NARRATIVE REVIEW

Lumbar Spinal Stenosis and Degenerative

Spondylolisthesis
A Review of the SPORT Literature
Haariss Ilyas, MD,*† Inyang Udo-Inyang Jr, MD,*† and Jason Savage, MD*†

observational cohorts. Primary outcomes were assessed

Abstract: The Spine Patient-Reported Outcomes Related Trial using the 36-Item Short-Form Health Survey (SF-36)
(SPORT) is arguably one of the most impactful and insightful bodily pain (BP) map and physical function (PF) scales
studies conducted in spine surgery. Designed as a prospective, and a modified Oswestry Disability Index (ODI). Sec-
multicenter study with randomized and observational cohorts, ondary measures included patient-reported improvement
SPORT has provided vast data on the pathogenesis, treatment and satisfaction. After the original publications, numerous
effects, clinical outcomes, cost effectiveness of disk herniation, subsequent analyses have been performed, including
lumbar spinal stenosis, and degenerative spondylolisthesis. With 8-year outcomes as well. In this article, we provide a brief
regards to spinal stenosis and degenerative spondylolisthesis, overview of LSS and DS as well as a review of subsequent
SPORT has demonstrated a sustained benefit from surgical in- literature based on SPORT (Table 1).
tervention at 2, 4, and 8 years postoperatively. Myriad subgroup
analyses have subsequently been performed that have also re-
sulted in clinically relevant findings. These analyses have assessed SPINAL STENOSIS AND DS
incidence and risk factors for reoperations and intraoperative LSS refers to degenerative changes in the spine that
complications, impact of patient comorbidities and host factors, result in central canal, lateral recess, and/or neural foraminal
influence of epidural injections, patient decision-making, and narrowing (Fig. 1). Because of its degenerative nature and
role of nonoperative therapy. This has resulted in significant significant impact on quality of life, it is the most common
findings that may allow spine surgeons to optimize patient out- indication for spine surgery in the geriatric population.25 A
comes while managing expectations appropriately. review of Medicare claims in 2007 noted a hospital bill of
$1.65 billion for lumbar stenosis surgery with an incidence of
Key Words: spinal stenosis, degenerative spondylolisthesis, lum- 135.5 cases per 100,000 Medicare beneficiaries.26 In patients
bar spine, SPORT, randomized trial refractory to nonoperative treatment, surgical treatment in
(Clin Spine Surg 2019;32:272–278) the form of decompression is well-accepted as the treatment
of choice.
Patients with LSS may also present with DS, defined as
“an acquired anterior displacement of one vertebra over the
T he Spine Patient-Reported Outcomes Trial (SPORT)
has become one of the most established and prolific
research studies in spine surgery. Designed to focus on
subjacent vertebra, associated with degenerative changes,
without an associated disruption or defect in the vertebral
ring”27 (Fig. 2). Traditionally, spondylolisthesis is thought to
intervertebral disk herniation (IDH), lumbar spinal represent instability and fusion surgery is added to decom-
stenosis (LSS) and degenerative spondylolisthesis (DS), pression in these patients to prevent further slippage.28 Spinal
this study has provided insight into nonoperative man- fusions embody a large portion of the healthcare sector, as it
agement, surgical treatment, postoperative complications, represents the highest costs of any surgical procedure in the
patient risk factors, cost-effectiveness, and the natural United States, with an aggregate sum of $12.8 billion in 2011.29
history of these diseases among other aspects.1 Although some recent studies have questioned the utility of
SPORT was conducted at 13 clinical sites across 11 fusion in all cases of spondylolisthesis,28–33 it is often considered
states starting in March 2000 and supported by a $15 the first-line surgical treatment option. Various techniques exist
million grant from the National Institute of Health.2 for fusion, including posterolateral and interbody methods,
Each trial was designed with 2 branches—randomized and depending on the patient’s pathology and surgeon’s experience
and preference. Several studies have analyzed these techniques,
Received for publication June 16, 2018; accepted April 19, 2019. without a clear consensus on the optimal method.27,34–39
From the *Center for Spine Health; and †Department of Orthopaedic
Surgery, Cleveland Clinic, Cleveland, OH.
The authors declare no conflict of interest. SPINAL STENOSIS RESULTS
Reprints: Haariss Ilyas, MD, Department of Orthopaedic Surgery,
Cleveland Clinic, 9500 Euclid Ave., A-40, Cleveland, OH 44195 In February 2008, 2-year results for the LSS portion
(e-mail: [email protected]). of the study were published in The New England Journal of
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved. Medicine.3 They reported on an aggregate of 654 patients,

272 | www.clinicalspinesurgery.com Clin Spine Surg  Volume 32, Number 7, August 2019

