Anesthesia for Cardiac Surgical Procedures

Electrocardiography

Standard monitoring of cardiac surgical patients by electrocardiography (ECG) involves using the five-lead electrode system. An electrode is placed on each extremity, and one precordial electrode is placed in the V ₅ position (on the left anterior axillary line at the fifth intercostal space). Ischemia detection is greatest (75%) with the V ₅ lead. This sensitivity increases to 80% when lead II is paired with a V ₅ lead. The addition of a second precordial lead, V ₄ , increases the sensitivity, thus making it possible to detect nearly 100% of ischemic episodes. Currently, most ECG monitors can perform ST-segment analysis automatically with high sensitivity and specificity for detecting ischemia.

Despite appropriate lead selection and the use of automated ST-segment monitoring, perioperative ECG monitoring has important limitations. Visual inspection of the ECG on the monitor has low sensitivity in diagnosing ischemic changes, and automated ST-segment analysis depends on the computer’s ability to set the isoelectric and J-point reference points accurately. During cardiac surgical procedures, the set points should be checked before and after bypass, especially persistent changes in heart rate, because the reference points chosen at the beginning of the case may not be accurate under conditions that may arise later.

Arterial and Central Venous Pressure Monitoring (see also Chapter 36 )

Invasive arterial cannulation and monitoring comprise a standard of care for cardiac surgical patients. Patients’ comorbidities often include labile hypertension, atherosclerotic CVD, or both. Furthermore, cardiac surgical technique frequently causes sudden, rapid changes in arterial blood pressure resulting from factors such as direct compression of the heart, impaired venous return caused by retraction and cannulation of the great vessels, and arrhythmias resulting from mechanical stimulation. In addition, sudden, significant blood loss may induce hypovolemia and hypotension. Finally, during nonpulsatile CPB, noninvasive blood pressure recordings are not accurate. Intraarterial monitoring provides continuous, real-time, beat-to-beat assessment of arterial perfusion pressure and waveform throughout the cardiac surgical procedure. Having an indwelling arterial catheter also enables the frequent drawing of arterial blood for laboratory analyses.

Although the radial artery is the most commonly accessed site, the femoral, brachial, ulnar, dorsalis pedis, posterior tibial, and axillary arteries can also be used. The pressures measured in the peripheral arteries are different from the central aortic pressure because the arterial waveform becomes progressively more distorted as the signal is transmitted down the arterial system. Although the mean arterial pressure (MAP) measured in the peripheral arteries is usually similar to the central aortic pressure, this may change after CPB is initiated. When cannulating a radial artery, one should consider the native state of the collateral circulation in the hand and the possible surgical use of a radial artery free graft as an arterial conduit. If a radial arterial graft is to be harvested, it is usually taken from the nondominant hand or arm; thus, the arterial line should be placed on the dominant side.

Central venous access is also standard of care during cardiac surgery. In addition to monitoring central venous pressure (CVP), central venous access provides a portal for intravascular volume replacement, pharmacologic therapy, and the insertion of other invasive monitors, such as pulmonary artery (PA) catheters. In addition, a CVP catheter can be used to measure the filling pressure of the right ventricle and estimate intravascular volume status. Although the CVP does not directly reflect left-sided heart filling pressure, it may be used as an estimate of left-sided pressures in patients with good left ventricular (LV) function. Following trends is more useful than relying on an individual measurement. In many patients, using a central venous catheter may offer a better risk-to-benefit ratio than using a PA catheter. For the accurate measurement of pressure, the distal end of the catheter must lie within one of the large intrathoracic veins or the right atrium.

An internal jugular vein is most commonly selected for catheterization because it provides ease of approach and optimal distance from the operative field. A femoral or subclavian vein can be considered as well, but groin access can be challenging in obese patients and may be inappropriate if femoral bypass cannula placement or vein graft harvesting is necessary. The subclavian approach is also imperfect because it is associated with an increased risk of catheter obstruction during sternal retraction.

The use of ultrasound is being rapidly adopted throughout the United States for facilitating the placement of central venous catheters and reducing the complications associated with them. Although ultrasound-guided central catheter placement is easily accomplished and appears to improve patients’ outcomes, concerns regarding the associated costs of the required hardware and training are reasons for the lack of universal adoption of this technique ( Box 54.1 ).

Pulmonary Artery Catheterization

The PA catheter is a flow-directed catheter typically placed through an introducer in the central venous compartment via the internal jugular, subclavian, or femoral vein (see also Chapter 36 ). This catheter can measure PA pressure (PAP), CVP, and pulmonary capillary wedge pressure (PCWP). However, the PCWP can both overestimate and underestimate LV filling pressure ( Box 54.2 ). Some PA catheters are designed with a thermistor to register blood temperature changes, which are used to calculate right-sided heart cardiac output (CO) or ejection fraction (EF) by thermodilution. PA catheters may also have oximetric capabilities to measure mixed venous oxygen saturation ( ${\text{S}\overline{\text{v}}\text{O}}_2$ ). Thus, a PA catheter can be used to assess intravascular volume status, measure CO, measure ${\text{S}\overline{\text{v}}\text{O}}_2$ , and derive hemodynamic parameters.

The CO, the amount of blood delivered to the tissues by the heart, is of particular interest to cardiac anesthesiologists. The product of stroke volume and heart rate, CO is affected by preload, afterload, heart rate, and contractility. PA catheters capable of measuring CO continuously were introduced into clinical practice in the 1990s. The correlation of continuous CO measurements with those measurements obtained by using the intermittent thermodilution method is good in physiologically and thermally stable prebypass and postbypass periods.

Continuous monitoring of ${\text{S}\overline{\text{v}}\text{O}}_2$ provides a means to estimate the adequacy of oxygen delivery relative to oxygen consumption. Decreases in ${\text{S}\overline{\text{v}}\text{O}}_2$ may indicate decreased CO, increased oxygen consumption, decreased arterial oxygen saturation, or decreased hemoglobin concentration. If it is assumed that oxygen consumption and arterial oxygen content are constant, changes in ${\text{S}\overline{\text{v}}\text{O}}_2$ should reflect changes in CO. However, London and colleagues found that continuous ${\text{S}\overline{\text{v}}\text{O}}_2$ monitoring did not lead to better outcomes than standard PA catheter monitoring.

Pacing PA catheters are also commercially available. Electrode PA catheters include five electrodes for atrial, ventricular, or atrioventricular (AV) sequential pacing. Paceport PA catheters (Edwards Lifesciences, Irvine, CA) have a port for the insertion of a ventricular wire or of both atrial and ventricular wires for temporary pacing.

