Anesthetic Techniques

Few trials have compared inhalation agents for FTCA. A single trial comparing sevoflurane and isoflurane in patients undergoing valve surgery was unable to demonstrate reductions in tracheal extubation times.¹ Several studies have examined the effectiveness of propofol versus inhalation agent, which demonstrated reductions in myocardial enzyme release (creatine kinase-MB, troponin I) and preservation of myocardial function in patients receiving inhalation agents.^2–5 Although this end point is a surrogate for myocardial damage and does not show improved outcome per se, creatine kinase-MB release post-CABG may be associated with poor outcome⁶ (Box 33-1).

The choice of muscle relaxant in FTCA is important to reduce the incidence of muscle weakness in the cardiac recovery area (CRA), which may delay tracheal extubation.⁷ Several randomized trials have compared rocuronium (0.5 to 1 mg/kg) versus pancuronium (0.1 mg/kg) and found significant differences in residual paralysis in the ICU.^8–11 Two studies found statistically significant delays in the time to extubation in the pancuronium group.^9,10 None of the trials used reversal agents, so the use of pancuronium appears acceptable as long as neostigmine or edrophonium is administered to patients with residual neuromuscular weakness.

Several trials have examined the use of different short-acting narcotic agents during FTCA. In these trials, fentanyl, remifentanil, and sufentanil all were found to be efficacious for early tracheal extubation.^12–14 The anesthetic drugs and their suggested dosages are listed in Table 33-1.

TABLE 33-1 Suggested Dosages for Fast-Track Cardiac Anesthesia

Evidence Supporting Fast-Track Cardiac Recovery

Several randomized trials and one meta-analysis of randomized trials have addressed the question of safety of FTCA.^15–21 None of the trials was able to demonstrate differences in outcomes between the fast-track group and the conventional anesthesia group (Figure 33-1). The meta-analysis of randomized trials demonstrated a reduction in the duration of intubation by 8 hours (Figure 33-2) and the ICU length of stay (LOS) by 5 hours in favor of the fast-track group. However, the length of hospital stay was not statistically different.

Figure 33-1 Forrest plot of mortality indicating no difference when fast-track cardiac anesthesia (FTCA) was compared with conventional high-dose narcotic anesthesia. TCA, traditional cardiac anesthesia.

Figure 33-2 Forrest plot showing the weighted mean difference in extubation times.

The overall effect was an 8.1-hour reduction in extubation times. CI, confidence interval; FTCA, fast-track cardiac anesthesia; TCA, traditional cardiac anesthesia.

One concern with FTCA is the potential for an increase in the incidence of adverse events, notably awareness. Awareness in patients undergoing FTCA was systematically investigated in a single trial, a prospective observational study of 617 FTCA patients. The reported incidence rate of explicit intraoperative awareness was 0.3% (2/608).²² This is comparable with the reported incidence during conventional cardiac surgery.²³ This suggests that FTCA does not increase the incidence of awareness compared with conventional cardiac surgery.

FTCA appears safe in comparison with conventional high-dose narcotic anesthesia. It reduces the duration of ventilation and ICU LOS considerably without increasing the incidence of awareness or other adverse events.^20,21 It appears effective at reducing costs and resource utilization.²⁴ As such, it is becoming the standard of care in many cardiac centers. The usual practice at many institutions is to treat all patients as fast-track candidates with the goal of allowing early tracheal extubation for every patient. However, if complications occur that would prevent early tracheal extubation, then the management strategy is modified accordingly. It has been demonstrated that the risk factors for delayed tracheal extubation (> 10 hours) are increased by age, female sex, postoperative use of intra-aortic balloon pump (IABP), inotropes, bleeding, and atrial arrhythmia. The risk factors for prolonged ICU LOS (> 48 hours) are those of delayed tracheal extubation plus preoperative MI and postoperative renal insufficiency.²⁵ Care should be taken to avoid excess bleeding (antifibrinolytics) and to treat arrhythmias either prophylactically or on occurrence (β-blockers, amiodarone).

