Intra-operative myocardial ischaemia causes significant alterations in cardiac function and is a harbinger of more serious untoward events such as cardiac dysrythmias, pulmonary oedema, myocardial infarction or even death . So far there is no reliable clinical 'gold standard' to establish the presence of myocardial ischaemia . Electrocardiogram (ECG) is still widely used for the detection of myocardial ischaemia in patients with ischaemic heart disease under-going cardiac and non-cardiac surgery . Acute coronary occlusion results in almost immediate cessation of myocardial contraction in the region supplied by the obstructed vessel . The characteristic association between loss of regional myocardial blood flow and muscular function permits segmental contraction abnormalities, detected by transoesophageal echocardiography (TOE), to be used as early sensitive markers for ischaemia and/or infarction [5–7]. Pulmonary capillary wedge pressure (PCWP) (based on the theory that ischaemia leads to a decrease in compliance, resulting in an increase in filling pressure) can also be used to detect intra-operative myocardial ischaemia in patients with ischaemic heart disease . Relatively recently, the pressure rate quotient (PRQ, mean arterial pressure/heart rate) less than one has undergone scrutiny in canine models and human applications . The concept underlying this index is that myocardial oxygen demand is directly related to the heart rate and blood pressure.
Intra-operative segmental wall motion abnor-malities (SWMA) detected by TOE have been shown to be a sensitive indicator of myocardial ischaemia, but are infrequently associated with ECG defined myocardial ischaemia  and haemodynamic changes [8,11]. It has been suggested that evaluation of apical or basal areas by biplane TOE may resolve this discrepancy [12,13]. To determine the incidence and characteristics of SWMA using biplane TOE and to define how changes in SWMA correlate with other proposed evidence of ischaemia: ECG changes, PCWP changes and PRQ <1, 62 consecutive patients with ischaemic heart disease undergoing non-cardiac surgery were monitored continuously.
Materials and methods
Patients selection and anaesthesia
The protocol for this study was approved by our institutional review boards and written informed consent was obtained from each patient. Sixty-four patients (44 men and 20 women) with coronary artery disease (CAD), at risk of CAD or hypertensive heart disease, ages 43–81 years (54 ± 11 years) who were in sinus rhythm with resting heart rate between 55 and 101 beats min−1 (mean 65 ± 14 beats min−1) were enrolled in this study. All patients were in ASA Class II of the multifactorial cardiac risk index for non-cardiac surgery with total predictive points 6–12 . The majority were scheduled for peripheral vascular surgery. Criteria for entry into this study group included the presence of one of the following: (a) defined CAD as indicated by previous myocardial infarction, typical angina, or atypical angina with an ischaemic ECG response to exercise or scintigraphic evidence of a myocardial perfusion defect; (b) high risk of CAD, suggested by previous or current vascular surgery or the presence of at least two of the following cardiac risk factors (in addition to male sex): age >65 years, hypertension, current smoker, serum cholesterol >6.2 mmol L−1, or diabetes mellitus. We defined a previous myocardial infarction using the ECG Minnesota Code criteria . Typical angina was defined as a history of chest pain with at least three of the following four characteristics: substernal location, precipitation by exercise or stress, duration of <15 min, and resolution after rest or nitroglycerin treatment. Atypical angina required two of the characteristics in addition to an ischaemic ECG response to exercise. Patients were excluded from the study if they had abnormal myocardial repolarization (e.g. left bundle branch block, digitalis effect and left ventricular hypertrophy) or non-ischaemic causes of abnormal wall motion (e.g. bundle branch block, ventricular pacing, prosthetic valve, myocarditis or infiltrative disorder of the left ventricle) . The patients were studied at the University Hospital of Vienna and The Glenfield Hospital, Leicester, between February 1993 and May 1996.
