The lumbar spine is the part of the spine that bears the most weight, consisting of five vertebrae. Each vertebra is connected by an intervertebral disc, and together they form a barrier to protect the spinal cord nerves within the spinal canal. Lumbar spinal stenosis refers to the reduction of space within the lumbar spinal canal, which compresses the spinal cord or nerve roots, leading to a series of symptoms. Depending on the cause, lumbar spinal stenosis is mainly divided into congenital lumbar spinal stenosis, which is due to abnormal development of the spinal canal, and acquired lumbar spinal stenosis, which is caused by degenerative changes such as lumbar disc degeneration and joint process hyperplasia. Lumbar spinal stenosis is a common cause of low back and leg pain, with a reported incidence rate as high as 11% in the general population and 19.4% in those over 60 years old, with a trend of increasing with age. With the aging of the population, the incidence of lumbar spinal stenosis is also on the rise.
The treatment of lumbar spinal stenosis mainly includes conservative treatment and surgical treatment. For patients whose symptoms severely affect their lives and do not respond well to conservative treatment, surgery should be performed as soon as possible to relieve nerve compression. The main goal of surgical treatment for lumbar spinal stenosis is to relieve nerve compression, stabilize the spine, and restore the normal volume of the spinal canal. Although traditional posterior lumbar decompression surgery can achieve definite decompression effects, it has issues such as severe destruction of posterior column structures and large surgical trauma. Minimally invasive techniques are a hot topic in various clinical fields, and microdiscectomy (MD) through a minimally invasive channel is a mature minimally invasive technique for treating lumbar spinal stenosis. As early as 1999, prospective randomized controlled studies confirmed that it has the same excellent clinical efficacy as open surgery but with less surgical trauma. It has now become a classic surgical method for treating lumbar spinal stenosis, with related articles published in the top spinal clinical journal JBJS.
In recent years, with the development of new equipment, endoscopic techniques have been gradually applied to the treatment of lumbar spinal stenosis. Unilateral biportal endoscopic discectomy (UBE) has the characteristics of convenient operation and flexibility because its endoscopic channel and working channel are separate and can use traditional open instruments. It can achieve good decompression effects and has been widely used in various medical centers. Percutaneous endoscopic lumbar discectomy (PELD), as a new generation of endoscopic technique, has less surgical trauma and shorter operation time. It has also been successfully applied to the clinical treatment of lumbar spinal stenosis and achieved good results. However, due to the limitations of the single-endoscopic channel field of view, its learning curve is steep, the surgical difficulty is high, and it is difficult to achieve complete decompression for bilateral spinal stenosis or dorsal nerve root compression. The efficiency of dealing with bony stenosis structures during surgery is low, leading to longer operation time and increased risk of nerve root and dural sac injury. Many clinical studies have compared the advantages and disadvantages of endoscopic techniques with traditional minimally invasive techniques and different endoscopic techniques, concluding that various endoscopic techniques can achieve surgical efficacy similar to traditional techniques with the advantages of less surgical trauma and faster postoperative recovery. Guilherme et al. believe that PELD has similar clinical efficacy to MD, but MD is superior for complex foraminal stenosis. In addition, a review by Ohyuk et al. suggests that UBE has better clinical efficacy and a lower complication rate compared to MD or PELD. However, most of these studies are small-scale, single-center, and retrospective, and the conclusions lack high-quality evidence support.
In recent years, single-channel decompression techniques under endoscopy have developed rapidly, especially large-channel endoscopes and related tools, which have significantly improved work efficiency and surgical safety. Many hospitals and medical research centers have introduced the Endo-Surgi Plus working channel, which has a larger working tube than traditional endoscopes, making it easier to treat certain types of lumbar disc herniation and lumbar spinal stenosis, further optimizing spinal endoscopic surgery techniques for lumbar spinal stenosis.
With the development of surgical instruments and approaches, these three minimally invasive spinal canal decompression techniques can effectively relieve pain symptoms in patients with various types of lumbar spinal stenosis. Compared with traditional open surgery, these techniques have reduced operation time, intraoperative blood loss, and postoperative pain. However, there are slight differences in the degree of trauma and ease of operation among these three techniques. Currently, there is still a lack of prospective validation of the clinical efficacy of these surgical techniques internationally, and the advantages and disadvantages of different endoscopic surgeries compared to traditional fenestration surgery.The purpose of this study is to conduct a multicenter comparison of the clinical efficacy of Endo-Surgi Plus endoscopic technique, UBE endoscopic technique, and microdiscectomy technique in the treatment of lumbar spinal stenosis.
Subjects were randomly divided into the Endo-Surgi Plus endoscopy group, the UBE group, and the fenestration group through the Quadrant channel. Randomization (1:1:1 grouping) was conducted within 2 weeks before surgery. Investigators used a block randomization design, stratified by 23 hospitals, with non-competitive enrollment between hospitals. The blocks were kept as small as possible (randomly selected block sizes of 3 and 6) to ensure that each hospital performed a similar number of all three types of surgeries. The computer-generated allocation output was sent via email to the local study coordinators (CRC), who did not participate in patient recruitment or treatment.
Due to the nature of the interventions, double blinding was not feasible. Randomization of patients aimed to reduce potential biases in preoperative planning and patient treatment. Throughout the study, outcome assessors and data analysts remained blinded to reduce biases in clinical outcome assessment and data analysis. Patients were informed that no treatment regimen was recorded as superior to another.
Determination of Sample Size: This study is a randomized controlled trial, with the postoperative 1-year ODI score of the subjects as the main outcome indicator. The trial required 324 patients to be included in total, with 80% statistical power to detect a difference of at least 12 points in ODI scores between treatment groups at a significance level of 0.05.Investigators conservatively chose a difference of 12 points because the U.S. Food and Drug Administration suggests that a 15-point reduction in ODI score indicates a minimally important improvement after spinal fusion surgery. Investigators conservatively set the standard deviation of the ODI score to 18, and estimated a dropout rate of 15%. Finally investigators used PASS to estimate that the trial would need to include a total of 324 patients.
Statistical and Analytical Plan: Patients who underwent the designated surgery and completed at least one year of follow-up were included. For continuous data that met normal distribution, analysis of variance (ANOVA) and post-hoc tests were used to compare differences among the three treatment groups; otherwise, the Kruskal-Wallis H-test was applied. Categorical variables were analyzed using chi-square or Fisher's exact test (expressed in numbers and percentages) to compare baseline characteristics among groups.
A mixed-effects model for repeated measures was used to compare changes in ODI and SF-36 scores relative to baseline across groups. Considering the correlation of repeated measures within patients, an unstructured covariance matrix was specified. The model included fixed effects for treatment (Endo-Surgi Plus endoscopy, UBE, minimally invasive fenestration), time (2 days, 1 month, 3 months, 1 year postoperatively), and time-treatment interaction, with results reported as least squares means and 95% confidence intervals. Robust standard errors and test statistics were calculated for the fixed effects. Appropriate comparisons were made within the mixed-effects model for repeated measures to compare least squares means between treatment groups at each time point and within each treatment group across time points. Considering the various etiologies of lumbar spinal stenosis such as disc degeneration, ligamentum flavum hypertrophy, and bone hyperplasia, this study will perform multivariate adjustments and subgroup analyses based on etiology. All analyses were conducted using R software version 4.3.3, primarily utilizing the lme4 and emmeans packages. A P-value of less than 0.05 was considered statistically significant.
