Chapter 77. Developing an Anesthetic Plan for a Critically Ill Patient

1. Critically ill patients often need surgery to correct the underlying cause of their illness or to deal with the complications of their illness.

2. Advanced planning and open communication between the anesthesiologists, surgeons, the critical care team, and the patient and the patient's family is crucial to understanding the goals and priorities of treatment.

3. Critically ill patients may have impaired function of multiple organ systems. Preoperative evaluation of the degree of organ dysfunction and optimization of the patient's condition ensures that the patient is in the best possible condition to undergo the additional stresses associated with surgery and anesthesia.

4. The simplest surgical procedure resulting in the least physiologic upset is generally the best option for the critically ill patient.

5. Planning for transport to the operating room is essential because patients are at high risk for adverse events during transport.

6. The anesthesiologist must decide which monitors are needed to assess the assessment of the patient's condition.

7. Although general anesthesia is most often necessary for surgery in the critically ill patient, regional anesthesia can play a valuable role helping achieve patient comfort and reduce physiologic stress.

8. There should be specific goals and end points defined in the management of the critically ill patient to optimize hemodynamics and minimize further end-organ damage.

More than 5 million patients are admitted annually to intensive care units (ICUs) across the United States. Many of these patients undergo a surgical procedure during their hospitalization either to correct the underlying cause of their illness or to deal with its complications. Similar to the healthy patient, the anesthesia plan for the critically ill patient should include a clear delineation of the goals of management, an assessment of the priorities of care, and a consideration of alternative strategies to avoid or treat complications.

Many studies indicate that the patient's outcome depends on the interaction of several factors, including the type and extent of the procedure, the physiologic reserves of the patient, the presence of chronic health problems, and the nature of the acute physiologic derangements.1 The number of complications attributable to anesthesia is 8 times greater for patients with American Society of Anesthesiologists (ASA) Physical Status grades P3 and P4 rather than for patients with ASA grade P1 or P2.2 One study reports that the presence of an anesthesiologist intraoperatively and certain characteristics of intraoperative and postoperative care were associated with a decreased risk of severe postoperative morbidity and mortality.2 Thus the plan for a critically ill patient should include preoperative optimization of the patient's condition, a chronological plan for intraoperative management, a plan to ensure that adequate support personnel and equipment are available, and a plan for postoperative care.

The critically ill patient often has many caregivers from various medical and surgical specialties. There should be open communication regarding the likely outcomes and realistic goals of treatment between the anesthesiologists, surgeons, critical care team, and the patient and the patient's family. Preoperatively, one person or team should be designated as the coordinator of the patient's care and should ensure that everyone taking care of the patient in the perioperative period understands the goals and priorities of treatment. Because these goals often change with time, frequent communication to update all is vital.

This chapter focuses on the key points for preoperative planning and some considerations for intraoperative management and optimization to ensure safe delivery of anesthetic care to the critically ill patient.

Critically ill patients may have impaired function of some or all their organ systems. An assessment of the degree of organ dysfunction and, whenever possible, optimization of the patient's condition should be undertaken before surgery to ensure that the patient is in the best possible condition to undergo the additional stresses associated with surgery and anesthesia.

Hemodynamic Considerations

A variety of cardiovascular abnormalities may be present in critically ill patients, including hemodynamic instability and dysrhythmias. Patients may manifest shock secondary to hypovolemia, vasodilatation, and/or myocardial dysfunction. Because shock represents the failure of the circulatory system to maintain adequate delivery of oxygen and other nutrients to tissues, cellular and organ dysfunction often ensues. Thus the ultimate goals of hemodynamic therapy in shock are to maintain or restore adequate tissue perfusion and organ function. Evaluation of the hemodynamic status of the critically ill patient should include the following: assessing the patient's volume status, determining whether pressors or inotropic agents are needed to maintain an adequate blood pressure and cardiac output (CO), and learning whether there are any clinical or laboratory indicators of hypoperfusion.

Respiratory Failure and Ventilator Management

Critically ill patients often have respiratory failure and require mechanical ventilation. A review of the ventilator's settings preoperatively should include the mode of ventilation, the ventilator rate, the fraction of inspired oxygen, the level of positive end-expiratory pressure (PEEP), and the tidal volume and airway pressures. Arterial blood gas values should be reviewed for adequacy of oxygenation and ventilation. Arrangements should be made to transport the ICU ventilator to the operating room if the patient's ventilation cannot be supported by the similar operating room ventilator.

Neurologic Assessment

The critically ill patient often has compromised neurologic function due to a myriad of causes including drugs, infection, metabolic derangements, trauma, and cerebrovascular accidents. Their preoperative evaluation should include an assessment of baseline neurologic function. Dosages and rates of infusions of sedative hypnotic and analgesic agents should be noted, and decisions should be made whether or not infusions of these agents should be continued intraoperatively.

Neuromuscular blocking agents are associated with critical illness polymyopathy/neuropathy, and in general they should be avoided in the septic patient.3,4 However, because neuromuscular blocking agents are often needed to facilitate surgery, monitoring the depth of blockade with a train-of-four monitor is advised.

Critical illness is associated with prolonged immobilization and a severe myopathy. Thus the use of succinylcholine in critically ill patients can precipitate severe hyperkalemia and/or rhabdomyolysis and should be avoided.

Evaluation of Renal Function

Critically ill patients often have renal dysfunction. Evaluation of the patient with renal failure should include an assessment of volume status, electrolyte balance, and acid-base status. Because renal ischemia is the primary cause of acute renal failure in the perioperative setting, intraoperative maintenance of adequate intravascular volume, mean arterial pressure, and CO are important measures to preserve renal perfusion. Furthermore, nephrotoxic drugs should be avoided to prevent further renal injury in these patients. Although the measurement of urine output is usually the primary means of evaluating renal function, invasive monitoring may be required to assure adequate volume repletion in the critically ill patient who is at risk for postoperative renal dysfunction.

Endocrine Abnormalities in the Critically Ill Patient

Recent clinical trials have examined 2 particular endocrine abnormalities in critically ill patients: adrenal insufficiency and glycemic control (see Chapter 13).

Adrenal Insufficiency

Patients with septic shock may have relative adrenal insufficiency. Thus intravenous (IV) corticosteroids may be administered to patients with septic shock who are poorly responsive to fluid replacement and vasopressor therapy. The dose of corticosteroids should equivalent to less than 300 mg hydrocortisone daily,5

Glucose Control

Critically ill patients are frequently hyperglycemic. Whereas a previous study showed that "intensive glucose control" (maintaining plasma glucose concentration between 80 and 100 mg/dL) was associated with decreased ICU mortality,6 a more recent study found that maintaining plasma glucose concentrations less than 180 mg/dL resulted in lower mortality and less risk of severe hypoglycemia.7 Because ICU patients often receive insulin infusions to achieve these goals, consideration should be given to continuing insulin infusions intraoperatively. Patients receiving insulin infusions must be given an external source of glucose to prevent hypoglycemia, and blood sugar levels should be checked frequently intraoperatively.

