Lightweight and effective tensor sensitivity for atomistic neural networks.

Atomistic machine learning focuses on the creation of models that obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here, we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 kcalmol for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature.

Adherence to the Obeid coronal malalignment classification and a residual malalignment below 20 mm can improve surgical outcomes in adult spine deformity surgery.

PURPOSE: Coronal balance is a major factor impacting the surgical outcomes in adult spinal deformity (ASD). The Obeid coronal malalignment (O-CM) classification has been proposed to improve the coronal alignment in ASD surgery. Aim of this study was to investigate whether a postoperative CM < 20 mm and adherence to the O-CM classification could improve surgical outcomes and decrease the rate of mechanical failure in a cohort of ASD patients. METHODS: Multicenter retrospective analysis of prospectively collected data on all ASD patients who underwent surgical management and had a preoperative CM > 20 mm and a 2-year follow-up. Patients were divided in two groups according to whether or not surgery had been performed in adherence to the guidelines of the O-CM classification and according to whether or not the residual CM was < 20 mm. The outcomes of interest were radiographic data, rate of mechanical complications and Patient-Reported Outcome Measures. RESULTS: At 2 years, adherence to the O-CM classification led to a lower rate of mechanical complications (40 vs. 60%). A coronal correction of the CM < 20 mm allowed for a significant improvement in SRS-22 and SF-36 scores and was associated with a 3.5 times greater odd of achieving the minimal clinical important difference for the SRS-22. CONCLUSION: Adherence to the O-CM classification could reduce the risk of mechanic complications 2 years after ASD surgery. Patients with a residual CM < 20 mm showed better functional outcomes and a 3.5 times greater odd of achieving the MCID for the SRS-22 score.

Distal junctional kyphosis in adult cervical deformity patients: where does it occur?

PURPOSE: To evaluate the impact of the lowest instrumented vertebra (LIV) on Distal Junctional kyphosis (DJK) incidence in adult cervical deformity (ACD) surgery. METHODS: Prospectively collected data from ACD patients undergoing posterior or anterior-posterior reconstruction at 13 US sites was reviewed up to 2-years postoperatively (n = 140). Data was stratified into five groups by level of LIV: C6-C7, T1-T2, T3-Apex, Apex-T10, and T11-L2. DJK was defined as a kyphotic increase > 10° in Cobb angle from LIV to LIV-1. Analysis included DJK-free survival, covariate-controlled cox regression, and DJK incidence at 1-year follow-up. RESULTS: 25/27 cases of DJK developed within 1-year post-op. In patients with a minimum follow-up of 1-year (n = 102), the incidence of DJK by level of LIV was: C6-7 (3/12, 25.00%), T1-T2 (3/29, 10.34%), T3-Apex (7/41, 17.07%), Apex-T10 (8/11, 72.73%), and T11-L2 (4/8, 50.00%) (p < 0.001). DJK incidence was significantly lower in the T1-T2 LIV group (adjusted residual = -2.13), and significantly higher in the Apex-T10 LIV group (adjusted residual = 3.91). In covariate-controlled regression using the T11-L2 LIV group as reference, LIV selected at the T1-T2 level (HR = 0.054, p = 0.008) or T3-Apex level (HR = 0.081, p = 0.010) was associated with significantly lower risk of DJK. However, there was no difference in DJK risk when LIV was selected at the C6-C7 level (HR = 0.239, p = 0.214). CONCLUSION: DJK risk is lower when the LIV is at the upper thoracic segment than the lower cervical segment. DJK incidence is highest with LIV level in the lower thoracic or thoracolumbar junction.

Comparable satisfaction and clinical outcomes after surgery for adolescent idiopathic scoliosis in the adult (AISA) between the US and Japan.

