Deep Learning Model to Classify and Monitor Idiopathic Scoliosis in Adolescents Using a Single Smartphone Photograph.

Importance: Adolescent idiopathic scoliosis (AIS) is the most common pediatric spinal disorder. Routine physical examinations by trained personnel are critical to diagnose severity and monitor curve progression in AIS. In the presence of concerning malformation, radiographs are necessary for diagnosis or follow-up, guiding further management, such as bracing correction for moderate malformation and spine surgery for severe malformation. If left unattended, progressive deterioration occurs in two-thirds of patients, leading to significant health concerns for growing children. Objective: To assess the ability of an open platform application (app) using a validated deep learning model to classify AIS severity and curve type, as well as identify progression. Design, Setting, and Participants: This diagnostic study was performed with data from radiographs and smartphone photographs of the backs of adolescent patients at spine clinics. The ScolioNets deep learning model was developed and validated in a prospective training cohort, then incorporated and tested in the AlignProCARE open platform app in 2022. Ground truths (GTs) included severity, curve type, and progression as manually annotated by 2 experienced spine specialists based on the radiographic examinations of the participants' spines. The GTs and app results were blindly compared with another 2 spine surgeons' assessments of unclothed back appearance. Data were analyzed from October 2022 to February 2023. Exposure: Acquisitions of unclothed back photographs using a mobile app. Main Outcomes and Measures: Outcomes of interest were classification of AIS severity and progression. Quantitative statistical analyses were performed to assess the performance of the deep learning model in classifying the deformity as well as in distinguishing progression during 6-month follow-up. Results: The training data set consisted of 1780 patients (1295 [72.8%] female; mean [SD] age, 14.3 [3.3] years), and the prospective testing data sets consisted of 378 patients (279 [73.8%] female; mean [SD] age, 14.3 [3.8] years) and 376 follow-ups (294 [78.2%] female; mean [SD] age, 15.6 [2.9] years). The model recommended follow-up with an area under receiver operating characteristic curve (AUC) of 0.839 (95% CI, 0.789-0.882) and considering surgery with an AUC of 0.902 (95% CI, 0.859-0.936), while showing good ability to distinguish among thoracic (AUC, 0.777 [95% CI, 0.745-0.808]), thoracolumbar or lumbar (AUC, 0.760 [95% CI, 0.727-0.791]), or mixed (AUC, 0.860 [95% CI, 0.834-0.887]) curve types. For follow-ups, the model distinguished participants with or without curve progression with an AUC of 0.757 (95% CI, 0.630-0.858). Compared with both surgeons, the model could recognize severities and curve types with a higher sensitivity (eg, sensitivity for recommending follow-up: model, 84.88% [95% CI, 75.54%-91.70%]; senior surgeon, 44.19%; junior surgeon, 62.79%) and negative predictive values (NPVs; eg, NPV for recommending follow-up: model, 89.22% [95% CI, 84.25%-93.70%]; senior surgeon, 71.76%; junior surgeon, 79.35%). For distinguishing curve progression, the sensitivity and NPV were comparable with the senior surgeons (sensitivity, 63.33% [95% CI, 43.86%-80.87%] vs 77.42%; NPV, 68.57% [95% CI, 56.78%-78.37%] vs 72.00%). The junior surgeon reported an inability to identify curve types and progression by observing the unclothed back alone. Conclusions: This diagnostic study of adolescent patients screened for AIS found that the deep learning app had the potential for out-of-hospital accessible and radiation-free management of children with scoliosis, with comparable performance as spine surgeons experienced in AIS management.

Supervised and unsupervised learning to classify scoliosis and healthy subjects based on non-invasive rasterstereography analysis.

The aim of our study was to classify scoliosis compared to to healthy patients using non-invasive surface acquisition via Video-raster-stereography, without prior knowledge of radiographic data. Data acquisitions were made using Rasterstereography; unsupervised learning was adopted for clustering and supervised learning was used for prediction model Support Vector Machine and Deep Network architectures were compared. A M-fold cross validation procedure was performed to evaluate the results. The accuracy and balanced accuracy of the best supervised model were close to 85%. Classification rates by class were measured using the confusion matrix, giving a low percentage of unclassified patients. Rasterstereography has turned out to be a good tool to distinguish subject with scoliosis from healthy patients limiting the exposure to unnecessary radiations.

Does the Use of Sanders Staging and Distal Radius and Ulna Classification Avoid Mismatches in Growth Assessment with Risser Staging Alone?

