A pilot randomized control trial comparing posterior paramedian versus midline incisions for interbody fusions of the lumbar spine.

Background context: The midline posterior approach to the lumbar spine remains the most commonly used approach for lumbar spine fusion. This approach, however, is not without downfalls and complications, most notably the risk of infection. The paramedian approach offers potential advantages but literature comparing these approaches is lacking.

Purpose: This pilot feasibility study is needed to establish the protocol and data necessary to determine the sample size required for a well-powered randomized control trial (RCT) to directly compare the midline and paramedian approaches to the lumbar spine.

Study design/setting: This was a single-centered, prospective, pilot randomized control trial conducted in Canada.

Patient sample: All consecutive patients deemed appropriate surgical candidates with single- or two-level degenerative conditions of the lumbar spine were approached for participation in the trial between December 2017 and November 2024. Inclusion criteria were a clinical history of buttock and/or leg pain, with symptoms for greater than three months, and failed conservative care.

Outcome measures: The primary clinical outcome measure was deep infection rate within 3 months of surgery, and the second outcome was re-operation at any time during follow up. Other clinical and safety outcomes include intra-operative factors (length of operation, estimated blood loss, length of stay), inpatient and outpatient adverse events, and patient reported outcomes (i.e. ODI, SF-12, VAS for back pain, leg pain, and overall health score, PHQ-9, and EQ-5D) recorded pre-operatively and at 6-18 week and one-year follow-ups.

Methods: Patients underwent a one or two-level lumbar instrumented fusion with interbody device, with the surgery being randomized to either a posterior midline or posterior paramedian approach.

Results: A total of 112 participants were assessed for eligibility in the trial of which 11 were excluded. Fifty-one patients were randomized to the midline group, and 50 to the paramedian group. Five patients (9.6%; 95% CI: 3.1-21.0%) in the midline group were diagnosed with a deep infection and two patients (4.1%; 95% 95% CI: 0.4-14.0%) in the paramedian group. A total of 13 (25%; 95% CI: 14.0-38.9%) patients underwent a revision procedure in the midline group, with a further three requiring a third procedure, while three revision procedures (6.1%; 95% CI: 1.2-16.9%) were performed in the paramedian group. The ODI, SF-12, VAS, and EQ-5D all showed improvement from pre-operative to one year post-operative in both the paramedian and midline groups.

Conclusion: The current pilot RCT demonstrates potential differences in post-operative deep wound infection at 3-months as well as different rates of re-operation between the midline and paramedian approach. Patient reported outcome measures were improved in each group, but differences in improvement, safety profile, and length of stay did exist. While clinical comparisons cannot be drawn, this pilot data will be utilized to refine the study protocol and ensure appropriate powering of a multi-centered RCT.