Advertisement

Predictors for failure after surgery for lumbar spinal stenosis, a prospective observational study

  • Ole Kristian Alhaug
    Correspondence
    Corresponding author. Innlandet Hospital Trust, The Research Center for Age-Related Functional Decline and Disease, PO Box 68, N-2313 Ottestad, Norway. Tel.: +47 41127443
    Affiliations
    Innlandet Hospital Trust, The Research Center for Age-Related Functional Decline and Disease, PO Box 68, N-2313 Ottestad, Norway

    Norwegian University of Science and Technology, NTNU PO Box 191, N-7491 Trondheim, Norway

    Orthopedic department, Akershus University hospital, PO Box 1000, N-1478 Loerenskog, Norway
    Search for articles by this author
  • Filip C. Dolatowski
    Affiliations
    Orthopedic department, Oslo University hospital, PO Box 4956, N-0424, Oslo, Norway
    Search for articles by this author
  • Tore K. Solberg
    Affiliations
    Neurosurgical department, University hospital of North Norway, N-9038, Tromsoe, Norway
    Search for articles by this author
  • Greger Lønne
    Affiliations
    Innlandet Hospital Trust, The Research Center for Age-Related Functional Decline and Disease, PO Box 68, N-2313 Ottestad, Norway

    Norwegian University of Science and Technology, NTNU PO Box 191, N-7491 Trondheim, Norway
    Search for articles by this author
Open AccessPublished:November 04, 2022DOI:https://doi.org/10.1016/j.spinee.2022.10.010

      Highlights

      • Thirty-three percent of patients operated for lumbar spinal stenosis report failure of treatment.
      • The strongest predictors for failure are long duration of symptoms, former surgery, and age> 70 years.
      • Socioeconomic factors are also associated with failure.

      Abstract

      Background/context

      Some patients do not improve after surgery for lumbar spinal stenosis (LSS), and surgical treatment implies a risk for complications and deterioration. Patient selection is of paramount importance to improve the overall clinical results and identifying predictive factors for failure is central in this work.

      Purpose

      We aimed to explore predictive factors for failure and worsening after surgery for LSS.

      Study design /setting

      Retrospective observational study on prospectively collected data from a national spine registry with a 12-month follow-up.

      Patient sample

      We analyzed 11,873 patients operated for LSS between 2007 and 2017 in Norway, included in the Norwegian registry for spine surgery (NORspine). Twelve months after surgery, 8919 (75.1%) had responded.

      Outcome measures

      Oswestry Disability Index (ODI) 12 months after surgery.

      Methods

      Predictors were assessed with uni- and multivariate logistic regression, using backward conditional stepwise selection and a significance level of 0.01. Failure (ODI>31) and worsening (ODI>39) were used as dependent variables.

      Results

      Mean (95%CI) age was 66.6 (66.4–66.9) years, and 52.1% were females. The mean (95%CI) preoperative ODI score was 39.8 (39.4–40.1). All patients had decompression, and 1494 (12.6%) had an additional fusion procedure. Twelve months after surgery, the mean (95%CI) ODI score was 23.9 (23.5–24.2), and 2950 patients (33.2%) were classified as failures and 1921 (21.6%) as worse. The strongest predictors for failure were duration of back pain > 12 months (OR [95%CI]=2.24 [1.93–2.60]; p<.001), former spinal surgery (OR [95%CI]=2.21 [1.94–2.52]; p<.001) and age>70 years (OR (95%CI)=1.97 (1.69–2.30); p<.001). Socioeconomic variables increased the odds of failure (ORs between 1.36 and 1.62). The strongest predictors for worsening were former spinal surgery (OR [95%CI]=2.04 [1.77–2.36]; p<.001), duration of back pain >12 months (OR [95%CI]=1.83 [1.45–2.32]; p<.001) and age >70 years (OR [95%CI]=1.79 [1.49–2.14]; p<.001). Socioeconomic variables increased the odds of worsening (ORs between 1.33–1.67).

      Conclusions

      After surgery for LSS, 33% of the patients reported failure, and 22% reported worsening as assessed by ODI. Preoperative duration of back pain for longer than 12 months, former spinal surgery, and age above 70 years were the strongest predictors for increased odds of failure and worsening after surgery.

      Keywords

      Introduction

      Lumbar spinal stenosis (LSS) is a common condition and represents an increasing burden for our health care system. Surgical treatment for LSS is a good option for patients failing nonoperative care [
      • Amundsen T*
      • Weber H
      • Nordal HJ
      • Magnaes B
      • Abdelnoor M
      • Lilleås F
      Lumbar spinal stenosis: conservative or surgical management?: a prospective 10-year study.
      ,
      • Atlas SJ
      • Keller RB
      • Robson DBS
      • Deyo RA
      • Singer DE.
      Surgical and nonsurgical management of lumbar spinal stenosis: four-year outcomes from the maine lumbar spine study.
      ,
      • Atlas SJ
      • Keller RB
      • Wu YA
      • Deyo RA
      • Singer DE.
      Long-term outcomes of surgical and nonsurgical management of lumbar spinal stenosis: 8 to 10 year results from the maine lumbar spine study.
      ,
      • Malmivaara A
      • Slätis P
      • Heliövaara M
      • Sainio P
      • Kinnunen H
      • Kankare J
      • et al.
      Surgical or nonoperative treatment for lumbar spinal stenosis?: a randomized controlled trial.
      ,
      • Weinstein JN
      • Tosteson TD
      • Lurie JD
      • Tosteson ANA
      • Blood E
      • Hanscom B
      • et al.
      Surgical versus nonsurgical therapy for lumbar spinal stenosis.
      ,
      • Kreiner DS
      • Shaffer WO
      • Baisden JL
      • Gilbert TJ
      • Summers JT
      • Toton JF
      • et al.
      An evidence-based clinical guideline for the diagnosis and treatment of degenerative lumbar spinal stenosis (update).
      ].
      Although the results after surgical treatment can vary, many studies report good results in 62%–75% of patients [
      • Nerland US
      • Jakola AS
      • Solheim O
      • Weber C
      • Rao V
      • Lønne G
      • et al.
      Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
      ,
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ,
      • Sigmundsson Freyr G
      • Kang Xiao P
      • Jönsson Bo
      • Strömqvist Björn
      Prognostic factors in lumbar spinal stenosis surgery.
      ,
      • Ogura Y
      • Kobayashi Y
      • Shinozaki Y
      • et al.
      Factors influencing patient satisfaction after decompression surgery without fusion for lumbar spinal stenosis.
      ,
      • Gunzburg R
      • Keller T
      • Szpalski M
      • et al.
      Clinical and psychofunctional measures of conservative decompression surgery for lumbar spinal stenosis: a prospective cohort study.
      ,
      • Kleinstück FS
      • Grob D
      • Lattig F
      • Bartanusz V
      • Porchet F
      • Jeszenszky D
      • et al.
      The influence of preoperative back pain on the outcome of lumbar decompression surgery.
      ,
      • Hébert JJ
      • Abraham E
      • Wedderkopp N
      • et al.
      Preoperative factors predict postoperative trajectories of pain and disability following surgery for degenerative lumbar spinal stenosis.
      ]. Furthermore, even though the surgical techniques constantly develop toward less invasiveness, the clinical results seem stable [
      • Nerland US
      • Jakola AS
      • Solheim O
      • Weber C
      • Rao V
      • Lønne G
      • et al.
      Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
      ,
      • Simpson AK
      • Lightsey IV
      • >Xiong GX
      • Crawford AM
      • Minamide A
      • Schoenfeld AJ
      Spinal endoscopy: evidence, techniques, global trends, and future projections.
      ].
      Of note, surgery also implies a risk for complications and deterioration, which underlines the importance of proper patient selection for surgery. It is tempting to consider a clinician's experience and judgment as the best prerequisite to selecting suitable patients for surgery. However, complex and subtle clinical pictures are difficult to perceive, and expert surgeons may overestimate the benefit of surgery and underestimate the risk of unfavorable outcomes [
      • Sacks GD
      • Dawes AJ
      • Ettner SL
      • Brook RH
      • Fox CR
      • Maggard-Gibbons M
      • et al.
      Surgeon perception of and benefit in the decision to operate.
      ]. Furthermore, risk factors may be positively or negatively associated with results, interact, and be subject to confounding. It may be challenging to overview a range of predictors in a clinical setting and assess their combined clinical relevance. However, previous studies have shown that a combined set of predictors perform better than one single in predicting outcomes and that predictors are superior to clinical judgment. Hence predictor analyses are central for informing and improving patient selection and clinical decision-making before surgery [
      • Kahneman D
      • Sibony O
      • Sunstein CR.
      Noise: a flaw in human judgment.
      ,
      • Meehl PE.
      Clinical versus statistical prediction: a theoretical analysis and a review of the evidence.
      ,
      • Grove WM
      • Zald DH
      • Lebow BS
      • Snitz BE.
      • Nelson C.
      Clinical versus mechanical prediction: a meta-analysis.
      ,
      • Yu MC.
      • Kuncel NR.
      Pushing the limits for judgmental consistency: comparing random weighting schemes with expert judgments.
      ].
      Many patients seem to understand the uncertainty regarding clinical outcomes and that not everybody improves. The risk of getting worse may be harder to accept. Former predictor studies are based on different databases (mandatory registers, voluntary registers, or a few treating centers) and include different variables or definitions of variables. Hence, the results are not necessarily transferable to other settings. Some former studies focus on improvement rather than failure or worsening. Risk assessment concerning unfavorable outcomes is crucial for informing patients and making clinical decisions and could aid in reducing the adverse effects of spine surgery [
      • Yap ZL
      • Summers SJ
      • Grant AR
      • Moseley GL
      • Karran EL.
      The role of the social determinants of health in outcomes of surgery for low back pain: a systematic review and narrative synthesis.
      ].
      We aimed to explore predictors for failure and worsening after surgery for LSS in a national comprehensive spine registry.

