Introduction

Surgical decompression with or without fusion is the standard surgical treatment for the patients with symptomatic lumbar spinal stenosis (LSS). Decompression has the potential for significant complications, especially when a fusion is performed. Post-operative complications may include the cardiovascular and pulmonary complications of general anesthesia, infection, iatrogenic instability, pseudarthrosis, hardware failure and the need for future surgery due to the development of new disease at the same or adjacent levels. Turner et al., in a 1992 meta-analysis of the spinal stenosis literature, have shown an overall complication rate of decompression surgery of about 13%.₁

The X STOP Interspinous Process Decompression (IPD) is a new surgical treatment option for a variety of spinal pathologies. It provides an unloading distractive force to the stenotic middle and posterior columns of the motion segment and has the potential to relieve the symptoms of intermittent neurogenic claudication associated with spinal stenosis. The X STOP (Kyphon, Sunnivale, CA, Figure 1) is the only FDA-approved IPD implant at the present time. It has been shown to be effective for the treatment of patients with neurogenic intermittent claudication secondary to LSS.₂

Figure 1

Figure 1: X STOP Device

The X STOP was designed to limit the terminal extension movement at the individual stenotic level(s) that provokes the symptoms, while allowing unrestricted movement in all the other motion axes of the treated as well as untreated level(s). Biomechanical studies have shown that the X STOP significantly increases the spinal canal, subarticular recess and neuroforaminal size; limits terminal extension; and reduces intradiscal pressure and facet loading.₃,₄,₅,₆ In a cadaveric disc pressure study, Swanson et al. reported that the pressures in the posterior annulus and nucleus pulposus were reduced by 63% and 41% respectively during extension, and by 38% and 20% respectively in the neutral, standing position.₄ Wiseman et al. performed a cadaveric facet loading study and reported that the mean facet force during extension decreased by 68% during extension.₆ In each of those studies, the adjacent level measurements were not significantly changed from the intact specimen state. Those pre-clinical studies indicate that the X STOP increases spinal canal and neural foramina space and also produces significant unloading of the disc and facets.

In addition to the pre-clinical data, there is also a growing body of clinical information. Zucherman et al. have demonstrated that IPD with the X STOP is superior to non-operative therapy in patients with neurogenic intermittent claudication secondary to spinal stenosis in the multicenter randomized study at 1 and 2 years post-operatively.₂,₇ Lee et al. have reported that 70% of the patients were satisfied with the surgical outcome following X STOP placement and concluded that it is effective in elderly LSS patients.₈ In a positional MRI study, Siddiqui et al. have demonstrated that the X STOP implant increases the cross-sectional area of the dural sac and foramina without causing changes in posture.₉ Kondrashov et al. have demonstrated that at 4-year follow-up X STOP interspinous decompression has effectiveness of about 78% in patients with neurogenic claudication.₁₀

Based on the available information, we hypothesized that IPD with the X STOP is clinically as effective as the conventional lumbar decompression without fusion. Since X STOP placement was performed under local anesthesia with sedation on the outpatient basis in a U.S. FDA trial, we have also hypothesized that it would provide significant cost savings compared to decompressive surgery, which is typically performed under general anesthesia and often requires that patients remain hospitalized for one or more days. The goal of the current study was to compare the clinical effectiveness and direct hospital costs of IPD with the X STOP implant to those of laminectomy for the treatment of patients with LSS.

Study Design

Methods

After obtaining Institutional Review Board (IRB) approval, eligible subjects were asked to participate in the study and to sign an informed consent document. In order to determine study eligibility a limited waiver of consent was obtained from the IRB to review patient records. Patient records were screened to identify potentially eligible subjects who underwent X STOP IPD surgery as part of an FDA Pivotal Trial conducted at our hospital from June, 2000 to July, 2001, or who underwent decompression without fusion during the same timeframe. The decompression encompassed laminectomy, hemi-laminectomy, laminotomy or foraminotomy, performed alone or in combination. All patients implanted with the X STOP were screened for eligibility (N=23). Four patients were not included due to lack of preoperative ODI data. One X STOP patient has died prior to initiation of the study from unrelated causes. An eligible group of 18 X STOP IPD patients was identified, all of whom elected to participate in the study.

The inclusion criteria for the original X STOP pivotal trial were as follows: Patients had to be at least 50 years old and have leg, buttock, or groin pain with or without back pain that was relieved during flexion. Patients had to be able to walk at least 50 feet and sit comfortably for 50 minutes. The exclusion criteria for the X STOP FDA pivotal trial were as follows:

Patients could not have a fixed motor deficit, cauda equina syndrome, previous lumbar surgery of the stenotic level or spondylolysthesis greater than grade I at the affected level.

