Postoperative complications among patients undergoing cardiac surgery

INTRODUCTION — The use of cardiopulmonary bypass distinguishes cardiac surgery from other types of surgery. It also introduces a unique set of potential postoperative complications. These include vasospasm, altered platelet-endothelial cell interactions, and a generalized inflammatory response due to blood contacting the synthetic surfaces of the bypass equipment. The result is low flow in the microcirculation of the heart, brain, and other organs, which may lead to organ dysfunction [1,2].

The postoperative management of patients following cardiac surgery is reviewed here. Preoperative pulmonary assessment and potential complications of coronary artery bypass grafting are discussed separately. (See "Evaluation of perioperative pulmonary risk" and "Coronary artery bypass surgery: Perioperative medical management" and "Early noncardiac complications of coronary artery bypass graft surgery".)

MONITORING — Routine monitoring following cardiac surgery typically includes continuous telemetry, measurement of the arterial blood pressure via an arterial catheter, measurement of the cardiac filling pressures via a pulmonary artery catheter (ie, right heart catheter, Swan Ganz catheter), continuous assessment of the arterial oxygen saturation via pulse oximetry, and continuous measurement of the mixed venous oxygen saturation via an oximetric pulmonary artery catheter. Such monitoring allows instantaneous assessment of cardiopulmonary physiology. The expected postoperative values are listed in the table (table 1).

Although pulmonary artery catheters are used routinely in most centers, there are conflicting data in support of this practice [3,4]. In a trial that randomly assigned 1094 patients undergoing coronary artery bypass grafting to receive either a central venous catheter or a pulmonary artery catheter to assist in perioperative management, there were no significant differences in the length of intensive care unit stay, occurrence of postoperative myocardial infarction, in-hospital death, major hemodynamic aberrations, or major noncardiac complications [5]. In a retrospective propensity matched cohort analysis of 6844 patients undergoing cardiac surgery, patients with pulmonary artery catheters had a decreased length of stay and decreased morbidity, with increased infections and no difference in 30-day mortality [6]. Most pulmonary artery catheters are removed within 12 to 24 hours of surgery if significant vasopressor, vasodilator, or inotropic therapy is no longer required. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults" and "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults" and "Pulmonary artery catheters: Insertion technique in adults".)

Fluid shifts should be closely monitored by frequent assessment of the central venous pressure and/or pulmonary artery occlusion pressure (ie, pulmonary capillary wedge pressure), chest and mediastinal tube drainage, urine output, and patient weight. Measurement of arterial blood gases, hemoglobin concentration, platelet count, coagulation parameters, serum electrolytes, and serum creatinine is routinely performed on a daily basis.

Most experts maintain serum glucose between 140 and 180 mg/dL (7.8 and 10 mmol/L) during the period following cardiac surgery, since hyperglycemia is associated with worse outcomes in this population [7-10]. Hyperglycemia should be avoided. Further details regarding glucose control is provided separately. (See "Glycemic control in critically ill adult and pediatric patients".)

Monitoring following cardiac surgery initially occurs in the intensive care unit (ICU). However, the use of clinical practice guidelines and clinical pathways, combined with improvements in cardiac surgical care, has decreased the length of stay in the intensive care unit [11]. Many patients are now ready to be transferred to "step-down" units within 24 hours of surgery.

CARDIAC DYSFUNCTION — Poor cardiac function during the early postoperative period is associated with an increased risk of death [12]. It is usually suspected when there is unexplained postoperative hypotension, tachycardia, or pulmonary edema. Evaluation of suspected cardiac dysfunction consists of the following [13-15]:

●Telemetry – Assess the patient's continuous telemetry to identify or exclude cardiac dysrhythmias.

●Echocardiography – Perform transthoracic or transesophageal echocardiography to assess biventricular function, to look for residual or unexpected valvular disease, and to exclude pericardial tamponade.

●Invasive hemodynamic assessment via a pulmonary artery catheter – Use a pulmonary artery catheter to measure cardiopulmonary pressures, evaluate pressure tracings, determine cardiac output via thermodilution or the Fick equation, and calculate the systemic and pulmonary vascular resistances. This information can be used to identify valve dysfunction, detect pericardial tamponade, determine whether pulmonary edema is likely cardiogenic or noncardiogenic, and distinguish whether hypotension is most likely due to cardiac dysfunction, hypovolemia, or vasodilation.

●Electrocardiography – Obtain a 12-lead electrocardiogram to look for myocardial ischemia and to assess in detail any dysrhythmias noted on telemetry.

