Study design: Retrospective case series of surgically treated patients with adult spine deformity (ASD).
Objective: To report the incidence of proximal junctional failure (PJF), characterize PJF and evaluate the outcome of revision surgery for PJF. A modified classification is also proposed.
Summary of background data: Although recent reports have shown the catastrophic results of PJF, few reports have shown the incidence, characteristics, and clinical outcomes of PJF in ASD.
Methods: This retrospective analysis reviewed data entered prospectively into a multicenter database. Surgically treated patients with ASD with a minimum 2-year follow-up were included. PJF was defined as any type of symptomatic proximal junctional kyphosis (PJK) requiring surgery. On the basis of our previous classification, the following modified PJK classification was established: grade A, proximal junctional increase of 10° to 19°; grade B, 20° to 29°; and grade C, 30° or more. Three types of PJK were also defined: ligamentous failure (type 1), bone failure (type 2), and implant/bone interface failure (type 3). An additional criterion was added for the presence or absence of spondylolisthesis above the upper instrumentation vertebra (UIV).
Results: PJF developed in 23 of the 1668 patients with ASD. The incidence of PJF was 1.4%. The mean age was 62.3 ± 7.9 years, and the mean follow-up was 4.0 ± 2.3 years. Seventeen patients had undergone prior surgical procedures. Six patients had UIV above T8, and 17 had UIV below T9. Six patients had associated spondylolisthesis above the UIV (PJF-S), whereas 17 patients did not (PJF-N). The radiographical data show a significant difference in the preoperative sagittal vertical axis between the PJF-S and PJF-N groups, whereas no significant difference was observed in the preoperative sagittal parameters (5.2 ± 3.9 cm vs. 11.4 ± 6.0 cm, P = 0.04). The most common type of PJF was type 2N. The PJF symptoms consisted of intolerable pain (n = 17), neurological deficits (n = 6), and progressive trunk deformity (n = 1). Eleven patients had additional PJK/PJF and 9 required additional revision surgical procedures.
Conclusion: The incidence of PJF among surgically treated patients with ASD was 1.4%. The most common type of PJF was 2N. Preoperative large sagittal vertical axis change and large amount of correction was a causative factor for spondylolisthesis above the UIV. After the revision surgery, further PJF was a commonly occurred event.