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Clin Spine Surg  Volume 32, Number 7, August 2019 Review of LSS and SPORT

treated analysis of both cohorts demonstrated significant

TABLE 1. Summary of SPORT Literature
treatment effects in favor of surgery. Specifically, at 2 years,
References Assessment Journal Year the treatment effect of surgery was 13.6 for SF-36 BP, 11.1 for
Index Trial Outcomes SF-36 PF, and -11.2 for ODI (Table 3). It is worth noting that
Weinstein 2-year outcomes of LSS New England Journal 2008 the treatment effects of surgery were significant at 6 weeks and
et al3 of Medicine
Weinstein 4-year outcomes of LSS Spine 2010
remained so, at the 2-year mark.
et al4 In 2010, 4-year results were subsequently published in
Lurie et al5 8-year outcomes of LSS Spine 2015 Spine.4 An as-treated analysis found significant treatment
Weinstein 2-year outcomes of DS New England Journal 2007 effects for surgery across all primary outcomes in both the
et al6 of Medicine observational and randomized cohorts (Table 2). Again,
Weinstein 4-year outcomes of DS Journal of Bone and 2009
et al7 Joint Surgery there was no difference in the magnitude of these treatment
Abdu et al8 8-year outcomes of DS Spine 2018 effects between the cohorts. A combined as-treated analysis
Reoperation factors found treatment effects of surgery to be 12.6 for SF-36 BP,
Gerling Reoperation risk factors Spine 2017 8.6 for SF-36 PF, and −9.4 for ODI (Table 3). Beyond
et al9 (DS)
Gerling Reoperation risk factors Spine 2016
reporting 4-year results, this study also demonstrated that
et al10 (LSS) the benefit obtained from surgical intervention persists at
Radcliff Reoperation at index or Spine 2013 4 years without any significant decline.
et al11 adjacent level More recently, Lurie et al5 published 8-year results.
Operative factors They reported a follow-up rate of 52% and 55% in the
Desai Impact of durotomy (DS) Spine 2012
et al12 observational and randomized cohorts, respectively. By
Desai Impact of durotomy (LSS) Neurosurgery 2011 8 years, 52% of patients randomized to nonoperative care
et al13 and underwent surgery. An intent-to-treat analysis found
Abdu Fusion method (DS) Spine 2009 no difference in primary outcomes between these patients.
et al14
Patient-specific factors
The as-treated analysis of the randomized cohort found
Mcguire Impact of extreme obesity Spine 2014 that the treatment effect of surgery diminished after the
et al15 fourth year and became insignificant after the fifth year.
Freedman Impact of diabetes Spine 2011 To contrast, the as-treated analysis of the observational
et al16 cohort found that the treatment effect remained statisti-
Radcliff Impact of duration of Spine 2011
et al17 symptoms cally significant for all 3 primary outcomes at 8 years. The
18
Fritz et al Evaluation of physical The Spine Journal 2013 authors noted that a combined as-treated analysis also
therapy demonstrated continued benefit from surgery.
Kasner Impact of epidural steroid Spine 2013
et al19 injections
Pearson Radiographic predictors of Spine 2008
et al20 outcomes (DS) DS RESULTS
Krebs Predictors of long-term Journal of Pain 2010 Two-year results were first published by Weinstein et al6
et al21 opioid use in 2007 in The New England Journal of Medicine. This study
Pearson Treatment effect predictors Spine 2012 compared operative and nonoperative treatment for patients
et al22
Health economics who had DS with isolated spinal stenosis exclusively at that
Tosteson Cost-effectiveness Annals of Internal 2008 level. Surgical treatment involved lumbar decompression with
et al23 Medicine or without posterolateral lumbar fusion. There were 304 and
Weinstein Costs effects Medical Care 2014 303 in the randomized and observational cohorts, respectively.
et al24
Similar to other cohorts, there were high rates of crossover in
DS indicates degenerative spondylolisthesis; LSS, lumbar spinal stenosis. the randomized cohort: 36% of the surgical cohort elected
nonoperative treatment and 49% of the nonoperative treat-
ment elected to undergo surgery by 2 years. An intent-to-treat
289 of which were in the randomized cohort and 365 in the analysis of the randomized cohort found no significant dif-
observational cohort. In the randomized cohort, 67% of ference among primary outcomes. These results differed
those assigned surgery and 43% of those assigned non- greatly with a combined as-treated analysis which found sta-
operative care underwent surgery by 2 years. An intent-to- tistically significant differences across all primary outcomes,
treat analysis found that there was a 7.8-point significant strongly in favor of surgical treatment (Table 4).
increase on the SF-36 scale for BP with surgical treatment. Four-year results similarly revealed no significant
No differences were seen in ODI or SF-36 PF. However, differences between operative and nonoperative cohorts in
the findings of the as-treated analysis were in notable an intent-to-treat analysis.7 The as-treated analysis found
contrast—there was significant improvement in all pri- significant differences in primary outcomes favoring surgi-
mary outcomes for the surgical group in both the cal intervention in both the observational and randomized
randomized and observational cohort (Table 2). Notably, cohort. The combined analysis (Table 4) demonstrated the
the authors also compared the treatment effect between the sustained treatment effect of surgical intervention at 4 years
randomized and observational cohorts and found there was postoperatively. Nonadherence and crossover remained
no statistical difference between them. The combined as- significant as 54% and 66% of patients randomized to