The risk-to-benefit ratio involved in using PA catheters has been a subject of controversy since the 1990s. Complications of PA catheter placement include those mentioned in the section on CVP placement, as well as transient arrhythmias, complete heart block, pulmonary infarction, endobronchial hemorrhage, thrombus formation, catheter knotting and entrapment, valvular damage, and thrombocytopenia. In addition, a common complication is incorrect interpretation of the data obtained from the PA catheter, with resultant incorrect treatment of the patient. Schwann and colleagues published a large, international, prospective observational study showing that using a PA catheter was associated with a more frequent risk of a composite mortality and morbidity outcome than using a CVP alone in patients undergoing CABG surgery. Smaller observational trials have also associated the PA catheter with increased morbidity and decreased survival in cardiac surgical patients.

Currently, the trend in the United States is to be selective in deciding which patients may benefit from a PA catheter, especially with the widespread use of transesophageal echocardiography (TEE). Absolute contraindications to PA catheter placement include tricuspid or pulmonic valvular stenosis, right atrial (RA) or right ventricular (RV) masses, and tetralogy of Fallot. Relative contraindications include severe arrhythmias and newly inserted pacemaker wires (which may be dislodged by the catheter during insertion). Clearly, patients undergoing low-risk cardiac surgical procedures can be managed safely without PA catheter placement. However, many cardiac surgeons and anesthesiologists still use the device in high-risk cardiac operations and in patients with right-sided heart failure (HF) or pulmonary hypertension, particularly to assist in postoperative management ( Box 54.3 ).

Transesophageal Echocardiography

TEE is used in many, if not most, cardiac surgical procedures in the current era. See Chapter 37 for a thorough discussion of this extraordinarily valuable diagnostic and monitoring modality.

Monitoring

Cerebral Oximetry

Cerebral oximetry uses near-infrared spectroscopy technology similar to that used in pulse oximeters. Light-emitting electrodes are placed on the patient’s forehead, lateral to the midline, and over both frontal cortices. Because the skull is translucent to infrared light, and because oxygenated and deoxygenated blood have different absorption characteristics when exposed to infrared light of two different wavelengths, regional cerebral oxygen saturation (rSO ₂ ) can be computed from the returning signal. The design of the cerebral oximeter enables it to measure change from the patient’s own established baseline and to monitor both frontal lobes simultaneously.

The use of near-infrared spectroscopy has been studied in the perioperative setting; a relative decrease in rSO ₂ to less than 80% of the preoperative baseline or to absolute levels less than 50% increased the incidence of adverse postoperative outcomes. These outcomes include POCD, stroke, organ dysfunction, mortality, and length of hospital stay. A physiologically derived treatment algorithm for management of perioperative cerebral oxygen desaturation has been proposed ( Fig. 54.3 ). As noted by Murkin, an important confounder in evaluating the role of monitoring cerebral oximetry is the efficacy of treatment for cerebral desaturation.

A baseline rSO ₂ that is low even though the patient is breathing supplemental oxygen (absolute value ≤50%) is an independent risk factor for 30-day and 1-year mortality. A baseline rSO ₂ could act as a refined marker for preoperative risk stratification and help clinicians identify patients who require more intensive monitoring and care in the postoperative period.

Glucose Control

Hyperglycemia in surgical patients is a consequence of the inflammatory or stress response to the trauma of surgery. Components of this response include an endocrine response (i.e., increased production of counterregulatory hormones such as cortisol, growth hormone, glucagon, and catecholamines) ( Fig. 54.4 ), an immune response resulting in increased cytokine production, an autonomic response resulting in increased sympathetic stimulation, and altered insulin signaling. These changes increase glucose production, decrease glucose elimination during CPB, and induce insulin resistance, thereby causing hyperglycemia.

All patients are at risk for developing hyperglycemia during cardiac surgery. Older patients, diabetic patients, and patients with CAD are particularly prone to perioperative hyperglycemia. Although cardiac surgery without CPB initiates a stress response, CPB increases this response manyfold. The degree of hyperglycemia depends on several variables associated with the use of CPB, such as the pump prime fluid selected and the degree of hypothermia induced. Epinephrine and other inotropic drugs may contribute to hyperglycemia after CPB by stimulating hepatic glycogenolysis and gluconeogenesis.

Impaired fasting glucose blood levels before cardiac surgery and persistently increased glucose levels during and immediately after surgical procedures are predictive of longer hospital stay and increased perioperative morbidity and mortality in both diabetic and nondiabetic patients. However, in diabetic patients undergoing cardiac operations, hyperglycemia may only partly explain the increased risk for adverse outcomes. Immunologic abnormalities common to diabetic patients, such as decreased chemotaxis, phagocytosis, opsonization, bacterial killing, and antioxidant defense, also promote adverse outcomes by increasing diabetic patients’ risk of infection.

Glucose blood levels should be controlled, beginning in the preoperative period and continuing until discharge. However, in a classic study, patients were randomly assigned to intensive intraoperative insulin treatment (to maintain glucose levels at 80-100 mg/dL) or to standard insulin treatment (to keep glucose levels <200 mg/dL). Surprisingly, the investigators reported a statistically nonsignificant increase in the incidence of death and stroke in the group receiving intensive insulin treatment. Current Society of Thoracic Surgeons (STS) guidelines state that blood glucose levels should be kept lower than 180 mg/dL throughout the perioperative period. Some cardiac surgical centers more aggressively administer continuous insulin infusions in an attempt to keep glucose levels lower than 150 mg/dL.

Thyroid Hormone

Fig. 54.5 shows the effect of thyroid hormone on cardiovascular hemodynamics. Abnormal thyroid function ultimately affects cardiac function in many ways ( Table 54.2 ). During cardiac surgery, the effect of CPB on thyroid hormone production is uncertain. Both increased and decreased levels of thyroid hormone can occur during or immediately after CPB. Free triiodothyronine (T ₃ ), the biologically active form of the thyroid hormone, is frequently reduced in cardiac patients because of the reduced activity of the 5′-monodeiodinase responsible for converting thyroxine (T ₄ ) into T ₃ in peripheral tissues. Patients with low T3 values are predisposed to low CO and therefore death as a complication of cardiac surgery. Patients should be profiled for T3 levels and labelled high risk if the levels are low preoperatively.