These four tests are likely to identify the cause of cardiac dysfunction. A chest radiograph may be helpful in the rare instances when these tests do not find the cause of cardiac dysfunction, since it can identify non-cardiac causes of cardiac dysfunction.

The most common causes of cardiac dysfunction following cardiac surgery are mechanical complications, physiologic complications (inadequate preload, excessive afterload, and poor ventricular inotropy), dysrhythmias, and myocardial infarction.

Mechanical complications — Mechanical complications of cardiac surgery are usually detected by echocardiography or invasive hemodynamic assessment. Examples include spasm or occlusion of a coronary artery graft, prosthetic valve paravalvular regurgitation, cardiac tamponade, hematoma, and systolic anterior motion of the mitral valve with left ventricular outflow tract obstruction. Treatment of the mechanical complications of cardiac surgery is usually surgical.

Not all mechanical complications of cardiac surgery involve the heart. Non-cardiac mechanical complications include pneumothorax, hemothorax, and endotracheal tube malposition. These complications can be readily identified on a chest radiograph. Pneumothorax and hemothorax often require tube thoracostomy, while endotracheal tube malposition requires repositioning.

Physiologic complications — Once mechanical complications have been excluded, an initial decision should be made about whether the patient's cardiac dysfunction (ie, low cardiac output) is most likely due to an insufficient stroke volume or heart rate. If it is likely due to diminished stroke volume, then it should be determined whether this is likely due to inadequate left ventricular preload (ie, intravascular hypovolemia), excessive left ventricular afterload (ie, hypertension), and/or poor inotropy (ie, cardiomyopathy). Data from the echocardiogram and/or invasive hemodynamic assessment can be used to inform these judgments. Treatment should be directed at the presumed abnormality, while the underlying cause is sought.

Inadequate preload — The best measure of left ventricular preload is the left ventricular end diastolic volume (LVEDV). It can be estimated by echocardiography or certain invasive hemodynamic measures: the pulmonary artery occlusion pressure (if a pulmonary artery catheter is in place) or the left atrial pressure (if a left atrial catheter has been placed). These hemodynamic measurements approximate the left ventricular end diastolic pressure (LVEDP), from which assumptions about the LVEDV can be made.

Left ventricular preload may be inadequate during the immediate postoperative period because of loss of vasomotor tone, increased capillary permeability, intraoperative and postoperative blood loss, or high urine output due to hypothermia. In addition, left ventricular compliance is frequently reduced following cardiac surgery, producing diastolic dysfunction; a higher LVEDP is required to maintain a given preload in affected patients [16]. The reduced compliance is the result of postischemic injury and "myocardial stunning," which results in inadequate myocardial diastolic relaxation. As an example, patients with significant left ventricular hypertrophy due to aortic stenosis or hypertrophic cardiomyopathy often require pulmonary artery occlusion pressures of 18 to 22 mmHg to maintain an adequate preload to support the cardiac output.

Inadequate preload should be corrected by administering intravascular volume. Such patients should be monitored closely because many will continue to exhibit signs of cardiac dysfunction even after their preload has been corrected. This suggests that there is coexisting poor inotropy and therapy should be redirected toward the poor inotropy as described below.

Excessive afterload — Postoperative hypertension is common; it can cause decreased stroke volume and increased myocardial oxygen demand [17,18]. The presumed etiology is systemic vasoconstriction, likely related to hypothermia induced during bypass [19]. The observation that rewarming does not closely correlate with resolution of the hypertension suggests that other factors may also be involved, such as humoral factors provoked by cardiopulmonary bypass [20,21]. Consequences of systemic vasoconstriction include tissue hypoxia in the skeletal muscles and secondary metabolic acidosis.

Patients judged to have excessive afterload should be treated with a vasodilator. Sodium nitroprusside is the vasodilator of choice to reduce excessive vasoconstriction in the immediate postoperative period. The blood pressure must be continuously monitored during therapy because vasoconstriction may improve quickly during rewarming from the hypothermia, leading to hypotension that requires immediate discontinuation of the sodium nitroprusside. Hypotension that persists despite the discontinuation of sodium nitroprusside should be initially treated with intravenous fluids, but may require vasopressors. Such hypotension is more common with normothermic bypass and longer cardiopulmonary bypass times, and less common in patients with diabetes, peripheral vascular disease, or a left ventricular ejection fraction of less than 40 percent [22,23].

Some patients will continue to exhibit signs of cardiac dysfunction even after their afterload has been reduced. This suggests there is coexisting poor inotropy and therapy should be redirected toward the poor inotropy as described below.