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved. www.clinicalspinesurgery.com | 273

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Ilyas et al Clin Spine Surg  Volume 32, Number 7, August 2019

FIGURE 1. T2-weighted magnetic resonance images sagittal (A) and axial (B) demonstrating severe lumbar stenosis at
L4-L5.

nonoperative care and operative care, respectively, under- all primary outcomes at 2 years (SF-36 BP: +11.7, SF-36 PF:
went surgery by 4 years. Recently, 8-year results were +8.3, and ODI: −7.5), without any occurrences of cauda
published by Abdu et al.8 Their as-treated analysis demon- equina. This demonstrates that patients who elect non-
strated sustained clinical improvement with surgical operative treatment will likely experience a stable clinical
intervention across all primary outcomes (Table 4). course. Although these patients will also likely improve with
This study was also significant for providing insight on nonoperative treatment, it will only be a fraction of that seen
the natural history of DS. On an average, patients who un- with surgery.
derwent nonoperative care experienced moderate increases in
REOPERATION AND INTRAOPERATIVE
COMPLICATIONS
In the DS cohort, the reoperation rate was 22% at
8 years.9 This included surgery at the prior or adjacent
level. A total of 54% of these reoperations occurred within
the first 2 years. A total of 45% of the reoperations were
due to late complications, including recurrent stenosis or
worsening spondylolisthesis and 36% were due to acute
local wound complications, including infection, de-
hiscence, and hematomas. Patients who had reoperations
still experienced significant improvement in primary out-
comes; however, this was blunted in comparison to those
who had no reoperations. The authors also found that
increased age, more advanced stenosis, predominant back
pain, lack of neurogenic claudication, no physical therapy,
and higher leg pain scores were associated with

TABLE 2. Surgical Treatment Effects From the As-treated

Analysis at 2 and 4 Years for Spinal Stenosis
2y 4y
Measured Randomized Observational Randomized Observational
Outcome Cohort Cohort Cohort Cohort
SF-36 BP 11.7 15.3 11.4 14.9
SF-36 PF 8.1 13.6 8 10.1
ODI −8.7 −13.1 −7.8 −11.5
Treatment effect—calculated as the difference between the mean changes in the
surgical and nonsurgical cohorts, respectively.
BP indicates bodily pain; ODI, Oswestry Disability Index; PF, physical func-
FIGURE 2. Lateral x-ray of the lumbar spine demonstrating a tion; SF-36, 36-Item Short-Form Health Survey.
grade 1 spondylolisthesis of L4 on L5.

274 | www.clinicalspinesurgery.com Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Clin Spine Surg  Volume 32, Number 7, August 2019 Review of LSS and SPORT

