CHAPTER 99: ANESTHESIA AND THE PATIENT WITH CARDIOVASCULAR DISEASE

Anesthetizing patients with cardiovascular disease is one of the greatest challenges facing the anesthesiologist. The constellation of anesthetic drug effects, the physiologic stresses of surgery, and underlying cardiovascular diseases complicate and limit the choice of anesthetic techniques for any particular procedure. The anesthesiologist’s approach to the patient with cardiovascular disease is to select agents and techniques that will optimize the patient’s cardiopulmonary function. The perioperative management of a patient with cardiovascular disease requires close cooperation between the cardiologist/internist, surgeon, and anesthesiologist. Each specialist has a unique knowledge base that complements that of the others. The approach should emphasize a continuum of care from the preoperative evaluation through the postoperative period.

The assessment of cardiac risk and preoperative optimization of the patient’s cardiovascular status are the traditional goals of the preoperative evaluation of patients with cardiovascular disease. In 1977, Goldman and associates¹ introduced the Cardiac Risk Index Score (CRIS) to guide more quantitatively the assignment of cardiac risk in patients undergoing noncardiac surgery. According to the CRIS, the risk for adverse cardiac events increases with the number of preexisting conditions such as heart failure, ischemic disease, cerebrovascular disease, diabetes, chronic kidney disease, as well as the type of planned procedure. This study had a major impact because clinicians concluded that preoperative improvements in some of these factors, such as heart failure symptoms, would decrease cardiac risk. Cardiac risk assessment continues to evolve, and is reviewed in detail in Chapter 98. The American College of Cardiology (ACC)/American Heart Association (AHA) Task Force on Practice Guidelines published “Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery,” which was last updated in 2014.² Preoperative evaluation should include a thorough history and physical examination, assessment of surgery specific risks, and advanced testing based on the severity of underlying cardiac disease as outlined in the AHA guidelines. The algorithmic approach to preoperative evaluation described in these guidelines is valuable in that more consistent clinical approaches have emerged (see Chapter 98, Fig. 98–1).

Specifically, the information derived from preoperative cardiac evaluation should address the following points:

1. Urgency assessment: If this is an emergency procedure, proceed to operating room and attempt risk modification (ie, surveillance/monitoring).

2. If surgery is not urgent, assess patient for signs of “active cardiac conditions.” These include unstable “acute” coronary syndrome (ACS) or recent myocardial infarction, decompensated heart failure, severe valvular disease, and significant arrhythmia. If present, consider further evaluation and treatment as per AHA guidelines before surgery.

3. In chronic, less severe cardiac disease (no active cardiac conditions), surgery-related risk assessment suggests advancing to surgery in low-risk surgical procedures.

4. If surgery-related risk is intermediate or high, functional status can give invaluable information about the cardiac reserve in response to surgery-related stress (eg, large fluid shifts, sudden hemodynamic changes). Functional capacity is described using metabolic equivalents (METs); one MET is the oxygen consumption of a 70-kg, 40-year-old man at rest. With moderate (eg, climbing a flight of stairs) or excellent functional status, proceeding to surgery is suggested.

5. With poor functional status (not able to perform daily activities such as eating, dressing, bathing), assessment for risk factors associated with cardiac disease is suggested. These include mild angina, prior (not recent) myocardial infarction, compensated heart failure, diabetes, renal insufficiency, advanced age, abnormal electrocardiogram (ECG), rhythm other than sinus, history of stroke, and uncontrolled hypertension.

6. In patients with poor or unknown functional status (< 4 METs) and three or more cardiac risk factors, additional testing should be considered for intermediate-risk or vascular surgery if testing will change management.

7. Patients with poor or unknown functional status (< 4 METs), but less than three cardiac risk factors, may proceed with planned surgery. In that case, risk modification, for example using β-blockade for heart rate control, is recommended unless contraindications apply.

8. Additional considerations may apply in patients with cardiac devices (pacemaker or implantable cardioverter-defibrillator) that may require interrogation/programming before going for surgery.

9. Patients on anticoagulant or antiplatelet therapy may need to have that adjusted prior to surgery

10. Patients may present contraindications for specific anesthetic techniques (see next paragraphs).

A cogent summary of the pertinent clinical, laboratory, radiologic, echocardiographic, radionuclide, and cardiac catheterization data comprises the ideal cardiac consultation for the anesthesiologist. With the benefit of this information, the two specialties can make intelligent decisions regarding the patient’s preoperative therapy and the optimal timing of surgery.

