Are the Benefits of Interbody Fusion Worth the Risks for Adult Deformity Correction?

2013

Background

Interbody fusions via an anterior, lateral, or transforaminal (TLIF) approach have often been advocated in deformity surgery to improve curve correction and obtain better clinical results. However, the actual clinical benefit of interbody use to assist with deformity correction in the adult population relative to posterior spinal fusion alone is uncertain.

Purpose

The goal of this study is to determine what comparative benefits and risks interbody fusion for the purpose of deformity correction have relative to posterior spinal fusion alone without interbodies. Interbody placement via transforaminal and thoraco-abdominal approaches are considered separately.

Study Design

Retrospective study of prospectively collected data.

Patient Sample

127 consecutive adult patients (> 21 years of age) with 2 year follow-up undergoing surgical correction for either adult idiopathic scoliosis or adult degenerative scoliosis.

Outcome Measures

Clinical outcomes include visual analog scale and Oswestry Disability Index (ODI) as well as perioperative complications. Radiographic outcomes include percent of deformity correction, Cobb angles, coronal balance, sagittal balance, and pelvic parameters.

Methods

127 consecutive adult patients with minimum 2 year follow-up underwent spinal deformity correction via anterior-posterior approach with anterior interbody cages (35 patients), posterior spinal fusion with TLIFs (48 patients), or posterior-only without interbodies (44 patients). Diagnoses were adult idiopathic scoliosis (60 patients) or degenerative scoliosis (67 patients). Outcomes were obtained prospectively by VAS and ODI at pre-op, 1 year, 2 year, and latest follow-up. Radiographs included posterior-anterior (PA) and lateral scoliosis films.

Results

Pre-operative demographics and clinical characteristics were similar between groups with the exception of initial curve magnitude, which was higher in the Anterior/PSF group (54.7º) than the TLIF (39.0º) or PSF (39.8º) groups (p=0.002). However, there was no difference in curve correction (Anterior: 66.8%, TLIF: 65.5%, PSF only: 61.4%, p=0.405) at 2 years post-op. There were also no significant differences between the groups in regards to lumbar lordosis, thoracic kyphosis, and coronal balance at 2 years. Sagittal balance was significantly better in the PSF group compared with the 2 interbody groups (Anterior: 4.3, TLIF: 1.9, PSF: 0.8, p=0.023), but all were within acceptable limits. Improvement in VAS and ODI scores at 2 years was similar between all groups and highly significant compared with pre-op (p<0.0001). Complications were greatest in the Anterior/PSF group and lowest in the PSF group and included revision surgery (anterior: 7 [20.0%], TLIF :10 [20.8%], PSF:5 [11.4%]), non-union (anterior: 4 [11.4%], TLIF: 4 [11.1%], PSF: 0 [0%]), adjacent level disease (anterior: 13 [37.1%], TLIF: 15 [31.3%], PSF: 16 [36.4%]), infection (anterior: 2 [5.8%], TLIF:6 [4.7%], PSF:0 [0%]), and imbalance (anterior: 8 [22.9%], TLIF: 8 [16.7%], PSF: 3 [6.8%]). Medical complications included 1 stroke and 1 death in anterior/PSF, 1 pulmonary failure in TLIF, and 1 myocardial infarction in PSF.

Conclusion

Compared with either anterior/PSF or TLIF-assisted correction, posterior spinal fusion alone achieves similar radiographic and clinical outcomes at 2 years in adult scoliosis patients while enjoying a substantially lower rate of complications and revision.