BACKGROUND: Although Risser stages are visible on the same radiograph of the spine, Risser staging is criticized for its insensitivity in estimating the remaining growth potential and its weak correlation with curve progression in patients with adolescent idiopathic scoliosis. Risser staging is frequently accompanied by other skeletal maturity indices to increase its precision for assessing pubertal growth. However, it remains unknown whether there is any discrepancy between various maturity parameters and the extent of this discrepancy when these indices are used concurrently to assess pubertal growth landmarks, which are important for the timing of brace initiation and weaning. QUESTIONS/PURPOSES: (1) What is the chronologic order of skeletal maturity grades based on the growth rate and curve progression rate in patients with adolescent idiopathic scoliosis? (2) What are the discrepancies among the grades of each maturity index for indicating the peak growth and start of the growth plateau, and how do these indices correspond to each other? (3) What is the effectiveness of Risser staging, Sanders staging, and the distal radius and ulna classification in assessing peak growth and the beginning of the growth plateau? METHODS: Between 2014 and 2017, a total of 13,536 patients diagnosed with adolescent idiopathic scoliosis were treated at our tertiary clinic. Of those, 3864 patients with a radiograph of the left hand and wrist and a posteroanterior radiograph of the spine at the same visits including initial presentation were considered potentially eligible for this study. Minimum follow-up was defined as 6 months from the first visit, and the follow-up duration was defined as 2 years since initial consultation. In all, 48% (1867 of 3864) of patients were eligible, of which 26% (485 of 1867) were excluded because they were prescribed bracing at the first consultation. These patients visited the subsequent clinics wearing the brace, which might have affected body height measurement. Six percent (117 of 1867) of eligible patients were also excluded as their major coronal Cobb angle reached the surgical threshold of 50° and had undergone surgery before skeletal maturity. Another 21% (387 of 1867) of patients were lost before minimum follow-up or had incomplete data, leaving 47% (878) for analysis. These 878 patients with 1139 skeletal maturity assessments were studied; 74% (648 of 878) were girls. Standing body height was measured in a standardized manner by a wall-mounted stadiometer. Several surgeons measured curve magnitude as per routine clinical consultation, skeletal maturity was measured according to the distal radius and ulna classification, and two raters measured Risser and Sanders stages. Reliability tests were performed with satisfaction. Data were collected for the included patients at multiple points when skeletal maturity was assessed, and only up to when brace wear started for those who eventually had bracing. The growth rate and curve progression rate were calculated by the change of body height and major coronal Cobb angle over the number of months elapsed between the initial visit and next follow-up. At each skeletal maturity grading, we examined the growth rate (in centimeters per month) and curve progression rate (in degrees per month) since the skeletal maturity assessment, as well as the mean age at which this maturity grading occurred. Each patient was then individually assessed for whether he or she was experiencing peak growth and the beginning of growth plateau at each timepoint by comparing the calculated growth rate with the previously defined peak growth rate of ≥ 0.7 cm per month and the beginning of growth plateau rate of ≤ 0.15 cm per month in this adolescent idiopathic scoliosis population. Among the timepoints at which the peak growth and the beginning of growth plateau occurred, the median maturity grade of each maturity index was identified as the benchmark grade for comparison between indices. We used the McNemar test to investigate whether pubertal growth landmarks were identified by specific maturity grades concurrently. We assessed the effectiveness of these skeletal maturity indices by the difference in proportions (%) between two benchmark grades in indicating peak growth and the growth plateau. RESULTS: For girls, the chronological order of maturity grades that indicated peak growth was the radius grade, ulna grade, Sanders stage, and Risser stage. Curve progression peaked between the age of 11.6 and 12.1 years at a similar timing by all maturity indices for girls but was inconsistent for boys. For both sexes, radius (R) grade 6, ulna (U) grade 5, Sanders stage (SS) 3, and Risser stage 0+ were the median grades for peak growth, whereas Risser stage 4, R8/9, U7/8, and SS6/7 indicated the beginning of the growth plateau. The largest discrepancy between maturity indices was represented by Risser stage 0+, which corresponded to six grades of the Sanders staging system (SS2 to SS7) and to R6 in only 41% (62 of 152) of girls in the whole cohort. Despite Risser stage 0+ corresponding to the wide range of Sanders and distal radius and ulna grades, none of the R6, U5, SS3, and Risser stage 0+ was found more effective than another grade in indicating the peak growth in girls. R6 most effectively indicated the peak growth in boys, and Risser stage 0+ was the least effective. For the beginning of the growth plateau in girls, SS6/7 was the most effective indicator, followed by U7/8. Risser stage 4 was the least effective because it indicated 29% (95% CI 21% to 36%; p < 0.001) fewer patients who reached the beginning of the growth plateau than did those with R8/9. Risser stage 4 also indicated 36% (95% CI 28% to 43%; p < 0.001) fewer patients who reached the beginning of the growth plateau than those indicated by U7/8, and it identified 39% fewer patients than SS6/7 (95% CI 32% to 47%; p < 0.001). For boys, similarly, R8/9, U7/8, and SS6/7 were all more effective than Risser stage 4 in identifying when the growth plateau began. CONCLUSION: Risser stage 0+ corresponds to a wide range of Sanders and distal radius and ulna grades. Risser stage 0+ is least effective in indicating the peak growth in boys, and Risser stage 4 is the least effective maturity grade for indicating when the growth plateau starts in both sexes. The concurrent use of R6 and SS3 can be useful for detecting the peak growth, and SS6/7 in conjunction with U7/8 is most effective in indicating the beginning of the growth plateau. Using a combination of specific grades of Sanders staging and the distal radius and ulna classification can indicate pubertal growth landmarks with reduced risk of underestimating or overestimating skeletal maturity. These findings may aid in refining clinical decision-making of brace initiation and weaning at a more precise timing. Among Risser stage 0, the appearance of R6, U5, and SS3 provide the most effective assessment of peak growth that can indicate the most effective bracing period within which curve progression occurs. For initiation of the growth plateau, Risser 4 is not useful, and SS6/7, R8/9 and U7/8 should be used instead. LEVEL OF EVIDENCE: Level III, diagnostic study.

Computerized image understanding system for reliable estimation of spinal curvature in idiopathic scoliosis.