Cardiovascular disease management frequently includes anticoagulation, as for stroke prophylaxis in patient, with atrial fibrillation, and potent antiplatelet agents, as are required following percutaneous coronary interventions (PCIs). Implications for surgery and anesthesia include timing of surgery and management of anticoagulation and antiplatelet therapy to minimize the risk of surgical bleeding without increasing the risk of stroke or increasing risk for stent thrombosis, respectively. Certain anesthetic techniques, such as neuraxial anesthesia, may be contraindicated in the presence of anticoagulant/antiplatelet regimens. Potentially life-threatening complications may occur if current guidelines and recommendations are not followed.

The increased risk of perioperative cardiac events following PCI is related to the perioperative hypercoagulability associated with surgical stress in the setting of non-endothelialized stent surfaces. Answering the questions of when and if anticoagulant/antiplatelet medications can be safely discontinued for surgery is an ongoing process, as stent technology and drug therapies continue to evolve.^3,4 Ideally, elective noncardiac surgery should be delayed until the stent surface is fully endothelialized. The 2014 ACC/AHA guidelines do not recommend elective noncardiac surgery within 2 weeks of coronary balloon angioplasty or 4 weeks after bare metal stent implantation.² Elective surgery should be delayed for at least 1 year following drug-eluting stent implantation. For nonelective surgery, the task force encourages a consensus decision among the clinicians involved regarding whether delaying surgery outweighs the risk of stent thrombosis. If surgery cannot be deferred and dual antiplatelet therapy (DAPT) must be interrupted in patients with recently placed coronary stents, aspirin should be continued if possible, and DAPT should be restarted as soon as possible after the surgical procedure. Recommended time intervals between discontinuation of DAPT and surgery depend on patient risk factors, type of surgery, and even the surgeon’s preference to accept an increased risk of surgical bleeding versus risking stent thrombosis with possibly catastrophic consequences. Many recommendations call for discontinuing clopidogrel 5 to 7 days, prasugrel 7 to 9 days, and ticagrelor 5 days before surgery. Bridging therapy in patients with long-term DAPT has been suggested in patients at high risk for cardiovascular events. However, unlike in patients on warfarin, unfractionated or low molecular weight heparin is not effective in patients who are depending on potent antiplatelet drugs. Alternatively, short-acting glycoprotein (GP) IIb/IIIa inhibitors and P2Y₁₂ inhibitors have been tried for bridging to surgery in patients on DAPT; however, experience is limited at this point. For certain procedures, including neurosurgery, ophthalmology, or spine surgery, the risk of even continuing aspirin perioperatively may outweigh the risks of discontinuing the drug. In emergency procedures, patients on long-acting antiplatelet drugs have a significantly increased risk of hemorrhage, and platelet transfusions may be necessary to achieve hemostasis. Based on the ACC/AHA guidelines summarized earlier, the American Society of Anesthesiologists (ASA) issued a Practice Alert for the perioperative management of patients with coronary artery stents.⁵

Anticoagulation therapy is indicated in patients at risk for thromboembolism (atrial fibrillation, mechanical heart valves, heart failure). Vitamin K antagonists (warfarin) or any of the novel oral anticoagulants (NOACs) are discontinued prior to surgery based on the increased risk of bleeding complications. Drugs with longer half-lives, such as warfarin, may require bridging to surgery either with unfractionated (intravenous) heparin or low-molecular-weight heparins. Bridging anticoagulation is usually considered in patients at very high thromboembolic risk (mechanical heart valves, recent [< 12 weeks] embolic stroke, or venous thromboembolism). If surgery cannot be delayed, the risk of bleeding is increased and must be weighed against the urgency of the procedure. Many of the NOACs have shorter half-lives than warfarin. For example, the current recommendation for dabigatran is to discontinue 1 to 2 days before the procedure if creatinine clearance is ≥ 50 mL/min, and 3 to 5 days if < 50 mL/min.

Cardiac risk assessment provides the clinician with an evidence-based approach to clinical decision making if a patient with known risk factors (according to the revised cardiac risk index [RCRI]) for major adverse cardiac events (MACE) should proceed for surgery or if prior optimization of risk factors should be attempted. However, if surgery in a patient with significant risk is unavoidable, the anesthesiologist must choose an anesthetic technique, continue implementing guideline-directed medical therapy, and choose appropriate monitoring to further reduce the risk of MACE.