BACKGROUND: The impact of ethnicity on the surgery outcomes of adolescent idiopathic scoliosis in the adult (AISA) is poorly understood. This study aimed to compare the surgery outcomes for AISA between the United States (US) and Japan (JP). METHODS: 171 surgically treated AISA (20-40y) were consecutively collected from 2 separate multicenter databases. Patients were propensity-score matched for age, gender, curve type, levels fused, and 2y postop spinal alignment. Demographic and radiographic parameters were compared between the US and JP at baseline and 2y post-op. RESULTS: A total of 108 patients were matched by propensity score (age; US vs. JP: 29 ± 6 vs. 29 ± 7y, females: 76 vs. 76%, curve type [Schwab-SRS TypeT; TypeD; TypeL; TypeN]: 35; 35; 30; 0 vs. 37; 33; 30; 0%)] levels fused: 10 ± 4 vs. 10 ± 4, 2y thoracic curve:17 ± 13 vs. 17 ± 12°, 2y CSVL: 10 ± 8 vs. 11 ± 9 mm). Similar clinical improvement was achieved between US and JP (function; 4.2 ± 0.9 vs 4.3 ± 0.6, p = 0.60, pain; 3.8 ± 0.9 vs 4.1 ± 0.8, p = 0.13, satisfaction; 4.3 ± 0.9 vs 4.2 ± 0.7, p = 0.61, total; 4.0 ± 0.8 vs 4.1 ± 0.5, p = 0.60). The correlation analyzes indicated that postoperative SRS-22 subdomains correlated differently with satisfaction (all subdomains moderately correlated with satisfaction in the US while only pain and mental health correlated moderately with satisfaction in JP ([function: r = 0.61 vs 0.29, pain: r = . 72 vs 0.54, self-image: r = 0.72 vs 0.37, mental health: r = 0.64 vs 0.55]). CONCLUSIONS: Surgery for AISA was similarly effective in the US and JP. Satisfaction for spinal surgery among patients in different countries may not be different unless the procedure limits an individual's unique lifestyle that the patient expected to resume.

The impact of lumbar alignment targets on mechanical complications after adult lumbar scoliosis surgery.

PURPOSE: The purpose of this study was to determine the discriminatory ability of age-adjusted alignment offset and the global alignment and proportion (GAP) score parameters to predict postoperative mechanical complications. METHODS: Surgical patients from the Adult Symptomatic Lumbar Scoliosis cohort were reviewed at 2 year follow up. Age-adjusted alignment offsets and GAP parameters were calculated for each patient. A series of nonlinear logistic regression models were fit, and the odds of mechanical complications were calculated. The discriminatory ability of the GAP score, GAP score parameters, and age-adjusted alignment offsets were determined plotting receiver operative characteristic (ROC) with the C statistic (AUC). RESULTS: A total of 165 patients were included. A total of 49 mechanical complications occurred in 41 patients (21 proximal junctional kyphosis and 28 pseudoarthrosis). The GAP score had no discriminatory ability in this cohort. Relative lumbar lordosis 15 degrees greater than ideal lumbar lordosis was associated with greater mechanical complications. A lumbar distribution index of 90% was associated with fewer mechanical complications compared to a lumbar distribution index of 65%. Age-adjusted offset alignment targets had no discriminatory ability to predict mechanical complications. CONCLUSION: Radiographic alignment targets using either age-adjusted alignment target offset or GAP score parameters had minimal ability to predict mechanical complications in isolation. Mechanical complications following adult spinal deformity surgery are complex, and patient factors play a critical role. Clinical trial registeration This study was registered at ClinicalTrials.gov (number NCT00854828) in March 2009.

Surgeons' risk perception in ASD surgery: The value of objective risk assessment on decision making and patient counselling.

BACKGROUND: Surgeons often rely on their intuition, experience and published data for surgical decision making and informed consent. Literature provides average values that do not allow for individualized assessments. Accurate validated machine learning (ML) risk calculators for adult spinal deformity (ASD) patients, based on 10 year multicentric prospective data, are currently available. The objective of this study is to assess surgeon ASD risk perception and compare it to validated risk calculator estimates. METHODS: Nine ASD complete (demographics, HRQL, radiology, surgical plan) preoperative cases were distributed online to 100 surgeons from 22 countries. Surgeons were asked to determine the risk of major complications and reoperations at 72 h, 90 d and 2 years postop, using a 0-100% risk scale. The same preoperative parameters circulated to surgeons were used to obtain ML risk calculator estimates. Concordance between surgeons' responses was analyzed using intraclass correlation coefficients (ICC) (poor < 0.5/excellent > 0.85). Distance between surgeons' and risk calculator predictions was assessed using the mean index of agreement (MIA) (poor < 0.5/excellent > 0.85). RESULTS: Thirty-nine surgeons (74.4% with > 10 years' experience), from 12 countries answered the survey. Surgeons' risk perception concordance was very low and heterogeneous. ICC ranged from 0.104 (reintervention risk at 72 h) to 0.316 (reintervention risk at 2 years). Distance between calculator and surgeon prediction was very large. MIA ranged from 0.122 to 0.416. Surgeons tended to overestimate the risk of major complications and reintervention in the first 72 h and underestimated the same risks at 2 years postop. CONCLUSIONS: This study shows that expert surgeon ASD risk perception is heterogeneous and highly discordant. Available validated ML ASD risk calculators can enable surgeons to provide more accurate and objective prognosis to adjust patient expectations, in real time, at the point of care.

