Penetrating diaphragmatic injury: Accuracy of 64-section multidetector CT with trajectography

David Dreizin, Maria J. Borja, Gary Danton, Kevin Kadakia, Kim Caban, Luis Rivas, Felipe Munera

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Purpose: To (a) determine the diagnostic performance of 64-section multidetector computed tomography (CT) trajectography for penetrating diaphragmatic injury (PDI), (b) determine the diagnostic performance of classic signs of diaphragmatic injury at 64-section multidetector CT, and (c) compare the performance of these signs with that of trajectography. Materials and Methods: This HIPAA-compliant retrospective study had institutional review board approval, with a waiver of the informed consent requirement. All patients who had experienced penetrating thoracoabdominal trauma, who had undergone preoperative 64-section multidetector CT of the chest and abdomen, and who had surgical confirmation of findings during a 2.5-year period were included in this study (25 male patients, two female patients; mean age, 32.6 years). After a training session, four trauma radiologists unaware of the surgical outcome independently reviewed all CT studies and scored the probability of PDI on a six-point scale. Collar sign, dependent viscera sign, herniation, contiguous injury on both sides of the diaphragm, discontinuous diaphragm sign, and transdiaphragmatic trajectory were evaluated for sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). Accuracies were determined and receiver operating characteristic curves were analyzed. Results: Sensitivities for detection of PDI by using 64-section multidetector CT with postprocessing software ranged from 73% to 100%, specificities ranged from 50% to 92%, NPVs ranged from 71% to 100%, PPVs ranged from 68% to 92%, and accuracies ranged from 70% to 89%. Discontinuous diaphragm, herniation, collar, and dependent viscera signs were highly specific (92%-100%) but nonsensitive (0%-60%). Contiguous injury was generally more sensitive (80%-93% vs 73%-100%) but less specific (50%-67% vs 83%-92%) than transdiaphragmatic trajectory when patients with multiple entry wounds were included in the analysis. Transdiaphragmatic trajectory was a much more sensitive sign of PDI than previously reported (73%-100% vs 36%), with NPVs ranging from 71% to 100% and PPVs ranging from 85% to 92%. Conclusion: Sixty-four-section multidetector CT trajectography facilitates the identification of transdiaphragmatic trajectory, which accurately rules in PDI when identified. Contiguous injury remains a highly sensitive sign, even when patients with multiple injuries are considered, and is useful for excluding PDI.

Original languageEnglish
Pages (from-to)729-737
Number of pages9
JournalRadiology
Volume268
Issue number3
DOIs
StatePublished - Sep 1 2013

Fingerprint

Multidetector Computed Tomography
Wounds and Injuries
Diaphragm
Viscera
Multiple Trauma
Health Insurance Portability and Accountability Act
Research Ethics Committees
Informed Consent
ROC Curve
Abdomen
Thorax
Software
Retrospective Studies
Tomography

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Penetrating diaphragmatic injury : Accuracy of 64-section multidetector CT with trajectography. / Dreizin, David; Borja, Maria J.; Danton, Gary; Kadakia, Kevin; Caban, Kim; Rivas, Luis; Munera, Felipe.

In: Radiology, Vol. 268, No. 3, 01.09.2013, p. 729-737.

