A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis

Forsth NEJM 2016;374:1413-23. doi:10.1056/NEJMoa1513721

Clinical Question

• In patients with lumbar spinal stenosis at 1 or 2 adjacent vertebral levels, with or without spondylolisthesis, what is the efficacy of decompression surgery compared with decompression and fusion surgery, on functional outcome assessed 2 years post-operatively?

Design

• Multi-centred, randomised, controlled trial
• Blinding of outcome assessors only
• Randomisation was stratified to the presence or absence of spondylolisthesis (forward displacement of a vertebra)
• Allocation concealment achieved by a web-based randomisation sequence
• Pragmatic design such that the surgical technique was determined by the surgeons, who were highly experienced in performing the two trial interventions
• The initial sample size was calculated assuming that the ratio of those with spondylolisthesis and those without was 1:3. However, the proportion of patients with spondylolisthesis was higher which meant that the goal of including 40 patients in each of the 4 stratum was achieved earlier than predicted. This allowed an 80% power to detect a difference in the ODI (Oswestry Disability Index) of 12 points. So following recalculation the number included in each group was more than 40 and 247 patients in total were randomised
• Per-protocol analysis, but an intention to treat analysis also conducted

Setting

• 7 Swedish hospitals
• 247 patients were enrolled between October 2006 and June 2012

Population

• Inclusion: Patients aged 50-80 years; pseudoclaudication in one or both legs and back pain; 1 or 2 adjacent stenotic segments (cross-section area of dural sac < 75mm2); Duration of symptoms > 6 months
• Exclusion: Spondylolysis (a defect or stress fracture in the pars interarticularis of the vertebral arch); degenerative lumbar scoliosis; history of lumbar spinal surgery for stenosis or instability; stenosis not due to degenerative changes; stenosis caused by a herniated disk; other specific spinal conditions (e.g cancer, ankylosing spondylitis); history of vertebral fractures in affected segments; psychological disorders (e.g dementia, drug abuse)
• 247 patients

Intervention

• Fusion Group = decompression and fusion

Control

• Decompression alone group

Outcome

• Primary outcome: per protocol analysis of Oswestry Disability Index (ODI) at 2 years.
  • no difference between the groups in the mean ODI
    • 27 in the fusion group vs 24 in decompression alone group, P=0.24
  • no statistical difference in the change of ODI
    • decrease from baseline of 15 in the fusion group and 17 in the decompression alone group, P = 0.36
  • no significant difference in outcome between type of treatment and the presence or not of spondylolithesis

ODI is a standard scoring system which measures the degree of disability and quality of life in patients with lower back pain. This score ranges from 0-100, with higher scores indicating more severe disability

• Secondary outcomes: no statistical difference in any of the outcome measures between type of treatment and the presence or not of spondylolithesis
  • European Quality of Life – 5 Dimensions; a score from 0-1 with higher scores indicating better quality of life
  • Visual analogues scales for back pain and leg pain (0-100, with higher scores indicating more severe pain)
  • Zurich Claudication Questionnaire (scores 1-4, with higher scores indicating more severe disability)
  • Questions regarding overall satisfaction, global assessment of back and leg pain and walking ability
  • 6 minute walk test (the distance in metres a person is able to walk in 6 minutes) at 2 years
  • Complications: Dural tears occurred in 11% in fusion group and 11% in decompression-alone group. Post-operative infections requiring antibiotics occurred in 10% in fusion group, 4% in decompression-alone group. AMI, CVA, thromboembolic complications were 3% in fusion group, 4% in decompression alone group
  • Reoperations: 22% in fusion group, 21% in decompression-alone group
  • 5 year follow up reported in a proportion of the patients: data available for 153 patients (7 had died, 1 had a stroke, 1 severe dementia, 85 within 5 years at time of publishing); 138 provided outcomes with no difference between fusion group and decompression-alone (with or without spondylolithesis) group in any of the primary or secondary outcome measures
  • CT, MRI and lateral radiography at 2 years: radiology findings not reported in this paper
  • Modified intention to treat analysis; 5 of the 14 patients who did not receive their assigned therapy were lost to follow-up. Of the 9 who were included in this analysis, results were similar between fusion and decompression-alone groups

• Resource use: Fusion increased hospital costs by US$6800 (from US$5400 to $12200) and length of stay from 4.1 days to 7.4 days

Authors’ Conclusions

• In patients with lumbar spinal stenosis, with or without degenerative spondylolithesis there is no difference in clinical outcome at 2 years after surgery with fusion and decompression compared with decompression alone

Strengths

• Randomised, multi-centred
• Allocation concealment
• Pragmatic design reflecting real world surgical practices
• Revision of sample size without impacting on power of study
• Blinding of the outcome assessor
• Outcomes were clinically meaningful
• Per-protocol analysis correlates well with intention to treat analysis
• Minimal loss to follow-up (93% had per-protocol analysis)

Weaknesses

• The trial was not powered to analyse differences in complication rates, however, the authors draw attention to a large analysis of registry data suggesting a doubling of severe adverse events with fusion surgery. This corresponds with a number needed to harm of 30 patients.
• Stratification by hospital and surgeon would have ensured equal allocation of subgroups to each variant

The Bottom Line

• If I were a surgeon and a patient presented to be with these inclusion criteria, I would offer decompression alone
• Unfortunately, the costs and morbidity to patients who have fusion in addition to decompression seem to be excessive AND there is no difference in clinically meaningful outcomes at 2 and 5 years and a potential to harm

External Links

• [article] A Randomized, Controlled Trial of Fusion Surgery for Lumbar Spinal Stenosis
• [further reading] Editorial
• [further reading] Quick take video summary

Metadata

Summary author: Celia Bradford
Summary date: June 28 2016
Peer-review editor: Steve Mathieu