Imaging evaluation of traumatic thoracolumbar spine injuries: Radiological review

Sensitivity and specificity of machine learning and deep learning algorithms in the diagnosis of thoracolumbar injuries resulting in vertebral fractures: A systematic review and meta-analysis

Introduction

Clinicians encounter challenges in promptly diagnosing thoracolumbar injuries (TLIs) and fractures (VFs), motivating the exploration of Artificial Intelligence (AI) and Machine Learning (ML) and Deep Learning (DL) technologies to enhance diagnostic capabilities. Despite varying evidence, the noteworthy transformative potential of AI in healthcare, leveraging insights from daily healthcare data, persists.

Research question

This review investigates the utilization of ML and DL in TLIs causing VFs.

Materials and methods

Employing Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA) methodology, a systematic review was conducted in PubMed and Scopus databases, identifying 793 studies. Seventeen were included in the systematic review, and 11 in the meta-analysis. Variables considered encompassed publication years, geographical location, study design, total participants (14,524), gender distribution, ML or DL methods, specific pathology, diagnostic modality, test analysis variables, validation details, and key study conclusions. Meta-analysis assessed specificity, sensitivity, and conducted hierarchical summary receiver operating characteristic curve (HSROC) analysis.

Results

Predominantly conducted in China (29.41%), the studies involved 14,524 participants. In the analysis, 11.76% (N = 2) focused on ML, while 88.24% (N = 15) were dedicated to deep DL. Meta-analysis revealed a sensitivity of 0.91 (95% CI = 0.86-0.95), consistent specificity of 0.90 (95% CI = 0.86-0.93), with a false positive rate of 0.097 (95% CI = 0.068-0.137).

Conclusion

The study underscores consistent specificity and sensitivity estimates, affirming the diagnostic test's robustness. However, the broader context of ML applications in TLIs emphasizes the critical need for standardization in methodologies to enhance clinical utility.

The AO Spine Thoracolumbar Injury Classification System and Treatment Algorithm in Decision Making for Thoracolumbar Burst Fractures Without Neurologic Deficit

STUDY DESIGN

Prospective Observational Study.

OBJECTIVE

To determine the alignment of the AO Spine Thoracolumbar Injury Classification system and treatment algorithm with contemporary surgical decision making.

METHODS

183 cases of thoracolumbar burst fractures were reviewed by 22 AO Spine Knowledge Forum Trauma experts. These experienced clinicians classified the fracture morphology, integrity of the posterior ligamentous complex and degree of comminution. Management recommendations were collected.

RESULTS

There was a statistically significant stepwise increase in rates of operative management with escalating category of injury (P < .001). An excellent correlation existed between recommended expert management and the actual treatment of each injury category: A0/A1/A2 (OR 1.09, 95% CI 0.70-1.69, P = .71), A3/4 (OR 1.62, 95% CI 0.98-2.66, P = .58) and B1/B2/C (1.00, 95% CI 0.87-1.14, P = .99). Thoracolumbar A4 fractures were more likely to be surgically stabilized than A3 fractures (68.2% vs 30.9%, P < .001). A modifier indicating indeterminate ligamentous injury increased the rate of operative management when comparing type B and C injuries to type A3/A4 injuries (OR 39.19, 95% CI 20.84-73.69, P < .01 vs OR 27.72, 95% CI 14.68-52.33, P < .01).

CONCLUSIONS

The AO Spine Thoracolumbar Injury Classification system introduces fracture morphology in a rational and hierarchical manner of escalating severity. Thoracolumbar A4 complete burst fractures were more likely to be operatively managed than A3 fractures. Flexion-distraction type B injuries and translational type C injuries were much more likely to have surgery recommended than type A fractures regardless of the M1 modifier. A suspected posterior ligamentous injury increased the likelihood of surgeons favoring surgical stabilization.

