Ebtesam Abdulla, Sabrina Rahman, and Md Moshiur Rahman
Sophia A. Doerr, Carly Weber-Levine, Andrew M. Hersh, Tolulope Awosika, Brendan Judy, Yike Jin, Divyaansh Raj, Ann Liu, Daniel Lubelski, Craig K. Jones, Haris I. Sair, and Nicholas Theodore
Damage to the thoracolumbar spine can confer significant morbidity and mortality. The Thoracolumbar Injury Classification and Severity Score (TLICS) is used to categorize injuries and determine patients at risk of spinal instability for whom surgical intervention is warranted. However, calculating this score can constitute a bottleneck in triaging and treating patients, as it relies on multiple imaging studies and a neurological examination. Therefore, the authors sought to develop and validate a deep learning model that can automatically categorize vertebral morphology and determine posterior ligamentous complex (PLC) integrity, two critical features of TLICS, using only CT scans.
All patients who underwent neurosurgical consultation for traumatic spine injury or degenerative pathology resulting in spine injury at a single tertiary center from January 2018 to December 2019 were retrospectively evaluated for inclusion. The morphology of injury and integrity of the PLC were categorized on CT scans. A state-of-the-art object detection region-based convolutional neural network (R-CNN), Faster R-CNN, was leveraged to predict both vertebral locations and the corresponding TLICS. The network was trained with patient CT scans, manually labeled vertebral bounding boxes, TLICS morphology, and PLC annotations, thus allowing the model to output the location of vertebrae, categorize their morphology, and determine the status of PLC integrity.
A total of 111 patients were included (mean ± SD age 62 ± 20 years) with a total of 129 separate injury classifications. Vertebral localization and PLC integrity classification achieved Dice scores of 0.92 and 0.88, respectively. Binary classification between noninjured and injured morphological scores demonstrated 95.1% accuracy. TLICS morphology accuracy, the true positive rate, and positive injury mismatch classification rate were 86.3%, 76.2%, and 22.7%, respectively. Classification accuracy between no injury and suspected PLC injury was 86.8%, while true positive, false negative, and false positive rates were 90.0%, 10.0%, and 21.8%, respectively.
In this study, the authors demonstrate a novel deep learning method to automatically predict injury morphology and PLC disruption with high accuracy. This model may streamline and improve diagnostic decision support for patients with thoracolumbar spinal trauma.
Brendan F. Judy, Hector Soriano-Baron, Yike Jin, Hesham M. Zakaria, Srujan Kopparapu, Mir Hussain, Connor Pratt, and Nicholas Theodore
Navigation and robotics are important tools in the spine surgeon’s armamentarium and use of these tools requires placement of a reference frame. The posterior superior iliac spine (PSIS) is a commonly used site for reference frame placement, due to its location away from the surgical corridor and its ability to provide solid fixation. Placement of a reference frame requires not only familiarity with proper technique, but also command of the relevant anatomy.
Cadaveric analysis demonstrates a significant difference in PSIS location in males versus females, and additionally provides average thickness for accurate placement.
In this technical note, the authors describe the precise technique for PSIS frame placement in addition to relevant anatomy and offer solutions to commonly encountered problems.
Yike Jin, Ann Liu, Jessica R. Overbey, Ravi Medikonda, James Feghali, Sonya Krishnan, Wataru Ishida, Sutipat Pairojboriboon, Ziya L. Gokaslan, Jean-Paul Wolinsky, Nicholas Theodore, Ali Bydon, Daniel M. Sciubba, Timothy F. Witham, and Sheng-Fu L. Lo
Treatment of primary spinal infection includes medical management with or without surgical intervention. The objective of this study was to identify risk factors for the eventual need for surgery in patients with primary spinal infection on initial presentation.
From January 2010 to July 2019, 275 patients presented with primary spinal infection. Demographic, infectious, imaging, laboratory, treatment, and outcome data were retrospectively reviewed and collected. Thirty-three patients were excluded due to insufficient follow-up (≤ 90 days) or death prior to surgery.