      Material/method

      This retrospective study on prospectively collected data includes adult patients operated on for lumbar spinal stenosis (LSS) between 2007 and 2017 in Norway. The Norwegian registry for spine surgery (NORspine) is a mandatory register with a coverage of 70% at the case level and 100% at the surgical unit level [
      ]. The registration process includes a preoperative form on socio-economical items and standard PROMs filled by the patients at admission for surgery (baseline). The surgeon fills out one postoperative form on diagnosis and surgical details. The patients complete two follow-up forms, one at three months and one at 12 months. They include common PROMs and a global perceived effect (GPE) transition scale. The PROMs are the Norwegian translation of the Oswestry Disability Index (ODI), a pain-related disability score ranging from 0 (no impairment) to 100 (bedbound) [
      • Fairbank JC
      • Pynsent PB.
      The oswestry disability index.
      ,
      • Grotle M
      • Brox JI
      • Vøllestad NK.
      Cross-cultural adaption of the Norwegian versions of the roland-morris disability questionnaire and the oswestry disability index.
      ], and Numerical Rating Scales (NRS) for back and leg pain (ranging from 0 = no pain to 10 = worst pain imaginable) [
      • McCaffery M.
      Pain, clinical manual for nursing practice.
      ]. The GPE scale have seven steps (1=completely recovered, 2=much improved, 3=somewhat improved, 4=unchanged, 5=somewhat worse, 6=much worse, 7=worse than ever) [
      • Kamper SJ
      • Ostelo RWJG
      • Knol DL
      • Maher CG
      • de Vet HCW
      • Hancock MJ.
      Global Perceived Effect scales provided reliable assessments of health transition in people with musculoskeletal disorders, but ratings are strongly influenced by current status.
      ].
      Primary outcome: We defined failure as ODI final score >31 points and worsening as ODI >39 points, in accordance with a recent study [
      • Alhaug OK
      • Dolatowski FC
      • Solberg TK
      • Lønne G.
      Criteria for failure and worsening after surgery for lumbar spinal stenosis: a prospective national registry observational study.
      ].
      Sensitivity analysis: To evaluate the robustness of the prediction we defined an ODI change of less than 8 points as failure and less than 4 points as worsening [
      • Alhaug OK
      • Dolatowski FC
      • Solberg TK
      • Lønne G.
      Criteria for failure and worsening after surgery for lumbar spinal stenosis: a prospective national registry observational study.
      ]. Finally, we also used GPE to assess the effect after surgery.

      Statistics

      We report central tendency in terms of mean (95% CI) for continuous data with normal distribution and number and proportions (%) for categorical data. We assessed predictors using uni- and multivariate logistic regression, with backward conditional stepwise selection with an entry and removal threshold of 0.01.
      ODI score 12 months after surgery of 31 for failure and 39 for worsening were used as dependent variables (outcome). Covariates in the predictor analyses were chosen according to previous literature: age, gender, smoking, ASA classification, BMI, educational level, civil status, Norwegian speakers, disability benefit, former spinal surgery, MRI findings, preoperative ODI score, duration of symptoms, multilevel surgery [
      ICHOM
      (International Consortium for Health Outcomes Measurements).
      ,
      • van hoof ML
      • Jacobs WC
      • Willems PC
      • Wouters MWJM
      • de Kleuver M
      • Peul WC
      • et al.
      Evidence and practice in spine registries . A systematic review, and recommendations for future design of registries.
      ,
      • van Hooff ML
      • van Loon J
      • van Limbeek J
      • de Kleuver M.
      The Nijmegen decision tool for chronic low back pain. Development of a clinical decision tool for secondary or tertiary spine care specialists.
      ]. Among the covariate variables, some were dichotomized to improve the data-to-model fit and facilitate interpretation of the analyses (age, BMI, ASA classification, and educational level). There was no strong (<0.7) correlation between the covariates, and linearity between continuous variables was checked against logit failure. Only preoperative ODI had nonlinearity, as displayed in Figure. Covariates were tested for relevant interactions using multivariate logistic regression, and no interactions had a statistically significant association with the outcomes.
      Figure
      FigureNon linear realtionship between preoperaitve ODI and logit failure.

      Subgroup analyses

      To explore the role of preoperative back pain, we dichotomized the population into those who reported more back than leg pain (yes/no). We reported the number (%) of failures in patients who received decompression and fusion versus those who received decompression only. We also explored the role of the number of levels operated by analyzing the number (%) of patients reporting failure who were operated on at one, two, three, and four levels. We performed subgroup analyses of patients with previous surgery at the same or another level. For secondary explorative analyses, we used simple crosstabulations.
      We did not impute any missing data. All statistical analyses were done with SPSS version 26 (IBM Corp. released in 2017. IBM SPSS Statistics for Windows, Version 25. Armonk, NY, USA).

      Ethics

      Participation in the registry is voluntary and includes written consent. The study was also approved by The Norwegian Regional Committee for medical and health research ethics ((2017/2157). The study was conducted in accordance with the Helsinki declaration and is presented according to the STROBE statement [
      • von Elm E
      • Altman DG
      • Egger M
      • Pocock SJ
      • Gotzsche PC
      • Vandenbroucke JP.
      The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies.
      ].

      Results

      Baseline

      We identified 11,873 patients operated for LSS between January 2007 and April 2017, 8,863 (74.6%) had completed three months follow-up, and 8,919 (75.1%) had completed 12 months follow-up. Table 1 displays patient characteristics at baseline for all patients and patients with treatment categorized as “failure” and “worsening” subgroups. The mean (95%CI) age was 66.6 (66.4–66.9) years, and 4,644 (52.1%) were females. The mean (95%CI) ODI was 39.8 (39.4–40.1). Patients with failure and worsening were elder, more often ASA >2, and had higher BMI and preoperative ODI. In addition, they more often had low education, comorbidities, disability benefit, and former surgery (Table 1). Patients lost to follow-up at 12 months were younger, more often smokers, and had higher preoperative ODI scores (Appendix Table 1). Table 2 displays the type of surgical treatment given. All patients had some kind of decompression, bilateral foraminotomy was the largest group, and 1,494 (12.6%) patients had an additional fusion procedure.
      Table 1Patient characteristics of 11,873 Norwegian patients with surgically treated lumbar spinal stenosis in a 10 years period (2007–2017)
      All patients, n=11,873 Mean (95%CI), or n (%)Failure (ODI>31), n=2,950 Mean (95%CI), or n (%)S
      Age (cont)65.8 (65.6–66.0)67.8 (67.2–68.0)67.8 (67.3–68.3)
      Age > 70 years4,442 (37.5%)1,352 (45.8%)1,016 (52.9%)
      Gender female6,204 (52.3%)1,714 (58.1%)1,115 (58.0%)
      Civil status, living alone3,169 (26.8%)937 (31.9%)619 (32.4%)
      Native Norwegian speaker11,353 (96.0%)2,796 (95.4%)1,910 (95.3%)
      ASA grade >2
      ASA = American Society of Anesthesiologists classification (1–5).
      2,462 (21.0%)848 (29.1%)601 (31.7%)
      Body mass index (cont)27.6 (27.5–27.7)28.0 (27.9–28.2)28.1 (27.9–28.3)
      Body mass index >302,920 (26.2%)853 (31.1%)573 (32.1%)
      Smoking2,518 (21.4%)682 (23.3%)470 (24.7%)
      Level of education below college8,209 (70.4%)2,281 (79.1%)1,501 (80.1%)
      Any comorbidity7,243 (67.2%)2,031 (75.2%)1,347 (76.0%)
      Receives disability benefit (all types)4,007 (34.8%))1,082 (38.1%)726 (39.3%)
      Previous lumbar spine surgery2,968 (25.3%)1,025 (35.3%)703 (37.1%)
      MRI central stenosis8,288 (69.8%)2,104 (71.3%)1,372 (71.4%)
      MRI lateral stenosis6,796 (57.2%)1,616 (54.8%)878 (45.7%)
      MRI foraminal stenosis1,225 (10.3%)337 (11.4%)218 (11.3%)
      X-ray degenerative olisthesis1,854 (15.6%)416 (14.1%)281 (14.6%)
      Leg pain > 12 months duration7,115 (65.1%)1,940 (72.9%)1,295 (73.8%)
      Back pain > 12 months duration8,415 (75.4%)2,267 (82.3%)1,507 (83.7%)
      Preoperative ODI
      ODI = Oswestry Disability Index (0–100), indicating increasing disability.
      40.3 (40.1–40.6)48.1 (47.6–48.6)50.8 (50.2–51.4)
      Preoperative leg pain (NRS)
      NRS = Numeric Rating Scale (0–10), indicating increasing pain.
      6.59 (6.55–6.63)7.05 (6.97–7.12)7.22 (7.12–7.32)
      Preoperative back pain (NRS)
      NRS = Numeric Rating Scale (0–10), indicating increasing pain.
      6.53 (5.49–6.57)7.24 (7.17–7.31)7.44 (7.35–7.52)
      Preoperative EQ-5D
      ED-5D = EuroQol`s quality of life, (-0.60 to 1.00), indicating increasing quality of life.
      0.363 (0.357–0.369)0.253 (0.241–0.265)0.205 (0.191–0.219)
      All patients, and patients reported as failure and worse.
      low asterisk ASA = American Society of Anesthesiologists classification (1–5).
      ODI = Oswestry Disability Index (0–100), indicating increasing disability.
      NRS = Numeric Rating Scale (0–10), indicating increasing pain.
      § ED-5D = EuroQol`s quality of life, (-0.60 to 1.00), indicating increasing quality of life.
      Table 2Surgical treatment for 11,873 Norwegian patients with lumbar spinal stenosis in a 10 years period (2007–2017)
      Completed 12 months follow up (n=8,919) Mean (95%CI), or n (%)Lost to follow up (n=2,954) Mean (95%CI), or n (%)
      Fusion surgery1,125 (12.6%)369 (12.5%)
       Fusion, TLIF
      Transforaminal Lumbar Interbody Fusion.
      309 (3.5%)120 (4.1%)
       Fusion, PLIF
      Posterior Lumbar Interbody Fusion.
      38 (0.4%)6 (0.2%)
       Fusion, PLF
      Posterolateral Lumbar Fusion.
      769 (8.6%)241 (8.2%)
       Fusion, other9 (0.2%)2 (0.1%)
      Decompression
       Unilateral foramenotomy1,973 (22.1%)732 (24.8%)
       Bilateral foraminotomy3,485 (39.1%)1,120 (37.9%)
       Cross over / “over the top”1,388 (15.6%)544 (18.4%)
       Laminectomy2,199 (24.7%)622 (21.1%)
      More than one level operated3,255 (36.9%)975 (33.3%)
      Patients completed 12 months follow-up, and patients lost to follow-up.
      low asterisk Transforaminal Lumbar Interbody Fusion.
      Posterior Lumbar Interbody Fusion.
      Posterolateral Lumbar Fusion.