Kickstand rods and correction of coronal malalignment in patients with adult spinal deformity.

PURPOSE: Coronal malalignment (CM) is a challenging spinal deformity to treat. The kickstand rod (KR) technique is powerful for correcting truncal shift. This study tested the hypothesis that the KR technique provides superior coronal alignment correction in adult deformity compared with traditional rod techniques. METHODS: A retrospective evaluation of a prospectively collected multicenter database was performed. A 2:1 matched cohort of non-KR accessory rod and KR patients was planned based on preoperative coronal balance distance (CBD) and a vector of global shift. Patients were subgrouped according to CM classification with a 30-mm CBD threshold defining CM, and comparisons of surgical and clinical outcomes among groups was performed. RESULTS: Twenty-one patients with preoperative CM treated with a KR were matched to 36 controls. KR-treated patients had improved CBD compared with controls (18 vs. 35 mm, P < 0.01). The postoperative CBD did not result in clinical differences between groups in patient-reported outcomes (P ≥ 0.09). Eight (38%) of 21 KR patients and 12 (33%) of 36 control patients with preoperative CM had persistent postoperative CM (P = 0.72). CM class did not significantly affect the likelihood of treatment failure (postoperative CBD > 30 mm) in the KR cohort (P = 0.70), the control cohort (P = 0.35), or the overall population (P = 0.31). CONCLUSIONS: Application of the KR technique to coronal spinal deformity in adults allows for successful treatment of CM. Compared to traditional rod techniques, the use of KRs did not improve clinical outcome measures 1 year after spinal deformity surgery but was associated with better postoperative coronal alignment.

Machine learned Hückel theory: Interfacing physics and deep neural networks.

The Hückel Hamiltonian is an incredibly simple tight-binding model known for its ability to capture qualitative physics phenomena arising from electron interactions in molecules and materials. Part of its simplicity arises from using only two types of empirically fit physics-motivated parameters: the first describes the orbital energies on each atom and the second describes electronic interactions and bonding between atoms. By replacing these empirical parameters with machine-learned dynamic values, we vastly increase the accuracy of the extended Hückel model. The dynamic values are generated with a deep neural network, which is trained to reproduce orbital energies and densities derived from density functional theory. The resulting model retains interpretability, while the deep neural network parameterization is smooth and accurate and reproduces insightful features of the original empirical parameterization. Overall, this work shows the promise of utilizing machine learning to formulate simple, accurate, and dynamically parameterized physics models.

The Scoli-RISK 1 results of lower extremity motor function 5 years after complex adult spinal deformity surgery.

INTRODUCTION: Neurologic complications after complex adult spinal deformity (ASD) surgery are important, yet outcomes are heterogeneously reported, and long-term follow-up of actual lower extremity motor function is unknown. OBJECTIVE: To prospectively evaluate lower extremity motor function scores (LEMS) before and at 5 years after surgical correction of complex ASD. DESIGN: Retrospective analysis of a prospective, multicenter, international observational study. METHODS: The Scoli-RISK-1 study enrolled 272 ASD patients undergoing surgery from 15 centers around the world. Inclusion criteria were Cobb angle of > 80°, corrective osteotomy for congenital or revision deformity and/or 3-column osteotomy. Among patients with 5-year follow-up, comparisons of LEMS to baseline and within each follow-up period were made via documented neurologic exams on each patient. RESULTS: Seventy-seven (28.3%) patients had 5-year follow-up. Among these 77 patients with 5-year follow-up, rates of postoperative LEMS deterioration were: 14.3% hospital discharge, 10.7% at 6 weeks, 6.5% at 6 months, 9.5% at 2 years and 9.3% at 5 years postoperative. During the 2-5 year window, while mean LEMS did not change significantly (-0.5, p = 0.442), eight (11.1%) patients deteriorated (of which 3 were ≥ 4 motor points), and six (8.3%) patients improved (of which 2 were ≥ 4 points). Of the 14 neurologic complications, four (28.6%) were surgery-related, three of which required reoperation. While mean LEMS were not impacted in patients with a major surgery-related complication, mean LEMS were significantly lower in patients with neurologic surgery-related complications at discharge (p = 0.041) and 6 months (p = 0.008) between the two groups as well as the change from baseline to 5 years (p = 0.041). CONCLUSIONS: In 77 patients undergoing complex ASD surgery with 5-year follow-up, while mean LEMS did not change from 2 to 5 years, subtle neurologic changes occurred in approximately 1 in 5 patients (11.1% deteriorated; 8.3% improved). Major surgery-related complication did not result in decreased LEMS; however, those with neurologic surgery-related complications continued to have decreased lower extremity motor function at 5 years postoperative. These results underscore the importance of long-term follow-up to 5 years, using individual motor scores rather than group averages, and comparing outcomes to both baseline and last follow-up.