Patients were premedicated with diazepam 0.15 mg kg−1 1 h prior to surgery. Anaesthesia was induced with fentanyl (3 μg kg−1) and thiopentone (4 mg kg−1). After orotracheal intubation, facilitated with vecuronium (100 μg kg−1), controlled ventilation was with nitrous oxide in oxygen (FiO2: 40%). Anaesthesia was maintained with increments of remifentanyl, vecuronium and isoflurane up to 1%. All monitoring data were collected prospectively in a standard manner and were then analysed after surgery (off-line).
Echocardiographic image acquisition and analysis
Investigations were carried out with an Hewlett Packard (1500 and 2500, Andover, MA) ultrasound system having a 5 MHz transoesophageal multiplane probe. After induction of anaesthesia intra-operative TOE investigations were performed as a part of routine cardiac monitoring in patients at high risk. To assess the wall motion pattern with a two-dimensional echocardiography both a standard left ventricular transgastric short-axis view at the level of the papillary muscles and transgastric long-axis view were employed.
A real time video tape was edited to obtain samples for analysis. Routine echocardiographic samples of 60s duration were obtained every 10 min throughout the entire intra-operative recording. Additional echocardiographic samples were obtained at certain prespecified times to detect whether anaesthetic and surgical stress had any immediate effect on regional wall motion. For example, in patients undergoing aortic or other arterial surgery, images were also analysed: (a) 5 min before vascular cross-clamping; (b) during cross-clamping; (c) after cross-clamping at 5, 10, 15, and 30 min; (d) during cross-clamp removal, and (e) after cross-clamp removal at 5, 10, 15, and 30 min.
Images in the short-axis and long-axis view were divided into the segments, according to the new recommendation of the American Society of Echocardiography on nomenclature and standards for identification of myocardial wall segments . (One of the justifications for making new segmental divisions is the fact that dividing the short-axis into six segments rather than eight produced a rational relation with other longitudinal views—four-chamber and two-chamber. Furthermore, one can relate the various segments to coronary artery distribution more conveniently). Figure 1 shows the new segment divisions chosen by the Society. Left ventricular transverse (T-scan) short-axis view (SAX) at mid-papillary muscle level and long-axis two-chamber view (2C) in longitudinal scan (L-scan) were used for wall motion analysis. The SAX was divided into the six segments, long-axis was divided into three portions (basal, mid, and apical) and each portions consisted of two segments (anterior, inferior). Mid portion of SAX was excluded from analysis, thus a total of 10 segments were examined. For the SAX radial cords between the central and both systolic and diastolic contours were constructed, where for the long-axis an alternative method was employed in which a combination of perpendicular and radial cords were constructed between a central reference line and the ventricular contour . Using a cinememory loop, biplane images were replayed side by side, within minutes of their acquisition, and were compared and recorded for later off-line analysis. The wall motion was graded by 1–4 score system (Table 1). A two-dimensional echo-episode suggestive of ischaemia was defined when wall motion of any segment deteriorated by two or more grades, lasting >1 min (Table 1) as described elsewhere [5,6,7,16]. Thus, a wall motion score of 1 implies normal systolic function, and increasing scores imply progressively worse ventricular function.
Two lead ECG monitoring and analysis. Continuous two lead ECG monitoring was performed with a two-channel Hewlett-Packard monitor. Two lead II and V5 were used, based on the work of Blackburn and Katigback and popularized by Kaplan and King . Ischaemia was defined as ST segment changes (lasting at least 1 min) from base-line of ≥0.1 mV depression (slope ≤1 mV sec−1) or ≥0.2 mV elevation at 60 msec after J point. The examination was performed continuously on-line by the attending anaesthesiologist. Routine ECG hard-copy samples of 60s duration were obtained every 10 min throughout the entire intra-operative recording. Additional ECG hard-copy samples were obtained at certain prespecified times to detect whether anaesthetic and surgical stress had any immediate effect on ECG.