Notably, hypothyroidism is more common in female than in male patients undergoing CABG. Zindrou and associates noted a 17% higher mortality rate in women who underwent CABG while receiving thyroxine replacement therapy for hypothyroidism. A review of the literature by Edwards and colleagues concluded that ensuring a perioperative euthyroid state in women with hypothyroidism who were undergoing CABG could be helpful in reducing the perioperative mortality of these patients.

Heparin as an Anticoagulant

Since its discovery by Jay McLean, MD, in 1915, heparin has stood the test of time and remains the primary anticoagulant used in cardiac operations that require CPB. The mechanism underlying heparin’s anticoagulant effect centers on the heparin molecule’s ability to bind simultaneously to antithrombin III and thrombin. Modern nomenclature refers to antithrombin III as antithrombin (AT). The binding process is mediated by a unique pentasaccharide sequence that binds to AT. The proximity of AT and thrombin, mediated by the heparin molecule, allows AT to inhibit the procoagulant effect of thrombin by binding to the active-site serine residue of the thrombin molecule. The inhibitory effect of AT is increased 1000-fold in the presence of heparin. The heparin-AT complex can affect several coagulation factors, but factor Xa and thrombin are the most sensitive to inhibition by heparin, and thrombin is 10 times more sensitive to the inhibitory effects of unfractionated heparin than is factor Xa.

Only approximately onethird of the heparin molecules in a dose of heparin contain the critical pentasaccharide segment that is needed for high-affinity binding to AT. Thus, relatively large doses are required to produce the anticoagulant effect necessary for CPB. In fact, dosing of heparin for CPB is somewhat empiric. After a baseline activated clotting time (ACT) is measured (the normal range is 80-120 seconds), a dose of 300 to 400 units/kg of heparin is given as an intravenous bolus. Commercially available assays are used for calculating the patient’s dose-responsiveness to heparin in vitro. Some practitioners administer heparin at the dose that is indicated by such an in vitro dose-response assay. Subsequent heparin dosing for extracorporeal circulation (ECC) is targeted at maintaining ACT values longer than 400 to 480 seconds. Also available is a heparin concentration monitor that uses protamine titration analysis for ex vivo calculation of the whole blood heparin concentration. This result is often used as an adjunct to the ACT value in confirming an adequate heparin concentration for CPB. Unfortunately, ACT test results vary substantially with clinical conditions and the particular platform used for measurement. Thus, the evidence supporting the use of a threshold of 400 or 480 seconds is almost entirely anecdotal.

The dose of heparin used in patients on CPB is based on early landmark work published by Bull and coauthors in 1975. A small study that sought evidence of thrombin activity during CPB in nonhuman primates and pediatric patients found results supporting a safe lower limit for ACT of 400 seconds. In 1979, Doty and colleagues proposed a simplified dosing regimen guided by ACT values without dose-response curves. The data and recommendations from these few studies constitute the primary basis for current heparin dosing protocols.

Despite heparin’s historic and continued role in anticoagulation for patients maintained on CPB, it is not a perfect anticoagulant. Intrinsic and extrinsic pathway coagulation occurs despite heparin administration, and platelets can still be activated by contact with bypass circuitry and by heparin directly. Alternative anticoagulants are discussed briefly in the section on heparin-induced thrombocytopenia (HIT).

Monitoring of Anticoagulation

Using ACT to monitor the effectiveness of heparin is not an exact science. Tremendous variability is observed in patients’ anticoagulation responses to a given dose of heparin; reasons for this variability include variations in levels of heparin-binding proteins and AT. Hence, ACT values correlate poorly with actual heparin concentrations. Nevertheless, since the publication of Bull and colleagues’ early work, ACT has been the mainstay of anticoagulation monitoring in cardiac operations that require CPB.

Many different ACT measurement devices are commercially available, and each uses a different platform for clot detection and endpoint signaling. However, they all involve the addition of whole blood to a tube or channel containing a contact-phase activator. The activator can be celite, kaolin, glass, or any combination thereof. The sample is warmed to 37°C before the measurement technique is performed. Clot formation occurs, and the measurement ends when a change in velocity, pressure, oscillation, electromagnetic forces, or even color is observed, depending on the particular platform of measurement.

Several clinical variables can affect the ACT ( Table 54.3 ). In addition to physiologic variations, the design of ACT measurement devices (which varies across their many manufacturers) also affects ACT normal and therapeutic values. ACTs also correlate poorly with whole blood and plasma heparin levels, and responses to heparin differ somewhat between adult and pediatric patients. Some authors argue that, because of its poor correlation with heparin concentration, ACT alone is not an adequate monitor of heparin efficacy, and that simultaneous or adjunct monitoring of heparin concentration should also be used during CPB. Prolongation of ACT by non–heparin-related clinical factors, such as hypothermia, hemodilution, or quantitative or qualitative platelet abnormalities, is a well-documented phenomenon, and the anesthesiologist must understand these factors to determine whether it is safe to reduce the heparin dose when the ACT is prolonged. The poor correlation between ACT and measured heparin concentration also makes it possible that a dose reduction will render the heparin concentration inadequate, even when the ACT remains within an acceptable range.

Some point-of-care (POC) monitors, such as the Hepcon HMS system (Medtronic Perfusion Systems, Minneapolis, MN), use a protamine titration assay to calculate the heparin concentration. Despotis et al. suggest that monitoring and maintaining heparin concentration—and thus giving larger doses of heparin—actually protect the hemostatic system and may decrease transfusion requirements. However, other investigators have not been able to confirm that higher doses of heparin confer better hemostatic protection; markers of ongoing coagulation are essentially the same whether traditional ACT monitoring or heparin concentration monitoring is used. The 2018 STS, Society of Cardiovascular Anesthesiologists (SCA), and the American Society of ExtraCorporeal Technology (AmSECT) Guidelines state: “Use of heparin concentration monitoring in addition to ACT might be considered, for the maintenance of CPB, as this strategy has been associated with a significant reduction in thrombin generation, fibrinolysis, and neutrophil activation. However, its effects on postoperative bleeding and blood transfusion are inconsistent (class IIb, Level of Evidence B).” An additional recommendation includes: “During CPB, routine administration of heparin at fixed intervals, with ACT monitoring, might be considered and offers a safe alternative to heparin concentration monitoring. (class IIb, Level of Evidence C).”