Coagulation Dysfunction and Anemia

Evaluation of the critically ill patient should include an assessment of their circulating hemoglobin levels and coagulation status.

Anemia is often present in patients with critical illness. Although current recommendations for critically ill patients state that packed red cell transfusions should be given only when the hemoglobin level decreases to 7.0 mg/dL or less,8,9 these recommendations should not be applied to individuals who are undergoing short-term resuscitation. In the operating room, the decision to transfuse must be based on how the patient responds to other interventions, including early, aggressive fluid resuscitation. In patients with evidence of hypoperfusion (eg, a low central venous oxygen saturation or lactic acidosis), a target hemoglobin of 10 g/dL has been suggested to maximize tissue oxygen delivery.10

Thrombocytopenia can occur in critically ill patients and may be a result of sepsis, drug therapy (Table 77-1), or hemodilution caused by massive transfusion. Platelet transfusions are given to patients who are bleeding or at risk of bleeding as a consequence of thrombocytopenia or impaired platelet function. Surgical bleeding may occur when platelet counts are less than 50,000/μL. Additional factors such as a coagulopathy, fever, and renal failure may increase the bleeding risk from severe thrombocytopenia. Therefore, the threshold platelet concentrations for triggering a prophylactic platelet transfusion must take all these considerations into account.8,11 Platelets should never be given to patients with type 2 heparin-induced thrombocytopenia because of the increased risk of thrombosis after platelet transfusions.

Coagulopathies can occur in the critically ill patient and are initially evaluated by measuring the prothrombin time, activated partial thromboplastin time, and fibrinogen levels. Fresh-frozen plasma or cryoprecipitate are infused to stop active bleeding associated with a coagulopathy or to correct a coagulopathy before surgical procedures. Vitamin K is administered to raise the levels of vitamin K–dependent clotting factors (II, VII, IX, and X).

Recombinant human coagulation factor VIIa (rFVIIa) is used to treat bleeding and promotes hemostasis by activating the extrinsic pathway of the coagulation cascade (Table 77-2). rFVIIa bypasses inhibitors of factors VIII and IX in patients with hemophilia A and B, and treats patients who have severe von Willebrand factor (vWF) deficiency caused by antibodies to vWF. Other uses of rFVIIa include the treatment of congenital or acquired factor VII deficiency, congenital factor XI or V deficiency, the coagulopathy of severe liver dysfunction, hemostatic changes that arise from extensive surgery, trauma, and bleeding, reversal of warfarin-induced excessive anticoagulation, certain inherited disorders of platelet function (eg, Glanzmann and Bernard-Soulier thrombasthenias), and bleeding from thrombocytopenia that is a result of antiplatelet glycoprotein antibodies that thwart the effects of platelet transfusion.12 The minimum effective rFVIIa dose for managing these hemorrhagic disorders is uncertain. The rFVIIa dose depends on the specific hemorrhagic disorder being treated; in various clinical studies the dose has ranged from 3 to 320 μg/kg.

Recombinant activated protein C (drotrecogin alfa) is a critical protein that reduces excessive thrombosis in the microcirculation and appears to reduce the mortality from severe sepsis associated with organ dysfunction in adult patients who are at high risk of dying (Acute Physiology and Chronic Health Evaluation II scores ≥25).13,14 Because of its anticoagulant effects, bleeding is the most common serious adverse effect associated with drotrecogin alfa therapy. Drotrecogin alfa should be discontinued 2 hours prior to undergoing an invasive surgical procedure or performing procedures with an inherent risk of bleeding. Once adequate hemostasis has been achieved, initiation of drotrecogin alfa may be considered 12 hours after major invasive procedures or surgery, or therapy may be restarted immediately following uncomplicated minor procedures.

The Elderly Critically Ill Patient

Advancing age is associated with an increased incidence of intercurrent diseases and declining physiologic reserve. Before the age of 60 years, both basal organ function and physiologic reserve (the difference between basal and maximal organ function) are generally well maintained. Subsequently, physiologic reserve diminishes. Both aging and a higher American Society of Anesthesiologists (ASA) Physical Status classification score are associated with an increased incidence of postoperative complications. In the healthiest patients, for example, those who are classified as ASA class P1 (free of systemic disease) and P2 (mild systemic disease), the frequency of major complications increases gradually with age until about 70 years of age. For ASA class P3 patients, an increasing incidence of major postoperative complications begins at an earlier age and increases rapidly with age. For ASA class P4 patients (a patient with disease that is a constant threat to the patient's life), the complication rate rises even more steeply. Many studies show that age and comorbidity are independent predictors of mortality in both the general surgical population and in the critically ill patient.15-17

Mortality and complication rates have been well-studied in the elderly (age ≥65 years) general surgical population.18-20 Characteristics of the elderly population include a reduced functional status, more emergency operations, a higher ASA classification (20% of patients ≥80 years of age were ASA P4), and more frequent do-not-resuscitate orders. Mortality rates vary widely across many types of operations but are higher for those 80 years and older. Postoperative complications were also more common in patients 80 years and older (20% had one or more complications), and mortality was higher in those who suffered complications postoperatively. One study reported a 5% increase in perioperative mortality risk for every year after age 80.19

Morbidity and mortality rates in elderly patients admitted to the ICU are higher than in younger patients. Compared with younger patients, elderly patients are more severely ill on admission, and they are more likely to have shock and renal dysfunction. One study reported that hospital mortality was more than doubled for patients 75 years and older in comparison with patients younger than 65 years.21-24

Critically ill patients undergo a myriad of surgical procedures, including a tracheostomy for respiratory failure, surgery to remove a septic focus, or procedures for traumatic injuries. Whether definitive surgery should be undertaken or whether it should be delayed and performed electively when the patient has recovered depends on the stability of the patient and the nature and complexity of the intervention that is needed. In general, the simplest intervention resulting in the least physiologic upset is the best option for the critically ill patient. The decision to proceed to surgery and the extent of the surgical procedure must weigh the benefits and risks of the specific intervention to cause complications, such as bleeding or inadvertent injury to other organs or tissues. Ideally, surgery should be undertaken following adequate resuscitation. However, timely and emergent intervention may be vital and lifesaving in certain conditions, for example, patients with necrotizing soft tissue infection or intestinal ischemia.

