Decompression alone or with fusion for degenerative lumbar spondylolisthesis (Nordsten-DS): five year follow-up of a randomised, multicentre, non-inferiority trial

Abstract

Objective: To assess whether decompression alone is non-inferior to decompression with instrumented fusion five years after primary surgery in patients with degenerative lumbar spondylolisthesis.

Design: Five year follow-up of a randomised, multicentre, non-inferiority trial (Nordsten-DS).

Setting: 16 public orthopaedic and neurosurgical clinics in Norway.

Participants: Patients aged 18-80 years with symptomatic lumbar spinal stenosis and a spondylolisthesis of 3 mm or more at the stenotic level.

Interventions: Decompression surgery alone and decompression with additional instrumented fusion (1:1).

Main outcome measures: The primary outcome was a 30% or more reduction in Oswestry disability index from baseline to five year follow-up. The predefined non-inferiority margin was a -15 percentage point difference in the proportion of patients who met the primary outcome. Secondary outcomes included the mean change in Oswestry disability index, Zurich claudication questionnaire, numeric rating scale for leg and back pain, and EuroQol Group 5-Dimension (EQ-5D-3L) questionnaire.

Results: From 12 February 2014 to 18 December 2017, 267 participants were randomly assigned to decompression alone (n=134) and decompression with instrumented fusion (n=133). Of these, 230 (88%) responded to the five year questionnaire: 121 in the decompression group and 109 in the fusion group. Mean age at baseline was 66.2 years (SD 7.6), and 69% were women. In the modified intention-to-treat analysis with multiple imputation of missing data, 84 (63%) of 133 people in the decompression alone group and 81 (63%) of 129 people in the fusion group had a at least a 30% reduction in Oswestry disability index, a difference of 0.4 percentage points. (95% confidence interval (CI) -11.2 to 11.9). The respective results of the per protocol analysis were 65 (65%) of 100 in the decompression alone group and 59 (66%) of 89 in the fusion group, a difference of -1.3 percentage points (95% CI -14.5 to 12.2). Both 95% CIs were higher than the predefined non-inferiority margin of -15%. The mean change in Oswestry disability index from baseline to five years was -17.8 in both groups (mean difference 0.02 (95% CI -3.8 to 3.9)). Results of the other secondary outcomes were in the same direction as the primary outcome. From two to five year follow-up, a new lumbar operation occurred in six (5%) of 123 people in the decompression group and 11 (10%) of 113 people in the fusion group, with a total from baseline to five years of 21 (16%) of 129 people and 23 (18%) of 125, respectively.

Conclusions: In participants with degenerative spondylolisthesis, decompression alone was non-inferior to decompression with instrumented fusion five years after primary surgery. Proportions of subsequent surgeries at the index level or an adjacent lumbar level were no different between the groups.

Trial registration: ClinicalTrials.gov NCT02051374.

Fig 1

Screening, randomisation, and follow-up of the trial participants. Patients who were referred with the diagnosis of lumbar spinal stenosis were screened for the presence of spondylolisthesis. 738 patients who were assessed for eligibility in the trial could have been excluded for more than one reason. The diagnosis of foraminal stenosis of grade 3 was made according to the classification of Lee et al. From 15 April 2014 (the start of inclusion) to 29 August 2015, a score of less than 25 on the Oswestry disability index (ODI) was an exclusion criterion. MRI=magnetic resonance imaging

Fig 2

Primary outcome. No of patients/total no (%) refers to the proportion of patients with 30% or more reduction in Oswestry disability index in each specified analysis. The between group differences in percentage points and its 95% confidence intervals (CIs) were calculated as decompression alone minus decompression with fusion. For patients in the modified intention-to-treat set with imputation of missing data at five year, the missing values were replaced by multiple imputation