Successful implementation of prophylactic veno-venoarterial extracorporeal membrane oxygenation in high-risk trauma surgery: A case report

INTRODUCTION

Extracorporeal Membrane Oxygenation (ECMO) is increasingly utilized in trauma care, yet its elective use during high-risk surgeries remains unreported.

CASE REPORT

We report a successful instance of prophylactic ECMO support via a Veno-Venoarterial (V-VA) configuration during high-risk surgery in a patient with extensive trauma, including severe thoracic damage and a highly unstable thoracic spine fracture. V-VA ECMO prevented complications such as hemodynamic and respiratory collapse associated with chest compression during the surgical procedure, as the patient should be in a prone position.

DISCUSSION

The potential of ECMO as prophylactic support in high-risk surgery amongst trauma patients underscores a novel application of this technology. Complex configurations must be evaluated to avoid associated ECMO complications.

CONCLUSION

Our case highlights the potential of prophylactic ECMO hybrid modes, indicating their safe application during high-risk procedures in select trauma patients.

Development of a Machine Learning-Based Model for Accurate Detection and Classification of Cervical Spine Fractures Using CT Imaging

Cervical spine fractures represent a significant healthcare challenge, necessitating accurate detection for appropriate management and improved patient outcomes. This study aims to develop a machine learning-based model utilizing a computed tomography (CT) image dataset to detect and classify cervical spine fractures. Leveraging a large dataset of 4,050 CT images obtained from the Radiological Society of North America (RSNA) Cervical Spine Fracture dataset, we evaluate the potential of machine learning and deep learning algorithms in achieving accurate and reliable cervical spine fracture detection. The model demonstrates outstanding performance, achieving an average precision of 1 and 100% precision, recall, sensitivity, specificity, and accuracy values. These exceptional results highlight the potential of machine learning algorithms to enhance clinical decision-making and facilitate prompt treatment initiation for cervical spine fractures. However, further research and validation efforts are warranted to assess the model's generalizability across diverse populations and real-world clinical settings, ultimately contributing to improved patient outcomes in cervical spine fracture cases.

Injury patterns of the spine following blunt trauma: A per-segment analysis of spinal structures and their detection rates in CT and MRI

Rationale and objectives

To provide a detailed analysis of injury patterns of the spine following blunt trauma and establish the role of supplementary MRI by evaluating discrepancies in the detection rates of damaged structures in CT and MRI.

Method

216 patients with blunt trauma to the spine who underwent CT followed by supplementary MRI were included in this study. Two board-certified radiologists blinded to clinical symptoms and injury mechanisms independently interpreted all acquired CT and MRI images. The interpretation was performed using a dedicated catalogue of typical findings associated with spinal trauma and assessed for spinal stability using the AO classification systems.

Results

Lesions to structures associated with spinal instability were present in 31.0% in the cervical spine, 12.3% in the thoracic spine, and 29.9% in the lumbar spine. In all spinal segments, MRI provided additional information regarding potentially unstable injuries. Novel information derived from supplementary MRI changed clinical management in 3.6% of patients with injury to the cervical spine. No change in clinical management resulted from novel information on the thoracolumbar spine. Patients with injuries to the vertebral body, intervertebral disc, or spinous process were significantly more likely to benefit from supplementary MRI.

Conclusion

In patients that sustained blunt spinal trauma, supplementary MRI of the cervical spine should routinely be performed to detect injuries that require surgical treatment, whereas CT is the superior imaging modality for the detection of unstable injuries in the thoracolumbar spine.

Single-Stage Posterior Vertebral Column Resection With Circumferential Reconstruction for Thoracic/Thoracolumbar Burst Fractures With or Without Neurological Deficit: Clinical Neurological and Radiological Outcomes

STUDY DESIGN

Retrospective study.

OBJECTIVE

The aim of this study is to evaluate the clinical, neurological, and radiological outcomes of posterior vertebral column resection (PVCR) technique for treatment of thoracic and thoracolumbar burst fractures.