      Clinical results

      Twelve months after the operation, 2950 (33.2%) patients were categorized as “failures,” including 1,921 (21.6%) classified as “worse” according to the ODI final score cut-offs. When we used the ODI change score cut–offs, , (32.8%) reported failure, and 2,132 (24.2%) reported worse. The mean (95%CI) ODI 12 months after surgery was 23.9 (23.5–24.2) and the mean (95%CI) improvement in ODI was 15.9 (15.5–16.3) points. When patients graded the effect of surgery by GPE, 1,829 (20.6%) perceived themselves as “unchanged” or any degree of worsening, and 521 (5.9%) reported “much worse” or “worse than ever.”

      Predictors

      Table 3 shows the results of the uni- and multivariate logistic regression analyses.
      Table 3Logistic regression for 8,919 patients operated for lumbar spinal stenosis and registered in NORspine during 2007–2017, using failure (ODI>31) and worsening (ODI>39) as dependent variables and potential predictors as explanatory variables
      Failure (ODI>31 points)Worsening (ODI>39 points)
      UnivariateMultivariateUnivariateMultivariate
      VariablesOR (95%CI)p valueOR (95%CI)p valueOR (95%CI)p valueOR (95%CI)p value
      Age >70 years1.50 (1.37–1.64)<.0011.99 (1.71–2.31)<.0011.50 (1.36–1.66)<.0011.93 (1.62–2.31)<.001
      Gender (female)1.44 (1.32–1.57)<.0011.36 (1.23–1.51)<.001
      Smoking1.46 (1.31–1.63)<.0011.40 (1.21–1.62)<.0011.52 (1.35–1.71)<.0011.53 (1.31–1.80)<.001
      Body mass index >301.54 (1.39–1.70)<.0011.34 (1.18–1.53)<.0011.53 (1.36–1.71)<.0011.33 (1.15–1.54)<.001
      ASA grade >2
      ASA = American Society of Anesthesiologists classification (1–5).
      2.05 (1.85–2.28)<.0011.34 (1.16–1.54)2.14 (1.91–2.40)<.0011.39 (1.19–1.62)<.001
      Education level below college1.99 (1.79–2.21)<.0011.54 (1.35–1.75)<.0011.95 (1.72–2.21)<.0011.51 (1.29–1.76)<.001
      Civil status, living alone1.62 (1.46–1.78)<.0011.33 (1.17–1.52)<.0011.52 (1.37–1.71)<.0011.26 (1.09–1.45).002
      Not Native Norw speakers1.58 (1.26–2.00)<.0011.66 (1.23–2.23).0011.49 (1.16–1.92)<.001
      Disability benefit (all types)
      All types of disability benefit, both full and partly supported.
      1.46 (1.33–1.60)<.0011.67 (1.44–1.94)<.0011.47 (1.32–1.63)<.0011.66 (1.40–1.98)<.001
      Former lumbar spine surgery (any)2.26 (2.05–2.50)<.0012.21 (1.94–2.51)<.0012.19 (1.96–2.44)<.0012.00 (1.74–2.30)<.001
      MRI central stenosis1.05 (0.95–1.15).3581.05 (0.94–1.17).428
      MRI lateral stenosis0.91 (0.83–1.00).0400.90 (0.81–1.00).044
      MRI foraminal stenosis1.18 (1.02–1.36).0241.14 (0.97–1.34).120
      RF degen olisthesis0.85 (0.75–0.97).0130.76 (0.64–0.89).0010.92 (0.80–1.06).255
      Pre opr ODI (cont)
      ODI = Oswestry Disability Index (0–100), indicating increasing disability.
      1.06 (1.06–1.07)<.0011.06 (1.05–1.06)<.0011.07 (1.07–10.7)<.0011.07 (1.06–1.07)<.001
      Duration leg pain >12 months1.68 (1.52–1.86)<.0011.74 (1.55–1.96)<.0011.29 (1.06–1.56).010
      Duration backpain >12months1.87 (1.68–2.10)<.0012.17 (1.88–2.50)<.0011.95 (1.70–2.24)<.0011.85 (1.47–2.32)<.001
      Multilevel surgery
      more than one level operated.
      1.21 (1.11–1.33)<.0011.19 (1.07–1.32).001
      low asterisk ASA = American Society of Anesthesiologists classification (1–5).
      All types of disability benefit, both full and partly supported.
      ODI = Oswestry Disability Index (0–100), indicating increasing disability.
      § more than one level operated.

      Failure

      The strongest independent risk factors for failure identified in the multivariate model were duration of back pain >12 months (OR=2.24 [1.93–2.60]; p<.001), former spinal surgery (OR=2.21 [1.94–2.52]; p<.001) and age >70 years (OR=1.97 [1.69–2.30]; p<.001). Socioeconomic variables, that is, receiving disability benefits, low educational level, not being a native Norwegian speaker, and living alone, all increased the odds of failure (OR between 1.36–1.62). Variables concerning general health, that is, smoking, BMI >30, and ASA>2, also increased the odds of failure (OR 1.32–1.40). The spine-related disability (ODI) and pain medication increased the odds (OR 1.06–1.29). Of the radiological variables, only the finding of degenerative olisthesis showed an effect on the odds for failure with decreased odds (OR=0.75).

      Worsening

      The strongest independent risk factors for worsening identified in the multivariate model were former spinal surgery (OR=2.04 [1.77–2.36]; p<.001), duration of back pain >12 months (OR=1.83 [1.45–2.32]; p<.001), and age >70 years (OR=1.79 (1.49–2.14); p<.001). Socioeconomic variables, that is, receiving a disability benefit, low educational level, and living alone, increased the odds of worsening (OR between 1.33 and 1.67). Variables concerning general health, that is, as BMI >30 and ASA >2 increased the odds for worsening (OR 1.28–1.38), and spine-related disability (ODI) and duration of leg pain > 12 mths increased the odds for worsening (OR 1.07–1.30). None of the preoperative radiological variables influenced the odds of worsening, except the finding of a degenerative olisthesis, which decreased the odds of failure (OR=0.76 [0.64–0.89]; p<.001)).

      Sensitivity analyses

      Appendix Table 2 displays the multiple regression using ODI change score to define failure and worsening; there were minor differences from the primary analysis.