Eighty decompressive surgery patients were found to meet the initial surgical criteria based on their clinical records. Following review of each patient's records, 17 of these patients met the remaining eligibility criteria and were invited to participate. Four patients declined and data were incomplete for 1 patient, yielding 12 eligible subjects who elected to participate (12/17). An absolute improvement of 15 ODI points was selected to define an individual patient success.

Surgical Technique

Results

Twelve of 18 X STOP patients (67%) were treated at one level and six of 18 (33%) at two levels (Table 2). Three of 12 laminectomy patients (25%) were treated at one level and nine of 12 (75%) at two levels. The average age was 68 years (SD 12.5) for X STOP patients and 69 years (SD 7.9) for laminectomy patients. Six of the 18 X STOP patients (33%) had grade I degenerative spondylolysthesis at the treated level versus two of 12 patients (17%) in the laminectomy group. The minimum follow-up was 45 months in the X STOP group and 43 months in the laminectomy group. The average follow-up was 51 months in the X STOP group (SD 5.3) and 52 months in the laminectomy group (SD 5.3). Pre-operative average ODI score in the X STOP group was 45 vs. 36 in the laminectomy group (p>0.1560). Post-operative average ODI score in the X STOP group was 15 vs. 24 in the laminectomy group (p>0.1159). The absolute ODI change was 29 points for the X STOP group, and 12 points for the laminectomy group (p<0.0186). The relative ODI change was 60% for the X STOP group, and 28% for the laminectomy group (p<0.0369). Based on the selected success criterion (ODI improvement of at least 15 points), fourteen of 18 (78%) X STOP procedures were considered successful and four of 12 (33%) laminectomy procedures were deemed successful (p<0.0243).

Figure 2

Table 1: Patient Demographics, Surgical Variables, and Outcomes

Figure 3

Table 2: Cost analysis of the X STOP and Decompressive Surgery Procedures

Figure 4

Table 3: Rates of additional surgery after laminectomy, X STOP placement and non-operative therapy

The X STOP procedure was performed on an outpatient basis, under local anesthesia, and required less operating time. The laminectomy, on the other hand, required an extended hospital stay, was performed under general anesthesia, and required more operating time. Average hospital costs for a single level X STOP implantation and a single level laminectomy were $15,980 and $45,302 respectively. Average hospital costs for a double level X STOP implantation and a double level laminectomy were $25,618 and $46,752 respectively (Table 2). The primary cost drivers were hospital charges, anesthesia charges, and operating room costs. The cost of the X STOP implant and higher radiology charges due to use of fluoroscopy during X STOP implantation were significantly offset by the cost drivers described above.

Discussion

Interspinous process decompression is a relatively new motion-preserving spinal procedure. A multi-center prospective, randomized controlled trial was performed in the U.S. comparing the outcomes of mild-to-moderate neurogenic intermittent claudication patients treated with the X STOP interspinous process decompression system to patients treated non-surgically.₂,₇ It was the first study to provide the level 1 data for surgical and non-surgical treatment of spinal stenosis. At 2-year follow-up, 57% of the patients reported a clinically significant improvement in their physical function compared to 15% of the controls, and 73% of the patients were at least somewhat satisfied compared to 36% of the controls. At all follow-up time points, the X STOP group scored significantly better than the control group in every physical domain. In other studies, the clinical success rate of the X STOP IPD at 2-year follow-up has ranged from 70 to 85%.₂,₇,₈. Zucherman et al. have reported a 6% re-operation rate for X STOP at 2-year follow-up.₂ Those results are quite comparable to those reported for laminectomy, even though there is no level 1 data on the outcomes of laminectomy. In a 3-6-year follow-up study of 88 patients post-laminectomy, Katz et al. have noted that 17% of the original eighty-eight patients had a repeat operation because of instability or stenosis.₁₂In a 7-10-year follow-up study of the same cohort of patients Katz et al. have noted that 75% patients were satisfied with the results of surgery and 23% of patients had undergone reoperation, suggesting even more dramatic deterioration of the results.₁₃ Jonnson et al. have conducted a prospective study of patients with lumbar spinal stenosis treated surgically with a 5-year follow-up by an independent observer.₁₄ Excellent results were reported by 63% to 67% of patients at 4-month and 2-year follow-ups compared with 52% at the 5-year follow-up. The re-operation rate was 18% within 5 years (Table 3). In another retrospective analysis of 58 LSS patients treated with decompression, Fokter and Yerby have reported that 63% of the patients were considered to be clinically successful based upon the clinical improvement in at least 2 out of 3 domains of Zurich Claudication Questionnaire (ZCQ) at a minimum 1-year follow-up (range 12-54 months) - 63% (37 of 58 patients) met two of the three ZCQ domain criteria and 43% (25 of 58 patients) met all three ZCQ domain criteria.₁₅ At the present time it is unclear whether the X STOP results will remain stable over long term. However in the intermediate term they seem to be stable. Our results have demonstrated that the success rate in the X STOP IPD group is 78% at an average of 4.2 years postoperatively and is consistent with the 2 year results reported for X STOP treatment arm of the FDA IDE Trial ₂ and those reported by Lee et al.₈