The high-dose thrombin time (HiTT) is a modification of the thrombin time that is designed to measure the high levels of heparin that are used during CPB. Unlike ACT, the HiTT correlates well with heparin concentration, both before and during CPB, and it is not affected by hemodilution and hypothermia. As a measure of thrombin inhibition, the HiTT is a more specific test of heparin’s effect on thrombin than ACT, and it appears to possess less artifactual variability. Preoperative heparin infusions do not affect HiTT values.

Protamine and Reversal of Anticoagulation

Protamine, which has been in clinical use for as long as heparin has, remains the heparin reversal drug of choice in cardiac surgery. The protamine dose required to reverse heparin is somewhat controversial. In the first published study to examine this question, Bull and associates chose a dose of 1.3 mg protamine for every 100 units of heparin to provide a slight excess of protamine from a projected needed amount of 1.2 mg for every 100 units of heparin. Protamine is usually administered according to the total amount of heparin given (i.e., 1-1.3 mg protamine per 100 units of heparin). This method may result in luxuriant protamine doses, which reduce any theoretic or real risks of heparin rebound but may put the patient at higher risk for bleeding events because of the anticoagulant effect of protamine. Guidelines for the practice of anticoagulation in CPB recommend the following:

• 1.

  “It is reasonable to limit the ratio of protamine/heparin to less than 2.6 mg protamine/100 Units of heparin, since total doses above this ratio inhibit platelet function, prolong ACT, and increase the risk of bleeding (class IIa, Level of Evidence C).”

• 2.

  Because of the risk of heparin rebound in patients requiring high doses of heparin and with prolonged CPB times, low-dose protamine infusion (25 mg/h) for up to 6 hours after the end of CPB may be considered as part of a multimodality blood conservation program (class IIb, Level of Evidence C).

Protamine can also be administered at a dose calculated from the heparin concentration, which is measured by a protamine titration assay. Guidelines do support this practice stating that “it can be beneficial to calculate the protamine reversal dose based upon a titration to existing heparin in the blood, since this technique has been associated with reduced bleeding and blood transfusion” (class IIa, Level of Evidence B). If the heparin concentration is not measured, the protamine dose can be graphically derived by plotting ACT values throughout the case and creating heparin dose-response curves. The amount of protamine used in this method is based on the circulating concentration of heparin in the patient at the time of reversal. Because, theoretically, no excess protamine exists, these patients may be at risk for heparin rebound and therefore may require additional protamine. In a small study conducted in patients undergoing valve surgery, administering protamine in two divided doses by titration resulted in a larger dose but reduced bleeding, presumably by treating heparin rebound.

Unique Hematologic Considerations in Cardiac Surgery

Hematologic Effects of Cardiopulmonary Bypass

The hematologic effects of CPB are complex. Exposure of blood to the surfaces of the extracorporeal circuit is a profound stimulus for inflammatory system upregulation, and activation of the hemostatic system is a component of the normal inflammatory response. According to traditional models of hemostasis, ECC activates both the intrinsic and extrinsic coagulation pathways and directly impairs platelet function. Intrinsic pathway activation can occur by contact activation and the conversion of factor XII to factor XIIa on the various surfaces of the CPB circuit. The tissue factor generated from the wound and the circulating tissue thromboplastin combine to cause the extrinsic activation of coagulation by cell-mediated hemostasis, which involves tissue factor–bearing leukocytes and activated endothelial cells. Tissue factor pathway generation of thrombin has a primary role in CPB-associated hemostasis abnormalities ( Fig. 54.6 ).

In addition to activating both extrinsic and intrinsic coagulation pathways, CPB directly impairs platelet function through a variety of mechanisms. Platelets express on their surface numerous glycoproteins that serve as receptors for several circulating ligands, such as fibrinogen, thrombin, and collagen ( Fig. 54.7 ). The components of the bypass circuit adsorb circulating proteins that can serve as foci for platelet attraction and adherence. These surface-bound platelets activate and release the contents of their cytoplasmic granules, which can then serve as localized sources of thrombin generation, or they may embolize to initiate microvascular thrombosis.

Fibrinolytic activity is also increased by CPB. Contact activation leads to the activation of factor XII, prekallikrein, and high-molecular-weight kininogen, which causes endothelial cells to produce tissue plasmin activator, and lysis of fibrin and fibrinogen ensues ( Fig. 54.8 ).

The vascular endothelium is itself an active substrate that is sensitive to circulating mediators, and it expresses and releases anticoagulant and procoagulant factors. When exposed to hypoxia or inflammatory mediators during CPB, the endothelium responds and can induce a relatively prothrombotic state marked by tissue factor upregulation, accelerated platelet adhesion, and increased expression of leukocyte adhesion molecules ( Fig. 54.9 ).

Heparin-Induced Thrombocytopenia

HIT is an immune-mediated prothrombotic disorder that occurs in patients exposed to heparin. Antibodies form against the protein platelet factor 4 (PF4) when PF4 has formed a complex with heparin. Although PF4 is found in only trace amounts in human plasma and is stored in platelet α granules, the presence of heparin increases plasma PF4 concentrations 15- to 30-fold by displacing bound PF4 on endothelial cell surfaces. PF4 is also expressed on the surface membrane of activated platelets by membrane fusion with the α-granule membrane; thus, PF4 is exposed and available to bind with heparin. The resulting PF4-heparin complex on the platelet surface is recognized by a specific immunoglobulin G (IgG), which binds to the complex and leads to immunologically mediated platelet activation. Hyperaggregability of these activated platelets is the hallmark of HIT and is responsible for its prothrombotic complications. Often a clinical score such as the 4 Ts score can be used to determine whether a heparin-platelet antibody test should be performed to diagnose HIT (class IIa, Level of Evidence B). Diagnosis can be difficult in the postcardiac surgery patient and often the 4 Ts score is unreliable.

A common presentation in patients with HIT is a reduction in the platelet count to less than 100,000/μL or to less than 50% of the baseline count. The incidence of seroconversion after CPB and heparin exposure is quite frequent (20%-50%). However, the reported prevalence of clinical HIT after CPB is only 1% to 3%. Thus, the risk of HIT in cardiac surgical patients with postoperative seroconversion is less than 10%.

The strength of the immunologic response, not the mere presence of PF4 or heparin antibodies, may determine which patients are prone to HIT and are at risk for thromboembolic complications. The presence of preoperative antibody, in addition to postoperative antibody, has been associated with increased morbidity after cardiac surgery. This morbidity takes the form of gut ischemia, renal dysfunction, limb ischemia, and other prothrombotic events.