Poor inotropy — Impaired left ventricular function is suggested if the echocardiogram shows poor ventricular contraction and a low ejection fraction, or if the invasive hemodynamic measurements include a low cardiac output accompanied by a normal or high pulmonary artery occlusion pressure and normal or high systemic vascular resistance. Impaired right ventricular function is suggested if the echocardiogram demonstrates a dilated and poorly contractile right ventricle with an underfilled left ventricle, or if invasive hemodynamic measurements include a low cardiac output with a high right atrial pressure and a comparatively normal or low left atrial pressure. Postoperative ventricular function may be decreased due to intraoperative events, postoperative events, and/or hibernating myocardium. Examples include inadequate myocardial protection during cross-clamping of the aorta, ischemic myocardial injury during off-pump operations, uncorrected valvular lesions, reduced or inadequate intraoperative coronary blood flow, cardiac tamponade, or ischemia or infarction due to coronary artery air embolus, coronary occlusion, coronary graft vasospasm, or coronary graft thrombosis.

Patients with poor inotropy due to ischemia, poor inotropy plus inadequate preload, or poor inotropy plus excessive afterload, should initially have their ischemia, inadequate preload, or excessive afterload corrected, respectively. Correction of the ischemia or excessive afterload may improve inotropy, rendering an inotropic agent unnecessary. Correction of the inadequate preload makes it less likely that hypotension will occur if an inotropic agent is started (most inotropic agents cause vasodilation at low doses and, thus, hypotension is common immediately following initiation of the inotropic agent).

Patients with persistent poor inotropy despite correction of potential underlying causes require pharmacologic inotropic support to augment contractility. The patients who are most likely to benefit from inotropic support are those with a cardiac index less than 2.0 liters/minute/m2 despite an optimized heart rate, rhythm, preload, afterload, and without evidence of tamponade (table 2). (See "Use of vasopressors and inotropes".)

Epinephrine is an effective inotrope following cardiac surgery, even though is rarely used as an inotrope in other settings. Epinephrine produces consistent increases in the cardiac output with variable effects on arterial blood pressure following cardiac surgery [24,25]. Dopamine and dobutamine also increase cardiac output and heart rate, and appear to have efficacies similar to epinephrine [24]. Dobutamine produces greater reduction in left ventricular preload than dopamine and it augments coronary blood flow. However, the importance of the latter phenomenon is unclear, because surgically revascularized patients rarely exhibit ischemia in the absence of mechanical compromise of the coronary circulation.

Phosphodiesterase inhibitors such as inamrinone (formerly known as amrinone), milrinone, enoximone, and vesnarinone increase myocardial contractility, enhance myocardial relaxation, improve coronary blood flow, and reduce the systemic vascular resistance. The effect is to increase the cardiac index and decrease left ventricular preload and afterload, with minimal change in myocardial oxygen demand. Such agents are probably beneficial when used briefly following cardiac surgery, despite concerns that they may increase mortality when used in chronic heart failure [26,27].Inamrinone is no longer available in North America and most/all other countries so the only intravenous phosphodiesterase-3 inhibitor used in the US is milrinone.

Calcium sensitizing agents have not proven to be beneficial, compared with placebo in the intraoperative or postoperative setting [28,29]. (See "Inotropic agents in heart failure with reduced ejection fraction".)

Patients with profound ventricular dysfunction who are unable to wean from cardiopulmonary bypass despite the use of inotropic drugs often require the addition of mechanical assist devices, such as an intraaortic balloon pump or ventricular assist device, until native ventricular function recovers from the stresses of surgery and cardiopulmonary bypass. These devices reduce ventricular wall stress and augment coronary and systemic perfusion. (See "Intraaortic balloon pump counterpulsation" and "Short-term mechanical circulatory assist devices".)

Rarely, patients may need temporary support with venoarterial extracorporeal membrane oxygenation. (See "Extracorporeal life support in adults: Management of venoarterial extracorporeal membrane oxygenation (V-A ECMO)", section on 'Refractory cardiogenic shock'.)

