When any of the following surgical procedures are covered for the treatment of discogenic back pain, only one invasive modality or procedure may be considered medically necessary at a surgical session.
Chemonucleolysis using chymopapain may be considered medically necessary as an alternative to surgery in patients when the following criteria are met:
- Physical (radicular symptoms) and diagnostic imaging evidence of an uncomplicated herniation of a single lumbar disc; and
- No evidence of a free fragment or sequestration of a portion of a disc; and
- Pain is not relieved by at least six weeks of conservative therapy.
Chemonucleolysis using chymopapain is considered not medically necessary in patients who demonstrate any of the following:
- Known or suspected allergy to papaya extracts, including chymopapain and related chemicals;
- A history of previous chemonucleolysis by chymopapain injections;
- Prior surgical treatment on the disc presently suspected to harbor a symptomatic herniation;
- Progressive neurological dysfunction;
- Impairment of bowel or bladder function;
- Evidence of a sequestered disc or free fragment of disc;
- Evidence of vertebral disease such as spinal stenosis or spondylolisthesis.
Chemonucleolysis is considered experimental, investigational and unproven for use in the cervical and thoracic disc regions.
AUTOMATED PERCUTANEOUS DISCECTOMY
Percutaneous discectomy by any means is considered experimental, investigational and unproven as a technique of intervertebral disc decompression in patients with back pain related to disc herniation in the lumbar, thoracic or cervical spine.
AXIAL LUMBOSACRAL INTERBODY FUSION (Axial LIF)
Axial lumbosacral interbody fusion (axial LIF) is considered experimental, investigational and unproven.
PERCUTANEOUS INTRADISCAL ELECTROTHERMAL (IDET) ANNULOPLASTY AND PERCUTANEOUS INTRADISCAL RADIOFREQUENCY ANNULOPLASTY
Percutaneous annuloplasty (e.g., intradiscal electrothermal annuloplasty, percutaneous intradiscal radiofrequency thermocoagulation, or intradiscal biacuplasty) for the treatment of chronic discogenic back pain is considered experimental investigational and unproven.
DECOMPRESSION OF THE INTERVERTEBRAL DISC USING LASER ENERGY (LASER DISCECTOMY) OR RADIOFREQUENCY COBLATION (NUCLEOPLASTY)
Laser discectomy and radiofrequency coblation (disc nucleoplasty) are considered experimental, investigational and unproven as techniques of disc decompression and treatment of associated pain.
Code 62287 describes any method of decompression of intervertebral disc. Therefore, based on this code alone, it might not be possible to distinguish among automated percutaneous discectomy, laser discectomy, or DISC nucleoplasty™.
Code 77002 may be used to describe the radiologic guidance.
A specific S code is available for the radiofrequency decompression procedure – S2348.
NOTE: Intradiscal electrothermotherapy or disc biacuplasty are being performed in the office setting.
A review of 105 consecutive cases of chymopapain chemonucleolysis for single level lumbar disc herniation was undertaken. Mean follow-up was 12.2 years (range 10—15.3). Patients were assessed using the Oswestry Disability Questionnaire. Eighty-seven patients were available for follow-up. An excellent or good response occurred in 58 patients (67%); four patients (4.5%) had a moderate response but were only minimally disabled. The treatment failed in 25 patients (28.5%) and 21 of these went on to surgery within a mean of 5.2 months (range 3 weeks—12 months). In 15 patients (71%) disc sequestration or lateral recess stenosis was found. Five of the remaining six cases had a large disc herniation at surgery. Surgery resulted in a significant improvement in nine cases. Discitis following chemonucleolysis occurred in six patients (5.7%). Chymopapain chemonucleolysis has a useful role in the management of lumbar intervertebral disc prolapse. However, its efficacy is dependent on careful clinical and radiological patient selection.
Chemonucleolysis is now an established treatment modality in the management of prolapsed lumbar intervertebral disc disease. The procedure has been in widespread use for over 30 years since its introduction by Lyman Smith in 1963. There is sufficient evidence to show that chymopapain is both safe and efficacious, although its use remains controversial. The authors report long-term results of patients treated with chymopapain for single level lumbar intervertebral disc prolapse at our institution.