Management of the cardiac surgical patient with HIT must include a careful risk-to-benefit analysis. The likelihood that a patient has true disease and is at increased risk for a thrombotic event must be weighed against the risks posed by using an alternative anticoagulant to heparin. The urgency of the surgical procedure is also an important factor that can affect decision making. It is preferable, when possible, to defer the operation until antibody titers have become undetectable or only weakly positive, which may occur after 90 days (class IIa, Level of Evidence C). If surgical postponement is not practical, then other therapeutic options must be considered ( Boxes 54.6 and 54.7 ). Currently, the direct thrombin inhibitors are used as the anticoagulants of choice. Hirudin and argatroban are approved by the U.S. Food and Drug Administration (FDA) for use in patients with HIT-related thrombosis. The use of these drugs as anticoagulants for CPB is fraught with hemorrhagic complications. Bivalirudin has been approved by the FDA for use in percutaneous interventions and, because of its short half-life, has been favored as an anticoagulant for CPB in patients with HIT. Guidelines suggest that in patients with a diagnosis of HIT and in need of an urgent operation requiring CPB, anticoagulation with bivalirudin is a reasonable option (class IIa, Level of Evidence B). However, no drug other than heparin has FDA approval for specific use as an anticoagulant in patients undergoing CPB. Bivalirudin undergoes renal elimination. Therefore, in seropositive HIT patients who have significant renal dysfunction, anticoagulation for urgent operations requiring CPB can be accomplished with argatroban, plasmapheresis prior to heparin to remove antibodies, or heparin with concomitant antiplatelet agents to prevent platelet activation (tirofiban, iloprost) (class IIb, Level of Evidence C). These latter two techniques have risk because they include heparin, and have been fraught with increased risks of bleeding. Boxes 54.6 and 54.7 summarize therapeutic options and alternative anticoagulant strategies for the patient with HIT whose operation cannot be deferred until seronegativity is documented. Fig. 54.10 delineates how each alternative drug inhibits factor Xa, thrombin, or fibrinogen.

Box 54.6

Therapeutic Options for Anticoagulation for Cardiopulmonary Bypass in Patients With Heparin-Induced Thrombocytopenia

Modified from Kaplan JA, Reich DL, Savino JS, eds. Kaplan’s Cardiac Anesthesia: The Echo Era. 6th ed. St. Louis: Saunders; 2011:966.

• 1.

  Ancrod

• 2.

  Low-molecular-weight heparin or heparinoid (test first)

• 3.

  Alternative thrombin inhibitor (hirudin, bivalirudin, argatroban)

• 4.

  Using a single dose of heparin, promptly neutralizing it with protamine, and

  • a.

    Delaying surgery so antibodies can regress or

  • b.

    Using plasmapheresis to decrease antibody levels or

  • c.

    Inhibiting platelets with iloprost, aspirin and dipyridamole (Persantine), abciximab, or RGD blockers

In all cases:

• 1.

  No heparin in flush solutions

• 2.

  No heparin-bonded catheters

• 3.

  No heparin lock intravenous ports

No agent is currently indicated for anticoagulation in cardiopulmonary bypass.

RGD, Receptor glycoprotein–derived.

Box 54.7

Potential Alternative Anticoagulants for Cardiopulmonary Bypass

From Kaplan JA, Reich DL, Savino JS, eds. Kaplan’s Cardiac Anesthesia: The Echo Era. 6th ed. St. Louis: Saunders; 2011:967.

• Ancrod

• Low-molecular-weight heparins

• Factor Xa inhibitors

• Bivalirudin or other direct thrombin inhibitors (hirudin, argatroban)

• Platelet receptor inhibitors

Protamine Reactions

Protamine is associated with several hemodynamic effects that can be categorized by their presentation and mechanism. Adverse reactions to protamine range from moderate hypotension to more profound and hemodynamically significant reactions that can increase in-hospital mortality risk. The clinical presentation of these reactions serves as a starting point for understanding their mechanistic relationships. Commonly, these reactions are classified as type I, type II, or type III. A type I protamine reaction involves isolated hypotension, with normal to low filling pressures and normal airway pressures. This reaction is usually mild and responds to volume infusion, slowing of protamine infusion, and the gentle titration of vasoactive medications. Type II reactions include moderate to severe hypotension and features of anaphylactoid reactions, such as bronchoconstriction. Anaphylactoid reactions include protamine sensitivity reactions that are classically immunologic or allergic in that they are immunoglobulin E (IgE) antibody mediated. Nonimmunologic mechanisms may involve IgG antibodies or complement activation. Type III reactions are thought to be caused by large heparin-protamine complexes that lodge in the pulmonary circulation, cause the release of mediators, and result in severe hypotension and elevated PA pressures that may lead to acute RV failure. This is obviously a profound response, resulting in global cardiovascular collapse or necessitating the reinstitution of CPB because of intractable RV failure. Fortunately, catastrophic hypotension and intractable RV failure are relatively rare events on the spectrum of protamine reactions.

Mechanistic explanations for protamine reactions include endothelial nitric oxide release, mast cell degranulation, and histamine release associated with rapid infusion. A study by Kimmel and colleagues found that neutral protamine Hagedorn insulin use, documented fish allergy, and a history of nonprotamine medical allergies were independent risk factors for protamine hypersensitivity reactions. In this study, 39% of patients who presented for cardiac surgery had one or more of these risk factors. Other possible but unconfirmed risk factors include prior exposure to protamine, a history of vasectomy, decreased LV function, and hemodynamic instability. The site of injection does not influence the incidence of protamine reactions. Pretreatment with histamine blockade is not preventive.

The following principles summarize treatment options for patients at risk for protamine reactions:

• 1.

  Protamine should be administered slowly (i.e., over ≥5 minutes). Limit protamine dose to less than 2.6 mg protamine/100 units of heparin, since total doses above this ratio inhibit platelet function, prolong ACT, and increase the risk of bleeding (class IIa, Level of Evidence C).

• 2.

  In patients with documented adverse events related to protamine, consideration should be given to not rechallenging the patient with protamine. Pharmacologic alternatives to protamine can be considered, or a decision can be made not to reverse heparin. Consideration may be given to using non–heparin-based CPB, performing off-pump coronary artery bypass (OPCAB) with an alternative to heparin, or, if heparin is used, administering nonprotamine heparin reversal drugs such heparinase, or simply waiting for heparin’s effects to dissipate.

• 3.