Analysis of scoliosis requires thorough radiographic evaluation by spinal curvature estimation to completely assess the spinal deformity. Spinal curvature estimation gives orthopaedic surgeons an idea of severity of spinal deformity for therapeutic purposes. Manual intervention has always been an issue to ensure accuracy and repeatability. Computer assisted systems are semi-automatic and is still influenced by surgeon's expertise. Spinal curvature estimation completely relies on accurate identification of required end vertebrae like superior end-vertebra, inferior end-vertebra and apical vertebra. In the present work, automatic extraction of spinal information central sacral line and medial axis by computerized image understanding system has been proposed. The inter-observer variability in the anatomical landmark identification is quantified using Kappa statistic. The resultant Kappa value computed between proposed algorithm and observer lies in the range 0.7 and 0.9, which shows good accuracy. Identification of the required end vertebra is automated by the extracted spinal information. Difference in inter and intra-observer variability for the state of the art computer assisted and proposed system are quantified in terms of mean absolute difference for the various types (Type-I, Type-II, Type-III, Type-IV, and Type-V) of scoliosis.

Effect of Etiology, Radiographic Severity, and Comorbidities on Baseline Parent-Reported Health Measures for Children with Early-Onset Scoliosis.

BACKGROUND: The Classification of Early-Onset Scoliosis (C-EOS) allows providers to differentiate patients, for clinical and research purposes, on the basis of the etiology of their disease as well as radiographic parameters. The Early Onset Scoliosis Questionnaire (EOSQ) is the first disease-specific, parent-reported HRQOL (health-related quality-of-life) outcome measure for this condition. We sought to determine the influence of the C-EOS etiology designation, radiographic parameters, and medical comorbidities on EOSQ scores to differentiate quality of life in this heterogeneous patient population. We hypothesized that baseline EOSQ scores for patients with EOS would be strongly affected by the C-EOS etiology designation. METHODS: The analysis included prospectively enrolled patients with EOSQ scores recorded in a multicenter EOS database prior to intervention for the EOS. EOSQ scores were compared across C-EOS etiologies, severity of disease based on radiographic measurements, and patient comorbidities prior to scoliosis intervention. RESULTS: Six hundred and ten patients with EOS were available for analysis; 119 had congenital, 201 had idiopathic, 156 had neuromuscular, and 134 had syndromic EOS. In multivariate analysis, neuromuscular and syndromic etiologies were associated with lower scores than congenital and idiopathic etiologies in many EOSQ domains including general health, transfer, daily living, fatigue/energy level, and emotion. Patients with neuromuscular EOS had the lowest EOSQ scores in general. Congenital and idiopathic EOS did not differ from each other in any EOSQ domain. Coronal Cobb and kyphosis angles had significant inverse but generally weak correlations with EOSQ domains. Individual medical comorbidities had a minor effect on certain domains while American Society of Anesthesiologists (ASA) class and total number of comorbidities had inverse correlations with most domains. CONCLUSIONS: The underlying etiology of EOS appears to have a significant influence on the parent-reported HRQOL outcomes of the disease. Specifically, syndromic and neuromuscular C-EOS diagnoses are associated with lower EOSQ scores before treatment compared with congenital and idiopathic diagnoses. Radiographic measurements of severity have a relatively small influence on EOSQ scores. These baseline differences in C-EOS-designated etiology should be accounted for in studies comparing outcomes of treatment for this heterogeneous patient population. LEVEL OF EVIDENCE: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.

Low Preoperative Lung Functions in Children With Early Onset Scoliosis Predict Postoperative Length of Stay.

INTRODUCTION: Children with early onset scoliosis (EOS) undergoing spine surgery often have significant respiratory disease. Preoperative risk assessments that predict an increased length of hospital stay (LOS) for this group have not been previously evaluated. METHODS: A voluntary protocol using preoperative lung function studies began among participants of a multicenter registry in 2016. Preoperative assessments were standardized to include spirometry, blood hemoglobin levels, serum bicarbonate, albumin and prealbumin; radiographic parameters of the spine, C-EOS classification and need for preoperative pulmonary assistance before initial growth friendly device insertion or "definitive" spine fusion. Primary outcome was LOS postoperatively. Data, including age, diagnosis, and type of surgery, was collected prospectively. Secondary outcomes measured included intensive care unit LOS, requirement for new pulmonary assistance on discharge, and pulmonary complications. Groups were compared using the Fisher exact tests. RESULTS: Of 525 children enrolled, 101 (20%) had preoperative spirometry. Median age was 8.9 years [interquartile range (IQR): 4.27]. Etiologies for EOS included 29 neuromuscular (28%), 33 idiopathic (32%), 19 syndromic (19%), and 22 congenital (21%) scoliosis. Eighty (78%) had growing rod (GR) insertions; 23 (22%) had spine fusion SF. Eighteen subjects (17%) were hospitalized ≥7 days (median=9 d); 83 had a LOS <7 days (median=3 d). Percentage of forced vital capacity (FVC%) predicted was inversely associated with LOS ≥7 days with a median of 75.3% (IQR: 41.7) for LOS <7 days and 51.7% (IQR: 41.6) (P=0.02). There were no detectable differences in LOS for other preoperative values. CONCLUSION: FVC predicted ≤50% preoperatively in children undergoing initial growth friendly rod insertion or definitive fusion after growth friendly treatment is associated with an increased risk of postoperative hospital stays ≥7 days. As demonstrated in previous studies, severe restrictive lung disease (FVC% predicted at or below 50%) is associated with increased risk of poorer outcomes for EOS patients.

Diagnostic yield and clinical impact of exome sequencing in early-onset scoliosis (EOS).

