Chapter 17. Management after Cardiac Surgery

Cardiac surgery is one of the most commonly performed major operative procedures in the United States. The indications for cardiac surgery include myocardial ischemia and infarction, heart failure, valvular dysfunction, aortic pathology, and surgery for dysrythmias. The management of patients following cardiac surgery requires a multifaceted approach and the involvement of a team of specialists. While the intensivist is often the point person for the management of the patients following open heart surgery, it is essential that the management involve the surgeon, cardiologist, and anesthesiologist, and a wide variety of other health care providers.

Successful postoperative management following cardiac surgery requires a clear understanding of the patient's preoperative conditions and the intraoperative events and management. The goal is to restore the patient's normal physiologic condition and homeostasis. As medical management and interventional cardiology procedures evolve and improve, the patients being referred for cardiac surgery are sicker and more debilitated than they were in the past. This trend is likely to continue in the years to come. Despite these increased challenges facing cardiac surgeons, patient outcomes remain very good in large part due to the postoperative management and ICU care. A systems-oriented approach is often necessary to deal with the multitude of problems facing these patients, and the cardiac system is generally the primary determinant of recovery.1

Hemodynamic Management

The goal of hemodynamic management is to maintain adequate oxygen delivery to the tissues and to minimize demands on a heart that has just undergone major surgery. Optimization of cardiac output is essential to maintain function of the brain, kidneys, gut, lungs, and other end organs necessary for optimal recovery. Postoperatively, contractility is almost always diminished, the magnitude of which is often related to the severity of chronic dysfunction, ischemia, and intraoperative events.2

Despite the wide array of patient disease and cardiac procedures, significant similarities exist in patient monitoring, evaluation, and management.2 The majority of patients have continuous monitoring of EKG, pulse oximetry, blood pressure, central venous pressure (CVP) monitoring, and in most instances a pulmonary artery catheter (PAC) for monitoring mixed venous O2 saturation (SvO2), pulmonary artery pressures, and continuous cardiac output. Although the use of a PAC in patients undergoing cardiac surgery has been a topic of some debate,3 it has been common practice to use PACs in almost every patient undergoing cardiac surgery. These modalities allow for measurement of oxygen consumption and mixed venous and arterial oxygen saturation, and an estimation of cardiac output. The goal is to maintain normal hemodynamic values if possible since it has been shown that normal oxygen transport and normal SvO2 (>70%) in the immediate postoperative period can improve outcome.4 Although achieving these targets can be challenging,5–7 adjustments to volume status, afterload, heart rate and rhythm, and cardiac output can help to maximize end-organ oxygen delivery.

Blood Pressure

Mean arterial pressure (MAP) is the most dynamic physiologic variable in the first hours following cardiac surgery.2 This can be due to a number of factors including decreased preload, vasodilatation, and cardiac contractility. Many patients' end organs are dependent on higher blood pressures; however, concerns about bleeding often lead to a struggle between maintaining a higher MAP for perfusion and keeping the pressure low to protect suture lines. Despite these concerns, the MAP should be maintained above 65 mm Hg. Volume resuscitation can be guided by CVP, and, although there is no gold standard for the type of fluid used for resuscitation, it has been our practice to use 5% albumin. There are concerns, however, about albumin extravasation,8 and no clinical difference has been shown to suggest that there is any benefit to colloid versus crystalloid.9 Once the patient is adequately volume resuscitated, there are a number of pharmacologic agents (Table 17-1) that can be useful for increasing vascular tone and cardiac contractility. It is essential that the intensivist have a thorough understanding of the mechanism of each of these agents and their interactions with one another.

In addition to hypotension, a small portion of patients may present with significant levels of hypertension.2,10 This may lead to excessive bleeding and an increase in afterload that can exacerbate low cardiac output and should be treated aggressively with vasodilators.