      Subgroup analyses

      Predominant preoperative back pain was reported by 1968 patients, of which 307 (16%) received decompression and fusion and 1,661 (84%) received decompression only. In the decompression and fusion group, 125 (41%) reported failure compared to 581 (35%) in the decompression only group. In the predominant leg pain group, 292 (36%) reported failure in the decompression and fusion group versus 1,921 (32%) in the decompression only group. Patients with predominant back pain had an increased risk ((RR) of 1.11 (1.04–1.19); p=.002) of failure.
      Appendix table 3 displays failure rates according to the number of levels operated. The proportion of patients that reported treatment failure increased by numbers of spinal levels operated. There were 48% failures reported by patients who had previously received surgery at the same spinal level, compared to 47% for those who were previously operated at another spinal level.

      Discussion

      In this register study, 33% of patients operated for lumbar spinal stenosis, were classified as failure after surgery, including 22% classified as worse. The strongest predictors for failure were preoperative duration of back pain for at least 12 months, previous spinal surgery, and age above 70 years. Both socioeconomic variables, general health variables, and spine-related variables affected the odds for failure. The same patterns were seen regarding the odds for worsening.
      The proportion of patients reported as failure and worsening seemed relatively high and may be partially explained by different outcome measures. For instance, the proportion of patients that perceived themselves as unchanged or worse was lower when patients used GPE, rather than ODI, to assess the effect of surgery. Similar results are reported in the literature. This is not surprising as GPE is conceptually different from a disease specific outcome measure. Previous studies reported success rates of about 62%–75% or failure rates of 25%–31% [
      • Nerland US
      • Jakola AS
      • Solheim O
      • Weber C
      • Rao V
      • Lønne G
      • et al.
      Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
      ,
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ,
      • Sigmundsson Freyr G
      • Kang Xiao P
      • Jönsson Bo
      • Strömqvist Björn
      Prognostic factors in lumbar spinal stenosis surgery.
      ,
      • Ogura Y
      • Kobayashi Y
      • Shinozaki Y
      • et al.
      Factors influencing patient satisfaction after decompression surgery without fusion for lumbar spinal stenosis.
      ,
      • Gunzburg R
      • Keller T
      • Szpalski M
      • et al.
      Clinical and psychofunctional measures of conservative decompression surgery for lumbar spinal stenosis: a prospective cohort study.
      ,
      • Kleinstück FS
      • Grob D
      • Lattig F
      • Bartanusz V
      • Porchet F
      • Jeszenszky D
      • et al.
      The influence of preoperative back pain on the outcome of lumbar decompression surgery.
      ,
      • Hébert JJ
      • Abraham E
      • Wedderkopp N
      • et al.
      Preoperative factors predict postoperative trajectories of pain and disability following surgery for degenerative lumbar spinal stenosis.
      ,
      • McIlroy S
      • Jadhakhan F
      • Bell D
      • Rushton A
      Prediction of walking ability following posterior decompression for lumbar spinal stenosis.
      ]. Moreover, the effect of surgery in our study with a mean ODI final score of 24, and a mean ODI improvement of 16 points is also in line with other studies [
      • Nerland US
      • Jakola AS
      • Solheim O
      • Weber C
      • Rao V
      • Lønne G
      • et al.
      Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
      ,
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ].

      Socioeconomics

      A short education, living alone, not being a native Norwegian speaker, and receiving disability benefits increased the odds of failure. The findings of associations between socioeconomic factors and odds for failure and worsening are known from the literature [
      • Yap ZL
      • Summers SJ
      • Grant AR
      • Moseley GL
      • Karran EL.
      The role of the social determinants of health in outcomes of surgery for low back pain: a systematic review and narrative synthesis.
      ]. The effect of these factors in our study was moderate (ORs between 1.33 and 1.67). One former study reported socioeconomic factors as more important than factors related to spine surgery and general health regarding return to work after spine surgery [
      • Asher A.L.
      • Devin C.J.
      • Archer K.R.
      • Chotai S.
      • Parker S.L.
      • Bydon M.
      • et al.
      An analysis from the quality outcomes database, part 2. predictive model for return to work after elective surgery for lumbar degenerative disease.
      ]. The impact of socioeconomics on the results of surgical treatment may seem surprising, but pain and disability are subjective feelings and functions. Hence, they may be affected by patient-related factors. Furthermore, communication is crucial in deciding on surgical treatment for a condition of pain and during clinical follow-up. Consequently, socioeconomic factors may impact the shared decision-making process between patients and health care personnel [
      • Stewart MA.
      Effective physician-patient communication and health outcomes: a review.
      ].
      Our findings of associations between socioeconomic factors and failure and worsening may contribute to a higher threshold to receive surgical treatment for some patients with socioeconomic challenges. Nevertheless, it is essential to consider equal rights to health care for all patients.

      General health

      Age >70 years was associated with almost doubled odds for failure and worsening (ORs between 1.93 and 1.99). However, the literature on the effect of age on clinical results after surgery for LSS is conflicting. Some studies find no or minimal association between age and clinical results [
      • Sobottke R
      • Herren C
      • Siewe J
      • Mannion AF
      • Röder C
      • Aghayev E.
      Predictors of improvement in quality of life and pain relief in lumbar spinal stenosis relative to patient age: a study based on the Spine Tango registry.
      ,
      • Claus CF
      • Tong D
      • Lytle E
      • Bahoura M
      • Garmo L
      • Li C
      • et al.
      Age as a predictor for complications and patient-reported outcomes in multilevel transforaminal lumbar interbody fusions: analyses from the michigan spine surgery improvement collaborative (MSSIC).
      ,
      • Nerland US
      • Jakola AS
      • Giannadakis C
      • Solheim O
      • Weber C
      • Nygaard ØP
      • et al.
      The risk of getting worse: predictors of deterioration after decompressive surgery for lumbar spinal stenosis: a multicenter observational study.
      ,
      • Giannadakis C
      • Solheim O
      • Jakola AS
      • Nordseth T
      • Gulati AM
      • Nerland US
      • et al.
      Surgery for lumbar spinal stenosis in individuals aged 80 and older: a multicenter observational study.
      ]. In contrast, others find decreased risk for success with increased age, or age <75 as a predictor of satisfaction [
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ,
      • Fritzell P
      • Mesterton J
      • Hagg O.
      Prediction of outcome after spinal surgery-using the dialogue support based on the Swedish national register.
      ]. Possible reasons for conflicting results include different outcomes, different ways of defining age groups (continuous data, age groups, or specific cut-offs), or differences in study populations. In addition, high age may be correlated with increased prevalence of other illnesses (ie, osteoarthritis), contributing to the association between age >70 years and increased odds for failure.
      Smoking, ASA>2, and BMI>30 also showed associations with failure and worsening (ORs between 1.33 and 1.53). Other studies have found ASA>2 more likely to have poor outcomes [
      • Hébert JJ
      • Abraham E
      • Wedderkopp N
      • et al.
      Preoperative factors predict postoperative trajectories of pain and disability following surgery for degenerative lumbar spinal stenosis.
      ]. The effect of BMI on the results is more uncertain in the literature. Mauro et al. reported worse outcomes with high BMI, while Onyekwelu et al. reported similar results for patients with BMI > 30 and BMI <30 [
      • Onyekwelu I
      • Glassman SD
      • Asher A
      • Shaffrey CI
      • Mummaneni PV
      • Carreon LY.
      Impact of obsiety on complications and outcomes: a comparison of fusion and nonfusion lumbar spine surgery.
      ,
      • Radcliff K.E.
      • Rihn J.
      • Hilibrand A.
      • DiIorio T.
      • Tosteson T.
      • Lurie J.D.
      • et al.
      Does the duration of symptoms in patients with spinal stenosis and degenerative spondylolisthesis affect outcomes?: analysis of the Spine Outcomes Research Trial.
      ].
      Using the final ODI score as the outcome, it seems natural that variables concerning general health (smoking, ASA>2, BMI>30) affect the outcome as they presumably reduce function.
      Some studies report frailty as a composite variable on general health, and frailty has shown an apparent effect on clinical results and complications after spine surgery and surgery in general [
      • Panayi AC
      • Orkaby AR
      • Sakthivel D
      • Endo Y
      • Varon D
      • Roh D
      • et al.
      Impact of frailty on outcomes in surgical patients: a systematic review and meta-analysis.
      ,
      • Veronesi F
      • Borsari V
      • Martini L
      • Visani A
      • Gasbarrini A
      • Brodano GB
      • et al.
      The impact of frailty on spine surgery: systematic review on 10 years clinical studies.
      ]. We believe the general health condition impacts the clinical outcome and disability after surgery for LSS. However, grading and recording this can be done in different ways. Hence, detecting it can be challenging, especially in a registry setting.

