A Practical Approach to Cardiac Anesthesia (Practical Approach Series) 5th Ed.

11 Anesthetic Management of Myocardial Revascularization

Michael S. Green, Gary S. Okum, and Jay C. Horrow


 1. The total annual cost of cardiovascular disease in the U.S. is close to 400 billion representing 17% of the total health care costs due to major illness.

 2. Risk factors for myocardial revascularization surgery include history of CHF, EF <30%, advanced age, obesity, emergency surgery, concomitant valve surgery, prior cardiac surgery, DM, and renal failure.

 3. Supply of oxygen to the myocardium is determined by arterial oxygen content of blood and coronary blood flow.

 4. A doubling of heart rate more than doubles the myocardial O2 demand. The reason is heart rate causes a concomitant small increase in contractility.

 5. Elevated wall stress also increases myocardial O2 demand. This is related by the formula: Wall stress = pressure × Radius/(2 × wall thickness).

 6. Regional wall motion abnormalities develop in less than 1 minute after the onset of ischemia: sooner than ECG evidence of ischemia.

 7. Off-pump coronary bypass compared to conventional CABG shows no difference in mortality and quality of life variables. However, duration of ventilation, hospital stay, and morbidity are decreased with off-pump coronary bypass.

 8. Multiple causes of myocardial ischemia can occur during the prebypass, bypass, and postbypass periods.

 9. A brief period of myocardial ischemia may protect against the damage caused by subsequent prolonged ischemia and tissue reperfusion. This ischemic preconditioning has been demonstrated in both animal and human studies.

I. Introduction

   A. Prevalence and economic impact of coronary artery disease. Although the death rate for coronary heart disease has declined over the last several decades, it is still the leading cause of death in the United States. In the United States, more than 16 million people have a history of angina pectoris, myocardial infarction, or both [1]. More than 400,000 coronary artery bypass graft (CABG) revascularization procedures are performed annually [1].


       The economic consequence of coronary heart disease is enormous. It is estimated that the total annual cost of cardiovascular disease in the United States is close to $400 billion, which represents almost 17% of the total health-care costs due to major illnesses [2]. Additionally, coronary heart disease accounts for almost a fifth of all disability disbursements by the Social Security Administration [1].

   B. Symptoms and progression of coronary artery disease. A complete description of angina pectoris and other symptoms related to coronary artery disease is given in Chapter 3. Unlike the usually predictable time course and progression of symptoms in patients with valvular heart disease, patients with coronary artery disease may have variable onset of symptoms as well as progression of disease characterized by discrete events such as angina or myocardial infarction. Many patients suffer ischemia without symptoms; these “silent” ischemic events require diligence for detection and prompt treatment before and after operation. Only 18% of myocardial infarctions are preceded by longstanding angina [1]. All aspects of preoperative evaluation of these patients (i.e., exercise stress testing and cardiac catheterization) are discussed in Chapter 3.

   C. Historical perspective of CABG. Early unsuccessful or suboptimal attempts at myocardial revascularization took place before the 1960s. In 1967, Favalaro and Effler at the Cleveland Clinic began performing reversed saphenous vein bypass grafting procedures. In 1968, Green performed an anastomosis of the internal mammary artery (IMA) directly to a coronary artery. There was a resurgence of interest in the IMA grafting procedure in the late 1970s and early 1980s after a number of studies showed far greater graft patency rates for IMA grafts compared with saphenous vein grafts. In addition, better long-term survival was evident in patients receiving IMA grafts, regardless of ventricular function.

       An interest in enhanced recovery (“fast track”) cardiac surgery, fueled perhaps by concern over resource utilization, led to recognized improvements in clinical outcome [3]. In select cases, the advent of new myocardial stabilization devices obviated the need for exposure to cardiopulmonary bypass for myocardial revascularization. In the 1990s, many centers experimented with minimally invasive revascularization techniques and transmyocardial revascularization. During this time, transesophageal echocardiography (TEE) established an increasing role during coronary revascularization.

       The extent of physiological derangement of patients presenting for cardiac surgery continues to increase, due to an aging patient population, increasing frequency of reoperations, and application of multivessel angioplasty and intracoronary stents to those with lesser degrees of disease.

   D. Evaluating risk of morbidity and mortality for CABG surgery

     1. Introduction. A comprehensive review of the myriad risk stratification schema for CABG surgery is beyond the scope of this text; however, a brief discussion of some of the more popular risk stratification tools follows.


     2. Risk factor models. Although risk stratification tools may differ in the specific weights assigned to certain risk factors and not all of the risk factors appear in every study, certain factors appear associated with increased risk of morbidity and mortality. These factors include poor left ventricular (LV) function (history of congestive heart failure or LV ejection fraction less than 30%), advanced age, obesity, emergency surgery, concomitant valve surgery, prior cardiac surgery, history of diabetes, and history of renal failure [4,5].

          The composite score of the individually weighted risk factors is associated with a certain predicted risk of short- and long-term morbidity and mortality, length of stay, and hospital costs. Table 11.1presents several risk stratification tools.

Table 11.1 Risk factor inclusion in various risk stratification models for coronary artery bypass grafting

     3. Model evaluation. No model can completely capture the risk of caring for sicker patients. The ACC/AHA assigns a level IIa recommendation (level of evidence C) to the use of models for prediction of morbidity and mortality [6]. Some models do not allow for adequate flexibility with regard to the dynamic nature of the patient’s physiology during the preoperative period. For example, a patient’s LV ejection fraction at the time of cardiac catheterization will be used in the evaluation process instead of that derived from TEE at the time of surgery.

II. Myocardial oxygen supply

   A. Introduction. The viability and function of the heart depend upon the relatively delicate balance of oxygen supply and demand. The cardiac anesthesiologist must understand the intricacies of this relationship so as to manipulate their determinants perioperatively. The myocardium maximally extracts O2 from arterial blood at rest: Coronary sinus blood PO2 is 27 mm Hg and its saturation is less than 50%. With exertion or hemodynamic stress, the only way the O2 supply can increase acutely to meet the myocardial energy demand is by increasing coronary blood flow (CBF). Ischemia occurs when CBF does not increase to a level sufficient to meet myocardial demand, leading to anaerobic metabolism. The following approach achieves the clinical goal of ensuring that O2 supply at least matches demand:

      1. Optimize the determinants of myocardial O2 supply and demand

      2. Select anesthetics and adjuvant agents and techniques according to their effects on O2 supply and demand

      3. Monitor for ischemia to detect its occurrence early and intervene rapidly

   B. Coronary anatomy

     1. Left main coronary artery. A thorough understanding of the coronary artery anatomy and the distribution of myocardial blood flow allows an understanding of the extent and degree of myocardium at risk for ischemia and infarction during surgery. The blood supply to the myocardium derives from the aorta through two main coronary arteries (see Fig. 11.2), the left and right coronary arteries. The left main coronary artery extends for a short distance (0 to 40 mm) before dividing between the aorta and the pulmonary artery into the left anterior descending artery and the circumflex coronary artery.

     2. Left anterior descending coronary artery. The left anterior descending artery begins as a continuation of the left main coronary artery and courses down the interventricular groove, giving rise to the diagonal and septal branches. The septal branches vary in number and size, and provide the predominant blood supply to the interventricular septum. The septal branches also supply the bundle branches and the Purkinje system. One to three diagonal branches of variable size distribute blood to the anterolateral aspect of the heart. The left anterior descending artery usually terminates at the apex of the LV.

     3. Circumflex coronary artery. The circumflex artery courses the left atrioventricular groove giving rise to one to three obtuse marginal branches, which supply the lateral wall of the LV. In 15% of the patients, the circumflex artery gives rise to the posterior descending coronary artery (“left dominant”). In 45% of the patients, the sinus node artery arises from the circumflex distribution.

     4. Right coronary artery. The right coronary artery traverses the right atrioventricular groove. It gives rise to acute marginal branches that supply the right anterior wall of the right ventricle (RV). In approximately 85% of the individuals with a right dominant system, the right coronary artery gives rise to the posterior descending artery to supply the posterior inferior aspect of the LV. Thus, in the majority of the population, the right coronary artery supplies a significant portion of blood flow to the LV, while in the other 15% of the population, the posterior-inferior aspect of the LV is supplied by the circumflex coronary artery (left dominant system) or both right coronary and circumflex arteries (codominant system). The sinus node artery arises from the right coronary artery in 55% of patients. The atrioventricular node artery derives from the dominant coronary artery and is responsible for blood supply to the node, the bundle of His, and the proximal part of the bundle branches.


   C. Determinants of myocardial oxygen supply. In broad terms, the supply of oxygen to the myocardium is determined by the arterial oxygen content of the blood and the blood flow in the coronary arteries.

     1. O2 content =(hemoglobin) (1.34) (% saturation) + (0.003) (Po2)

          Ensuring maximal O2 content therefore involves having a high hemoglobin level, highly saturated blood, and a high PO2. Normothermia, normal pH, and high levels of 2,3-diphosphoglyceric acid all favor tissue release of O2.

     2. Determinants of blood flow in normal coronary arteries. CBF varies directly with the pressure differential across the coronary bed (coronary perfusion pressure [CPP]) and inversely with coronary vascular resistance (CVR): CBF = CPP/CVR. However, CBF is autoregulated (i.e., resistance varying directly with perfusion pressure) so that flow is relatively independent of CPP between 50 and 150 mm Hg but is pressure dependent outside of this range. Metabolic, autonomic, hormonal, and anatomic parameters alter CVR, and hydraulic factors influence CPP. Coronary stenoses also increase CVR.

        a. Control of CVR. Factors affecting CVR are outlined in Table 11.2.