Choice of Anesthetic Technique

The choice of anesthetic technique is inherently a difficult one because multiple factors must be considered. These include patient preferences, the requirements of the surgical procedure, and the patient’s underlying medical condition(s). There is little scientific evidence that any particular anesthetic approach is superior to reasonable alternatives, or that anesthetic technique per se influences patient outcome.^{6,7,8,9,10,11,12}

Most recently, Leslie and coworkers analyzed patients registered in the Perioperative Ischemic Evaluation 2 (POISE-2) trial regarding the effect of a neuraxial technique (spinal, lumbar epidural, or thoracic epidural) alone or in combination with general anesthetic on 30-day adverse cardiovascular outcomes (death, stroke, myocardial infarction).¹³ Neuraxial block and postoperative epidural analgesia were not associated with adverse cardiovascular outcomes. Poeran and associates retrospectively analyzed 98,290 elective colectomies.¹⁴ General anesthesia was used in the majority of cases (93.9%), while the remainder received a combined anesthetic that included a neuraxial technique. Although the addition of a neuraxial technique was associated with a decreased risk for thromboembolism and cerebrovascular events, myocardial infarction and other morbid outcomes, such as urinary tract infection, postoperative ileus, blood transfusion, and admission to the intensive care unit, were observed more frequently in the neuraxial group. Overall, the findings were inconsistent, not showing a clear advantage for either anesthetic technique. In a systematic review, Guay and colleagues¹⁵ inferred that there is a moderate level of evidence that the addition of a neuraxial technique to a general anesthetic may reduce 30-day mortality and the risk of pneumonia, but does not impact rate of myocardial infarction.

Overall, there are no definite recommendations for choosing a particular anesthetic to decrease the risk of perioperative cardiac events. The most effective strategies remain preoperative optimization and vigilant perioperative monitoring to allow for early detection of hemodynamic compromise. In certain defined surgical procedures (eg, transurethral resection of prostate), a neuraxial technique may be advantageous in enabling early recognition of complications. This is not applicable in all circumstances, however, and clinical judgment must be exercised to make the best choices in individual circumstances. Regional anesthetics and monitored anesthesia care are not infrequently converted to general anesthetics intraoperatively as a result of unexpectedly long surgery, patient discomfort, or changes in the surgical plan. No practitioner can be certain that a particular technique will be adequate for the surgical procedure, given the unpredictability of the situation, and the anesthesiologist must have flexibility to alter the technique, as needed. Therefore, it is essential that the cardiologist/internist does not recommend excluding specific anesthetic technique(s) during a preoperative consultation.

General Anesthesia

General anesthesia is defined as a reversible state consisting of amnesia, analgesia, immobility, and the prevention of undesirable reflexes. The general anesthetics include many drugs, almost all of which have cardiovascular side effects. Intravenous agents are nearly always used for the induction of anesthesia in adults. Anesthesia is maintained using inhalational agents, intravenous agents, or a combination of the two.

Neuromuscular blocking drugs (muscle relaxants) are used to facilitate tracheal intubation, to lower the requirements for anesthetic agents, and to prevent involuntary muscular activity in surgical cases where complete paralysis is mandatory. In children, the induction of anesthesia is highly individualized according to patient needs, the practitioner’s preferences, and institutional standards. With the exception of brief operations, most general anesthetics include tracheal intubation and mechanical ventilation. As an alternative to tracheal intubation, supraglottic devices, such as the laryngeal mask airway, may be used to secure a patient’s airway. Loss of consciousness is usually accompanied by a decrease in sympathetic tone. This, as well as the effects of positive pressure ventilation and the cardiac depressant properties of inhalational, and most intravenous, anesthetic agents, causes a moderate decrease in cardiac output.

General anesthesia masks many of the symptoms of cardiovascular decompensation, such as angina, dyspnea, dizziness, and palpitations. Other signs of cardiovascular disease, such as tachycardia, are nonspecific and may be misinterpreted as being cause by hypovolemia or light anesthesia. Depending on the type of surgery, large fluid shifts and decreased venous return may occur, sometimes unpredictably. It is for these reasons that appropriate monitoring and selection of anesthetic agents is vital to the intraoperative management of the patient with cardiovascular disease.

