Non-instrumented correction of cervicothoracic kyphosis in ankylosing spondylitis: a critical analysis on the results of open-wedge osteotomy C7-T1 with gradual Halo-Thoracic-Cast based correction.

INTRODUCTION: With progression of cervicothoracic kyphosis (CTK), ankylosing spondylitis (AS) patients suffer functional disability. Surgical correction still poses neurologic risks, while evidence of an ideal technique preventing its complications is weak.
MATERIALS AND METHODS: We report our results with non-instrumented correction in perspective of a review of literature, serving as an important historical control. Database review identified 18 AS patients with CTK correction. After application of a Halo-Thoracic-Cast (HTC) patients underwent posterior non-instrumented open-wedge osteotomy at C7/T1 and osteotomy closure by threaded HTC-rod adjustments. Postoperative gradual HTC correction was continued for 2-4 weeks. Patients were invited for follow-up and medical charts were reviewed for demographics, surgical details, complications and outcomes. The patients' preoperative, postoperative, before HTC removal and follow-up photographs were analyzed for the Chin-Brow-Vertical-Angle (CBVA), radiographs for the CTK angle.
RESULTS: Patients' age was 50 ± 11 years, follow-up was 37 ± 47 months and CBVA correction was 25° ± 9° (p < 0.000001). The final radiographic correction at follow-up was 20° ± 11° (p = 0.00002). At the latest follow-up, three patients judged their outcome as excellent, nine good, three moderate and one poor. Upon invitation, seven patients appeared with follow-up averaging 87 months. Neck-pain disability index was 8 ± 14 %. Two patients died, three were lost, one had revision elsewhere and five just had a routine follow-up. Six patients sustained a minor and ten a major complication. Revisions were indicated in five patients including infection, C8-radiculopathy and neurologic events by translation at the osteotomy. A total of 44 % of patients showed translation at the osteotomy indicating acute surgery with instrumentation twice after osteotomy closure, three patients had a revision posterior decompression and instrumented fusion for sequels related to translation.
CONCLUSION: With the non-instrumented HTC-based technique, average CBVA correction of 25° was achieved and all patients were ambulatory at follow-up. However, regarding translation at the osteotomy, loss of correction, morbidity of the HTC and lack of control at the osteotomy instrumentation-based correction and instrumented fusion seem to be preferable.