Have an account? Please log in.
Text size: Small font Default font Larger font
.
Radiology Daily
Radiology Daily PracticalReviews.com Radiology Daily

Dural Tears in Spinal Burst Fractures on MRI

February 12, 2009
Written by: , Filed in: Uncategorized
  • Comments
.

Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened interpedicular distance. Objective
Dural tears are an important complication associated with burst fractures of the thoracic and lumbar spine. They can lead to persistent cerebrospinal fluid leak, meningitis, and nerve entrapment. Short of myelography, dural tears are very difficult to diagnose.

Objective
To determine indirect MRI findings that suggest the presence of dural tears in patients with spinal burst fractures.

Design
Retrospective trial.

Participants
Twenty-one patients with surgically proven dural tears associated with burst fractures were evaluated; 33 patients with burst fractures, but without dural tear, were used as controls. All fractures occurred from T12 to L5.

Methods
Sagittal T1- and T2-images of 4- to 5-mm thickness and axial images of 3- to 6-mm thickness were obtained on a 1.5T MRI. Images were evaluated for evidence of a meningocele. Measurements were made of the widest interpedicular (IP) distance on axial images, the angle of the retropulsed segment of the vertebral body (angle), ratio of the narrowest central canal AP diameter to the normal AP diameter, and extent of epidural hemorrhage.

Lamina were classified as 0=no fracture; 1= fracture without gap; 2=fracture with gap; and 3=displaced fracture. Receiver operating characteristic (ROC) curves were created for each parameter.

Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened interpedicular distance. Objective Dural tears are an important complication associated with burst fractures of the thoracic and lumbar spine. They can lead to persistent cerebrospinal fluid leak, meningitis, and nerve entrapment. Short of myelography, dural tears are very difficult to diagnose. Objective To determine indirect MRI findings that suggest the presence of dural tears in patients with spinal burst fractures. Design Retrospective trial. Participants Twenty-one patients with surgically proven dural tears associated with burst fractures were evaluated; 33 patients with burst fractures, but without dural tear, were used as controls. All fractures occurred from T12 to L5. Methods Sagittal T1- and T2-images of 4- to 5-mm thickness and axial images of 3- to 6-mm thickness were obtained on a 1.5T MRI. Images were evaluated for evidence of a meningocele. Measurements were made of the widest interpedicular (IP) distance on axial images, the angle of the retropulsed segment of the vertebral body (angle), ratio of the narrowest central canal AP diameter to the normal AP diameter, and extent of epidural hemorrhage. Lamina were classified as 0=no fracture; 1= fracture without gap; 2=fracture with gap; and 3=displaced fracture. Receiver operating characteristic (ROC) curves were created for each parameter. [text_ad] Results No pseudomeningoceles were detected in any patient. ROC curves revealed that all parameters except extent of hemorrhage were statistically significant in differentiating cases with and without dural tears, with areas under the curve of 0.7 to 0.8. Threshold values generated by the ROC curves were as follows: canal ratio <0.5; fracture angle <135°; IP distance >28 mm; and laminar fracture grade ?2. Odds ratio also demonstrated that canal ratio was the best measurement. Sensitivities using these measurements were as follows: laminar fracture grade 82%; angle of retropulsed segment 86%; ratio of central canal diameter 77%; and IP distance 55%. Specificity for these values was not provided. Conclusions Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened IP distance. Reviewer's Comments The paper and subsequent discussion serve as a guide as to which parameters suggest presence of dural tears in the setting of burst fractures, but I was somewhat disappointed with the provided statistics. The authors suggest using a combination of factors given the relatively poor sensitivities, but did not provide any data regarding which combinations may be most effective. Also, specificity was not provided, but from the ROC graph, it seems that the suggested thresholds sacrifice specificity considerably. Low thresholds would have been useful to know at what point the risk of dural tear is highly unlikely. (Reviewer–Yaron Lebovitz, MD). Reference Lee IS, Kim HJ, et al: AJNR Am J Neuroradiol; 2009;30 (January): 142-146: Dural Tears in Spinal Burst Fractures: Predictable MR Imaging Findings. [text_ad]

Results
No pseudomeningoceles were detected in any patient. ROC curves revealed that all parameters except extent of hemorrhage were statistically significant in differentiating cases with and without dural tears, with areas under the curve of 0.7 to 0.8.