METHODS

Fifty-one patients (18 male, 33 female) with thoracic/thoracolumbar burst fractures who had been treated with PVCR technique were retrospectively reviewed. Preoperative and most recent radiographs were evaluated and local kyphosis angle (LKA), sagittal and coronal spinal parameters were measured. Neurological and functional results were assessed by the American Spinal Injury Association (ASIA) Impairment Scale, visual analogue scale score, Oswestry Disability Index, and Short Form 36 version 2.

RESULTS

The mean age was 49 years (range 22-83 years). The mean follow-up period was 69 months (range 28-216 months). Fractures were thoracic in 16 and thoracolumbar in 35 of the patients. AO spine thoracolumbar injury morphological types were as follows: 1 type A3, 15 type A4, 4 type B1, 23 type B2, 8 type C injuries. PVCR was performed in a single level in 48 of the patients and in 2 levels in 3 patients. The mean operative time was 434 minutes (range 270-530 minutes) and mean intraoperative blood loss was 520 mL (range 360-1100 mL). The mean LKA improved from 34.7° to 4.9° (85.9%). For 27 patients, the initial neurological deficit (ASIA A in 8, ASIA B in 3, ASIA C in 5, and ASIA D in 11) improved at least 1 ASIA grade (1-3 grades) in 22 patients (81.5%). Solid fusion, assessed with computed tomography at the final follow-up, was achieved in all patients.

CONCLUSION

Single-stage PVCR provides complete spinal canal decompression, ideal kyphosis correction with gradual lengthening of anterior column together with sequential posterior column compression. Anterior column support, avoidance of the morbidity of anterior approach and improvement of neurological deficit are the other advantages of the single stage PVCR technique in patients with thoracic/thoracolumbar burst fractures.

Classification and Radiological Diagnosis of Thoracolumbar Spine Fractures: WFNS Spine Committee Recommendations

The aim of this review to determine recommendations for classification and radiological diagnosis of thoracolumbar spine fractures. Recommendation was made through a literature review of the last 10 years. The statements created by the authors were discussed and voted on during 2 consensus meetings organized by the WFNS (World Federation Neurosurgical Societies) Spine Committee. The literature review was yielded 256 abstracts, of which 32 were chosen for full-text analysis. Thirteen papers evaluated the reliability of a classification system by our expert members and were also chosen in this guideline analysis. This literature review-based recommendation provides the classification and radiologic diagnosis in thoracolumbar spine fractures that can elucidate the management decision-making in clinical practice.

Clinical and Radiological Factors Affecting Thoracolumbar Fractures Outcome: WFNS Spine Committee Recommendations

To obtain a list of recommendations about clinical and radiological factors affecting outcome in thoraco-lumbar fractures with the aim of helping spine surgeons in daily practice. A systematic literature search in PubMed and Google Scholar database was done from 2010 to 2020 on the topic "thoracolumbar fracture AND radiology AND surgical outcomes" and "thoracolumbar fracture AND radiology AND surgical outcomes." A total of 58 papers were analyzed and WFNS (World Federation of Neurosurgical Societies) Spine Committee organized 2 consensus meetings to formulate the specific recommendations the first in Peshawar in December 2019 and in a subsequent virtual meeting in June 2020 to reach an agreement. Both meetings utilized the Delphi method to analyze preliminary literature review statements based on the current evidence levels to generate recommendations through a comprehensive voting session. Eight statements were presented and reached the consensus about this topic. A variety of clinical factors is known to influence outcome of patients with thoracolumbar fractures. Some of these are well-known established factors such as blood pressure augmentation and patient age, while some are not well studied. Overall, the quality of evidence is low and we need more randomized controlled studies to validate our results. Similarly, radiological factors that can predict outcome are well stated and there is a high accordance worldwide. In reverse, still under debate is the application to choose which surgical treatment is advisable based on them.

