Department of Diagnostic Radiology, University of Maryland Medical Center, 22 S Greene St, Baltimore, MD 21201, USA.
Radiology. 2013 Jul;268(1):79-88. doi: 10.1148/radiol.13121370. Epub 2013 Feb 28.
To retrospectively compare the diagnostic performance of arterial, portal venous, and dual-phase computed tomography (CT) for blunt traumatic splenic injury.
Informed consent was waived for this institutional review board-approved, HIPAA-compliant study. Retrospective record review identified 120 blunt trauma patients (87 male [72.5%] 33 female [27.5%]; age range, 18-94 years) who had undergone dual-phase abdominal CT within 5 years, including 30 without splenic injury, 30 with parenchymal injury only, 30 with splenic active bleeding, and 30 with intrasplenic pseudoaneurysm. Six radiologists each performed blinded review of 20 different cases, and scored the presence of pseudoaneurysm, active bleeding, parenchymal injury, and hematoma; 20 cases were interpreted by all radiologists. Data analysis included calculation of diagnostic performance measures with confidence intervals, areas under receiver operating characteristic curves, and interobserver agreement/variability.
For intrasplenic pseudoaneurysm, arterial phase imaging was more sensitive (70% [21 of 30] vs 17% [five of 30]; P < .0002) and more accurate (87% [78 of 90] vs 72% [65 of 90]; P = .0165) than portal venous phase imaging. For active bleeding, arterial phase imaging was less sensitive (70% [21 of 30] vs 93% [28 of 30]; P = .0195) and less accurate (89% [80 of 90] vs 98% [88 of 90]; P = .0168) than portal venous phase imaging. For parenchymal injury, arterial phase CT was less sensitive (76% [68 of 90] vs 93% [84 of 90]; P = .001) and less accurate (81% [nine of 120] vs 95% [114 of 120]; P = .0008) than portal venous phase CT. For all injuries, dual-phase review was equivalent to or better than single-phase review.
For CT evaluation of blunt splenic injury, arterial phase is superior to portal venous phase imaging for pseudoaneurysm but inferior for active bleeding and parenchymal disruption; dual-phase CT provides optimal overall performance.
回顾性比较动脉期、门静脉期和双期 CT 对钝性脾损伤的诊断性能。
本研究经机构审查委员会批准，并符合 HIPAA 规定，获得了豁免知情同意。回顾性病历审查确定了 120 例钝性创伤患者（87 例男性[72.5%]，33 例女性[27.5%]；年龄 18-94 岁），他们在 5 年内接受了双期腹部 CT 检查，其中 30 例无脾损伤，30 例仅脾实质损伤，30 例脾实质活动性出血，30 例脾内假性动脉瘤。6 名放射科医生分别对 20 例不同的病例进行了盲法评估，并对假性动脉瘤、活动性出血、实质损伤和血肿的存在进行了评分；20 例由所有放射科医生进行了评估。数据分析包括计算置信区间内的诊断性能指标、受试者工作特征曲线下面积以及观察者间的一致性/变异性。
对于脾内假性动脉瘤，动脉期成像的敏感性（70%[30 例中的 21 例] vs 17%[30 例中的 5 例]；P<.0002）和准确性（87%[90 例中的 78 例] vs 72%[90 例中的 65 例]；P=.0165）均优于门静脉期成像。对于活动性出血，动脉期成像的敏感性（70%[30 例中的 21 例] vs 93%[30 例中的 28 例]；P=.0195）和准确性（89%[90 例中的 80 例] vs 98%[90 例中的 88 例]；P=.0168）均低于门静脉期成像。对于实质损伤，动脉期 CT 的敏感性（76%[90 例中的 68 例] vs 93%[90 例中的 84 例]；P=.001）和准确性（81%[120 例中的 9 例] vs 95%[120 例中的 114 例]；P=.0008）均低于门静脉期 CT。对于所有损伤，双期检查与单期检查相当或优于单期检查。
对于钝性脾损伤的 CT 评估，动脉期成像在诊断假性动脉瘤方面优于门静脉期成像，但在诊断活动性出血和实质破坏方面则逊于门静脉期成像；双期 CT 提供了最佳的整体性能。
