Todderud, Julia E

Background: Scoliosis has various etiologies, ranging from idiopathic, congenital, to those with spinal cord abnormalities. Advanced imaging is not always feasible and limited by cost, low-yield, and anesthesia risks in children. Given artificial intelligence's potential, we hypothesize that a deep learning (DL)–based image classifier will demonstrate superior performance in radiographic classification of scoliosis etiology compared witht experienced spine surgeons based on spine radiographs. Methods: One thousand thirty-six pediatric patients from single institution with scoliosis diagnosis and paired anterior-posterior - lateral images were included. Patients were manually classified based on their scoliosis etiology from chart review, including previous spine imaging studies when available. Categories included idiopathic, congenital, and spinal cord pathology. Images were randomized and assigned for training, validation, and testing. A DL-classifier using EfficientNet B4 architecture was trained on the radiographs. Accuracy and positive predictive value as defined by precision, recall, and F1-score were calculated to assess final performance metrics. Results: The trained classifier performed well at identifying correct etiologies, with F1-Score (harmonic mean of precision-recall) of 0.97. Model precision was 0.99 for adolescent idiopathic scoliosis, 0.89 for congenital, and 0.78 for spinal cord pathology. Performance was higher on more common classes, with lower performance observed in class with fewer images. The algorithm has highest overall precision (0.96), recall (0.96), and F1 (0.96), while Surgeons 1 and 2 have lower accuracies (precision 0.80, recall 0.79, and F1 0.79 and precision 0.76, recall 0.67, and F10.71, respectively). There was no clear pattern identified regarding the surgeons' errors and poor agreement on which X-rays corresponded to underlying spinal cord pathology. Conclusion: A DL-convolutional-neural-network classifier has been trained to high degree of accuracy to distinguish between 3 scoliosis etiologies on pediatric spine radiographs. This model could provide a novel tool to clinicians to decide when to refer for axial imaging. It had superior diagnostic performance compared with experienced spine surgeons. After further validation and refinement, a clinician could put a patient's X-rays into the predictive tool for analysis and use the results to discuss with the family the risk of finding spinal cord pathology, helping them determine together the best timing for an MRI. Level of Evidence: Level III . See Instructions for Authors for a complete description of levels of evidence.

Background: The purpose of this study was to report on 2-year results of vertebral body tethering (VBT), performed under a Food and Drug Administration protocol, to obtain insight into outcomes and complications. Methods: Forty prospectively enrolled patients with adolescent idiopathic scoliosis (AIS) who had a Sanders score of ≤4 or a Risser score of ≤2 underwent VBT for curves between 40° and 70°. Surgical, radiographic, and patient-reported outcomes were reviewed at a minimum 2-year follow-up. Results: Mean age at surgery was 13 (range, 10 to 16) years. The 40 patients were 90% female; 95% White, 2.5% other, and 2.5% unreported; and 92.5% non-Hispanic, 5% Hispanic, and 2.5% unreported. A mean of 8 (range, 5 to 12) levels were instrumented. Most patients were at Sanders 4 (65%) and Risser 0 (63%). Mean length of stay was 3 ± 1 days, estimated blood loss was 236 ± 158 (range, 25 to 740) mL, and operative time was 4.4 ± 1.4 hours. Mean correction of the major curve was 44% (range, 22% to 95%) on the 3-month standing radiograph, 49% at 1 year, and 46% (range, −10% to 93%) at 2 years. The mean major Cobb angle improved from 51° ± 8° (range, 40° to 70°) preoperatively to 27° ± 11° (range, 3° to 56°) at 2 years. Success at 2 years, defined by a Cobb angle of <35° and no reoperation, was seen in 30 patients (75%) and was associated with a mean Cobb angle of <35° on the first postoperative standing radiograph (p < 0.001). Twelve patients (30%) demonstrated improvement in the curve with growth. By 2 years, 2 (5%) of the patients underwent repeat surgery (1 release for overcorrection, 1 lumbar VBT for lumbar curve progression after thoracic VBT). The Scoliosis Research Society (SRS) satisfaction score improved 2 years following surgery (p < 0.001), but other SRS domains only remained stable over time. Beyond 2 years, 1 additional lumbar tether was required after thoracic VBT, 1 implant was removed, and 3 fusions were performed, for a 10% fusion rate and overall 20% reoperation rate at a mean of 3.8 ± 1.1 years of follow-up. The rate of cord breakage in the study population was 20%. Conclusions: In skeletally immature patients treated in the U.S. under a prospective Investigational Device Exemption, there was a 75% rate of successful outcomes at 2 years. Most correction was obtained at the time of surgery, and inadequate intraoperative curve correction was associated with a higher Cobb angle on the first postoperative standing radiograph and failure by 2 years. Level of Evidence: Therapeutic Level II . See Instructions for Authors for a complete description of levels of evidence.