concluded that baseline ODI scores <56, being a nonsmoker,

TABLE 3. Surgical Treatment Effects in the Combined As-
treated Analysis for Spinal Stenosis presence of neuroforaminal stenosis, predominant leg pain, and
not lifting at work were independent factors for greater treat-
Measured Outcome 2y 4y 8y
ment effects. Regarding ODI scores, this study suggested that
SF-36 BP 13.6 12.6 10.9 patients with the greatest amount of disability responded sig-
SF-36 PF 11.1 8.6 10.6 nificantly better to nonoperative treatment than those with less
ODI −11.2 −9.4 −11.2
disability, strengthening the support for initial nonoperative
Treatment effect—calculated as the difference between the mean changes in the treatment in even the high baseline ODI population.
surgical and non-surgical cohorts, respectively.
BP indicates bodily pain; ODI, Oswestry Disability Index; PF, physical func-
Freedman et al16 evaluated the impact of diabetes
tion; SF-36, 36-Item Short-Form Health Survey. mellitus on patient complications and outcomes. They
found that the stenosis patients with diabetes mellitus were
mildly older, had a higher body mass index (BMI), and
reoperation.9 This analysis did not find obesity, smoking, higher incidence of hypertension. On the aggregate, dia-
or diabetes to be associated with reoperation. betic patients were found to have a greater incidence of
Similarly, the reoperation was 13% and 18% at 4 and postoperative medical and surgical complications, without
8 years in the spinal stenosis group, respectively.10,11 A total a significantly greater incidence of infections. With regards
of 42% and 70% of these reoperations were within the first 2 to all 3 primary outcomes, there was no difference in the
and 4 years, respectively. Recurrent stenosis or further treatment effect for surgery between diabetic and non-
spondylolisthesis was the surgical indication in 52% of the diabetic patients at 4 years postoperatively. Interestingly,
cases. Although patients who underwent reoperation expe- the authors found that nonoperative treatment was more
rienced clinically significant improvement, their treatment effective in nondiabetic patients than diabetic patients
effect was also blunted when compared with the cohort who with regards to SF-36 PF and ODI.16
did not undergo reoperation. Patients who underwent re- In the DS cohort, patients with diabetes were found to
operation were less likely to have a neurological deficit on have greater comorbidities and a higher BMI.16 No difference
their index presentation (P = 0.04); however, no other patient in medical or surgical complications was noted. Diabetic
factors (such as smoking, diabetes, or presence of workmen’s patients were found to have significant improvement with
compensation) were found to be associated with reoperation. surgical treatment when compared with their nonoperative
Radcliff and colleagues did note that at 4 years, the only counterparts, however, this improvement was less than that
discernable risk factor for reoperation was duration of seen in nondiabetic patients. In contrast to spinal stenosis,
symptoms for > 12 months. there was no difference in the effectiveness of nonoperative
A review of the DS patients found an incidence of treatment between the diabetic and nondiabetic patients.
intraoperative durotomy to be 10.5%.12 Notably, they found Radcliff et al17 also performed an analysis comparing
no increased risk of postoperative complications or difference LSS patients with greater and less than 12 months of symp-
in clinical outcomes, annually, up to 4 years postoperatively in toms. They found that patients with longer duration of
these patients. Among the spinal stenosis patients, the rate of symptoms experienced less improvement than those with
incidental durotomy was found to be 9% and associated with <12 months of symptoms with regards to all 3 primary out-
a longer operative time and mildly increased length of stay comes (P = 0.002 to 0.007). These findings were mirrored in the
(4.3 vs. 3.1 d, P = 0.003).13 Similarly to the DS cohort, this nonoperative cohort where patients with <12 months of
study found no difference in clinical outcomes, up to 4 years symptoms did significantly better than their counterparts with
postoperatively. > 1 year of symptoms, similarly in all primary outcomes. To
contrast, no clinical differences were found in the DS cohorts,
PATIENT FACTORS irrespective of the patients’ duration of symptoms. The authors
Pearson et al22 performed a multivariate model analysis concluded that LSS patients experienced less improvement,
evaluating 53 variables to see which were significant factors in with both surgical and nonsurgical treatment, if they had
predicting treatment effect of ODI scores in the spinal stenosis symptoms for > 12 months. Clinically, this suggests that pa-
cohort. By assessing treatment effect, this study also accounted tients who seek care and formal treatment for spinal stenosis
for the effect of those variables on nonoperative care. They earlier in their course may have better clinical outcomes.
McGuire et al15 analyzed the impact of extreme obesity,
defined as a BMI > 35, in patients enrolled in SPORT. In the
TABLE 4. Surgical Treatment Effects in the Combined As- spinal stenosis cohort, they found the extreme obesity cohort to
treated Analysis for Degenerative Spondylolisthesis
have significantly more comorbidities and worse SF-36 BP
Measured Outcome 2y 4y 8y scores at baseline. Notably, no difference in wound complica-
SF-36 BP 18.1 15.3 11.8 tions, reoperations, or treatment effect was noted. In the DS
SF-36 PF 18.3 18.9 10.3 cohort, extremely obese patients were found to be significantly
ODI −16.7 −14.3 −10.3 different than their counterparts, including with regards to age,
Treatment effect—calculated as the difference between the mean changes in the comorbidities, socioeconomic status, sex composition, and
surgical and non-surgical cohorts, respectively. baseline clinical outcomes. Reoperations rates were higher in
BP indicates bodily pain; ODI, Oswestry Disability Index; PF, physical func-
tion; SF-36, 36-Item Short-Form Health Survey. the extremely obese group at 3 and 4 years postoperatively.
Treatment effect was numerically greater for the extremely

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved. www.clinicalspinesurgery.com | 275