AUTOMATED PERCUTANEOUS DISCECTOMY
Originally based on a 1990 Blue Cross Blue Shield Association (BCBSA) Technology Evaluation Center Program (TEC) Evaluation, which concluded that percutaneous discectomy met the TEC criteria. Therefore, the conclusion was made that percutaneous discectomy was considered medically necessary in carefully selected patients. Since the 1990 TEC Evaluation, the methodology of evidence-based medicine in general has grown in sophistication. Specifically, it is recognized that randomized clinical trials are extremely important to assess treatments of painful conditions and low back pain in particular, due both to the expected placebo effect, the subjective nature of pain assessment in general, and also the variable natural history of low back pain that often responds to conservative care. Disc decompression using lasers or radiofrequency energy to ablate disc material is a relatively new minimally invasive technique that is considered an alternative to percutaneous discectomy, and both techniques are considered investigational, in part due to the lack of controlled trials.
A specific focus was placed on controlled clinical trials comparing percutaneous discectomy to either open discectomy or conservative therapy. The literature search identified a large number of case series, but only five controlled trials, of which four were reviewed in a 2000 Cochrane report. (All of these controlled trials were published after 1990 and thus were not reviewed as part of the TEC Evaluation.) The Cochrane review concluded, “Three trials of percutaneous discectomy provided moderate evidence that it produces poorer clinical outcomes than standard discectomy or chymopapain.” For example, Chatterjee reported on the results of a study that randomized 71 patients with lumbar disc herniation to undergo either percutaneous discectomy or lumbar microdiscectomy. A successful outcome was reported in only 29% of those undergoing percutaneous discectomy compared to 80% in the microdiscectomy group. The trial was halted early due to this inferior outcome. In a 1993 randomized study, Revel and colleagues compared the outcomes of percutaneous discectomy to chymopapain injection in 141 patients with disk herniation and sciatica. Treatment was considered successful in 61% of patients in the chymopapain group compared to 44% in the percutaneous discectomy group. Another trial cited in the Cochrane review, Mayer et al., is not applicable since the technique used modified forceps in addition to a suction probe. Finally, the last trial cited in the Cochrane review, Hermantin et al., provided insufficient data to allow detailed analysis of results.
The only additional controlled study published since the 2000 Cochrane review was the LAPDOG study, a randomized trial designed to compare percutaneous and open discectomy in patients with lumbar disc herniation. This trial was designed to recruit 330 patients, but was only able to recruit 36 patients, for reasons that were not readily apparent to the authors. Of the evaluable 27 patients, 41% of the percutaneous discectomy patients and 40% of the conventional discectomy patients were assessed as having successful outcomes at six months. The authors concluded that this trial was unable to enroll sufficient numbers of patients to reach a definitive conclusion. The authors state, “It is difficult to understand the remarkable persistence of percutaneous discectomy in the face of a virtually complete lack of scientific support for its effectiveness in treating lumbar disc herniation.”
The most recent literature search was performed for the period September 2009 through December 2010. The following is a summary of the key literature to date:
In 2007, Gibson and Waddel published an updated Cochrane review of surgical interventions for lumbar disc prolapse, concluding that there is insufficient evidence on percutaneous discectomy techniques to draw firm conclusions. In the same year, a task force of the American Society of Interventional Pain Physicians reports that percutaneous disc decompression remains controversial; although all observational studies were positive, the evidence from four of four randomized published studies was negative. Questions also remain about the appropriate patient selection criteria (particularly related to the size and migration of the disc herniation) for this procedure.
Freeman and Mehdian assessed the current evidence for three minimally invasive techniques used to treat discogenic low back pain and radicular pain: electrothermal therapy (IDET), percutaneous discectomy, and nucleoplasty in a 2008 paper. They reported that trials of automated percutaneous discectomy suggest that clinical outcomes are at best fair and often worse when compared with microdisctomy.
