Search Results

You are looking at 1 - 2 of 2 items for

  • Author or Editor: Carly Weber-Levine x
  • Refine by Access: all x
Clear All Modify Search
Restricted access

Carly Weber-Levine, Brendan F. Judy, Andrew M. Hersh, Tolulope Awosika, Yohannes Tsehay, Timothy Kim, Alejandro Chara, and Nicholas Theodore

OBJECTIVE

The authors systematically reviewed current evidence for the utility of mean arterial pressure (MAP), intraspinal pressure (ISP), and spinal cord perfusion pressure (SCPP) as predictors of outcomes after traumatic spinal cord injury (SCI).

METHODS

PubMed, Cochrane Reviews Library, EMBASE, and Scopus databases were queried in December 2020. Two independent reviewers screened articles using Covidence software. Disagreements were resolved by a third reviewer. The inclusion criteria for articles were 1) available in English; 2) full text; 3) clinical studies on traumatic SCI interventions; 4) involved only human participants; and 5) focused on MAP, ISP, or SCPP. Exclusion criteria were 1) only available in non-English languages; 2) focused only on the brain; 3) described spinal diseases other than SCI; 4) interventions altering parameters other than MAP, ISP, or SCPP; and 5) animal studies. Studies were analyzed qualitatively and grouped into two categories: interventions increasing MAP or interventions decreasing ISP. The Scottish Intercollegiate Guidelines Network level of evidence was used to assess bias and the Grading of Recommendations, Assessment, Development, and Evaluation approach was used to rate confidence in the anticipated effects of each outcome.

RESULTS

A total of 2540 unique articles were identified, of which 72 proceeded to full-text review and 24 were included in analysis. One additional study was included retrospectively. Articles that went through full-text review were excluded if they were a review paper (n = 12), not a full article (n = 12), a duplicate paper (n = 9), not a human study (n = 3), not in English (n = 3), not pertaining to traumatic SCI (n = 3), an improper intervention (n = 3), without intervention (n = 2), and without analysis of intervention (n = 1). Although maintaining optimal MAP levels is the current recommendation for SCI management, the published literature supports maintenance of SCPP as a stronger indicator of favorable outcomes. Studies also suggest that laminectomy and durotomy may provide better outcomes than laminectomy alone, although higher-level studies are needed. Current evidence is inconclusive on the effectiveness of CSF drainage for reducing ISP.

CONCLUSIONS

This review demonstrates the importance of assessing how different interventions may vary in their ability to optimize SCPP.

Free access

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

OBJECTIVE

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.

METHODS

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.

RESULTS

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.

CONCLUSIONS

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.