The effectiveness of selective thoracic fusion for treatment of adolescent idiopathic scoliosis: a systematic review

Abstract

Scoliosis curves have a proven complex deformity, consisting of a three-dimensional deformity involving the coronal, sagittal and rotational planes. For many years, spinal surgeons have been debating whether a more rigid and straighter spine or a mobile and less straight spine provides better outcomes. The premise of selective thoracic fusion is that after fixation of the primary thoracic curve, there is spontaneous coronal correction of the unfused lumbar curve. Thus, the thoracic curve can be exclusively fused to allow for a more mobile lumbar spine. The objective of this review was to assess the effectiveness of selective thoracic fusion as a form of treatment in adolescent idiopathic scoliosis (AIS). This was compared with all other forms of operative management for major structural thoracic curves. A comprehensive and exhaustive literature search was conducted for studies that included children aged 10-18 years with adolescent idiopathic scoliosis curve with a thoracic component that is described as structural, treated with selective fusion of the thoracic curve with no distal fusion lower than L1. Congenital, neuromuscular or syndromic causes were excluded. All studies needed a minimum follow-up of 2 years. Radiological outcomes measured were main thoracic curve, compensatory lumbar curve, coronal balance, thoracic kyphosis, lumbar lordosis, sagittal balance, thoracic apical vertebral rotation, lumbar apical vertebral rotation. Clinical outcomes included quality of life surveys, pulmonary function and complications. A total of 373 studies were retrieved for review with 339 studies excluded after reading the full article for clearly not meeting the inclusion criteria of the review. Two reviewers independently assessed the 34 studies for methodological quality. Eight studies were eligible for inclusion in a meta-analysis comparing selective thoracic fusion via the anterior or posterior approach. There was no significant difference between approaches for the outcomes measured except for post-operative lumbar lordosis. The anterior approach had a 4.29 (95% CI: 1.5, 7.05) degree lower post-operative lumbar lordosis than the posterior approach. Two studies were eligible for inclusion for descriptive analysis of comparing compensated curves against decompensated or imbalanced curves post-operatively with no obvious difference between groups with the exception of a worsening sagittal balance in the coronally decompensated group. Two studies were eligible for inclusion for meta-analysis of comparing selective thoracic fusion in Lenke B vs Lenke C curves, with no difference in groups between the outcomes measured. Thirty-three studies were eligible for inclusion for meta-analysis of effectiveness of selective thoracic fusion in adolescent idiopathic scoliosis. Selective thoracic fusion was significant in changing the main thoracic curve, compensatory lumbar curve, and thoracic kyphosis post-operatively. Selective thoracic fusion did not have a significant effect on changing the coronal balance, lumbar lordosis, sagittal balance, thoracic apical vertebral rotation or lumbar apical vertebral rotation. The highest reported complication was coronal decompensation which was reported in 23.1% (95% CI: 15.1, 32.1%). Pulmonary function and quality of life were poorly reported and therefore little conclusions could be made, besides a return to respiratory baseline and adequate quality of life following surgery. Unfortunately, due to the lack of high level evidence in the form of RCTs and using the best available evidence which mainly consisted of retrospective case series, only weak conclusions can be drawn into the true effect of selective thoracic fusion. Further prospective uniform trials will be needed to increase the level of evidence available in this topic area.