PCWP monitoring and analysis. A 7.5 fibre optic thermodilution catheter (Swan-Ganz, Baxter, Irvine, CA) was inserted by subclavian vein puncture and advanced to the distal pulmonary artery. A correct position for recording PCWP was confirmed by observing a clear change in the pressure waveform when the balloon was inflated, and by measuring the oxygen saturation of blood at the tip of the catheter. Before induction of anaesthesia, PCWP was measured during normal respiration, and after induction, it was measured while the patient was temporarily disconnected from the ventilator. Routine PCWP samples of 60s duration were obtained every 10 min throughout the entire intra-operative recording. Additional PCWP were obtained at certain prespecified times to detect whether anaesthetic and surgical stress had any immediate effect on PCWP. Analysis of PCWP was performed using changes in PCWP calculated for each patient from one measurement to the next because our aim was to study the relation between acquired ischaemia and changes in PCWP rather than absolute values of PCWP. Values of ≥3 mmHg were considered to be suggestive for ischaemia .
Pressure rate quotient analysis. To clarify the definition of predictors and indicators of myocardial ischaemia, the term predictor is employed whenever onset of PRQ <1 occurs within 15 min of TOE and/or ECG evidence of ischaemia. This would suggest a temporal relation between the occurrence of TOE and/or ECG ischaemia and PRQ <1. An indicator of myocardial ischaemia identifies simultaneous occurrence of ischaemia by PRQ <1 with TOE and/or ECG evidence of ischaemia. Additionally routine PRQ samples were calculated every 10 min throughout the entire intra-operative recording and at certain prespecified times to detect whether anaesthetic and surgical stress had any immediate effect on PRQ.
All haemodynamic data were stored on a Hewlett-Packard Vectra Computer on Microsoft (Redmond, WA) Excel spreadsheet.
Statistical analysis. Data are presented as means ±SD. The haemodynamic parameters were analysed statistically using repeated-measurements one-way analysis of variance. Mean values were compared with the results of Dunnet's multiple range test. Significance was defined as P<0.05. Calculations: Equations 1 and 2Table 2
New SWMAs were detected in 16 of 64 patients (25%) using biplane TOE and in 12 patients (19%) using transverse plane TOE. The results are shown in Table 1: in the comparison of monoplane TOE and biplane TOE, there were 12 true positives, 48 true negatives, 0 false positive, and 4 false negatives. This yielded a sensitivity of 75%, a specificity of 100%, a positive predictive value of 100% and a negative predictive value of 92%.
In the comparison of biplane TOE and two lead ECG, there were 9 true positives, 47 true negatives, 1 false positive and 7 false negatives. This yielded a sensitivity of 56%, a specificity of 98%, a positive predictive value of 90% and a negative predictive value of 87%.
In the comparison of biplane TOE and PCWP in detecting myocardial ischaemia, there were 4 true positives, 45 true negatives, 3 false positives and 12 false negatives. This yielded a sensitivity of 25%, a specificity of 93%, a positive predictive value of 57% and a negative predictive value of 79%.
Finally, in the comparison of biplane TOE and PRQ <1 for the detection of myocardial ischaemia, there were 3 true positives, 44 true negatives, 4 false positives and 13 false negatives. This yielded a sensitivity of 43%, a specificity of 92%, a positive predictive value of 19%, and a negative predictive value of 77% (Table 1).
The detailed examination showed the following three characteristics: (a) New SWMA detected only in the apical area were not associated with ECG ischaemia (0%), however, those detected in both views were significantly (P<0.01) associated with ECG ischaemia (82%); (b) in a greater number of segments was associated with ECG ischaemia (P<0.01); (c) in the anteroseptal segment was (P<0.01) related to V5 ECG changes. Examples of acute myocardial ischaemia using transgastric left ventricular short-axis and transgastric long-axis view are presented in Fig. 2 and Fig. 3, respectively. Importantly, SWMA precede the changes in ECG, pulmonary artery catheter and PRQ by 3.0 ± 1.2, 4.5 ± 2.1 and 4.5 ± 2.5 min, respectively.