Artificial intelligence clustering of adult spinal deformity sagittal plane morphology predicts surgical characteristics, alignment, and outcomes.

PURPOSE: AI algorithms have shown promise in medical image analysis. Previous studies of ASD clusters have analyzed alignment metrics-this study sought to complement these efforts by analyzing images of sagittal anatomical spinopelvic landmarks. We hypothesized that an AI algorithm would cluster preoperative lateral radiographs into groups with distinct morphology. METHODS: This was a retrospective review of a multicenter, prospectively collected database of adult spinal deformity. A total of 915 patients with adult spinal deformity and preoperative lateral radiographs were included. A 2 × 3, self-organizing map-a form of artificial neural network frequently employed in unsupervised classification tasks-was developed. The mean spine shape was plotted for each of the six clusters. Alignment, surgical characteristics, and outcomes were compared. RESULTS: Qualitatively, clusters C and D exhibited only mild sagittal plane deformity. Clusters B, E, and F, however, exhibited marked positive sagittal balance and loss of lumbar lordosis. Cluster A had mixed characteristics, likely representing compensated deformity. Patients in clusters B, E, and F disproportionately underwent 3-CO. PJK and PJF were particularly prevalent among clusters A and E. Among clusters B and F, patients who experienced PJK had significantly greater positive sagittal balance than those who did not. CONCLUSIONS: This study clustered preoperative lateral radiographs of ASD patients into groups with highly distinct overall spinal morphology and association with sagittal alignment parameters, baseline HRQOL, and surgical characteristics. The relationship between SVA and PJK differed by cluster. This study represents significant progress toward incorporation of computer vision into clinically relevant classification systems in adult spinal deformity. LEVEL OF EVIDENCE IV: Diagnostic: individual cross-sectional studies with the consistently applied reference standard and blinding.

Clinical characteristics and long-term outcomes for patients who undergo cytoreductive surgery for thoracic meningiomas: a retrospective analysis.

OBJECTIVE: Primary spinal meningiomas represent a rare indolent neoplasm usually situated in the intradural-extramedullary compartment. They have a predilection for afflicting the thoracic spine and most frequently present with sensory and/or motor symptoms. Resection is the first-line treatment for symptomatic tumors, whereas other clinical factors will determine the need for adjuvant therapy. In this study, the authors aimed to elucidate clinical presentation, functional outcomes, and long-term outcomes in this population in order to better equip clinicians with the tools to counsel their patients. METHODS: This is a retrospective analysis of patients treated at the authors' institution between 1998 and 2018. All patients with thoracic meningiomas who underwent resection and completed at least one follow-up appointment were included. Multiple preoperative clinical variables, hospitalization details, and long-term outcomes were collected for the cohort. RESULTS: Forty-six patients who underwent resection for thoracic meningiomas were included. The average age of the cohort was 59 years, and the median follow-up was 53 months. Persistent sensory and motor symptoms were present in 29 patients (63%). Fifteen lesions were ventrally positioned. There were 43 WHO grade I tumors, 2 WHO grade II tumors, and 1 WHO grade III tumor; the grade III tumor was the only case of recurrence. The median length of hospitalization was 4 days. Seventeen patients (37%) were discharged to rehabilitation facilities. Thirty patients (65.2%) experienced resolution or improvement of symptoms, and there were no deaths within 30 days of surgery. Only 1 patient developed painful kyphosis and was managed medically. Ventral tumor position, new postoperative deficits, and length of stay did not correlate with disposition to a facility. Age, ventral position, blood loss, and increasing WHO grade did not correlate with length of stay. CONCLUSIONS: Outcomes are overall favorable for patients who undergo resection of thoracic meningiomas. Symptomatic patients often experience improvement, and patients generally do not require significant future operations. Tumors located ventrally, while anatomically challenging, do not necessarily herald a significantly worse prognosis or limit the extent of resection.

