The prevalence of coronary artery disease substantially affects both cardiac and noncardiac surgery. Assuming that biometric data reported from North America are representative for Germany, the following incidences can be estimated: around 1 million out of 8 million patients operated upon each year will suffer from coronary artery disease, and 15,000 of these patients will have a perioperative myocardial infarction. Since a close relationship has been shown between pre-, intra-, and postoperative myocardial ischaemia and postoperative cardiac morbidity and mortality, early diagnosis and therapy of acute perioperative myocardial ischaemia is warranted. The purpose of this review is to weigh critically the various methods for diagnosis of myocardial ischaemia in view of their practicability and cost/benefit relationship in the perioperative setting. The symptoms of angina pectoris are unreliable in the perioperative period, since patients are premedicated preoperatively, without symptoms during anaesthesia, and usually receive analgesics postoperatively. Intraoperative detection of myocardial ischaemia focuses on standard electrocardiography (ECG) with on-line registration of the ST-segment in two leads (usually leads II and V5) and automatic analysis of ST-segment deviation, achieving a sensitivity of 80% in the detection of myocardial ischaemia. Measurement of regional wall motion abnormalities with trans-esophageal echocardiography (TEE) is a more sensitive method of myocardial ischaemia detection compared to ECG. However, several reasons preclude the broader application of this method in the perioperative phase: (1) it lacks validation by an accepted and independent gold standard; (2) there is a wide spectrum of false-positive findings (considerable interindividual variations in left ventricular contraction, bundle branch blocks, hypertension, hypervolemia); (3) changes in the inferior and apical segments of the left ventricle cannot be detected by single-plane TOE. Detection in these segments might be achieved with biplane echocardiography, but few data on this improved technique are presently available; (4) the method is semi-invasive and might be not applicable during periods with a high incidence of myocardial ischaemia, e.g., intubation, the end of anaesthesia, and extubation; (5) anaesthetists seldom fulfil standard guidelines in echocardiography training; and (6) the method is expensive, which also limits its broader application. Cardiokymography, a noninvasive technique, allows analog representation of anterior wall motion. However, this method also has some limitations that restrict its application in the peri-operative period: (1) wall motion registration is only possible during intermittent cessation of ventilation; (2) probe positioning may be difficult; (3) the method cannot be applied in patients with previous anterior myocardial infarction and concomitant wall motion abnormalities; (4) only anterior wall motion abnormalities of the left ventricle can be detected; and (5) the method cannot be applied in patients subjected to thoracic or upper abdominal surgery. The pulmonary artery catheter lacks sensitivity in diagnosing acute myocardial ischaemia, and therefore should not be used in view of the cost/benefit ratio. Decreases in left ventricular ejection fraction during acute myocardial ischaemia as measured by radioactive-labeled erythrocytes and a small gamma-camera (nuclear stethoscope) have not achieved widespread use, mainly for ethical and cost reasons. The rate-pressure product and pressure-rate quotient are too insensitive in predicting myocardial ischaemia compared to TOE (positive predictive values of 34% and 4.5%, respectively). Finally, measurement of metabolic variables (lactate, inosine, hypoxanthine) are of limited value in detecting acute myocardial ischaemia during anaesthesia (invasive character, delayed availability of results). Diagnosis of perioperative myocardial infarction is also difficult for the reasons mentioned above, with more than 60% being asymptomatic. Technetium 99mpyrophosphate scanning with single-photon-emission computed tomography might overcome these difficulties, but this technique is rarely available. Conclusion. Weighing the cost, benefit, and risk ratio, improved detection of perioperative myocardial ischaemia might be achieved by correctly using standard ECG with at least two leads (II and V5) but avoiding common mistakes such as high skin resistance beneath the electrodes, lack of a left precordial lead, lack of stringent criteria of ischaemia, lack of a registration and calibration unit, and insufficient filters.
