Dupuis J Y, Nathan H J, Wynands J E
Department of Anaesthesia, University of Ottawa Heart Institute, Ottawa Civic Hospital, Ontario, Canada.
Can J Anaesth. 1991 Nov;38(8):1055-64. doi: 10.1007/BF03008627.
Perioperative cardiac complications are associated with a high mortality rate. During the preoperative assessment the anaesthetist must determine if the patient is at risk so that appropriate interventions can be made to ensure the best possible outcome. Over the past three decades several indices have been proposed to identify the surgical patient at risk of cardiac morbidity or mortality. These indices can be thought of as diagnostic tests which determine the likelihood of perioperative cardiac complication. In this article the authors review the epidemiological basis of diagnostic tests and apply this theory to three currently available indices of perioperative risk: the American Society of Anesthesiologists physical status classification, the Goldman multifactorial cardiac risk index and a modified version devised by Detsky. A case history describing a surgical patient with unstable angina is first presented to show how different indices can lead to different predictions. These indices are then analysed as if they were diagnostic tests to predict cardiac complications after surgery. The characteristics of diagnostic tests such as sensitivity, specificity, positive and negative predictive values, and receiver operating characteristic curves are described and applied to the indices. Basic principles such as odds ratio, likelihood ratio and their relation to prevalence and post-test probability are also presented. This overview will help the anaesthetist to understand the limitations of risk indices as predictors of adverse cardiac outcome and to apply these scales more effectively to individual patients at different institutions.
围手术期心脏并发症与高死亡率相关。在术前评估期间，麻醉医生必须确定患者是否处于风险中，以便能够采取适当干预措施以确保获得尽可能好的结果。在过去三十年里，已经提出了几种指标来识别有心脏发病或死亡风险的外科手术患者。这些指标可以被视为确定围手术期心脏并发症可能性的诊断测试。在本文中，作者回顾了诊断测试的流行病学基础，并将该理论应用于目前可用的三种围手术期风险指标：美国麻醉医师协会身体状况分类、戈德曼多因素心脏风险指数以及由德茨基设计的一个修改版本。首先呈现一个描述患有不稳定型心绞痛的外科手术患者的病例史，以展示不同指标如何导致不同的预测结果。然后将这些指标当作预测术后心脏并发症的诊断测试进行分析。描述了诊断测试的特征，如敏感性、特异性、阳性和阴性预测值以及受试者工作特征曲线，并将其应用于这些指标。还介绍了诸如比值比、似然比及其与患病率和检验后概率的关系等基本原理。这一概述将有助于麻醉医生理解风险指标作为不良心脏结局预测指标的局限性，并更有效地将这些量表应用于不同机构的个体患者。