Intravenous Anesthetics

Intravenous anesthetic induction drugs are composed of lipophilic molecules that have an affinity for neuronal tissue or specific receptors. Their anesthetic actions (eg, hypnosis) are generally terminated by redistribution from the vessel-rich tissues (brain, heart, liver, and kidneys) to other tissues (muscle, fat, and skin). Elimination via hepatic metabolism and renal excretion is typically much slower and takes place over several hours. Most intravenous anesthetics exhibit some degree of cardiovascular depression in the form of reduced cardiac output or vasodilation. Reduced doses and slower injection of the drug (titrating to effect) will markedly decrease these cardiovascular effects. The use of synthetic opioids (fentanyl, sufentanil, or remifentanil), etomidate, or ketamine may be indicated in patients with severely compromised cardiac function, because they tend to maintain hemodynamic stability. Table 99–1 summarizes commonly used intravenous anesthetics.

Inhalational Anesthetics

Inhalational anesthetics include nitrous oxide and the potent volatile anesthetic agents. Nitrous oxide has analgesic properties, but it is not a potent anesthetic. Concentrations up to 75% may be given safely (so as to maintain an adequate FiO₂), but incomplete amnesia, postoperative nausea, and movement in response to painful stimuli are likely. Thus, nitrous oxide is nearly always administered with other anesthetic agents, such as opioids or potent volatile anesthetics, and neuromuscular blockers. Many anesthesiologists opt to avoid nitrous oxide altogether, because of the minimal analgesic benefits and increasingly recognized undesirable side effects.

The use of inhalational anesthesia with potent volatile anesthetics is the most common anesthetic technique because of its relatively low cost, ease of use, reliable amnesia, and overall safety record of these agents. The effect of these agents is rapidly changed when the inspiratory concentration is adjusted. The ability to easily titrate inhalational anesthetics is an advantage compared with intravenous drugs, because the duration of surgical procedures and the degree of surgical stimulation are often unpredictable. All volatile agents are myocardial depressants at higher doses and cause vasodilation resulting in some degree of hypotension. Table 99–2 summarizes commonly used inhalational anesthetics in current practice.

There is evidence that potent volatile anesthetics offer some degree of myocardial protection from ischemic insult. The mechanisms of anesthetic drug-induced protection from myocardial injury are complex, and related to adaptive and protective cellular processes that confer protection to recurrent ischemic events. Myocardial preconditioning to ischemic events (ie, temporary coronary occlusion) elicits these processes; however, some volatile anesthetic agents and certain anesthetic drugs, such as opioids, mimic ischemic preconditioning.^{16,17,18,19,20,21,22,23,24} Substantial literature has been published on this topic. In clinical practice, however, no meaningful differences in major outcome parameters have been found. The time course of administration of potent volatile agents (pre- vs post-conditioning), other comorbidities (eg, diabetes mellitus), and drugs that may abolish protective effects (eg, ketamine, β-blockers, and certain sulfonylureas) must be considered.^25,26 Currently, there are no conclusive data regarding which patients may benefit from anesthetic preconditioning or postconditioning.

Neuromuscular Blockade

Two major classes of nondepolarizing neuromuscular blocking drugs are currently used: benzylisoquinolinium compounds (eg, cisatracurium) and aminosteroid derivates (eg, vecuronium and rocuronium). Cisatracurium undergoes a unique form of spontaneous degradation that is organ-independent (Hofmann elimination) and, therefore, is frequently used in patients with renal insufficiency or hepatic failure. Pancuronium is the classic aminosteroid nondepolarizing neuromuscular blocking drug. Its effects, however, are long lasting and it has been supplanted by shorter-acting compounds. Vecuronium is structurally very similar to pancuronium, but with an intermediate duration of action. Vecuronium does not have clinically significant cardiovascular adverse effects. Rocuronium has a more rapid onset of action, as a result of its lower potency and slightly increased heart rate. Although pancuronium elimination is almost entirely renal, the newer aminosteroid compounds are mainly degraded by the liver. These nondepolarizing neuromuscular blocking drugs require reversal at the end of a procedure to restore normal muscle function and respiration. The pharmacologic reversal is usually produced with a combination of neostigmine and glycopyrrolate, which may lead to bradycardia, tachycardia, or other arrhythmias in patients with cardiovascular disease, and has been associated with cardiac arrest even in patients with prior heart transplants. Recently, a new reversal agent, sugammadex, has been approved that produces none of these adverse cardiovascular effects.