Postcardiac Surgical Recovery Models

The failure of many randomized FTCA trials to show reductions in resource utilization likely stems from the traditional ICU models used by these centers during the study period. Even when trials were combined in a meta-analysis, the ICU LOS was reduced only by 5 hours despite patients being extubated a mean of 8 hours earlier.²¹ Typically, patients who are extubated within the first 24 hours of ICU admission are transferred to the ward on postoperative day 1 in the morning or early afternoon. This allows the following daytime cardiac cases to have available ICU beds but prevents patient transfers during nighttime hours. Two models have been proposed to deal with this issue: the parallel model and the integrated model (Figure 33-3). In the parallel model, patients are admitted directly to a CRA, where they are monitored with 1:1 nursing care until tracheal extubation. After this, the level of care is reduced to reflect reduced nursing requirements with ratios of 1:2 or 1:3. Any patients requiring overnight ventilation are transferred to the ICU for continuation of care. The primary drawback with the parallel model is the physical separation of the CRA and ICU, which leads to two separate units and, thus, does not eliminate the requirement to transfer patients. The integrated model overcomes these limitations because all patients are admitted to the same physical area, but postoperative management such as nursing-to-patient ratio is variable based on patient requirements.^26–28 Because nursing care accounts for 45% to 50% of ICU costs, reducing the nursing requirements where possible creates the greatest saving. Other cost savings from reductions in arterial blood gases (ABGs) measurement, use of sedative drugs, and ventilator maintenance are small. The goal is a postoperative unit that allows variable levels of monitoring and care based on patient need.²⁸ Furthermore, FTCA has been demonstrated to be a safe and cost-effective practice that decreases resource utilization after patient discharge from the index hospitalization up to 1-year follow-up.²⁹

Figure 33-3 Postcardiac surgical recovery models.

CRA, cardiac recovery area; ICU, intensive care unit; PCSU, postcardial surgical unit.

Ventilation Management: Admission to Tracheal Extubation

Ventilatory requirements should be managed with the goal of early tracheal extubation in patients (Table 33-4). ABGs are initially drawn within 1/2 hour after admission and then repeated as needed. Patients should be awake and cooperative, be hemodynamically stable, and have no active bleeding with coagulopathy. Respiratory strength should be assessed by hand grip or head lift to ensure complete reversal of neuromuscular blockade. The patient’s temperature should be more than 36° C, preferably normothermic. When these conditions are met and ABG results are within the reference range, tracheal extubation may take place. ABGs should be drawn about 30 minutes after tracheal extubation to ensure adequate ventilation with maintenance of Pao₂ and Paco₂. Inability to extubate patients as a result of respiratory failure, hemodynamic instability, or large amounts of mediastinal drainage will necessitate more complex weaning strategies (see Chapter 35). Some patients may arrive after extubation in the OR. Careful attention should be paid to these patients because they may subsequently develop respiratory failure. The patient’s respiratory rate should be monitored every 5 minutes during the first several hours. An ABG should be drawn on admission and 30 minutes later to ensure the patient is not retaining carbon dioxide. If the patient’s respirations become compromised, ventilatory support should be provided. Simple measures such as reminders to breathe may be effective in the narcotized/anesthetized patient. Low doses of naloxone (0.04 mg intravenously) also may be beneficial. Trials of continuous positive airway pressure or bilevel positive airway pressure may provide enough support to allow adequate ventilation. Reintubation should be avoided because it may delay recovery; however, it may become necessary if the earlier mentioned measures fail, resulting in hypoxemia, hypercarbia, and a declining level of consciousness.

TABLE 33-4 Ventilation Management Goals during the Initial Trial of Weaning from Extubation

Regulation of Hemoglobin Level

Anemia is common during and after cardiac surgery as a result of both dilutional changes and bleeding. Although a hemoglobin transfusion threshold of 10 g% was once common, increasing evidence suggests that a threshold of 7 g% is reasonably safe.³² However, in the post-CPB period, patients with incomplete revascularization or with poor target vessels may require a higher transfusion threshold.³² As a result, blood transfusions should be individualized for each patient but certainly should be used to maintain a minimal hemoglobin level of 7 g%.