Use of rhBMP-2 for adult spinal deformity surgery: patterns of usage and changes over the past decade.

OBJECTIVE: Recombinant human bone morphogenetic protein-2 (rhBMP-2) has been shown to increase fusion rates; however, cost, limited FDA approval, and possible complications impact its use. Decisions regarding rhBMP-2 use and changes over time have not been well defined. In this study, the authors aimed to assess changes in rhBMP-2 use for adult spinal deformity (ASD) surgery over the past decade. METHODS: A retrospective review of the International Spine Study Group prospective multicenter database was performed to identify ASD patients treated surgically from 2008 to 2018. For assessment of rhBMP-2 use over time, 3 periods were created: 2008-2011, 2012-2015, and 2016-2018. RESULTS: Of the patients identified, 1180 met inclusion criteria, with a mean age 60 years and 30% of patients requiring revision surgery; rhBMP-2 was used in 73.9% of patients overall. The mean rhBMP-2 dose per patient was 23.6 mg. Patients receiving rhBMP-2 were older (61 vs 58 years, p < 0.001) and had more comorbidities (Charlson Comorbidity Index 1.9 vs 1.4, p < 0.001), a higher rate of the Scoliosis Research Society-Schwab pelvic tilt modifier (> 0; 68% vs 62%, p = 0.026), a greater deformity correction (change in pelvic incidence minus lumbar lordosis 15° vs 12°, p = 0.01), and more levels fused (8.9 vs 7.9, p = 0.003). Over the 3 time periods, the overall rate of rhBMP-2 use increased and then stabilized (62.5% vs 79% vs 77%). Stratified analysis showed that after an overall increase in rhBMP-2 use, only patients who were younger than 50 years, those who were smokers, those who received a three-column osteotomy (3CO), and patients who underwent revision sustained an increased rate of rhBMP-2 use between the later two periods. No similar increases were noted for older patients, nonsmokers, primary surgery patients, and patients without a 3CO. The total rhBMP-2 dose decreased over time (26.6 mg vs 24.8 mg vs 20.7 mg, p < 0.001). After matching patients by preoperative alignment, 215 patients were included, and a significantly lower rate of complications leading to revision surgery was observed within the 2012-2015 period compared with the 2008-2011 (21.4% vs 13.0%, p = 0.029) period, while rhBMP-2 was increasingly used (80.5% vs 66.0%, p = 0.001). There was a trend toward a lower rate of pseudarthrosis for patients in the 2012-2015 period, but this difference did not reach statistical significance (7% vs 4.2%, p = 0.283). CONCLUSIONS: The authors found that rhBMP-2 was used in the majority of ASD patients and was more commonly used in those with greater deformity correction. Additionally, over the last 10 years, rhBMP-2 was increasingly used for ASD patients, but the dose has decreased.

TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials.

This paper presents TorchANI, a PyTorch-based program for training/inference of ANI (ANAKIN-ME) deep learning models to obtain potential energy surfaces and other physical properties of molecular systems. ANI is an accurate neural network potential originally implemented using C++/CUDA in a program called NeuroChem. Compared with NeuroChem, TorchANI has a design emphasis on being lightweight, user friendly, cross platform, and easy to read and modify for fast prototyping, while allowing acceptable sacrifice on running performance. Because the computation of atomic environmental vectors and atomic neural networks are all implemented using PyTorch operators, TorchANI is able to use PyTorch's autograd engine to automatically compute analytical forces and Hessian matrices, as well as do force training without requiring any additional codes. TorchANI is open-source and freely available on GitHub: https://github.com/aiqm/torchani.

Ventilator Mode Does Not Influence Blood Loss or Transfusion Requirements During Major Spine Surgery: A Retrospective Study.