Succinylcholine is a depolarizing short-acting neuromuscular blocker that is associated with rapid onset, and short duration of action. Its cardiovascular effects depend on whether nicotinic or muscarinic receptor effects predominate in a given patient. Thus, tachycardia and hypertension or bradycardia and hypotension may occur. Vagal effects tend to predominate with repeated doses or in children. In patients with various disorders (including neuromuscular diseases, recent burns, and massive trauma), hyperkalemic cardiac arrest may occur because of exaggerated release of intracellular potassium from myocytes. It is, therefore, not recommended for routine use, particularly in children.

Regional Anesthesia and Neuraxial Anesthesia

Cushing coined the term regional anesthesia for operations where local anesthetics were used to operate on localized areas of the body without loss of consciousness. Regional anesthesia typically involves the use of a local anesthetic to block sensation and pain from one or several large peripheral nerves and associated area(s) of distribution. Neuraxial anesthesia infers placement of a local anesthetic around the nerves of the central nervous system (spinal or epidural anesthesia). The advantages of either technique include simplicity, low cost, and minimal equipment requirements. Many of the adverse effects of general anesthesia are avoided, such as myocardial and respiratory depression. There are many surgical procedures, however, that are not amenable to a regional or neuraxial anesthetic technique. Serious side effects, such as hypotension, may be long-lasting and not reversed as easily compared to a general anesthetic. Additional contraindications include use of systemic anticoagulation, use of select antiplatelet therapy, patient refusal to be awake in the operating room, and local or systemic infections. The combination of a regional or neuraxial technique with a general anesthetic can be advantageous in providing adequate analgesia extending into the postoperative period with the added advantages of allowing for earlier mobilization and likely reducing thromboembolic risk.

Local Anesthetic Agents

Local anesthetics are classified based on their chemical structure as esters or amides. The esters are hydrolyzed by esterases in the plasma, and the amides are metabolized in the liver. The duration of action of local anesthetic agents is affected by the protein-binding characteristics of the molecule. Epinephrine and phenylephrine, which produce local vasoconstriction, may be added in small doses to local anesthetic solutions to prolong their duration of action. The systemic absorption of epinephrine occurs very slowly, and the β-adrenergic effects predominate. This results in slight tachycardia and diastolic hypotension, which is undesirable in patients with certain cardiovascular diseases. Toxic reactions to local anesthetics are generally characterized by central nervous system excitation (seizures), which may be followed by central nervous system depression and cardiovascular collapse. Table 99–3 provides an overview of commonly used local anesthetics in current practice.

Spinal Anesthesia

The injection of a relatively small dose of local anesthetic into the subarachnoid space produces profound motor and sensory blockade. The level of the anesthetic block is usually controlled by the choice of local anesthetic and the position of the patient. Spinal anesthesia also produces blockade of preganglionic sympathetic fibers, resulting in a sympathetic blockade that is generally two dermatomal segments higher than the sensory dermatomal level. If the dermatomal level of sympathetic blockade reaches T1, then a complete sympathectomy is present until the block recedes. Higher levels of sympathetic blockade are associated with profound hypotension from arterial and venous vasodilatation, as well as bradycardia from the loss of cardiac accelerator fiber function. Blockade above a T2 level also produces respiratory insufficiency as a result of intercostal and phrenic nerve root blockade. Cardiovascular collapse and respiratory insufficiency (or apnea) are the signs of a “total spinal block,” a life-threatening situation that must be treated with tracheal intubation and aggressive vasoactive support. Because of the risk of sudden severe vasodilation and hypotension, spinal anesthesia is contraindicated in patients with severe valvular stenosis or hypertrophic obstructive cardiomyopathy.

Epidural Anesthesia

The epidural space lies immediately external to the dura mater and is filled with loose areolar tissue and a venous plexus. An indwelling catheter is usually placed percutaneously for intermittent bolus injections or continuous infusions of local anesthetic or opioids. The epidural space may be entered by thoracic, lumbar, or caudal approaches. The hemodynamic effects of epidural anesthesia are similar to those of spinal anesthesia, except that the onset of sympathetic blockade is more gradual. Thus, with appropriate monitoring, cautious administration of epidural anesthetics can be performed safely, even in patients with valvular stenosis, or hypertrophic obstructive cardiomyopathy. In patients with cardiovascular disease, advanced monitoring techniques should be strongly considered. Beat-to-beat blood pressure monitoring via an intra-arterial catheter, and a secure route of central drug administration, for example via a central venous catheter, allows the practitioner to monitor and promptly treat changes in hemodynamic parameters.