Management of Bleeding

Chest tube drainage should be checked every 15 minutes after ICU admission to assess a patient’s coagulation status. Although blood loss is commonly divided into two types, surgical or medical, determining the cause of bleeding is often difficult. When bleeding exceeds 400 mL/hr during the first hour, 200 mL/hr for each of the first 2 hours, or 100 mL/hr over the first 4 hours, returning to the OR for chest reexploration should be considered. The clinical situation must be individualized for each patient, however, and in the face of a known coagulopathy, more liberal blood loss before chest re-exploration may be acceptable. There are numerous medical causes for bleeding after cardiac surgery. Platelet dysfunction after cardiac surgery is common. The CPB circuit itself leads to contact activation and degranulation of platelets, resulting in their dysfunction. Residual heparinization is common postcardiac surgery and frequently occurs when either heparinized pump blood is transfused after CPB or insufficient protamine is administered. Fibrinolysis is also common after CPB, predominantly caused by a host of activated inflammatory and coagulation pathways. Coagulation factors may decrease from activation at air–blood interfaces or from dilution with the CPB pump-priming solution. Hypothermia also may aggravate the coagulation cascade and lead to further bleeding. Conventional coagulation tests are helpful to identify the coagulation abnormality contributing to the bleeding. Common laboratory testing includes activated partial thromboplastin time, international normalized ratio (INR), platelet count, fibrinogen level, and d-dimers. Unfortunately, most conventional measures take 20 to 40 minutes before results are available. This has led to the development of new methods to help guide treatment. These bedside point-of-care tests are providing more rapid, clinically relevant results compared with laboratory testing. The use of point-of-care testing such as thromboelastography has been demonstrated to reduce transfusion requirements without increasing blood loss and is commonly used, especially following difficult cardiac cases^33,34 (see Chapters 17 and 28 to 31).

Initial medical treatment of excessive blood loss consists of 50 to 100 mg intravenous protamine to ensure complete heparin reversal. This may need to be repeated if heparinized CPB pump blood has been administered after protamine reversal. Although the reinfusion of chest tube blood was common to avoid exposure to donor packed red blood cells, it is no longer used routinely in practice because this blood is known to contain activated coagulation and inflammatory mediators that may predispose to an increased risk for infection.³⁵

Fresh-frozen plasma usually is given in the setting of an increased INR (> 1.5). Platelet levels of less than 100,000/mm³ may warrant platelet transfusion, but caution must be exercised when considering this course. Platelet transfusions carry the greatest risk for transfusion-related complications of any blood component, typically sepsis from bacterial contamination. Platelets should be used only when platelet counts are low or the patient has a known platelet dysfunction, secondary to the use of acetylsalicylic acid, glycoprotein IIb/IIIa inhibitors, or clopidogrel.³⁶ Certain physical measures should be instituted, including warming of the hypothermic patient. The benefit of positive end-expiratory pressure on postoperative bleeding is equivocal and likely has little benefit in the face of surgical bleeding or in patients who are coagulopathic.^37,38 The use of antifibrinolytics after cardiac surgery is likely of little benefit because several randomized trials were unable to demonstrate the efficacy of antifibrinolytics used after surgery^39,40 (Box 33-3).

Factor VIIa recently has become available and initially was introduced for treatment of hemophiliacs who present with bleeding. It was introduced in cardiac surgery as rescue therapy in patients with uncontrolled bleeding, usually in the presence of normal coagulation results and no surgical evidence of bleeding.⁴¹ Although frequently used in the OR before returning to the ICU, it is still given frequently in the ICU setting. Doses initially were in the range of 75 to 100 μg/kg, but concern over thrombotic complications has led to dosage reductions ranging down to as little as 17 μg/kg.^41–43