Transporting the Critically Ill Patient

Unless the surgical procedure will occur in the ICU, the critically ill patient must be transported from the ICU to the operating room. These patients are at high risk for complications en route. Adverse events during transport include both minor and serious mishaps such as the loss of IV access, accidental tracheal extubation, occlusion of the endotracheal tube, and exhaustion of the oxygen supply, as well as physiologic deterioration of the patient's state (eg, worsening hypotension or hypoxemia). Studies report adverse events rates in 5.9% to 66% of the transports of critically ill patients.25,26 One study reported a high incidence of hemodynamic changes requiring therapeutic intervention during and after transport from the operating room to the ICU.27 Guidelines for the transport of critically ill patients have been published.28 Having a plan for transport to minimize potential complications is an important area of patient safety (Table 77-3).

The patients should be hemodynamically stable before transport. All critically ill patients undergoing transport should receive the same level of basic physiologic monitoring during transport as they had in the ICU and should include, at a minimum, continuous electrocardiographic monitoring, continuous pulse oximetry, and periodic measurement of blood pressure, pulse rate, and respiratory rate. Selected patients may benefit from capnography, continuous intra-arterial blood pressure, or intracranial pressure monitoring.

The patient should have a secured airway to allow safe transport. Patients should be assessed for their need for intubation before transport. Intubated patients are usually taken off mechanical ventilation and switched to manual ventilation with a self-inflating Ambu bag during transport. Adequate PEEP should be applied during manual ventilation. The oxygen tank should be checked to ensure an adequate supply at the required flow rate for transport.

Before transport the anesthesiologist should check the size and placement of the patient's IV access. Patency of the IV routes should be assessed. All drug infusions (eg, sedatives, pressors, insulin), their rates, and their IV routes should be noted. There should be an adequate volume of drug to last during transport. If total parenteral nutrition is stopped, an alternative source of glucose must be provided. If a sedative and/or muscle relaxant is needed for transport, it should be administered in the ICU and the patient monitored for hemodynamic and respiratory side effects (eg, hypotension) before transport. The anesthesiologist should have common drugs needed for resuscitation, such as epinephrine and atropine, readily available in the event of cardiac arrest or arrhythmia. As well, a mask, laryngoscope, and extra endotracheal tubes should be available should an accidental extubation occur during transport. Standardized transport kits can ensure that the necessary drugs and equipment are always present and available (Fig. 77-1).

All infusion pumps should be securely attached to the bed or to specialized transfer devices. Battery-operated equipment should be fully charged and capable of functioning for the duration of the transport. There should be adequate assistance to transport the patient and treat minor problems. The operating room should be ready to receive the patient prior to leaving the ICU.

Standard Monitors

Basic standards for monitoring have been established by the ASA (Table 77-4) and should be used for the critically ill patient.29 The patient's systemic oxygenation (SpO2), ventilation (ETCO2), circulation (blood pressure and heart rate), and core temperature should be continuously monitored and evaluated during each anesthetic procedure.

An oxygen analyzer to measure the oxygen concentration in the patient breathing system and a quantitative assessment of systemic blood oxygenation (eg, SpO2 via pulse oximetry) should be used to monitor adequate oxygenation. Ventilation should also be evaluated by clinical signs (eg, chest or abdominal excursion and auscultation of breath sounds) as well as quantitative continuous monitoring of the level of expired carbon dioxide. Monitoring the tidal volume of expired gas is strongly encouraged. There should be a rapid audible alarm if any of the components of the breathing system are disconnected.

Each critically ill patient should have their electrocardiogram (ECG) continuously displayed, and blood pressure and heart rate should be determined and assessed every 5 minutes at a minimum. To ensure the adequacy of the patient's circulatory function, the patient should be continuously evaluated by either palpation of an arterial pulse (at a minimum), auscultation of heart sounds, observing the trace of intraarterial systemic pressure, ultrasound peripheral pulse velocity monitoring, pulse plethysmography, or oximetry.

Core temperature monitoring should be used when clinically significant changes of body temperature are likely.

Because of possibly rapid changes of hemodynamic status occurring during anesthesia, there must be qualified anesthesia personnel to monitor the patient and provide care throughout all general and/or regional anesthetics and during monitored anesthesia care.

Invasive Monitors

Additional monitors may be needed to provide an accurate picture of the patient's hemodynamic status and to guide therapy to optimize the patient's condition. However, there are limitations to each monitor and some risk to the patient when using invasive monitors. Thus the anesthesiologist must be cognizant of these limitations and weigh the risks and benefits of invasive monitoring to choose appropriate monitors for each critically ill patient (Table 77-5).

Invasive Arterial Pressure Monitoring

Because shock represents failure of the circulatory system to maintain adequate delivery of blood flow to tissues, and the goal of hemodynamic therapy is to restore adequate tissue perfusion, systemic pressure measurement is the most frequently used parameter to indirectly assess perfusion.

In healthy individuals, blood pressure as determined by noninvasive blood pressure (NIBP) monitors is, on average, within 5 mm Hg of measurements obtained by direct arterial pressure monitoring. Sources of error in NIBP measurements include using the incorrect cuff size or highly irregular or rapid cardiac rhythms. However, in shock states, NIBP measurements are often inaccurate. Therefore, inserting an arterial catheter provides a more accurate and reproducible measurement of systemic arterial pressure. Invasive arterial monitoring allows beat-to-beat display so decisions regarding therapy can be based on continuous blood pressure analysis.

Blood pressure does not directly equate to tissue blood flow, and the level of mean arterial pressure (MAP) to aim for is not necessarily the same in all patients. Below a MAP of 60 mm Hg, autoregulation of the coronary, renal, and central nervous system vascular beds is compromised and organ flow becomes linearly dependent on pressure. Thus in adults, maintenance of a MAP 65 mm Hg or higher is recommended to maintain and optimize flow.4 However, because the loss of autoregulation may occur at different levels in different organs, some patients require higher blood pressures to maintain adequate tissue perfusion. Furthermore, the degree to which flow autoregulation remains intact in septic patients is uncertain. Thus it may be necessary at times to supplement blood pressure measurement with other means of assessing regional and global perfusion (eg, urine output; see section on downstream monitoring).

