Anterior Fusion after Thoracoscopic Disc Excision: Analysis of 103 Consecutive Deformity Cases with Greater Than Two Year Follow Up

Purpose: The early results of thoracoscopic anterior release and fusion procedures have been reported; however, the actual rate of anterior arthrodesis with this minimally invasive approach has not been studied. The purpose of this analysis was to examine the rate and quality of anterior thoracic fusion following thoracoscopic disc excision and bone grafting procedures in patients with spinal deformity.

Methods: A consecutive series of 103 spinal deformity patients treated with a combined thoracoscopic anterior release/fusion and posterior instrumentation/fusion procedure performed between 1994 and 2000 were reviewed. The outcomes were assessed with special attention given to quality of interbody fusion anteriorly for each interspace treated thoracoscopically. The most recent PA and lateral radiographs (all > 2 years postoperative) were interpreted by two of the authors. A previously reported grading system for anterior lumbar fusion was modified as follows: Grade I - Definite fusion with bridging trabecular bone; Grade II - Probably fused, bone >50% of the width on AP or lateral radiograph; Grade III - Probably not fused, bone <50 % of the width; Grade IV - Definitely not fused, no bridging bone.

Results: Of the 103 patients (45 neuromuscular, 39 idiopathic scoliosis/kyphosis, 18 other) 66 had radiographs available for review > 2 years post-operatively (average follow-up: 3 years). The anterior bone graft was allogenous in the vast majority of cases. The mean preoperative scoliosis of 62 degrees (range 32 - 105) was corrected to 27 degrees (range 1 - 65) (58% correction; range 5 - 98%), while those with kyphosis were corrected from 86 degrees (range 70 - 120) to 57 degrees (range 30 - 108). A total of 413 disc spaces were rated for fusion. Of these, there were 32% Grade I, 36% Grade II, 18% Grade III, 9% Grade IV, and 5% that were not ratable. Intraobserver reliability was moderate º=.47. Of the Grade IV fusion ratings, the majority (57%) were at the thoracolumbar junction (T11-L2).

Conclusion: Thoracoscopic disc excision with primarily allogenous bone grafting resulted in a radiographic appearance of fusion (Grade I or II) in 72% of the ratable disc spaces when combined with a posterior procedure. The region of the spine most difficult to achieve anterior arthrodesis appeared to be the thoracolumbar junction.