Table 11.2 Control of coronary vascular resistance

           (1) Metabolic factors. When increased coronary flow is required secondary to increased myocardial workload, metabolic control factors are primarily responsible. Hydrogen ion, CO2, lactate, and adenosine all may play a role in metabolic regulation of CBF by inducing changes in CVR [7].

           (2) Autonomic nervous system. The coronary arteries and arterioles possess α- and β-receptors. In general, α1-receptors are responsible for coronary vasoconstriction, whereas β-receptors mediate a vasodilatory effect. α2-Receptors on endothelial cells and muscarinic signaling appear to be involved in a nitric oxide-mediated decrease in coronary vascular tone [8]. An increased population of α-receptors may cause episodes of coronary spasm in individuals with nonobstructed coronaries. α1-Mediated constriction of the coronary circulation may counter some of the metabolic vasodilation, especially in the resting basal state. However, under most circumstances such as increasing demand or ischemia, metabolic control factors will over-ride α-mediated vasoconstriction.

           (3) Hormonal factors. Two stress hormones, vasopressin (antidiuretic hormone) and angiotensin, are potent coronary vasoconstrictors. However, blood levels of these hormones during major stress may be insufficient to produce significant coronary vasoconstriction. Thromboxane participates in thrombosis and coronary vasospasm during myocardial infarction. Prostaglandin I2 (PGI2) decreases coronary vascular tone.

           (4) Endothelial modulation. Nitric oxide triggers a cyclic guanosine monophosphate (c-GMP)–mediated vasodilatory effect on vascular smooth muscle, and may also contribute to vessel patency and blood flow by inhibiting platelet adhesion. PGI2 and endothelium-derived hyperpolarizing factor also cause relaxation of vascular smooth muscle.

           (5) Anatomic factors

             (a) Capillary/myocyte ratio. Only three- to four-fifths of available myocardial capillaries function during normal conditions. During exercise, episodes of hypoxia, or extreme myocardial O2demand, the additional unopened capillaries are recruited and increase blood flow, causing a decrease in CVR and in diffusion distance of O2 to a given myocyte. This adaptation, along with coronary vasodilation, contributes to coronary vascular reserve.

             (b) Coronary collaterals. Coronary collateral channels exist in the human myocardium. Under most circumstances, they are nonfunctional. However, in the presence of impeded CBF, these coronary channels may enlarge over time and become functional.

           (6) Other factors affecting CVR. CVR may be partly regulated by myogenic control of vessel diameter, which dynamically responds to the distending pressure inside the vessel. CVR increases with the increased blood viscosity caused by high hematocrit or hypothermia. Thus, hemodilution should accompany induced hypothermia. There is a transmural gradient of vascular tone, with vascular resistance being lower in the subendocardium than in the subepicardium [7].

        b. Hydraulic factors and subendocardial blood flow

           (1) LV subendocardial blood flow. Unlike CBF in the low-pressure RV system, LV subendocardial blood flow is intermittent and occurs only during diastole. Because of the increased intracavitary pressure and excessive subendocardial myocyte shortening, subendocardial arterioles are essentially closed during systole. Of the total LV coronary flow, 85% occurs during diastole and 15% occurs in systole (primarily in the epicardial region). Thus, the majority of blood flow to the epicardial and middle layers of the LV and all the blood flow to the endocardium occur during diastole.

           (2) CPP. Equals the arterial driving pressure less the back-pressure to flow across the coronary bed. For the LV, the driving pressure is the aortic blood pressure during diastole. The back-pressure to flow depends on the area of myocardium under consideration. Because the endocardium is the area most prone to ischemia, attention focuses on its flow, and thus the usual formula for CPP uses left ventricular end-diastolic pressure (LVEDP) as back-pressure instead of right atrial pressure, despite the fact that most blood returns to the heart via the coronary sinus:

CPP = aortic diastolic pressure - LVEDP

              Because diastole shortens relative to systole as heart rate increases, subendocardial blood flow is decreased at extremely rapid heart rates. Figure 11.1 demonstrates the total time per minute spent in diastole as a function of heart rate. Elevations in LVEDP (e.g., heart failure, ischemia) also will impede subendocardial blood flow. Thus, to optimize CPP, one should aim for normal-to-high diastolic blood pressure, low LVEDP, and low heart rate.

Figure 11.1 Total time spent in diastole each minute is plotted as a function of heart rate in beats per minute (beats/min). The reduction in diastolic interval leads to diminished LV blood flow as heart rate increases.

     3. Determinants of myocardial blood flow in stenotic coronaries. In addition to the physiologic determinants of myocardial blood flow in normal coronary arteries, stenotic vessels add pathologic determinants of myocardial blood flow. Stenoses increase CVR and decrease CBF. Reduction in CBF in stenotic vessels is a function of the length and degree of stenosis, presence or absence of collaterals, pattern of stenosis, and presence of certain coexisting disease states, such as diabetes mellitus and hypertension, which cause predisposition to microcirculatory pathology and ventricular hypertrophy, respectively.

          In particular, some patients present with a vasospastic component, which may aggravate a fixed lesion or even create anginal symptoms in patients with angiographically clear vessels.

        a. Poiseuille’s law determines the hemodynamic significance of a coronary obstruction in long (segmental) lesions. Given the same decrease in cross-sectional area, a longer segmental stenosis of a coronary artery reduces flow more than a short one.

        b. Because CBF is reduced in proportion to the fourth power of the vessel diameter, a 50% diameter decrease in lumen size decreases flow to 1/16th its initial value, which is hemodynamically consistent with symptoms of angina on exertion. A 75% reduction in diameter at angiography corresponds to a greater than 98% reduction in flow, and corresponds clinically to symptoms of angina at rest.

        c. Sequential lesions in the same coronary artery impact coronary flow in an additive fashion.

        d. With longstanding coronary obstruction, collateral circulation often develops. For low-grade obstructive lesions, these channels supply enough blood flow to prevent ischemia. However, as the degree of coronary stenosis increases, the collateral channels may not be adequate.

        e. Certain patterns of stenoses have important clinical implications related to the amount of myocardium supplied and placed in jeopardy by the stenotic lesion(s). A left main coronary stenosis limits blood flow to a large amount of the LV muscle mass. High-grade, very proximal stenotic lesions of both the circumflex and left anterior descending systems have the same physiologic implications as does a left main stenosis. Prognostically, however, a left main stenosis is grimmer because rupture of a single atheroma will compromise a large amount of myocardium. In addition, similar “left main equivalent” situations may exist when a severely stenosed coronary provides collateral blood flow to a region with a totally occluded vessel (Fig. 11.2). In addition to discrete focal and segmental coronary lesions in graftable vessels, diffuse distal disease may lessen the effectiveness of bypassing proximal coronary obstructions.

Figure 11.2 Two examples of possible left main “equivalency.”  Two-vessel coronary disease with an occluded left anterior descending coronary artery (LAD) and myocardium jeopardized by a right coronary artery (RCA) stenosis (A) or an occluded RCA and myocardium jeopardized by a stenotic LAD (B). Cx, circumflex coronary artery; LCA, left coronary artery. (From Hutter AM Jr. Is there a left main equivalent? Circulation. 1980;62:209, with permission.)

III. Myocardial oxygen demand.

Direct measurement of myocardial oxygen demand is not feasible in the clinical setting. The three major determinants of myocardial O2 demand are heart rate, contractility, and wall stress.

   A. Heart rate. If a relatively fixed amount of O2 were consumed per heartbeat, one would expect the O2 demand per minute to increase linearly with heart rate. Thus, a doubling of heart rate would yield a doubling of O2 demand. In fact, demand more than doubles with a two-fold increase in heart rate. The source of this additional O2 demand is the staircase phenomenon, in which increased heart rate causes a small increase in contractility and increases in contractility mean more consumption of O2 (see Section B below).


   B. Contractility. More O2 is used by a highly contractile heart compared to a more relaxed heart.

     1. Quantitative assessment. Strictly defined, the contractile state of the heart is a dynamic intrinsic characteristic that is not influenced by preload or afterload. Previous attempts to measure contractility using physiologic variable include the rate of rise of LV pressure, (dP/dt), and its value normalized to chamber pressure, ([dP/dt]/P). Neither succeeds. Clinically, it is possible to quantify dP/dtechocardiographically by measuring the rate of rise in the velocity of the mitral regurgitant jet using Doppler technology. Unfortunately, loading conditions and chamber compliance significantly affect the acceleration of the mitral regurgitant jet. Additionally, although mitral regurgitation is frequently observed echocardiographically, it is not universally present. Contractility can be approximated in a load-independent fashion using the slope of the end-systolic pressure–volume relationships of a family of LV pressure–volume loops. This method is usually not available in clinical settings.

     2. Qualitative measures. One can easily observe the contractile state of the heart when the pericardium is open. Remember, though, that the RV is more easily and most often viewed this way, whereas the left is more obscured. TEE provides a means for qualitative estimation of LV contractility. Clinically, we infer good contractility when the arterial pressure tracing rises briskly. However, the shape of the radial arterial tracing is heavily influenced by the system’s resonant frequency, damping by air bubbles, compliance of the arterial tree, reflections of pressure waves from arterioles, and other confounders.