Two systematic reviews published in 2009 analyzed the literature for different devices. Hirsch and colleagues reviewed four randomized controlled trials (RCTs) and 76 observational studies in their analysis of studies in which the Nucleotome was used. In two of the RCTs, the comparator was chemonucleolysis. One of those RCTs was reviewed for a previous update. The second did not meet Cochrane review criteria for randomized controlled trials. The other two RCTs compared automated percutaneous discectomy with microdiscectomy and also failed to meet study quality criteria. Singh et al. performed a systematic analysis of studies in which the Dekompressor device was used; no RCTs were identified.
All of the trials reviewed here focused on lumbar disc herniation. There were no clinical trials of percutaneous discectomy of cervical or thoracic disc herniation.
The National Institute for Health and Clinical Excellence (NICE) published guidance in 2005 indicating that there is limited evidence of efficacy based on uncontrolled case series of heterogeneous groups of patients and evidence from small randomized controlled trials shows conflicting results. The guidance states that in view of uncertainty about the efficacy of the procedure, it should not be done without special arrangements for consent and for audit or research.
There is insufficient evidence obtained from well-designed and executed randomized controlled trials to evaluate the impact of automated percutaneous discectomy on net health outcome; thus, the procedure is considered experimental, investigational and unproven.
PERCUTANEOUS INTRADISCAL ELECTROTHERMAL (IDET) ANNULOPLASTY AND PERCUTANEOUS INTRADISCAL RADIOFREQUENCY ANNULOPLASTY (PIRFT)
The most recent literature update was performed in July 2010. As with any therapy for pain, a placebo effect is anticipated, and thus randomized placebo-controlled trials are necessary to investigate the extent of the placebo effect and to determine whether any improvement with annuloplasty exceeds that associated with a placebo. Therefore, evidence reviewed for this policy focuses on randomized controlled trials.
In 2007, a systematic review of IDET and PIRFT was published that followed the criteria recommended by the Cochrane Back Review Group. Four randomized and two nonrandomized studies, totaling 283 patients, were included in the review (the relevant studies are described below). The report concluded that the available evidence does not support the efficacy or effectiveness of IDET or PIRFT and that these procedures are associated with potentially serious side effects.
Intradiscal Electrothermal Annuloplasty (IDET™)
In 2003, Pauza and colleagues published the results of a randomized study, which was the focus of discussion in the 2003 BCBSA TEC Assessment. The study included 64 patients with low back pain of greater than six months’ duration who were randomly assigned to receive either IDET™ or a sham procedure. Visual analogue scale (VAS) pain was reduced by an average of 2.4 cm in the IDET group, compared with 1.1 cm in the sham group, a significant difference between groups (p=0.045). The mean change in the Oswestry Disability Scale (ODS) was also significantly greater for the IDET group compared with the sham group. The improvement on the SF-36 Bodily Pain subscale was significantly higher for the IDET group. The authors stated that per-protocol; analyses were conducted, which excluded data from eight patients, five from the IDET group and three from the sham group. One patient died, one was lost to follow-up, one had unsatisfactory electrode placement, one had post-treatment bone fracture, and two had new injuries unrelated to low back pain and were excluded due to compensation claims or opioids. Besides failing to perform intent-to-treat analyses, there are additional concerns about statistical methods used by Pauza et al. The report noted that the analysis of SF-36 Role Physical scores adjusted for differences at baseline, but whether the comparison used adjustment and statistical techniques was not specified. The technique for comparing group scores on continuous variables was described only as a t-test, suggesting simple comparison of mean change at follow-up. More appropriate techniques for comparing changes between groups include analysis of covariance and repeated measure analysis of variance. The comparison of means on the VAS for pain and the ODS for disability do not readily reveal how often patients achieve a clinically significant improvement. Minimally significant improvement in VAS has been estimated at 1.8–1.9 cm, and by this estimate, the mean change in VAS of 2.4 cm for IDET would be considered clinically significant. However, a small number of extreme values can influence this measure. The study also reported the percentage with a change in VAS of more than 2.0 cm, which is greater than the minimally clinically significant improvement of 1.8–1.9. When the VAS is dichotomized in this way, a relative risk of 1.5 is observed with a 95% confidence interval of 0.82–2.74.