Dysrhythmias — The trauma of cardiac surgery predisposes patients to atrial and ventricular arrhythmias:

●Atrial fibrillation – Atrial fibrillation can disturb normal atrioventricular synchrony and result in a 15 to 25 percent reduction in cardiac output [30]. Initial management involves slowing the ventricular heart rate using negative chronotropic agents. (See "Control of ventricular rate in patients with atrial fibrillation who do not have heart failure: Pharmacologic therapy".) The optimal heart rate usually occurs at rates between 80 and 100 beats per minute [31]. Once the ventricular heart rate is controlled, restoration of sinus rhythm with electrical or pharmacologic cardioversion may be considered. (See "Atrial fibrillation: Cardioversion".) Postoperative prophylactic therapy with beta blockers or amiodarone can help prevent postoperative atrial fibrillation and may have a role in the management of some patients. (See "Atrial fibrillation and flutter after cardiac surgery", section on 'Prevention of atrial fibrillation and complications'.)

Patients who develop atrial fibrillation after surgery are at increased risk of stroke, however anticoagulation in this patient population for stroke prevention has not been well studied. (See "Atrial fibrillation and flutter after cardiac surgery", section on 'Anticoagulation'.)

●Ventricular arrhythmias – The hemodynamic instability of patients with ventricular arrhythmias is variable and depends upon the rate of the tachyarrhythmia and left ventricular systolic and diastolic function. Sustained ventricular tachyarrhythmias should be promptly converted chemically or electrically. (See "Cardioversion for specific arrhythmias".)

●Bradyarrhythmias – Bradyarrhythmias are particularly common after valve surgery and are probably a consequence of direct surgical injury and local edema. If the bradycardia is symptomatic, temporary electrical pacing may be required. In some cases, permanent pacing may be necessary. (See "Permanent cardiac pacing: Overview of devices and indications".)

Myocardial infarction — Perioperative myocardial infarction (MI), defined as new Q waves on the postoperative electrocardiogram, occurs in 4 to 5 percent of patients undergoing coronary artery bypass grafting [32,33]. It is usually due to poor distal perfusion after grafting of the more proximal coronary arteries. The incidence of perioperative MI after other types of cardiac surgery is uncertain. It is important for clinicians to realize that the diagnostic accuracy of elevations in the serum creatine kinase (CK), CK-MB, and troponin is reduced after cardiac surgery because the enzymes are released as a routine sequela of the procedure. The diagnosis and management of perioperative MI are reviewed separately. (See "Early cardiac complications of coronary artery bypass graft surgery", section on 'Perioperative MI'.)

VASODILATORY SHOCK — Cardiopulmonary bypass can be complicated by vasodilatory (distributive) shock, which is a consequence of severely decreased systemic vascular resistance [34]. The cardiac output is typically increased in an effort to compensate for the diminished systemic vascular resistance, but may be normal or low if there is coexisting left ventricular dysfunction. (See "Definition, classification, etiology, and pathophysiology of shock in adults", section on 'Distributive'.)

The incidence of vasodilatory shock has been estimated to be 5 to 8 percent following cardiopulmonary bypass [34,35], and even higher among patients who are undergoing insertion of a ventricular assist device for end-stage heart failure (42 percent) [36] or who have a left ventricular ejection fraction <35 percent (27 percent) [34]. Risk factors include a reduced ejection fraction, prolonged aortic cross-clamp times, male gender, and preoperative therapy with angiotensin converting enzyme inhibitors [34,37]. Patients who develop vasodilatory shock following cardiac surgery have an increased likelihood postoperative bleeding, renal and liver injury, neurologic dysfunction, and respiratory failure [38,39].

Most patients with vasodilatory shock respond to hemodynamic-guided intravenous fluid therapy and/or intravenous norepinephrine, often at low doses [35,40,41]. However, the selection of agent should be individualized (table 2) with one small randomized trial suggesting that vasopressin may be a reasonable alternative to norepinephrine [41]. There are limited data about the outcome of patients who develop vasodilatory shock after cardiopulmonary bypass. Small studies have reported mortality rates between 5 and 15 percent [34,35,40,41]. (See "Use of vasopressors and inotropes".)

The pathogenesis of vasodilatory shock following cardiac surgery is uncertain. One hypothesis is that vasodilation is a consequence of a systemic inflammatory response to ischemia, reperfusion, surgical trauma, endotoxin release, or blood contact with the bypass circuit [38,39]. Alternative proposals include a relative deficiency of arginine vasopressin (AVP) or increased production of nitric oxide (NO):

●Patients with post-bypass vasodilatory shock have inappropriately low serum AVP concentrations [34,42]. Moreover, in a retrospective cohort study of 40 patients with norepinephrine-resistant hypotension following cardiac surgery, treatment with AVP at doses ranging from 0.01 to 0.10 U/min was associated with an increase in the mean arterial pressure from 58 to 82 mmHg, as well as an increase in the systemic vascular resistance from 771 to 1192 dyne-sec/cm5 [34].

