Etiological analysis and clinical significance of proximal junctional kyphosis after osteotomy in ankylosing spondylitis-related kyphosis

English Keywords:Ankylosing spondylitis  Thoracolumbar kyphosis  Proximal junctional kyphosis  Pedicle subtraction osteotomy  Smith-Petersen osteotomy

English Abstract:

  【Abstract】 Objectives: To investigate the risk factors and clinical features of proximal junctional kyphosis (PJK) after osteotomy in thoracolumbar kyphosis secondary to ankylosing spondylitis(AS). Methods: A retrospective review of 133 AS patients with thoracolumbar kyphosis who underwent correction surgery between January 2002 and December 2015 with a minimum of 2-year follow-up performed, including 118 males and 15 females. The mean age was 35.1±9.8 years(range, 18-63 years). Clinical data and radiographic parameters were assessed for the risk factors of PJK. Clinical data included age, sex, body mass index, types of osteotomy and fusion levels. Radiographic parameters were measured preoperatively, postoperatively and at each follow-up including sagittal vertical axis(SVA), thoracolumbar kyphosis(TK), lumbar lordosis(LL), sacral slope(SS), pelvic tilt(PT) and proximal junctional angle(PJA). Patients were divided into PJK group and no PJK group according to the change in PJA. Comparison of clinical and radiographic data were performed between the 2 groups using Student′s t-test and Chi-square test. Results: Among the 133 patients with an average of 3.6±2.2 years(range, 2 to 15 years) follow-up, PJK was observed in 11 patients, at a mean of 3.7±4.1 years(range, 0.3 to 15 years) with PJA increased from 5.5°±9.7° preoperatively to 21.2°±9.5° at the time of PJK. Compared with the no PJK group, PJK group showed lower age(P<0.01) and higher proportion of patients undergoing SPO(P<0.01). Preoperative PJA and SVA were greater in the no PJK group(P<0.05). There were no significant differences in TK, LL, SS, PT and PI between the two groups(P＞0.05). There were 3 types of PJK, including pseudoarthrosis(n=2), compression fracture(n=3) and progressing thoracic kyphosis due to the natural history of AS(n=6). At final follow-up, spontaneous fusion of pseudoarthrosis was found in 1 of the 2 patients with PJK of pseudoarthrosis, and the other was lost follow-up. Revision surgery was performed in 1 of the patients with PJK of progressing thoracic kyphosis due to the natural history of AS. Conclusions: PJK could be presented in the characteristics of pseudoarthrosis, compression fracture and progressing thoracic kyphosis due to the natural history of AS. The age at initial surgery and types of osteotomy were the risk factors of PJK. The patients with lower preoperative PJA and SVA were found to have a higher risk of PJK.