  Hypotension associated with protamine reversal of heparin often is ameliorated by simply slowing or pausing protamine infusion while volume is infused through intravenous lines or the aortic cannula. Vasoactive medications, such as phenylephrine or ephedrine, use of calcium chloride, or increased inotropic support may be necessary.

• 4.

  Severe or intractable hypotension, with or without evidence of pulmonary circulatory involvement, bronchospasm, or overt RV failure, demands immediate aggressive attention, intervention, and planning for a potential return to CPB. Steps to consider include the following:

  • a.

    Reheparinization to prepare the patient for a return to CPB and to reduce heparin-protamine complex size. If hemodynamics permit, a low dose of heparin (70 units/kg) may be tried first while supportive treatment continues, followed by a full CPB dose of heparin (300 units/kg) if it becomes necessary to return the patient to CPB. (class I, Level of Evidence C).

  • b.

    Inotropic support, either by infusion or by intermittent bolus administration, is warranted. Epinephrine and norepinephrine are acceptable options, and milrinone may be considered if the patient’s hemodynamic status permits.

  • c.

    If the patient’s hemodynamic status allows, nebulized albuterol is helpful in the management of bronchospasm and elevated airway pressures.

Antifibrinolytic Therapy: Prophylaxis for Bleeding

Prophylactic use of antifibrinolytic drugs before CPB reduces bleeding and transfusion requirements in cardiac surgical patients in randomized trials and in multiple metaanalyses. The most well-known antifibrinolytic drugs include the synthetic lysine analogues ε-aminocaproic acid (EACA) and tranexamic acid (TA) and the serine protease inhibitor aprotinin. Presumably, the blood-sparing effects of the synthetic drugs result from the inhibition of fibrinolysis by their binding to the lysine-binding sites on plasmin. This also has platelet protective properties because plasmin’s antiplatelet effects are also inhibited by antagonist binding.

Aprotinin is a direct enzymatic inhibitor of plasmin and has other protease-inhibiting properties that confer its antiinflammatory and antikallikrein effects. However, aprotinin was notably associated with increases in post-CPB creatinine values and other adverse organ system outcomes in large-scale observational trials. When a randomized prospective trial showed an increase in mortality in the aprotinin group, despite a reduction in bleeding, the drug was removed from the global market. Although the causes of death in the aprotinin-treated patients were not found to be related to thrombosis or other drug-related effects, aprotinin was rendered unavailable for commercial use for years after publication of this study. The decision to remove aprotinin from clinical use was revisited on reevaluation of these study data, and aprotinin has been reintroduced in Canada and other countries specifically for use as labeled in CABG surgery.

Hypercoagulable States

The use of antifibrinolytics has become common in cardiac operations that require CPB. The use of antifibrinolytic drugs and prohemostatic drugs and blood products to treat bleeding has brought to light the risk of thrombosis in CPB, during which feedback mechanisms are critical and homeostasis is perturbed. All patients incur this risk of thrombosis as consumptive coagulopathy increases, but the risk is greatly increased in patients with congenital or acquired thrombophilic states. These states may be important in view of current practices that involve pharmacologically inhibiting fibrinolytic pathways in cardiac surgical patients.

The factor V Leiden (FVL) mutation is the most common inherited thrombophilic disorder; its prevalence is 3% to 7% in white populations. Mechanistically and clinically, FVL has been implicated in thrombotic complications in cardiac operations, most often those involving a period of circulatory arrest and the use of antifibrinolytics. In a review of FVL mutation, Donahue gave the following summation and recommendations :

• ▪

  Cardiac surgical patients who are heterozygous for the FVL mutation bleed less than do noncarriers.

• ▪

  The risk of early graft thrombosis may be increased in patients with FVL deficiency.

• ▪

  The use of antifibrinolytic drugs may increase thrombotic risk in patients with FVL.

• ▪

  Anecdotal evidence suggests that in patients with FVL mutation who are exposed to deep hypothermic circulatory arrest, antifibrinolytics increase thrombotic risk.

Cardiac Surgical Patients Taking Anticoagulant or Antithrombotic Drugs

Antithrombotic therapy in cardiac surgical patients has many roles and applications. Patients with ischemic heart disease can be managed short- or long-term with pharmacologic drugs that may include aspirin, AT inhibitors (heparins), direct thrombin inhibitors, or an array of platelet inhibitors (adenosine diphosphate [ADP] receptor inhibitors and glycoprotein IIb/IIIa [Gp IIb/IIIa] receptor inhibitors). Patients with a history of peripheral vascular disease, valvular heart disease, or low ventricular ejection states can similarly be managed with some form of antithrombotic therapy that may also include warfarin. Frequently, patients arrive for surgery while receiving multiple antithrombotic medications. Thus, postoperative bleeding is a common but challenging complication of cardiac surgery, especially when the bleeding risk posed by CPB itself is considered.

The use of percutaneous coronary interventions such as angioplasty and intracoronary stent deployment for ischemic heart disease has led to the use of antithrombotic medications to maintain stent patency and to prevent stent thrombosis. The American College of Cardiology and AHA (ACC/AHA) initial guidelines for percutaneous coronary intervention recommended (on the basis of class I evidence) the use of both aspirin and clopidogrel for at least 1 year after drug-eluting stent placement. However, percutaneous coronary intervention data with the second generation drug-eluting stents indicate that shorter periods of dual anti-platelet therapy are equally effective in preventing in-stent thrombosis, and thus allow for earlier cessation of a single anti-platelet agent after 6 months. This opens the door for interventions and surgical procedures to be performed sooner and with less risk of bleeding. The administration of thienopyridine antiplatelet drugs in addition to aspirin increases the risk of bleeding after cardiac surgery ; however, it is unclear whether aspirin alone increases this risk. Abundant evidence (mostly level B evidence from small, retrospective, nonrandomized studies) suggests that the ADP receptor antagonist clopidogrel (Plavix) has been associated with excessive perioperative bleeding in patients who undergo CABG. This trend has even been reported in the OPCAB population, although not consistently. Earlier recommendations included a 5- to 7-day delay after discontinuation of clopidogrel in patients who require CABG. However, guidelines suggest that a 3-day delay may be sufficient to lessen bleeding risk and provide safe outcomes. It is likely that a 5- to 7-day delay is not necessary, but some interval between the discontinuation of clopidogrel and CABG is supported by the available evidence.