     3. Increased subendocardial myocyte shortening. Myocytes in the subendocardial region undergo more shortening than those in other areas because of geometric factors. Because they operate at a higher contractile state, they have greater oxidative metabolism. Subendocardial vessels, already maximally dilated, cannot respond to increased demand and intermittent limitations of blood flow in the subendocardial region. Thus, myocardial O2 tension falls first here (Fig. 11.3), and this region is more susceptible to ischemia.

Figure 11.3 Relationship of subendocardial O2 supply (represented by myocardial O2 tension) to reductions in CBF. Demonstrated is the increased vulnerability of the subendocardial zone compared to the epicardial zone. (Modified from Winbury MM, Howe BB. Stenosis: regional myocardial ischemia and reserve. In: Winbury MM, Abiko Y, eds. Ischemic Myocardium and Antianginal Drugs. New York, NY: Raven; 1979:59.)


   C. Wall stress. The stress in the ventricular wall depends on the pressure in the ventricle during contraction (afterload), the chamber size (preload), and the wall thickness. The calculation for a sphere (which we shall assume for the shape of the ventricle, for the sake of simplicity) comes from LaPlace’s Law:

Wall stress = pressure . radius/(2 . wall thickness)

     1. Chamber pressure. Oxygen demand increases with chamber pressure. Doubling the pressure doubles the O2 demand. Systemic blood pressure usually reflects the chamber pressure; thus, we equate systemic blood pressure with LV afterload. The heart’s true afterload is more complex because elastic and inertial components also affect ejection. Mean systemic pressure, not peak systolic pressure, correlates with O2 demand. In aortic stenosis, however, the LV experiences very high chamber pressures despite more modest systemic pressures. The clinical goal is to keep afterload (and thus wall stress) low.

     2. Chamber size. Doubling the ventricular volume increases the radius by only 26% (volume varies with the radius cubed). Thus, increased chamber size is associated with more modest increases in O2demand. Nevertheless, because preload determines ventricular size, we desire a low preload to keep wall stress (and thus O2 demand) low. Much of the benefit of nitroglycerin stems from venodilation and its attendant decrease in preload.

     3. Wall thickness. A thicker wall means less stress over any part of the wall. Ventricular hypertrophy serves to decrease wall stress, although the additional tissue requires more O2 overall. Hypertrophy occurs in response to the elevated afterload that occurs in chronic systemic hypertension or aortic stenosis. Although wall thickness is essentially uncontrollable clinically, its effects should be considered. LV aneurysms, seen after transmural infarction, increase wall stress because of their effect on LV volume (radius) and reduced wall thickness.

   D. Summary. The factors that increase O2 demand are increases in heart rate, chamber size, chamber pressure, and contractility. Table 11.3 and Figure 11.4 summarize the relationship between myocardial supply and demand. Note that tachycardia and increases in LVEDP both lead to increased demand and decreased supply of oxygen.

Table 11.3 Regulation of O2 supply and demand

Figure 11.4 Summary of factors that affect myocardial oxygen supply and demand. (Adapted from Crystal GJ. Cardiovascular physiology. In: Miller RD, ed. Atlas of Anesthesia: Vol. VIII. Cardiothoracic Anesthesia. Philadelphia, PA: Churchill Livingstone; 1999;1:1, with permission.)

IV. Monitoring for myocardial ischemia

   A. Introduction. Monitoring for cardiac surgery is discussed in Chapter 4. Typical monitoring for CABG surgery includes the standard American Society of Anesthesiology (ASA) monitors and invasive arterial blood pressure monitoring. Although TEE is used more frequently in current practice, the most recent ASA/SCA practice guidelines uphold guidelines published in 1996 classifying use in patients with increased risk for myocardial infarction as a category II indication [9]. The use of pulmonary arterial (PA) catheters for routine CABG is also controversial. Detection and treatment of intraoperative ischemia are critically important because intraoperative ischemia is an independent predictor of postoperative myocardial infarction [10]. Only half of the intraoperative ischemic events can be related to a hemodynamic alteration and none can be detected by the presence of angina in anesthetized patients. Lactate extraction of a regional myocardial circulatory bed, while diagnostic of ischemia, cannot be routinely measured. Thus, we seek clues that ischemia leaves in its wake: Changes on the electrocardiogram (ECG), PA pressure (PAP) changes, and myocardial wall-motion abnormalities.

   B. ECG monitoring

     1. Introduction. The detection of wall-motion abnormalities by TEE has not led to the replacement of continuous multilead ECG monitoring as a standard monitor of intraoperative ischemia. ECG monitoring is inexpensive, easy to use and read, can be automated, and is available before and during the induction of anesthesia, when the TEE probe is not in place, and may be carried through to the ICU setting, where TEE monitoring is impractical. ECG changes are less sensitive to ischemia: They occur later in the temporal cascade of events that follow myocardial ischemia, especially with less dramatic coronary supply/demand inequality.

     2. ST-segment analysis. Depression of the ST segment of the ECG denotes endocardial ischemia and elevation denotes transmural ischemia. ST-segment changes occur at least 60 to 120 s after the start of ischemia. The reference for the ST segment is usually taken as 80 ms after the J-point, which is the end of the QRS wave (Fig. 11.5). Significant changes are usually defined as 0.1 mV or 1 mm of ST-segment elevation or depression at normal gain. ECG monitoring systems include automated real-time ST-segment analysis. Although this feature constitutes a definite advance in the “human engineering” aspects of ischemia monitoring, the machine is only as smart as the person interpreting its data. Beware of intraventricular conduction delays, bundle branch blocks, and ventricular pacing, all of which can render ST-segment analysis invalid. Check the machine’s determination of where the ST segment occurs: 80 ms after the J-point is not always appropriate.

Figure 11.5 A: ST-segment depression, an indicator of subendocardial ischemia. B: Transmural ischemia, one cause of ST-segment elevation, produces the pattern appearing in the lower tracing.

        a. Differential diagnosis of ST-segment changes. ST-segment elevation may arise from the several causes of transmural ischemia (atherosclerotic disease, coronary vasospasm, intracoronary air), or from pericarditis or ventricular aneurysm (Fig. 11.5). One must also consider improper lead placement, particularly reversal of limb and leg leads, and improper selection of electronic filtering. The diagnostic mode should always be chosen on machines equipped with a diagnostic–monitor mode selection switch.

     3. T-wave changes. New T-wave alterations (flipped or flattened) may indicate ischemia. These may not be detected by viewing the ST segment alone. Likewise, pseudonormalization of the ST segment or T wave (an ischemic-looking tracing in a patient without ischemia changing to a more normal-looking one) may indicate a new onset of ischemia and should be treated appropriately.

     4. Multilead ECG monitoring. Simultaneous observation of an inferior lead (II, III, or aVF) and an anterior lead (V5) provides detection superior to single-lead monitoring, detecting approximately 90% of ischemic events. Ischemia limited to the posterior of the heart is difficult to detect with standard ECG monitoring. Modified chest leads may be necessary when the surgical incision precludes usual placement.

   C. PAP monitoring

     1. General indications for a PA catheter for revascularization procedures. PA catheters provide a conduit for infusions, measurement of blood temperature and chamber pressures, and calculations of cardiac output, vascular resistance, and RV ejection fraction (with special catheters equipped with particularly fast thermistors). Some catheters also measure mixed venous oxygen saturation. Observational studies concluding that PA catheterization does not affect outcome [11,12] remain unconvincing until randomized studies provide validation. Although perhaps unnecessary for the intraoperative management of routine coronary bypass surgery, PA catheters achieve utility postoperatively, when TEE cannot be utilized.

     2. Detection of ischemia: PAPs. The absolute PAP is not diagnostic of ischemia. Pulmonary hypertension, whether primary or secondary to chronic ischemia, hypertension, or to valvular heart disease, is not uncommon. PAPs or pulmonary capillary occlusion pressures commonly exceed upper levels of normal due to chronic obstructive pulmonary disease (COPD), dependent catheter locations in the lung, or mitral stenosis, making measurement less reliable in the absence of ischemia [13]. The shape of the transduced pressure waveform, however, is more predictive. Appearance of a new V wave on the pulmonary capillary occlusion pressure waveform indicates functional mitral regurgitation, which is due to “new” ischemic papillary muscle dysfunction. It may occur before or even in the absence of ECG changes. However, detection of changes on the pulmonary capillary occlusion waveform requires frequent balloon inflation, introducing additional risk because of the possibility of vessel rupture. Often ischemia may be detected by a change in the shape of the PA tracing, obviating the need for frequent wedging.

   D. TEE

     1. General indications for TEE during revascularization procedures. TEE can assess ventricular preload and contractility, detect myocardial ischemia-induced regional wall-motion abnormalities (RWMAs), evaluate the aortic cannulation site, detect concomitant valve pathology, detect the presence and pathophysiologic effect of pericardial effusion, aid the placement of intra-aortic balloon catheters and coronary sinus catheters, and detect the presence of ventricular aneurysms and ventricular septal defects. TEE has become an invaluable clinical tool and has achieved routine use at many institutions. New RWMAs after bypass correlate with adverse outcomes. See Chapter 5 for a detailed discussion of intraoperative TEE.