Natural changes of traumatic vertebral compression fractures during the first 6 months in patients visiting for disability certificates: A retrospective observational study

To identify the natural changes of traumatic vertebral compression fractures during the first six months in patients visiting for disability certificates after conservative treatment.Data of patients who visited the rehabilitation medicine department of a university hospital for disability certificates concerning traumatic vertebral compression fractures from 2015 to 2018 were reviewed. Those who visited 180 to 210 days after injuries were included, and those who received invasive procedures for compression fractures were excluded. The anterior and posterior heights, local kyphotic angle of compression fractures, and upper and lower vertebrae on initial and follow-up images were measured and compared. Compression ratio was calculated by vertebral body compression ratio and anterior vertebral body compression percentage. Thoracic and lumbar traumatic fractures were also compared.Among 110 patients, 61 patients met the criteria. After six months, the anterior height of compression fractures decreased more than 4 mm, which implies the development of new compression fractures. The compression ratio and local kyphotic angle increased significantly without affecting the upper and lower vertebrae. Thoracic and lumbar compression fractures showed similar changes.Traumatic vertebral compression fractures change significantly during the first six months. This study could warrant 6 months of waiting for issuance of disability certificates for patients with traumatic vertebral compression fractures.

Rare Case of Bilateral Pure Facet Joint Dislocation of Upper Lumbar Spine without Facet Fracture: A Case Report

Introduction

Bilateral facetal dislocation without facet fracture, although common in cervical spine, is a very rare entity in lumbar spine with <15 cases reported so far. Such injuries are very unstable involving all the three columns. Neurological insult and visceral affection are commonly associated with bilateral facetal dislocation.

Case Report

A 22-year-old gentleman presented with ASIA Aparaplegia following road traffic accident. Radiographs/computed tomography scan revealed pure facetal dislocation L1-L2 with no evidence of facet fracture. The patient also had liver laceration. The patient underwent open instrumented reduction along with left-sidedtransforaminal removal of damaged disc and inter body fusion. The patient improved significantly to ASIA C neurological status at 6-month follow-up.

Conclusion

Pure facetal dislocation, although rarely seen in lumbar region, is a very unstable injury. Prompt recognition and early intervention facilitate nursing care and neurological recovery. Recognition of associated injuries is also important.

The role of diffusion tensor imaging as an objective tool for the assessment of motor function recovery after paraplegia in a naturally-occurring large animal model of spinal cord injury

Background

Traumatic spinal cord injury (SCI) results in sensory and motor function impairment and may cause a substantial social and economic burden. For the implementation of novel treatment strategies, parallel development of objective tools evaluating spinal cord (SC) integrity during motor function recovery (MFR) is needed. Diffusion tensor imaging (DTI) enables in vivo microstructural assessment of SCI.

Methods

In the current study, temporal evolvement of DTI metrics during MFR were examined; therefore, values of fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were measured in a population of 17 paraplegic dogs with naturally-occurring acute SCI showing MFR within 4 weeks after surgical decompression and compared to 6 control dogs. MRI scans were performed preoperatively and 12 weeks after MFR was observed. DTI metrics were obtained at the lesion epicentre and one SC segment cranially and caudally. Variance analyses were performed to compare values between evaluated localizations in affected dogs and controls and between time points. Correlations between DTI metrics and clinical scores at follow-up examinations were assessed.

Results

Before surgery, FA values at epicentres were higher than caudally (p = 0.0014) and control values (p = 0.0097); ADC values were lower in the epicentre compared to control values (p = 0.0035) and perilesional (p = 0.0448 cranially and p = 0.0433 caudally). In follow-up examinations, no significant differences could be found between DTI values from dogs showing MFR and control dogs. Lower ADC values at epicentres correlated with neurological deficits at follow-up examinations (r = − 0.705; p = 0.0023).

Conclusions

Findings suggest that a tendency to the return of DTI values to the physiological situation after surgical decompression accompanies MFR after SCI in paraplegic dogs. DTI may represent a useful and objective clinical tool for follow-up studies examining in vivo SC recovery in treatment studies.

