Publications
APS Bulletin • Volume 11, Number 4, July/August 2001
Innovations in Practice
Norman Harden, M.D., Department Editor
Intradiscal Electrothermal Annuloplasty for the Management of Chronic Discogenic Pain: A Review of Current Concepts and the Literature
Gregory N. Arends, MD
Department Editor’s Note: When encountering new technology like annuloplasty I have to remind myself that 15 years ago biofeedback therapy was considered very unusual, somewhat “alternative.” Of course now biofeedback is the standard of care and is supported by a wealth of evidence-based medicine. In the early going the techniques such as annuloplasty may seem very aggressive, interventional, and experimental. In fact, until evidence is generated, it must be considered experimental. At the moment we must go on empirical and anecdotal evidence, and the fact that mechanically the technique at least makes sense. It also is the absolute responsibility of all doctors who would offer this clinically to properly inform their patients that this is an unproven and experimental technique, and it is questionable whether it is the responsibility of insurance companies to pay for all techniques of this stature. It also is those doctors’ responsibility to generate randomized, controlled trials to support the technique within a reasonable period. That being said, we must keep an open mind; who knows which of the range of odd-looking therapies today will be the standard of care tomorrow.
Discogenic pain is believed to be the etiology of many cases of chronic persistent lumbar pain (Schwarzer, Aprill, & Derby, 1995). The fact that the posterior outer annulus is innervated and a potential pain generator has been well established (Coppes & Marani, 1997). Diagnosis of discogenic pain relies on clinical impression, imaging techniques, and provocative discography (Moneta & Videman, 1994). Treatment has long been limited to conservative measures and open surgery. It is apparent that a vacuity has existed between the two forms of treatment.
Minimally invasive techniques have been developed to span this gap in the treatment armamentarium. Advances in imaging and technique allow for safe intradiscal access. Intradiscal injection of corticosteroids has been performed for years (Feffer, 1975). This technique lacks evidence-based validation and may not yield lasting effects due to the limited biologically active life of the injectate. Further, it does not significantly alter the suspect patho-anatomy. Radiofrequency (RF) lesioning of the posterior annulus from within the disk has been performed with limited success (Houpt & Conner, 1996). RF energy is neuroablative, and the heat produced may cause some collagen modification (Letcher & Goldring, 1968). The treatment zone of an RF probe is small, and high focal temperatures are produced that may injure healthy tissues. The technique of intradiscal electrothermal annuloplasty (IDEA), or intradiscal electrothermal therapy (IDET) as it is commercially known, was developed to allow for more precise and controlled thermal delivery and allows for potential definitive treatment for this condition through a minimally invasive procedure. Other minimally invasive techniques are being developed and used such as Coblation-assisted NucleoplastyTM. With this technique, RF energy is used to molecularly disintegrate tissue thereby debulking a pressure-sensitive disk. Bipolar RF coagulation further denatures proteoglycans, altering the internal milieu of the affected nucleus. This technique purports to treat axial and radicular pain (contained lumbar disks only) and is currently undergoing clinical trials. IDEA was introduced in 1997, and studies for its efficacy and safety are numerous. A large, prospective, controlled clinical trial is under way in Australia but has not been completed. The current concepts and literature on IDEA are reviewed herein.
IDEA candidates
Candidates for IDEA have had a clinical diagnosis of discogenic low-back pain for more than 6 months and have failed an exhaustive conservative treatment regimen. Ideal IDEA candidates have uni-segmental disease, or a single affected disk level as determined by magnetic resonance imaging (MRI) and provocative discography (Aprill, 1997) with normal control disks. A “classic” presentation would include a history of axial low-back pain without radicular symptoms and pain exacerbated with lumbar flexion and/or prolonged sitting. Pain typically increases with maneuvers that increase intradiscal pressure (e.g., Valsalva’s manuever). Physical examination should disclose a lack of “hard” neurologic findings in the lower limbs. A positive spring-test (subjective reproduction of axial back pain with direct pressure over the suspect spinous process) at the affected lumbar segment is often present on prone examination. Significant zygapophysial disease should be ruled out with imaging and clinical examination. An MRI would demonstrate single-level “dark-disk disease” (disk dehydration or desiccation), some loss of disk height (not greater than 50%), and possibly a high-intensity zone posteriorly on saggital views (Aprill & Bogduk, 1992). Provocative discography would then be used to rule in the suspect level of pathology and rule out multisegmental disease. IDEA may be performed on multiple levels, but there is less likelihood of a successful result than with a single-level procedure.