●Production of NO is increased in many types of vasodilatory shock, including post-bypass vasodilatory shock [42,43]. Because NO causes vasodilation via activation of soluble guanylate cyclase, methylene blue was studied as a potential therapeutic agent (methylene blue is an inhibitor of guanylate cyclase) [35,44]. A trial randomly assigned 56 patients with vasodilatory shock despite norepinephrine therapy following cardiopulmonary bypass to receive methylene blue (a single dose of 1.5 mg/kg infused over one hour) or placebo [44]. In the methylene blue group, the mean duration of vasodilatory shock was significantly shorter (less than two hours versus greater than 24 hours) and mortality was significantly lower (0 versus 21 percent). The patients who died during the trial all had positive blood cultures.

HEMATOLOGIC DYSFUNCTION — Patients who undergo cardiac surgery are at increased risk for both bleeding and thrombosis.

Bleeding — Postoperative bleeding is common, with severe bleeding (requiring transfusion of >10 units of packed red blood cells) occurring in 3 to 5 percent of patients who have undergone cardiopulmonary bypass [45]. Such extensive bleeding is usually due to one or more of the following factors: incomplete surgical hemostasis, residual heparin effect after cardiopulmonary bypass, clotting factor depletion (eg, hypofibrinogenemia), hypothermia, postoperative hypotension, hemodilution (dilutional thrombocytopenia and coagulopathy), or platelet abnormalities (platelet dysfunction and thrombocytopenia).

The association of cardiopulmonary bypass with both thrombocytopenia and platelet dysfunction was illustrated by an observational study of 60 patients undergoing coronary artery bypass graft surgery (CABG) [46]. The study compared patients who underwent CABG using cardiopulmonary bypass to those who underwent off-pump surgery. It found that those who underwent cardiopulmonary bypass had more severe thrombocytopenia and abnormal platelet activation postoperatively. This may explain why surgery using cardiopulmonary bypass is associated with a greater risk of bleeding than off-pump surgery.

Postoperative bleeding frequently requires fresh frozen plasma, cryoprecipitate, fibrinogen, and platelets to correct the coagulation abnormalities. Transfusion of packed red blood cells may also be necessary to replace blood loss. Extensive bleeding can be mitigated in some cases by the administration of the antifibrinolytic agent, epsilon-aminocaproic acid [47,48]. Transfusion thresholds for treating anemia from blood loss and strategies to minimize transfusions in the perioperative cardiac surgery population are discussed separately. (See "Use of blood products in the critically ill" and "Perioperative blood management: Strategies to minimize transfusions" and "Indications and hemoglobin thresholds for RBC transfusion in adults", section on 'Cardiac surgery'.)

Thrombosis — Patients who undergo cardiopulmonary bypass appear to be at increased risk of developing thrombosis, probably due to increased platelet activity. This was demonstrated by a trial that randomly assigned 80 patients to undergo CABG with or without cardiopulmonary bypass [49]. Platelet activity during the early postoperative period was higher among patients who underwent cardiopulmonary bypass than among those who did not. This platelet activity may be resistant to aspirin, as suggested by another study that reported that the antiplatelet effect of aspirin was impaired after CABG with cardiopulmonary bypass, but not after CABG without cardiopulmonary bypass [50].

It has been hypothesized that the aspirin resistance noted in these studies may be due to the increased platelet turnover that follows cardiopulmonary bypass. Data looking at the long-term outcome of patients who develop aspirin resistance after cardiopulmonary bypass are lacking. However, there is evidence that the platelets of patients who develop graft thrombosis are more likely to be resistant to aspirin than patients without thrombotic events [51]. Different aspirin dosing regimens have been examined, with a small study suggesting that higher dosing of aspirin at 325 mg daily, and more frequent dosing of aspirin at 162 mg twice daily may lower risk of clotting [52], but further clinical correlation is needed. Dual antiplatelet therapy has been compared to aspirin therapy post CABG with a meta-analysis suggesting lower rates of cardiovascular mortality in dual antiplatelet therapy, but this advantage did not hold in the subanalysis of the randomized controlled data. Dual antiplatelet therapy was associated with significantly higher major bleeding rates [53]. Further pharmacological approaches to improve antithrombotic therapy in patients who develop aspirin resistance after CABG are still being explored.

Individuals with thrombosis and thrombocytopenia may require testing for heparin-induced thrombocytopenia (HIT), depending on the pretest probability of HIT. This subject is discussed separately. (See "Clinical presentation and diagnosis of heparin-induced thrombocytopenia", section on 'Diagnosis in special populations'.)