      Disease-related factors

      Duration of symptoms of >12 months strongly predicted both failure and worsening. Still, long-lasting back pain had a more negative impact than prolonged leg pain. In the multivariate analyses, leg pain > 12 months was either not detectable or moderated. We found an increased risk for failure for patients with preoperative predominant back pain. Surgery for LSS aims to increase the cross-sectional area of the spinal canal to relieve leg pain and improve walking capability. Patients who reporting predominately back prior to surgery could therefore be expected to benefit less of surgery. Previous studies have shown better outcomes among patients with short symptom duration [
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ,
      • ChaoT PK
      • Steward O
      • Gupta R
      Chronic nerve compression injury induces a phenotypic switch of neurons within the dorsal root ganglia.
      ]. One possible explanation is that prolonged symptoms may lead to biochemical differences in the nerve cells and chronic pain, hence poorer treatment effect [
      • Gupta R
      • Rowshan K
      • Chao T
      • Mozaffar T
      • Steward O.
      Chronic nerve compression induces local demyeliniation and remyelination in a rat model of carpat tunnel syndrome.
      ,
      • Fauzi A
      • Gatham R
      • Phedy P
      • Ajiantoro A
      • Gatam L
      • Mahadipta H
      • et al.
      Multivariate analysis of factors that influence the clinical outcome oswestry disability index in patients with lumbar spinal stenosis post percutaneous stenoscopic lumbar decompression.
      ].
      Former spinal surgery was a significant predictor of failure and worsening, doubling the odds for these outcomes (ORs 2.00–2.21), dichotomized irrespective of level, that is, including surgery at the same or another segment. In the subgroup analysis, we analyzed previous surgery at the same spinal level versus previous surgery at another spinal level. Interestingly, we found no differences in failure rates between these subgroups. Nerland et al. reported similar ORs for worsening associated with previous surgery in the same segment [36?]. Furthermore, Aalto et al. reported an association between no previous surgery and increased odds for good results (OR=3.65) [
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ]. Possible explanations for this association can be that previous surgery may lead to scar tissue resulting in technical difficulties in surgery. In addition, patients undergoing repeated surgery may be nonresponsive to surgical treatment, often achieving failure and worsening.
      Preoperative ODI showed increased OR for failure and worsening. The effect may seem small (OR=1.06–1.07 per ODI point). However, marked differences in preoperative ODI will significantly affect the odds of failure and worsening. Association between preoperative ODI scores and clinical outcomes have been reported before [
      • Sigmundsson Freyr G
      • Kang Xiao P
      • Jönsson Bo
      • Strömqvist Björn
      Prognostic factors in lumbar spinal stenosis surgery.
      ,
      • Park D.K.
      • An H.S.
      • Lurie J.D.
      • Zhao W.
      • Tosteson A.
      • Tosteson T.D.
      • et al.
      Does multilevel lumbar stenosis lead to poorer outcomes?: a subanalysis of the Spine Patient Outcomes Research Trial (SPORT) lumbar stenosis study.
      ]. As we define failure and worsening by final ODI scores, the preoperative ODI score seems as a natural predictive factor; it is less likely to achieve a postoperative cut-off with a higher preoperative disability.
      Patients who had surgery in more than one level had slightly increased odds for failure and worsening (OR = 1.19–1.21), although not significant in the multivariate analyses. The proportion of patients reporting treatment failure increased by number of spinal levels operated. Two former studies reported no statistically significant differences in outcomes for one and multilevel LSS treated surgically [
      • Nerland US
      • Jakola AS
      • Giannadakis C
      • Solheim O
      • Weber C
      • Nygaard ØP
      • et al.
      The risk of getting worse: predictors of deterioration after decompressive surgery for lumbar spinal stenosis: a multicenter observational study.
      ,
      • Aaen J
      • Austevoll IM
      • Hellum C
      • Storheim K
      • Myklebust TÅ
      • Banitalebi H
      • et al.
      Clinical and MRI findings in lumbar spinal stenosis: baseline data from the NORDSTEN study.
      ]. If one level operation has a specific risk for failure, adding the chance for failure per level could be a reasonable way to estimate the risk for failure in multilevel surgery. However, our study did not support such findings.
      In our study, radiological findings showed no or negligible associations with failure or worsening. That is in line with previous studies, showing no clinically relevant association between MRI findings and preoperative disability and no or minor association between MRI findings and clinical outcome [
      • Sigmundsson Freyr G
      • Kang Xiao P
      • Jönsson Bo
      • Strömqvist Björn
      Prognostic factors in lumbar spinal stenosis surgery.
      ,
      • Weber C
      • Giannadakis C
      • Rao V
      • Jakola A
      • Nerland U
      • Nygaard Ø
      • et al.
      Is there an association between radiological severity of lumbar spinal stenosis and disability, pain, or surgical outcome?.
      ]. Radiologic findings were only recorded as yes /no, and no grading of the radiologic findings was recorded. The validity of radiological data in NORspine has not been reported.
      When we examined failure and worsening as defined by ODI change score in the sensitivity analysis, we found that previous surgery, preoperative back pain lasting longer than 12 months, and age above 70 years were the strongest predictors of failure and worsening (Appendix Table 2)
      In patients with predominant back pain, fusion, in addition to decompression, did not improve the results.

      Future perspectives

      There is a need for better instruments predicting outcomes after surgery for LSS. Prediction models have been developed to assist in patient selection. However, Staartjes et al. reported only a moderate ability to identify patients likely to benefit from surgery for degenerative spine disorder. Therefore, they concluded that prediction models should only play a minor role in decision-making [
      • Staartjes V.E.
      • Stumpo V.
      • Ricciardi L.
      • Maldamer N
      • Eversdijk HAJ
      • Vieli M
      • et al.
      FUSE-ML: development and external validation of a clinical prediction model for mid-term outcomes after lumbar spinal fusion for degenerative disease.
      ]. The Swedish spine registry (Swespine) has also developed a prediction tool based on a prediction model to aid in patient selection for spine surgery [
      • Onyekwelu I
      • Glassman SD
      • Asher A
      • Shaffrey CI
      • Mummaneni PV
      • Carreon LY.
      Impact of obsiety on complications and outcomes: a comparison of fusion and nonfusion lumbar spine surgery.
      ]. From a future perspective, it could be interesting to develop a parallel prediction model based on NORspine data. The prediction model might help select suitable patients for surgery.