Cardiac Contractility

Myocardial contractility following cardiac surgery can be a variable and dynamic process. Patients with low cardiac output following cardiac surgery are at significant risk for end-organ hypoperfusion if the cardiac index (CI) is below 2.2. It is essential that the source for low CI be identified and treated rapidly. Once hypovolemia, bleeding, and tamponade have been ruled out, the focus should be on instituting pharmacologic and mechanical support for the failing heart. The most useful drugs (Table 17-1) have significant inotropic and vasodilatory activity.11 Epinephrine has both α and β effects and is very useful in the postoperative period. The β2 effects are seen predominantly at lower doses, whereas the α effects predominate at higher doses. Both dobutamine and dopamine are effective β-agonists with broad dosage-dependent effects, but dobutamine has a superior effect on cardiac contractility. Milrinone, which is a cyclic phosphodiesterase inhibitor, works by increasing cAMP levels leading to increase in calcium flux and myocardial contractility. In addition to its inotropic affect, milrinone decreases vascular tone, particularly in the pulmonary bed. A randomized trial by Feneck et al11 compared the effects of milrinone and dobutamine on low-output syndrome following cardiac surgery. They found that patients receiving dobutamine had a higher CI, heart rate, and left ventricular (LV) stroke work index than those receiving milrinone, whereas milrinone led to a greater decrease in pulmonary capillary wedge pressure. Dobutamine, however, was associated with a higher incidence of hypertension and conversion from sinus rhythm to atrial fibrillation (AF).

Mechanical Support for Low Cardiac Output

Pharmacologic support is generally effective in helping separate from cardiopulmonary bypass (CPB) and in supporting CI in the postoperative period. There are situations, however, where mechanical support is necessary such as when the CI does not increase above 2.0 despite maximum inotropic support. The intra-aortic balloon pump (IABP) is a mechanical circulatory assist device developed in 1968 by Kantrowitz et al.12 The balloon is placed in the descending thoracic aorta just distal to the subclavian artery. The balloon inflates during diastole, increasing coronary perfusion, and deflates during systole, decreasing afterload. Its use has evolved from circulatory support of patients in cardiogenic shock to now being used as an aid to weaning patients from CPB and for postoperative low CI. It is now the most commonly used mechanical assist device with over 100,000 implanted annually.

Despite adequate volume loading, maximum inotropic support, and the use of IABP, a small percentage of patients will not be able to be weaned from CPB or will develop severe cardiogenic shock in the perioperative period. Over the last 20 years, the use of ventricular assist devices (VAD) has helped aid the failing heart after cardiac surgery. Studies have shown that patients requiring two or more high-dose inotropes to wean from CPB had better outcomes with early VAD insertion.13 While the details of the use of VADs are beyond the scope of this chapter, it is important to understand their utility in the management of the postcardiac surgery patient.

Rate and Rhythm

Following cardiac surgery, patients are prone to develop some form of dysrhythmia. Heart rate and conduction abnormalities are common following valvular and coronary surgery. Most surgeons place epicardial pacing leads on the atrium, ventricle, or both that are brought out in the subxiphoid region and can be helpful in assisting in management of bradycardia and atrial dysrhythmias. Augmentation of the heart rate, even when normal, to rates of 90–100 can be useful to increase CI. Conversely, patients may be persistently tachycardic following surgery with rates up to 120, which may be due to a variety of drugs used in the postoperative period and which often requires no intervention.

Ventricular arrhythmias such as nonsustained ventricular tachycardia (VT) are not uncommon following cardiac surgery and should prompt an examination and correction of any electrolyte disturbance. Frequent episodes or periods of sustained VT may require more aggressive therapy with antiarrhythmic drugs such as amiodarone or lidocaine and electrical cardioversion if the patient is hemodynamically compromised. These episodes should also prompt a thorough investigation of possible ischemia, particularly in coronary artery bypass graft (CABG) patients.

Atrial Fibrillation

AF occurs in 15–45% of patients undergoing cardiac surgery with the highest incidence being in patients undergoing valve and combined CABG/valve procedures. The cause for postoperative AF is not well understood but is likely due to reentry of multiple waves of excitation throughout the atria.14 Several factors have been associated with increased risk for postoperative AF including advanced age, concomitant valve surgery, history of AF, congestive heart failure, COPD, and decreased LV function.15 A number of strategies have been implemented to reduce the incidence of AF. β-Adrenergic blockers have been extensively studied and have shown significant efficacy in reducing AF in the postoperative setting.15 β-Blockers should be started as early as possible in the postoperative period to reduce the incidence of AF. For patients who cannot be started on β-blockers, antiarrhythmic agents such as amiodarone and sotalol are both safe and efficacious in decreasing postoperative AF risk.