The invasive arterial pressure waveform can also be used to estimate blood volume status and assess fluid responsiveness in mechanically ventilated patients. A positive pressure breath reduces venous return and decreases left ventricular filling resulting in a decreased stroke volume, CO, and blood pressure. This effect can be especially marked in hypovolemic patients. Marked variations in pulse pressure, which is proportional to stroke volume, and systolic pressure tracing variations usually predict that an increase in CO will occur with volume loading.30,31

Central Venous Pressure Catheter

Systemic hypotension is the most common reason to initiate invasive central hemodynamic monitoring in the critically ill patient. The central venous pressure (CVP) reflects pressure in the large systemic veins. Although the CVP also reflects intravascular volume, it does not measure blood volume directly and is influenced by right-heart function, venous return, right-heart compliance, intrathoracic pressure, and the patient's positioning. Consequently, the CVP level should be interpreted with other measures of cardiac function and circulating volume (eg, pulse, blood pressure, urine output). A single CVP value may not be as important as serial CVP measurements or their change with therapy.

Measurement of the CVP may provide useful information in patients with normal cardiac function who are hypotensive because of blood loss or widespread vasodilatation because a decreased venous return will result in a falling right atrial pressure and CVP. A central venous catheter (CVC) should be inserted when an infusion of vasopressors or inotropes is planned.

Pulmonary Artery Catheter

In addition to the CVP, a flow-directed pulmonary artery catheter (PAC) allows measurement of the pulmonary artery pressure, the pulmonary artery occlusion pressure (PAOP), the CO, and the mixed-venous oxygen saturation (SVO2). The PAOP reflects the pulmonary venous pressure and the left atrial and left ventricular end-diastolic pressure and therefore can provide a crude indirect estimate of left ventricular end-diastolic volume (LVEDV). Indications for placing a pulmonary artery catheter are related to the planned procedure and the patient's state of health (Table 77-6). The pulmonary artery catheter may provide useful information in any procedure associated with acute, severe changes of preload, afterload, or myocardial contractile state. For example, procedures where blood loss may be massive or where partial or complete caval occlusion might occur can cause acute changes of cardiac preload. Similarly, proximal aortic cross-clamping can cause a rapid increase of left ventricular afterload. Patient factors that can lead to the insertion of a PAC include septic shock or other significant cardiac, respiratory, or renal disease. In septic shock, both hypovolemia and myocardial dysfunction may contribute to impaired tissue perfusion. Pulmonary artery catheter placement can be used both to monitor CO and assess ventricular stroke volume.

Unjustified use of the pulmonary artery catheter has been suppressed by the publication of an observational study demonstrating an increased mortality associated with the use of the pulmonary artery catheter in critically ill patients during the first 24 hours of intensive care as compared with case-matched controls.32 Multiple randomized controlled trials evaluating the use of the PAC in many different patient populations have not demonstrated a decreased overall mortality or obvious benefit.33-35 In all of these studies, the PAC has proven primarily to be a diagnostic tool. The studies also highlight our lack of consensus and data about our therapeutic interventions that will improve the outcome of the critically ill patient in shock.

A major pitfall of central pressure monitoring occurs when pressure alone is used to estimate volume. Important variables in addition to the volume of the systemic and pulmonary circulations can affect the central pressure measurement. The relationship between pressure and volume is controlled by the compliance of the chamber. In patients with an abnormal left ventricular compliance, the PAOP may over- or underestimate the LVEDV. Right ventricular (RV) dysfunction is common in critically ill patients and can result from pulmonary embolism, severe acute respiratory distress syndrome (ARDS), or other conditions that increase RV afterload, such as high levels of PEEP or an increased pulmonary vascular resistance as a consequence of other vascular, cardiac, metabolic, or pulmonary causes. Because the right and left ventricles are both enclosed within the relatively stiff pericardium, pressure and volume overload of the RV can lead to abnormal motion of the interventricular septum and will impair left ventricular (LV) relaxation. In this situation, the pressure–volume relationship of the left ventricle is altered, and information obtained from the PAC may be misleading. Other factors that can affect the ability of central pressures to reflect ventricular filling volumes include increased intrathoracic pressures and valvular lesions. Increased intrathoracic pressure caused by positive pressure ventilation or increased intra-abdominal pressures can also elevate central vascular pressures. Stenotic lesions of the atrioventricular valves can also elevate central vascular pressures.

Other potential pitfalls include problems with the transducer system (eg, improper transducer placement or reference level zeroing), improper interpretation of waveforms, and erroneous thermodilution CO measurements due to errors of injectate volume or temperature. The presence of marked tricuspid valve regurgitation, which can occur in the critically ill patient as a consequence of high pulmonary artery pressures, can also produce erroneous CO measurements. Mixed venous blood gas tensions and saturation measurements may be invalid if the blood gas analysis is inaccurate or if the pulmonary artery specimen is "arterialized" by being withdrawn from a wedged or partially wedged PAC.

There are major but uncommon complications associated with the presence of a PAC. Rupture of the pulmonary artery is a rare occurrence (<1%) but a potentially catastrophic one with a mortality rate approaching 50%. Patients who have pulmonary hypertension, are older than 60 years, or are receiving anticoagulation therapy are at greater risk. The sudden onset of hemoptysis (especially after inflation of the pulmonary artery catheter balloon) is a sign of possible pulmonary artery rupture. Immediate management includes lateral decubitus positioning the bleeding side down, intubation with a double-lumen endotracheal tube, and increasing PEEP. Embolization via angiography, or even a lobectomy, may become necessary if pulmonary bleeding continues or is massive. The incidence of pulmonary infarction associated with the use of the pulmonary artery catheter is less than 7% and usually caused by unintentional distal migration of the PAC tip. Catheter-related thrombi may also cause pulmonary infarction. Infection related to the PAC is fairly common with a risk for clinical sepsis of less than 0.5% per day of catheter use.

Thus hemodynamic measurements obtained from a PAC should be interpreted with the full knowledge of possible confounding factors. It may often be more appropriate to respond to trends of change in central hemodynamic measurement (eg, a slowly falling pulmonary capillary wedge pressure or CVP) rather than to the absolute values themselves.

Echocardiography

In view of the risks and benefits of pulmonary artery catheterization, other noninvasive methods for improving monitoring and resuscitation have been developed and evaluated. Echocardiography has been used in the operating room since the 1970s. Transesophageal echocardiography (TEE) is preferred in the operating room because the acoustic images of transthoracic echocardiography (TTE) are generally poorer than those of TEE.36 Furthermore, factors such as patient positioning, the surgical field, and surgical equipment, drapes, or other monitors may block access to the chest, limiting the usefulness of TTE.

There are some important limitations to TEE.36 Some regions of the heart and great vessels cannot be well visualized. Insertion and manipulation of the TEE probe can produce pharyngeal and/or laryngeal trauma, dental injuries, esophageal trauma, arrhythmias, and hemodynamic effects. The inaccurate interpretation of TEE images may result in improper clinical decisions by the anesthesiologist or surgeon. The performance of TEE will require the anesthesiologists' time and attention and may thus detract from other intraoperative responsibilities or delay other important interventions.