To determine the effects of inter-observer variability over time, 160 scrambled samples from patients were reread by the two observers, by consensus. There was 92% agreement in correctly identifying an ischaemic episode. Inter- and intra-observer variability during the study was determined by having both observers independently analyse a preselected tape of 40 samples twice. A discrepancy between observations was defined by a difference of two or more grades in scoring a segment. The degree of inter-observer variability was 4%. Intra-observer variability was 3% (for both observers).
Relatively recent work has highlighted the common and deleterious, yet often unrecognized, nature of peri-operative myocardial ischaemia . The major reason for seeking aggressively ischaemic events is to prevent these deleterious effects on myocardial ischaemia: pulmonary oedema, haemodynamic instability, dysrythmias and myocardial infarction. Although not proved, it is hypothesized that earlier recognition of myocardial ischaemia leads to earlier treatment which than prevents the above mentioned consequences. Of the ischaemia monitoring commonly employed in the operating room (ECG, pulmonary artery catheter and TOE), the ECG is dominant. Certain prerequisites must be met to use the ECG effectively. The diagnostic mode allows detection of ST changes which are filtered out by the monitoring mode . The number of ECG leads and their location affect detection of ischaemic events . The third requirement is that of immediate availability of a hard copy of the ECG for more detailed analysis. Studies have reported that even trained observers recognize only 15–40% of ECG ischaemic events displayed on an oscilloscope .
Our data are in agreement with other authors using ECG and monoplane (T-scan) [1,5,7] and biplane [12,13] TOE during anaesthesia. Smith et al. compared surface electrocardiography and two-dimensional echocardiography for detection of myocardial ischaemia in 50 anaesthetized patients known to have coronary artery disease. Twenty-four of the 50 patients (48%) developed new segmental wall motion abnormalities intra-operatively; however, only six (12%) developed ST changes on ECG. ST changes were always accompanied by wall motion abnormalities, which developed before or simultaneously with the ST changes. Of the three patients who developed peri-operative myocardial infarction, only one had intraoperative ST changes whereas all had persistent new wall motion abnormalities.
In non-cardiac surgical patients at risk for CAD, investigators comparing monoplane TOE, two-lead Holter, and 12-lead continuous ECG monitoring found that two-lead Holter detected almost twice as many episodes of ischaemia as the 12-lead ECG or monoplane TOE . They found that the value of monoplane TOE was small (odds ratio, 2.6; 95% confidence interval, 1.2–5.7; P = 0.02) and concluded that monitoring for myocardial ischaemia with (monoplane) TOE during non-cardiac surgery has little increased clinical value in identifying patients at high risk for peri-operative ischaemic. However, concordance in the diagnosis of myocardial ischaemia was poor, with only 10% concordance for all three monitors, 16% between TOE and two-lead Holter or 12-lead ECG, and 37% between two-lead Holter and 12-lead ECG. The apparently superior sensitivity of two-lead Holter and greater concordance between the two ECG techniques are likely to be because of differences between the monoplane TOE and ECG protocols and the patient population. That is, the two-lead Holter and 12-lead ECG were monitored continuously whereas monoplane TOE was evaluated only intermittently, and TOE images were analysed by review of video tape recordings, not by the more reliable technique of cine loop analysis. Additionally, only about one-half of these patients had proved CAD—a factor critical in interpreting the results, because as the prevalence of a disease decreases in a population, the risk of false/positive findings increases. For example, two-lead Holter reveals ST segment changes diagnostic of ischaemia in 7% of healthy young adults and in 26% of healthy young adults given digitalis [23,24]. Thirteen per cent of patients in the study just described were receiving digitalis.