Patients with a history of Gp IIb/IIIa receptor blockade as a part of their acute coronary syndrome management are at risk of increased bleeding and blood component use when they are given abciximab, especially within 12 hours of cardiac surgery. Shorter-acting Gp IIb/IIIa receptor antagonists do not seem to increase bleeding or adverse outcomes and in fact may improve myocardial outcomes when Gp IIb/IIIa blockers are in use. The interval between the discontinuation of antiplatelet therapy and cardiac or noncardiac surgery is critical to preventing thrombotic events without increasing the risk of bleeding. These decisions regarding the cessation of therapy can be guided by knowledge of drug pharmacology and testing of antiplatelet drug efficacy.

Enoxaparin, a low-molecular-weight heparin, increases transfusion rates and the risk of surgical reexploration when it is used in association with CPB. Low-molecular-weight heparin may also decrease heparin responsiveness.

Patients who present for cardiac surgery with residual warfarin effects may benefit by having enhanced anticoagulation during CPB. Postoperatively, if excessive bleeding is noted and is confirmed by hemostasis testing, factor replacement can be performed with blood products or pharmacologic prothrombin complex concentrates (PCCs).

Patients with atrial fibrillation may be maintained on new antithrombotic therapeutic drugs, direct oral anticoagulants (DOACs) that include thrombin inhibitors (dabigatran) and factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban). These drugs are potent and long-acting, and they have no antidote, so one would expect that treating cardiac surgical patients with these drugs would increase their bleeding risk. When compared to vitamin K antagonists, the DOACs have similar thromboprophylaxis efficacy yet fewer bleeding complications. An additional benefit is that DOACs have a predictable pharmacodynamic profile and routine monitoring is often unnecessary. Routine monitoring tests such as the INR and aPTT do not even accurately assess anticoagulant activity of the DOACs and thus a thrombin time or a direct measure of anti-Xa activity would be considered more accurate.

In summary, patients who present for cardiac surgery with preexisting, pharmacologically induced inhibition of the hemostatic system may have undesirable post-CPB bleeding. The diagnosis and treatment of this complication should be the same whether the derangement is a function of CPB itself, coexisting pharmacologic inhibition, or both. The management and treatment of persistent postoperative bleeding are discussed in the section on problems in the postoperative period.

Atrial Arrhythmias

Atrial fibrillation (AF) is the most common postoperative arrhythmia (27%-40%). Patients are at highest risk for new-onset atrial fibrillation 2 to 3 days after cardiac surgery. This arrhythmia can potentially prolong the patient’s hospital stay and increase management costs by causing hemodynamic compromise or thromboembolic complications in the postoperative period.

Many potential risk factors have been studied in efforts to predict the occurrence of postoperative atrial fibrillation. With advancing age, dilatation of the atrium interrupts cell-to-cell electrical coupling between atrial muscle fibers. Other preoperative risk factors for postoperative atrial fibrillation include a history of atrial fibrillation, chronic obstructive pulmonary disease, valve surgery, and postoperative withdrawal of a β-blocker or ACEI.

Increased preoperative hemoglobin A1c, physical activity of low intensity during 1 year prior to surgery, Caucasian race, obesity, and electrolyte disturbances (hypokalemia, hypomagnesemia) have also been shown to contribute towards increased risk of atrial fibrillation.

Perioperative factors include inadequate atrial protection during surgery, pericardial inflammation, autonomic imbalance during the postoperative period, change in atrial size with fluid shifts, electrolyte (potassium and magnesium) abnormalities, and excessive production of catecholamines. Reduced risk has been associated with the postoperative administration of β-blockers, ACEIs, potassium supplementation, and nonsteroidal antiinflammatory drugs (NSAIDs).

Treatment for atrial fibrillation includes both pharmacologic agents and electrical stimulation. Many studies have shown that β blockade significantly reduces the incidence of postoperative atrial fibrillation and that withdrawal of β-blockers increases risk. Synchronized cardioversion is reserved for patients who show signs of hemodynamic instability during atrial fibrillation. In the absence of hemodynamic instability, pharmacologic agents are used with the goal of preventing a rapid ventricular response. Drugs usually employed for this purpose include calcium channel blockers, β-blockers, magnesium, and amiodarone. Expert consultation should be obtained before treatment is initiated, especially in a stable patient.

Ventricular Arrhythmias

Although ventricular arrhythmias occur after cardiac surgical procedures, sustained ventricular arrhythmias are relatively uncommon. Associated factors may include hemodynamic instability, electrolyte abnormalities, hypoxia, hypovolemia, ischemia or infarction, acute graft closure reperfusion, and the use of inotropic agents.

Ventricular arrhythmias can range from simple premature ventricular complexes (PVCs) to VT. Simple PVCs do not pose a significant risk of life-threatening ventricular arrhythmia. Conversely, complex ventricular arrhythmias, including both frequent PVCs (>30/h) and nonsustained VT, may make patients prone to sudden death, especially in the long term. Sudden death is even more likely if ventricular function is also compromised. A study of 126 patients with postoperative complex ventricular ectopy found a mortality rate of 75%. Patients with sustained ventricular arrhythmias have a poor prognosis in both the short and the long term.

Although hemodynamically unstable VT should be treated by synchronized cardioversion, patients with PVCs or short runs of nonsustained VT with hemodynamic stability do not need to be treated. Any reversible causes should be sought and corrected. Amiodarone is reserved for hemodynamically stable patients with VT or an uncertain rhythm. Ventricular fibrillation should be treated promptly with electrical defibrillation. For long-term management of ventricular arrhythmias, in addition to antiarrhythmic agents, electrophysiologic studies or placement of an ICD should be considered.

Bradyarrhythmias

Bradyarrhythmias are not uncommon in the immediate postoperative period. In most cases, a temporary epicardial pacemaker is sufficient. In a small percentage of patients, a permanent pacemaker may be necessary, especially in patients with sinus node dysfunction or AV conduction disturbances after either CABG or valve repair. Patients who need a permanent pacemaker may receive either a single-chamber or a dual-chamber pacemaker. Multiple factors dictate which particular device would most benefit an individual patient.

