Roos Justus E, Hilfiker Paul, Platz Andreas, Desbiolles Lotus, Boehm Thomas, Marincek Borut, Weishaupt Dominik
Institute of Diagnostic Radiology, University Hospital Zurich, Raemistrasse 100, Zurich 8091, Switzerland.
AJR Am J Roentgenol. 2004 Oct;183(4):959-68. doi: 10.2214/ajr.183.4.1830959.
The objective of our study was to assess the diagnostic performance of a standardized 4-MDCT trauma protocol for the evaluation of the thoracic and lumbar spine in patients with multiple injuries.
Eighty-two patients with multiple injuries underwent MDCT for the chest and abdomen using a standardized 4-MDCT trauma protocol (collimation, 4 x 2.5 mm). Secondary reconstructions targeted to the spine were performed (slice width, 3 mm; reconstruction interval, 1.5 mm). All spinal fractures were additionally scanned using a collimation of 4 x 1 mm, and these images served as the standard of reference for fracture classification. An additional 50 patients with no spinal fracture served as the control group. A total of 65 major spinal fractures were present in 55 of the patients with multiple injuries. Two observers (observer 1 and observer 2) independently evaluated all CT data for spinal fractures using a 5-point confidence scale, classified the different fracture types, and rated the image quality of spinal structures on axial images and multiplanar reformations.
Image quality for axial images was excellent in 80% and in 68% using 4 x 1 mm and 4 x 2.5 mm collimation, respectively. Image quality of the multiplanar reformations was excellent in 75% and good in 65% using 4 x 1 mm and 4 x 2.5 mm collimation, respectively. Spinal fractures were detected by observer 1 and observer 2 with a sensitivity and specificity of 98% and 97% and of 97% and 97%, respectively. Interobserver agreement regarding the confidence scale for fracture detection was substantial (kappa = 0.80), and agreement between the different imaging protocols for fracture classification was excellent for observer 1 (kappa = 0.95) and observer 2 (kappa = 0.97).
Accurate evaluation of the thoracolumbar spine is possible with targeted image reconstruction based on a standardized 4-MDCT trauma protocol of the chest and abdomen.
本研究的目的是评估标准化的4层螺旋CT(4-MDCT)创伤扫描方案在评估多发伤患者胸腰椎方面的诊断性能。
82例多发伤患者采用标准化的4-MDCT创伤扫描方案(准直,4×2.5mm)进行胸部和腹部的MDCT检查。针对脊柱进行了二次重建(层厚,3mm;重建间隔,1.5mm)。所有脊柱骨折均额外采用4×1mm准直进行扫描,这些图像用作骨折分类的参考标准。另外50例无脊柱骨折的患者作为对照组。82例多发伤患者中有55例共存在65处主要脊柱骨折。两名观察者(观察者1和观察者2)使用5级置信度量表独立评估所有CT数据中的脊柱骨折情况,对不同骨折类型进行分类,并对轴位图像和多平面重建图像上脊柱结构的图像质量进行评分。
使用4×1mm和4×2.5mm准直时,轴位图像的图像质量分别有80%为优和68%为优。使用4×1mm和4×2.5mm准直时,多平面重建图像的图像质量分别有75%为优和65%为良。观察者1和观察者2检测脊柱骨折的敏感度分别为98%和97%,特异度分别为97%和97%。观察者间在骨折检测置信度量表方面的一致性较强(kappa=0.80),观察者1(kappa=0.95)和观察者2(kappa=0.97)在不同成像方案的骨折分类方面的一致性极佳。
基于标准化的胸部和腹部4-MDCT创伤扫描方案进行针对性图像重建,能够准确评估胸腰椎情况。