Risk Factors for Reoperation in Patients Treated Surgically for Lumbar Stenosis
A Subanalysis of the 8-Year Data From the SPORT Trial
Abstract and Introduction
Study Design. A retrospective subgroup analysis was performed on surgically treated patients from the lumbar spinal stenosis (SpS) arm of the Spine Patient Outcomes Research Trial (SPORT), randomized, and observational cohorts.
Objective. To identify risk factors for reoperation in patients treated surgically for SpS and compare outcomes between patients who underwent reoperation with those who did not.
Summary of Background Data. SpS is one of the most common indications for surgery in the elderly; however, few long-term studies have identified risk factors for reoperation.
Methods. A post-hoc subgroup analysis was performed on patients from the SpS arm of the SPORT, randomized and observational cohorts. Baseline characteristics were analyzed between reoperation and no-reoperation groups using univariate and multivariate analysis on data 8 years postoperation.
Results. Of the 417 study patients, 88% underwent decompression only, 5% noninstrumented fusion, and 6% instrumented fusion. At the 8-year follow-up, the reoperation rate was 18%; 52% of reoperations were for recurrent stenosis or progressive spondylolisthesis, 25% for complication or other reason, and 16% for new condition. Of patients who underwent a reoperation, 42% did so within 2 years, 70% within 4 years, and 84% within 6 years. Patients who underwent reoperation were less likely to have presented with any neurological deficit (43% reop vs. 57% no reop, P = 0.04). Patients improved less at follow-up in the reoperation group (P < 0.001).
Conclusion. In patients undergoing surgical treatment for SpS, the reoperation rate at 8-year follow-up was 18%. Patients with a reoperation were less likely to have a baseline neurological deficit. Patients who did not undergo reoperation had better patient reported outcomes at 8-year follow-up compared with those who had repeat surgery.
Spinal stenosis (SpS) is one of the most common indications for surgery in the elderly and was the fastest growing indication for spine surgery in the last three decades.[1–5] Although SpS may lead to significant disability, several well-designed prospective studies have documented the efficacy of timely surgical intervention for carefully selected patients.[1–4,6,7] Recent changes in our healthcare system, technological advancements, and societal pressures have fueled the demand for cost-effective, clinically effective, and durable surgical options. A recent report compared long-term health-related quality of life (HRQoL), and cost per quality adjusted life years (QALY) in patients who had undergone surgery for SpS with patients who had undergone total knee or hip arthroplasty (TKA, THA).[8,9] Despite a higher revision rate, significant improvement was observed in the HRQoL and QALY of the SpS patients, comparable to the TKA and THA patients. Furthermore, long-term benefits for SpS patients were sustained up to a mean 7 to 8-year follow-up. Collectively, the literature supports the efficacy of surgery for SpS. However, the need for reoperation has been reported and concerns exist over potential increased costs and inferior outcomes among patients who require a revision.
Most of the current literature regarding risk factors for reoperation is from retrospective data, heterogenous patient groups, or limited demographic regions.[5,10–13] Most studies comparing surgical outcomes for SpS include patients with underlying deformity or instability.[6,10] An improved understanding of potential risk factors for revision surgery will serve to improve predictability of patient outcomes and cost effectiveness.
The Spine Patient Outcomes Research Trial (SPORT) is a large, multicenter, prospective study with strict inclusion criteria across three arms consisting of intervertebral disc herniation, degenerative spondylolisthesis, and SpS.[1,2] The SPORT database is unique in that it allows for analysis of patients with SpS without degenerative spondylolisthesis, as these patients were studied separately. The purpose of this study was to perform a subanalysis of the 8-year SPORT data to determine if patient baseline characteristics would emerge as risk factors for reoperation in patients treated surgically for SpS and compare outcomes between patients who underwent reoperation with those who did not.
Materials and Methods
This investigation was a retrospective subanalysis of prospectively collected data from the SPORT trial. The data were based on data collected from enrollment in the initial SPORT study through 8-year follow-up.