      Limitations

      There are several limits to this study. One is whether the NORspine registry records relevant variables to predict clinical outcomes. The study design does not allow conclusions regarding causality; only associations can be discovered. Some associations might be confounders or mediators connected to causal variables left unobserved. For instance, the NORspine did not record data on spinal alignment during the study period.
      Our study had a loss to follow-up of 24.9%; Appendix Table 1 and Table 2 display no significant differences between responders and nonresponders in baseline data or surgical treatment. Although according to former studies, loss to follow-up in national spine registers does not affect the clinical outcome, one of these studies examined the NORspine population [
      • Endler P.
      • Ekman P.
      • Hellström F.
      • Mõller H
      • Gerdhem P
      • et al.
      Minor effect of loss to follow-up on outcome interpretation in the Swedish spine register.
      ,
      • Højmark K
      • Støttrup C
      • Carreon L
      • Andersen MJESJ.
      Patient-reported outcome measures unbiased by loss of follow-up. Single-Center study based on Dane Spine.
      ,
      • Solberg TK
      • Sørlie A
      • Sjaavik K
      • Nygaard ØP
      • Ingebrigtsen T.
      Would loss to follow-up bias the outcome evaluation of patients operated for degenerative disorders of the lumbar spine?.
      ].
      Appendix table 1Patient characteristics of 11,873 Norwegian patients with surgically treated lumbar spinal stenosis in a 10 years period (2007-2017)
      Completed 12 months follow up (n=8919) Mean (95%CI), or n (%)Lost to follow up (n=2954) Mean (95%CI), or n (%)
      Age (cont)66.6 (66.4–66.9)63.2 (62.7–63.6)
      Female sex4,644 (52.1%)1,560 (52.8%)
      Civil status, living alone2,282 (25.7%)887 (30.2%)
      Native Norwegian speaking8,565 (96.5%)2,788 (94.7%)
      ASA grade >2
      ASA = American Society of Anesthesiologists classification (1–5).
      1,840 (20.8%)622 (21.3%)
      Body Mass Index (cont)27.5 (27.4–27.6)27.8 (27.6–28.0)
      Smoking1,697 (19.2%)2,100 (28.1%)
      Education level below college6,145 (70.1%)2,064 (71.4%)
      Comorbidity, any5,410 (66.9%)1,833 (68.0%)
      Patient not working7,443 (86.0%)2,425 (85.1%)
      Receives Disability benefit (uføret)1,421 (15.9%)511 (17.9%)
      Previous spinal surgery, any level2,173 (24.7%)795 (27.3%)
      Leg pain > 12 months duration5,284 (64.3%)1,831 (67.5%)
      Back pain > 12 months duration6,280 (74.9%)2,135 (76.8%)
      Preoperative ODI
      ASA = American Society of Anesthesiologists classification (1–5).
      39.8 (39.4–40.142.1 (41.6–42.7)
      Presoperative NRS leg pain
      ODI = Oswestry Disability Index (0–100).
      6.57 (6.52–6.61)6.7 (6.6–6.8)
      Pre operative NRS back pain6.50 (6.45–6.55)6.6 (6.5–6.7)
      Pre operative EQ-5D
      NRS = Numeric Rating Scale (0–10) ⁎⁎⁎⁎ED-5D = EuroQol`s quality of life, (-0.60–1.00)
      0.376 (0.369–0.383)0.323 (0.311–0.335)
      Patients completed 12 months follow-up, and patients lost to follow-up.
      low asterisk ASA = American Society of Anesthesiologists classification (1–5).
      ODI = Oswestry Disability Index (0–100).
      NRS = Numeric Rating Scale (0–10)⁎⁎⁎⁎ED-5D = EuroQol`s quality of life, (-0.60–1.00)
      Appendix table 2Sensitivity analysis, multivariable logistic regression for 8,919 patients operated for lumbar spinal stenosis and registered in NORspine during 2007–2017, using failure (ODI change <8 points) and worsening (ODI change <4 points) as dependent variables and potential predictors as explanatory variables
      Failure (ODI change <8 points)Worsening (ODI change <4 points)
      OR95%CIpOR95%CIp
      Age >701.681.47–1.93<.0011.611.39–1.87<.001
      Smoking1.391.22–1.60<.0011.491.29–1.72<.001
      BMI>301.251.11–1.42<.001
      ASA>21.371.19–1.56<.0011.491.22–1.63<.001
      Education level below college1.451.28–1.63<.0011.531.34–1.75<.001
      Civil status, living alone1.261.11–1.42<.0011.261.20–1-44<.001
      Disability benefit (all types)1.491.30–1.71<.0011.501.29–1.75<.001
      Previous surgery,any1,901.68–2.14<.0012.011.77–2.29<.001
      MRI: central stenosis0.790.71–0.89<.0010.810.71–0.92<.001
      Degenerative olisthesis (x-ray)0.690.59–0.81<.0010.700.59–0.84<.001
      Preoprerative ODI (cont)0.960.96–0.97<.0010.960.96–0.97<.001
      Back pain >12 mnts1.891.65–2.15<.0011.711.06–1.49<.001
      Leg pain >12 mhts1.261.06–1.49.008
      Appendix table 3Cross tabulation of number of operated levels and rates of treatment failure defined by ODI final score >31
      Number of levels operatedFailureNonsssfailureTotalProportion failure
      11,7473,7965,54332%
      29411,7402,68135%
      319930150039%
      423355839%
      One can discuss the choice of the cut-offs for failure and worsening. The cut-offs were assessed in a former study on LSS patients using a transitional scale as an anchor and are in concordance with another survey of cut-offs for success and with the PASS score of 22 proposed by van Hooff [
      • Alhaug OK
      • Dolatowski FC
      • Solberg TK
      • Lønne G.
      Criteria for failure and worsening after surgery for lumbar spinal stenosis: a prospective national registry observational study.
      ,
      • Austevoll I.M.
      • Gjestad R.
      • Grotle M.
      • Solberg T
      • Brox JI
      • Hermansen E
      • et al.
      Follow-up score, change score or percentage change score for determining clinical important outcome following surgery? An observational study from the Norwegian registry for Spine surgery evaluating patient reported outcome measures in lumbar spinal stenosis and lumbar degenerative spondylolisthesis.
      ,
      • van Hoof ML
      • Mannion AM
      • Staub LP
      • Osteolo RWJG
      • Fairbank JCT.
      Determination of the oswestry disability index score equivalent to a “satisfactory symptom state” in patients undergoing surgery for degenerative disorders of the lumbar spine – a Spine Tango registry based study.
      ]. Our cut-offs result in proportions classified as failure and worsening, similar to other studies [
      • Nerland US
      • Jakola AS
      • Solheim O
      • Weber C
      • Rao V
      • Lønne G
      • et al.
      Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
      ,
      • Aalto T
      • Sinikallio S
      • Kröger H
      • et al.
      Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
      ,
      • Sigmundsson Freyr G
      • Kang Xiao P
      • Jönsson Bo
      • Strömqvist Björn
      Prognostic factors in lumbar spinal stenosis surgery.
      ,
      • Ogura Y
      • Kobayashi Y
      • Shinozaki Y
      • et al.
      Factors influencing patient satisfaction after decompression surgery without fusion for lumbar spinal stenosis.
      ,
      • Gunzburg R
      • Keller T
      • Szpalski M
      • et al.
      Clinical and psychofunctional measures of conservative decompression surgery for lumbar spinal stenosis: a prospective cohort study.
      ,
      • Kleinstück FS
      • Grob D
      • Lattig F
      • Bartanusz V
      • Porchet F
      • Jeszenszky D
      • et al.
      The influence of preoperative back pain on the outcome of lumbar decompression surgery.
      ,
      • Hébert JJ
      • Abraham E
      • Wedderkopp N
      • et al.
      Preoperative factors predict postoperative trajectories of pain and disability following surgery for degenerative lumbar spinal stenosis.
      ,
      • McIlroy S
      • Jadhakhan F
      • Bell D
      • Rushton A
      Prediction of walking ability following posterior decompression for lumbar spinal stenosis.
      ]. Failure and worsening have been defined differently in other studies, but the use of MCID regarding increasing ODI score and worsening is not well supported [
      • Nerland US
      • Jakola AS
      • Giannadakis C
      • Solheim O
      • Weber C
      • Nygaard ØP
      • et al.
      The risk of getting worse: predictors of deterioration after decompressive surgery for lumbar spinal stenosis: a multicenter observational study.
      ]. In the sensitivity analysis, our main findings were confirmed. There are, however, no explicit definitions of failure and worsening after spinal surgery.
      Twelve months follow-up might be short for a chronic illness. On the other hand, several studies show no clinically significant differences between 12 and 24 months. We, therefore, consider 12 months as sufficient in LSS patients [
      • Endler P.
      • Ekman P.
      • Hellström F.
      • Mõller H
      • Gerdhem P
      • et al.
      Minor effect of loss to follow-up on outcome interpretation in the Swedish spine register.
      ,
      • Weinstein JN
      • Tosteson TD
      • Lurie JD
      • Tosteson A
      • Blood E
      • Herkowitz H
      • et al.
      Surgical versus nonoperative treatment for lumbar spinal stenosis four-year results of the Spine Patient Outcomes Research Trial.
      ,
      • Parai C
      • Hägg O
      • Lind B
      • Brisby H.
      Follow-up of degenerative lumbar spine surgery-PROMs stabilize after 1 year: an equivalence study based on Swespine data.
      ,
      • Weinstein JN
      • Lurie JD
      • Tosteson TD
      • Zhao W
      • Blood A
      • Tosteson ANA
      • et al.
      Surgical compared with nonoperative treatment for lumbar degenerative spondylolisthesis. four-year results in the Spine Patient Outcomes Research Trial (SPORT) randomized and observational cohorts.
      ,
      • Lønne G
      • Johnsen LG
      • Rossvoll I
      • Andresen H
      • Storheim K
      • Zwart JA
      • et al.
      Minimally invasive decompression versus x-stop in lumbar spinal stenosis: a randomized controlled multicenter study.
      ]. Twelve months follow-up is also recommended in a systematic review with recommendations for spine registries in 2015 [
      • van hoof ML
      • Jacobs WC
      • Willems PC
      • Wouters MWJM
      • de Kleuver M
      • Peul WC
      • et al.
      Evidence and practice in spine registries . A systematic review, and recommendations for future design of registries.
      ].
      Different surgical techniques were used in this study; both decompression methods and fusion methods varied. Naturally, this introduces heterogeneity in our material, but on the other hand, it reflects the everyday practice and increases the external validity.
      Different findings between predictor studies can result from differences in patient selection, surgical techniques, and the selection and recording of possible predictive variables.

      Strengths

      The study population is large and recruited in a national register obligatory to all treating centers in Norway. Therefore, the patient population reflects everyday practice, and we consider the external validity good. The large sample size also allowed for strict thresholds for entry and removal of covariates into the model of 0.01, improving the power of the analyses.
      Interpreting changes in PROM scores, that is, ODI, in groups is not straightforward, and especially in large datasets, one can find statistically significant findings that do not reflect clinical importance. Therefore, we chose to use dichotomous outcomes to define failure and worsening, emphasized in a review article as favorable regarding clinical important change [
      • Katz NP
      • Paillard FC
      • Ekman E.
      Determining the clinical importance of treatment benefits for interventions for painful orthopedic conditions.
      ].

      Conclusion

      In this prospective observational spine register study, 33% of patients reported treatment failure, including a worsening rate of 22%, after surgery for LSS. Associated with increased odds for failure and worsening were duration of back pain of more than 12 months, former spinal surgery, and age >70 years. This information can assist in patient information and patient selection for surgery.

      Declaration of Competing Interest

      The authors declare that they have no known potential conflicts of interest influencing the work with this paper. None of the authors have received financial support to complete this study.

      Acknowledgments

      We would like to thank Milada Cvancarova Småstuen, Professor at Oslo Metropolitan University, for her invaluable help with the statistics in the study.