Despite the use of the preventive measures, AF remains a significant problem in the postoperative period. In most patients, AF is self-limiting and the treatment options vary based on the patient's clinical condition. One management strategy generally utilized in postoperative AF is demonstrated in Figure 17-1. For patients who are hemodynamically unstable, difficult to manage, or have a contraindication to anticoagulation, rhythm control is the preferred approach. However, in patients who are tolerating their AF, rate control and anticoagulation is the preferred approach since many of these patients will return to sinus rhythm within 3 months.

Bleeding, both intraoperatively and postoperatively, presents a significant challenge to the surgeon and intensivist in the management of the cardiac surgery patient. Excessive bleeding requires significant blood product usage that both is costly and adds significant morbidity and mortality to the patient.16 The needs of intraoperative anticoagulation combined with the effects of platelet dysfunction by the CPB circuit lead to postoperative coagulopathy. In addition, many patients now come to the operating room with some exposure to potent antiplatelet agents such as glycoprotein IIb/IIIa inhibitors or clopidogrel (Plavix). These agents significantly disrupt platelet aggregation and may lead to increased postoperative coagulopathy.17 In general, postoperative coagulopathy is usually due to a number of factors including thrombocytopenia, fibrinolysis, hypothermia, hemodilution, and residual or rebound heparin. This generally leads to some degree of chest tube output in the range of 50–100 cm3/h.

The treatment of postoperative bleeding is summarized in Table 17-2 and is dependent on the judgment of the surgeon and intensivist to determine if the patient is bleeding from surgical problems or coagulopathy. Chest tube inspection and management is essential to understand the pathophysiology of the bleeding. Generally, in coagulopathic bleeding there is no clot in the chest tubes and this can often be managed with correction of the coagulopathy and administration of blood products. Some other maneuvers that can be useful include increasing positive end-expiratory pressure (PEEP), use of epsilon aminocaproic acid, and warming. In addition, some have advocated the use of activated recombinant factor VII (rFVIIa) that has been widely reported in the management of bleeding after cardiac surgery. However, concerns about the safety, particularly in bypass patients with fresh grafts, have been raised.18

Mediastinal Re-Exploration

Patients in whom surgical bleeding is suspected often require re-exploration. This should be considered when bleeding exceeds 400 mL/h for the first hour, 300 mL/h for 2–3 hours, and 200 mL/h for 4 hours (Table 17-2). Chest tube output is not the only indicator of significant surgical bleeding since chest tubes can become clotted and blood can collect and clot in the pericardium. Hemodynamic instability not responsive to inotropic support or signs of tamponade such as elevated filling pressures or equalization of pressures should initiate a return to the operating room for re-exploration. Widening of the mediastinum on chest x-ray (CXR) or signs of tamponade on transesophageal echocardiogram can be helpful in patients where the diagnosis is questionable. On rare occasions, the patient may require re-exploration at the bedside if there is massive sudden hemorrhage or cardiac arrest is imminent. In those situations, the goal is to relieve the tamponade, restore cardiac contractile function, and temporarily control bleeding while the patient is returned to the operating room.

Postoperative pulmonary care is aimed at restoring normal pulmonary capillary permeability and interstitial lung volume, preventing or treating atelectasis, maintaining normal arterial blood gases, and preventing infection.

Early versus Delayed Extubation

Early extubation may be defined as extubation of a patient within 3–6 hours of arrival in the ICU. The goals, as with any extubation, are to have a patient who is awake enough to protect his or her airway and adequately oxygenate and ventilate. Weaning of sedation should begin early after admission to the ICU as long as the patient is hemodynamically stable with manageable chest tube output.

There are a variety of approaches to rapid weaning including the rapid shallow breathing index (RSBI) that is a calculated by dividing a patient's tidal volume into the observed respiratory rate during a spontaneous breathing trial of 10–30 minutes. Statistically, an RSBI of <105 is thought to be highly predictive of a successful extubation (See Chapter 5).