BACKGROUND: Early-onset scoliosis (EOS), defined by an onset age of scoliosis less than 10 years, conveys significant health risk to affected children. Identification of the molecular aetiology underlying patients with EOS could provide valuable information for both clinical management and prenatal screening. METHODS: In this study, we consecutively recruited a cohort of 447 Chinese patients with operative EOS. We performed exome sequencing (ES) screening on these individuals and their available family members (totaling 670 subjects). Another cohort of 13 patients with idiopathic early-onset scoliosis (IEOS) from the USA who underwent ES was also recruited. RESULTS: After ES data processing and variant interpretation, we detected molecular diagnostic variants in 92 out of 447 (20.6%) Chinese patients with EOS, including 8 patients with molecular confirmation of their clinical diagnosis and 84 patients with molecular diagnoses of previously unrecognised diseases underlying scoliosis. One out of 13 patients with IEOS from the US cohort was molecularly diagnosed. The age at presentation, the number of organ systems involved and the Cobb angle were the three top features predictive of a molecular diagnosis. CONCLUSION: ES enabled the molecular diagnosis/classification of patients with EOS. Specific clinical features/feature pairs are able to indicate the likelihood of gaining a molecular diagnosis through ES.

A New Global Spinal Balance Classification Based on Individual Pelvic Anatomical Measurements in Patients With Adult Spinal Deformity.

STUDY DESIGN: A retrospective study. OBJECTIVE: This study aims to establish a new, easy-to-implement global spinal balance (GSB) classification based on individual pelvic anatomical measurements in adult spinal deformity patients, and to validate this classification in terms of patients' quality of life (QOL). SUMMARY OF BACKGROUND DATA: Coronal (central sacral vertical line) and sagittal (C7 sagittal vertical axis [C7 SVA]) lines are standard parameters to evaluate GSB. However, these parameters do not consider individual differences in height and physique. Therefore, a new classification to more simply evaluate GSB related to QOL would be a beneficial tool. METHODS: This study included 279 patients with adult spinal deformity. The enrollment criteria included age >18 years and scoliosis >30°. GSB was evaluated based on C7 coronal vertical axis and C7 SVA, and was graded with our sagittal (Grades 1-3) and coronal classifications (Grades 1-3), where larger values represented worse GSB. In addition, total global balance classification (Grades 1-3) was also established based on both coronal and sagittal classifications. Patient self-assessment measures of health status were collected using the Scoliosis Research Society patient questionnaire (SRS-22). The relationships between responses to the SRS-22 and GSB were evaluated using partial correlation coefficients, adjusted by age. RESULTS: A newly described coronal classification correlated with the preoperative function (P < 0.02), pain (P = 0.001), subtotal (P < 0.001) domains in the SRS-22, whereas a newly described sagittal classification was correlated with the preoperative subtotal domain (P < 0.001). There were also significant correlations between total global balance and the preoperative function (P = 0.03) and subtotal domains (P < 0.001) in the SRS-22. CONCLUSION: We present a simple new classification for GSB that strongly correlates with QOL in patients with adult spinal deformity, thereby translating clinical data into meaningful patient outcomes and livelihoods.Level of Evidence: 4.

Postoperative T1 tilt is a risk factor for postoperative distal adding-on in Lenke type 1 adolescent idiopathic scoliosis: A preliminary report.

Retrospective comparable study.Postoperative loss of correction, which is referred to as the distal adding-on phenomenon, sometimes occurs during the postoperative course in Lenke type 1 adolescent idiopathic scoliosis (AIS). Selection of the lowest instrumented vertebra (LIV) has been reported to be one of the significant factors for preventing distal adding-on. However, proximal parameters, such as the Cobb angle of the proximal thoracic (PT) curve, radiographic shoulder height, and T1 tilt, were rarely described in previous reports. This study aimed to identify the risk factors for postoperative distal adding-on, including proximal radiographic parameters, in Lenke type 1 AIS.Preoperative and postoperative radiographs of 34 consecutive patients with Lenke type 1 curve who underwent selective thoracic fusion were analyzed. The patients were divided into an adding-on group and a no-adding-on group according to the presence of adding-on at a 2-year follow-up. The 2 groups were compared with regard to age at surgery, Lenke lumbar modifier, Risser grade, instrumentation type, and radiographic parameters.Distal adding-on was noted in 10 patients (29%). The adding-on group had significant variables including preoperative larger PT Cobb angle (P = .002), larger main thoracic (MT) flexibility (P = .006), smaller thoracolumbar (TL) Cobb angle (P = .012), larger LIV shift (P < .001), larger T1 tilt (P = .001), postoperative larger PT Cobb angle (P = .012), smaller MT Cobb angle (P = .016), smaller TL Cobb angle (P < .001), larger PT-MT mismatch (P < .001), larger LIV shift (P = .026), and larger T1 tilt (P = .006) when compared with the findings in the no-adding-on group. Postoperative T1 tilt was significantly correlated with PT-MT mismatch.Our findings suggest that not only the LIV but also proximal parameters, including T1 tilt and PT-MT mismatch, are associated with postoperative distal adding-on in Lenke type 1 AIS. Strategies to reduce postoperative T1 tilt and PT-MT mismatch are required to prevent distal adding-on.

Does Lateral Lumbar Interbody Fusion Decrease the Grading of Lenke-Silva Classification and Determine the Optimal Fusion Level in Severe Adult Degenerative Scoliosis?