It is clear that ischaemia confined to the apex or base of the left ventricle will be missed if only the midpapillary muscle cross-section is monitored [12,25]. At the same time, the apex is an extremely important region to evaluate accurately because apical wall motion abnormalities are common with disease of the left anterior descending coronary artery. Ischaemia confined to the apex or base of the ventricle will be missed if only the SAX is monitored because ECG leads II and V5 are not sensitive for detecting apical ischaemia. Hegger et al. correlated regional wall asynergy with the site of myocardial infarction in 37 patients. The most frequently involved segments were those in the mid-papillary muscle at the short-axis view, followed by the apical segment. Biplane TOE using transverse and longitudinal scans provides an advantage for accurate evaluation regional wall motion abnormalities. Matrix biplane real time imaging (in which elements are arranged and electronically steered in a way permitting simultaneous image acquisition in both planes), is not currently available, but looks promising.
In comparing changes in pulmonary capillary wedge pressure with new-onset SWMA as viewed on TOE, van Daele et al. showed that, in 98 patients anaesthetized for CABG, the onset of ischaemia was associated with a small yet significant increase in pulmonary capillary wedge pressure. Owing to wide variations in pulmonary capillary wedge pressure changes, however, its value as a monitoring technique was deemed limited. A number of other studies report that pulmonary artery catheter is an insensitive monitor and should not be inserted with this as a primary indication .
In a study by Harris et al. an evaluation of PRQ and its relation to TOE-defined myocardial ischaemia, PRQ <1 was compared with the presence of SWMA perioperatively. Despite a difference in patients selection and in methodology, they found PRQ <1 associated with only 20% SWMA episodes occurring during their entire study. The original work by Buffington , which popularized the PRQ, was performed in a canine model in which alterations in blood flow in the distribution of the left circumflex artery were evaluated by sonomicrometry. An experimental stenosis was created, blood pressure was maintained with a phenylephrine infusion and heart rate was controlled with ventricular pacing. During varying combinations of blood pressure and heart rate, systolic wall thickening was determined for the area under question. He suggested that ischaemic dysfunction was unlikely if the mean blood pressure exceeded the heart rate (PRQ >1). However, extrapolation of these findings to the clinical situation should be taken with caution. In many patients CAD can involve multiple coronary arteries and variable aetiologies for stenosis. Lastly, it has been shown that most intra-operative ischaemia is related to decreased supply rather than increased demand. That is, most intra-operative ischaemic events occur with relatively normal haemodynamic values. Thus, indices which are based on increased demand (PRQ or pressure rate product) will have little sensitivity for intra-operative ischaemia.
We conclude that neither changes in electro-cardiography leads II and V5 nor in haemodynamic performance match those of SWMA for the detection of the onset of myocardial ischaemia. Whether intra-operative biplane TOE monitoring alters patient out-comes needs to be addressed by a prospective, randomized trial examining the effect of each intra-operative monitor on clinical management and out-come.
1 Hollenberg M, Mangano DT, Browner WS et al.
Predictors of postoperative myocardial ischemia in patients under-going noncardiac surgery. JAMA
2 Lowenstein E. Review of recent information on myocardial ischemia. In: ASA 1996 Annual Refresher Course Lectures.
Philadelphia: Lippincott-Raven, 1996: 234.
3 Eisenberg MJ, London MJ, Lening JM. Monitoring for myocardial ischemia during noncardiac surgery. A technology assessment of transesophageal echocardiography and 12-lead electrocardiography
4 Waters DD, da Luz P, Wyatt HL, Swan HLC, Forrester JS. Early changes in regional and global left ventricular function induced by graded reduction in regional coronary perfusion. Am J Cardiol
5 Leung JM, O'Kelly BF, Mangano DT, SPI Research group. Relationship of regional wall motion abnormalities to hemodynamic indices of myocardial oxygen supply and demand in patients undergoing CABG surgery. Anesthesiology
6 Voici P, Bilotta F, Aronson S et al.
Echocardiographic analysis of dysfunctional and normal myocardial segments before and immediately after coronary artery bypass graft surgery. Anesth Analg
7 Öwall A, Echrenberger J, Brodin A. Myocardial ischaemia as judged from transesophageal echocardiography and ECG in the early phase after coronary artery bypass surgery. Acta Anaesth Scand
8 van Daele MERM, Sutherland GR, Mitchell MM et al.