     2. Detection of myocardial ischemia with TEE


        a. RWMAs

           (1) Introduction. The heart develops abnormal motion in less than 1 min following perfusion defects [14]. RWMAs resulting from myocardial ischemia temporally precede both ECG and PAP changes. TEE simultaneously interrogates regions of the heart representative of all three major coronary arteries, including the posterior wall, which are not easily monitored with ECG. This is, perhaps, most easily accomplished in the short-axis, midpapillary muscle view. Ischemia confined to nonvisualized regions will escape detection. Proper interrogation of the heart requires a comprehensive examination in multiple planes.

           (2) Limitations of monitoring RWMA

             (a) Tethering. Nonischemic tissue that is adjacent to ischemic tissue may move abnormally simply because it is attached to tissue exhibiting an RWMA. This tends to exaggerate the RWMA.

             (b) Pacing/bundle branch blocks. Abnormal ventricular depolarization sequences not only affect ST-segment analysis (see Section IV.B.2), but also alter wall motion and can mimic RWMA.

             (c) Interventricular septum. Normal septal motion depends on appropriate ventricular loading conditions, the presence of pericardium, and normal electrical conduction.

             (d) Stunned myocardium. Adequately perfused myocardium may exhibit RWMA if recovering from recent ischemia, thus prompting inappropriate therapeutic intervention.

             (e) Induction/ICU. TEE use is not practical during induction of anesthesia or for continuous ICU monitoring.

        b. Diastolic LV filling patterns. Unfortunately, ventricular filling patterns depend on ventricular loading conditions and the site of Doppler interrogation within the ventricular inflow tract, limiting utility to monitor ischemia.

        c. Intraoperative stress TEE. Low-dose dobutamine (2.5 μg/kg/min), when administered for 3 to 5 min, improves coronary flow without significantly affecting demand. The resultant improved myocardial energetics improves existing wall-motion abnormalities. Demonstration of contractile reserve using intraoperative stress echo then directs revascularization efforts to regions of myocardium able to benefit from the additional blood supply.

        d. Contrast echocardiography. Intracoronary injection of sonicated albumin allows imaging of perfused myocardium, providing a tool to identify stunned myocardium, thus avoiding needless therapeutic intervention. At present, contrast agents are not approved by the U.S. Food and Drug Administration for this indication. Additionally, there are some technical imaging issues that must be resolved before this technique can become a real clinical tool. A microbubble contrast injection technique has also been described [15].

        e. Detection of infarction complications. TEE can detect complications of ischemia/infarction such as acute mitral insufficiency, ventricular septal defect, and pericardial effusion.

V. Anesthetic effects on myocardial oxygen supply and demand.

Outcome studies in cardiac surgery fail to reveal an effect of particular anesthetic agents [16,17]. Cardiac anesthesiologists, keenly aware of the impact of anesthetic agents on myocardial oxygen supply/demand dynamics, effectively monitor for and treat myocardial ischemia, thus accommodating for any such effects.

   A. Intravenous (IV) nonopioid agents

     1. Thiopental and thiamylal. Induction doses of ultrashort-acting barbiturates decrease systemic vascular resistance (SVR) and cardiac contractility and increase heart rate. Oxygen demand decreases from the first two effects and increases from the third. All three decrease oxygen supply. The net effect on myocardial O2 balance depends on the initial conditions. The hyperdynamic, hypertensive patient benefits from the restoration of more physiologic blood pressure and contractility, whereas a patient with a normal heart rate may respond to the resultant tachycardia with ischemia.

     2. Propofol. The cardiovascular effects of induction doses of propofol are similar to those of the thiobarbiturates: Systemic blood pressure, SVR, and cardiac contractility decrease. Heart rate may increase less with propofol compared to thiopental.

     3. Ketamine. Ketamine increases sympathetic tone, leading to increases in SVR, filling pressures, contractility, and heart rate. Myocardial O2 demand strongly increases, whereas O2 supply only slightly augments, thus producing ischemia. However, the patient already maximally sympathetically stimulated responds with decreased contractility and vasodilation. Ketamine is not recommended for routine use in patients with ischemic heart disease. However, it is sometimes used in patients with cardiac tamponade, because of its ability to preserve heart rate, contractility, and SVR; see Chapter 18.

     4. Etomidate. Induction doses of etomidate (0.2 to 0.3 mg/kg) do not alter heart rate or cardiac output, although mild peripheral vasodilation may lower blood pressure slightly. As such, it is an ideal drug for rapid induction of anesthesia in patients with ischemic heart disease. Etomidate offers little protection from the increases in heart rate and blood pressure that accompany intubation. It is usually necessary to supplement etomidate with other agents (e.g., opioids, benzodiazepines, volatile agents, β-blockers, and nitroglycerin) in order to control the hemodynamic profile and prevent myocardial oxygen supply/demand imbalance. An induction dose blocks adrenal steroidogenesis for 6 to 8 h.

     5. Benzodiazepines. Midazolam (0.2 mg/kg) or diazepam (0.5 mg/kg) may be used to induce anesthesia. Although both agents are compatible with the goal of maintaining hemodynamic stability, blood pressure may decrease more with midazolam because of more potent peripheral vasodilation. Negative inotropic effects are inconsequential. Blood pressure and filling pressures decrease with induction, whereas the heart rate remains essentially unchanged. Addition of induction doses of a benzodiazepine to a moderate-dose opioid technique, however, may result in profound peripheral vasodilation and hypotension.

      6. a2-Adrenergic agonists. Centrally acting α2-adrenergic agonists result in a reduction in stress-mediated neurohumoral responses and therefore are associated with decreases in heart rate and blood pressure [18]. These agents are typically used during maintenance of anesthesia or postoperatively. Preoperative oral clonidine reduces perioperative myocardial ischemia in patients undergoing CABG surgery, but occasionally results in significant intraoperative hypotension. Dexmedetomidine possesses greater α2 selectivity than clonidine. Both agents have sedative and antinociceptive properties. Use of α2-adrenergic agonists is associated with a reduced opioid requirement. Additionally, α2-adrenergic agonists do not result in respiratory depression.

   B. Volatile agents. In general, volatile anesthetics decrease both O2 supply and demand. The net effect on the myocardial supply/demand balance depends upon the hemodynamic profile that prevails at the time of administration.

     1. Heart rate. Sevoflurane has negligible effect on heart rate. Desflurane and isoflurane often increase heart rate, although isoflurane decreases heart rate if its associated decrease in SVR is not profound, if the carotid baroreceptor function is impaired, or if the patient is fully β-blocked. In the steady state, the cardiovascular actions of desflurane are similar to those of isoflurane. However, during induction without opioids, heart rate and systemic and PA blood pressures may increase and require therapeutic intervention.

          Desflurane use for inhalation induction is unwise due to a significant increase in heart rate, particularly with rapid escalation of the inspired concentration. Junctional rhythms may occur with any volatile agent; they deprive the heart of an atrial kick, leading to decreased stroke volume, cardiac output, and CBF, offsetting the salubrious effects of low heart rate.

     2. Contractility. All volatile anesthetics decrease contractility, lowering O2 demand. However, isoflurane, desflurane, and sevoflurane cause less depression than halothane or enflurane. In decompensated hearts, all volatile anesthetics impair ventricular function.

     3. Afterload. Decreases in cardiac output and SVR with volatile anesthesia result in decreased systemic blood pressure. Venodilation and blunted contractility account for the decrease in cardiac output. SVR decreases most with isoflurane and desflurane, moderately with enflurane and sevoflurane, but not at all with halothane administration. Both O2 supply and O2 demand decrease.

     4. Preload. Volatile agents maintain filling pressures. Therefore, CPP (diastolic aortic pressure minus LVEDP) may decrease during volatile anesthesia.

     5. Coronary steal. A coronary “steal” phenomenon has been described in which dilation of normal vascular beds diverts blood away from other beds that are ischemic and thus maximally dilated (Fig. 11.6). Steal-prone anatomy may exist in 23% of the patients undergoing CABG [19]. Coronary steal has been observed in canine models of steal-prone coronary anatomy with isoflurane administration under circumstances that caused collateral flow which is pressure dependent. It is doubtful that isoflurane-induced coronary steal is of significant clinical importance to patients undergoing coronary revascularization surgery as long as hypotension and consequent pressure-dependent coronary artery perfusion are avoided. Coronary steal has not been observed with halothane, enflurane, or desflurane.

Figure 11.6 Theoretical basis of coronary steal. The shaded, marginally ischemic area normally receives barely enough flow (left). A potent vasodilator improves flow to the normal myocardium but does not affect the already maximally dilated area in jeopardy. This process decreases flow through the collateral from the nonischemic bed. This further impoverishment of the marginally ischemic area produces frank ischemia (right). (Modified from Becker LC. Conditions for vasodilator-induced coronary steal in experimental myocardial ischemia. Circulation. 1978;57:1108.)

     6. Preconditioning. Volatile anesthetics may confer a degree of preconditioning-like protective effect against ischemia-reperfusion injury in human myocardial tissue [20] and improve late cardiac events following cardiac surgery [21] (see Section VII.D.6).

   C. Nitrous oxide. The mild negative inotropic effects of nitrous oxide decrease contractility, producing a reduction in both O2 supply and demand. Nitrous oxide inhibition of norepinephrine uptake in the lung may explain the increased plasma norepinephrine levels and associated increase in pulmonary vascular pressures and resistance seen with nitrous oxide administration [22]. Adding nitrous oxide to an opioid-oxygen anesthetic decreases SVR due, in part, to the removal of the vasoconstrictive effects of 100% O2. The sympathomimetic effects of nitrous oxide counterbalance any direct depression of contractility except in patients with poor LV function in whom the myocardium already is highly stimulated intrinsically.