Nevertheless, the role of TEE is expanding both in the operating room and the ICU. Current indications for the use of intraoperative TEE include the diagnosis of myocardial ischemia, confirmation of the adequacy of valve reconstruction and other surgical repairs, and determining the causes of hemodynamic instability and other intraoperative complications.36 TEE can readily provide information on biventricular volumes and contractility, and valvular and wall-motion abnormalities.

The perioperative period is a time of increased risk of myocardial ischemia as a result of hemodynamic and other physiologic stresses associated with anesthesia and surgery. Wall-motion abnormalities generally precede ECG changes during myocardial ischemia37 and may allow earlier detection of ischemia. The incidence of regional ventricular dysfunction detected by TEE ranges from 10% to 60% in various surgical populations.36 Intraoperative TEE detection of ischemia may permit corrective interventions such as altering surgery and/or anesthetic management, initiating infusions, and postoperative triage, which may prevent perioperative complications. There is a lack of studies demonstrating that the detection and treatment of regional ventricular dysfunction or other TEE evidence of ischemia can improve perioperative clinical outcome or increase long-term survival.

Perioperative TEE is often used emergently to determine the cause of acute persistent, life-threatening hemodynamic disturbances. However, elective use of the TEE should be considered in the care of the critically ill surgical patient. TEE has become a useful tool for assessing hemodynamic function qualitatively and for imaging the heart (eg, to diagnose a hemopericardium or cardiac tamponade). It provides an assessment of LV function and an indirect measurement of CO, contractility, and left, and often right, ventricular volume. Because of the limitations in using catheter-derived pressure data to estimate LVEDV, TEE may determine the precise causes of hemodynamic instability (eg, a low CO) in patients with left ventricular dysfunction in a more useful fashion than the PAC.

Esophageal Doppler

Esophageal Doppler monitoring measures aortic flow velocity in the thoracic aorta. A fixed relationship between aortic blood flow and CO is assumed, and thus CO can be calculated using this relationship. The esophageal Doppler is smaller than an ordinary transesophageal echocardiographic probe and thus less invasive. There have been no case reports of esophageal perforation and only reports of minor complications such as mucosal trauma and endobronchial placement. However, proper positioning of the probe is necessary to get an optimal waveform. Although the correlation between CO measured by esophageal Doppler and PAC is modest, there is an excellent correlation with the change in CO with therapeutic interventions.38 One meta-analysis found that use of the esophageal Doppler to guide intraoperative fluid management resulted in an increase of the amount of fluid administered and a reduced length of hospital stay, time to resume full oral diet, and postoperative morbidity or complications.39

Downstream Markers of Organ Perfusion

Derangement of blood flow at the capillary and tissue level is one of the critical pathogenic events in sepsis and is associated with multiorgan failure and mortality.40 The relationship between global hemodynamics as measured by blood pressure, CVP, and CO and microcirculatory blood flow is incompletely understood. Thus, although practice parameters suggest maintaining a MAP of 65 mm Hg or more,4 it is unclear this pressure ensures adequate tissue perfusion. In practice, indices of organ function, such as ECG evidence of myocardial ischemia, urine output, chemical measurements of blood urea nitrogen and creatinine, and liver function tests, are used to indirectly assess the adequacy of regional perfusion.

Blood lactate levels are a product of anaerobic metabolism and an indirect marker of tissue perfusion and the adequacy of resuscitation. An elevated blood lactate concentration (>4 mEq/L) is associated with a high risk of death, and the rate of blood lactate clearance is a good marker of outcome.38 Because lactate clearance lags therapeutic interventions, lactate levels are not suited for immediate assessment of resuscitation. A recent multi-institutional study reported no survival difference in patients with severe sepsis who were resuscitated to normalize CVP, MAP, and either a lactate clearance of 10% or a normalized central venous oxygen saturation (ScVO2).41

Gastric tonometry is a method used to assess regional perfusion of the gut. A balloon is placed in the stomach to measure the intramucosal partial pressure of carbon dioxide (Pco2). This in turn is used to indirectly determine gastrointestinal mucosal pH (pHim). Hypoperfusion causes mucosal carbon dioxide (CO2) to increase and produces gut tissue acidosis. Because CO2 readily diffuses across membranes, the PCO2 in the gut lumen leads to an increase in the gradient between arterial and luminal PCO2. Tonometry provides an indicator of the gut's blood flow-to-CO2 production ratio because a decreased pHim may arise as a result of either decreased blood flow or increased CO2 production. A low gastric pHim and increased gastric luminal PCO2 are highly predictive of postoperative complications.42 Although tonometry is a reasonably good predictor of mortality in critically ill patients,43 its usefulness as a therapeutic guide in patients with sepsis and septic shock has not been proven.

Mixed venous oxygen saturation (SVO2) is measured either in the pulmonary artery with a PAC or estimated via the right atrium with a CVC. It can be used as an end point for therapeutic interventions.10 A low SVO2 may be due to increased oxygen consumption or a decreased hemoglobin, CO, or arterial oxygen saturation. Its major disadvantage is that it is a global measurement, and thus blood from vital organs with low tissue PO2 will be diluted by blood from organs with lower metabolic requirements and a higher PO2. Furthermore, in sepsis, there may be organ or tissue shunting at the microcirculatory level resulting in a higher SVO2.

There are myriad indications for endotracheal intubation of the critically ill patient. These patients may lose their airway, potentially develop an inadequate respiratory drive because of central nervous system disease or the administration of sedative drugs, suffer disruption of their rib cage as a consequence of trauma, develop lung parenchymal disease secondary to ARDS or aspiration, or have an impaired ability to cough or protect their airway against the aspiration of gastric or pharyngeal contents.

As in the healthy patient, preparation for endotracheal intubation should begin with an assessment of airway anatomy. If a difficult intubation is anticipated, fiberoptic intubation should be considered. Airway adjuncts such as a laryngeal mask airway and oral and nasal airways of various sizes should be immediately available. Additional personnel should be on hand and a surgeon should be available to perform a cricothyrotomy or tracheostomy should endotracheal intubation be unsuccessful and ventilation inadequate via mask or laryngeal mask airway.

Studies show that an increased ASA classification and emergency surgery are associated with an increased risk of aspiration.44 Information such as recent vomiting, bowel obstruction, morbid obesity, diabetes mellitus, or a depressed mental status should be factored into the assessment of aspiration risk in the critically ill patient. Prior to intubation, nothing by mouth status should be confirmed if possible. Consider placing a nasogastric tube to empty liquid gastric contents. In the conscious patient, an awake intubation is preferred, although a rapid sequence intubation may also be considered.