Management of graft thrombosis in the postoperative cardiac surgery patient is discussed separately. (See "Early cardiac complications of coronary artery bypass graft surgery", section on 'Early graft occlusion'.)

PULMONARY DYSFUNCTION — Pulmonary dysfunction is a significant cause of morbidity following cardiac surgery. Common types of pulmonary dysfunction include the following:

●Pleural effusion – Pleural effusions are common postoperative findings in patients who undergo various cardiac surgical procedures [54]. Most are a consequence of the surgery itself (nonspecific pleural effusions) and follow a benign course. However, pleural effusions may also occur with post-cardiac injury syndrome (PCIS) or may be the initial manifestation of a potentially serious complicating event. Causes of pleural effusion after cardiac surgery are listed in the table (table 3). (See "Evaluation and management of pleural effusions following cardiac surgery".)

●Pneumonia – Sternotomy and thoracotomy incisions produce pain, which impairs the ability to cough and breathe deeply. This increases the risk for pneumonia. The incidence of pneumonia following cardiac surgery is estimated to be 1 to 5 percent and it may be associated with increased mortality [55-57]. Risk factors include: chronic obstructive pulmonary disease, smoking, and older age [55,58]. The prevention and treatment of hospital-acquired pneumonia and assessment of the risk of postoperative pneumonia are discussed in detail separately. (See "Risk factors and prevention of hospital-acquired and ventilator-associated pneumonia in adults" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults".)

●Atelectasis – Atelectasis occurs in up to 70 percent of patients following cardiac surgery, usually as a result of single lung ventilation and intentional lung collapse during the surgery [59,60]. Management consists of incentive spirometry and early physical mobilization. (See "Atelectasis: Types and pathogenesis in adults".)

●Decreased thoracic compliance – Chest wall and lung compliance decrease postoperatively. This effect peaks approximately three days after surgery and may complicate the extubation of patients, especially those with underlying lung disease. (See "Weaning from mechanical ventilation: Readiness testing" and "Initial weaning strategy in mechanically ventilated adults".)

●Difficulty weaning – Some patients prove difficult to wean from mechanical ventilation following cardiac surgery. The prognosis of such patients varies widely across series, but is not uniformly poor. As an example, a series of 124 patients who received greater than seven days of mechanical ventilation following cardiac surgery found that 85 percent of the patients survived until discharge and 99 percent of the survivors were successfully weaned from mechanical ventilation [61].

●Diaphragmatic dysfunction – Phrenic nerve injury during surgery is rare, but may cause diaphragmatic dysfunction or paralysis if it occurs. (See "Diagnosis and management of nontraumatic unilateral diaphragmatic paralysis (complete or partial) in adults" and "Treatment of bilateral diaphragmatic paralysis in adults".)

●Acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) – ALI and ARDS are types of hypoxemic respiratory failure characterized by acute onset, bilateral infiltrates, a decreased ratio of arterial oxygen tension to fraction of inspired oxygen, and lack of evidence of elevated left atrial pressure. They differ only in the severity of hypoxemia, with ARDS being more severe than ALI. ALI and ARDS complicate less than two percent of cardiac surgeries that used cardiopulmonary bypass [62]. (See "Acute respiratory distress syndrome: Clinical features, diagnosis, and complications in adults".)

Patients undergoing cardiac surgery for structural or congenital heart disease frequently have pulmonary hypertension, which can complicate postoperative management. As an example, pulmonary complications (eg, pneumonia, atelectasis) may cause hypoxic vasoconstriction, worsened pulmonary hypertension, and hypoxemia. The result may be hypoxemia that seems out of proportion to the severity of the pulmonary complication. Inhaled pulmonary vasodilators, including nitric oxide (NO) and epoprostenol causes selective pulmonary vasodilation and, therefore, can mitigate the exacerbation of pulmonary hypertension and stabilize hemodynamically compromised patients with severe pulmonary hypertension. These cannot be used long-term but may be useful while the acute process that led to the deterioration is reversed. Inhaled pulmonary vasodilators require a special device to administer. Complications of use of inhaled nitric oxide include rebound pulmonary hypertension during weaning. Oral sildenafil may facilitate inhaled NO withdrawal by preventing rebound pulmonary hypertension in patients who have failed previous attempts to wean the inhaled NO [63,64]. When compared head to head in cardiac surgery patients there is no significant difference in outcomes between the use of NO and epoprostenol [65]. (See "Inhaled nitric oxide in adults: Biology and indications for use" and "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)" and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