In patients with known coronary artery disease, epidural anesthesia and analgesia, especially thoracic epidural catheters, have been shown to reduce intraoperative and early postoperative ischemic events, lower hormonal stress response, and reduce the incidence of atrial fibrillation.^27,28,29,30 Patient outcome data, however, were inconclusive when epidural anesthesia was compared with general anesthesia in patients at risk for perioperative cardiac events undergoing vascular surgery.^31,32

When compared with spinal anesthesia, epidural anesthesia requires higher doses of local anesthetic. This increases the potential for complications and adverse effects from inadvertent intravascular injection or absorption that can cause severe arrhythmias, including ventricular fibrillation and cardiovascular collapse. Potent local anesthetics such as bupivacaine and, to a lesser degree, ropivacaine are more cardiotoxic than lidocaine. The slower rate of recovery of fast Na⁺ channels in the conduction system and myocardium seen with bupivacaine is frequently identified as the reason for its increased cardiotoxicity. Cardiac resuscitation following inadvertent intravenous administration and cardiac arrest is difficult. Intravenous lipid emulsion has been successfully used to reverse local anesthetic toxicity,^33,34 and is now recommended to treat cardiovascular collapse from local anesthetic toxicity.³⁵ The hemodynamic consequences of inadvertent intravenous injections of epinephrine-containing solutions may be significant for patients who cannot tolerate tachycardia. Epidural infusions of opioids for postoperative analgesia may be complicated by pruritus, nausea, urinary retention, somnolence, and respiratory depression. Thus, appropriate monitoring and nursing care are required.

Nerve Blocks and Infiltration of Local Anesthetic

Nerve blocks and local anesthetic infiltration may be performed to facilitate surgery of localized areas of the body. For upper extremity surgery, the brachial plexus can be blocked by various approaches. The lower extremity may be anesthetized by blocking the femoral, obturator, and sciatic nerves. Local anesthetic infiltration (“field block”) is performed in defined regions, such as the inguinal area. These blocks, when properly performed, have minimal cardiovascular effects. They do, however, require large volumes of local anesthetic solution, which result in toxic reactions if inadvertent intravascular injection occurs. Intercostal blocks are associated with high blood concentrations even without intravascular injection, because the neurovascular bundle enhances absorption of the local anesthetic, and multiple blocks are required for clinical efficacy. Recently, thoracic paravertebral blocks have been used more frequently for patients having unilateral surgery to avoid the hemodynamic effects of epidural blockade.

Regional Anesthesia and Anticoagulation Therapy

Appropriate patient selection is crucial to avoid potentially catastrophic complications (eg, paraplegia) when planning for an intraoperative central neuraxial anesthesia or postoperative neuraxial analgesia in patients with anticoagulation or potent antiplatelet therapy. A careful evaluation of drug history and bleeding diathesis provides essential information, because the risk of bleeding associated with many antiplatelet therapies will not be detected by standard preoperative coagulation tests.³⁶ The establishment of guidelines for the use of neuraxial anesthesia and analgesia in patients who have or will receive anticoagulants is an evolving process. Recommendations from the American Society of Regional Anesthesia and Pain Medicine for appropriate withdrawal of anticoagulant and antiplatelet therapy prior to neuraxial anesthesia can be found at the American Society of Regional Anesthesia and Pain Medicine website site (http://www.asra.com).³⁷ It is prudent to avoid neuraxial manipulation in patients who are receiving potent antiplatelet drugs that include GP IIb/IIIa antagonists, adenosine diphosphate inhibitors, and low molecular weight heparins at the time of the planned anesthesia. There are currently also very limited data on the safe use of new oral anticoagulants, such as the direct thrombin inhibitors and factor Xa inhibitors, in patients considered for neuraxial anesthesia. Safe time intervals between discontinuation of a specific drug and safe administration of a neuraxial anesthesia are an evolving process. Table 99–4 summarizes these guidelines and the current literature. Although guidelines can help in risk assessment, it is important to recognize that the decision to perform a neuraxial technique must be made on an individual basis. Frequently, a combination of drugs is administered, and the response to certain drugs may vary between individuals, for example based on genetic variances, with unpredictable effects on the coagulation system. For patient safety, it is also crucial to monitor neurologic status carefully after administration of spinal or epidural anesthesia; rapid diagnosis and treatment of neuraxial hematoma probably improves outcome.³⁸