       If nitrous oxide is used in a technique that provides a “light” anesthetic that is inadequate to cover attendant stimulation, increases in SVR and afterload ensue.

   D. Opioids

     1. Heart rate. All opioids except meperidine decrease heart rate by centrally mediated vagotonia; meperidine has an atropine-like effect. The dose of drug and speed of injection affect the degree of bradycardia. The result is decreased O2 demand. By releasing histamine, morphine or meperidine may elicit a reflex tachycardia that decreases O2 supply and increases O2 demand.

     2. Contractility. Aside from meperidine, which decreases contractility, the opioids have little effect on contractility in clinical doses.

     3. Afterload. In compromised patients, who often depend on elevated sympathetic tone to maintain cardiac output and systemic resistance, the loss of sympathetic tone associated with opioid induction of anesthesia may result in a sudden drop in blood pressure and consequent decreases in both O2 supply and demand. Concomitant midazolam use augments the decreased SVR with opioid induction.

     4. Preload. Despite a lack of histamine-releasing properties, fentanyl and sufentanil reduce preload when administered in either moderate doses (25 μg/kg for fentanyl) or larger doses by decreasing intrinsic sympathetic tone. Oxygen demand is decreased.

     5. Hyperdynamic state. Elevations of heart rate, blood pressure, and cardiac output with or without decreased filling pressures are common during pure opioid-oxygen anesthetic techniques in patients with good ventricular function. This high-supply/high-demand state may be less preferable than the low-supply/low-demand state achieved with volatile anesthesia. Additional opioid, which often fails to treat the hypertension associated with a hyperdynamic cardiac state, frequently decreases systemic blood pressure excessively when hypertension originates from increased SVR alone.

   E. Muscle relaxants

     1. Succinylcholine. This drug may cause a variety of dysrhythmias (bradycardia, tachycardia, extrasystoles), which negatively affect myocardial O2 balance.

     2. Pancuronium. Heart rate increases 20% when pancuronium accompanies a volatile anesthetic. With high-dose opioid anesthesia, heart rate remains stable. Occasionally, a patient will develop tachycardia and ischemia during induction or intubation. The latest available survey data indicate pancuronium as the neuromuscular blocker most commonly used in cardiac anesthesia [23].

     3. Vecuronium. Vecuronium has a flat cardiovascular profile that is ideal with a low- or moderate-dose opioid anesthetic supplemented by volatile agent. Bradycardia occurs when it accompanies the rapidinjection of high doses of the highly lipid-soluble opioids.

     4. Rocuronium. Vagolytic action lies intermediate to that of vecuronium and pancuronium. Consequently, tachycardia is possible but less likely than with pancuronium. Rocuronium administration does not result in significant perturbations in other hemodynamic parameters.

     5. Cisatracurium. An isomer of atracurium, lacks significant cardiovascular effects with manifestations of histamine release not seen. The majority of metabolism is via Hofmann elimination with ester hydrolysis making a minor contribution. Evidence supports faster neuromuscular recovery with cisatracurium use versus rocuronium or vecuronium in both adults and children [24,25].

   F. Summary. Volatile anesthesia provides a low-supply/low-demand environment. The opioid-oxygen technique provides a high-supply/high-demand environment. Success with either technique depends on maintaining proper balance, with O2 supply exceeding demand.

VI. Anesthetic approach for myocardial revascularization procedures

   A. Fast-track cardiac anesthesia

     1. Historical perspective

        a. High-dose narcotic technique. Until the 1990s, anesthesia for cardiac surgery largely utilized a high-dose narcotic technique compatible with minimal impairment of cardiac function and, therefore, conducive to the maintenance of perioperative hemodynamic stability. Additionally, it was believed that the delayed awakening associated with high-dose narcotic technique allowed time for the heart to recover from the obligatory ischemia suffered during cardiopulmonary bypass, decreased the release of stress hormones and attendant myocardial ischemia in the immediate postoperative period, and allowed time requisite for temperature and hemostatic homeostasis [26].

           (1) Anesthetic management for high-dose narcotic technique. Typical premedication consists of morphine (0.1 mg/kg administered intramuscularly [IM]) and scopolamine (0.3 to 0.4 mg, IM). Fentanyl (25 to 100 μg/kg) induces anesthesia. Maintenance consists of fentanyl (total cumulative dose of approximately 100 μg/kg) and long-acting muscle relaxant, supplemented as needed with volatile agent, benzodiazepine, or both.

           (2) Indications for high-dose narcotic technique. This technique applies when early extubation is not a realistic goal and when precipitous changes in hemodynamic conditions must be scrupulously avoided, such as when there are surgical concerns about anastomotic or aortic cannulation site integrity.

        b. Impetus for change. Responding to pressure to reduce hospital stays and costs, clinicians developed a technique based on sharply reduced doses of opioids supplemented by volatile agents or short-acting IV agents [27]. Heavy postoperative sedation, a feature of earlier schema thought necessary to avoid myocardial ischemia, has likewise fallen into disfavor [28]. New surgical techniques, improved myocardial protection, warmer bypass temperatures (32 to 34°C), improvements in anesthetic technique, management of perioperative coagulopathy, and temperature homeostasis (e.g., warmer operating room temperatures and use of active warming devices) have supported development of this “fast-track” cardiac anesthesia.

     2. Inclusion guidelines. Early fast-track initiatives excluded patients with obesity, moderate-to-severe pulmonary disease, emergency operations, poor ventricular function, combined procedures, redo operations, or advanced age—about 40% to 60% of surgical revascularization patients. Application to sicker and riskier patients has progressed the technique, so that most patients now have an opportunity for early extubation, the exceptions usually being those with hemodynamic compromise or difficult airways.

     3. Anesthetic management (Table 11.4)

Table 11.4 Typical fast-track cardiac anesthetic at Robert Wood Johnson Medical School

        a. Premedication. Long-acting agents such as scopolamine are preferentially avoided. Same-day admission patients may not have adequate time for proper sedation with slow-onset agents. Midazolam (0.03 to 0.07 mg/kg IV) usually suffices to allay anxiety.

        b. Intraoperative anesthetic agent management

           (1) Induction of anesthesia. Etomidate, propofol, or thiopental accomplishes induction. Etomidate often requires adjuvant agents, such as volatile agent or low-dose opioid, to avoid tachycardia and hypertension with intubation. Esmolol or nitroglycerine can help prevent or treat a hyperdynamic state that breaks through low doses of opioids.

   Intrathecal or epidural analgesia, placed preoperatively, facilitates recovery by decreasing the perioperative opioid requirements [29]. Concern about neuroaxial hematoma formation in the presence of systemic anticoagulation limits the clinical application of regional anesthesia.

           (2) Maintenance of anesthesia. Volatile agents are used freely to limit the total opioid dose to 10 to 15 μg/kg fentanyl IV. The ultrashort-acting opioid remifentanil provides good hemodynamic stability, adequate attenuation of the neurohumoral stress response, and early awakening. However, its short half-life necessitates providing analgesia in the postoperative period. Reduce reliance on a volatile agent in advance of patient transport to the ICU, lest the patient becomes acutely hyperdynamic upon arrival, prompting the use of long-duration sedatives by the ICU staff.

   α2-Agonists (e.g., dexmedetomidine and clonidine) are used adjunctively to reduce the neurohumoral stress response and for their sedative and antinociceptive properties. These agents reduce the anesthetic requirement and thus facilitate more rapid emergence.

        c. Intraoperative awareness. Significant intraoperative awareness occurs in 0.3% of fast-track patients, similar to that observed in general surgery [30]. When using moderate doses of opioids, matching the depth of anesthesia to the operative stimulus at different times will avoid harmful hemodynamic responses; volatile agent provides this flexibility. A vaporizer attached to the cardiopulmonary bypass circuit facilitates appropriate anesthetic depth during bypass with moderate hypothermia. The role of the bispectral index (BIS, Aspect Medical Systems, Natick, MA, USA) remains unclear, although many centers have embraced its use.

        d. Temperature homeostasis. Early extubation is unwise in the hypothermic patient: Arrhythmia, coagulopathy, shivering, and increased oxygen consumption complicate postoperative care and delay discharge [31]. Fluid-filled warming blankets and heated-humidified breathing circuits fail to preserve heat in the face of the huge caloric deficits of cardiac surgical patients. IV fluid warmers significantly impact body heat during cardiac surgery where infused volumes are substantial [32]. The cardiopulmonary bypass circuit heat exchanger provides the best means of restoring body temperature. Target a PA blood temperature greater than 37°C and a bladder temperature greater than 34°C. Forced hot-air convective warming helps during off-bypass revascularization, despite the minimal body surface available. Maintain operating room temperature as warm as is reasonably tolerated by operative personnel to prevent heat loss.

        e. Hemostatic homeostasis. To prevent bleeding after operation, supplement a scrupulous surgical attention to hemostasis with pharmacologic hemostatic agents such as aminocaproic acid or aprotinin (see Chapter 17). Use of these agents is generally restricted to revascularization that involves the use of bypass.

        f. Intraoperative extubation. In the absence of ventricular impairment, hemodynamic instability, hypothermia, or significant coagulopathy, consider extubation in the operating room prior to transport. In one nonrandomized, underpowered study, intraoperative extubation did not reduce the length of stay in the ICU [33]. A more recent study showed that with careful patient selection, immediate extubation offers the opportunity for shorter ICU and hospital stays [34]. Strategic use of opioids at induction and judicious use after sternal wiring produce adequate ventilation and analgesia. One technique combines intrathecal morphine (300 to 500 μg) prior to induction using sufentanil (150 to 250 μg IV) with morphine (0 to 10 mg IV) at sternal reapposition. Rapid awakening depends upon discontinuation of isoflurane or sevoflurane upon placement of sternal wires with substitution by nitrous oxide or desflurane. ICU personnel must tolerate the mild to moderate respiratory acidosis seen in patients with adequate analgesia immediately after extubation. Patients remaining intubated need adequate sedation/anesthesia to last through the transport and for at least 30 min beyond, to avoid hypertension, tachycardia, and bucking in the semicontrolled environment of tangled access lines and rolling beds.