Electronic supplementary material

The online version of this article (10.1186/s12967-018-1630-4) contains supplementary material, which is available to authorized users.

Prediction of MRI findings including disc injury and posterior ligamentous complex injury in neurologically intact thoracolumbar burst fractures by the parameters of vertebral body damage on CT scan

OBJECTIVE

To formulate radiological indexes based on CT for further MRI examination to detect posterior ligamentous complex injury (PLC) or disc injury in thoracolumbar burst fractures without neurological deficit in the emergent setting.

MATERIALS AND METHODS

Patients with a single thoracolumbar burst fracture and no neurological deficit were included into this study. Radiological indexes on CT included canal compromise (CC), anterior and posterior vertebral height ratio (PVH and AVH ratio), local kyphosis (LK) and regional kyphosis (RK). PLC and disc injury were assessed on MRI. Statistical analysis was performed to identify the predictive power for radiological indexes for any MRI findings either or both disc and PLC injury.

RESULTS

Eighty-four patients were included in this study. According to MRI, patients with no PLC and disc injury were allocated into MRI finding negative group, others were defined as positive group. There was no significant difference in AVH ratio, PVH ratio and RK between these two groups. The CC and LK were significant higher in positive group than that in negative group (p < 0.001).The areas under receiver operating characteristic curve were 0.826 and 0.893 for CC and LK respectively and without significant difference. The best thresholds for CC and LK were 0.19 (sensitivity: 69.4%; specificity: 87.5%) and 14.00° (sensitivity: 83.3%; specificity: 83.3%), respectively.

CONCLUSION

The presence of CC > 0.19 and/or LK > 14.00° on CT scan can predict MRI findings including PLC and disc injury. These thresholds may be the guideline for MRI examination in patients with neurologically intact thoracolumbar burst fracture in the emergent condition.

Normal aging of the lumbar spine in women

BACKGROUND

Lumbar lordosis is required for bipedalism.

OBJECTIVES

To investigate age-related changes in lumbar lordosis and to clarify the relationships between lumbar lordosis and vertebral wedging and disc degeneration.

METHODS

A total of 300 women were included in this retrospective study, 50 in each of six age groups (20-, 30-, 40-, 50-, 60- and 70-year-olds). Patients with vertebral collapse, instable fracture or disc sequestration were excluded. In each patient, lumbar lordosis angle, posterior vertebral wedging, L5-S1 intervertebral disc angle, L5/L1 vertebral height ratio and L5-S1 intervertebral disc/L1-L2 intervertebral disc height ratio were examined. Significance level was set at p< 0.05 and two-sided tests were used.

RESULTS

Significant differences were found in lumbar lordosis according to age group (p< 0.001). Lumbar lordosis correlated most strongly with L4 posterior vertebral wedging, L5 posterior vertebral wedging and L5-S1 intervertebral disc angle, in that order (r= 0.50, r= 0.40, r= 0.32, respectively; p< 0.001).

CONCLUSION

In this cohort, strong spinal structure was maintained during physiological aging from 20 to 40 years of age; lumbar lordosis increased by 50 years of age. Increased lordosis correlated with increased posterior vertebral wedging and loss of posterior disc height.

Imaging of Spine Trauma

Imaging with computed tomography and magnetic resonance imaging is fundamental to the evaluation of traumatic spinal injury. Specifically, neuroradiologic techniques show the exact location of injury, evaluate the stability of the spine, and determine neural element compromise. This review focuses on the complementary role of different radiologic modalities in the diagnosis of patients with traumatic injuries of the spine. The role of imaging in spinal trauma classifications will be addressed. The importance of magnetic resonance imaging in the assessment of soft tissue injury, particularly of the spinal cord, will be discussed. Last, the increasing role of advanced imaging techniques for prognostication of the traumatic spine will be explored.