OBJECTIVE: To testify whether lateral lumbar interbody fusion (LLIF) decreases the Lenke-Silva classification grading and determines the optimal fusion level in adult degenerative scoliosis (ADS) of global imbalance. METHODS: Thirty-seven patients with ADS of level V and VI based on Lenke-Silva classification were included. After the first-stage LLIF, patients received reassessment and were divided into group A (the Lenke-Silva classification grading changed) and group B (the grading was unchanged). Posterior fixation was performed according to the reassessment. The demographic, operative, radiographic, and clinical data were compared between the 2 groups. RESULTS: Twenty-five patients of level V and 12 patients of level VI were included, with a mean follow-up of 29.6 months. After first-stage LLIF, the Lenke-Silva classification grading changed in 22 patients (group A), with the remaining 15 patients unchanged (group B). There were significant differences in preoperative Lenke-Silva classification grading, use of anterior column realignment and hyperlordotic cage and high-grade cage subsidence between the 2 groups. The posterior fusion levels in the second surgery were less and the rate of fusion to thoracic spine region was lower in group A. The visual analog scale and Oswestry Disability Index were significantly improved and restorations of coronal and sagittal balance were found at the latest follow-up in both groups. CONCLUSIONS: LLIF decreased the Lenke-Silva classification grading and determined the optimal fusion level in patients with severe ADS. Change of Lenke-Silva classification may be associated with preoperative Lenke-Silva classification grading, use of anterior column realignment, and hyperlordotic cage.

Correlation between surgical site infection and classification of early onset scoliosis (C-EOS) in patients managed by rib-based distraction instrumentation.

OBJECTIVES: The aim of this study is to determine risk factors for infection among EOS patients treated by rib-based distraction instrumentation, and to further assess the incidence of infection among C-EOS categories and sub-types. Despite the heterogonous nature of early onset scoliosis, the classification of early onset scoliosis (C-EOS) has proven to have excellent reliability across its major categories. C-EOS's reliability has been verified; however, little data exist on the utility of this categorization in clinical decision-making and risk assessment. METHODS: After institutional review board approval, data for EOS patients treated by rib-based distraction instrumentation were collected between 2013 and 2017 in a single institution. Data collection included: major categories of early onset scoliosis classification (etiology, major curve and kyphosis), BMI, height, weight, procedure type, site of procedure, presence of tracheostomy, and bowel/urinary incontinence. RESULTS: 156 EOS patients underwent 843 rib-based distraction instrumentation procedures. 22.4% of patients (35/156 patients, 42 procedures) developed infections, 30/35 requiring irrigation and debridement. Type of procedure was significantly associated with infection rate, with rib-based distraction instrumentation insertion corresponding with the highest incidence of infection, as compared to instrumentation revisions or expansions (p = 0.006). Infection rates were also more common in shorter and lighter weight children (p = 0.001 and 0.03; respectively). Patients with a neuromuscular etiology had the highest rate of infection in comparison to congenital, syndromic, and idiopathic (5.7% vs, 4.9%, 4.7%, and 2.6%; respectively). Notably, high infection rates occurred neuromuscular hyper-kyphotic subjects (M+), occurring in all major curve C-EOS subgroups and at a rate of 8.3% for all procedures. CONCLUSION: Neuromuscular, larger magnitude major curve, and larger magnitude kyphotic angle C-EOS categories appear to be at a higher risk of infection. Such information potentiates the usefulness of C-EOS in surgical decision-making and in the informed consent process. LEVEL OF EVIDENCE: Level III therapeutic.

Towards a new 3D classification for adolescent idiopathic scoliosis.

STUDY DESIGN: Retrospective analysis of consecutive cases. OBJECTIVES: To identify clinically relevant three-dimensional (3D) sub-groups for adolescent idiopathic scoliosis (AIS). Classifications for AIS are developed to assist surgeons in surgical planning and therapeutic management. However, current systems are based on two-dimensional (2D) parameters that do not completely describe the 3D deformity. Hence, variations in surgical results based on pre-operative 2D classifications may be attributed to the lack of 3D description. METHODS: Subjects from a multicenter database of AIS patients were included in this study. All patients had bi-planar radiographs and 3D reconstruction of the entire spine. A clustering algorithm based on fuzzy c-means was utilized to identify sub-groups based on the following ten parameters measured on 3D reconstructions of the spine: Cobb angle, orientation of the plane of maximum curvature of the proximal thoracic, mid-thoracic (MT) and thoracolumbar (TLL) levels, axial rotation of the apical vertebra of the MT and TLL segments, T4-T12 thoracic kyphosis, and L1-S1 lumbar lordosis. Da Vinci views were also generated and analyzed for each patient in the study. A panel of four experienced spine surgeons from the SRS 3D Scoliosis Committee reviewed and evaluated each group to determine if cluster groups were clinically distinct from each other. RESULTS: The clustering algorithm was able to detect 11 sub-groups. The population size for each cluster varied from 11 to 290. Statistically significant differences were seen between the parameters for each group. Four spine surgeons reviewed the three most representative cases of each group and unanimously agreed that each cluster group represents a sub-group that was not defined in current classifications. CONCLUSIONS: This study presents a new method of classifying AIS based on a fuzzy clustering algorithm using parameters describing the 3D characteristics of the deformity. Further clinical validation is needed to confirm the usefulness of this classification system. LEVEL OF EVIDENCE: IV.

Reproducibility of the classification of early onset scoliosis (C-EOS).