A T-tube trial is another way of performing a spontaneous breathing trial. The patient has supplemental oxygen administered while remaining intubated. There is no ventilatory support. Patients who tolerate a T-piece trial have a high chance of successful extubation presumably because breathing through a T-piece is more difficult than normal breathing.

Other clinicians rely on the traditional method of converting the ventilation mode to a partial support (assist control or intermittent mandatory ventilation with pressure-supported [PS] spontaneous breaths) or a total spontaneous setting when the patient is fully awake. The decision to extubate is then based on clinical observation, O2 saturation, and blood gas results. Minimum settings before extubation typically are PS of 10, with a PEEP of 5. This amount of PS helps to overcome endotracheal tube resistance, and PEEP helps to maintain and recruit alveoli. No particular weaning technique has been shown to be superior to any other, and the literature does not implicate any technique as being more commonly associated with adverse outcome.

The postoperative CXR should also factor into the decision-making process. Pulmonary edema and pleural effusions are common in the early postoperative period, but are usually not large enough to be significant. Patients who do not meet the criteria for early extubation may develop larger effusions over time that may need to be drained in order to maximize the likelihood of a successful extubation. Cardiac surgery patients are also at risk for worsening pulmonary edema; therefore, diuresis may be helpful in patients who cannot be extubated early.

Definitive outcome studies are only beginning to be presented, but the data so far suggest that early extubation is not associated with any increased risk of mortality or morbidity.19 An audit by Akhtar and Hamid of 388 postoperative CABG patients found that 196 (49.5%) patients could be extubated within 6 hours of arrival in the cardiac ICU. Of the remaining patients, reasons for prolonged intubation included deep sedation in 80 patients (46.5%), confusion in 44 (25%), excessive bleeding in 20 (11.3%), and high inotropic support in 10 (5.68%).20

Factors that may adversely impact successful early extubation include the effects of anesthesia on postoperative hemodynamics, stress responses and awareness, altered management in the control of pain, shivering and ischemia in the early postoperative period, and the risks of reintubation in patients who might require reoperation for bleeding. Risk factors for delayed extubation of postoperative CABG patients include increased age, female gender, postoperative use of IABP, inotrope requirement, bleeding, and atrial arrhythmia.21

Pulmonary Management Following Extubation

Once extubated, patients will need supplementary oxygen. This is usually given as a 40% face mask, and then gradually weaned to a nasal cannula. The use of bedside incentive spirometry and chest physiotherapy will help reduce atelectasis and the risk of pneumonia. In addition, short-term β-agonists help in the postextubation recovery phase, even in patients without a history of COPD or reactive airway disease.

A median sternotomy or thoracotomy incision is associated with significant pain, splinting, and decreased chest wall compliance, resulting in shallow breaths, atelectasis, and an increased risk of pneumonia.22 Adequate pain control reduces the likelihood of postoperative splinting, atelectasis, and pneumonia. Opiates combined with rapid-acting NSAIDs such as ketorolac have been used with success.23 Patient-controlled analgesia (PCA) may have the advantage of improved pain control as well as a reduction in the occurrence of atelectasis when compared with nurse-controlled analgesia.

Pulmonary Complications

Atelectasis and pleural effusion are among the more common complications following cardiac surgery.24 In addition, there are a number of other complications (see Table 17-3) that may complicate postoperative recovery, leading to longer stays in the ICU and hospital and increasing overall morbidity and mortality. Recognizing these potential complications and treating them early is essential to ensuring a successful postoperative outcome.

Postoperative Pulmonary Dysfunction

Postoperative pulmonary dysfunction (PPD) is a constellation of events that occurs in patients undergoing cardiac surgery. It may delay extubation, factor into extubation failure, and delay functional recovery. PPD involves alterations in pulmonary function such as increased work of breathing, shallow respiration, ineffective cough, and relative hypoxemia. All postoperative cardiac surgery patients develop some degree of PPD.

The underlying basis for PPD is the development of abnormal gas exchange, and alterations in lung mechanics. Gas exchange abnormalities include a widened alveolar–arterial oxygen gradient, increased microvascular permeability in the lung, increased pulmonary vascular resistance, increased pulmonary shunt fraction, and intrapulmonary aggregation of leukocytes and platelets.25 Alterations in the mechanical properties of the lung lead to reductions in vital capacity, functional residual capacity, and static and dynamic lung compliance.

