The Neurosurgery Signal — Signal, Not Noise

Rapid review note

Journal Club is a rapid, AI-assisted appraisal layer. It highlights study design, effect estimates, and practice relevance, but it is still a briefing, not a replacement for the paper.

For education only. Not medical advice.

Paper snapshot

Rapid study overview

Open paper

DOI

10.1056/NEJMoa0707136

PMID

18287602

PICO

Population

Surgical candidates with at least 12 weeks of symptoms and spinal stenosis without spondylolisthesis confirmed on imaging

Intervention

Decompressive surgery

Comparator

Usual nonsurgical care

Outcomes

SF-36 bodily pain and physical function scores, modified Oswestry Disability Index at multiple time points up to 2 years

Design

Type

Randomized controlled trial with concurrent observational cohort

Randomized

Yes

Multicenter

Yes

Blinded

Not applicable: patients and clinicians not blinded to treatment assignment

Follow-up

2 years

Primary endpoint

SF-36 bodily pain and physical function scores, modified Oswestry Disability Index

Secondary endpoints

Treatment adherence rates
As-treated outcomes analysis

Practice impact

What this means

This SPORT trial compared surgery versus nonsurgical care for lumbar spinal stenosis without spondylolisthesis. Despite high crossover (67% surgery, 43% nonsurgical groups), as-treated analysis showed surgery provided significantly greater improvement in pain and function by 3 months sustained to 2 years. Intention-to-treat analysis showed benefit only for bodily pain. Consider surgical referral for appropriate candidates after failed conservative management.

Bottom line

Surgery provides greater improvement in pain and function than nonsurgical care for spinal stenosis without spondylolisthesis, but high crossover rates complicate interpretation.

Strength of evidence

moderate

Recommendation

consider change

Why it matters

As-treated analysis shows consistent surgical benefit across all outcomes
Intention-to-treat analysis limited by high crossover
Largest randomized study on this population to date

What would change my mind

Longer-term follow-up showing sustained benefit or complications
Randomized trial with lower crossover rates
Cost-effectiveness analysis comparing treatments

Critical appraisal

How strong is the paper?

Methods critique

Risk of bias

High crossover: 67% of surgery group and 43% of nonsurgical group crossed over by 2 years, limiting intention-to-treat interpretation

Confounding

As-treated analysis adjusted for potential confounders but observational cohort introduces selection bias

Missing data

Not explicitly reported in excerpt; analysis appears complete for primary outcomes at 2 years

Multiplicity

Multiple primary outcomes assessed at multiple time points without explicit adjustment

Notes

Combined randomized and observational cohorts in as-treated analysis
High nonadherence to assigned treatment

Stats check

NNT

Not reported: no dichotomous outcomes with NNT calculation provided

Effect sizes

Mean difference in SF-36 bodily pain change: 7.8 (95% CI 1.5 to 14.1) favoring surgery
289 patients in randomized cohort, 365 in observational cohort

Absolute effects

67% of surgery group underwent surgery by 2 years
43% of nonsurgical group underwent surgery by 2 years

Concerns

Intention-to-treat analysis showed significant benefit only for bodily pain, not physical function or Oswestry
As-treated analysis showed significant benefit for all outcomes but is observational

External validity

Who it applies to

Surgical candidates with symptomatic lumbar spinal stenosis without spondylolisthesis at U.S. spine centers

Who it does not

Patients with degenerative spondylolisthesis, acute symptoms <12 weeks, or non-surgical candidates

Generalizability notes

13 U.S. spine clinics improve generalizability
Observational cohort reflects real-world treatment preferences

Evidence trace

Source trace and metadata

Citations (3)

claim_id

methods_critique.risk_of_bias

locator

p. 794 Results

quote

At 2 years, 67% of patients who were randomly assigned to surgery had undergone surgery, whereas 43% of those who were randomly assigned to receive nonsurgical care had also undergone surgery.

claim_id

stats_check.effect_sizes

locator

p. 794 Results

quote

mean difference in change from baseline of 7.8 (95% confidence interval, 1.5 to 14.1)

claim_id

practice_impact.bottom_line

locator

p. 794 Conclusions

quote

patients who underwent surgery showed significantly more improvement in all primary outcomes than did patients who were treated nonsurgically.

Metadata

Generated at

2026-03-06T13:41:29.251Z

Version

top 100 cited in past 20 years