Sensory dysfunction in HTLV-I-associated myelopathy/tropical spastic paraparesis. A comprehensive neurophysiological study.

BACKGROUND

Direct lateral (transpsoas) lumbar interbody fusion (LLIF) reportedly achieves union by 1 year postoperatively, but how soon fusion occurs after these minimally invasive procedures is unclear. This study investigated LLIF fusion progression at 6 months and 1 year in a large-scale cohort using bone morphogenetic protein (BMP) graft and examined risk factors associated with failed fusion.

METHODS

Patients undergoing primary LLIF with a single surgical team from 2015 through 2016 with polyetheretherketone (PEEK) iimplants and BMP graft were identified. Retrospective chart review included demographics and medical history, construct length and location, and concurrent L5-S1 fusion. Inclusion criteria included minimum 1-year follow-up and postoperative lumbar computed tomography at 6 months and 1 year, which was independently assessed for bony union at each level.

RESULTS

166 patients underwent LLIF at a total of 312 levels. Seventy-nine patients (48%) underwent 1-level fusion; 45 (27%), 2 levels; and 42 (25%), 3 or more levels. At 6 months, 160 (51%) levels showed fusion. At 1 year, 70% of the remainder were fused, and total fusion rate was 85%. Fusion rates from L1 through L4 were similar (84%-87%). Nonunion was not significantly associated with construct length (P = .19), concurrent anterior L5-S1 interbody fusion (P = .50), age (P = .70), BMI (P = .15), or comorbidities such as diabetes (P = .86) or thyroid disease (P = .46).

CONCLUSIONS

This large retrospective cohort study corroborates prior 1-year LLIF fusion rate reports (85%) independent of construct length or location or medical comorbidities. Significantly, half showed fusion by 6 months, earlier than previously described and validating the efficacy of LLIF.

LEVEL OF EVIDENCE: 5.

CLINICAL RELEVANCE: This study presents a large cohort of patients to support effective lumbar fusion after LLIF with BMP-2.