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Ilyas et al Clin Spine Surg  Volume 32, Number 7, August 2019

obese cohort in all primary outcomes; however, it was only or postoperative complications. With regards to outcomes
statistically significant for SF-36 PF scores. These results averaged over the first 4 years, the authors found that
demonstrate that not only is surgery effective in this cohort, but patients who underwent an ESI and then had surgery had
it may play an even more significant role due to limited success significantly less improvement in SF-36 PF than their
of nonoperative treatment in these patients. counterparts who did not have surgery (difference of 7.7,
One of the benefits of the observational cohort is P = 0.025). Statistically, the changes in SF-36 BP and ODI
that it provided insight into the decision-making process were not inherently significant, although they trended to
from a patient’s perspective. Kurd et al40 analyzed the favor the non-ESI cohort.
observational cohort of patients with spinal stenosis to Interestingly, patients who underwent nonsurgical
assess factors influencing the patient’s decision. They treatment and received an ESI also fared worse than their
found that patients who elected surgical treatment were nonsurgical counterparts with regards to SF-36 BP (dif-
younger and averaged 13.5, 14.6, and 12.9 points worse on ference of 9.4, P = 0.007) and SF-36 PF (difference of 9.7,
the SF-36 BP and PF and ODI scores, respectively. Pa- P = 0.009). These findings were in contrast to the IDH arm
tients with lateral recess stenosis were also significantly of SPORT, which found receiving an ESI to have no im-
more likely to proceed with surgical intervention (0.022). pact on clinical outcomes over 4 years, as well as a sig-
This in combination with preoperative scores may suggest nificantly increased likelihood for crossover from surgical
that lateral recess stenosis presents with more severe to nonsurgical treatment.43 Given the prevalent use of
symptoms and potentially more clinical disability, leading ESIs, these findings are particularly interesting and war-
to a greater preference for surgical intervention. rant further investigations into the effects, pathophysio-
logy, and utility of ESIs.

NONOPERATIVE TREATMENT COST-EFFECTIVENESS

Although physical therapy is a first-line treatment
A cost-analysis evaluation at 2 years postoperatively
for spinal stenosis, it’s impact and role has not been clearly
found the average total cost, including direct and indirect
defined. Fritz et al18 evaluated patients who underwent
costs, was $26,222 for spinal stenosis and $42,081 for
nonoperative treatment from a combined cohort. They DS.23 These were calculated as comprehensive costs and
found that the 37% of patients who underwent physical
included surgery, imaging, diagnostic tests, therapy, and
therapy in the first 6 weeks within enrollment had a lower
indirect costs of loss of productivity due to missing work.
rate of crossover into surgical treatment than their coun- In comparison with nonoperative treatment, the authors
terparts who did not undergo physical therapy (21% vs.
found the marginal cost per quality adjusted life years as
33%, P = 0.045). The authors also found that PT was as-
$77,600 for spinal stenosis and $115,600 for DS in the first
sociated with significant improvements in SF-36 PF and
2 years. Because of sustained improvement from surgical
self-related progress at 6 months and 1 year. No difference
intervention, this cost per quality adjusted life years im-
in ODI, SF-36 BP, or opioid use was noted.
proved to $59,400 and $64,300, respectively for the 2
Opioid use has become an increasingly controversial
groups at 4 years postoperatively.24 These results dem-
topic across all medical and surgical specialties.41 Krebs
onstrated the feasibility of surgical treatment for both
et al21 evaluated patients in all cohorts of SPORT (spinal
lumbar stenosis and DS.
stenosis, DS, and IDH) and found a baseline opioid use of
42.3%. In total, 25% and 21% of these patients continued
opioid use at 1 and 2 years postoperatively, respectively. FUSION TECHNIQUES AND RADIOGRAPHIC
In an adjusted analysis, the authors found that smoking FINDINGS
and nonsurgical treatment were the strongest and most Patients with DS who underwent fusion were treated
significant predictors of long-term opioid use, suggesting with either uninstrumented fusion, pedicle screw fixation,
that surgical treatment was less likely to continue opioid or pedicle screw fixation with interbody fusion (either
use when compared with nonsurgical treatment. anterior, posterior, or transformational).14 Abdu et al14
The role of epidural steroid injections (ESI) has been evaluated these different fusion modalities and compared
well-contested in the literature.42 Although their ther- their outcomes at 4 years postoperatively. Different mo-
apeutic role and cost-effectiveness are not agreed upon, dalities were associated with better outcomes transiently,
most agree that epidural injections provide a diagnostic however, no difference was appreciated at 3 or 4 years.
value in determining a patient’s pain generator. Radcliff This analysis was limited by small cohorts, insufficient
and colleagues performed a subgroup analysis evaluating information, and loss to follow-up, but certainly estab-
patients who underwent epidural injections before lished the need for further investigation.
surgery.19 Sixty-nine patients received an ESI within the DS presents with a spectrum of disease, including
first 3 months of enrollment compared with 207 who did mobility and extent of disk degeneration. Pearson et al20
not. There were no differences between these groups with performed a retrospective analysis comparing clinical
regards to baseline scores or severity of symptoms. Inter- outcomes in groups stratified by severity of disease. They
estingly, the authors found that the ESI group was asso- found that grade 2 slips had greater treatment effects than
ciated with greater operative time (26.5 min) and length of grade 1 slips at 1 year; however, this difference became
stay (0.9 d) (P < 0.05) with no difference in intraoperative insignificant at 2 years. No significant differences were

