Development and Validation of an Online Calculator to Predict Proximal Junctional Kyphosis After Adult Spinal Deformity Surgery Using Machine Learning.

OBJECTIVE: Although adult spinal deformity (ASD) surgery aims to restore and maintain alignment, proximal junctional kyphosis (PJK) may occur. While existing scoring systems predict PJK, they predominantly offer a generalized 3-tier risk classification, limiting their utility for nuanced treatment decisions. This study seeks to establish a personalized risk calculator for PJK, aiming to enhance treatment planning precision. METHODS: Patient data for ASD were sourced from the Korean spinal deformity database. PJK was defined a proximal junctional angle (PJA) of ≥ 20° at the final follow-up, or an increase in PJA of ≥ 10° compared to the preoperative values. Multivariable analysis was performed to identify independent variables. Subsequently, 5 machine learning models were created to predict individualized PJK risk post-ASD surgery. The most efficacious model was deployed as an online and interactive calculator. RESULTS: From a pool of 201 patients, 49 (24.4%) exhibited PJK during the follow-up period. Through multivariable analysis, postoperative PJA, body mass index, and deformity type emerged as independent predictors for PJK. When testing machine learning models using study results and previously reported variables as hyperparameters, the random forest model exhibited the highest accuracy, reaching 83%, with an area under the receiver operating characteristics curve of 0.76. This model has been launched as a freely accessible tool at: (https://snuspine.shinyapps.io/PJKafterASD/). CONCLUSION: An online calculator, founded on the random forest model, has been developed to gauge the risk of PJK following ASD surgery. This may be a useful clinical tool for surgeons, allowing them to better predict PJK probabilities and refine subsequent therapeutic strategies.

Anatomical feasibility for safe occipital condyle screw fixation.

PURPOSE: The occipital condyle (OC) screw can be a viable alternative option for the occipito-cervical fixation. However, the risk of vertebral artery (VA) injury during the direct OC screw fixation has not been adequately assessed. The purpose of this study was to establish the course of the VA (V3) relative to the nearby osseous structures to estimate the feasibility of OC screw fixation and describe its anatomical relationship depending on patient's age and sex. METHODS: A total of 387 three-dimensional computed tomographic angiograms (3D-CTA) were used and compared between two age groups. The vertebral artery diameter and two kinds of bony space were measured. The occipito-C1 arch space (O-C1S) and VA-occipital bone distance (VOD, six entry points) were measured on both sides. The feasibility of direct OC screw fixation can be represented by the VOD value; the minimum feasible value was determined to be 4 mm. Angular measurements (O-C1A and O-C2A) were also taken to assess their relationship to the bony space. RESULTS: The mean value of the O-C1S ranged from 9.0 to 9.9 mm. The mean value of the VOD ranged from 3.2 to 3.5 mm, and the proportion of individuals for which direct OC screw fixation was considered feasible ranged from 32 to 42 % in both age groups and there was no significant difference between two age groups. The VOD value was not affected by laterality or by gender (P > 0.05). The mean kyphosis of the O-C1A (-5° ± 5.2°, range -22° to 8°) was significantly smaller in the young age group compared to the older age group (-7.6° ± 5.3°, range -26° to 2°). The mean lordosis of the O-C2A (12.4° ± 6.4°, range 3°-33°) was significantly higher in the young age group compared to the older age group (10.4° ± 6.4°, range 0°-36°). CONCLUSIONS: The direct OC screw fixation was not possible in a considerable number of cases due to the limited space and the position of the VA regardless of age group. Only about one quarter (21-24 %) of all patients was affordable to place the direct OC screw bilaterally. There was more space available to place the direct OC screw when the angle of the upper cervical spine is more kyphotic.

Sequential alignment change of the cervical spine after anterior cervical discectomy and fusion in the lower cervical spine.

PURPOSE: The cervical spine has a linear chain of correlation or reciprocal relationship regionally (within the cervical spine) and globally (head to whole spine). The purpose of this study was to assess the sequential alignment change of the regional and global cervical spine after two-level anterior cervical discectomy and fusion (ACDF) performed on the lower cervical spine. METHODS: This study included 61 patients (mean age 56 ± 8.6 years; range 35-70 years) who underwent ACDF at C5-6-7 with a plate-cage construct and whose C-spine neutral lateral radiographs showed an identical degree of horizontal gaze (occipital slope) peri-operatively. We compared the change in cervical curvature from the occiput to C7 with the absolute value (slope angle) and relative value (between two different slopes). We also investigated the correlated change in multiple angular parameters according to the change in the occipital slope. RESULTS: The occipital slope was significantly correlated with the value of the C1-slope (r = 0.33) and C2- slope (r = 0.51). The value of the center of the sellar turcica-C7 sagittal vertical axis (St-SVA) was very closely related to the C1-slope (r = -0.83), C2-slope (r = -0.8), C2-7 angle (r = -0.43), and C2-5 angle (r = -0.46). The amount of angular change at the surgical level (C5-7A) was 5.8° (2.9° -> 8.5°), and the sum of the change in the C5-slope and C7-slope was 6° (3.1° + 2.9°). In general, the C2-5 angle decreased about 3°, in proportion to the upward inclination of C5-slope (3.1°), because the C2-slope was fixed. However, patients who showed improvement in cervical alignment (greatly increased C5-7 lordosis or greatly decreased St-SVA after surgery) often had upper cervical slope change (C1-s and C2-s). CONCLUSIONS: The ACDF procedure itself can induce regional slope change (C5-s and C7-s) directly at the surgical level and can also influence upper cervical slope change (C1-s and C2s) indirectly. Then the change in the upper cervical spine can induce a change in the St-CVA and spino-cranial angle (SCA).

Quantification of pediatric cervical spine growth at the cranio-vertebral junction.

OBJECTIVE: The purpose of this study was to investigate morphological change at the craniovertebral junction (CVJ) region using computed tomography. METHODS: A total of 238 patients were included in this study, and mean age was 47.8±21.3 months. Spinal canal diameter, Power's ratio, McRae line, antero-posterior C1 ring height, atlantoaxial joint space, C2 growth, epidural space from the dens (M-PB-C2) and longitudinal distance (basion to C2 lower margin, B-C2) were measured. The mean value of each parameter was assessed for individual age groups. The cohorts were then divided into three larger age groups : infancy (I) (≤2 years), very early (VE) childhood (2-5 years) and early (E) childhood (5≥ years). RESULTS: Spinal canal diameter increased with age; however, this value did not increase with statistical significance after VE age. A significant age-related difference was found for all C2 body and odontoid parameters (p<0.05). Mean McRae line was 8.5, 8, and 7.5 mm in the I, VE, and E groups, respectively. The M-PB-C2 line showed up-and-down dynamic change during early pediatric periods. CONCLUSION: Expansion of the spinal canal was restricted to the very early childhood period (less than 5 years) in the CVJ region; however, the C2 body and odontoid process increased continuously with age. The above results induced a dynamic change in the M-PB-C2 line. Although C2 longitudinal growth continued with age, the McRae line showed relatively little change.