Patient Population: The SPORT trial is a multicenter study carried out among 13 institutions in 11 states across the United States. The study includes both randomized and observational cohorts using validated primary and secondary outcome measures (Short-Form 36 [SF-36], AAOS/Modems version of the Oswestry Disability Index [ODI], patient-reported improvement, satisfaction with current symptoms and care, Stenosis Bothersomeness Index [SBI], and the Leg pain and Low Back Pain Bothersomeness Scale). Further details of the original SpS arm of the SPORT study have been reported elsewhere.[3,4]
Patients included in the SpS arm of the SPORT trial had neurogenic claudication or radicular leg symptoms for at least 12 weeks and spinal stenosis seen on cross-sectional imaging at one or more levels. Exclusion criteria included patients with fixed or unstable lumbar spondylolisthesis (translation of more than 4 mm or 10° of angular motion between flexion and extension on upright lateral radiographs), or spondylolysis.
A research nurse at each site identified participants and verified eligibility with use of a shared decision-making video. Enrollment began in March of 2000 and ended in November 2004. Follow-up was at 6 weeks, 3 months, 1 year, 2 years, 4 years, 6 years, and 8 years postenrollment.
The protocol surgical intervention was a standard posterior decompressive laminectomy. If, at the time of surgery, the surgeon believed that the patient required a procedure that differed significantly from the protocol (e.g., the addition of an arthrodesis) the patient remained a study patient and the specifics of the procedure performed were recorded on the surgical treatment form.
Primary endpoints were the Short-Form 36 bodily pain and physical function scores (SF-36 BP and PF), as well as the Oswestry Disability Index (ODI). Secondary outcomes included patient reported self-improvement, Stenosis Frequency Index (SFI), Stenosis Bothersomeness index (SBI), Low Back Pain Bothersome Index, Leg Pain Bothersome Index, satisfaction with current symptoms and care, and work status.
Patient characteristics included age, sex, work lift demand, race, education, marital status, work status, compensation, body mass index (BMI), smoking status, hypertension, diabetes, osteoporosis, depression, heart problem, joint problem, time since most recent episode began, expectation of being free of pain with surgery, expectation of being free of pain with nonoperative treatment, opioid use, injections, physical therapy, use of antidepressants, taking NSAID, missed work, predominant back pain, and low back pain severity.
Operative characteristics measured included procedure type (decompression only, noninstrumented fusion, instrumented fusion), multilevel fusion, decompression level, number of levels decompressed, operative time, blood loss, blood replacement, length of hospital stay, and complications.
Intraoperative complications included dural tear or spinal fluid leak or «other» intraoperative complications. Postoperative complications were measured up to 8 weeks postoperatively and included wound hemotoma and wound infection, or «other» postoperative complications. Bone graft complications, CSF leaks, cauda equina injury, wound dehiscence, and pseudoarthrosis were assessed but did not occur in this group.
A subgroup analysis was performed on surgically treated patients in the SpS arm of the SPORT trial, including both randomized and observational cohorts. In this analysis, patients were stratified into no-reoperation versus reoperation groups and statistical analysis compared the two groups.
Baseline characteristics, operative factors, complications, and medical events in the reoperation and no-reoperation groups were compared using a χ2 test for categorical variables and t tests for continuous variables. To analyze the risk factors for reoperation, Cox proportional hazards model was used to explore which variables maintained significance after adjusting for other variables in the model. Variables significant at P < 0.10 were candidates for inclusion in the final multivariable regression model. Final selection for the model was done using the stepwise method as implemented in SAS, which sequentially enters the most significant variable with P < 0.10 and then after each entered variable removes variables that do not maintain significance at P < 0.05. Age and sex were forced to be in the model.
Primary outcomes analyses compared the reoperation and no-reoperation groups using changes from baseline at each follow-up, with a mixed effects longitudinal regression model including a random individual effect to account for correlation between repeated measurements within individuals. The analyses were adjusted for age, sex, compensation, baseline score of stenosis bothersomeness index, income, smoking status, stomach problem, joint problem, diabetes, duration of most recent episode, treatment preference, baseline score (for SF-36 and ODI), and center. Across the 8-year follow-up, overall comparisons of «area-under-the-curve» between groups were made using a Wald test. Computations were done using SAS procedure PROC MIXED for continuous data and PROC GENMOD for binary outcome (SAS version 9.2, SAS Institute Inc, Cary, NC). Statistical significance was defined as P < 0.05 based on a two-sided hypothesis test with no adjustments made for multiple comparisons.