276 | www.clinicalspinesurgery.com Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Clin Spine Surg  Volume 32, Number 7, August 2019 Review of LSS and SPORT

found based on the extent of disk height loss. The authors CONCLUSIONS
evaluated mobility based on extension-flexion radiographs SPORT has been one of the most influential studies
and found that patients with hypermobile segments re- in providing insight into the natural progression and
sponded better to nonoperative treatment than the stable treatment effect of various operative and nonoperative
spondylolisthesis group. This yielded lower treatment ef- therapies in the treatment of spinal stenosis and DS.
fects for the hypermobile patients. This effect was sub- Current data continue to demonstrate the effectiveness
sequently found to be insignificant at 2 years. Although of operative intervention for these conditions (Tables 2
this was limited by sample size (n = 44), these results and 3). These data have also allowed the understanding
suggest the feasibility of attempting nonoperative treat- and stratification of treatment effectiveness, based on a variety
ment in hypermobile patients as well. of factors, including, presenting pathology, symptomatology,
as well as patient-host factors.
LIMITATIONS
Despite the profound results of the SPORT, the trial REFERENCES
has been criticized for some shortcomings. Although SPORT 1. Asghar FA, Hilibrand AS. The Impact of the Spine Patient
was designed as a randomized trial, there was significant Outcomes Research Trial (SPORT) Results on Orthopaedic Practice.
patient crossover in the randomized groups. By 2 years, the J Am Acad Orthop Surg. 2012;20:160–166.
crossover rates were 33% and 43% in the spinal stenosis 2. Birkmeyer NJO, Weinstein JN, Tosteson ANA, et al. Design of the
Spine Patient Outcomes Research Trial (SPORT). Spine (Phila Pa
surgical and nonsurgical cohorts, respectively.3 In the DS 1976). 2002;27:1361–1372.
group, crossover rates were 36% and 49% in the surgical and 3. Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonsurgical
nonsurgical cohorts, respectively.6 The phenomenon of therapy for lumbar spinal stenosis. N Engl J Med. 2008;358:794–810.
crossover is inevitable in a surgical trial, irrespective of 4. Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus
randomization as patient autonomy is a well-established nonoperative treatment for lumbar spinal stenosis four-year results
of the Spine Patient Outcomes Research Trial. Spine (Phila Pa
value in medicine that trumps other priorities. Statistically, 1976). 2010;35:1329–1338.
this clouds the utility of an intent-to-treat analysis as there is 5. Lurie JD, Tosteson TD, Tosteson A, et al. Long-Term Outcomes of
a misrepresentation of each group. However, significant Lumbar Spinal Stenosis: Eight-Year Results of the Spine Patient
value is still found in each cohort’s as-treated analysis, as Outcomes Research Trial (SPORT) NIH Public Access. Spine
January. 2015;15:63–76.
well as the combined as-treated analysis. 6. Weinstein J, Lurie J, Tosteson T, et al. Surgical versus nonsurgical
Another concerning aspect of SPORT is the variability in treatment for lumbar degenerative spondylolisthesis. N Engl J Med.
both the specific treatment modalities of operative and non- 2007;356:2257–2270.
operative treatment. Patients who underwent nonoperative 7. Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical compared with
treatment were as an minimum, provided with physical ther- nonoperative treatment for lumbar degenerative spondylolisthesis:
four-year results in the Spine Randomzied and Observational
apy, home exercises, and nonsteroidal anti-inflammatories, as Cohorts. J Bone Joint Surg Am. 2009;91:1295–1304.
medically tolerated.2 Beyond this, there was great variability in 8. Abdu WA, Sacks OA, Tosteson ANA, et al. Long-term results of
terms of the type and extent of physical therapy, different types surgery compared with nonoperative treatment for lumbar degener-
of injections, including trigger points, facet blocks, and epidural ative spondylolisthesis in the Spine Patient Outcomes Research Trial
(SPORT). Spine (Phila Pa 1976). 2018;43:1619–1630.
injections, as well as different types of medications, such as 9. Gerling MC, Leven D, Passias PG, et al. Risks factors for reoperation
membrane-stabilizers, antidepressants, oral steroids, etcetera. in patients treated surgically for degenerative spondylolisthesis. Spine
This provides a challenge in generalizing the results of the (Phila Pa 1976). 2017;42:1559–1569.
nonoperative cohort, particularly assessing treatment effect, as 10. Gerling M, Leven D, Passias P, et al. Risk factors for reoperation in
they may not have consistent protocols within that cohort. patients treated surgically for lumbar stenosis. Spine (Phila Pa 1976).
2016;41:901–909.
Similar variability was found in the surgically treated 11. Radcliff KE, Curry P, Hilibrand A, et al. Risk for adjacent segment
cohorts as well; the operative guidelines of DS permitted the and same segment reoperation after surgery for lumbar stenosis: a
use of autogenous iliac bone graft, both instrumented and subgroup analysis of the Spine Patient Outcomes Research Trial
noninstrumented fusion, as well as the use of an interbody (SPORT). Spine (Phila Pa 1976). 2014;38:531–539.
12. Desai A, Ball PA, Bekelis K, et al. Surgery for lumbar degenerative
device.2,6 The definition and variability of “spondylolisthesis” spondylolisthesis in Spine Patient Outcomes Research Trial: does incidental
allows provides a similar challenge. In SPORT, a lateral durotomy affect outcome? Spine (Phila Pa 1976). 2012;37:406–413.
standing x-ray was used to assess for spondylolisthesis. Of the 13. Desai A, Ball PA, Bekelis K, et al. SPORT: does incidental
included cohort, 84% of the DS cohort had a grade 1 slip.6 durotomy affect longterm outcomes in cases of spinal stenosis?
Unfortunately, there was no further distinction of the stability Neurosurgery. 2015;76:S57–S63.
14. Abdu WA, Lurie JD, Spratt KF, et al. Degenerative spondylolis-
of these slips, as to whether there was motion or reduction on thesis: does fusion method influence outcome? Four-year results of
bending films. This heterogeneity in surgical cohorts and the spine patient outcomes research trial. Spine (Phila Pa 1976).
treatment options may limit the applicability of the reported 2009;34:2351–2360.
clinical effectiveness, clinical outcomes, complication rates, 15. Mcguire KJ, Khaleel MA, Rihn JA, et al. The effect of extreme
obesity on outcomes of treatment for lumbar spinal conditions:
and cost-effectiveness for treatment of DS. subgroup analysis of the Spine Patient Outcomes Research Trial
(SPORT). Spine (Phila Pa 1976). 2014;39:1975–1980.
AUTHORS’ PREFERRED TREATMENT 16. Freedman MK, Hilibrand AS, Blood EA, et al. The impact of
diabetes on the outcomes of surgical and nonsurgical treatment of
Our preferred treatment of LSS is open decompression patients in the Spine Patient Outcomes Research Trial. Spine (Phila
with laminectomy. Pa 1976). 2011;36:290–307.