     4. ICU management. Maintain a degree of sedation with a relatively short-acting agent devoid of significant hemodynamic effect until the patient is ready for extubation. For an uncomplicated patient not extubated in the operating room, tailor the level of sedation for extubation 4 to 6 h after arrival, during which time the patient recovers from inadequate myocardial protection during bypass, achieves normothermia, and demonstrates perioperative hemostasis (Fig. 11.7).

Figure 11.7 Early and late complications seen in patients undergoing coronary revascularization in the immediate postoperative period. (Adapted from Higgins T. Pro: Early endotracheal extubation is preferable to late extubation in patients following coronary artery surgery. J Cardiothorac Vasc Anesth.1992;6:488–493.)

          Propofol provides a smooth transition from intraoperative anesthesia to postoperative sedation. It is easy to titrate, with quick onset and offset of action, and minor hemodynamic effects. Supplemental nitroglycerin or nitroprusside infusions titrate blood pressure rapidly.

          Dexmedetomidine is an effective medication in the postoperative period. When administered as a continuous infusion, dexmedetomidine produces sedation that mimics natural sleep. It significantly reduces the use of analgesics, β-blockers, antiemetics, epinephrine, and diuretics [35].

          The literature provides conflicting evidence regarding the relative safety and efficacy of dexmedetomidine compared to propofol for ICU sedation, including endpoints such as analgesic requirement reduction and hemodynamic changes [35,36]. Studies indicate no difference in ventilator weaning time [35].

          Special versions of sedation and weaning protocols for fast-track patients facilitate the goal of early ICU discharge. Point-of-care arterial blood gas monitoring permits rapid decisions regarding weaning from mechanical ventilation. Many institutions use the fast-track anesthesia program as a continuing performance improvement project.

     5. Clinical and economic benefits

        a. Clinical benefits of enhanced recovery. Early extubation may be associated with reduced postoperative lung atelectasis and improved pulmonary shunt fraction [37]. Positive-pressure ventilation has deleterious effects on cardiac output and organ perfusion that may be minimized by early extubation. Early chest tube removal facilitates patient mobilization. Early extubation yields greater patient satisfaction as long as appropriate analgesia is maintained.

        b. Economic benefits. Early extubation protocols reduce ICU stay, hospital stay, and the cost of hospitalization by as much as 25% [34,38,39].

   B. Off-bypass revascularization

  Off-pump coronary artery bypasses (OPCAB) utilize venous or arterial conduits to the coronary arteries via sternotomy but without CPB. See Section VI.D.2 for a discussion of port-access surgery and Section VI.D.3 for minimally invasive directed coronary artery bypasses (MIDCAB).

     1. Technique. Following sternotomy and anticoagulation, the surgeon places distal anastomoses using myocardial stabilizers. With the heart continuing to fill, beat, and contract, a section is immobilized sufficiently to permit vascular sutures for distal anastomoses. An aortic side clamp permits placement of proximal anastomoses. Reapproximation of the sternum and layered closure complete the surgical procedure.

     2. Advantages and disadvantages

        a. Graft patency. Randomized studies comparing OPCAB to conventional CABG with CPB yield differing results [40,41] with respect to short- and long-term graft patency. Grafts to the right coronary circulation appear more likely to lose patency with OPCAB.


        b. Level-IA evidence-based criteria suggest that mortality and quality-of-life markers do not differ, whereas the duration of ventilation, length of stay in the hospital, and overall morbidity decrease with OPCAB.

        c. Somewhat weaker evidence supports decreased blood loss and transfusion requirements, less myocardial enzyme release up to 24 h, less renal dysfunction, coagulation abnormalities, incidence of atrial fibrillation, and impairment of cognitive function with OPCAB [42]. More large randomized studies will better define outcomes.

        d. Hemodynamic instability. Despite frequent manipulation of patient position, volume status, and vasopressors, maintaining a regular rhythm, cardiac output, and blood pressure during performance of circumflex or inferior anastomoses may prove difficult owing to torsion of the heart or geometry incompatible with effective ejection. Ventricular fibrillation may occur suddenly during this time.

      3. Patient selection for OPCAB varies widely among institutions:

        a. Patients with anterior lesions only, thus limiting the risks of hemodynamic compromise and potential subsequent graft occlusion associated with inferior anastomoses

        b. Avoidance of CPB to minimize the risk for stroke and renal failure (i.e., for those patients with large amounts of aortic plaque, severe vascular disease in general, or with renal insufficiency)

        c. None or all. Selection occurs when the patient or referring physician selects the surgeon, who performs all procedures as OPCAB or conventionally.

     4. Preoperative assessment. Infinitely more so than with conventional CABG, with OPCAB, knowledge of the intended placement of grafts and functional reserve of myocardium prepares the anesthesia team for the extent of physiologic trespass and expected impact on homeostasis.

        a. Interruption of flow when grafting proximal stenoses can impact a large amount of myocardium.

        b. Interruption of flow to vessels with high-grade lesions may have little impact owing to formation of collateral circulation.

      5. Monitoring of a given patient for OPCAB should be at least as extensive as for conventional CABG due to the added stresses of cumulative myocardial ischemia during vessel occlusion and grafting. Twisting or flipping the heart changes its electrical axis, making difficult the determination of ischemia via ECG criteria, and also thwarts acquisition of standard images using TEE. Thermodilution will still accurately measure cardiac output.

     6. Fast-track criteria apply. See Section VI.A regarding choice of anesthetic agents and techniques to facilitate early extubation, awakening, and safe transfer from intensive care environments. Occasionally, an OPCAB case converts to one using CPB; for this reason, some caregivers avoid techniques they would not employ in the presence of full heparinization, such as epidural anesthesia.

     7. Temperature. Without the heat exchanger of the CPB machine to transfer calories to the patient, temperature management focuses on prevention of heat loss and use of forced air on available surface, even if only head and shoulders. See Section VI.A.3.d for additional guidance.

      8. Heparin management varies among institutions. Some target an ACT twice that of baseline, as for noncoronary vascular surgery. Others aim for a minimum of 300 s, representing twice the upper limit of normal, whereas some administer doses as they would for the conduct of CPB. Protamine doses should aim to neutralize the heparin administered, because bleeding with OPCAB remains as much a threat to successful surgery as it does with conventional CABG. However, CPB affects platelet function adversely and activates the fibrinolytic system more than does OPCAB.

      9. Specific maneuvers to provide adequate hemodynamics

        a. Intravascular volume loading and head-down tilt (modified Trendelenburg’s position) augment preload to counter obstructed venous return from torqued or flipped cardiac positioning during performance of distal anastomoses.

        b. Vasoconstrictors (phenylephrine or norepinephrine) likewise maintain CPP, maintaining collateral flow during vessel occlusion.

        c. Controlled hypotension. When the surgeon sutures the proximal anastomoses using an aortic side clamp, systolic pressures less than 100 mm Hg will help prevent aortic dissection. Use vasodilators and volatile anesthetics with normovolemia and head-up tilt to accomplish this.

   C. Regional anesthesia adjuncts

     1. Intrathecal opioid. Placement of 200 to 400 μg of preservative-free morphine via lumbar puncture immediately prior to induction of general anesthesia provides lasting, preemptive analgesia. Larger doses increase the risks of postoperative pruritus, nausea, and ventilatory depression, peaking around 6 to 10 h after administration. Minimize the possibility of neuraxial bleeding and compression as follows:

        a. Time the administration of heparin to occur at least 1 h following puncture, especially if blood returns in the spinal needle. Some centers delay the operation by 24 h if a large-bore needle has been used.

        b. Use a small-gauge spinal needle—24 G or smaller.

        c. Do not persist if identification of the intrathecal space proves difficult.

     2. Epidural opioid. After identifying the epidural space, up to 5 mg of preservative-free morphine can provide preemptive analgesia. The larger epidural needle increases the risk of bleeding and of significant hematoma formation.

     3. Precautions. Potent antiplatelet therapy (e.g., clopidogrel) not halted at least 7 days before the day of surgery raises concern in performing neuraxial procedures, as does a therapeutic international normalized ratio (greater than 1.5) from vitamin K antagonist therapy.

      4. Neuraxial preemptive analgesia may facilitate early extubation by facilitating adequate ventilation. Nevertheless, expect a patient breathing spontaneously immediately after a very-early-extubation technique to have an arterial Pco2 of 50 mm Hg, which is similar to that of noncardiac surgical patients with adequate opioid-based analgesia in the PACU.

   D. Special circumstances

     1. Radial artery conduits. Confirm before placing monitoring catheters whether or not the surgeon intends to utilize this conduit, and, if so, mark the operative side clearly to avoid placing either IV or radial arterial catheters there. Some plans call for diltiazem infusions as prophylaxis against arterial spasm in the radial graft; be aware of the accompanying systemic vasodilation.