Premorbid conditions such as COPD and undiagnosed preoperative pneumonia can contribute to the development of pulmonary dysfunction after cardiac surgery. Postoperatively, these patients may develop pulmonary edema, atelectasis, or pneumonia. The risk of developing pneumonia increases with delayed extubation. In addition, shallow breathing, a weak cough, and splinting from inadequate pain control can also contribute to the development of PPD, leading to respiratory failure and reintubation.

CPB can contribute to pulmonary dysfunction by increasing left atrial or pulmonary venous pressure. These effects combined with reduced plasma oncotic pressure can increase extravascular lung water.26,27 During CPB, cytotoxic and vasoactive mediators of the inflammatory response28–32 and circulating microemboli33 may reach the lung via bronchial arteries. This inflammatory response has been termed “pump lung” or “postpump syndrome.” These agents increase pulmonary capillary permeability, perivascular edema, and bronchial secretions. Once CPB commences, the cessation of pulmonary ventilation results in collapsed lungs and insufficient alveolar distention to activate the production of surfactant, a situation that potentiates alveolar collapse. Abnormal pulmonary mechanics, retention of secretions, and atelectasis can also occur.25

The cumulative effect of premorbid conditions and CPB contributes to the development of an increased work of breathing, postoperative atelectasis, pulmonary edema, as well as an increased susceptibility to infection.

Pleural Effusions

Pleural effusions are common after cardiac surgery including CABG and can be categorized as: perioperative (within the first week), early (within 1 month), late (2–12 months), or persistent (after 6 months).34 Among patients undergoing CABG, the prevalence of pleural effusions in the immediate postoperative period is high. In the week after CABG, the reported prevalence of pleural effusions has ranged from 40% to 75%.35–39 Most effusions are small, unilateral, left-sided, and asymptomatic. In a study by Labidi et al, almost 7% of patients had a clinically significant pleural effusion in the 30 days postsurgery.40 Peng et al conducted a similar study of 356 patients who were available for evaluation 1 month after undergoing CABG. The initial diagnosis of a newly developed symptomatic large pleural effusion was made in 11 patients (3.1%) within 30 days of CABG. Eight had a pleural effusion predominantly on the left side and three on the right.41 When the presence of a pleural effusion hinders extubation or causes pulmonary dysfunction, tube thoracostomy, thoracentesis, or placement of a pig-tail catheter may be indicated.

Pulmonary Edema

CPB can cause cardiogenic pulmonary edema from hemodilution, volume overload, and reduction in oncotic pressure. It can also cause noncardiogenic pulmonary edema (NCPE) by producing a systemic inflammatory response syndrome (SIRS). This involves an increase in capillary permeability and accumulation of extravascular lung water. Surfactant production is also decreased resulting in atelectasis. Other potential causes of postoperative NCPE include blood transfusions, the administration of fresh frozen plasma to control bleeding, and preexisting lung conditions.42 Additionally, protamine sulfate, which is often given to reverse the effects of heparin intraoperatively, has been associated with causing NCPE on rare occasions.43

Most cases of postoperative pulmonary edema are mild and can be treated with early diuresis. Fulminant NCPE, although rare, is associated with a high mortality. SIRS can progress to full-blown adult respiratory distress syndrome (ARDS). This syndrome is diagnosed by the presence of bilateral patchy infiltrates on CXR, normal cardiac filling pressures, relative hypoxemia, and a Pao2/FiO2 ratio of <200. Management of NCPE or ARDS involves mechanical ventilation with as low an FiO2 as possible to maintain a Po2 of 60–70. Higher settings of PEEP may be needed and must be balanced with the effect of reducing cardiac output by reducing venous return. Ventilator settings with a tidal volume of 6 mL/kg of predicted body weight have been shown to improve survival when compared with higher tidal volume.