NEUROLOGIC DYSFUNCTION — The incidence of postoperative neurologic sequelae after coronary artery bypass graft surgery (CABG) is approximately 2 to 6 percent, with the frequency increasing among older patients [66]. A prospective cohort study followed 2108 patients undergoing CABG for postoperative neurologic complications [67]. The complications were placed into two categories: type I neurological complications were defined as focal injury (eg, stroke), stupor, or coma at discharge, while type II neurological complications were defined as deterioration in intellectual function, memory deficits, or seizures. The incidence of all adverse cerebral events was six percent. Patients with cerebral complications had higher in-hospital mortality, longer hospitalizations, and a higher rate of requiring discharge to a chronic care facility than did those without neurologic sequelae. Type I complications occurred in three percent, primarily consisting of nonfatal strokes. Type II complications also occurred in three percent and consisted primarily of deterioration in intellectual function (figure 1).

Early recognition of neurological complications is important. Although treatment is largely supportive, prompt initiation of therapy may prevent worsening of the complication. (See "Initial assessment and management of acute stroke" and "Diagnosis of delirium and confusional states" and "Delirium and acute confusional states: Prevention, treatment, and prognosis".)

Neurological complications following cardiac surgery may be a result of a variety of processes. These include atheroembolism of aortic debris, embolism of left atrial or left ventricular thrombus, cerebral hypoperfusion, air embolism, and microembolism of granulocyte aggregates, fibrin, and platelets [68].

RENAL DYSFUNCTION — Acute renal failure occurs in up to 30 percent of patients who have undergone cardiac surgery, when defined as a 50 percent increase in the serum creatinine concentration above baseline [69]. It is severe enough to require dialysis in 1 to 5 percent of patients and it appears to be associated with increased mortality. In a prospective cohort study of 43,642 patients who had undergone coronary artery bypass graft (CABG) or valve surgery, acute renal failure was associated with a 30-day mortality of 64 percent, compared to four percent without renal failure [70]. More recent data suggest the incidence of renal failure following cardiac surgery has increased, while the associated mortality has decreased [71]. This probably reflects an increase in comorbid disease and improvements in postoperative care.

●Risk factors – Postoperative risk factors for acute renal failure include poor cardiac performance and perioperative hemodynamic instability [60,70]. Risk factors that cannot be controlled postoperatively include advanced atherosclerotic vascular disease, reduced creatinine clearance, a long duration of cardiopulmonary bypass, and the use of radiocontrast agents immediately before surgery.

●Mechanism – Mechanisms of perioperative renal failure include renal artery vasoconstriction, hypothermia, loss of pulsatile flow during cardiopulmonary bypass, and atheroembolic disease.

●Prevention – The best preventive strategy is to optimize renal perfusion (ie, avoid hypotension and hypovolemia) and to avoid potentially nephrotoxic agents (eg, aminoglycoside antibiotics, angiotensin converting enzyme inhibitors, and radiologic contrast agents) in the immediate postoperative period. There is no clear evidence supporting the efficacy of pharmacologic therapy (eg, low-dose dopamine, loop diuretics) to prevent acute renal failure after major surgery. In addition to lack of proven efficacy, there is some concern about toxicity (eg, arrhythmias, myocardial ischemia, and intestinal ischemia with dopamine). (See "Possible prevention and therapy of ischemic acute tubular necrosis".)

●Treatment – There is no convincing evidence of benefit from early and/or aggressive dialysis, and there is some concern that renal function might be impaired by this approach [72]. Thus, the decision to perform dialysis generally should be based upon the presence of uremic symptoms, fluid overload, or electrolyte abnormalities, rather than a specific blood urea nitrogen or serum creatinine concentration. (See "Dialysis-related factors that may influence recovery of kidney function in acute kidney injury (acute renal failure)".)

A prediction model has been proposed using basic metabolic panel laboratory values data (including changes in serum creatinine from preoperative values, serum potassium, bicarbonate, sodium, albumin, and blood urea nitrogen, adjusted for the time between conclusion of the surgical procedure to the first postoperative blood draw) from a retrospective observational cohort of over 58,000 adult patients who underwent cardiac surgery [73]. The derivation model had excellent predictive discrimination for moderate to severe acute kidney injury within 72 hours after surgery (area under the curve [AUC] 0.876, 95% CI 0.869-0.883) and 14 days (AUC 0.854, 95% CI 0.850-0.861) after surgery. The model also performed well in the validation cohort of nearly 5000 patients with an AUC of 0.860 (95% CI 0.838-0.882) within 72 hours and 0.842 (95% CI 0.820-0.865) within 14 days following cardiac surgery. Further data are need to determine whether such a model improves clinical outcomes.