The Pauza et al. trial is well-designed with respect to randomization, clear description of intervention, and use of valid and reliable outcomes measures. However, this single center trial does not permit conclusions about the relative effects of IDET and placebo. The study did not conduct intent to treat analysis, and it is unclear whether IDET achieves clinically and statistically significant improvements in measures of pain, disability, and quality of life.
A 2005 double-blinded randomized controlled trial (RCT) with 57 patients (38 IDET, 19 placebo) found IDET™ to be no more effective than sham stimulation, and no subject in either group achieved a successful outcome. In another study, comparison of 21 electrothermal (IDET) and 21 radiofrequency procedures found significant improvements in a majority of IDET patients but not in matched radiofrequency-treated patients at one-year follow-up; the study did not have a placebo-control group. An industry funded meta-analysis and systematic review were recently published that support the use of IDET. However, the quality of the studies included in these reviews was poor; 14 of the 18 studies reviewed did not have appropriate controls.
Evidence-based guidelines from the American Society of Interventional Pain Physicians in 2007 concluded that the evidence is moderate for management of chronic discogenic low back pain with IDET. Complications include catheter breakage, nerve root injuries, post-IDET disc herniation, cauda equine syndrome, infection, epidural abscess, and spinal cord damage.
Percutaneous Intradiscal Radiofrequency Thermocoagulation (PIRFT)
There is relatively minimal published data on PIRFT. In 2001, Barendse and colleagues reported on a double-blind trial that randomly assigned 28 patients with chronic low back pain to undergo PIRFT or to a sham control group. The primary outcome was the percentage of success at eight weeks, as measured by changes in pain level, impairment, ODS, and analgesics taken. At the end of eight weeks, there were two treatment successes in the sham group compared to one in the treatment group. The authors concluded that PIRFT was not better than the placebo procedure in reducing pain and disability.
Evidence-based guidelines from the American Society of Interventional Pain Physicians in 2007 found the evidence for radiofrequency posterior annuloplasty (PIRFT) to be limited, with complications similar to IDET (catheter breakage, nerve root injuries, post-IDET disc herniation, cauda equina syndrome, infection, epidural abscess, and spinal cord damage).
In 2009, Kvarstein and colleagues published 12-month follow-up from a RCT of intra-annular radiofrequency thermal disc therapy using the discTRODE™ probe from Radionics. Recruitment was discontinued when blinded interim analysis of the first 20 patients showed no trend toward overall effect or difference in pain intensity between active and sham treatment at six months. At 12 months, there was a reduction from baseline pain, but no significant difference between the two groups. Two patients from each group reported an increase in pain. Although this controlled study did not find evidence for a benefit of PIRFT, it may not have been powered to detect a small or moderate effect of the procedure.
One case report of transdiscal radiofrequency annuloplasty using two transdiscal probes (biacuplasty) was identified in 2007; the authors indicate this to be the first publication with this procedure.
A search of ClinicalTrials.gov. in July 2010 identified a small Phase I randomized double-blind placebo-controlled trial of transdiscal radiofrequency annuloplasty using two transdiscal probes (NCT00750191) by the same principal investigator as above. The posting lists an estimated enrollment of eight subjects with a final study collection date of September 2010 for the primary 12-month outcome measure.
The United Kingdom’s National Institute for Health and Clinical Excellence (NICE) guidance, published in 2004, indicates that the current evidence on safety and efficacy of percutaneous intradiscal percutaneous radiofrequency thermocoagulation for lower back pain does not appear adequate to support its use. NICE guidance on electrothermal annuloplasty was updated in 2009. NICE considers current evidence on the safety and efficacy of percutaneous intradiscal electrothermal therapy for low back pain to be inconsistent. NICE recommends that this procedure only be used with special arrangements for clinical governance, consent, and audit or research.
In addition to the systematic reviews described above, a number of other reviews from 2008 and 2009 have varying conclusions about the evidence for IDET annuloplasty; these reviews found no evidence to support a role for radiofrequency annuloplasty. Evidence is insufficient to conclude that these procedures improve health outcomes. Therefore, annuloplasty (i.e., IDET™, PIRFT, and biacuplasty) are considered experimental, investigational and unproven.