The somewhat lower scores in the laminectomy group at 4 years could be due to the limited sample size, as well as known deterioration of the success of surgical decompression with time.₁₂,₁₃,₁₄ In our study the pre-operative average ODI score in the X STOP group was 45 vs. 36 in the laminectomy group, indicating somewhat more severe symptoms in the X STOP group pre-operatively, though the difference was not statistically significant (p>0.1560). Fokter and Yerby have shown that patients with more severe pre-operative symptoms and more physical function restrictions had better success results than those patients with milder symptoms and less restrictive physical function in a retrospective study describing the outcome predictors of decompressive surgery for LSS.₁₅

In our economic analysis, we were able to demonstrate a dramatic difference in direct hospital costs between the X STOP implantation and decompressive surgery. Less invasive approach did translate into very significant cost savings as well. Those numbers can translate into very significant savings on the national level, especially in the light of ever increasing rates of spine surgery in the U.S. and the aging population. Our study was limited to direct hospital costs, however one can extrapolate those findings to indirect costs as well. In our experience, the recovery after X STOP procedure is much quicker than from laminectomy and patients usually limit their activities only briefly after the surgery. This can translate into less time off work for the patients and, more commonly, for their family and caregivers. Another advantage of X STOP would be in revision surgery. It has been our experience that the rates of complications, most notably dural tears in revision laminectomy are much greater than during primary procedures. Since the IPD does not involve entering the spinal canal, revising an X STOP to laminectomy would theoretically pose much less risks of, at least, dural tears and, thus, offer even more cost savings in the long term by avoiding those complications altogether.

The primary advantage of our study is that it is the first study to directly compare IPD with lumbar decompression both clinically and economically. It does provide some important insights into the spectrum of approaches to the surgical management of spinal stenosis.

Our study does have some important shortcomings. First, the sample size was relatively small and there was some potential for sampling error. Second, we have utilized hospital charges as opposed to estimated costs, which, in some instances, may be less accurate. It would have been difficult, if not impossible, to obtain some of the estimated costs at the time of the study. For instance, the hospital costs for X STOP placement could only be assessed based on actual hospital charges, since there was no reimbursement data during the X STOP study. Third, our analysis does not represent a true cost-effectiveness or cost-utility study, since we have not adjusted the costs of interventions for changes in quality of life or life expectancy. Our assumption was that both interventions would have the same impact on the life expectancy, even though general anesthesia used for laminectomy does historically have a significantly higher mortality rate compared to sedation used for X STOP placement. Fourth, our analysis is not a randomized double-blinded clinical trial and therefore could have been subjected to a number of potential biases. Lastly, some clinicians would argue, that a more accurate measure of success is return to work and cessation of narcotic usage. However, in our elderly study population (average age 68-69 years) a substantial number of patients are retired or working part-time and, therefore, the rate of return to work would be less meaningful. Our patients also have multiple comorbidities some of which necessitate the use of narcotics (e.g. osteoarthritis) and some of which, on the contrary, preclude the use of any narcotics (e.g. confusion and drowsiness caused by opiates in the elderly). Therefore, using the reduction in narcotic usage as a measure of clinical success in our study population is probably less than ideal. There is a need for more studies of similar scope with larger groups of patients, longer follow-up and, possibly, randomization to minimize the bias.

In summary, interspinous process decompression with the X STOP device is a new effective treatment option for surgical treatment of lumbar spinal stenosis with or without low-grade degenerative spondylolysthesis. It appears to be at least as effective as lumbar decompressive surgery at 4 year follow-up in the studied patient population and offers significant savings in direct hospital costs over standard laminectomy.