BACKGROUND: Blood loss during adult spinal deformity surgery is multifactorial. Anesthetic-related factors, such as mode of mechanical ventilation, may contribute to intraoperative blood loss. The aim of this study was to determine the influence of ventilator mode and ventilator parameters on intraoperative blood loss and transfusion requirements in patients undergoing prone position spine surgery. METHODS: This single-center retrospective study examined electronic medical records of patients ≥18 years of age who underwent elective prone position spine surgery between May 2015 and June 2016. Associations between ventilator mode and ventilator parameters with intraoperative estimated blood loss (EBL), packed red blood cells (PRBCs), fresh-frozen plasma (FFP), cryoprecipitate and platelet transfusions, and subfascial drain output were examined using multiple linear regression models controlling for age, sex, American Society of Anesthesiologist (ASA) physical status score, body mass index (BMI), preoperative blood coagulation parameters and laboratory values, operative levels, cage constructs, osteotomies, transforaminal lumbar interbody fusions, laminectomies, reoperation, spine surgery invasiveness index, and operative time. In a secondary analysis, EBL, blood product transfusions, and postoperative drain output were compared between pressure-controlled ventilation (PCV) and volume-controlled ventilation (VCV) propensity score-matched cohorts. RESULTS: Nine hundred forty-six records were reviewed, and 822 were included in the analysis. After adjusting for confounding, no statistically significant associations were observed between mode of ventilation and intraoperative EBL (estimate, -2; 95% confidence interval [CI], -248 to 245; P = .99) or blood product transfusions (PRBC: estimate, -9; 95% CI, -154 to 135; P = .90; FFP: estimate, -3; 95% CI, -59 to 54; P = .93; cryoprecipitate: estimate, -14; 95% CI, -70 to 43; P = .63; platelets: -7; 95% CI, -39 to 24; P = .64). After propensity score matching (n = 27 per group), no significant differences were observed in EBL (mean difference, 525 mL; 95% CI, -15 to 1065; P = .056) or blood transfusions (PRBC: mean difference, 208 mL; 95% CI, -23 to 439; P = .077; FFP (mean difference, 34 mL; 95% CI, -17 to 84; P = .19); cryoprecipitate (mean difference, 55 mL; 95% CI, -24 to 133; P = .17); or platelets (mean difference, 26 mL; 95% CI, -12 to 64; P = .18) between PCV and VCV groups. CONCLUSIONS: In prone position spine surgery, neither mode of mechanical ventilation nor airway pressure is associated with intraoperative blood loss or need for allogeneic transfusion. Use of modern ventilation strategies using lung protective techniques may mitigate differences in blood loss previously observed between PCV and VCV modes.

The spino-pelvic ratio: a novel global sagittal parameter associated with clinical outcomes in adult spinal deformity patients.

PURPOSE: Analysis of interactions of spinal alignment metrics may uncover novel alignment parameters, similar to PI-LL. This study utilized a data-driven approach to hypothesis generation by testing all possible division interactions between spinal alignment parameters. METHODS: This study was a retrospective cohort analysis. In total, 1439 patients with baseline ODI were included for hypothesis generation. In total, 666 patients had 2-year postoperative follow-up and were included for validation. All possible combinations of division interactions between baseline metrics were assessed with linear regression against baseline ODI. RESULTS: From 247 raw alignment metrics, 32,398 division interactions were considered in hypothesis generation. Conceptually, the TPA divided by PI is a measure of the relative alignment of the line connecting T1 to the femoral head and the line perpendicular to the sacral endplate. The mean TPA/PI was 0.41 at baseline and 0.30 at 2 years postoperatively. Higher TPA/PI was associated with worse baseline ODI (p < 0.0001). The change in ODI at 2 years was linearly associated with the change in TPA/PI (p = 0.0172). The optimal statistical grouping of TPA/PI was low/normal (≤ 0.2), medium (0.2-0.4), and high (> 0.4). The R-squared for ODI against categorical TPA/PI alone (0.154) was directionally higher than that for each of the individual Schwab modifiers (SVA: 0.138, PI-LL 0.111, PT 0.057). CONCLUSION: This study utilized a data-driven approach for hypothesis generation and identified the spino-pelvic ratio (TPA divided by PI) as a promising measure of sagittal spinal alignment among ASD patients. Patients with SPR > 0.2 exhibited inferior ODI scores. LEVEL OF EVIDENCE: III.

Mini-open lateral retropleural/retroperitoneal approaches for thoracic and thoracolumbar junction anterior column pathologies.