STUDY DESIGN: Reproducibility study. OBJECTIVES: Assess the agreement and reliability of the classification of early onset scoliosis (C-EOS). C-EOS is a promising tool for patients with early onset scoliosis (EOS). However, the reliability has only been examined without measuring radiographs and not including the annual progression rate (APR) modifier. METHODS: We included a single-center consecutive cohort of patients diagnosed with EOS seen in our outpatient clinic. Patients had no previous spine surgery. Four raters rated 60 cases. Two anterior-posterior full-spine radiographs, taken minimum 6 months apart, and one sagittal radiograph were measured twice by all raters in a blinded test-retest setup. Results were assessed using crude frequency of overall agreement (OA), intra- and inter-rater Fleiss kappa (κ) statistics, and intraclass correlation coefficient (ICC). We calculated the 95% limits of agreement (LOA) for major curve angle (MCA), kyphosis, and APR using a linear mixed-effects model. Inter- and intra-rater LOA were analyzed for each etiology separately. RESULTS: Mean age was 8.7 ± 3.4 years and the etiology were congenital/structural (n = 20), idiopathic (n = 19), neuromuscular (n = 13), or syndromic (n = 8). For etiology, OA was 75.8% and κ = 0.80. For major curve angle, OA was 84.2%, κ = 0.86, ICC = 0.97, and LOA = 12.8°. For kyphosis, OA was 55.8%, κ = 0.52, ICC = 0.87, and LOA = 20.6°. For APR, OA was 76.7%, κ = 0.61, ICC = 0.77, and LOA = 17.4°/year. Inter- and intra-rater LOA were generally largest for neuromuscular and smallest for idiopathic patients. CONCLUSIONS: We found substantial agreement for etiology, however, with disagreement in certain cases. The reliability of MCA was excellent; however, somewhat lower for kyphosis and APR with less accuracy. The measurement errors of MCA, kyphosis, and APR depended largely on the etiology. Regarding APR, LOA exceeded the 10°/year increments proposed in the C-EOS, suggesting a revision of this optional modifier. LEVEL OF EVIDENCE: Diagnostic study level 1.

High Risk of Mismatch Between Sanders and Risser Staging in Adolescent Idiopathic Scoliosis: Are We Guiding Treatment Using the Wrong Classification?

BACKGROUND: Despite known limitations, Risser staging has traditionally been the primary marker of skeletal maturity utilized in decision-making for treatment of adolescent idiopathic scoliosis (AIS). The purpose of this study is to assess the incidence and factors associated with mismatch between Risser Staging and Sanders classification, and determine interobserver reliability. METHODS: We reviewed the medical records of consecutive patients aged 10 to 18 referred to our institution for evaluation of AIS from January to June 2016 with a closed triradiate cartilage. Data collected included sex, age, race, height, weight, body mass index percentile, menarchal status, Risser stage, Sanders classification, and major curve. Risser and Sanders stage was determined by 2 fellowship-trained pediatric spine surgeons and 1 pediatric orthopaedic nurse practitioner. Mismatch was defined as Risser stage 2 to 4 corresponding to Sanders 3 to 5, and Risser 0 to 1 corresponding to Sanders 6 to 7. RESULTS: A total of 165 consecutive patients were identified (mean age: 13.9±1.7 y, major curve 28.2±15.4 degrees, 76% female). The risk of skeletal maturity mismatch, based on the criteria of Risser 2 to 5 (limited growth remaining) corresponding to Sanders 3 to 5 (significant growth remaining) was 21.8%, indicating that 1 of 5 patients would be undertreated if managed by Risser criteria. Conversely, the mismatch risk for Risser 0 to 1 corresponding to Sanders 6 to 7 was 3.6%, leading such patients to be treated conservatively longer than necessary. Males and those of Hispanic ethnicity were at a higher risk of mismatch (23.1% vs. 11.9%, P=0.08; 33.3% vs. 8.8%, P=0.04, respectively). Body mass index percentile, race, and major curve were not associated with mismatch. The unweighted and weighted interobserver κ for Risser staging was 0.74 and 0.82, respectively, and 0.86 and 0.91 for Sanders classification, respectively. CONCLUSION: Given the limited sensitivity of Risser staging during peak growth velocity, high mismatch risk, and lower interobserver reliability, the Sanders classification should be utilized to guide treatment options in patients with AIS. Compared with Sanders, utilizing Risser staging results in mistreatment in a total of 1 of 4 patients, with the vast majority being undertreated. LEVEL OF EVIDENCE: Level II.

Expert Consensus for Early Onset Scoliosis Surgery.

BACKGROUND: Despite a validated classification system, high-quality multicenter research databases (CSSG/GSSG), and a recent proliferation in publications, early-onset scoliosis (EOS) surgeons have no consensus on standards for surgical treatment. The 21st-century revolution in EOS care has only accelerated, with the arrival of a classification system, magnetically controlled growing rod, nusinersen, and improved nonoperative care (Mehta or Risser casting and compliance-monitored braces). This dizzying pace of change may have outstripped our ability to develop best-practice standards for EOS surgical indications. To learn where consensus is best (and worst) at this moment, we surveyed EOS world thought-leaders on a collection of representative cases. METHODS: A 6-case survey was constructed and sent to 20 EOS world thought-leaders. The cases were selected to be representative of the major treatment categories: idiopathic, neuromuscular, syndromic, congenital, thoracic dysplasia, and spinal muscular atrophy (specifically to assess the impact of nusinersen and parasol deformity on surgical planning). Respondents were queried regarding treatment with specific attention to instrumentation and construct when surgery was selected. Responses regarding surgical timing and technique were analyzed for consensus (defined as >80%). χ analysis was performed to evaluate for differences in treatment preferences based on years of experience. RESULTS: The survey response was 100%. Clinical experience ranged from 8 to 40 years (average 23.9 y). There was no consensus on any case. The greatest variability was on the congenital case; the closest to consensus was on the spinal muscular atrophy case. Three or more approaches were selected for all 6 cases; >4 approaches were selected for 5 cases. There is a trend towards screw fixation for proximal anchors. The management of thoracic dysplasia and parasol deformity is far from consensus. CONCLUSION: The lack of consensus for surgical treatment of 6 representative EOS cases demands a renewed effort and commitment to develop best-practice guidelines based on multicenter outcome data. LEVEL OF EVIDENCE: Level V-Expert Opinion.