      References

        • Amundsen T*
        • Weber H
        • Nordal HJ
        • Magnaes B
        • Abdelnoor M
        • Lilleås F
        Lumbar spinal stenosis: conservative or surgical management?: a prospective 10-year study.
        Spine. 2000; 25: 1424-1436
        • Atlas SJ
        • Keller RB
        • Robson DBS
        • Deyo RA
        • Singer DE.
        Surgical and nonsurgical management of lumbar spinal stenosis: four-year outcomes from the maine lumbar spine study.
        Spine. 2000; 25: 556-562
        • Atlas SJ
        • Keller RB
        • Wu YA
        • Deyo RA
        • Singer DE.
        Long-term outcomes of surgical and nonsurgical management of lumbar spinal stenosis: 8 to 10 year results from the maine lumbar spine study.
        Spine. 2005; 30: 936-943https://doi.org/10.1097/01.brs.0000158953.57966.c0
        • Malmivaara A
        • Slätis P
        • Heliövaara M
        • Sainio P
        • Kinnunen H
        • Kankare J
        • et al.
        Surgical or nonoperative treatment for lumbar spinal stenosis?: a randomized controlled trial.
        Spine. 2007; 32: 1-8https://doi.org/10.1097/01.brs.0000251014.81875.6d
        • Weinstein JN
        • Tosteson TD
        • Lurie JD
        • Tosteson ANA
        • Blood E
        • Hanscom B
        • et al.
        Surgical versus nonsurgical therapy for lumbar spinal stenosis.
        N Engl J Med. 2008; 358: 794-810https://doi.org/10.1056/NEJMoa0707136
        • Kreiner DS
        • Shaffer WO
        • Baisden JL
        • Gilbert TJ
        • Summers JT
        • Toton JF
        • et al.
        An evidence-based clinical guideline for the diagnosis and treatment of degenerative lumbar spinal stenosis (update).
        Spine J. 2013; 13: 734-743
        • Nerland US
        • Jakola AS
        • Solheim O
        • Weber C
        • Rao V
        • Lønne G
        • et al.
        Minimally invasive decompression versus open laminectomy for central stenosis of the lumbar spine: pragmatic comparative effectiveness study.
        BMJ. 2015; 350: h1603https://doi.org/10.1136/bmj.h1603
        • Aalto T
        • Sinikallio S
        • Kröger H
        • et al.
        Preoperative predictors for good postoperative satisfaction and functional outcome in lumbar spinal stenosis surgery — a prospective observational study with a two-year follow-up.
        Scandinavian J Surg. 2012; : 255-260https://doi.org/10.1177/145749691210100406
        • Sigmundsson Freyr G
        • Kang Xiao P
        • Jönsson Bo
        • Strömqvist Björn
        Prognostic factors in lumbar spinal stenosis surgery.
        Acta Orthopaedica. 2012; 83 (To link to this article: 10.3109/17453674.2012.73391): 536-542https://doi.org/10.3109/17453674.2012.733915
        • Ogura Y
        • Kobayashi Y
        • Shinozaki Y
        • et al.
        Factors influencing patient satisfaction after decompression surgery without fusion for lumbar spinal stenosis.
        Global Spine J. 2020; 10: 627-632https://doi.org/10.1177/2192568219868205
        • Gunzburg R
        • Keller T
        • Szpalski M
        • et al.
        Clinical and psychofunctional measures of conservative decompression surgery for lumbar spinal stenosis: a prospective cohort study.
        Eur Spine J. 2003; 12: 197-204https://doi.org/10.1007/s00586-002-0479-6
        • Kleinstück FS
        • Grob D
        • Lattig F
        • Bartanusz V
        • Porchet F
        • Jeszenszky D
        • et al.
        The influence of preoperative back pain on the outcome of lumbar decompression surgery.
        Spine. 2009; 34: 1198-1203https://doi.org/10.1097/BRS.0b013e31819fcf35
        • Hébert JJ
        • Abraham E
        • Wedderkopp N
        • et al.
        Preoperative factors predict postoperative trajectories of pain and disability following surgery for degenerative lumbar spinal stenosis.
        Spine (Phila Pa 1976). 2020; 45: E1421-E1430https://doi.org/10.1097/BRS.0000000000003587
        • Simpson AK
        • Lightsey IV
        • >Xiong GX
        • Crawford AM
        • Minamide A
        • Schoenfeld AJ
        Spinal endoscopy: evidence, techniques, global trends, and future projections.
        Spine J. 2022; 22: 64-74
        • Sacks GD
        • Dawes AJ
        • Ettner SL
        • Brook RH
        • Fox CR
        • Maggard-Gibbons M
        • et al.
        Surgeon perception of and benefit in the decision to operate.
        Ann Surg. 2016; : 896-903https://doi.org/10.1097/SLA.0000000000001784
        • Kahneman D
        • Sibony O
        • Sunstein CR.
        Noise: a flaw in human judgment.
        London William Collins, 2021
        • Meehl PE.
        Clinical versus statistical prediction: a theoretical analysis and a review of the evidence.
        University of Minnesota Press, 1954https://doi.org/10.1037/11281-000
        • Grove WM
        • Zald DH
        • Lebow BS
        • Snitz BE.
        • Nelson C.
        Clinical versus mechanical prediction: a meta-analysis.
        Psycholo Assess. 2000; 12: 19-30https://doi.org/10.1037/1040-3590.12.1.19
        • Yu MC.
        • Kuncel NR.
        Pushing the limits for judgmental consistency: comparing random weighting schemes with expert judgments.
        Personnel Assess Decisions. 2020; Number 6 (Article 2. DOI. Available at:)https://doi.org/10.25035/pad.2020.02.002
        • Yap ZL
        • Summers SJ
        • Grant AR
        • Moseley GL
        • Karran EL.
        The role of the social determinants of health in outcomes of surgery for low back pain: a systematic review and narrative synthesis.
        Spine J. 2022; 22: 793-809https://doi.org/10.1016/j.spinee.2021.11.013
      1. Norwegian registry for Spine Surgery (NORspine) Annual report 2017. 2SKDE. Tore K Solberg, Lena Ringstad olsen Universtitetssykehuset I Nord Norge (UNN, 2018 (Available at:) (Accessed July 17, 2022)
        • Fairbank JC
        • Pynsent PB.
        The oswestry disability index.
        Spine. 2000; 25: 2940-2952
        • Grotle M
        • Brox JI
        • Vøllestad NK.
        Cross-cultural adaption of the Norwegian versions of the roland-morris disability questionnaire and the oswestry disability index.
        J Rehabil Med. 2003; 35: 241-247
        • McCaffery M.
        Pain, clinical manual for nursing practice.
        Mosby, 1989
        • Kamper SJ
        • Ostelo RWJG
        • Knol DL
        • Maher CG
        • de Vet HCW
        • Hancock MJ.
        Global Perceived Effect scales provided reliable assessments of health transition in people with musculoskeletal disorders, but ratings are strongly influenced by current status.
        J Clin Epidemiol. 2010; 63: 760-766https://doi.org/10.1016/j.jclinepi.2009.09.009
        • Alhaug OK
        • Dolatowski FC
        • Solberg TK
        • Lønne G.
        Criteria for failure and worsening after surgery for lumbar spinal stenosis: a prospective national registry observational study.
        Spine J. 2021; 21 (criteria DOI:): 1489-1496https://doi.org/10.1016/j.spinee.2021.04.008
        • ICHOM
        (International Consortium for Health Outcomes Measurements).
        Standard Set for Low Back Pain. 2014; (Available at:) (Accessed July 17, 2022. [Google Scholar])
        • van hoof ML
        • Jacobs WC
        • Willems PC
        • Wouters MWJM
        • de Kleuver M
        • Peul WC
        • et al.
        Evidence and practice in spine registries . A systematic review, and recommendations for future design of registries.
        Acta orthopaedic. 2015; 86: 534-544https://doi.org/10.3109/17453674.2015.1043174
        • van Hooff ML
        • van Loon J
        • van Limbeek J
        • de Kleuver M.
        The Nijmegen decision tool for chronic low back pain. Development of a clinical decision tool for secondary or tertiary spine care specialists.
        Plos one. 2014; https://doi.org/10.1371/journal.pone.0104226
        • von Elm E
        • Altman DG
        • Egger M
        • Pocock SJ
        • Gotzsche PC
        • Vandenbroucke JP.
        The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies.
        Ann Intern Med. 2007; 147 (PMID: 17938396): 573-577
        • McIlroy S
        • Jadhakhan F
        • Bell D
        • Rushton A
        Prediction of walking ability following posterior decompression for lumbar spinal stenosis.
        Eur Spine J. 2021; 30: 3307-3318https://doi.org/10.1007/s00586-021-06938-6
        • Asher A.L.
        • Devin C.J.
        • Archer K.R.
        • Chotai S.
        • Parker S.L.
        • Bydon M.
        • et al.
        An analysis from the quality outcomes database, part 2. predictive model for return to work after elective surgery for lumbar degenerative disease.
        J Neurosurg. 2017; 27 (Available at) (Accessed July 28, 2022): 370-381https://doi.org/10.3171/2016.8.SPINE16527
        • Stewart MA.
        Effective physician-patient communication and health outcomes: a review.
        CMAJ. 1995; 152 (PMID: 7728691): 1423-1433
        • Sobottke R
        • Herren C
        • Siewe J