How IDEA works
The IDEA procedure involves gaining intradiscal access via a trochar and introducing a flexible catheter into the nucleus that is positioned in a circumferential manner along the inner annulus. When optimal positioning is achieved, the catheter is heated to a target catheter temperature of 90 C in a gradually increasing 17-minute heating protocol. The theory behind IDEA is to affect the disk by two means: The high temperature is transmitted through the posterior annulus and causes thermal destruction of the afferent nociceptors (branches off the sinuvertebral nerve) that are abundant in the posterior outer third of the annulus (Bogduk & Tynan, 1981). This theoretically blocks further discogenic-pain transmission. Thermal-mapping studies have been performed that demonstrate that the average outer temperature of the annulus during IDEA is about 43 C (Ashley, Gharpuray, Saal, & Saal, 1999). Studies have shown that thermal destruction of these small nociceptors occurs at 42 C–45 C (Houpt & Conner, 1996). The second mechanism of action is that the high temperatures cause collagen contraction by denaturing the protein molecules of the annulus (Naseef & Foster, 1997). This causes a debulking of the disk volume by as much as 12% and may afford greater structural stability to the functional spinal unit in this contracted state. It has been determined that collagen is denatured and contracts at a temperature of 65 C (Naseef & Foster). This temperature is easily achieved in the annular tissue that is in contact with the 90C catheter used for IDEA.
The IDEA procedure
The procedure begins with the patient prepared in sterile fashion prone on the procedure table. Anatomical landmarks are identified under fluoroscopy from an oblique projection, and local anesthetic is applied. Access to the nucleus is gained with fluoroscopic guidance via a standard extra-pedicular approach with a 17-gauge trochar. Avoiding the segmental nerves is extremely important. The patient must be only lightly sedated so as to be able to inform the practitioner of any radicular pain or irritation. Entry to the disk is made from the side opposite the suspect pathology. This is to ensure optimal positioning of the catheter over the region of the annulus suspect for producing the painful syndrome. Once the annulus has been breeched, the flexible thermal-resistive catheter is advanced under fluoroscopy from a lateral view to make certain the peritoneal cavity is not embarrassed. The catheter should “hit” the anterior annulus and then “turn” posteriorly following the confines of the inner annulus until the catheter has spanned a majority of the posterior annulus and positively the area of suspect pathology. The catheter is then heated for approximately 17 minutes following manufacturer-provided protocol guidelines. It is imperative that the patient remains conscious during the thermal delivery to relate any untoward radicular symptomatology. The catheter is then removed. When performing this procedure, I also administer intradiscal antibiotics before the removing of the trochar. The patient is recovered and discharged to home with activity guidelines when stable. A lumbar support brace may be issued for a brief postoperative period. Relative rest is recommended for 1–3 days postprocedurally, with return to normal activities of daily living at 1 week. A postoperative rehabilitation program promoting flexibility and stabilization is provided and may require therapist guidance and is usually initiated by the second postoperative week. Some patients respond well and experience rapid resolution of symptoms. Other patients may take as long as 3–6 months postprocedurally to appreciate maximum benefit. These cases may warrant re-evaluation at that time.
Though IDEA is a percutaneous, minimally invasive technique, it is not without patient risk and complications. As with any interventional procedure, infection is possible, including the potential for discitis. I routinely prophylaxes with preintravenous, postintravenous, and intradiscal antibiotics. Thermal injury to the segmental nerves also is possible and could lead to irreversible nerve damage if not detected promptly. This is the reason for light sedation and constant patient monitoring. A case of iatrogenic cauda equina syndrome from IDEA has been reported (Hsia & Isaac, 2000).
Because this is a fairly new interventional tool, no large, randomized controlled trials have been published to date. Several case-control and retrospective studies have been published. The largest and most recent completed studies to date are presented here for review.
Literature review
Karasek and Bogduk (2000) recently conducted a well-designed case-control study. Of 110 patients evaluated with lumbar pain, 53 patients were diagnosed with internal disk disruption (IDD) with provocative discography and postdiscography computed tomography scan identifying a pressure-sensitive disk and a radial tear extending to the outer third of the annulus (but not through the annulus). Thirty-six patients were “cleared” by their insurance carriers for IDEA. The control group comprised the 17 patients who also were “denied” the treatment. This presents somewhat of a selection bias but is a fair and ethical method of semi-randomization. The control group underwent a comprehensive, conservative treatment regimen, including physical therapy, strengthening and conditioning exercises, and education and counseling, and was followed for 3 months. The treatment arm underwent IDEA, and these patients were followed for 1 year.