Threshold values generated by the ROC curves were as follows: canal ratio <0.5; fracture angle <135°; IP distance >28 mm; and laminar fracture grade ?2. Odds ratio also demonstrated that canal ratio was the best measurement.

Sensitivities using these measurements were as follows: laminar fracture grade 82%; angle of retropulsed segment 86%; ratio of central canal diameter 77%; and IP distance 55%. Specificity for these values was not provided.

Conclusions
Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened IP distance. Reviewer’s Comments
The paper and subsequent discussion serve as a guide as to which parameters suggest presence of dural tears in the setting of burst fractures, but I was somewhat disappointed with the provided statistics. The authors suggest using a combination of factors given the relatively poor sensitivities, but did not provide any data regarding which combinations may be most effective.

Also, specificity was not provided, but from the ROC graph, it seems that the suggested thresholds sacrifice specificity considerably. Low thresholds would have been useful to know at what point the risk of dural tear is highly unlikely.

(Reviewer–Yaron Lebovitz, MD).

Reference
Lee IS, Kim HJ, et al: AJNR Am J Neuroradiol; 2009;30 (January): 142-146:
Dural Tears in Spinal Burst Fractures: Predictable MR Imaging Findings.

Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened interpedicular distance. Objective Dural tears are an important complication associated with burst fractures of the thoracic and lumbar spine. They can lead to persistent cerebrospinal fluid leak, meningitis, and nerve entrapment. Short of myelography, dural tears are very difficult to diagnose. Objective To determine indirect MRI findings that suggest the presence of dural tears in patients with spinal burst fractures. Design Retrospective trial. Participants Twenty-one patients with surgically proven dural tears associated with burst fractures were evaluated; 33 patients with burst fractures, but without dural tear, were used as controls. All fractures occurred from T12 to L5. Methods Sagittal T1- and T2-images of 4- to 5-mm thickness and axial images of 3- to 6-mm thickness were obtained on a 1.5T MRI. Images were evaluated for evidence of a meningocele. Measurements were made of the widest interpedicular (IP) distance on axial images, the angle of the retropulsed segment of the vertebral body (angle), ratio of the narrowest central canal AP diameter to the normal AP diameter, and extent of epidural hemorrhage. Lamina were classified as 0=no fracture; 1= fracture without gap; 2=fracture with gap; and 3=displaced fracture. Receiver operating characteristic (ROC) curves were created for each parameter. [text_ad] Results No pseudomeningoceles were detected in any patient. ROC curves revealed that all parameters except extent of hemorrhage were statistically significant in differentiating cases with and without dural tears, with areas under the curve of 0.7 to 0.8. Threshold values generated by the ROC curves were as follows: canal ratio <0.5; fracture angle <135°; IP distance >28 mm; and laminar fracture grade ?2. Odds ratio also demonstrated that canal ratio was the best measurement. Sensitivities using these measurements were as follows: laminar fracture grade 82%; angle of retropulsed segment 86%; ratio of central canal diameter 77%; and IP distance 55%. Specificity for these values was not provided. Conclusions Radiologists can suggest the possibility of a dural tear in the setting of burst fracture when the central canal is narrowed by more than half, acute angle <135°, more unstable laminar fracture, or widened IP distance. Reviewer's Comments The paper and subsequent discussion serve as a guide as to which parameters suggest presence of dural tears in the setting of burst fractures, but I was somewhat disappointed with the provided statistics. The authors suggest using a combination of factors given the relatively poor sensitivities, but did not provide any data regarding which combinations may be most effective. Also, specificity was not provided, but from the ROC graph, it seems that the suggested thresholds sacrifice specificity considerably. Low thresholds would have been useful to know at what point the risk of dural tear is highly unlikely. (Reviewer–Yaron Lebovitz, MD). Reference Lee IS, Kim HJ, et al: AJNR Am J Neuroradiol; 2009;30 (January): 142-146: Dural Tears in Spinal Burst Fractures: Predictable MR Imaging Findings. [text_ad]
.

Permalink: http://www.radiologydaily.com/?p=2781

Tags: , , , , , , , , , , , , , , , , , , ,

  • Comments
.

Would you like to keep current with radiological news and information?

Post Your Comments and Responses

Comments are closed.