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved. www.clinicalspinesurgery.com | 277

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.

Ilyas et al Clin Spine Surg  Volume 32, Number 7, August 2019

17. Radcliff KE, Rihn J, Hilibrand A, et al. Does the duration of 31. Ulrich NH, Burgstaller JM, Pichierri G, et al. Decompression
symptoms in patients with spinal stenosis and degenerative surgery alone versus decompression plus fusion in symptomatic
spondylolisthesis affect outcomes? Analysis of the Spine Outcomes lumbar spinal stenosis. Spine. 2017;42:E1077–E1086.
Research Trial. Spine (Phila Pa 1976). 2011;36:2197–2210. 32. Försth P, Ólafsson G, Carlsson T, et al. A randomized, controlled
18. Fritz JM, Lurie JD, Zhao W, et al. Associations between physical trial of fusion surgery for lumbar spinal stenosis. N Engl J Med. 2016;
therapy and long-term outcomes for individuals with lumbar spinal 374:1413–1423.
stenosis in the SPORT study. Spine J. 2014;14:1611–1621. 33. Donnarumma P, Tarantino R, Nigro L, et al. Decompression versus
19. Kasner E, Hunter CA, Kariko K. Epidural steroid injections are associated decompression and fusion for degenerative lumbar stenosis: analysis
with less improvement in the treatment of lumbar spinal stenosis: a of the factors influencing the outcome of back pain and disability.
subgroup analysis of the SPORT. Spine (Phila Pa 1976). 2013;38:279–291. J Spine Surg (Hong Kong). 2016;2:52–58.
20. Pearson AM, Lurie JD, Blood EA, et al. Spine patient outcomes 34. Asil K, Yaldiz C. Retrospective comparison of radiological and
research trial: radiographic predictors of clinical outcomes after clinical outcomes of PLIF and TLIF techniques in patients who
operative or nonoperative treatment of degenerative spondylolis- underwent lumbar spinal posterior stabilization. Medicine (Baltimore).
thesis. Spine (Phila Pa 1976). 2008;33:2759–2766. 2016;95:e3235.
21. Krebs EE, Lurie JD, Fanciullo G, et al. Predictors of long-term 35. McAnany SJ, Baird EO, Qureshi SA, et al. Posterolateral fusion
opioid use among patients with painful lumbar spine conditions. versus interbody fusion for degenerative spondylolisthesis. Spine
J Pain. 2010;11:44–52. (Phila Pa 1976). 2016;41:E1408–E1414.
22. Pearson A, Lurie J, Tosteson T, et al. Who should have surgery for spinal 36. Sakaura H, Miwa T, Yamashita T, et al. Posterior lumbar interbody
stenosis? Treatment effect predictors in SPORT. Spine (Phila Pa 1976). fusion with cortical bone trajectory screw fixation versus posterior
2012;37:1791–1802. lumbar interbody fusion using traditional pedicle screw fixation for
23. Tosteson ANA, Lurie JD, Years SC. Surgical treatment of spinal degenerative lumbar spondylolisthesis: a comparative study. J Neuro-
stenosis with and without degenerative spondylolisthesis: cost- surg Spine. 2016;25:1–6.
effectiveness after 2 years. Ann Intern Med. 2008;149:845–853. 37. Mobbs RJ, Phan K, Malham G, et al. Lumbar interbody fusion:
24. Weinstein JN, Tosteson ANA, Tosteson TD, et al. The SPORT techniques, indications and comparison of interbody fusion options
value compass: do the extra costs of undergoing spine surgery including PLIF, TLIF, MI-TLIF, OLIF/ATP, LLIF and ALIF.
produce better health benefits? Med Care. 2014;52:1055–1063. J Spine Surg. 2015;1:2–18.
25. Bae HW, Rajaee SS, Kanim LE. Nationwide trends in the surgical 38. Lee N, Kim KN, Yi S, et al. Comparison of outcomes of anterior,
management of lumbar spinal stenosis. Spine (Phila Pa 1976). 2013; posterior, and transforaminal lumbar interbody fusion surgery at a
38:916–926. single lumbar level with degenerative spinal disease. World Neurosurg.
26. Deyo RA, Mirza SK, Martin BI, et al. Trends, major medical 2017;101:216–226.
complications, and charges associated with surgery for lumbar spinal 39. Campbell RC, Mobbs RJ, Lu VM, et al. Posterolateral fusion versus
stenosis in older adults. JAMA. 2010;303:1259. interbody fusion for degenerative spondylolisthesis: systematic
27. Matz PG, Meagher RJ, Lamer T, et al. Guideline summary review: review and meta-analysis. Glob Spine J. 2017;7:482–490.
an evidence-based clinical guideline for the diagnosis and treatment 40. Kurd MF, Lurie JD, Zhao W, et al. Predictors of treatment choice in
of degenerative lumbar spondylolisthesis. Spine J. 2016;16:439–448. lumbar spinal stenosis. Spine (Phila Pa 1976). 2012;37:1702–1707.
28. Inui T, Murakami M, Nagao N, et al. Lumbar degenerative 41. Manchikanti L, Ii SH, Fellows B, et al. Opioid epidemic in the
spondylolisthesis. Spine (Phila Pa 1976). 2017;42:E15–E24. United States. Pain Physician. 2012;15:9–38.
29. Ghogawala Z, Dziura J, Butler WE, et al. Laminectomy plus fusion 42. Chou R, Hashimoto R, Friedly J, et al. Epidural corticosteroid
versus laminectomy alone for lumbar spondylolisthesis. N Engl J Med. injections for radiculopathy and spinal stenosis: a systematic review
2016;374:1424–1434. and meta-analysis. Ann Intern Med. 2015;163:373–381.
30. Ahmad S, Hamad A, Bhalla A, et al. The outcome of decompression 43. Radcliff K, Hilibrand A, Lurie JD, et al. The impact of epidural
alone for lumbar spinal stenosis with degenerative spondylolisthesis. steroid injections on the outcomes of patients treated for lumbar disc
Eur Spine J. 2017;26:414–419. herniation. J Bone Joint Surg. 2012;94:1353–1358.

278 | www.clinicalspinesurgery.com Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Copyright r 2019 Wolters Kluwer Health, Inc. All rights reserved.