At 3 months, only 1 patient in the control group had dramatic improvement, 3 had modest improvement, 4 had no improvement, and 9 had worsening of their condition. Pain scores as determined by a visual analog pain scale (VAS) were not improved. Of the five patients originally not working, only one returned to work. Of the 10 patients working prior to treatment, three stopped working. Analgesic use was not significantly affected. Because most of the control group faired poorly with comprehensive, conservative treatment, further follow-up past 3 months was not possible because these patients sought alternative treatment.
At 12 months for the IDEA group, 32 of the 36 patients had statistically significant (p < .001) decreases in their VAS scores, with a median decrease of three points. This was clinically and statistically significantly (p < .001) different from the control group. Eight percent of the treatment group experienced a more than 90% decrease in reported pain, whereas no patients in the control group had this amount of relief. Half of the treatment group experienced at least a 50% overall reduction in reported pain compared with only 6% of the control group. Sixty percent of the treated patients achieved at least a 50% reduction in their pain to a (VAS) level of 4 or less, with 23% experiencing complete relief from their pain.
All of the treated patients either continued working or returned to work during the study period.
Of the 19 patients who received at least 50% pain reduction at 12 months, 11 discontinued opioid medication use. Five continued to take opioids but at reduced doses. Three continued opioid usage (for reported non-back related pathology).
Though not a prospective study, and lacking true randomization, this otherwise well-designed study demonstrated that in carefully selected cases a skilled practitioner can perform IDEA and substantially reduce or eliminate discogenic pain, with a success rate as high as 60%, and improve function and reduce opioid consumption. It is superior to conventional conservative care.
Saal and Saal (2000) completed another study recently with 1-year follow-up. Their prospective case series showed results similar to Kareasek and Bogduk’s. In this study, 62 patients were selected for IDEA from a pool of 1,116 low-back pain sufferers having failed 6 months of aggressive-conservative care and met the rigid inclusion/exclusion criteria for the study. Follow-up was at 1 year and outcomes were analyzed using preprocedural and postprocedural VAS scores, SF-36 scores, and return-to-work status. None of the treated patients experienced complications from IDEA.
Of the 62 treated patients, 44 patients (71%) improved their pain rating on a VAS, showing a 3.0 mean decrease of the VAS score (p < .001). This value improved if the few patients with multisegmental disease were excluded from the analysis. The uni-segmental group VAS score showed a mean change (decrease) of 3.4 (p < .001) and was 2.6 (p < .001) in the multilevel group.
The Health Status Questionnaire Short Form (SF)-36 Physical Function subscale and Bodily Pain subscales were analyzed (Gatchel & Polatin, 1998). Physical function scores improved in 71% of the patients and demonstrated a mean change for the entire cohort of 20 (p < .001). Again, when uni-level disease was isolated the scores increased to a mean change of 23.6 (p < .001) and decreased to 17 (p < .010) in the multilevel group. The bodily pain scores improved in 74% of the patients. The mean change in the bodily pain scores was 17.4 (p < .001): 16.8 (p < .001) for the single-level group, and 18 (p < .001) for the multilevel group. Six patients did not improve on either SF-36 subscale.
Ninety-two percent of the patients returned to work: 97% of the patients with private insurance and 83% of the patients on workers’ compensation. All patients who had private insurance and were working at baseline continued to work after treatment, and of those not working at baseline, 19 of 20 returned to work. Patients working sedentary jobs at the time of the procedure returned to work in 2 weeks after the procedure. Patients with more strenuous jobs returned to work in 4–6 months, or were retrained to lighter job duties and returned to work in 1–3 months.
This prospective study demonstrated that discogenic pain unresponsive to aggressive-conservative care may be successfully treated with IDEA with statistically significant improvements in pain and function (as great as 70%), and return to work at 1-year follow-up. Limitations of this study were that it lacked a control group and that there was no randomization of treated subjects. The study has clinical significance, but is deficient in delivering true evidence-based science.
