Management of progressive spine deformity in a child with pseudoachondroplasia

Short-limbed dwarfism, knock knees and joint hypermobility are the classical clinical criteria encountered in patients with pseudoachondroplasia. Progressive kyphoscoliosis was the main clinical presentation in our patients with pseudoachondroplasia. Preoperative spine radiographs in an 13-years-old-girl showed severe kyphoscoliotic deformity (kyphosis - 70o and scoliosis of Cobbs angle – 55o ). Keywords: Pseudoachondroplasia; COMP gene mutation.

2-step deep learning model for landmarks localization in spine radiographs

In this work we propose to use Deep Learning to automatically calculate the coordinates of the vertebral corners in sagittal x-rays images of the thoracolumbar spine and, from those landmarks, to calculate relevant radiological parameters such as L1–L5 and L1–S1 lordosis and sacral slope. For this purpose, we used 10,193 images annotated with the landmarks coordinates as the ground truth. We realized a model that consists of 2 steps. In step 1, we trained 2 Convolutional Neural Networks to identify each vertebra in the image and calculate the landmarks coordinates respectively. In step 2, we refined the localization using cropped images of a single vertebra as input to another convolutional neural network and we used geometrical transformations to map the corners to the original image. For the localization tasks, we used a differentiable spatial to numerical transform (DSNT) as the top layer. We evaluated the model both qualitatively and quantitatively on a set of 195 test images. The median localization errors relative to the vertebrae dimensions were 1.98% and 1.68% for x and y coordinates respectively. All the predicted angles were highly correlated with the ground truth, despite non-negligible absolute median errors of 1.84°, 2.43° and 1.98° for L1–L5, L1–S1 and SS respectively. Our model is able to calculate with good accuracy the coordinates of the vertebral corners and has a large potential for improving the reliability and repeatability of measurements in clinical tasks.

Automated and Precise Bone Mineral Density Prediction and Fracture Risk Assessment using Hip/Lumbar Spine Plain Radiographs via Learning Deep Image Signatures and Correlations

Dual-energy X-ray absorptiometry (DXA) and the Fracture Risk Assessment Tool are recommended tools for osteoporotic fracture risk evaluation, but are underutilized. We present a novel and fully-automated tool to identify fractures, predict bone mineral density (BMD), and evaluate fracture risk using plain pelvis and lumbar spine radiographs. The performance of this tool were evaluated in 1639 and 11908 patients with pelvis or lumbar spine radiographs and DXA, respectively. The model was well calibrated for hip and spine BMD assessments with minimal or no bias. The area under the curve and accuracy were 0.89 and 92.4% for hip osteoporosis, 0.87 and 86.8% for spine osteoporosis, 0.92 and 94.6% for high 10-year major fracture risk, and 0.92 and 92.2% for high hip fracture risk, respectively. The success rates of our automated algorithm a real-world test were 85.3% and 90.4% for hip and spine, respectively. The clinical use of this automated tool may increase the likelihood of identifying high-risk patients in previously unscreened populations.

Hoarseness After Shoulder Surgery in Beach-chair Position

Background Isolated vocal cord palsy resulting hoarseness after shoulder surgery in beach-chair position had not been reported in literature to date. The purpose of this study was to review its incidence in our patient cohort, and identify any risk factors that may predispose the patient to the injury.Methods There were 10215 operative shoulder cases from January 2010 to December 2017. Inclusion criteria was any post-operative patients, whose operation was performed under general anesthesia in beach-chair position, who had the related symptoms, but the diagnoses had to be confirmed by otorhinolaryngologists with laryngoscopy studies. The affected patients’ clinical notes were retrospectively reviewed with the particular interest in the operative times, and the peri-operative cervical spine radiographs. The degree of cervical spine lordosis was assessed using a method described in literature, in which ‘absolute rotation angle’ (ARA) was measured. Results There were 8 reported cases of vocal cord injury in total (0.08%). Four were male patients and four were arthroscopic cases. The mean age was 59.4 ± 11.9 years old. No particular difficulties with positioning or intubation were documented. The average duration of anesthetic times was 141 minutes. On peri-operative cervical spine radiographs, the average lordosis was 8.2° (1.5° kyphosis - 21° lordosis), and except for one patient, all had ‘non-lordotic’ type curvatures. All but one patient had recovered fully with observation and expectant management, with the average recovery time being 19 weeks (range: 2 weeks to 1 year). Only patient who had not recovered during our 2-year follow-up period, had a ‘sigmoidal’ type cervical spine and was also managed with observation only.Conclusions The incidence of vocal cord injury with beach-chair positioning at our institution was low at 0.08%. The possible risk factors include long duration of the procedure and ‘non-lordotic’ cervical spine, as demonstrated by the trend in our study. Although rare, vocal cord injury has varying duration of recovery time, in the worst-case scenario being permanent, therefore it needs to be avoided by taking utmost care during positioning of the patient in beach-chair. Level of Evidence Level IV, case series

The Effect of Prolonged Pre-Operative Halo Gravity Traction for Severe Spinal Deformities on the Cervical Spine Radiographs

Study Design: Retrospective review of consecutive series. Objective: The study sought to assess the effect of prolonged pre-operative halo gravity traction (HGT) on the c-spine radiographs Methods: Data of 37 pediatric and adult patients who underwent ≥ 12wks pre-op HGT prior to definitive spine surgery from 2013-2015 at a single site in West Africa was reviewed. Radiographic assessment of the c-spine including ADI, SVA and C2-C7 Lordosis were done at pre HGT and at 4 weekly intervals. Paired T-Test was performed to evaluate changes in these parameters during HGT. Results: 37pts, 18/19 (F/M). Average age 18.2yrs. Diagnoses: 22 idiopathic, 6 congenital, 3 Post TB, 2 NM and 4 NF. Average duration of HGT: 125 days. Baseline coronal Cobb:130 deg, corrected 30% in HGT; baseline sagittal Cobb:146 deg, corrected 32% post HGT. Baseline ADI (3.17 ± 0.63 mm) did not change at 4wks ( P > 0.05) but reduced at 8wks (2.80 ± 0.56 mm) and 12wks (2.67 ± 0.51 mm) post HGT ( P < 0.05). Baseline HGT SVA (20.7 ± 14.98 mm) significantly improved at 4wks (11.55 ± 10.26 mm), 8wks (7.54 ± 6.78 mm) and 12wks (8.88 ± 4.5 mm) ( P < 0.05). Baseline C2-C7 lordosis (43 ± 20.1 deg) reduced at 4wks (26 ± 16.37 deg), 8wks (17.8 ± 14.77 deg) and 12wks (16.7 ± 11.33 deg) post HGT ( P < 0.05). There was no incidence of atlanto-axial instability on flexion extension radiographs at any interval. Conclusion: Prolonged HGT, while providing partial correction of severe spine deformities, also appeared to have no adverse effect on atlanto-axial stability or cervical alignment. Therefore, HGT can be safely applied for several weeks in the preoperative management of severe spine deformities in pediatric/adult patients.