Perioperative Monitoring

Perioperative monitoring has significantly contributed to attainment of the level of safety provided by modern anesthesia, allowing for more complex surgeries in even the highest-risk patients. The ASA established standards for basic intraoperative monitoring in 1986.³⁹ Intraoperative monitoring that is required based on these guidelines include the following: (1) heart rate, (2) ECG, (3) blood pressure, (4) pulse oximetry, (5) capnometry, and (6) body temperature. The indications for the use of more invasive monitors, such as intra-arterial and central venous monitoring, vary and depend on patient and surgery specific factors, as well as institution and practitioner preferences (Tables 99–5 and 99–6).⁴⁰ It must be recognized, though, that monitoring guidelines are almost entirely based on observational cohort analyses and expert opinion. Although adequate monitoring is essential for early detection of hemodynamic disturbances, a particular monitoring technique will not result in improved outcome unless timely and effective treatment is initiated and available. The indications for pulmonary arterial catheter (PAC) monitoring, for example, are especially controversial. Large randomized prospective studies of PAC use in various clinical settings have failed to demonstrate improved patient outcomes,^41,42,43,44 and some have even suggested potential harm.⁴⁵ As with all monitoring devices, the caregiver’s competency in interpreting PAC-derived data and instituting appropriate treatment is essential to derive maximal benefit and avoid complications.^46,47,48 The ASA published practice parameters to guide practitioners in the appropriate use of the PAC.⁴⁹ The decision to use perioperative PAC monitoring should be based on a combination of patient risk factors, surgical risk, and the experience of the practitioner. Many clinicians believe that patients with clinically significant pulmonary hypertension, significant valvular disease (eg, aortic stenosis), or severe cardiac dysfunction, undergoing surgery with expected large fluid shifts or hemodynamic instability will benefit from PAC monitoring.⁵⁰

Transesophageal echocardiography (TEE) has acquired a much larger role in perioperative management. The more widespread availability of these devices in the operating room and intensive care unit setting and the development of newer modalities, such as three-dimensional echocardiography and tissue Doppler, have enhanced the ability of anesthesiologists, cardiologists, and surgeons to make intraoperative diagnoses, evaluate hemodynamic aberrations, and assess the quality of cardiac surgical interventions. Standardized intraoperative examination guidelines for multiplane TEE⁵¹ and training guidelines^52,53 have been published. The National Board of Echocardiography administers a certification process. The ACC/AHA/American Society of Echocardiography Task Force on Practice Guidelines has published practice guidelines that address perioperative TEE.⁵⁴ The ASA and the Society of Cardiovascular Anesthesiologists released the most recent update.⁵⁵ Based on these recommendations, the use of TEE is recommended for all cardiac or thoracic aortic surgery, and may be used for all patients undergoing transcatheter intracardiac procedures. In noncardiac surgery, TEE may be used if the planned procedure or the patient’s known or suspected cardiovascular disease might result in severe hemodynamic, pulmonary, or neurologic compromise. During unexplained life-threatening circulatory instability, and in critical care patients, TEE should be used when diagnostic information that is expected to change management cannot be obtained by transthoracic echocardiography or other modalities in a timely manner. TEE may also be used for patients with oral, esophageal, or gastric disease, if the expected benefit outweighs the potential risk. These are guidelines only, and the practitioner should decide on intraoperative TEE monitoring based on his or her level of training and experience, as well as patient- and surgery-related factors.

Less invasive and noninvasive methods of cardiovascular monitoring are continually being developed. Cardiac output can be estimated using arterial pressure waveform analysis (pulse contour analysis), indicator dilution technique, electrical bioimpedance, and esophageal Doppler ultrasound. Parameters, such as intrathoracic blood volume and extravascular lung water, can also be estimated by some of these devices. Some of the less invasive cardiac output devices require calibrating against more invasive measurements. Additional limitations include poor correlation with more invasive methods in patients with rapidly changing hemodynamics and irregular rhythm.