     2. Port-access surgery. This approach utilizes a small thoracotomy incision, endoscopic instruments inserted through small ports, and TEE placement of endovascular devices to facilitate CPB. This technique is more frequently employed when valvular surgery is performed rather than coronary revascularization.

     3. MIDCAB. The anterior thoracotomy incision permits access for harvesting of the left IMA and its anastomosis to the LAD coronary artery without CPB.

     4. Urgent CABG

        a. Patient profile. Patients who present for urgent coronary artery bypass surgery usually have ongoing ischemia, including actively infarcting tissue, frequently with hemodynamic instability.

        b. Induction and monitoring. If the patient’s condition requires rapid establishment of CPB, utilize existing femoral catheters for monitoring and IV access rather than delayed surgery.

        c. Anesthetic management. In the hypotensive patient, consider etomidate for induction and potent amnestics like scopolamine as adjuncts. Make every effort to institute CPB without delay.

      5. Patients with poor ventricular function (ejection fraction less than 35%) receive reduced or no premedication. Severe ventricular compromise may require inotrope administration before induction. Popular induction techniques employ etomidate or slow titration of opioids (150 to 250 μg fentanyl every 30 s).

     6. Left main disease or its equivalent. These patients generally fare better with a high-supply/high-demand anesthetic technique rather than a low-supply/low-demand technique (see Section V.F). Maintain diastolic blood pressure during induction and until revascularization is completed with phenylephrine or other vasopressors to ensure adequate oxygen balance.

     7. CABG for patients previously treated with antiplatelet medications. Patients treated with potent antiplatelet agents (platelet glycoprotein receptor antagonists or clopidogrel) are at risk for significant coagulopathy and consequent bleeding. Empirical platelet infusions can avoid catastrophic blood loss and high transfusion requirements.

     8. CABG for patients with heparin-induced thrombocytopenia. See Chapter 17.

     9. Redo CABG procedures. More bleeding, perioperative ischemia, infarction, and pump failure accompany repeat operations. Adherence of the RV to the sternum may result in sudden, extensive hemorrhage during surgical dissection. Table 11.5 summarizes these special concerns, their causes, and appropriate perioperative anesthetic management.

Table 11.5 Perioperative management of the myocardial revascularization reoperation patient

     10. Concomitant acute ischemic mitral regurgitation arises from ischemic papillary muscle dysfunction or ruptured chordae tendineae, with hemodynamic instability. Operation is nearly always urgent (see Section VI.D.4). The accompanying tachycardia, high LVEDP, and increased contractility all worsen myocardial oxygen supply/demand balance. Intra-aortic balloon counterpulsation frequently stabilizes these patients sufficiently to permit survival until the institution of CPB.

VII. Causes and treatment of perioperative myocardial ischemia (Table 11.6)

   A. Causes of ischemia in the prebypass period

     1. Specific high-risk anesthetic-surgical events. Events precipitating ischemia in the prebypass period include endotracheal intubation, surgical stress (skin incision, sternal split), cannulation, and initiation of bypass [10]. Episodes of ischemia may occur during these high-risk events even in the absence of hemodynamic changes [10].

     2. Hemodynamic abnormalities. Some episodes of ischemia during high-risk periods arise from hemodynamic abnormalities, especially tachycardia (greater than 100 beats/min). Minimize hemodynamic alterations to prevent myocardial ischemia. Intraoperative ischemia triples the likelihood of perioperative infarction.

     3. Coronary spasm. Coronary spasm in a normal vessel or around an atherosclerotic lesion causes ischemia. Intense sympathetic stimulation, light levels of anesthesia, and surgical manipulation may theoretically trigger coronary vasospasm.


Table 11.6 Causes

     4. Spontaneous thrombus formation. Atherosclerotic plaque rupture with thrombus formation occludes coronary vessels. Although uncommon, this may occur at any time, including in the operating room prior to revascularization.

   B. Causes of ischemia during bypass

     1. Periods without aortic cross-clamp. Hemodynamic alterations rarely lead to ischemia during CPB. However, mechanical factors and ventricular fibrillation can sharply decrease oxygen supply or increase its demand, respectively. Air or particulate microemboli (thrombus, plastic, and other foreign materials) present in the CPB circuit may embolize to the coronary circulation. Whenever the heart or aorta is opened, air embolism to the native coronary circulation can occur.

     2. During aortic cross-clamp. Ischemia occurs regardless of the myocardial preservation technique. The potential for ischemic injury and subsequent infarction increases with cross-clamp time. Washout of cold cardioplegic solution owing to excessive noncoronary collateral flow may lead to ischemia. Electrical and mechanical quiescence during this period precludes monitoring for ischemia.

     3. After cross-clamp removal

        a. Surgical and technical complications

           (1) Inadvertent incision of the coronary back wall, leading to coronary dissection

           (2) Improper handling of the vein graft with endothelial cell loss, leading to graft thrombus formation

           (3) Twisting of vein grafts

           (4) Anastomosis of the vein graft to the coronary vein

           (5) Suturing the artery closed while grafting or poor-quality anastomoses

           (6) Inadequate vein graft length, leading to stretching of the vein when the heart is filled

           (7) Excess length of vein graft, leading to vein kinking

        b. Etiology of ST-segment elevation after cross-clamp removal. This arises from residual electrophysiologic effects of cardioplegia, coronary air or atheromatous debris embolus, or coronary artery spasm. Appearance in the inferior leads (i.e., right coronary distribution) implicates air embolism, because air seeks the high location of the right coronary ostium. Ventricular aneurysm and pericarditis also cause persistent ST-segment elevations.

   C. Causes of ischemia in the postbypass period

     1. Incomplete revascularization

        a. Ungraftable vessels. The region of myocardium supplied by the unrevascularized stenotic vessel may develop ischemia after CPB because of the added insult of ischemic arrest.

        b. Diffuse distal disease and diabetes. Revascularization rarely restores effective flow in this scenario. Early vein graft closure owing to poor runoff in the small distal vessels further complicates care.

        c. Stress. Ischemia occurrence depends on attention to oxygen supply and demand balance, rather than the degree of analgesia or sedation provided to patients [28]. This includes inappropriate inotrope use, including calcium, during separation from CPB or thereafter.

     2. Coronary spasm. Coronary spasm can occur in the postbypass period, most commonly in the right coronary arteries that are not diseased. Surgical manipulation and exogenous and endogenous catecholamines contribute to this problem.

     3. Mechanical factors. These include vein graft kinking or stretching, or occlusion of IMA flow secondary to overinflation of the lungs.

     4. Thrombus formation. See Section VII.A.4. Postsurgical hypercoagulability may contribute to clot formation.

   D. Treatment of myocardial ischemia (Table 11.7)

Table 11.7 Treatment of ischemia

     1. Treatment of ischemia secondary to hemodynamic abnormalities

        a. Increase or decrease anesthetic depth.

        b. Administer vasodilators or vasopressors when SVR is either high or low, respectively, assuming euvolemia.

        c. Administer a β-blocker, specifically esmolol, to treat tachycardia.

        d. Use atrioventricular sequential pacing. (Specifically in the postbypass period, this can be extremely beneficial to improve rate, rhythm, and hemodynamic stability.)

        e. Use inotropes for ventricular failure, diagnosed by decreased cardiac output and increased ventricular filling pressures. (Pump failure leads to severe decreases in CPP because diastolic blood pressure is decreased and LVEDP is increased.) Note: Indiscriminate use of inotropes may aggravate ischemia. Therefore, preload and heart rate and rhythm should be optimized before using inotropes.

     2. Correction of surgical complications and mechanical problems

        a. Avoid overinflation of the lungs when an IMA graft is present.

        b. Increasing systemic blood pressure with a vasoconstrictor such as phenylephrine can push intracoronary air through the vasculature, and restore blood flow to regions not perfused because of the intravascular air lock.

     3. Treatment of coronary spasm. IV nifedipine, nitroglycerin, diltiazem, and nicardipine can treat coronary spasm. For specific doses of each drug for this indication, refer to Chapter 2.

     4. Specific pharmacologic treatment of ischemia. This treatment includes (a) nitroglycerin, (b) β-blockers, and (c) calcium-channel blockers. See Chapter 2. Prophylactic IV nitroglycerin finds use after CPB in incompletely revascularized patients, patients with severe distal coronary disease, and diabetic patients. Because ischemia frequently occurs from thrombus formation on atheroma, many nonsurgical patients receive heparin as acute therapy and antiplatelet agents (most commonly aspirin) for long-term prophylaxis. These agents should be withheld after bypass until the threat of surgical hemorrhage ceases.

     5. Mechanical support. Refer to Chapter 20 for a complete discussion of circulatory assist devices.

        a. Intra-aortic balloon pumps. Intra-aortic balloon counterpulsation increases CPP and decreases afterload for LV ejection. Patients with impaired ventricular function benefit from improved pump performance in addition to relief of ischemia.

        b. Right and LV assist devices. These devices may be useful for treating severe ischemia caused by myocardial failure or ischemia that led to myocardial failure. Data on their use in treatment of ischemia are still in question.