Other maneuvers that may be useful in the management of ARDS include prone positioning that improves oxygenation but poses safety concerns in the poststernotomy patient, and high levels of PEEP (35–40 cm H2O) that have not been shown to improve survival. High-frequency oscillatory ventilation is, in theory, the ideal “lung-protective” method, but its benefits have not been proven. No drug therapy, including corticosteroids, has been shown to improve survival in patients with ARDS. Inhaled nitric oxide, although useful in decreasing pulmonary vascular resistance and decreasing right heart failure, has no substantial impact on the duration of ventilatory support or mortality.

Patients undergoing cardiac surgery often have some degree of peripheral vascular disease, diabetes, or other predisposing factors that impact renal function. Patients at risk for developing postoperative acute kidney injury (AKI) include those of increasing age, as well as those with a history of hypertension, diabetes mellitus, and CHF. Additionally, patients who require lengthy procedures on bypass are also at risk.44–46 AKI is a major complication of CABG surgery that is strongly associated with in-hospital mortality.47 The incidence of AKI may be increasing despite a trend of decreasing in-hospital mortality. Some have suggested that this is due to broader criteria for the diagnosis.48 Some degree of renal injury almost always occurs during CPB, and postperfusion proteinuria occurs in many patients.49 Up to 30% of patients who undergo CABG develop some degree of acute renal impairment.50

The Society of Thoracic Surgeons National Cardiac Surgery database defines postoperative new renal failure as a serum creatinine of >2.0 mg/dL, doubling of peak preoperative creatinine, or requirement of dialysis. The Acute Dialysis Quality Initiative formulated the Risk, Injury, Failure, Loss, and End-Stage Kidney (RIFLE) classification.51 RIFLE defines three grades of increasing severity of AKI—risk (class R), injury (class I), and failure (class F)—and two outcome classes (loss and end-stage kidney disease). The RIFLE classification grading is based on changes in either serum creatinine or urine output from the baseline condition (see Table 17-4).

Renal Protection

AKI can be classified on the basis of the underlying pathology. The majority of postoperative AKI is due to acute tubular necrosis (ATN) or prerenal azotemia. Postobstructive uropathy and glomerulonephritis can also occur. Prerenal azotemia often develops from hypoperfusion and ischemia, but is effectively managed by restoration of normal blood flow. This may be accomplished with volume supplementation with fluid or blood, or increasing cardiac output with inotropes. ATN is thought to arise from a variety of insults to the kidney in these patients, including ischemia, general anesthesia, radiocontrast dyes, and heart failure. Blood flow decreases in the early phase of ATN, and therefore vasodilators theoretically should reduce necrosis in the kidney by restoring blood flow to the tubules. However, a prospective single-center, randomized, double-blind trial involving 80 patients undergoing cardiac surgery studied the effects of fenoldopam at 0.05 μg/kg/min or dopamine at 2.5 μg/kg/min after the induction of anesthesia for a 24-hour period. Peak postoperative serum creatinine level, intensive care unit and hospital stay, and mortality were similar in the two groups. They concluded that there was no renal protective benefit afforded by the vasodilator fenoldopam in a high-risk population undergoing cardiac surgery.53

CPB and cardioplegic arrest are associated with free radical formation. These free radicals are thought to cause damage to various organs including the kidneys. Several studies have looked at free radical scavenging with agents such as N-acetylcysteine as a way to preserve renal function. Sisillo et al showed no difference in mortality regarding the use of N-acetylcysteine, although their data did suggest a decrease in the incidence of postoperative acute renal failure and ventilator days.54 Barr and Kolodner studied 79 patients with preexisting renal disease who underwent cardiac surgery.55 Their data suggested a protective effect of fenoldopam, N-acetylcysteine, or the combination when given before surgery, compared with controls. However, there was no decrease in length of ICU or hospital stay and fenoldopam was associated with perioperative hypotension.

