The Lumbar Adjacent Segment Stenosis Trial

Globally, musculoskeletal disorders are the leading causes of years lived with disability, with spinal disorders contributing to 70% of this burden. The consequences for patients, their families and society are significant.

Lumbar spinal stenosis is a common condition with an increasing prevalence, largely due to an aging population. It is caused by degenerative changes in the supporting structures of the lumbar spine leading to symptoms like low-back pain and radiating pain to the lower extremities as well as numbness and weakness causing walking restrictions. Annually, about 200,000 to 250,000 new cases are diagnosed in the United States alone and this condition is the most frequent indication for spinal surgery in patients older than 65 years of age. Approximately 10% of the population over 50 years suffer from symptomatic lumbar spinal stenosis.

Lumbar surgery may be indicated in patients with various degenerative spinal conditions when non-surgical treatment has failed, and radiological findings are consistent with clinical symptoms. Decompression (removal of stenotic structures to relieve nerve pressure) and spinal fusion (the use of screws, rods, plates, or other devices to assist in achieving fusion between vertebral bodies) remain the cornerstone of operative treatment. The volume of lumbar spine surgery has increased considerably over the past decades, particularly for more complex procedures such as instrumented fusion, and especially among the elderly. In the United States, the hospital costs of lumbar instrumented fusion procedures is reported to be higher than for any other surgical procedure such as percutaneous coronary angioplasty and knee- and hip arthroplasty. In Norway, approximately 700 lumbar spinal fusions are performed every year, with about half of them being reoperations.

Lumbar spinal fusion has the highest complication rates among all spinal surgeries. Complications involving either the segment above or below (=adjacent to) a previously fused segment can arise years after the initial fusion. An annual incidence of 2,5-4% for new subsequent surgery after initial lumbar fusion is reported, 13.6% and 22.2% were re-operated within 5 and 10 years respectively. The majority of reoperations following an initial fusion are due to spinal stenosis adjacent to a previously fused segment (Lumbar Adjacent Segment Stenosis - LASS), and these patients constitute a substantial subgroup within the extensive and growing patient population with spinal stenosis.

At present, there are no guidelines or prospective studies that can advise clinicians on how patients with LASS should be treated, and this issue is emerging as one of the most relevant and debated questions within degenerative spine surgery. According to register data, the most common treatments are either extending the fusion beyond the symptomatic adjacent stenotic segment or just decompression of the stenotic structures, with a tendency of performing just decompression in patients with cranial LASS and decompression with extended fusion in patients with caudal LASS. However, there is large variation in surgical practices both between and even within hospitals. Considering the increasing prevalence of LASS, costs, and risks associated with spinal fusions, there is need for randomized trials to evaluate surgical practice and the relative efficacy of the two most common surgical treatments.

Statistical Hypothesis: The primary objective is to evaluate the effect of Decompression alone versus Decompression with extended Fusion at two-year FU. Thus, the null hypothesis to be tested in relation to the primary endpoint is as follows:

• Null hypothesis: The proportions of participants reaching a minimum 30% ODI improvement from baseline to two-year FU is equal in the Decompression alone and the Decompression with extended Fusion treatment groups.
• Alternative hypothesis: The proportions of participants reaching a minimum 30% ODI improvement from baseline to two-year FU is greater in one of the treatment groups than the other Secondary, the researchers in the current project aim to conduct a cost-utility analysis and evaluate the impact of radiological parameters. Main endpoint is at 2-year follow-up, but each participant will be followed for 5 years after surgery.

Patients will be randomly assigned in a 1:1 ratio to undergo decompression surgery alone or decompression combined with extended instrumented fusion. Procedure for decompression will be mid-line preserving technique, and with the use of vision-enhancing aid (microscope magnifying glasses). The procedure for extended fusion will involve bilateral pedicle screws and with an aim to transplant a minimal amount of bone volume. All surgeons need to have at least senior surgeon experience.

Statistical analysis: The effect measure for the primary endpoint will be the difference between the treatment groups in the probability of reaching a minimum 30% ODI improvement from baseline to one-year FU. It will be estimated by the difference in observed proportions with a 95% Newcombe hybrid score confidence interval. A two-sided P-value for the null hypothesis of a zero difference will also be calculated with the Fisher mid-P test. Safety endpoints will be tabulated and presented with descriptive statistics.

Further details of the statistical methods will be provided in the SAP. The SAP will be finalized before the database lock and will include a more technical and detailed description of the statistical analyses.

Interim analysis: An interim analysis for safety will be performed when 50% of the patients in each group have completed the 12-month FU. If one of the proposed stop criteria is fulfilled the study will be terminated:

1. The proportion of patients needing reoperation due to any condition in the operated level(s) is statistically significantly (p<0.05) higher in one of the groups.
2. The change in NRS leg pain from baseline to 12-month FU is statistically significantly (p<0.05) lower in one of the groups.

An independent statistician blinded for treatment adherence will perform the interim analysis. Only data on reoperations and on NRS for leg pain will be available to the statistician. Following the analysis, the statistician will inform the steering committee, via the central monitor (at Clinical Trial Unit, Oslo University Hospital) whether the study can continue. Further information about the analysis will not be disclosed or available by anyone else than the independent statistician until the main analysis at 2-year FU.

Sample Size Determination: This study is powered for the main outcome (ODI) to be able to detect at least a 20% absolute difference between the proportions of responders (ODI improvement of minimum 30% from baseline) assessed at two-year follow-up between the two treatment groups. For the sample size determination, we anticipated that 55% from the decompression with extended fusion arm and 35% from the decompression alone arm will reach the primary outcome. The statistical power selected is 80%, and the accepted type I error rate 5% (β =0.2, α = 0.05). Considering these assumptions with a superiority two-sided test, 96 patients are required in each group (192 total). The significance level will be set at 0.05 and all estimates will be reported with 95% CI and the exact p-value.