Direct laryngoscopy to facilitate tracheal intubation produces a marked stress response.45 Although these responses are short lived, they may produce detrimental effects on the coronary or cerebral circulation of high-risk patients.46 Thus patients should first be assessed for the presence of angina or ischemia, dysrhythmias, and congestive heart failure. The patient's neurologic status including the presence of an increased intracranial pressure, intracranial aneurysms, and hemorrhage should be determined. In these patients, hypertension should be avoided and heart rate and blood pressure should be maintained within a narrow range during laryngoscopy and intubation. Before intubation, adjuncts such as airway blocks with local anesthetics, use of adequate β-adrenergic blockade, and deep anesthesia with opioids or barbiturates should be considered.

In trauma patients, the presence of cervical and mandibular fractures and instability should be sought. All patients with multiple trauma, head, or facial injury should be presumed to have a cervical spine injury unless it has been excluded by a thorough radiographic and physical evaluation. A second skilled person should be present during intubation to provide inline stabilization to maintain the head and neck in a neutral position. Nasotracheal intubation is relatively contraindicated in patients with oropharyngeal and facial trauma because of the possibility of cranial vault disruption.

Patients with spinal cord denervation injuries, crush injuries, or burns should not be given depolarizing muscle relaxants because of the risk of life-threatening hyperkalemia. The patient's coagulation status should be evaluated because mucosal trauma and bleeding associated with laryngoscopy can impair visualization of the airway and increase the risk of aspiration.

General anesthesia is most often used for surgery in the critically ill patient because of multiple factors, among them surgical and hemodynamic considerations. However, a regional anesthetic can be a valuable adjunct to general anesthesia and in the management of postoperative pain in the critically ill patient to enhance patient comfort and to reduce physiologic stress.47

Epidural analgesia is a regional analgesic technique commonly used in the ICU. It is used to manage pain after chest trauma, thoracic and abdominal surgery, major orthopedic surgery, and intractable anginal pain. Studies show that to manage patients after chest trauma, thoracic epidural analgesia provides superior analgesia and improves lung function.48 High levels of thoracic epidural analgesia (T1-T4) can provide effective treatment of myocardial ischemia refractory to conventional medical therapy.49 Benefit probably occurs through both increased myocardial oxygen delivery as a result of coronary artery vasodilatation and decreased myocardial oxygen demand because of a reduced heart rate and blood pressure. Issues such as local or systemic infection and coagulopathy can limit or prevent the use of epidural analgesia. Furthermore, there remains a controversy over the safety of placing epidural catheters in sedated patients, and confirmation of catheter position can be difficult when sensory-level testing is unreliable.

The use of peripheral nerve blocks to produce anesthesia in the critically ill patient has not been evaluated by randomized controlled trials. Continuous interscalene, infraclavicular, and axillary catheters may provide excellent postoperative analgesia after shoulder or upper extremity surgery. Similarly, a femoral nerve catheter in combination with a sciatic block can provide pain relief for the entire leg. These techniques can be used to provide surgical anesthesia for procedures such as external fixation, painful dressing changes, or debridement of burns and large soft tissue wounds. Regional analgesia may be particularly advantageous in the patient with brain injury where opioid analgesic agents might mask their neurologic examination.47 Although there are concerns about placing regional blocks in patients with an impaired mental status because of neurologic injury or sedation, using ultrasonography and nerve stimulation to guide the placement of the needle or catheter may minimize the risk of complications.

Specific goals and end points should be defined for the hemodynamic management of the critically ill patient. Therapies such as fluid administration and vasopressor and inotropic support should be titrated to reach those end points. The results of these interventions should be evaluated on an ongoing basis by monitoring a combination of variables reflecting both global and regional tissue perfusion.

Hemodynamic Management

Early Goal-Directed Therapy

The care of the critically ill patient is largely supportive. Based on the old observation that improved survival was associated with the ability to sustain a hypermetabolic state by increasing CO and oxygen delivery, the hypothesis was proposed that by proactively increasing the systemic oxygen delivery rate, oxygen debt could be reversed and mortality from multiple organ failure might be reduced. Unfortunately, a large randomized controlled trial investigating the effect of increasing oxygen transport on the outcome of the critically ill patient found no overall benefits.50 However, some studies have found that the perioperative hemodynamic optimization of high-risk surgical patients appears to be associated with a reduction in morbidity and mortality.51,52 These studies are consistent with a report that early goal-directed therapy in patients with septic shock reduced in-hospital mortality.10

Studies on the intraoperative use of goal-directed fluid therapy have focused on patients who are not critically ill. Goal-directed fluid administration in this less severely ill population has been shown to decrease hospital length of stay, reduce complication rates, and shorten the time to recovery of gut function.53,54

One of the challenges is to define the precise goal in goal-directed therapy. Some studies have targeted a systemic oxygen delivery rate of 600 mL/min per m2 body surface area.52 Others have used a central venous oxygen saturation greater than 70% as the goal.10 More recently, an individualized approach to fluid resuscitation using each patient's response to fluid administration to guide resuscitation has been advocated.53-55

Another challenge is determining the best therapeutic maneuvers to achieve the targeted goal. Most studies have used fluid (either crystalloid or colloid) infusion and inotropic agents (dobutamine and dopexamine). One study required blood transfusions to a hematocrit of 30% or more to help achieve the goal of achieving a central venous oxygen saturation greater than 70% (Fig. 77-2).10

Although the goals of treatment in the critically ill and the ideal therapy needed to achieve those goals remain uncertain, it seems clear that early optimization of hemodynamic status will provide a significant benefit in reducing the mortality rate in patients with severe sepsis and septic shock.

Fluid Management

Fluid resuscitation consists of infusing natural or artificial colloids or crystalloids. Meta-analyses of clinical studies comparing crystalloid and colloid resuscitation in surgical patient populations show no difference of clinical outcome between infusing colloids or crystalloids.56 This has also been demonstrated in the critically ill patient.57 Although no human studies have been done, a goal-directed colloid therapy resulted in increased tissue oxygen in animals suggesting that one benefit of colloid use could be an improved microcirculatory blood flow.58 Resuscitation with crystalloids also results in more edema because the volume of distribution is much larger for crystalloids than for colloids thus requiring more fluid to achieve the same end points.

Pharmacologic Management

Although hypovolemia is the most common factor contributing to shock in the patient with sepsis, below some mean systemic arterial pressure, perfusion becomes linearly dependent on pressure because of the loss of autoregulation. Therefore, it is frequently necessary to infuse vasopressors to maintain an adequate blood pressure in patients with severe shock. Patients with a low CO despite adequate fluid resuscitation (eg, too high CVP or pulmonary capillary wedge pressure) may require an inotropic agent to increase their CO.

