Proximal junctional kyphosis (PJK) is a recognized postsurgical phenomenon after spinal fusion in both the treatment of adolescent [1], [2], [3] and adult [1], [4] idiopathic scoliosis, as well as after long-segment instrumented fusion in treating adult spinal deformity (ASD). Proximal junctional kyphosis occurs at the rostral junction between the fixed instrumented spinal segment and the mobile vertebral levels above. The prevalence of PJK varies greatly, owed in part to the disagreement in minimum kyphotic angulation that meets the criteria for PJK [5], [6]. One recent study by Yagi et al. defined PJK as greater than 10° of kyphosis between the lower end plate of the uppermost instrumented vertebra (UIV) and the upper end plate of the adjacent vertebral segment above (UIV+1) [6]. Junctional stress resulting in mild ventral wedging of the vertebral body, loss of disc height, and disruption of the posterior tension band all contribute to PJK. Postoperative refractory pain and loss of correction resulting in reoperation is often termed proximal junctional failure (PJF) [7]. A prevalence of 17% to 40% has been reported for PJK, owing to minor variations in definition of PJK and study design [1], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18]. In one large study, Kim et al. found a 39% prevalence of PJK across a 7-year duration, although the vast majority of all cases occur early in the first year [4].
The risk factors for PJF identified in prior observational cohort studies include old age, increased body mass index, and osteoporosis [4], [7], [19], [20]. More recently, radiographic features have been the primary focus: large corrections in the sagittal vertical axis (SVA), increased preoperative and postoperative thoracic kyphosis (TK), increased postoperative lumbar lordosis (LL) [13], and larger corrections in the coronal plane are part of a growing list of risk factors for PJK. Surgical techniques linked to a stiffer construct have been associated with a higher PJK rate: fusion to the sacrum, pedicle screw use, combined anteroposterior (AP) fusion, and pedicle screw use at the UIV [13]. Anteroposterior fusion has been observed as a greater risk factor at the proximal end of the construct, and is likely a result of the increased difference in stiffness between the screw-rod construct and the rostral adjacent vertebra (UIV+1), resulting in larger junctional forces [4].
Importantly, PJK may not achieve clinical relevance and go untreated. Loss of correction or development of a compression fracture resulting in surgical therapy is a PJF and treated by proximal extension of the instrumented fusion [4], [12], [14], [21], [22], [23], [24]. Lengthening of the fusion construct either above the apex of the TK or to the upper thoracic spine proximally [25], the use of hooks at the UIV [26], selection of a proximal vertebral level with a small UIV angle [27], and using preoperative imaging to determine a stable vertebral level are sound methods to lower the incidence of PJK [28].
One surgical technique that may mitigate the development of PJK and PJF is the use of prophylactic vertebroplasty (PV) with polymethylmethacrylate (PMMA) [29], [30], [31], [32]. The placement of PMMA in the UIV increases pedicle screw pullout strength and increases the overall construct stiffness [33]. However, fusion alone in a long-segment construct has been thought to increase stress at the UIV+1. Cement augmentation at the UIV alone increases the stiffness of the proximal construct even further, further loading the UIV+1 [7]. This can increase the risk of PJK or compression fractures at the UIV+1 or more rostral levels [34]. In contrast, PV at the UIV and UIV+1 may buffer axial compression forces and protect the UIV+1 from fracture by spreading adjacent-level forces over an additional segment, thereby stress-shielding the UIV+1 and intervening disc [7].
Moreover, recent biomechanical evidence confirms that PV at the UIV alone places the UIV+1 at a higher risk of compression fracture by redundantly reinforcing the already rigid and immobile UIV level, contributing to additional fracture risk [7]. Another biomechanical study demonstrates the biomechanical advantage to PV at the UIV and UIV+1 [7]. Two-level PV was shown to be 17% more effective, observing significantly fewer fractures [7].
In prior clinical practice, Theologis and Burch evaluated cement augmentation at the UIV and UIV+1 and showed that PV lowered the rate of revision surgery for PJK compared with controls [35]. Similarly, Hart et al. found a 15% reduction of PJF compared with patients not treated with PV [23]. More recently, Martin et al. found that patients treated with two-level PV at the UIV and UIV+1 developed PJK and PJF in 8% and 5% of respective cases, significantly lower than historical controls.
The authors provide a retrospective cohort analysis of patients treated with PV at the UIV and UIV+1 in long-segment spinal fusions for ASD compared with controls.