DECOMPRESSION OF THE INTERVERTEBRAL DISC USING LASER ENERGY (LASER DISCECTOMY) OR RADIOFREQUENCY COBLATION (NUCLEOPLASTY)
Randomized, placebo-controlled trials are considered particularly important when assessing treatment of low back pain to control not only for the expected placebo effect but to also control for the variable natural history of low back pain, which may resolve with conservative treatment alone.
Laser discectomy has been practiced for more than 20 years, and there is fairly extensive literature that describes different techniques using different types of lasers. In 2003, Gibson and colleagues published a Cochrane review of surgery for lumbar disc prolapse, which included a review of laser discectomy. The review aimed to determine the relative treatment effectiveness of laser discectomy compared to either no treatment, discectomy, or automated percutaneous discectomy. The review also included chemonucleolysis and open surgical discectomy. In their overall review of all surgeries, 27 randomized controlled clinical trials were identified, but none addressed laser discectomy. This review concluded that unless or until better scientific evidence is available, laser discectomy should be regarded as a research technique. In a 2007 paper, Goupille et al. reviewed the literature on laser disc decompression and concluded that “although the concept of laser disc nucleotomy is appealing, this treatment cannot be considered validated for disc herniation-associated radiculopathy resistant to medical treatment.” They cite the lack of consensus regarding technique, the questionable methodology and conclusions of published studies, and the absence of a controlled study in their discussion.
A 2007 updated Cochrane review of surgical interventions for lumbar disc prolapse included two comparative studies reported in congress proceedings and abstracts. One study, comparing two types of lasers, did not report comparative outcome results, and the other, which compared laser discectomy with chemonucleolysis, reported limited results favoring chemonucleolysis. The authors concluded that clinical outcomes following automated discectomy and laser discectomy “are at best fair and certainly worse than after microdiscectomy, although the importance of patient selection is acknowledged.” Singh et al. reported in 2009 that, based on a systematic review of current evidence, there is Level II-2 evidence for percutaneous laser disc decompression for short- and long-term relief of pain “which is equivalent to automated percutaneous lumbar disc decompression.” Evidence was rated according to U.S. Preventive Services Task Force (USPSTF) criteria; Level II-2 describes studies from well-designed cohort or case-control analytic studies, preferably from more than one center or research group.
Other than the comparative studies mentioned above, the evidence for laser discectomy is limited to case series. Choy described the largest series of 1,275 patients treated with 2,400 procedures (including cervical, thoracic, and lumbar discs) over a period of 18 1/2 years, reporting an overall success rate, according to the MacNab criteria (measuring pain and function) of 89%. “The complication rate (only infectious discitis) was 0.4%; all 10 patients with complications were cured with appropriate antibiotics. The recurrence rate was 5% and usually due to reinjury.” Ahn and colleagues reported on the outcomes of a case series of 43 consecutive patients who underwent laser lumbar disc decompression for recurrent herniation. Two case series described laser disc decompression of cervical discs, reporting successful results, particularly in patients with pain radiating down their arms. One case series of 16 patients also reported on combined treatment of IDET and nucleoplasty at multiple levels.
A retrospective review reported outcomes from 500 patients with discogenic pain and herniated discs treated with microdiscectomy (1997–2001 by six surgeons) and 500 patients treated with percutaneous laser disc decompression (2002–2004 by a single surgeon). Patients with sequestered discs were excluded. This retrospective review found that the hospital stay (six vs. two days), overall recovery time (60 vs. 35 days), and repeat procedure rates (7% vs. 3%) were lower in the laser group; these were not compared statistically. The percentage of patients with overall good/excellent outcomes (MacNab criteria) was found to be similar in the two groups (85.7% vs. 83.8%, respectively) at the two-year assessment; quantitative outcome measures were not reported.
Ishiwata et al. investigated the clinical results of their magnetic resonance-guided percutaneous laser disc decompression practice with reference to the site of the needle tip in the disc. They divided the disc on axial image into four quadrants and three concentric zones and evaluated clinical results by MacNab’s critera in each subdivided area six months after the procedure. The authors report an overall success rate of 68.8% in their series of 32 patients with low back pain and conclude that targeting certain zones seems to result in better outcomes.