Neuroprotection Strategies

Numerous strategies have been tried to decrease the incidence and severity of neurologic injury in cardiac surgical patients. The most common nonpharmacologic approaches emphasize the reduction of macroemboli and microemboli. As described in earlier sections of this chapter, these strategies include avoiding aortic atheroma by using transesophageal or epiaortic echocardiography, optimizing the placement of the aortic cannula in the aorta, avoiding partial occlusion clamping of the aorta for proximal anastomoses by using a single cross-clamp, and, in selected patients, avoiding all cross-clamping of the aorta (the “no touch technique”). Other strategies to minimize particulate microemboli include the routine use of arterial filtration in the cardiopulmonary circuit and use of the cell saver before reinfusing blood suctioned into the cardiotomy, to remove particulate and lipid material. Strategies to minimize air microemboli include carefully removing air after any procedure involving opening a cardiac chamber, and flooding the field with carbon dioxide to minimize embolism of air entrained into the heart from the surgical field. Eliminating CPB by performing off-pump CABG surgery has been touted as a means to reduce emboli in selected patients, but this approach has not decreased POCD rates at 1 or 5 postoperative years. Many of the late-occurring strokes are probably caused by atrial fibrillation. Early pharmacologic or electrical correction and adequate anticoagulation are mandatory (see the section on arrhythmias in the postoperative period).

Other nonpharmacologic approaches include perioperative temperature control, as well as blood gas management (α-stat or pH-stat) during hypothermic CPB. These considerations are discussed in the section on CPB.

Diabetes is recognized as a risk factor for stroke and delirium after cardiac operations. Even in nondiabetic patients, hyperglycemia is extremely common during cardiac surgical procedures because of the stress response to surgery (as well as CPB), the resultant increases in circulating catecholamines and cortisol, and hypothermia-induced reductions in insulin effectiveness. Experimental evidence has revealed a relationship between hyperglycemia and worse outcome after different types of neurologic injury. However, attenuating the hyperglycemic response to cardiac surgery or CPB has proved difficult. Furthermore, concerns exist regarding inducing even a single episode of severe hypoglycemia with aggressive attempts to control glucose levels. The aim of perioperative glucose control for most patients is a level between 140 and 180 mg/dL.

The influence of intraoperative hemodynamics may influence neurologic and other outcomes. The prospective, randomized trial conducted by Gold and colleagues compared maintaining “normal” MAP (minimum of 50 mm Hg) with keeping the MAP elevated (target of 80-100 mm Hg). These investigators found that the overall incidence of combined cardiac and neurologic complications was significantly lower in the high-MAP group. In another study, investigators noted a higher frequency of hypoperfusion-type “watershed” strokes in patients who underwent cardiac surgical procedures with a MAP that was at least 10 mm Hg lower during CPB than before CPB. The current recommendation is to maintain higher targets for MAP (i.e., >50 mm Hg) in patients at high risk for neurologic injury. Because high-risk patients can be identified and a defined therapeutic window exists, pharmacologic protection would be very desirable. However, no definitively proven drug therapy currently exists for the prevention or treatment of neurologic injury after cardiac surgery.

Cerebral oximetry has been proposed to be of benefit in preventing neurologic injury in cardiac surgery, but the evidence is still lacking in this regard.

Postoperative Management of Central Nervous System Injury or Dysfunction

Postoperative stroke may be suggested by the patient’s inability to follow commands or move all extremities shortly after the surgical procedure. Neurologic consultation should be obtained, including diagnostic imaging. Diffusion-weighted magnetic resonance imaging is the most sensitive and accurate imaging technique for postoperative cardiac surgical patients. It detects more microembolic lesions than does conventional magnetic resonance imaging and is better able to find multiple watershed lesions.

Management of CNS injury or dysfunction after cardiac operations consists of general supportive measures. Hypotension should be avoided, and consideration should be given to using volume expansion, inotropic augmentation (pharmacologic or mechanical), or vasoactive medications (vasopressin or phenylephrine) to support blood pressure and brain perfusion. The brain oxygen supply-demand balance should be optimized through adequate oxygenation, sedation, and strict temperature control. Hyperthermia (fever) should be aggressively controlled. Both hyperglycemia and hypoglycemia should be avoided. Lytic therapy is of little use in cardiac surgery because it poses a risk of postoperative surgical bleeding.

POCD can manifest subtly, being evident only with psychometric testing, or, at the other extreme, it can manifest as postoperative delirium. Delirium is defined as an acute and overt change in cognition and attention, which may include alterations in consciousness and disorganized thinking. In the cardiac surgical literature, the reported incidence of delirium depends greatly on the methods used for delirium assessment, varying from 3% (chart review only) to 8% (interviews with nurses); however, with rigorous daily mental status testing and the application of a validated diagnostic algorithm, the incidence may be as high as 53%.

Risk factors include preexisting cognitive impairment, poor preoperative functional status, prior stroke or transient ischemic attack, depression, alcohol abuse, and abnormal preoperative laboratory values (glucose, sodium, potassium, and albumin). Precipitating factors for delirium in the postoperative period include intraoperative and postoperative medications, particularly sedatives and analgesics. The postoperative environment in the ICU often results in sleep deprivation and overstimulation, contributing to the onset of delirium. Complications of hospitalization and surgical procedures, including those that result in prolonged controlled ventilation and reduced mobility, also contribute to the development and severity of delirium.

Nonpharmacologic postoperative strategies to prevent delirium are listed in Table 54.9 . Medications, including those used to control pain and anxiety, are a common cause. Debate continues in the literature for a better drug to minimize the incidence of delirium between benzodiazepines and dexmedetomidine. For patients who develop agitation, the mainstay of treatment is a thorough review of medications, as well as removing other precipitating factors, such as low CO or perfusion state, metabolic disorders (e.g., hyperglycemia), fluid and electrolyte disturbances (hypoglycemia or hyperglycemia and uremia), constipation, urinary retention, and environmental noise. For patients in whom these nonpharmacologic interventions are not sufficient, an antipsychotic, usually haloperidol, is considered first-line therapy for agitation associated with delirium.

Delirium may accelerate cognitive decline in patients with Alzheimer disease or cause a type of posttraumatic stress disorder in younger patients. The long-term mental health implications of delirium have not been fully studied but may include impairment of functional recovery.

Peripheral Nerve Injuries

Peripheral nerve injuries are not uncommon in cardiac surgical patients and are often self-limited. These are the result of positioning of the upper extremities with inadequate padding of the ulnar nerve. In addition, the brachial plexus is also prone to injury from excessive stretch by the sternal retractor. Usual presenting symptoms are numbness, weakness, pain, decreased reflexes, and diminished coordination.

Phrenic nerve, recurrent laryngeal nerve, and sympathetic chain injuries have also been reported. Saphenous nerve injury during saphenous vein harvesting is also known to have been reported.