The mean age of the 242 patients was 58.8 ± 13.6 years. The majority of the patients were male (n = 130, 53.7%), White (n = 150, 62.0%), and never smokers (n = 132, 54.5%). Fifty-four patients (22.3%) were intravenous drug users. One hundred fifty-four patients (63.6%) ultimately required surgery while 88 (36.4%) never needed surgery during the duration of follow-up. There was no significant difference in age, gender, race, BMI, or comorbidities between the surgery and no-surgery groups. On univariate analysis, the presence of an epidural abscess (55.7% in the no-surgery group vs 82.5% in the surgery group, p < 0.0001), the median spinal levels involved (2 [interquartile range (IQR) 2–3] in the no-surgery group vs 3 [IQR 2–5] in the surgery group, p < 0.0001), and active bacteremia (20.5% in the no-surgery vs 35.1% in the surgery group, p = 0.02) were significantly different. The cultured organism and initial laboratory values (erythrocyte sedimentation rate, C-reactive protein, white blood cell count, creatinine, and albumin) were not significantly different between the groups. On multivariable analysis, the final model included epidural abscess, cervical or thoracic spine involvement, and number of involved levels. After adjusting for other variables, epidural abscess (odds ratio [OR] 3.04, 95% confidence interval [CI] 1.64–5.63), cervical or thoracic spine involvement (OR 2.03, 95% CI 1.15–3.61), and increasing number of involved levels (OR 1.16, 95% CI 1.01–1.35) were associated with greater odds of surgery. Fifty-two surgical patients (33.8%) underwent decompression alone while 102 (66.2%) underwent decompression with fusion. Of those who underwent decompression alone, 2 (3.8%) of 52 required subsequent fusion due to kyphosis. No patient required hardware removal due to persistent infection.
At time of initial presentation of primary spinal infection, the presence of epidural abscess, cervical or thoracic spine involvement, as well as an increasing number of involved spinal levels were potential risk factors for the eventual need for surgery in this study. Additional studies are needed to assess for risk factors for surgery and antibiotic treatment failure.
Tadatsugu Morimoto, Takaomi Kobayashi, Masaya Ueno, Hirohito Hirata, and Masaaki Mawatari
Ann Liu, Yike Jin, Ethan Cottrill, Majid Khan, Erick Westbroek, Jeff Ehresman, Zach Pennington, Sheng-fu L. Lo, Daniel M. Sciubba, Camilo A. Molina, and Timothy F. Witham
Augmented reality (AR) is a novel technology which, when applied to spine surgery, offers the potential for efficient, safe, and accurate placement of spinal instrumentation. The authors report the accuracy of the first 205 pedicle screws consecutively placed at their institution by using AR assistance with a unique head-mounted display (HMD) navigation system.
A retrospective review was performed of the first 28 consecutive patients who underwent AR-assisted pedicle screw placement in the thoracic, lumbar, and/or sacral spine at the authors’ institution. Clinical accuracy for each pedicle screw was graded using the Gertzbein-Robbins scale by an independent neuroradiologist working in a blinded fashion.
Twenty-eight consecutive patients underwent thoracic, lumbar, or sacral pedicle screw placement with AR assistance. The median age at the time of surgery was 62.5 (IQR 13.8) years and the median body mass index was 31 (IQR 8.6) kg/m2. Indications for surgery included degenerative disease (n = 12, 43%); deformity correction (n = 12, 43%); tumor (n = 3, 11%); and trauma (n = 1, 4%). The majority of patients (n = 26, 93%) presented with low-back pain, 19 (68%) patients presented with radicular leg pain, and 10 (36%) patients had documented lower extremity weakness. A total of 205 screws were consecutively placed, with 112 (55%) placed in the lumbar spine, 67 (33%) in the thoracic spine, and 26 (13%) at S1. Screw placement accuracy was 98.5% for thoracic screws, 97.8% for lumbar/S1 screws, and 98.0% overall.
AR depicted through a unique HMD is a novel and clinically accurate technology for the navigated insertion of pedicle screws. The authors describe the first 205 AR-assisted thoracic, lumbar, and sacral pedicle screws consecutively placed at their institution with an accuracy of 98.0% as determined by a Gertzbein-Robbins grade of A or B.