In a 2009 paper, a design for a randomized controlled trial was described by investigators in the Netherlands. No trials involving laser discectomy were found in a search of the ClinicalTrials.gov on-line database in April 2010.
Disc nucleoplasty™ is a relatively new technology, and the literature consists of case series only with no controlled trials. In 2009, Chou et al. published a review of the evidence for nonsurgical interventions for low back pain for an American Pain Society guideline. The authors noted that one lower quality systematic review identified no randomized controlled trials and insufficient evidence from small case series to evaluate efficacy. Singh and colleagues reported clinical outcome data from an uncontrolled case series of 67 patients with contained disc herniation and low back pain who underwent DISC nucleoplasty™. Improvement was reported in approximately 60% of patients at 12 months. Li and colleagues report on a prospective study of 126 patients from China with contained cervical disc herniations that underwent nucleoplasty. Visual analog scale (VAS) pain scores were significantly improved at 1, 3, 6, and 12 months’ follow-up. Two smaller studies also report statistically significant reduction in pain. Calisaneller et al. reported on 29 patients who had lumbar nucleoplasty. Mean pre-operative VAS score was 6.95, and postoperative scores were 2.45, 4.0, and 4.53 at 24 hours, and three and six months, respectively. In a retrospective study from a U.S. center of 22 patients with 12 months of follow-up after lumbar nucleoplasty, statistically significant improvement on measures of pain, functional status, and medication use were reported. Al-Zain and colleagues report outcomes for 69 patients for whom 12-month data were available from a cohort of 96 patients who underwent nucleoplasty for back pain and/or radiating pain in the lower extremities. (Seven patients were lost to follow-up, 11 were excluded due to secondary disc sequestration at the treated segment or elsewhere, and data for eight patients are available only up to six months.) Seventy-three percent (73%) of patients improved more than 50% in early postoperative VAS score; this was reduced to 61% of patients at six months and to 58% after one year.
A prospective study assessed outcomes in 52 consecutive patients treated with radiofrequency nucleoplasty of lumbar discs. Included in the study were patients younger than 60 years of age with radicular pain that was resistant to at least three months of conservative treatment, combined with magnetic resonance imaging (MRI) evidence of small and medium-sized herniated discs (less than 6 mm) that correlated with the patient’s symptoms. Patients with a disc height of less than 50% of adjacent discs, severe degenerated or fractured disc material, or evidence of extruded disc herniation were excluded. Independent assessment at two weeks, six months, and one year (94% follow-up) found a decrease in VAS pain scores from 7.5 to 2.1, a change from 42 to 21 on the Oswestry Disability Index, and a reduction or complete stopping of use of analgesics in 94% of patients. Birnbaum reports on a series of 26 patients with cervical disc herniation (29 discs) treated with disc nucleoplasty who had two years of follow-up. He compares their outcomes with a group of 30 patients who received conservative treatment. It does not appear that patients were randomly assigned to either treatment group but that the control patients were randomly chosen. Conservatively treated patients received perineural injections with bupivacaine and prednisolone acetate during the first week of treatment. Baseline visual analog scores were 8.4 in the control group and 8.8 in the nucleoplasty group. At one week, scores were 7.3 and 3.4, respectively, and at 24 months, 5.1 and 2.3 respectively. No other outcome data are provided.
Query of ClinicalTrials.gov in April 2010 found two trials of disc nucleoplasty; one, a sham controlled study, is completed and the other, comparing radiofrequency nucleoplasty versus percutaneous nucleotomy (Dekompressor) versus decompression catheter for the treatment of painful contained lumbar disc herniation, is not yet recruiting.
The National Institute for Clinical Excellence (NICE) published guidance on laser lumbar discectomy in 2003, stating that current evidence “does not appear adequate to support the use of this procedure without special arrangement for consent and for audit or research” and that patients should understand the uncertainty about the safety and efficacy of the procedure. Guidance on percutaneous disc decompression using coblation for lower back pain was published in 2006 stating that there is some evidence of short-term efficacy; however “this is not sufficient to support the use of this procedure without special arrangements for consent and audit or research.”