Prognosis

Ryckwaert et al, in a study of 591 patients, determined that a postoperative 20% increase in plasma creatinine after cardiac surgery was associated with an increase in mortality, especially when accompanied by multiple organ dysfunction.56 When acute renal failure is severe enough to require renal replacement therapy (RRT), mortality rates are 50–90% compared with <3% for patients without AKI.50,57 Possible explanations for the increased mortality associated with AKI are salt and water retention resulting in volume overload, hyperkalemia, and acid–base derangements.58 These derangements may result in hypertension, hypotension, changes in cardiac output, as well as changes in blood flow to the liver and other organs. There is evidence that AKI may lead to insulin resistance and protein breakdown, and immunocompromise.59 Patients with AKI also have a high incidence of infectious complications60 and frequently develop anemia. Finally, AKI itself can lead to a noninfectious, proinflammatory response with activation of leukocytes, secretion of proinflammatory cytokines, and recruitment of neutrophils and macrophages with resultant lung injury, as has been demonstrated in animal models of ischemia–reperfusion-induced acute renal failure.61,62

Gastrointestinal (GI) maladies are infrequent but serious complications of cardiac surgery, with high rates of morbidity and mortality. The incidence is low ranging from 0.41% to 2.0%63–67; however, the mortality rate has been reported to be as high as 63%.65–68 Patients most at risk for dying include those with New York Heart Association (NYHA) class IV and unstable symptoms, an increased need for preoperative IABP support, the need for GI surgical intervention, and patients with ischemic bowel. Zacharias et al identified eight parameters that predicted GI complications: age greater than 70 years, long duration of CPB, need for blood transfusions, reoperation, triple vessel disease, NYHA functional class IV, peripheral vascular disease, and congestive heart failure. They suggest that intra-abdominal injury is usually ischemic in nature due to low cardiac output, hypotension, blood loss, or intra-abdominal atheroemboli.69 Other patients at risk include those with combined CABG–valve operations, prolonged ventilation time, female sex, need for vasopressors, sternal wound infection, and a history of peptic ulcer disease.65,67,70

The list of potential complications includes GI bleeding (most common), acute pancreatitis, perforated peptic ulcer, intestinal ischemia, cholecystitis, and small bowel obstruction.67,71 Interestingly, Mangi et al, in their study of 8,709 patients, found the most frequent serious GI complication to be mesenteric ischemia, which developed in 67% of patients who suffered a serious GI complication.64

Postoperative Care

Postoperatively, it is important to maintain GI perfusion and an adequate CVP. The common etiological factor in developing GI complications of any kind, after cardiac surgery, seems to be postoperative splanchnic hypoperfusion with visceral ischemia.72 CPB has been shown to decrease gastric pH, increasing the risk of erosion and bleeding. Stress ulcer prophylaxis with proton pump inhibitors reduces this risk. Other postoperative preventive measures that have had variable results in the literature include selective gut decontamination, early enteral feeding, and adjuvants to promote gut function such as glutamine, fiber, and growth hormone.73

Ideally, patients are extubated early in the postoperative period and are started on oral diets. For patients who are unstable or in whom complications develop that necessitate a prolonged intubation, early enteral feeding should be considered. Enteral nutritional support for critically ill patients is thought to maintain GI mucosal integrity and barrier function, and stimulate splanchnic and GI-associated lymphoid tissue blood flow. Further, when compared with parenteral nutrition, enteral nutrition improves substrate utilization, reduces the risk of sepsis, and decreases cost.74

Incidence

Stroke and other neurologic impairments are among the most dreaded complications of open heart surgery for both the surgeon and the patient. The likelihood of perioperative stroke varies between 1% and 5% in most published series and is dependent on a multitude of risk factors.75 Cognitive deterioration after cardiac surgery is far more common, although the incidence varies widely. It may affect as many as 80% of patients a few days after surgery and may persist in up to one third of patients.76

A retrospective report from the Society of Thoracic Surgery National Cardiac Database, of over 400,000 cardiac surgeries between 1996 and 1997, reported an overall incidence of a new neurologic event (stroke, transient ischemic attack, or unexplained coma lasting more than 24 hours) of 3.3%.77 A prospective study evaluated 2,108 patients undergoing CABG at 24 hospitals in the United States between 1991 and 1993. Overall, 6.1% suffered a cerebral complication, roughly equally divided between stroke and encephalopathy.78

Neurologic injury after cardiac surgery can be classified into two types. Type I includes stroke, seizures, stupor, or coma. Type II is more common and includes intellectual deterioration and memory deficit. These complications are most often caused by microemboli and/or hypoperfusion. There may also be a contribution from the effects of general anesthesia. Some have theorized that the degree of aortic manipulation and clamping during cardiac surgery may be the predominate cause of neurologic injury later.79,80

