Ont Health Technol Assess Ser. 2005;5(5):1-57. Epub 2005 Apr 1.
Computed tomography (CT) scanning continues to be an important modality for the diagnosis of injury and disease, most notably for indications of the head and abdomen. (1) According to a recent report published by the Canadian Institutes of Health Information, (1) there were about 10.3 scanners per million people in Canada as of January 2004. Ontario had the fewest number of CT scanners per million compared to the other provinces (8 CT scanners per million). The wait time for CT in Ontario of 5 weeks approaches the Canadian median of 6 weeks. This health technology and policy appraisal systematically reviews the published literature on multidetector CT (MDCT) angiography as a diagnostic tool for the newest indication for CT, coronary artery disease (CAD), and will apply the results of the review to current health care practices in Ontario. This review does not evaluate MDCT to detect coronary calcification without contrast medium for CAD screening purposes.
Compared with conventional CT scanning, MDCT can provide smaller pieces of information and can cover a larger area faster. (2) Advancing MDCT technology (8, 16, 32, 64 slice systems) is capable of producing more images in less time. For general CT scanning, this faster capability can reduce the time that patients must stay still during the procedure, thereby reducing potential movement artefact. However, the additional clinical utility of images obtained from faster scanners compared to the images obtained from conventional CT scanners for current CT indications (i.e., non-moving body parts) is not known. There are suggestions that the new fast scanners can reduce wait times for general CT. MDCT angiography that utilizes a contrast medium, has been proposed as a minimally invasive replacement to coronary angiography to detect coronary artery disease. MDCT may take between 15 to 45 minutes; coronary angiography may take up to 1 hour. Although 16-slice and 32-slice CT scanners have been available for a few years, 64-slice CT scanners were released only at the end of 2004.
There are many proven, evidence-based indications for conventional CT. It is not clear how MDCT will add to the clinical utility and management of patients for established CT indications. Therefore, because cardiac imaging, specifically MDCT angiography, is a new indication for CT, this literature review focused on the safety, effectiveness, and cost-effectiveness of MDCT angiography compared with coronary angiography in the diagnosis and management of people with CAD. This review asked the following questions: Is the most recent MDCT angiography effective in the imaging of the coronary arteries compared with conventional angiography to correctly diagnose of significant (> 50% lumen reduction) CAD?What is the utility of MDCT angiography in the management and treatment of patients with CAD?How does MDCT angiography in the management and treatment of patients with CAD affect longterm outcomes?The published literature from January 2003 to January 31, 2005 was searched for articles that focused on the detection of coronary artery disease using 16-slice CT or faster, compared with coronary angiography. The search yielded 138 articles; however, 125 were excluded because they did not meet the inclusion criteria (comparison with coronary angiography, diagnostic accuracy measures calculated, and a sample size of 20 or more). As screening for CAD is not advised, studies that utilized MDCT for this purpose or studies that utilized MDCT without contrast media were also excluded. Overall, 13 studies were included in this review.
The published literature focused on 16-slice CT angiography for the detection of CAD. Two abstracts that were presented at the 2005 European Congress of Radiology meeting in Vienna compared 64-slice CT angiography with coronary angiography. The 13 studies focussing on 16-slice CT angiography were stratified into 2 groups: Group 1 included 9 studies that focused on the detection of CAD in symptomatic patients, and Group 2 included 4 studies that examined the use of 16-slice CT angiography to detect disease progression after cardiac interventions. The 2 abstracts on 64-slice CT angiography were presented separately, but were not critically appraised due to the lack of information provided in the abstracts. 16-SLICE COMPUTED TOMOGRAPHY ANGIOGRAPHY: The STARD initiative to evaluate the reporting quality of studies that focus on diagnostic tests was used. Overall the studies were relatively small (fewer than 100 people), and only about one-half recruited consecutive patients. Most studies reported inclusion criteria, but 5 did not report exclusion criteria. In these 5, the patients were highly selected; therefore, how representative they are of the general population of people with suspicion if CAD or those with disease progression after cardiac intervention is questionable. In most studies, patients were either already taking, or were given, β-blockers to reduce their heart rates to improve image quality sufficiently. Only 6 of the 13 studies reported interobserver reliability quantitatively. The studies typically assessed the quality of the images obtained from 16-slice CT angiography, excluded those of poor quality, and compared the rest with the gold standard, coronary angiography. This practice necessarily inflated the diagnostic accuracy measures. Only 3 studies reported confidence intervals around their measures. Evaluation of the studies in Group 1 reported variable sensitivity, from just over 60% to 96%, but a more stable specificity, at more than 95%. The false positive rate ranged from 5% to 8%, but the false negative rate was at best under 10% and at worst about 30%. This means that up to one-third of patients who have disease may be missed. These patients may therefore progress to a more severe level of disease and require more invasive procedures. The calculated positive and negative likelihood ratios across the studies suggested that 16-slice CT angiography may be useful to detect disease, but it is not useful to rule out disease. The prevalence of disease, measured by conventional coronoary angiography, was from 50% to 80% across the studies in this review. Overall, 16-slice CT angiography may be useful, but there is no conclusive evidence to suggest that it is equivalent to or better than coronary angiography to detect CAD in symptomatic patients. In the 4 studies in Group 2, sensitivity and specificity were both reported at more than 95% (except for 1 that reported sensitivity of about 80%). The positive and negative likelihood ratios suggested that the test might be useful to detect disease progression in patients who had cardiac interventions. However, 2 of the 4 studies recruited patients who had been asymptomatic since their intervention. As many of the patients studied were not symptomatic, the relevance of performing MDCT angiography in the patient population may be in question. 64-SLICE COMPUTED TOMOGRAPHY ANGIOGRAPHY: An analysis from the interim results based on 2 abstracts revealed that 64-slice CT angiography was insufficient compared to coronary angiography and may not be better than 16-slice CT angiography to detect CAD.