Pressor Support

Norepinephrine and dopamine are the first-choice vasopressor agents to reverse the hypotension of septic shock.4 Both human and animal studies suggest some advantages for norepinephrine over epinephrine and phenylephrine. Epinephrine can produce a tachycardia and may have disadvantageous effects on the splanchnic circulation because of vasoconstriction. Use of phenylephrine may result in a decreased stroke volume because it is a pure α-agonist, but it is least likely to cause tachycardia. Norepinephrine increases mean systemic arterial pressure due to its vasoconstrictor effects with little change of heart rate and stroke volume, and it has only weak β-adrenergic effects. Dopamine increases the mean arterial pressure and CO primarily as a consequence of an increased stroke volume and heart rate. Its use may be limited because it produces tachycardia and can be arrhythmogenic. Norepinephrine is more potent than dopamine and may be more effective at reversing the hypotension of patients suffering septic shock. A multicenter randomized trial comparing norepinephrine with dopamine in patients with shock demonstrated no significant difference between the 2 groups in mortality at 28 days although subgroup analysis showed that dopamine was associated with an increased rate of death in patients with cardiogenic shock. Dopamine was also associated with more arrhythmic events than norepinephrine.59

Vasopressin infusion should be considered in patients with refractory shock after infusing adequate fluid volumes and administering high doses of conventional vasopressors.4 Its vasoconstrictor effects are mediated by stimulating peripheral vasopressin receptors. In contrast to catecholamine-mediated vasoconstriction, the effects of vasopressin are preserved despite hypoxia and severe acidosis. Exogenous infusion of vasopressin can reverse the natural arginine vasopressin deficiency that occurs in prolonged shock.60 However, whether this is important to its mechanism of action is unclear. Additional vasopressin mechanisms that might be helpful in septic shock include facilitation of myocyte depolarization and vasoconstriction of vascular smooth muscle, attenuation of nitric oxide generation by cytokines and inflammatory mediators, enhanced adrenergic responsiveness, and stimulation of synthesis of endothelin-1, a potent endogenous vasoconstrictor.61 When infused into adults, vasopressin should be administered at infusion rates of 0.01 to 0.04 units/min. Doses of vasopressin greater than 0.04 units/min are associated with myocardial ischemia, significant decreases of CO, and splanchnic hypoperfusion.

Inotropic Support

Dobutamine is the first-choice inotropic agent for critically ill patients with a measured or suspected low CO in the presence of an adequate left ventricular filling pressure and adequate fluid resuscitation.4 Because a low arterial blood pressure may be produced by both a low CO as well as systemic vasodilatation, infusion of both an inotropic agent such as dobutamine to augment CO and a vasopressor such as norepinephrine to achieve an adequate MAP may be needed.

Dopexamine is a dopamine receptor agonist developed to treat heart failure and low CO states. It is not available for clinical use in the United States. Dopexamine increases splanchnic blood flow and may possess intrinsic anti-inflammatory properties. These additional actions may contribute to the reported reduction in sepsis-related mortality associated with the infusion of dopexamine. Additional controlled trials are needed to demonstrate these clinical effects of dopexamine.

Preservation of Renal Function

Patients with sepsis, cirrhosis, jaundice, hepatorenal syndrome, congestive heart failure, malignant hypertension, preeclampsia and toxemia, hypotension as a consequence of hemorrhage, or recent exposure to IV radiocontrast agents are predisposed to develop acute renal failure when exposed to a subsequent intraoperative ischemic insult. A 2008 meta-analysis found no evidence that pharmacologic or other interventions used to protect the kidneys during surgery are of benefit to patients. However, the authors noted that there is a lack of high-quality studies on this topic and further research is needed to establish the usefulness of interventions during surgery to protect the kidneys from adverse effects.62

Sodium Bicarbonate and N-Acetylcysteine

The most extensive literature on renal failure protection comes from studies of radiocontrast nephropathy. Radiocontrast dye can induce severe changes of intrarenal hemodynamics leading to ischemic injury. Studies show that hydration before the administration of contrast is effective in lowering the rate of acute renal injury.63-65 Furthermore, hydration with sodium bicarbonate has been shown to be more effective than hydration with sodium chloride for the prophylaxis of contrast-induced renal failure.66 The purported mechanism is by inhibition of free radical formation at the higher pH. The prophylactic administration of N-acetylcysteine, an antioxidant, improves renal outcome after radiocontrast administration.67 Although there is no evidence that these maneuvers improve the outcome of patients at risk for other types of acute renal injuries, these studies are often extrapolated to high-risk surgical procedures.

Dopamine

Many studies demonstrate that "renal dose" dopamine infusion (1-2 μg/kg/min) has no beneficial effect on renal outcome perioperatively.68 Urine flow rate frequently increases during dopamine administration. However, urine output may not correlate with adequate oxygenation of the renal medulla. Dopamine infusion causes an increase in renal cortical blood flow, leading to increased glomerular filtration, solute excretion, and urine output. These actions increase renal oxygen consumption. Dopamine also has a diuretic and natriuretic effect, believed to be a result of inhibition of sodium-potassium adenosine triphosphatase in the proximal tubule and medullary thick ascending limb. This inhibitory effect would decrease tubular energy requirements and medullary oxygen requirements. Thus the net effect of a dopamine infusion on renal medullary oxygenation is unclear.

Mannitol

Mannitol has traditionally been given IV to patients considered at high risk for acute renal failure. It increases tubular diameter and decreases tubular resistance to fluid flow by decreasing endothelial cell swelling by dehydration. The enhanced urine flow is believed to help prevent tubular obstruction and further renal injury. Mannitol is a weak free radical scavenger. Although mannitol is beneficial in animal studies and small clinical trials,69 there are no large prospective controlled trials demonstrating any benefit in surgical patients who are at high risk of renal damage.

Mannitol is usually administered as a single IV dose (0.5-1.0 g/kg) intraoperatively before the application of the cross-clamp in aortic surgery patients, or it is included in the cardiopulmonary bypass prime solution for cardiac surgery. Volume status and electrolyte balance, especially plasma potassium shifts, should be monitored after the administration of mannitol.

Furosemide

Furosemide inhibits energy-dependent reabsorptive work in the medullary thick ascending limb and thus ameliorates hypoxia in the renal medulla. Because it can cause cortical vasodilatation resulting in medullary hypoperfusion, prolonged administration of furosemide may be more deleterious than protective to renal function. Thus it may be prudent to only administer furosemide as a single large-bolus dose shortly before the anticipated ischemic stress. There are no randomized studies evaluating furosemide as a sole renal protective agent in surgical patients.