Surgical outcome differences between the 3D subtypes of right thoracic adolescent idiopathic scoliosis.

BACKGROUND: The current classifications of adolescent idiopathic scoliosis (AIS) aim to guide surgical decision making. However, variance exists within treatment recommendations and suboptimal outcomes have been observed while following these guidelines based on two-dimensional images. We used previously developed 3D classification for right thoracic AIS patients and aimed to determine the variation in surgical decision making and the risk of suboptimal outcomes in each subtype according to our classification. METHODS: Seventy-six right thoracic AIS patients with 2-year follow-up were included retrospectively. Five 3D preoperative subgroups were determined based on a previous classification system. The upper and lower instrumented vertebrae (UIV and LIV) and the radiographic surgical outcomes at 2-year [frontal balance (FB), proximal junctional kyphosis (PJK), and adding on] were compared between the subtypes. RESULTS: The fusion length and the rate of radiographic suboptimal outcomes were statistically different between the five groups. LIV at T12 in Type 1 and UIV at T2 in Type 2 were associated with improved FB and lower PJK, respectively. Type 3 had the highest rate of suboptimal FB and developing PJK. Type 4 had the longest fusion, and suboptimal FB was observed in 42% of the patients independent from the LIV level. Type 5 had the lowest rate of unsatisfactory radiographic outcomes at 2 years. CONCLUSION: Following the preoperative 3D classification of the AIS patients, we showed that the UIV and LIV selection has a different impact on the surgical outcomes in each of the five subtypes. The proposed 3D classification has the potential for risk stratification following a posterior spinal surgery in right thoracic AIS. These slides can be retrieved under Electronic Supplementary Material.

Clinical photography in severe idiopathic scoliosis candidate for surgery: is it a useful tool to differentiate among Lenke patterns?

PURPOSE: Clinical photography has proven to be reliable for posture assessment in adolescents and young adults with idiopathic scoliosis. This paper attempts to elucidate whether clinical photography is capable of distinguishing the distinctive characteristics in trunk deformity of the different Lenke patterns in patients with severe scoliosis candidate for surgery. METHODS: One hundred and seventy-three patients (82% women), average age of 20.8 years and average largest curve magnitude of 58.7° were included. PA standing full-spine X-rays and digital photographs from the back of the trunk were measured. RADIOLOGICAL MEASUREMENTS: It is used to measure magnitude of the proximal thoracic (PTC), main thoracic (MTC) and thoracolumbar/lumbar (TL/LC) curves, T1 tilt and the clavicle-rib intersection angle. PHOTOGRAPHIC MEASUREMENTS: It is used to measure shoulder height angle, axilla height angle, waist height angle (WHA), right and left waist angles and trunk areas. STATISTICAL ANALYSIS: One-way ANOVA to test mean differences among Lenke types for radiological and photographic measurements was performed. ROC curve analysis was conducted to find out cutoff values in photographic measurements to differentiate among curve patterns. RESULTS: Most radiological and photographic measurements differ among curve patterns. On ROC curve analysis, solid cutoff values were found for WHA (AUC = 0.8), left waist angle (AUC = 0.81), right waist angle (AUC = 0.81) and the difference between left and right waist angles (AUC = 0.86) to differentiate between types 1 and 2 and the other three types (3, 5 and 6). CONCLUSIONS: Clinical photography is a valid method for assessing trunk asymmetry in severe idiopathic scoliosis. Specifically, for waist area measurements, robust cutoff values can be determined to discriminate among different curve patterns according to Lenke classification. These slides can be retrieved under Electronic Supplementary Material.

Immediate postoperative coronal imbalance in Lenke 5 and Lenke 6 adolescent idiopathic scoliosis: Is it predictable?