Patients at Risk

A study by Kolkka and Hilberman examined the characteristics of patients with neurologic complications after cardiac surgery. When compared with the 169 patients with no evidence of neurologic or neuropsychological dysfunction at discharge, the 35 patients with complications were older (65 ± 10 years vs. 55 ± 12 years), had a lower incidence of CABG as the sole surgical procedure (29% vs. 55%), a higher mortality rate (11.4% vs. 1.8%), and prolonged CPB (140 ± 45 minutes vs. 107 ± 38 minutes).81

Hammon identified several risk factors for neurologic damage (Figure 17-2) including proximal aortic atherosclerosis, history of neurologic disease, diabetes mellitus, use of IABP, history of hypertension, history of pulmonary disease, history of unstable angina, age (per additional decade), systolic blood pressure >180 mm Hg at admission, history of excessive alcohol consumption, history of CABG, dysrhythmia on the day of surgery, antihypertensive therapy, perioperative hypotension, ventricular venting, congestive heart failure on the day of surgery, and a history of peripheral vascular disease.82

Postoperative Care and Prevention

The observation of a patient's neurologic decline after surgery is most distressing to family members. Early neurocognitive dysfunction (within 3 months of operation) is most likely related to a combination of factors that include microemboli, relative hypotension, general anesthesia, and the overall inflammatory condition initiated by CPB. Neurocognitive deficits that are present after 3 months are often permanent.82

There is evidence that late cognitive changes are more likely related to the presence of preoperative neurologic conditions. Patients with atherosclerotic disease that has progressed enough to require CABG often have a similar degree of cerebrovascular disease. Many have had silent cerebrovascular events. Perioperative carotid endarterectomy in patients with significant carotid artery disease helps to reduce the incidence of postoperative neurologic complications. Studies have demonstrated that neurocognitive outcomes in patients who underwent standard CABG did not differ from those in a comparable control group without surgery, at both 1 and 3 years suggesting that neurocognitive decline is not due to the surgical procedure or CPB.83–85 On-pump versus off-pump procedures also do not seem to make a difference. The Best Bypass Surgery trial compared neurocognitive outcomes between off-pump patients and those patients who underwent on-pump CABG with CPB. They concluded that “in elderly high-risk patients, no significant difference was found in the incidence of cognitive dysfunction 3 months after either off-pump or on pump CABG.”86

Hyperglycemia in hospitalized patients has been shown to increase both morbidity and mortality, even in nondiabetic patients.87 Several trials have shown the benefits of intensive insulin therapy in critically ill patients, and particularly in patients undergoing cardiac surgery.88 A trial by van den Berghe et al showed significant decreases in mortality, bloodstream infections, acute renal failure, blood transfusions, and critical care polyneuropathy in patients managed with strict glucose control.89 Many centers have since tried to achieve strict glycemic control with a goal of blood glucose levels between 80 and 110 mg/dL.

The benefits of intensive glucose control in critically ill patients have recently come into question. Some trials have shown no benefit to intensive glucose control90 as well as an increase in the incidence of hypoglycemia.90,91 The Normoglycemia in Intensive Care Evaluation—Survival Using Glucose Algorithm Regulation (NICE-SUGAR) trial showed that intensive glucose control in critically ill adult patients actually increased the risk for death by 10%.92 The investigators reported that intensive glucose control increased the absolute risk of death at 90 days by 2.6 percentage points, and that the difference in mortality was significant even after adjusting for potential confounders. Hypoglycemia was “significantly more common” in the intensive control group. The authors concluded that intensive glucose control increased mortality among adults in the ICU and that a blood glucose target of 180 mg/dL or less resulted in lower mortality than did a target of 81–108 mg/dL. Intensively lowering blood glucose to a target of 81–108 mg/dL does not benefit critically ill patients and may well increase their risk of dying.

The risks of intensive insulin therapy may outweigh the benefits. Intensive glucose control involves insulin infusions that require close monitoring. This may also increase expenditures as well as the workload for the intensive care unit staff. At this time, a reasonable approach to hyperglycemia should have the goals of maintaining blood glucose levels as close to normal as possible with minimal fluctuations, hypoglycemia, or hypokalemia.