A 2009 American Pain Society Clinical Practice Guideline on nonsurgical interventions for low back pain states that “there is insufficient (poor) evidence from randomized trials (conflicting trials, sparse and lower quality data, or no randomized trials) to reliably evaluate” a number of interventions including coblation.
Practice Guidelines published in 2009 by the American Society of Interventional Pain Physicians report USPSTF Level II-2 evidence of short-term and long-term relief of pain for percutaneous laser discectomy, citing the review by Singe and making a strong recommendation. The guidelines report Level II-3 evidence for disc nucleoplasty in managing predominantly lower extremity pain due to contained disc herniation and state that there is no evidence available for axial low back pain. The guidelines make a weak recommendation for radiofrequency disc nucleoplasty in managing radicular pain due to contained disc herniation. No recommendation for nucleoplasty is given regarding managing axial low back pain.
While a number of authors state that laser discectomy and disc nucleoplasty are effective treatments, the lack of well-designed and conducted controlled trials limits interpretation of reported data. Reconsideration of the policy position awaits multicenter randomized trials with adequate follow-up (at least one year) that control for the placebo effect and the natural variability of the course of low back pain. These procedures are considered experimental, investigational and unproven.
AXIAL LUMBOSACRAL INTERBODY FUSION
The literature on axial lumbosacral interbody fusion (axial LIF) consists of case series. No controlled trials have been identified that compare outcomes of axial LIF with other approaches to lumbosacral interbody fusion.
The largest case series published to date is a 2011 retrospective analysis of 156 patients from four clinical sites in the U.S. Patients were selected for inclusion if they underwent a L5-S1 interbody fusion via the axial approach and had both presurgical and two-year radiographic or clinical follow-up. The number of patients who underwent axial LIF but were not included in the analysis was not reported. The primary diagnosis was degenerative disc disease (61.5%), spondylolisthesis (21.8%), revision surgery (8.3%), herniated nucleus pulposus (8.3%), spinal stenosis (7.7%) or other (8.3%). Pain scores on a numeric rating scale improved from a mean of 7.7 to 2.7 (n=155), while the Oswestry Disability Index (ODI) improved from a mean of 36.6 preoperatively to 19.0 (n=78) at two-year follow-up. Clinical success rates, based on an improvement of at least 30%, were 86% for pain (n=127/147) and 74% for the ODI (n=57/77). The overall radiographic fusion rate at two years was 94% (145 of 155). No vascular, neural, urologic, or bowel injuries were reported in this study group. Limitations of this study include the retrospective analysis, lack of controls, and potential for selection bias by only reporting on the patients who had two years of follow-up.
In 2010, Patil and colleagues reported a retrospective review of 50 patients treated with axial LIF. Four patients (8%) underwent two-level axial LIF, and 16 patients (32%) underwent a combination of axial LIF with another procedure for an additional level of fusion. There were three reoperations due to pseudoarthrosis (n=2) and rectal injury (n=1). Other complications included superficial infection (n=5), hematoma (n=2), and irritation of a nerve root by a screw (n=1). At 12- to 24-month follow-up, VAS scores had decreased from 8.1 to 3.6 (n=48). At an average 12-month follow-up, 47 of 49 patients (96%) with postoperative radiographs achieved solid fusion. There were no significant differences between pre- and postoperative disk space height and lumbar lordosis angle.
Aryan and colleagues reported on a series of 35 patients with average follow-up of 17.5 months in 2008. These patients had pain secondary to lumbar degenerative disc disease, degenerative scoliosis, or lytic spondylolisthesis. In 21 of the patients, the axial LIF procedure was followed by percutaneous pedicle screw-rod fixation; two patients had extreme lateral interbody fusion (XLIF) combined with posterior instrumentation, and 10 had a standalone procedure. Two patients had axial LIF as part of a larger construct after unfavorable anatomy prevented access to the L5-S1 disc space during open lumbar fusion. Radiographic evidence of stable cage placement and fusion was found in 32 patients at last follow-up.