Cardiac imaging is a relatively new indication for CT. A systematic review of the literature was performed from 2003 to January 2005 to determine the effectiveness of MDCT angiography (16-slice and 64-slice) compared to coronary angiography to detect CAD. At the time of this report, there was no published literature on 64-slice CT for any indications. Based on this review, the Medical Advisory Secretariat concluded that there is insufficient evidence to suggest that 16-slice or 64-slice CT angiography is equal to or better than coronary angiography to diagnose CAD in people with symptoms or to detect disease progression in patients who had previous cardiac interventions. An analysis of the evidence suggested that in investigating suspicion of CAD, a substantial number of patients would be missed. This means that these people would not be appropriately treated. These patients might progress to more severe disease and possibly more adverse events. Overall, the clinical utility of MDCT in patient management and long-term outcomes is unknown. Based on the current evidence, it is unlikely that CT angiography will replace coronary angiography completely, but will probably be used adjunctively with other cardiac diagnostic tests until more definitive evidence is published. If multi-slice CT scanners are used for coronary angiography in Ontario, access to the current compliment of CT scanners will necessarily increase wait times for general CT scanning. It is unlikely that these newer-generation scanners will improve patient throughput, despite the claim that they are faster. Screening for CAD in asymptomatic patients and who have no history of ischemic heart disease using any modality is not advised, based on the World Health Organization criteria for screening. Therefore, this review did not examine the use of multi-slice CT for this purpose.
计算机断层扫描（CT）仍是诊断损伤和疾病的重要手段，对头部和腹部疾病的诊断尤为重要。（1）根据加拿大卫生信息研究所最近发布的一份报告，（1）截至2004年1月，加拿大每百万人中约有10.3台CT扫描仪。安大略省每百万人口中的CT扫描仪数量与其他省份相比最少（每百万人口中有8台CT扫描仪）。安大略省CT检查的等待时间为5周，接近加拿大6周的中位数。本卫生技术与政策评估系统回顾了关于多排CT（MDCT）血管造影作为CT最新适应症——冠状动脉疾病（CAD）诊断工具的已发表文献，并将评估结果应用于安大略省当前的医疗实践。本综述不评估用于CAD筛查目的的无造影剂MDCT检测冠状动脉钙化的情况。