Study Design Prospective Cohort Study. Objective The present study aims to determine if the racial representation of patients enrolled in a large prospective scoliosis registry is reflective of the general United States population. Further, we studied whether there was an association between race, pre-operative parameters, outcomes and loss to follow-up. Methods Prospectively collected data for patients who underwent spinal fusion for adolescent idiopathic scoliosis (AIS) was reviewed, including self-reported race/ethnicity. The U.S. pediatric population and U.S. patients enrolled in the prospective registry were compared. The data obtained was analyzed for variations between races, for pre-operative variables and follow-up. Results Of the 2210 included patients in the registry 66% of patients reported as White, while 52% of the 2018 U.S. pediatric population reported as White. 15% of the registry reported as Hispanic/Latino compared to 22% of the U.S. pediatric population, 13% Black compared to 14% of the U.S. pediatric population, and 4% Asian compared to 5% of the U.S. pediatric population. Asian and White patients had statistically significant higher 2-year follow-up in all but one of six enrollment sites ( P < 0.001). Native American, Other, and Hispanic/Latino patients had the highest BMIs. Native American and Black patients had the highest pre-op thoracic Cobb angles. Pre-op ages of Black, Hispanic, and Native American patients were statistically lower ( P < 0.01). Conclusion This study demonstrates the association between race and patient follow-up and pre-operative factors in patients who underwent surgery for AIS. Black, Native American, and Hispanic populations were underrepresented both at pre-op and follow-up when compared to their relative proportion in the U.S. pediatric population.

Background: Motion-sparing scoliosis surgeries such as the posterior dynamic distraction device (PDDD) are slowly increasing in use. However, there is limited clinical data documenting postoperative motion across the PDDD construct. With this cohort study, we aim to measure sagittal and coronal motion following PDDD. We hypothesize coronal and sagittal spinal motion will be partially preserved across the construct. Methods: Retrospective review of prospectively collected data. Preoperative and minimum 1-year postoperative coronal range of motion across the instrumented levels was compared. Available flexion/extension radiographs were evaluated postoperatively to assess sagittal arc of motion. Radiographs from latest follow-up were used. Results: At a mean of 1.9 years (1 to 5 y), flexibility radiographs were available on 29 patients treated with PDDD (17 thoracic, 12 lumbar). Mean age at surgery was 16 years (12 to 25). Postoperative coronal arc of motion in PDDD patients was 11 degrees (3 to 19 degrees) in the thoracic spine and 10 degrees (0 to 28 degrees) in the lumbar spine. Compared with preoperative motion, the thoracic arc of motion was maintained by 33% (35 to 11 degrees) and lumbar motion was maintained by 30% (34 to 10 degrees). Flexion-extension radiographs were available on 7 patients. Sagittal arc for the upper instrumented vertebral end plate to the lower instrumented vertebral endplate of the cohort was 10 degrees in the thoracic spine (6 to 18) and 14 degrees in the lumbar spine (5 to 21). Sagittal measurements for the changes in the arc of the upper and lower screws on the construct were 4 degrees in the thoracic group (2 to 8) and 9 degrees in the lumbar group (2 to 17). By latest follow-up 11 patients (38%) underwent reoperation, with most cases due to implant breakage (N=4, 14%), extender misalignment (N=2, 7%), and screw misplacement (N=2, 7%). Conclusion: At mean 1.9 years postoperatively, PDDD preserves measurable spinal motion over the construct both in the coronal and the sagittal plane without evidence for autofusion. Coronal arc of motion averages 10 to 12 degrees and sagittal arc of motion ranged from 4 to 14 degrees, although this varies by patient. This study confirms that PDDD for pediatric scoliosis preserves a measurable degree of postoperative flexibility both in the sagittal and coronal planes. Level of Evidence: Level II—therapeutic study.