MORTALITY — The perioperative and in-hospital mortality rate after cardiac surgery ranges from 1 to 5 percent [74]. The type of cardiac surgery, preoperative frailty (eg, poor mobility, disability, and nutrition), and the presence of comorbid disease influence the mortality rate [75]. This topic is discussed in detail separately. (See "Operative mortality after coronary artery bypass graft surgery".)

SUMMARY AND RECOMMENDATIONS

●Routine monitoring following cardiac surgery typically includes continuous telemetry, measurement of the arterial blood pressure via an arterial catheter, measurement of the cardiac filling pressures via a pulmonary artery catheter (ie, right heart catheter, Swan Ganz catheter), continuous assessment of the arterial oxygen saturation via pulse oximetry, and continuous measurement of the mixed venous oxygen saturation via an oximetric pulmonary artery catheter. Such monitoring allows instantaneous assessment of cardiopulmonary physiology. (See 'Monitoring' above.)

●Postoperative cardiac dysfunction is usually suspected when there is unexplained postoperative hypotension, tachycardia, or pulmonary edema. Evaluation consists of reviewing the patient's telemetry, bedside echocardiography, invasive hemodynamic assessment via a pulmonary artery catheter, and a 12-lead electrocardiogram. These four tests are likely to identify the cause of cardiac dysfunction. A chest radiograph may be helpful if these tests do not find the cause of cardiac dysfunction. The most common causes of cardiac dysfunction following cardiac surgery are mechanical complications, physiologic complications (inadequate preload, excessive afterload, and poor ventricular inotropy), dysrhythmias, and myocardial infarction. (See 'Cardiac dysfunction' above.)

●Cardiac surgery using cardiopulmonary bypass can be complicated by vasodilatory (distributive) shock. Most patients with vasodilatory shock respond to intravenous norepinephrine, often at low doses. The pathogenesis of vasodilatory shock following cardiac surgery is uncertain. (See 'Vasodilatory shock' above.)

●Patients who undergo cardiac surgery are at increased risk for both bleeding and thrombosis.

•Bleeding is usually due to incomplete surgical hemostasis, residual heparin effect after cardiopulmonary bypass, clotting factor depletion, hypothermia, postoperative hypotension, hemodilution (dilutional thrombocytopenia and coagulopathy), and/or platelet abnormalities (platelet dysfunction and thrombocytopenia). Postoperative bleeding frequently requires fresh frozen plasma and platelets to correct the coagulation abnormalities. Transfusion of packed red blood cells may also be necessary to replace blood loss. (See 'Bleeding' above.)

•The increased risk of developing thrombosis is probably due to increased platelet activity caused by cardiopulmonary bypass. This platelet activity may be resistant to aspirin. Pharmacological approaches to improve antithrombotic therapy in patients to develop aspirin resistance after cardiopulmonary bypass are being explored. (See 'Thrombosis' above.)

●Pulmonary dysfunction is a significant cause of morbidity following cardiac surgery. Common types of pulmonary dysfunction include pleural effusion, pneumonia, atelectasis, decreased thoracic compliance, difficulty weaning from mechanical ventilation, diaphragmatic dysfunction, acute lung injury, and acute respiratory distress syndrome. (See 'Pulmonary dysfunction' above.)

●Neurological complications may follow cardiac surgery. Patients with cerebral complications have higher in-hospital mortality, longer hospitalizations, and a higher rate of requiring discharge to a chronic care facility than those without neurologic sequelae. Early recognition of neurological complications is important. Although treatment is largely supportive, prompt initiation of therapy may prevent worsening of the complication. (See 'Neurologic dysfunction' above.)

●Acute renal failure occurs in up to 30 percent of patients who have undergone cardiac surgery and it appears to be associated with increased mortality. The best preventive strategy is to optimize renal perfusion (ie, avoid hypotension and hypovolemia) and to avoid potentially nephrotoxic agents (eg, aminoglycoside antibiotics, angiotensin converting enzyme inhibitors, and radiologic contrast agents) in the immediate postoperative period. There is no convincing evidence of benefit from early and/or aggressive dialysis. (See 'Renal dysfunction' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Frank E Silvestry, MD, who contributed to earlier versions of this topic review.