No significant association was seen between type of index operation and need for revision surgery among patients with degenerative lumbar spinal stenosis (DLSS), according to a cohort study.
Among 328 patients, the cumulative incidence of revision operations required within 3 years of index surgery was 11.3% for those who underwent decompression alone compared with 13.9% for those who underwent decompression with fusion (log-rank P=0.60), reported Nils H. Ulrich, MD, of the University of Zurich in Switzerland, and colleagues
Moreover, there was no significant difference between the two groups over time (adjusted HR 1.40, 95% CI 0.63-3.13), the group noted in JAMA Network Open.
As for secondary outcomes, the number of revisions was significantly associated with higher Spinal Stenosis Measure (SSM) symptom severity scores (β=0.171, 95% CI 0.047-0.295, P=0.007) and with lower EuroQol Health-Related Quality of Life 5-Dimension 3-Level questionnaire (EQ-5D-3L) summary index scores (β = -0.061, 95% CI -0.105 to -0.017, P=0.007), but was not linked with higher SSM physical function scores (β=0.068, 95% CI -0.036 to 0.172, P=0.20).
While surgical decompression has long been the gold-standard treatment for spinal stenosis with and without spondylolisthesis not responding to non-surgical treatment, “the use of additional fusion in surgery for degenerative spondylolisthesis has been a controversial issue,” wrote Ivar Magne Austevoll, MD, of Haukeland University Hospital in Bergen, Norway, and Eira Ebbs, MSS, of Oslo University Hospital in Norway, in an accompanying editorial.
Ulrich and colleagues pointed out “there has been a remarkable increase in decompression and fusion surgery for lumbar spine disorders during the past 3 decades; numbers have increased by up to 3 fold, with the largest increase in patients aged 65 years or older.”
For DLSS, the most frequent indication for spine surgery in older adults, rates of decompression without fusion procedures are inconsistent, with a slight decrease in the U.S. according to two studies and an increase in Australia, the group noted.
“In contrast, all 3 studies reported up to a 15-fold increase in fusion procedures,” the authors wrote. This trend could lead to increased rates of revision surgery, generally “an undesired event owing to intraoperative or postoperative complications, bone fusion failures, persistence of pain, or additional progressive degeneration such as adjacent segments disease.”
Austevoll and Ebbs observed that “current evidence indicates that the more invasive fusion procedure is associated with increased costs but not clinical benefits.” Given the frailty of many older people with lumbar spinal stenosis, they suggested that “the risk of further surgery should be kept as low as possible because results after revision operations may be worse than those after primary operations.”
While cautioning that the findings cannot be generalized to all spinal stenosis patients, the editorialists noted that the study “showed that decompression alone was associated with more revision operations owing to restenosis at the index level, whereas fusion was associated with more subsequent operations for replacement of screws and rods or a secondary decompression.”
Ulrich and team called for further research to establish “what constitutes an unfavorable outcome in a revised surgical decompression with or without fusion … linking it to patient-level factors [that] would aid in the development of guidelines to evaluate patients before surgery, stratify risks, predict outcomes, and make shared decisions.”
For this cohort study, Ulrich and colleagues analyzed data from the Lumbar Stenosis Outcome Study on 328 patients with DLSS treated at eight spine surgery and rheumatology units in Switzerland from December 2010 to December 2015. Median age was 73 years, and 50.3% were men. The two cohorts differed somewhat, with the 22% who underwent fusion being a few years younger and having a slightly longer duration of symptoms than the 78% who received decompression alone. Patients were followed for 3 years.
Study limitations included the heterogeneous patient sample, a lack of randomized treatment allocation, and possible selection bias.
The study authors and editorialists reported no conflicts of interest.