Summary
As these studies show, IDEA seems to be an efficacious treatment option for carefully selected patients with recalcitrant lumbar discogenic pain who have failed conservative treatment. The recent data suggests an efficacy of roughly 60%–70%, which is in accordance with my clinical results. Clearly, IDEA is in its infancy and demands the scrutiny of prospective, double-blinded, placebo-controlled studies. IDEA is not without risk and needs to be performed only by skilled hands and on carefully selected patients. The overall success of IDEA seems to be similar to, if not superior to, fusion surgery (Blumenthal & Baker, 1988; Knox & Chapman, 1993) for select patients. This is without the morbidity and mortality of a large operation and the likely development of junctional disease above or below the fusion site (Kostuik, 1997). It is not a replacement for fusion techniques, but does fill the gap that has existed between conservative and surgical options.
References
Aprill, C. (1997). Diagnostic disc injection II: Diagnostic lumbar injection. In J.W. Frymoyer (Ed.), The adult spine: Vol. 1. Principles and practice. (2nd ed., pp. 39–62). Philadelphia: Lippincott-Raven.
Aprill, C., & Bogduk, N. (1992). High intensity zone: A diagnostic sign of painful lumbar disc on MRI. British Journal of Radiology, 65, 361–369.
Ashley, J.E., Gharpuray, V.M., Saal, J.S., & Saal, J.A. (1999). Temperature distribution in the intervertebral disc: A comparison of intranuclear radio-frequency needle to a novel heating catheter. BED Vol. 42, 1999 Bioengineering Conference ASME.
Blumenthal, S.L., & Baker, J. (1988). The role of anterior lumbar fusion for internal disc disruption. Spine, 12, 566–569.
Bogduk, N., & Tynan, W. (1981). The nerve supply to the human lumbar intervertebral disc. Journal of Anatomy, 132, 39–55.
Coppes, M.H., & Marani, E. (1997). Innervation of “painful” lumbar discs. Spine, 22 (20), 2342–2349
Feffer, H.L. (1975). Regional use of steroids in the management of lumbar intervertebral disc disease. Orthopedic Clinics of North America, 6(1), 249–253.
Gatchel, R.J., & Polatin, P.B. (1998). Use of the SF-36 health status survey with a chronically disabled back pain population: Strengths and limitations. Journal of Occupational Rehabilitation, 8, 237–246.
Houpt, J., & Conner, E. (1996). Experimental study of temperature distributions and thermal transport during radiofrequency current therapy and the intervertebral disc. Spine, 21, 1808–1813.
Hsia, A.W., & Isaac, K. (2000). Cauda equina syndrome from intradiscal electrothermal therapy. Neurology, 55(2), 320.
Karasek, M., & Bogduk, N. (2000). Twelve-month follow-up of a controlled trial of intradiscal thermal annuloplasty for back pain due to internal disc disruption. Spine, 25 (20), 2601–2607.
Knox, B.D., & Chapman, T.M. (1993). Anterior lumbar interbody fusion for discogram concordant pain. Journal of Spinal Disorders, 6, 242–244.
Kostuik, J. (1997). Failures after spinal fusion. In J.W. Frymoyer (Ed.), The adult spine: Vol. 1. Principles and practice. (2nd ed., pp. 2277–2328). Philadelphia: Lippincott-Raven.
Letcher, F., & Goldring, S. (1968). The effect of radiofrequency current and heat on peripheral nerve action potential in the cat. Journal of Neurosurgery, 29, 42–47.
Moneta, G.B., & Videman, T. (1994). Reported pain during lumbar discography as a function of annular ruptures and disc degeneration: A re-analysis of 833 discograms. Spine, 17, 1968–1974.
Naseef, G.S., & Foster, T.E. (1997). The thermal properties of bovine joint capsule: The basic science of laser and radiofrequency-induced capsular shrinkage. American Journal of Sports Medicine, 25, 670–674.
Saal, J., & Saal, J. (2000). Intradiscal electrothermal treatment (IDET) for chronic discogenic low back pain: A prospective outcome study with one year follow-up. Spine, 25 (20), 2262–2267
Schwarzer, A., Aprill, C., & Derby, R. (1995). The prevalence and clinical features of internal disc disruption in patients with chronic low back pain. Spine, 20, 1878–1883.
Gregory Arends is a member of the Chicago Institute of Neurosurgery and Neuroresearch (CINN), and a founding member of the CINN Institute for Spine Care. He is a board-certified physician with an expertise in the diagnosis and treatment of patients with occupational- and sports-related injuries.