Axial LIF with percutaneous pedicle screw reduction has also been described for grade 2 spondylolisthesis in a case series of three patients.
The United Kingdom’s National Institute for Health and Clinical Excellence (NICE) provided guidance on transaxial interbody fusion in the lumbar spine in 2011. The guidance states that current evidence on the efficacy of transaxial interbody lumbosacral fusion is limited in quantity but shows symptom relief in the short term in some patients. Evidence on safety shows that there is a risk of rectal perforation. Therefore this procedure should only be used with special arrangements for clinical governance, consent, and audit or research. NICE encourages further research into transaxial interbody lumbosacral fusion. Research outcomes should include fusion rates, pain and functional scores, quality-of-life measures and the frequency of both early and late complications. NICE may review this procedure on publication of further evidence.
The American Association of Neurological Surgeons published guidelines for interbody techniques for lumbar fusion in 2005. There was insufficient evidence to recommend a treatment standard. Minimally invasive procedures were not reviewed.
An industry-sponsored five-year voluntary postmarketing surveillance study of 9,152 patients was reported by Gundanna et al. in 2011. A single-level L5-S1 fusion was performed in 8,034 patients (88%) and a two-level (L4-S1) fusion was performed in 1,118 patients (12%). A pre-defined database was designed to record device- or procedure-related complaints through spontaneous reporting. Several procedures, including the presence of a TransS1 representative during every case, were implemented to encourage complication reporting. The complications that were recorded included bowel injury, superficial wound and systemic infections, transient intraoperative hypotension, migration, subsidence, presacral hematoma, sacral fracture, vascular injury, nerve injury, and ureter injury, (pseudoarthrosis was not included). The follow-up period ranged from three months to five years three months. Complications were reported in 120 patients (1.3%) at a median of five days (mean, 33 days; range, 0-511 days). Bowel injury was the most commonly reported complication (0.6%), followed by transient intraoperative hypotension (0.2%). All other complications had an incidence of 0.1% or lower. There were no significant differences in complication rates for single-level (1.3%) and two-level (1.6%) fusion procedures. Although this study includes a large number of patients, it is limited by the dependence on spontaneous reporting, which may underestimate the true incidence of complications.
Lindley and colleagues found high complication rates in a retrospective review of 68 patients who underwent axial LIF between 2005 and 2009. Patient diagnoses included degenerative disc disease, spondylolisthesis, spinal stenosis, degenerative lumbar scoliosis, spondylolysis, pseudoarthrosis, and recurrent disc herniation. Ten patients underwent two-level axial LIF (L4-S1) and 58 patients underwent a single level axial LIF (L5-S1). A total of 18 complications in 16 patients (23.5%) were identified with a mean 34 months follow-up (range 17-61 months). Complications included pseudoarthrosis (8.8%), superficial infection (5.9%), sacral fracture (2.9%), pelvic hematoma (2.9%), failure of wound closure (1.5%), and rectal perforation (2.9%). Both of the patients with rectal perforation underwent emergency repair and were reported to have no long-term sequelae. The patients with non-union underwent additional fusion surgery with an anterior or posterior approach. The two patients with sacral fractures had pre-existing osteoporosis; one was treated with long iliac screws. Because of the potential for these complications, the authors recommend full bowel preparation and preoperative magnetic resonance imaging (MRI) prior to an axial LIF procedure used to assess the size of the presacral space, determine rectal adherence to the sacrum, rule out vascular abnormalities, and determine a proper trajectory.
A search of the FDA’s MAUDE database (available online in October 2011 identified over 100 adverse event reports for axial LIF, including possible and confirmed bowel injuries.
The available published evidence on axial LIF consists of case series. This evidence is insufficient to evaluate whether axial LIF is as effective or as safe as other surgical approaches to lumbosacral interbody fusion, due to the variable natural history of the disorder and the subjective nature of the main outcomes. In addition, there are a relatively large number of adverse event reports in the MAUDE database for axial LIF, which raises the possibility of an increased risk of complications. Due to limited evidence and concerns about the safety and efficacy of the axial approach, axial LIF is considered investigational.
Disclaimer for coding information on Medical Policies
Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.
The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.
Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.