PURPOSE: Immediate postoperative coronal imbalance (IPCIB) is a common reason for worse postoperative appearance in adolescent idiopathic scoliosis (AIS) patients and rarely improves on its own at follow-up, thereby greatly influencing the patients' health-related quality of life. However, no studies have been performed to detect the primary risk factors for IPCIB and it remains unclear whether the condition can be predicted preoperatively. The purpose of this study is to detect the primary risk factors for IPCIB in Lenke 5 and Lenke 6 AIS patients and to explore whether IPCIB can be predicted preoperatively. METHODS: Medical records of Lenke 5 and Lenke 6 AIS patients who underwent correction surgery in our hospital from June 2017 to October 2018 were analyzed. Anteroposterior films were evaluated before and after surgery. Patients were divided into two groups, i.e., occurrence and non-occurrence of IPCIB. The risk factors for IPCIB were analyzed, and an IPCIB index was proposed and verified. RESULTS: Thirty-seven Lenke 5/Lenke 6 AIS patients with IPCIB and 48 patients without IPCIB were recruited. Univariate analysis showed that there were significant differences between the two groups in the number of unfused vertebrae, preoperative thoracic Cobb angle, preoperative lumbar Cobb angle, preoperative translation of lumbar apex, preoperative coronal balance, preoperative L5 tilt, preoperative bending L5 tilt, postoperative translation of thoracic apex, postoperative lumbar Cobb angle, postoperative translation of lumbar apex, postoperative radiographic shoulder height, and postoperative L5 tilt. Logistic regression analysis showed that the preoperative bending L5 tilt, postoperative translation of the thoracic apex, and postoperative lumbar Cobb angle were the primary risk factors for IPCIB. The IPCIB index was defined as 1.3 * preoperative bending L5 tilt + 1.5 * postoperative translation of thoracic apex - 0.8 * postoperative lumbar Cobb angle. The receiver operating characteristics curve indicated that the occurrence rate of IPCIB was 88% and the non-occurrence rate was 90% when the IPCIB index was greater than 16. CONCLUSION: The preoperative bending L5 tilt, postoperative translation of the thoracic apex, and the postoperative lumbar Cobb angle were the primary risk factors for IPCIB in Lenke 5 and Lenke 6 AIS patients. The IPCIB index can be used to predict the occurrence of IPCIB with high accuracy. Our results indicate that the thoracic curve should be adequately corrected during surgery; however, moderate correction of the lumbar curve is recommended. These slides can be retrieved under Electronic Supplementary Material.

A novel radiographic classification of severe spinal curvatures exceeding 100°: the Omega (Ω), gamma (γ) and alpha (α) deformities.

PURPOSE: For spine curvatures with Cobb angles > 100°, curve classification and characterization become more difficult with conventional radiographs. 3-D computerized axial tomography scans add relevant information to categorize and describe a new classification to aid preoperative assessment in communication and patient evaluation. The purpose of this study is to describe a radiographic classification system of curves exceeding 100°. METHODS: A consecutive series of patients with curves exceeding 100° underwent a full spine radiographic review using conventional radiographs and 3-D CT. A descriptive analysis was performed to categorize curves into 4 main types (1, 2, 3 and 4) and 6 subtypes (1C, 1S, 1CS, 2P, 2D and 2PD) based on the location of the Cobb angle of the major scoliotic and kyphotic deformity as well as the location of the upper/lower end vertebra relative to the apical vertebra. RESULTS: A total of 98 patients met the inclusion criteria. There were 51 males and 47 females with an average age of 17.8 ± 4.5 years. The diagnosis included idiopathic (48); congenital (24); neuromuscular (4); and neurofibromatosis (2). The mean major coronal and sagittal Cobb (kyphosis) were 131.2° ± 23.4° and 154 ± 45.6, respectively. The classification scheme yielded 4 main types (1, 2, 3 and 4) and 6 subtypes under types 1 and 2 (1C, 1S, 1CS, 2P, 2D and 2PD). CONCLUSIONS: Our study describes a novel method of classifying severe spinal curvatures exceeding 100° using erect AP/lateral radiographs and 3-D CT reconstructive images. We hope that the descriptive analysis and classification will expand our understanding of these complex deformities. These slides can be retrieved under Electronic Supplementary Material.

Modified PUMC classification for adolescent idiopathic scoliosis.

BACKGROUND: The original Peking Union Medical College (PUMC) classification of adolescent idiopathic scoliosis (AIS) is one system to combine each type with corresponding surgical fusion guidance, presenting comparable interobserver reliability, and reproducibility with Lenke classification. However, during its application in previous over 10 years, we found 2 main problems of this classification, which required modification. PURPOSE: (1) To evaluate the interobserver and intraobserver reliability, (2) to assess the effects of the added fusion criteria of proximal thoracic (PT) curve on improving postoperative shoulder balance of the modified PUMC classification of AIS. STUDY DESIGN/SETTING: Retrospective analysis of our AIS cohort and prospective validation of its effectiveness. PATIENT SAMPLE: Fifty sets of preoperative radiographs of AIS patients were randomly chosen from our AIS database. Furthermore, 46 consecutive AIS cases with PT curve were enrolled who underwent surgeries in our center from July 2007 to July 2013, with at least 2-year follow-up. OUTCOME MEASURES: The classification results of 50 sets of preoperative radiographs by 5 surgeons. The shoulder balance was evaluated using radiographic shoulder height. METHODS: Five surgeons independently evaluated and classified presurgical radiographs of 50 AIS patients based on the modified PUMC classification. Inter- and intraobserver reliabilities were calculated. Furthermore, the post-op shoulder balance was investigated in 46 consecutive cases of AIS with PT curve who were treated strictly based on the modified PUMC classification. RESULTS: The Kappa coefficients of inter- and intraobserver reliability of the overall modified PUMC classification are 0.889 and 0.865, respectively. The Kappa coefficients of inter- and intraobserver reliability for the type II are 0.791 and 0.746, respectively. In addition, the shoulder balance rate of the 46 AIS patients with PT curve at the final follow-up was 95.7%. CONCLUSIONS: Modified PUMC classification presents incremental improvement compared to our original published version, with high interobserver and intraobserver reliability and better performance of postoperative shoulder balance. Furthermore, the modified PUMC classification provides corresponding surgical fusion guidance for each subtype. Multicenter prospective studies with larger samples are still needed to further improve this system.