Electrolyte Management

Hypokalemia is common after cardiac surgery, especially if diuretics were given intraoperatively. Hypokalemia contributes to increased automaticity and may lead to ventricular arrhythmias, ventricular tachycardia, or ventricular fibrillation. Treatment consists of potassium infusions (20 mEq potassium in 50 mL D₅W infused over 1 hour) until the potassium exceeds 3.5 mEq/mL. In patients with frequent premature ventricular contractions caused by increased automaticity, 5.0 mEq/mL potassium may be desirable. Hypomagnesemia contributes to ventricular pre-excitation and may contribute to atrial fibrillation (AF). It is common in malnourished and sick patients, a frequent occurrence in the cardiac surgical setting. Management consists of intermittent boluses of magnesium—1 to 2 g over 15 minutes. Hypocalcemia also is frequent during cardiac surgery and may reduce cardiac contractility. Intermittent boluses of calcium chloride or calcium gluconate (1 g) may be required (Table 33-5).

TABLE 33-5 Common Electrolyte Abnormalities and Possible Treatment Options

Glucose Management

Diabetes is a common comorbidity (up to 30%) and is a known risk factor for adverse outcome in patients presenting for cardiac surgery.^44–46 Hyperglycemia itself is common during CPB. The risk factors for hyperglycemia include diabetes, administration of steroids before CPB, volume of glucose-containing solutions administered, and use of epinephrine infusions.⁴⁷ Poor perioperative glucose control is associated with increases in mortality and morbidity, including an increased risk for infection and a prolonged duration of ventilation.^48–52 In a large prospective, randomized, controlled trial of tight glucose control (blood glucose levels of 4.1 to 6.5 mmol/L) during postoperative ICU stay, reductions in mortality were shown by the authors compared with more liberal glucose control (blood glucose levels of 12 mmol/L).⁵² This trial enrolled both diabetic and nondiabetic hyperglycemic patients who underwent cardiothoracic surgery and demonstrated that tight management of glucose is beneficial in the CRA. However, another recent multicenter trial, as well as a meta-analysis of tight glucose control in the ICU, suggest an increase in harm, likely related to an increase in episodic hypoglycemia.^53,54 Therefore, it may be prudent to accept a more liberal blood sugar level (< 10.0 mmol/L) to reduce hypoglycemic episodes.

Pain Control

Pain control after cardiac surgery has become a concern as narcotic doses have been reduced to facilitate fast-track protocols. Intravenous morphine is still the mainstay of treatment for postcardiac surgery patients. The most common approach is patient-demanded, nurse-delivered intravenous morphine, and this treatment remains popular because of 1:1 to 1:2 nursing typically provided during cardiac recovery. However, with a change to more flexible nurse coverage and, therefore, higher nurse-to-patient ratios, patient-controlled analgesia morphine is becoming increasingly popular. Several studies have examined patient-controlled analgesia morphine use in patients after cardiac surgery.^55–61 A meta-analysis looking at patient-controlled analgesia morphine for postoperative pain showed small incremental benefits. However, young patients, those who use narcotics before surgery or are transferred to a regular ward within 24 hours, may benefit from patient-controlled analgesia for pain management⁶² (Table 33-6; see Chapter 38).

TABLE 33-6 Pain Management Options after Cardiac Surgery

Medications for Risk Reduction after Coronary Artery Bypass Graft Surgery

CABG surgery itself reduces the risk for mortality and angina recurrence, but several medical management issues may help maintain the long-term benefit after CABG surgery. Specifically, the use of aspirin, β-blockers, and lipid-lowering agents has been demonstrated to prolong survival or reduce graft restenosis, or both (Box 33-4).

Anticoagulation for Valve Surgery

Anticoagulation should be started in the early postoperative period for patients who have undergone valve replacement, with either a mechanical or bioprosthesis and also should be considered when AF complicates the postoperative course.¹⁰⁸ The recommended prophylactic regimens for patients with both mechanical and bioprosthetic heart valves are shown in Table 33-7. AF management is discussed later.

TABLE 33-7 Suggested Antithrombotic Therapy for Heart Valve Prophylaxis