Index Patient Population
Overall, in the SpS arm of the SPORT trial, 289 patients were enrolled to the randomized cohort and 365 were enrolled to the observational cohort. By 8 years, 70% of those randomized to surgery and 52% of those randomized to nonoperative treatment had undergone surgery, constituting a total of 428 patients, of which 422 had at least one follow-up and were included in the analysis (). Of these, 417 had adequate data on surgery to be included in the analysis of operative treatment and complications (). Patients were lost to follow-up because of dropping out of the study, missed visits, or death
Of the 417 patients who had surgery data, 88% underwent decompression alone, 5% noninstrumented fusion, and 6% instrumented fusion. At 8-year follow-up, the reoperation rate was 18% (n = 77). Of these patients, 32 underwent reoperation within 2 years (42%), 54 within 4 years (70%), and 65 within 6 years (84%). The indication for reoperation was recurrent stenosis or progressive spondylolisthesis in 52% (40/77), a complication or other reasons in 15% (19/77), and a new condition in 16% (12/77). One patient was indicated for a pseudoarthrosis or fusion exploration over the 8-year period. The remaining patients had incomplete information with regard to indication for reoperation.
Of the 77 patients who underwent reoperation, 63 had complete information with regard to spinal level (as recorded in the Nurse Repeat Spine Surgery Survey). Twenty-one of 63 of whom had reoperation at the same level as the index operation, 29 of 63 had reoperation at a different level, and 13 of 63 with reoperation level unspecified. Recurrent stenosis or progressive spondylolisthesis accounted for 11 of 21 (same level), 23 of 29 (different level), and 2 of 13 (unspecified) of the reoperations, respectively
Cox’s proportional hazards model was used to explore which variables maintained significance after adjusting for other variables in the model. Variables significant at P<0.10 were candidates for inclusion in the final multivariable regression model. Final selection for the model was done using the stepwise method as implemented in SAS Release 9.2 (SAS Institute, Cary, NC), which sequentially enters the most significant variable with P <0.10 and then after each entered variable removes variables that do not maintain significance at P<0.05. Age and sex were forced to be in the model.
Candidate predictor variables: age, sex, race, education, marital status, work status, BMI, smoking status, work lift demand, hypertension, diabetes, osteoporosis, depression, heart problem, joint problem, time since most recent episode, patient’s self-assessment of health trend, patient dissatisfied with symptoms, expectation of free of pain with surgery, expectation of free of pain with nonoperative treatment, opioid use, injections, had physical therapy, taking antidepressants, taking NSAID, predominant back pain, low back pain bothersomeness scale, back pain bothersomeness scale, stenosis bothersomeness index, Oswestry disability index, bodily pain, physical function, mental component summary, physical component summary, neurogenic claudication, pain on straight-leg raising or femoral-nerve tension sign, dermatomal pain radiation, any neurologic deficit, asymmetric reflex depression, asymmetric sensory decrease, asymmetric motor weakness, moderate or severe stenotic levels, location of stenosis, and severity of stenosis.
Laminectomy level for the index procedure was L2–L3 in 149 patients (36%), L3–L4 in 286 patients (69%), L4–L5 in 379 patients (90%), and L5–S1 in 157 patients (38%). Four percent of patients underwent a multilevel fusion and the percent of patients with zero, one, two, and three or more levels decompressed were 2%, 23%, 31%, and 45%, respectively.
Patients undergoing reoperation were less likely to have presented with any neurological deficit upon enrollment (43% vs. 57% in the no- reoperation group, P = 0.04). There were no other significant differences in demographic variables or incidence of comorbidities, baseline HRQOL, or baseline clinical information.
Analysis of operative treatments, complications, and adverse events revealed patients in the reoperation group had a slightly longer length of hospital stay during their index surgery though this was not statistically significant (2.7 vs. 3.3,P = 0.058) (). No significant differences between the two groups were found in decompression levels, multilevel fusions, instrumentation, number of levels decompressed, operative time, blood loss, use of blood replacement, or rate of intraoperative complications.
Although the SPORT trial data demonstrated significant benefits of surgical intervention for the treatment of SpS, a better understanding of risk factors for poor outcomes is critical for use during the decision-making process and counseling for surgery. This subanalysis of the SPORT SpS data offers unique insights into clinical factors associated with reoperation after surgical treatment of spinal stenosis. Specifically, this study investigated the incidence and causes of reoperation, risk factors for reoperation, and differences in outcomes in patients undergoing reoperation versus patients not undergoing reoperation by 8-year follow-up. This study represents the first investigation of its kind, using a large prospective database focused on lumbar stenosis that excludes instability or deformity cases.