Fenoldopam Mesylate

Fenoldopam is a selective dopamine-1 (DA1) receptor agonist used to treat hypertension.70 Unlike dopamine, it has no activity at dopamine-2 (DA2) or at α- or β-adrenergic receptors, and thus it does not cause either tachycardia or hypertension. Fenoldopam reduces blood pressure in a dose-dependent manner while preserving renal perfusion and glomerular filtration rate.71 It inhibits sodium reabsorption and thus may attenuate medullary oxygen demand while enhancing its supply. Large prospective trials evaluating fenoldopam as a renal protectant in humans have not yet been published.

Dopexamine

Dopexamine, a dopaminergic (DA1 and DA2) agonist with β2-adrenergic activity, has generated interest as a renal protectant based on the hypothesis that it could improve renal perfusion without causing α-adrenergic stimulation.72 Dopexamine lacks the myocardial side effects of dopamine, although it may cause hypotension. Large well-designed trials are needed to assess its usefulness as a renal protectant.

Intraoperative Glucose Management

Anesthesia and surgery are associated with increased catecholamine levels resulting in increased glycogenolysis, gluconeogenesis, and lipogenesis. Furthermore, glucagon levels increase and insulin secretion decreases. Thus patients undergoing anesthesia and surgery are at risk for hyperglycemia caused by the combination of increased insulin resistance and decreased insulin secretion. Data regarding the outcome of intraoperative glucose control in noncardiac surgery are lacking. Based on evidence in cardiac surgery patients, the Society of Thoracic Surgeons recommended maintaining intraoperative blood glucose less than 180 mg/dL. They also note that glycemic control is best achieved with continuous insulin infusions rather than intermittent subcutaneous insulin injections or intermittent IV insulin boluses.73

Awareness

Awareness is the postoperative recollection of events occurring during general anesthesia. The incidence of awareness is rare in the general surgical population (0.1%-0.2%).74 However, the incidence of awareness is greater in certain at-risk populations, especially patients who receive light anesthesia, usually because of hemodynamic instability. For example, the incidence of awareness in patients undergoing surgery for major trauma is reported to be as high as 43%. Awareness can be difficult to detect clinically because the typical indicators of "light" anesthesia such as an increased heart rate, hypertension, or movement may be masked by drugs or the patient's overall status. More than half of the patients who report experiencing intraoperative awareness have postoperative mental distress, including posttraumatic stress syndrome. If light anesthesia is required because of hemodynamic considerations, then amnestic drugs such as midazolam, scopolamine, or subanesthetic doses of ketamine should be administered. When available, neuromonitoring technologies to detect the presence of awareness should be considered. One randomized controlled study found that Bispectral Index Sensor monitoring of the electroencephalogram (EEG) of high-risk patients (defined as patients undergoing caesarian delivery or high-risk surgery, patients with a history of chronic benzodiazepine, opioid, or heavy alcohol use, and patients with acute trauma and hypovolemia) reduced their risk of awareness by 82%.75

Ideally, the critically ill patient should be transported to the ICU from the operating room by an anesthesiologist. The considerations for transport to the ICU should be similar to those for transport to the operating room. The patient should be hemodynamically stable and have a secure airway before transportation. Emergency medications, ample oxygen, and supplies for airway management must be available. There should be functional IV access. All infusions should be running well and sufficient medications to last the time needed for transportation to and restabilization in the ICU should be available.

The anesthesia team is responsible for the care of the patient until a full verbal report is given to the ICU team and the ICU team accepts caring for the patient. At this time, the anesthesiologist is the caregiver who is most aware of the patient's status and therefore should be available to assist with any problems that arise. A full report, including the patient's medical history, location of monitors and lines, anesthesia technique, including the amounts and types of drugs administered, surgical procedure, other medications administered, estimated fluid and blood volume loss and replacement, anesthetic and surgical complications, and any special issues (eg, allergies, isolation precautions) should be provided to the nurses and physicians who will be caring for the patient in the ICU. The report should include specific problems (eg, oxygenation, ventilation, hemodynamic instability) that were encountered intraoperatively, the maneuvers, whether successful or unsuccessful, made to resolve the problems, and the rationale behind these maneuvers. A well-considered plan for postoperative pain management should be conveyed. The ICU team should be informed of any special care plans for the postoperative period as well as any potential postoperative problems.

Anesthesiologists are involved in the care of critically ill patients in the operating room because they often need surgery to correct the underlying cause of their illness or to deal with the complications of their illness. Planning for the anesthetic management of the critically ill patient begins with a careful preoperative evaluation because the patient may have impaired function of multiple organ systems. Ideally, optimization of the patient's condition should occur preoperatively to ensure that the patient is in the best possible condition to undergo the additional stresses associated with surgery and anesthesia.

Preoperative planning should include a well-considered plan for transport to the operating room because patients are at high risk of adverse events during transport. The anesthesiologist must decide which monitors are necessary to assess the patient's condition, taking into account the advantages and pitfalls of the various options. General anesthesia is most often planned for surgery in the critically ill patient. The anesthesiologist may need to plan a total IV anesthetic technique if the patient has severe respiratory failure and will require an ICU ventilator intraoperatively for adequate oxygenation and ventilation. Regional anesthesia should be given due consideration because it can play a valuable adjunctive role to achieve optimum patient comfort and reduce physiologic stress both intraoperatively and postoperatively.

There should be specific goals and end points defined in the intraoperative management of the critically ill patient to optimize hemodynamics and minimize further end-organ damage. Advanced planning and communication between the anesthesiologists, surgeons, critical care team, and the patient and the patient's family are vital to understanding the goals and priorities of treatment.

• Arbous MS, Meursing AE, van Kleef JW, et al. Impact of anesthesia management characteristics on severe morbidity and mortality. Anesthesiology. 2005;102:257.
• Bundgaard-Nielsen M, Holte K, Secher NH, et al. Monitoring of peri-operative fluid administration by individualized goal-directed therapy. Acta Anaesthesiol Scand. 2007;51:331.
• Dellinger RP, Levy MM, Carlet JL, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Intensive Care Med. 2008;34:17.
• Perel P, Roberts I, Pearson M. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev. 2009:CD000567.
• Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368.
• The NICE-SUGAR Study Investigators. Intensive versus conventional glucose control in critically ill Patients. N Engl J Med. 2009;360:1283.
• Zacharias M, Conlon NP, Herbison GP, et al. Interventions for protecting renal function in the perioperative period. Cochrane Database Syst Rev. 2008:CD003590.