     6. Ischemic preconditioning

        a. Myocardial injury. Tissue injury results from ischemia, and subsequent tissue reperfusion may involve a reduction in adenosine triphosphate (ATP) levels, free oxygen radicals, calcium-mediated injury, nitric oxide, heat shock proteins, a form of protein kinase C, mitogen-activated protein kinases, and mitochondrial ATP-dependent potassium channels.


        b. A brief period of tissue ischemia may protect against the damage caused by subsequent prolonged ischemia and tissue reperfusion. This ischemic preconditioning occurs in both experimental animals and humans. Endogenously produced adenosine may mediate ischemic preconditioning via enhanced preservation of ATP, inhibition of platelet and neutrophil-mediated inflammatory tissue injury, vasodilation, and decreased basal cellular energy requirements related to intracellular hyperpolarization [20].

        c. Early versus late preconditioning. The early, classic form lasts 2 to 3 h after the ischemic event. It is followed by a secondary period that begins 24 h hence and lasts up to 2 to 3 days. Typically, the period of ischemia and subsequent period of reperfusion each last about 5 min. Four such cycles reduce the infarct size from a subsequent 40-min ischemic period by 75% [41]. The required 40 min of preconditioning time per anastomosis to mitigate sequelae from 15 min of vessel occlusion create logistic challenges in clinical practice.

        d. Role of inhalational anesthetics. Halothane, isoflurane, desflurane, and sevoflurane decrease the deleterious effects of ischemia, in a manner similar to ischemic preconditioning.


 1. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics-2011 update. A report from the American Heart Association. Circulation. 2011;123:e18–e209.

 2. National Institutes of Health. Disease Statistics, Fact Book, Fiscal Year 2000. Bethesda, MD: National Heart, Lung, and Blood Institute.

 3. Cheng DC, Wall C, Djaiani G, et al. Randomized assessment of resource use in fast-track cardiac surgery 1-year after hospital discharge. Anesthesiology. 2003;98:651–657.

 4. Higgins T. Quantifying risk and assessing outcome in cardiac surgery. J Cardiothorac Vasc Anesth. 1998;12:330–340.

 5. Dupuis J, Feng W, Nathan H, et al. The cardiac anesthesia risk evaluation score. Anesthesiology. 2001;94:194–204.

 6. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: Summary article: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2004;110:1168–1176.

 7. Dole W. Autoregulation of the coronary circulation. Prog Cardiovasc Dis. 1987;29:293–323.

 8. Berkowitz DE. Vascular function: From human physiology to molecular biology. In: Schwinn DA, ed. New Advances in Vascular Biology and Molecular Cardiovascular Medicine. Baltimore, MD: Williams & Wilkins; 1998:25–47.

 9. Thys DM, Brooker RF, Cahalan MK, et al. Practice guidelines for perioperative transesophageal echocardiography. A report by the American Society of Anesthesiologists and the Society of Cardiovascular Anesthesiologists Task Force on Transesophageal Echocardiography. Anesthesiology. 2010;112(5):1084–1096.

10. Slogoff S, Keats A. Does perioperative ischemia lead to postoperative myocardial infarction? Anesthesiology. 1985;62:107–114.

11. Tuman KJ, McCarthy RJ, Spiess BD, et al. Effect of pulmonary artery catheterization on outcome in patients undergoing coronary artery surgery. Anesthesiology. 1989;70:199–206.

12. Connors A, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA. 1996;276:889–896.

13. American Society of Anesthesiologists Task Force on Pulmonary Artery Catheterization. Practice guidelines for pulmonary artery catheterization: An updated report by the American Society of Anesthesiologists Task Force on Pulmonary Artery Catheterization. Anesthesiology. 2003;99:988–1014.

14. Shanewise JS. How to reliably detect ischemia in the intensive care unit and operating room. Semin Cardiothorac Vasc Anesth. 2006;10:101–109.

15. Ward R, Parker MD. Myocardial contrast echocardiography in acute coronary syndromes. Curr Opin Cardiol. 2002;17:455–463.

16. Slogoff S, Keats A. Randomized trial of primary anesthetic agents on outcome of coronary artery bypass operations. Anesthesiology. 1989;70:179–188.

17. Tuman KJ, McCarthy RJ, Spiess BD, et al. Does choice of anesthetic agent significantly affect outcome after coronary artery surgery? Anesthesiology. 1989;70:189–198.

18. Mukhtar AM, Obayah EM, Hassona AM. The use of dexmedetomidine in pediatric cardiac surgery. Anesth Analg. 2006;103: 52–56.

19. Buffington CW, Davis KB, Gillispie S, et al. The prevalence of steal-prone anatomy in patients with coronary artery disease: An analysis of the Coronary Artery Surgery Study Registry. Anesthesiology.1988;69:721–727.

20. Lee T. Mechanisms of ischemic preconditioning and clinical implications for multiorgan-reperfusion injury. J Cardiothorac Vasc Anesth. 1999;13:78–91.

21. Garcia C, Julier K, Bestmann L, et al. Preconditioning with sevoflurane decreases PECAM-1 expression and improves one-year cardiovascular outcome in coronary artery bypass graft surgery. Br J Anaesth. 2005;94:159–165.

22. Rorie DK, Tyce GM, Sill JC. Increased norepinephrine release from dog pulmonary artery caused by nitrous oxide. Anesth Analg. 1986;65:560–564.

23. Murphy GS, Szokol JW, Vender JS, et al. The use of neuromuscular blocking drugs in adult cardiac surgery: Results of a national postal survey. Anesth Analg. 2002;95(6):1534–1539.

24. Reich DL, Hollinger I, Harrington DJ, et al. Comparison of cisatracurium and vecuronium by infusion in neonates and small infants after congenital heart surgery. Anesthesiology. 2004;101(5):1122–1127.

25. Jellish WS, Brody M, Sawicki K, et al. Recovery from neuromuscular blockade after either bolus and prolonged infusions of cisatracurium or rocuronium using either isoflurane or propofol-based anesthetics. Anesth Analg.2000;91:1250.

26. Mangano DT, Siliciano D, Hollenberg M, et al. Postoperative myocardial ischemia: Therapeutic trials using intensive analgesia following surgery. Anesthesiology. 1992;76:342–353.

27. Prakash O, Johson B, Meij S, et al. Criteria for early extubation after intracardiac surgery in adults. Anesth Analg. 1977;56:703–708.

28. Hall R, MacLaren C, Smith M, et al. Light versus heavy sedation after cardiac surgery: Myocardial ischemia and the stress response. Anesth Analg. 1997;85:971–978.

29. Scott NB, Turfrey DJ, Ray D, et al. A prospective randomized study of the potential benefits of thoracic epidural anesthesia and analgesia in patients undergoing coronary artery bypass grafting. Anesth Analg. 2001;93:528–535.

30. Dowd N, Cheng D, Karski J, et al. Intraoperative awareness in fast track cardiac anesthesia. Anesthesiology. 1998;89:1068–1073.

31. Leslie K, Sessler D. The implications of hypothermia for early extubation following cardiac surgery. J Cardiothorac Vasc Anesth. 1998;12:30–34.

32. Ginsberg S, Solina A, Papp D, et al. A prospective comparison of three heat preservation methods for patients undergoing hypothermic cardiopulmonary bypass. J Cardiothorac Vasc Anesth.2000;14:501–505.

33. Her DL, Sum-Ping ST, England M. ICU sedation after coronary artery bypass graft surgery: Dexmedetomidine-based versus propofol-based sedation regimens. J Cardiothorac Vasc Anesth.2003;17:576.

34. Anger KE, Szumita PM, Baroletti SA, et al. Evaluation of dexmedetomidine versus propofol-based sedation therapy in mechanically ventilated cardiac surgery patients at a tertiary academic medical center. Crit Path Cardiol. 2010;9(4):221–226.

35. Montes F, Sanchez S, Giraldo J, et al. The lack of benefit of tracheal extubation in the operating room after coronary artery bypass surgery. Anesth Analg. 2000;91:776–780.

36. Chamchad D, Horrow J, Nachamchik L, et al. The impact of immediate extubation in the operating room after cardiac surgery on intensive care and hospital lengths of stay. J Cardiothorac Vasc Anesth.2010;24:780–784.

37. Cheng D, Karski J, Peniston C, et al. Morbidity outcome in early versus conventional tracheal extubation after coronary artery bypass grafting: A prospective randomized trial. J Thorac Cardiovasc Surg. 1996;112:755–764.

38. Hawkes CA, Dhileepan S, Foxcroft D. Early extubation for adult cardiac surgical patients. Cochrane Database Syst Rev. 2003;4:CD003587.

39. Cheng D, Karski J, Peniston C, et al. Early tracheal extubation after coronary artery bypass graft surgery reduces costs and improves resource use. Anesthesiology. 1996;85:1300–1310.

40. Khan NE, DeSouza A, Mister R, et al. A randomized comparison of off-pump and on-pump multivessel coronary-artery bypass surgery. N Engl J Med. 2004;350:21–28.

41. Puskas JD, Williams WH, Mahoney EM, et al. Off-pump vs conventional coronary artery bypass grafting. Early and 1-year graft patency, cost, and quality-of-life outcomes. A randomized trial. JAMA.2004;291:1841–1849.

42. Sellke FW, DiMaio JM, Caplan LR, et al. American Heart Association. Comparing on-pump and off-pump coronary artery bypass grafting: Numerous studies but few conclusions: A scientific statement from the American Heart Association Council on cardiovascular surgery and anesthesia in collaboration with the interdisciplinary working group on quality of care and outcomes research. Circulation.2005;111(21):2858–2864.