Research output: Contribution to journalArticle

Dreizin, David ; Borja, Maria J. ; Danton, Gary ; Kadakia, Kevin ; Caban, Kim ; Rivas, Luis ; Munera, Felipe. / Penetrating diaphragmatic injury : Accuracy of 64-section multidetector CT with trajectography. In: Radiology. 2013 ; Vol. 268, No. 3. pp. 729-737.
@article{653a28eb1d824034912b1a62f0ff6391,
title = "Penetrating diaphragmatic injury: Accuracy of 64-section multidetector CT with trajectography",
abstract = "Purpose: To (a) determine the diagnostic performance of 64-section multidetector computed tomography (CT) trajectography for penetrating diaphragmatic injury (PDI), (b) determine the diagnostic performance of classic signs of diaphragmatic injury at 64-section multidetector CT, and (c) compare the performance of these signs with that of trajectography. Materials and Methods: This HIPAA-compliant retrospective study had institutional review board approval, with a waiver of the informed consent requirement. All patients who had experienced penetrating thoracoabdominal trauma, who had undergone preoperative 64-section multidetector CT of the chest and abdomen, and who had surgical confirmation of findings during a 2.5-year period were included in this study (25 male patients, two female patients; mean age, 32.6 years). After a training session, four trauma radiologists unaware of the surgical outcome independently reviewed all CT studies and scored the probability of PDI on a six-point scale. Collar sign, dependent viscera sign, herniation, contiguous injury on both sides of the diaphragm, discontinuous diaphragm sign, and transdiaphragmatic trajectory were evaluated for sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). Accuracies were determined and receiver operating characteristic curves were analyzed. Results: Sensitivities for detection of PDI by using 64-section multidetector CT with postprocessing software ranged from 73{\%} to 100{\%}, specificities ranged from 50{\%} to 92{\%}, NPVs ranged from 71{\%} to 100{\%}, PPVs ranged from 68{\%} to 92{\%}, and accuracies ranged from 70{\%} to 89{\%}. Discontinuous diaphragm, herniation, collar, and dependent viscera signs were highly specific (92{\%}-100{\%}) but nonsensitive (0{\%}-60{\%}). Contiguous injury was generally more sensitive (80{\%}-93{\%} vs 73{\%}-100{\%}) but less specific (50{\%}-67{\%} vs 83{\%}-92{\%}) than transdiaphragmatic trajectory when patients with multiple entry wounds were included in the analysis. Transdiaphragmatic trajectory was a much more sensitive sign of PDI than previously reported (73{\%}-100{\%} vs 36{\%}), with NPVs ranging from 71{\%} to 100{\%} and PPVs ranging from 85{\%} to 92{\%}. Conclusion: Sixty-four-section multidetector CT trajectography facilitates the identification of transdiaphragmatic trajectory, which accurately rules in PDI when identified. Contiguous injury remains a highly sensitive sign, even when patients with multiple injuries are considered, and is useful for excluding PDI.",
author = "David Dreizin and Borja, {Maria J.} and Gary Danton and Kevin Kadakia and Kim Caban and Luis Rivas and Felipe Munera",
year = "2013",
month = "9",
day = "1",
doi = "10.1148/radiol.13121260",
language = "English",
volume = "268",
pages = "729--737",
journal = "Radiology",
issn = "0033-8419",
publisher = "Radiological Society of North America Inc.",
number = "3",

}

TY - JOUR

T1 - Penetrating diaphragmatic injury

T2 - Accuracy of 64-section multidetector CT with trajectography

AU - Dreizin, David

AU - Borja, Maria J.

