-S1 disc was treated in 52% of the patients and L4-5 was treated in 38%. The L3-4 disc was treated in 8% of the patients. Only 1 level was treated in 35 (41%) patients. The remaining 50 (58%) patients had two levels treated. No patient had three or more levels treated.

Results

We do not have concise outcome data at this time. However, a great many of these patients experienced significant to complete relief of their presenting complaints. There was concordant improvement in ability to perform activities of daily living and the need for pain medication declined or was able to be discontinued.

Conclusion

We do not wish to suggest that such treatment represents a cure for discogenic pain, nor do we believe that a “cure” currently exists. As clinicians and surgeons, we do not believe that we can successfully replace, at present, disc and associated tissues, which have become degenerated and dysfunctional, due to progressive or traumatic aberrations in mechanical and chemical function. However, we do believe that treatment of the type described, herein, is part of a number of advancements, aimed at controlling pain symptoms over the relative short-term. We have observed absolutely no untoward events during the rendering of this treatment. The treatment does typically need to be repeated about every 10-18 months, based upon current estimates.

We believe that there is significant merit to further study in this area of treatment. We recognize that this data does not represent results from a more controlled and blinded, randomized clinical trial. While such a study is now being designed, we did not want to withhold this information, even in its current form, as it is both an exciting and encouraging advancement in the realm of minimally invasive treatment of discogenic pain.

Keywords: Degenerative Disc Disease, Discogenic Pain, Discopathy, Epidural Steroid Injection, Minimally Invasive, Dextrose 25%, Low Back Pain, Osteoarthritis

References

1. Groen G, Baljet B, Drukker J: Nerves and nerve plexuses of the human vertebral column. Am J Anat 188:282-296, 1990

2. Freemont AJ, Peacock TE, Goupille P, et al.: Nerve ingrowth into diseased intervertebral disc in chronic low back pain. Lancet 3350:178-181, 1997

3. Naylor A: Intervertebral disc prolapse and degeneration: The biochemical and biophysical approach. Spine 1:108-115, 1978

4. Buckwalter J: Spine Update; Aging and Degeneration of the Human Intervertebral Disc. Spine 11:1307-1313, 1995

Prolotherapy^1-4 is defined as injecting dextrose into joints and ligaments to produce a proliferating response and healing process. The origins date back to Hippocrates who employed a version of the technique for torn shoulder ligaments. Various conditions including temporomandibular joint syndrome, carpal tunnel syndrome, sacroiliitis, osteoarthritis of the spine, and instability of ligaments of the knee have all been treated. Hacket¹ in 1958 claimed a cure rate of 82% in 1600 patients with low back pain. Further studies of the basic mechanism and reports of clinical trials should be undertaken.

1. Hackett GS: Ligament and Tendon Relaxation Treated by Prolotherapy. 3^rd Edition, Charles C. Thomas, Springfield, IL, 1958.

2. Klein RG, Dorman TA, Johnson CE: Proliferant injections for low back pain. J. Neurol Orthop med Surg 10:123-126, 1989

3. Leedy RF: Basic techniques of sclerotherapy. Oseteop Med 9:1-9, 1987

4. Onigey MJ, Klein RG, Dorman TA, et al. A new approach to the treatment of low back pain. Lancet 11:143-146, 1989