OBJECTIVE: Advancements in less invasive lateral retropleural/retroperitoneal approaches aim to address the limitation of posterolateral approaches and avoid complications associated with anterior open thoracotomy or thoracoabdominal approaches. METHODS: Consecutive patients treated with a mini-open lateral approach for thoracic or thoracolumbar anterior column pathologies were analyzed in a retrospective case series including clinical and radiographic outcomes. Special attention is given to operative techniques and surgical nuances. RESULTS: Eleven patients underwent a mini-open lateral retropleural or combined retropleural/retroperitoneal approach for thoracic or thoracolumbar junction lesions. Surgical indications included chronic fracture/deformity (n = 5), acute fracture (n = 2), neoplasm (n = 2), and osteomyelitis (n = 2). The mean length of postoperative hospital stay was 7.2 days (range 2-19 days). All patients ultimately had successful decompression and reconstruction with a mean follow-up of 16.7 months (range 6-29 months). Axial back pain assessed by the visual analog scale improved from a mean score of 8.2 to 2.2. Complications included 1 patient with deep venous thrombosis and pulmonary embolism and 1 with pneumonia. One patient developed increased leg weakness, which subsequently improved. One patient undergoing corpectomy with only lateral plate fixation developed cage subsidence requiring posterior stabilization. CONCLUSIONS: Mini-open lateral retropleural and retroperitoneal corpectomies can safely achieve anterior column reconstruction and spinal deformity correction for various thoracic and thoracolumbar vertebral pathologies.

Mental health, social support, and HIV-related sexual risk behaviors among HIV-negative adolescent sexual minority males: three U.S. cities, 2015.

We examined the association between mental health issues, social support, and HIV among adolescent sexual minority males (SMM), who are disproportionally affected by HIV. National HIV Behavioral Surveillance among Young Men Who Have Sex with Men (NHBS-YMSM) data among SMM aged 13-18 years were collected in three cities (Chicago, New York City, and Philadelphia). Separate log-linked Poisson regression models were used to estimate associations between mental health issues and social support (general and family), and 3 HIV-related sexual risk behavior outcomes: past-year condomless anal intercourse (CAI) with a male partner, past-year sex with ≥ 4 partners, and first vaginal or anal sex before age 13. Of 547 adolescent SMM, 22% reported ever attempting suicide and 10% reported past-month suicidal ideation. The majority (52%) reported depression and anxiety. Thirty-nine percent reported CAI, 29% reported ≥ 4 sex partners and 22% reported first sex before age 13. Ever attempting suicide, suicidal ideation, and depression and anxiety were associated with CAI. Separately, ever attempting suicide and lack of family support were associated with ≥ 4 sex partners. None of the mental health or support measures were associated with having sex before age 13. General social support was not associated with any sexual risk behaviors. Mental health issues are common among adolescent SMM and associated with sexual risk behaviors. Including mental health support in comprehensive HIV prevention for adolescent SMM could potentially reduce HIV risk in this population.

Pseudarthrosis in adult and pediatric spinal deformity surgery: a systematic review of the literature and meta-analysis of incidence, characteristics, and risk factors.

We conducted a systematic review with meta-analysis and qualitative synthesis. This study aims to characterize pseudarthrosis after long-segment fusion in spinal deformity by identifying incidence rates by etiology, risk factors for its development, and common features. Pseudarthrosis can be a painful and debilitating complication of spinal fusion that may require reoperation. It is poorly characterized in the setting of spinal deformity. The MEDLINE, EMBASE, and Cochrane databases were searched for clinical research including spinal deformity patients treated with long-segment fusions reporting pseudarthrosis as a complication. Meta-analysis was performed on etiologic subsets of the studies to calculate incidence rates for pseudarthrosis. Qualitative synthesis was performed to identify characteristics of and risk factors for pseudarthrosis. The review found 162 articles reporting outcomes for 16,938 patients which met inclusion criteria. In general, the included studies were of medium to low quality according to recommended reporting standards and study design. Meta-analysis calculated an incidence of 1.4% (95% CI 0.9-1.8%) for pseudarthrosis in adolescent idiopathic scoliosis, 2.2% (95% CI 1.3-3.2%) in neuromuscular scoliosis, and 6.3% (95% CI 4.3-8.2%) in adult spinal deformity. Risk factors for pseudarthrosis include age over 55, construct length greater than 12 segments, smoking, thoracolumbar kyphosis greater than 20°, and fusion to the sacrum. Choice of graft material, pre-operative coronal alignment, post-operative analgesics, and sex have no significant impact on fusion rates. Older patients with greater deformity requiring more extensive instrumentation are at higher risk for pseudarthrosis. Overall incidence of pseudarthrosis requiring reoperation is low in adult populations and very low in adolescent populations.

Development of predictive models for all individual questions of SRS-22R after adult spinal deformity surgery: a step toward individualized medicine.

