Bach D S, Hepner A, Marcovitz P A, Armstrong W F
Department of Internal Medicine, University of Michigan Medical Center, Ann Arbor.
Am Heart J. 1993 May;125(5 Pt 1):1257-61. doi: 10.1016/0002-8703(93)90992-i.
The problems of population referral bias in the calculation of specificity in diagnostic testing for coronary artery disease have been previously described. Previous studies investigating the sensitivity and specificity of dobutamine stress echocardiography (DSE) have been subject to pretest and posttest referral biases, largely as a result of the requirement for coronary arteriography. This study determines the normalcy rate for DSE by examining a population at statistically low risk for coronary artery disease. The probability of significant coronary artery disease was determined for 828 consecutive patients referred for DSE at the University of Michigan, and groups were identified with < 10% and < 5% probability of disease. Four of 72 patients (5.6%) with a normal baseline echocardiogram and a probability of coronary artery disease of < 10%, and three of 38 (7.9%) with a probability of < 5% were found to have an abnormal DSE, yielding normalcy rates of 94.4% and 92.1%, respectively. The area of abnormality involved the posterior circulation in three of four patients (75%). This study demonstrates that DSE has a normalcy rate of 92% to 94% and is an accurate test for excluding the presence of significant coronary artery disease.
冠状动脉疾病诊断测试中特异性计算的人群转诊偏倚问题此前已有描述。以往研究多巴酚丁胺负荷超声心动图（DSE）敏感性和特异性时，受到了检验前和检验后转诊偏倚的影响，这主要是由于冠状动脉造影的要求所致。本研究通过检查冠状动脉疾病统计学低风险人群来确定DSE的正常率。对密歇根大学连续转诊进行DSE检查的828例患者确定了显著冠状动脉疾病的概率，并识别出疾病概率<10%和<5%的组。在72例基线超声心动图正常且冠状动脉疾病概率<10%的患者中，有4例（5.6%）DSE异常；在38例疾病概率<5%的患者中，有3例（7.9%）DSE异常，正常率分别为94.4%和92.1%。4例患者中有3例（75%）异常区域累及后循环。本研究表明，DSE的正常率为92%至94%，是排除显著冠状动脉疾病存在的准确检查。
