Diab Hadi Mahmoud Haider, Rasmussen Lars Melholt, Duvnjak Stevo, Diederichsen Axel, Jensen Pia Søndergaard, Lindholt Jes Sanddal
Department of Cardiothoracic and Vascular Surgery, Odense University Hospital, Sdr. Boulevard 29, 5000, Odense C, Denmark.
Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Sdr. Boulevard 29, 5000, Odense C, Denmark.
BMC Med Imaging. 2017 Dec 13;17(1):61. doi: 10.1186/s12880-017-0233-5.
Primary to validate a commercial semi-automated computed tomography angiography (CTA) -software for vulnerable plaque detection compared to histology of carotid endarterectomy (CEA) specimens and secondary validating calcifications scores by in vivo CTA with ex vivo non-contrast enhanced computed tomography (NCCT).
From January 2014 to October 2016 53 patients were included retrospectively, using a cross-sectional design. All patients underwent both CTA and CEA. Sixteen patients had their CEA specimen NCCT scanned. The semi-automated CTA software analyzed carotid stenosis using different HU values defining plaque components. The predictive values of CTA based detection of vulnerable plaques were calculated. Quantification of calcifications on CTA using region of interest (ROI)-function and mathematical equations was done manually, and validated by NCCT of the CEA specimen.
The semi-automated CTA software had a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 89.1% (95% CI, 73.6% - 96.4%), 31.3% (95% CI, 12.1% - 58.5%), 75% (95% CI, 59.3% - 86.2%) and 55.6% (95% CI, 22.6% - 84.6%). Strong correlation between in vivo CTA and ex vivo NCCT in quantification of calcification was observed, but CTA systematically underestimated calcificationsscore (CALS) with increasing calcification.
The CTA-software cannot be used in risk assessment of patients, due to poor specificity and NPV. The correlation between in vivo CTA and ex vivo NCCT was strong, proposing it to be used in both scientifically and clinical settings, but studies with larger sample sizes are needed.
主要目的是验证一款用于检测易损斑块的商用半自动计算机断层血管造影（CTA）软件与颈动脉内膜切除术（CEA）标本的组织学结果相比的准确性，次要目的是通过体内CTA与体外非增强计算机断层扫描（NCCT）来验证钙化评分。
采用横断面设计，回顾性纳入2014年1月至2016年10月的53例患者。所有患者均接受了CTA和CEA检查。16例患者的CEA标本进行了NCCT扫描。半自动CTA软件使用不同的HU值定义斑块成分来分析颈动脉狭窄情况。计算基于CTA检测易损斑块的预测值。使用感兴趣区域（ROI）功能和数学方程手动对CTA上的钙化进行定量，并通过CEA标本的NCCT进行验证。
半自动CTA软件的灵敏度、特异度、阳性预测值（PPV）和阴性预测值（NPV）分别为89.1%（95%CI，73.6% - 96.4%）、31.3%（95%CI，12.1% - 58.5%）、75%（95%CI，59.3% - 86.2%）和55.6%（95%CI，22.6% - 84.6%）。观察到体内CTA与体外NCCT在钙化定量方面有很强的相关性，但随着钙化增加，CTA系统性地低估了钙化评分（CALS）。
由于特异度和NPV较差，该CTA软件不能用于患者的风险评估。体内CTA与体外NCCT之间的相关性很强，建议在科研和临床环境中使用，但需要更大样本量的研究。