        • Mannion AF
        • Röder C
        • Aghayev E.
        Predictors of improvement in quality of life and pain relief in lumbar spinal stenosis relative to patient age: a study based on the Spine Tango registry.
        Eur Spine J. 2017; 26 (Epub 2015 Jul 3. PMID: 26138216): 462-472https://doi.org/10.1007/s00586-015-4078-8
        • Claus CF
        • Tong D
        • Lytle E
        • Bahoura M
        • Garmo L
        • Li C
        • et al.
        Age as a predictor for complications and patient-reported outcomes in multilevel transforaminal lumbar interbody fusions: analyses from the michigan spine surgery improvement collaborative (MSSIC).
        Spine (Phila Pa 1976). 2021; 46 (15PMID: 33620179): 356-365https://doi.org/10.1097/BRS.0000000000003792
        • Nerland US
        • Jakola AS
        • Giannadakis C
        • Solheim O
        • Weber C
        • Nygaard ØP
        • et al.
        The risk of getting worse: predictors of deterioration after decompressive surgery for lumbar spinal stenosis: a multicenter observational study.
        World Nerosrug. 2015; 84: 1095-1102https://doi.org/10.1016/j.wneu.2015.05.055
        • Giannadakis C
        • Solheim O
        • Jakola AS
        • Nordseth T
        • Gulati AM
        • Nerland US
        • et al.
        Surgery for lumbar spinal stenosis in individuals aged 80 and older: a multicenter observational study.
        J Am Geriatr Soc. 2016; 64 (Epub 2016 Sep 9. PMID: 27611928): 2011-2018https://doi.org/10.1111/jgs.14311
        • Fritzell P
        • Mesterton J
        • Hagg O.
        Prediction of outcome after spinal surgery-using the dialogue support based on the Swedish national register.
        Eur Spine J. 2021; (Epub ahead of print. PMID: 34837113): 27https://doi.org/10.1007/s00586-021-07065-y
        • Onyekwelu I
        • Glassman SD
        • Asher A
        • Shaffrey CI
        • Mummaneni PV
        • Carreon LY.
        Impact of obsiety on complications and outcomes: a comparison of fusion and nonfusion lumbar spine surgery.
        J Neurosurg. 2017; 26: 158-162
        • Radcliff K.E.
        • Rihn J.
        • Hilibrand A.
        • DiIorio T.
        • Tosteson T.
        • Lurie J.D.
        • et al.
        Does the duration of symptoms in patients with spinal stenosis and degenerative spondylolisthesis affect outcomes?: analysis of the Spine Outcomes Research Trial.
        Spine. 2011; 36: 2197-2210https://doi.org/10.1097/BRS.0b013e3182341edf
        • Panayi AC
        • Orkaby AR
        • Sakthivel D
        • Endo Y
        • Varon D
        • Roh D
        • et al.
        Impact of frailty on outcomes in surgical patients: a systematic review and meta-analysis.
        Am J Surg. 2019; 218: 393-400https://doi.org/10.1016/j.amjsurg.2018.11.020
        • Veronesi F
        • Borsari V
        • Martini L
        • Visani A
        • Gasbarrini A
        • Brodano GB
        • et al.
        The impact of frailty on spine surgery: systematic review on 10 years clinical studies.
        Aging Dis. 2021; 12 (Published 2021 Apr 1): 625-645https://doi.org/10.14336/AD.2020.0904
        • ChaoT PK
        • Steward O
        • Gupta R
        Chronic nerve compression injury induces a phenotypic switch of neurons within the dorsal root ganglia.
        J Comparat Neurol. 2008; 506: 180-193https://doi.org/10.1002/cne.21537
        • Gupta R
        • Rowshan K
        • Chao T
        • Mozaffar T
        • Steward O.
        Chronic nerve compression induces local demyeliniation and remyelination in a rat model of carpat tunnel syndrome.
        Experimental Neuro. 2004; 187: 500-508https://doi.org/10.1016/j.expneurol.2004.02.009
        • Fauzi A
        • Gatham R
        • Phedy P
        • Ajiantoro A
        • Gatam L
        • Mahadipta H
        • et al.
        Multivariate analysis of factors that influence the clinical outcome oswestry disability index in patients with lumbar spinal stenosis post percutaneous stenoscopic lumbar decompression.
        Biomed Pharmacol J. 2021; 14https://doi.org/10.13005/bpj/2125
        • Park D.K.
        • An H.S.
        • Lurie J.D.
        • Zhao W.
        • Tosteson A.
        • Tosteson T.D.
        • et al.
        Does multilevel lumbar stenosis lead to poorer outcomes?: a subanalysis of the Spine Patient Outcomes Research Trial (SPORT) lumbar stenosis study.
        Spine. 2010; 35: 439-446https://doi.org/10.1097/BRS.0b013e3181bdafb9
        • Aaen J
        • Austevoll IM
        • Hellum C
        • Storheim K
        • Myklebust TÅ
        • Banitalebi H
        • et al.
        Clinical and MRI findings in lumbar spinal stenosis: baseline data from the NORDSTEN study.
        Eur Spine J. 2022; 31: 1391-1398https://doi.org/10.1007/s00586-021-07051-4
        • Weber C
        • Giannadakis C
        • Rao V
        • Jakola A
        • Nerland U
        • Nygaard Ø
        • et al.
        Is there an association between radiological severity of lumbar spinal stenosis and disability, pain, or surgical outcome?.
        Spine. 2016; 41: E78-E83https://doi.org/10.1097/BRS.0000000000001166
        • Staartjes V.E.
        • Stumpo V.
        • Ricciardi L.
        • Maldamer N
        • Eversdijk HAJ
        • Vieli M
        • et al.
        FUSE-ML: development and external validation of a clinical prediction model for mid-term outcomes after lumbar spinal fusion for degenerative disease.
        Eur Spine J. 2022; https://doi.org/10.1007/s00586-022-07135-9
        • Endler P.
        • Ekman P.
        • Hellström F.
        • Mõller H
        • Gerdhem P
        • et al.
        Minor effect of loss to follow-up on outcome interpretation in the Swedish spine register.
        Eur Spine J. 2020; 29: 213-220https://doi.org/10.1007/s00586-019-06181-0
        • Højmark K
        • Støttrup C
        • Carreon L
        • Andersen MJESJ.
        Patient-reported outcome measures unbiased by loss of follow-up. Single-Center study based on Dane Spine.
        Danish Spine Surg Registry. 2016; 25: 282-286https://doi.org/10.1007/s00586-015-4127-3
        • Solberg TK
        • Sørlie A
        • Sjaavik K
        • Nygaard ØP
        • Ingebrigtsen T.
        Would loss to follow-up bias the outcome evaluation of patients operated for degenerative disorders of the lumbar spine?.
        Acta Orthopaedica. 2011; 82: 56-63https://doi.org/10.3109/17453674.2010.548024
        • Austevoll I.M.
        • Gjestad R.
        • Grotle M.
        • Solberg T
        • Brox JI
        • Hermansen E
        • et al.
        Follow-up score, change score or percentage change score for determining clinical important outcome following surgery? An observational study from the Norwegian registry for Spine surgery evaluating patient reported outcome measures in lumbar spinal stenosis and lumbar degenerative spondylolisthesis.
        BMC Musculoskelet Disord. 2019; 20: 31https://doi.org/10.1186/s12891-018-2386-y
        • van Hoof ML
        • Mannion AM
        • Staub LP
        • Osteolo RWJG
        • Fairbank JCT.
        Determination of the oswestry disability index score equivalent to a “satisfactory symptom state” in patients undergoing surgery for degenerative disorders of the lumbar spine – a Spine Tango registry based study.
        Spine J. 2016; 16: 1221-1230https://doi.org/10.1016/j.spinee.2016.06.010
        • Weinstein JN
        • Tosteson TD
        • Lurie JD
        • Tosteson A
        • Blood E
        • Herkowitz H
        • et al.
        Surgical versus nonoperative treatment for lumbar spinal stenosis four-year results of the Spine Patient Outcomes Research Trial.
        Spine (Phila Pa 1976). 2010; 35: 1329-1338https://doi.org/10.1097/BRS.0b013e3181e0f04d
        • Parai C
        • Hägg O
        • Lind B
        • Brisby H.
        Follow-up of degenerative lumbar spine surgery-PROMs stabilize after 1 year: an equivalence study based on Swespine data.
        Eur Spine J. 2019; 28: 2187-2197https://doi.org/10.1007/s00586-019-05989-0
        • Weinstein JN
        • Lurie JD
        • Tosteson TD
        • Zhao W
        • Blood A
        • Tosteson ANA
        • et al.
        Surgical compared with nonoperative treatment for lumbar degenerative spondylolisthesis. four-year results in the Spine Patient Outcomes Research Trial (SPORT) randomized and observational cohorts.
        J Bone Joint Surg Am. 2009; 91: 1295-1304https://doi.org/10.2106/JBJS.H.00913
        • Lønne G
        • Johnsen LG
        • Rossvoll I
        • Andresen H
        • Storheim K
        • Zwart JA
        • et al.
        Minimally invasive decompression versus x-stop in lumbar spinal stenosis: a randomized controlled multicenter study.
        Spine (Phila Pa 1976). 2015; 40: 77-85https://doi.org/10.1097/BRS.0000000000000691
        • Katz NP
        • Paillard FC
        • Ekman E.
        Determining the clinical importance of treatment benefits for interventions for painful orthopedic conditions.
        J Orthop Surg. 2015; 10: 24https://doi.org/10.1186/s13018-014-0144-x