Do Changes in pulmonary capillary wedge pressure adequately reflect myocardial ischemia during anesthesia. Circulation
9 Buffington CW. Hemodynamic determinants of myocardial dysfunction in the presence of coronary stenosis in dogs. Anesthesiology
10 Leung JM, O'Kelly B, Browner WS et al.
Prognostic importance of postbypass regional wall motion abnormalities in patients undergoing coronary artery bypass graft surgery. Anesthesiology
11 Harris SN, Gordon MA, Urban MK, O'Konor TZ, Barash PG. The pressure rate quotient is not an indicator of myocardial ischemia in humans. An echocardiographic study. Anesthesiology
12 Rouin-Rapp K, Ionescu P, Balea M, Foster E, Cahalan MK. Detection of intraoperative segmental wall motion abnormalities by transesophageal echocardiography: The incremental value of additional cross sections in the trans-verse and longitudinal planes. Anesth Analg
13 Shah PM, Kyo S, Matsumura M, Omoto R. Utility of biplane transesophageal echocardiography in left ventricular wall motion analysis
. J Cardiothorac Vasc Anesth
14 Goldman L, Caldera DL, Nussbaum SR. Multifactoral index of cardiac risk in noncardiac surgical procedures. N Engl J Med
15 Blackburn H, Keys A, Simonson E. The electrocardiogram in population studies: a classification system. Circulation
16 Kolev N, Zimpfer M. Transesophageal Echocardiography.
Wien, New York: Springer, 1997: 136–145.
17 Schiller N, Shah P, Crawford M et al.
Recommendations for quantification of the left ventricle by two-dimensional echocardiography
. J Am Soc Echocardiogr
18 Kaplan JA, King SB. The precordial electrocardiographic lead V5 in patient who have coronary disease. Anesthesiology
19 Slogoff S, Keats AS, David Y. Incidence of perioperative myocardial ischemia detected by different electrocardio-graphic systems. Anesthesiology
20 Holleneberg MJ, London MJ, Leung JM et al.
Monitoring myocardial ischemia during noncardiac surgery: A technology assessment of transesophageal echocardiography and 12-lead electrocardiography
21 Biagini A, Labbate A, Testa R. Unreliability of conventional visual electrocardiographic monitoring for detection of transient ST segment changes in a coronary care unit. Eur Heart J
22 Smith JS, Cahalan MK, Benefiel DJ. Intraoperative detection of myocardial ischemia in high-risk patients: electrocardiography
vs. two-dimensional transoesophageal echocardiography. Circulation
23 Voller H, Andraseu D, Brügginanu T, Jeveezek M, Becker B, Schröder R. Transient ST segment depression during Holter monitoring: how to avoid false positive findings? Am Heart J
24 Mooss AN, Prevedel J, Mohiuddius, Hilleman DE, Sketch M. Effect of digoxin on ST segment changes detected by ambulatory electrocardiographic monitoring in healthy subjects. Am J Cardiol
25 Kolev N, Huemer G, Ihra G, Spiss CK, Hammerle A, Zimpfer M. Improved detection of perioperative myocardial ischemia with multiplane (Hewlett Packard) transesophageal scanning: Two-dimensional biplane and transmitral Doppler echocardiography. J Cardiovasc Technol
26 Hegger JJ, Weyman AL, Wann LS, Dilon JC, Feigenbaum H. Cross sectional echocardiography in acute myocardial infarction: detection and localization of regional left ventricular asynergy. Circulation
27 Haggmark S, Hohner P, Östman M et al.
Comparison of hemodynamic, electrocardiographic mechanical and metabolic indicators of intraoperative ischemia in vascular surgical patients with coronary artery disease. Anesthesiology
Keywords:© 1997 European Academy of Anaesthesiology
Two-dimensional echocardiography; Wall motion analysis; Electrocardiography; Haemodynamics