与传统CT扫描相比，MDCT能够提供更详细的信息，且能更快地覆盖更大的区域。（2）不断发展的MDCT技术（8排、16排、32排、64排系统）能够在更短的时间内生成更多图像。对于普通CT扫描，这种更快的扫描能力可以减少患者在检查过程中必须保持静止的时间，从而减少潜在的运动伪影。然而，与传统CT扫描仪获取的图像相比，更快的扫描仪所获取的图像对于当前CT适应症（即身体静止部位）的额外临床效用尚不清楚。有迹象表明，新型快速扫描仪可以缩短普通CT的等待时间。利用造影剂的MDCT血管造影已被提议作为冠状动脉造影的微创替代方法来检测冠状动脉疾病。MDCT检查可能需要15至45分钟；冠状动脉造影可能需要长达1小时。虽然16排和32排CT扫描仪已经上市数年，但64排CT扫描仪直到2004年底才推出。
传统CT有许多经过验证的、基于证据的适应症。目前尚不清楚MDCT将如何增加已确立的CT适应症患者的临床效用和管理效果。因此，由于心脏成像，特别是MDCT血管造影，是CT的一项新适应症，本综述重点关注MDCT血管造影与冠状动脉造影相比在CAD患者诊断和管理中的安全性、有效性和成本效益。本综述提出了以下问题：与传统血管造影相比，最新的MDCT血管造影在冠状动脉成像中对显著（管腔狭窄>50%）CAD的正确诊断是否有效？MDCT血管造影在CAD患者的管理和治疗中有何效用？MDCT血管造影在CAD患者的管理和治疗中如何影响长期预后？检索了2003年1月至2005年1月31日发表的文献，寻找聚焦于使用16排或更快CT与冠状动脉造影相比检测冠状动脉疾病的文章。检索到138篇文章；然而，125篇被排除，因为它们不符合纳入标准（与冠状动脉造影比较、计算诊断准确性指标、样本量为20或更多）。由于不建议对CAD进行筛查，因此利用MDCT进行此目的的研究或使用无造影剂MDCT的研究也被排除。总体而言，本综述纳入了13项研究。
已发表的文献主要关注16排CT血管造影检测CAD。在2005年维也纳欧洲放射学会会议上发表的两篇摘要比较了64排CT血管造影与冠状动脉造影。专注于16排CT血管造影的13项研究分为两组：第1组包括9项关注有症状患者CAD检测的研究，第2组包括4项研究16排CT血管造影在心脏介入后检测疾病进展的研究。关于64排CT血管造影的两篇摘要单独列出，但由于摘要中提供的信息不足，未进行严格评估。
16排计算机断层扫描血管造影：采用了STARD倡议来评估专注于诊断测试的研究的报告质量。总体而言，这些研究规模相对较小（少于100人），只有约一半的研究招募了连续患者。大多数研究报告了纳入标准，但有5项未报告排除标准。在这5项研究中，患者是经过高度选择的；因此，他们对于怀疑患有CAD的一般人群或心脏介入后疾病进展患者的代表性如何值得怀疑。在大多数研究中，患者要么已经在服用，要么被给予β受体阻滞剂以降低心率，从而充分提高图像质量。13项研究中只有6项定量报告了观察者间的可靠性。这些研究通常评估从16排CT血管造影获得的图像质量，排除质量差的图像，并将其余图像与金标准冠状动脉造影进行比较。这种做法必然夸大了诊断准确性指标。只有3项研究报告了其测量结果的置信区间。对第1组研究的评估报告显示，敏感性从略高于60%到96%不等，但特异性更稳定，超过95%。假阳性率在5%至8%之间，但假阴性率最高不到10%，最差约为30%。这意味着多达三分之一患有疾病的患者可能会被漏诊。因此，这些患者的病情可能会发展到更严重的程度，需要更具侵入性的治疗。各项研究计算出的阳性和阴性似然比表明，16排CT血管造影可能有助于检测疾病，但排除疾病方面并无用处。通过传统冠状动脉造影测量的疾病患病率在本综述的各项研究中为50%至80%。总体而言，16排CT血管造影可能有用，但没有确凿证据表明它在检测有症状患者的CAD方面等同于或优于冠状动脉造影。在第2组的4项研究中，敏感性和特异性均报告超过95%（有1项报告的敏感性约为80%除外）。阳性和阴性似然比表明，该测试可能有助于检测心脏介入患者的疾病进展。然而，4项研究中有2项招募的是介入后无症状的患者。由于许多研究的患者没有症状，因此在该患者群体中进行MDCT血管造影的相关性可能存在疑问。
64排计算机断层扫描血管造影：基于两篇摘要的中期结果分析表明，与冠状动脉造影相比，64排CT血管造影不足，在检测CAD方面可能并不优于16排CT血管造影。
心脏成像是CT相对较新的适应症。对2003年至2005年1月的文献进行了系统综述，以确定MDCT血管造影（16排和64排）与冠状动脉造影相比检测CAD的有效性。在本报告发布时，尚无关于64排CT任何适应症的已发表文献。基于本综述，医学咨询秘书处得出结论，没有足够的证据表明在诊断有症状的CAD患者或检测既往有心脏介入的患者的疾病进展方面，16排或64排CT血管造影等同于或优于冠状动脉造影。对证据的分析表明，在调查CAD疑似病例时，会有大量患者被漏诊。这意味着这些患者得不到适当的治疗。这些患者的病情可能会发展到更严重的程度，并可能出现更多不良事件。总体而言，MDCT在患者管理和长期预后方面的临床效用尚不清楚。基于目前的证据，CT血管造影不太可能完全取代冠状动脉造影，但在有更确凿的证据发表之前，可能会与其他心脏诊断测试一起辅助使用。如果安大略省将多层CT扫描仪用于冠状动脉造影，那么使用现有的CT扫描仪必然会增加普通CT扫描的等待时间。尽管新一代扫描仪声称速度更快，但不太可能提高患者的检查通量。根据世界卫生组织的筛查标准，不建议对无症状且无缺血性心脏病病史的患者使用任何方式进行CAD筛查。因此，本综述未研究多层CT用于此目的的情况。