The reoperation rate for patients with SpS at 8 years postoperatively was 18% in this cohort, with 70% of these occurring within the first 4 years. As found in other studies, the causes for reoperation in the first 4 years were inadequate decompression, disease progression, technical errors, adjacent segment disease, persistent pain, and postoperative complications.[1,2,14,15] In 52% of cases, the reason for reoperation was documented as recurrent stenosis or progressive spondylolisthesis, which may occur as a result of disease progression or related to surgical technique.[10,16] In our study, a reoperation was most likely because of progressive degenerative changes or recurrent stenosis at the index or adjacent levels, though some were a result of complications or a new condition.
Whether or not arthrodesis augments or blunts the effects of surgical decompression has also not been substantiated.[5,10,13,15,17] In our study, surgery included arthrodesis in 12% of patients and there was no difference in reoperation rates when comparing fusion versus nonfusion procedures. Stratification by surgical technique showed similar reoperation rates for decompression alone, noninstrumented fusion, or instrumented fusion (P < 0.84). Though these findings are consistent with prior reports, there were insufficient numbers of patients with fusion to power further analyses. Selection bias may also explain our findings as more severe cases may have been selected for fusion. Previous studies have shown similar findings, though most reports include heterogenous data including patients with other spinal pathology.[5,13,18]
In contrast with prior studies, multilevel laminectomy (more than two levels) was not associated with an increased risk of reoperation (P < 0.71), though our sample size was small.[11,12] Fox et al correlated the development of postoperative instability in patients undergoing a decompression of greater than one level, though there was no correlation between patient outcomes and late instability after spinal decompression alone. Several other reports have also questioned this debate of whether arthrodesis is indicated when extensive decompressions are performed and demonstrated no correlation between patient outcomes and decompression with arthrodesis versus decompression alone.[10,18,19] In a study by Deyo et al, patients undergoing decompression alone versus arthrodesis at 4-year follow-up had similar reoperation rates, though they included patients with spondylolisthesis, scoliosis, and previous lumbar surgery. Martin et al found higher reoperation rates in patients undergoing fusion compared with decompression alone (21.5% versus 18.8%, P < 0.008), though they also included a heterogenous patient sample.
Interestingly, smoking, depression, workman’s compensation, other medical comorbidities, and obesity were not associated with an increased risk for reoperation. This result differs from other reports that have shown increasing rates of reoperation with specific comorbidities.[5,10,13,20] Patients undergoing reoperation were less likely to have presented with a neurologic deficit at enrollment, which underscores the importance of patient selection and that the presence of neurologic deficits may help identify more appropriate surgical candidates. To our knowledge, this is the first report of these specific findings, though the exact significance remains to be substantiated, and the presence of neurological deficit was not significant in the multivariate prediction model.
As expected, the patient-reported pain and functional outcomes at 8 years were more favorable for patients who did not require a reoperation. Patients going on to have a reoperation had less improvement in patient outcomes scores. Other studies have shown similar findings with worse outcomes scores reported in patients who required multiple lumbar spine operations. In this time of a changing healthcare climate, decreasing reoperation rates through a better understanding of risk factors and surgical indications will enhance patient care and long-term patient outcomes.
Limitations of our study: The relatively small sample size of patients who underwent reoperation was a limitation to this subanalysis (n = 77) and there was a small degree of heterogeneity in the surgical techniques. That being said, this cohort is larger and more homogenous, with longer follow-up than found in other published works. The high drop out rate with any long-term study can also confound the study findings as patients with worse outcomes may seek treatment elsewhere or not at all, and patients doing well after surgical or nonsurgical intervention may also be less likely to return for follow-up.
