Background context:

The indication to perform a fusion and decompression surgery as opposed to decompression alone for lumbar degenerative spondylolisthesis (LDS) remains controversial. A variety of factors are considered when deciding on whether to fuse, including patient demographics, radiographic parameters, and symptom presentation. Likely surgeon preference has an important influence as well.

Purpose:

The aim of this study was to assess factors associated with the decision of a Canadian academic spine surgeon to perform a fusion for LDS.

Study design/setting:

This study is a retrospective analysis of patients prospectively enrolled in a multi-center Canadian study that was designed to evaluate the assessment and surgical management of LDS.

Patient sample:

Inclusion criteria were patients with: radiographic evidence of LDS and neurogenic claudication or radicular pain, undergoing posterior decompression alone or posterior decompression and fusion, performed in one of seven, participating academic centres from 2015-2019.

Outcome measures:

Patient demographics, patient-rated outcome measures (ODI, NRS back pain and leg pain, SF-12), and imaging parameters were recorded in the Canadian Spine Outcomes Research Network (CSORN) database. Surgeon factors were retrieved by survey of each participating surgeon and then linked to their specific patients within the database.

Methods:

Univariate analysis was used to compare patient characteristics, imaging measures, and surgeon variables between those that had a fusion and those that had decompression alone. Multivariate backward logistic regression was used to identify the best combination of factors associated with the decision to perform a fusion.

Results:

This study includes 241 consecutively enrolled patients receiving surgery from 11 surgeons at 7 sites. Patients that had a fusion were younger (65.3±8.3 vs. 68.6±9.7 years, P=0.012), had worse ODI scores (45.9±14.7vs.40.2±13.5, P=0.007), a smaller average disc height (6.1±2.7 vs. 8.0±7.3 mm, P=0.005), were more likely to have Grade II spondylolisthesis (31% vs. 14%, P=0.008), facet distraction (34%vs.60%, P=0.034), and a non-lordotic disc angle (26% vs. 17%, P=0.038). The rate of fusion varied by individual surgeon and practice location (P<0.001, respectively). Surgeons that were fellowship trained in Canada more frequently fused than those who fellowship trained outside of Canada (76% vs. 57%, P=0.027). Surgeons on salary fused more frequently than surgeons remunerated by fee-for-service (80% vs. 64%, P=0.004). In the multivariate analysis the clinical factors associated with an increased odds of fusion were decreasing age, decreasing disc height, and increasing ODI score; the radiographic factors were grade II spondylolisthesis and neutral or kyphotic standing disc type; and the surgeon factors were fellowship location, renumeration type and practice region. The odds of having a fusion surgery was more than two times greater for patients with a grade II spondylolisthesis or neutral/kyphotic standing disc type (opposed to lordotic standing disc type). Patients whose surgeon completed their fellowship in Canada, or whose surgeon was salaried (opposed to fee-for-service), or whose surgeon practiced in western Canada had twice the odds of having fusion surgery.

Conclusion:

The decision to perform a fusion in addition to decompression for LDS is multifactorial. Although patient and radiographic parameters are important in the decision-making process, multiple surgeon factors are associated with the preference of a Canadian spine surgeon to perform a fusion for LDS. Future work is necessary to decrease treatment variability between surgeons and help facilitate the implementation of evidence-based decision making.