AU - Danton, Gary

AU - Kadakia, Kevin

AU - Caban, Kim

AU - Rivas, Luis

AU - Munera, Felipe

PY - 2013/9/1

Y1 - 2013/9/1

N2 - Purpose: To (a) determine the diagnostic performance of 64-section multidetector computed tomography (CT) trajectography for penetrating diaphragmatic injury (PDI), (b) determine the diagnostic performance of classic signs of diaphragmatic injury at 64-section multidetector CT, and (c) compare the performance of these signs with that of trajectography. Materials and Methods: This HIPAA-compliant retrospective study had institutional review board approval, with a waiver of the informed consent requirement. All patients who had experienced penetrating thoracoabdominal trauma, who had undergone preoperative 64-section multidetector CT of the chest and abdomen, and who had surgical confirmation of findings during a 2.5-year period were included in this study (25 male patients, two female patients; mean age, 32.6 years). After a training session, four trauma radiologists unaware of the surgical outcome independently reviewed all CT studies and scored the probability of PDI on a six-point scale. Collar sign, dependent viscera sign, herniation, contiguous injury on both sides of the diaphragm, discontinuous diaphragm sign, and transdiaphragmatic trajectory were evaluated for sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). Accuracies were determined and receiver operating characteristic curves were analyzed. Results: Sensitivities for detection of PDI by using 64-section multidetector CT with postprocessing software ranged from 73% to 100%, specificities ranged from 50% to 92%, NPVs ranged from 71% to 100%, PPVs ranged from 68% to 92%, and accuracies ranged from 70% to 89%. Discontinuous diaphragm, herniation, collar, and dependent viscera signs were highly specific (92%-100%) but nonsensitive (0%-60%). Contiguous injury was generally more sensitive (80%-93% vs 73%-100%) but less specific (50%-67% vs 83%-92%) than transdiaphragmatic trajectory when patients with multiple entry wounds were included in the analysis. Transdiaphragmatic trajectory was a much more sensitive sign of PDI than previously reported (73%-100% vs 36%), with NPVs ranging from 71% to 100% and PPVs ranging from 85% to 92%. Conclusion: Sixty-four-section multidetector CT trajectography facilitates the identification of transdiaphragmatic trajectory, which accurately rules in PDI when identified. Contiguous injury remains a highly sensitive sign, even when patients with multiple injuries are considered, and is useful for excluding PDI.

AB - Purpose: To (a) determine the diagnostic performance of 64-section multidetector computed tomography (CT) trajectography for penetrating diaphragmatic injury (PDI), (b) determine the diagnostic performance of classic signs of diaphragmatic injury at 64-section multidetector CT, and (c) compare the performance of these signs with that of trajectography. Materials and Methods: This HIPAA-compliant retrospective study had institutional review board approval, with a waiver of the informed consent requirement. All patients who had experienced penetrating thoracoabdominal trauma, who had undergone preoperative 64-section multidetector CT of the chest and abdomen, and who had surgical confirmation of findings during a 2.5-year period were included in this study (25 male patients, two female patients; mean age, 32.6 years). After a training session, four trauma radiologists unaware of the surgical outcome independently reviewed all CT studies and scored the probability of PDI on a six-point scale. Collar sign, dependent viscera sign, herniation, contiguous injury on both sides of the diaphragm, discontinuous diaphragm sign, and transdiaphragmatic trajectory were evaluated for sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). Accuracies were determined and receiver operating characteristic curves were analyzed. Results: Sensitivities for detection of PDI by using 64-section multidetector CT with postprocessing software ranged from 73% to 100%, specificities ranged from 50% to 92%, NPVs ranged from 71% to 100%, PPVs ranged from 68% to 92%, and accuracies ranged from 70% to 89%. Discontinuous diaphragm, herniation, collar, and dependent viscera signs were highly specific (92%-100%) but nonsensitive (0%-60%). Contiguous injury was generally more sensitive (80%-93% vs 73%-100%) but less specific (50%-67% vs 83%-92%) than transdiaphragmatic trajectory when patients with multiple entry wounds were included in the analysis. Transdiaphragmatic trajectory was a much more sensitive sign of PDI than previously reported (73%-100% vs 36%), with NPVs ranging from 71% to 100% and PPVs ranging from 85% to 92%. Conclusion: Sixty-four-section multidetector CT trajectography facilitates the identification of transdiaphragmatic trajectory, which accurately rules in PDI when identified. Contiguous injury remains a highly sensitive sign, even when patients with multiple injuries are considered, and is useful for excluding PDI.

UR - http://www.scopus.com/inward/record.url?scp=84883196076&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883196076&partnerID=8YFLogxK

U2 - 10.1148/radiol.13121260

DO - 10.1148/radiol.13121260

M3 - Article

C2 - 23674790

AN - SCOPUS:84883196076

VL - 268

SP - 729

EP - 737

JO - Radiology

JF - Radiology

SN - 0033-8419

IS - 3

ER -