This subanalysis reports mid-term reoperation rates, risk factors, and outcomes of surgically treated patients from the lumbar SpS arm of the SPORT. In this study, patients with any neurological deficit upon enrollment were less likely to undergo a subsequent reoperation. To our surprise, patient comorbidities and socioeconomic factors did not correlate with higher reoperation rates. In addition, patient reported outcomes at 8 years were more favorable for patients who did not undergo a reoperation. These findings further support the role of surgical intervention in the treatment paradigm for SpS. The overall low reoperation rate in this trial reflects the fact that this study represented a homogenous group of patients with lumbar spinal stenosis without degenerative spondylolisthesis, or degenerative scoliosis. With strict adherence to these patient selection criteria, surgeons may achieve similar results for their patients.
- Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonoperative treatment for lumbar spinal stenosis four-year results of the Spine Patient Outcomes Research Trial. Spine 2010;35:1329–1338.
- Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008;358:794–810.
- Weinstein J, Birkmeyer J. The Dartmouth Atlas of Musculoskeletal Health Care. Chicago, IL: American Hospital Association; 2000;60.
- Weinstein JN, Lurie JD, Olson P, et al. United States trends and regional variations in lumbar spine surgery: 1992–2003. Spine 2006;31:2707.
- Atlas SJ, Keller RB, Robson D, et al. Surgical and nonsurgical management of lumbar spinal stenosis: four-year outcomes from the maine lumbar spine study. Spine 2000;25:556–62.
- Boden SD, Davis D, Dina T, et al. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg 1990;72:403–8.
- Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med 1994;331:69–73.
- Rampersaud YR, Lewis SJ, Davey JR, et al. Comparative outcomes and cost-utility after surgical treatment of focal lumbar spinal stenosis compared with osteoarthritis of the hip or knee-part 1: long-term change in health-related quality of life. Spine J 2014;14:234–43.
- Rampersaud YR, Tso P, Walker KR, et al. Comparative outcomes and cost-utility following surgical treatment of focal lumbar spinal stenosis compared with osteoarthritis of the hip or knee: part 2-estimated lifetime incremental cost-utility ratios. Spine J 2014;14:244–54.
- Martin BI, Mirza SK, Comstock BA, et al. Reoperation rates following lumbar spine surgery and the influence of spinal fusion procedures. Spine 2007;32:382–7.
- Deyo RA, Ciol MA, Cherkin DC, et al. Lumbar spinal fusion: a cohort study of complications, reoperations, and resource use in the Medicare population. Spine 1993;18:1463–70.
- Deyo RA, Gray DT, Kreuter W, et al. United States trends in lumbar fusion surgery for degenerative conditions. Spine 2005;30:1441–5.
- Atlas SJ, Deyo RA, Keller RB, et al. The Maine Lumbar Spine Study, Part III: 1-year outcomes of surgical and nonsurgical management of lumbar spinal stenosis. Spine 1996;21:1787–94.
- Hazard RG. Failed back surgery syndrome: surgical and nonsurgical approaches. Clin Orthop Relat Res 2006;443:228–32.
- Radcliff K, Curry P, Hilibrand A, et al. Risk for adjacent segment and same segment reoperation after surgery for lumbar stenosis: a subgroup analysis of the Spine Patient Outcomes Research Trial (SPORT). Spine (Phila Pa 1976)2013;38:531–9.
- Thomé C, Zevgaridis D, Leheta O, et al. Outcome after lessinvasive decompression of lumbar spinal stenosis: a randomized comparison of unilateral laminotomy, bilateral laminotomy, and laminectomy. J Neurosurg Spine2005;3:129–41.
- Ghogawala Z, Benzel EC, Amin-Hanjani S, et al. Prospective outcomes evaluation after decompression with or without instrumented fusion for lumbar stenosis and degenerative Grade I spondylolisthesis. J Neurosurg Spine2004;1:267–72.
- Fischgrund JS, Mackay M, Herkowitz HN, et al. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective, randomized study comparing decompressive laminectomy and arthrodesis with and without spinal instrumentation. Spine (Phila Pa 1976) 1997;22:2807–12.
- Fox MW, Onofrio BM, Hanssen AD. Clinical outcomes and radiological instability following decompressive lumbar laminectomy for degenerative spinal stenosis: a comparison of patients undergoing concomitant arthrodesis versus decompression alone. J Neurosurg 1996;85:793–802.
- Freedman MK, Hilibrand AS, Blood EA, et al. The impact of diabetes on the outcomes of surgical and nonsurgical treatment of patients in the spine patient outcomes research trial. Spine 2011;36:290–307.