PURPOSE: Health-related quality of life (HRQL) instruments are essential in value-driven health care, but patients often have more specific, personal priorities when seeking surgical care. The Scoliosis Research Society-22R (SRS-22R), an HRQL instrument for spinal deformity, provides summary scores spanning several health domains, but these may be difficult for patients to utilize in planning their specific care goals. Our objective was to create preoperative predictive models for responses to individual SRS-22R questions at 1 and 2 years after adult spinal deformity (ASD) surgery to facilitate precision surgical care. METHODS: Two prospective observational cohorts were queried for ASD patients with SRS-22R data at baseline and 1 and 2 years after surgery. In total, 150 covariates were used in training machine learning models, including demographics, surgical data and perioperative complications. Validation was accomplished via an 80%/20% data split for training and testing, respectively. Goodness of fit was measured using area under receiver operating characteristic (AUROC) curves. RESULTS: In total, 561 patients met inclusion criteria. The AUROC ranged from 56.5 to 86.9%, reflecting successful fits for most questions. SRS-22R questions regarding pain, disability and social and labor function were the most accurately predicted. Models were less sensitive to questions regarding general satisfaction, depression/anxiety and appearance. CONCLUSIONS: To the best of our knowledge, this is the first study to explicitly model the prediction of individual answers to the SRS-22R questionnaire at 1 and 2 years after deformity surgery. The ability to predict individual question responses may prove useful in preoperative counseling in the age of individualized medicine. These slides can be retrieved under Electronic Supplementary Material.

Rotational thromboelastometry-guided transfusion during lumbar pedicle subtraction osteotomy for adult spinal deformity: preliminary findings from a matched cohort study.

OBJECTIVESignificant blood loss and coagulopathy are often encountered during adult spinal deformity (ASD) surgery, and the optimal intraoperative transfusion algorithm is debatable. Rotational thromboelastometry (ROTEM), a functional viscoelastometric method for real-time hemostasis testing, may allow early identification of coagulopathy and improve transfusion practices. The objective of this study was to investigate the effect of ROTEM-guided blood product management on perioperative blood loss and transfusion requirements in ASD patients undergoing correction with pedicle subtraction osteotomy (PSO).METHODSThe authors retrospectively reviewed patients with ASD who underwent single-level lumbar PSO at the University of Virginia Health System. All patients who received ROTEM-guided blood product transfusion between 2015 and 2017 were matched in a 1:1 ratio to a historical cohort treated using conventional laboratory testing (control group). Co-primary outcomes were intraoperative estimated blood loss (EBL) and total blood product transfusion volume. Secondary outcomes were perioperative transfusion requirements and postoperative subfascial drain output.RESULTSThe matched groups (ROTEM and control) comprised 17 patients each. Comparison of matched group baseline characteristics demonstrated differences in female sex and total intraoperative dose of intravenous tranexamic acid (TXA). Although EBL was comparable between ROTEM versus control (3200.00 ± 2106.24 ml vs 3874.12 ± 2224.22 ml, p = 0.36), there was a small to medium effect size (Cohen's d = 0.31) on EBL reduction with ROTEM. The ROTEM group had less total blood product transfusion volume (1624.18 ± 1774.79 ml vs 2810.88 ± 1847.46 ml, p = 0.02), and the effect size was medium to large (Cohen's d = 0.66). This difference was no longer significant after adjusting for TXA (ß = -0.18, 95% confidence interval [CI] -1995.78 to 671.64, p = 0.32). More cryoprecipitate and less fresh frozen plasma (FFP) were transfused in the ROTEM group patients (cryoprecipitate units: 1.24 ± 1.20 vs 0.53 ± 1.01, p = 0.03; FFP volume: 119.76 ± 230.82 ml vs 673.06 ± 627.08 ml, p < 0.01), and this remained significant after adjusting for TXA (cryoprecipitate units: ß = 0.39, 95% CI 0.05 to 1.73, p = 0.04; FFP volume: ß = -0.41, 95% CI -772.55 to -76.30, p = 0.02). Drain output was lower in the ROTEM group and remained significant after adjusting for TXA.CONCLUSIONSFor ASD patients treated using lumbar PSO, more cryoprecipitate and less FFP were transfused in the ROTEM group compared to the control group. These preliminary findings suggest ROTEM-guided therapy may allow early identification of hypofibrinogenemia, and aggressive management of this may reduce blood loss and total blood product transfusion volume. Additional prospective studies of larger cohorts are warranted to identify the appropriate subset of ASD patients who may benefit from intraoperative ROTEM analysis.