Cardiac output as well as other monitoring modalities are used in an attempt to intervene and optimize target parameters early in the disease process. This early goal-directed therapy has been suggested to improve outcomes in some studies,⁵⁶ although clear evidence is still lacking.⁵⁷ The lack of outcomes evidence in support of invasive monitoring and the established problems (eg, central line–associated bloodstream infections) give further impetus for improvements and innovations in less invasive and noninvasive monitoring.

Various “brain function” monitors using proprietary electroencephalographic analysis have been developed for the purpose of monitoring depth of sedation and level of consciousness.⁵⁸ Incomplete amnesia leading to intraoperative awareness is rare with current anesthetic techniques, with a reported incidence of 0.1% to 0.2%.^59,60 The ASA currently does not recommend routine brain function monitoring in patients undergoing general anesthesia.⁶¹ Elevated risk of intraoperative awareness is associated with a prior history of intraoperative awareness, morbid obesity, substance abuse, chronic pain patients with opioid tolerance, and certain procedures (eg, trauma surgery). Brain function monitoring should thus be used on a case-by-case basis.

Near-infrared spectroscopy (NIRS) for measurement of cerebral oximetry is gaining increasing popularity for monitoring the oxygen supply-demand ratio of the brain. The cerebral tissue saturations (derived from the outer cortex of the frontal lobes) reflect mainly the venous component, continuously, in real time. Thus, it can be viewed as a correlate of the jugular bulb saturation, with a constant bias towards higher readings because of the approximately 25% to 30% admixture of arterial blood in cerebral tissue.⁶² Variables affecting cerebral saturations include cardiac output, cerebral perfusion pressure, oxygen-carrying capacity, PaCO₂, anesthetic depth, and temperature. Technical problems, such as cardiopulmonary bypass cannula malpositioning in the superior vena cava or aorta, could lead to acute changes. Trend monitoring, compared with baseline values obtained during hemodynamically stable conditions, as well as absolute lower thresholds, have been recommended in clinical practice for decision making. Although there is increasing evidence that low cerebral saturations are associated with adverse outcomes, few studies show that interventions based on cerebral oximetry monitoring improve outcome. Murkin and colleagues monitored cerebral oximetry and targeted therapy to optimize cerebral tissue oxygenation in patients undergoing coronary artery bypass grafting.⁶³ They found that a composite measure of postoperative organ dysfunction was substantially improved in the treatment group where goal-directed therapy aimed to maintain the cerebral tissue oxygen saturation values within 25% of the pre-induction baseline. One of the strengths of cerebral oximetry monitoring is the detection of catastrophic events, particularly during periods of nonpulsatile flow, such as during cardiopulmonary bypass, with certain left ventricular assist devices, or extracorporeal membrane oxygenation.^64,65,66 Additional clinical settings where this technology is increasingly deployed include monitoring collateral cerebral blood flow during carotid endarterectomy,⁶⁷ ventricular tachycardia ablation in the electrophysiology lab,⁶⁸ and real-time assessment of advanced cardiac life support/resuscitation efforts.⁶⁹ Cerebral oximetry combined with invasive blood pressure monitoring has been used to determine cerebral autoregulation blood pressure limits in real time, allowing individualized blood pressure treatment.^70,71

Emergence from anesthesia is frequently accompanied by hypertension and tachycardia, which may result from incomplete analgesia, awakening with airway devices in the oropharynx or trachea, withdrawal from antihypertensive drugs, hypoxemia, hypercarbia, delirium, or bladder distension. Patients with multiple risk factors are particularly prone to postoperative complications, such as myocardial infarction, pulmonary edema, malignant ventricular arrhythmia, and cardiac death.⁷² Inadequate pain control and shivering both cause tachycardia and high catecholamine levels, resulting in increased oxygen demand. Hypercoagulability from immobilization and surgery, discontinuation of antiplatelet therapy in patients with intracoronary stents, hypovolemia, anemia, and intravascular volume shifts all probably contribute to the high incidence of complications in the postoperative period.

The anesthesiologist plays a vital role in the postoperative period, providing adequate pain control through the development of multidisciplinary pain services that administer epidural analgesia, peripheral nerve blockade infusion catheters, patient-controlled analgesia, and multimodal therapies. Additionally, hemodynamic parameters must be optimized, especially in patients at increased risk for MACE. If tachycardia, hypertension, or hypotension persists, and an underlying modifiable cause is not identified, then intravenous drugs, such as nitroglycerin, labetalol, or esmolol, can be used to control hemodynamics in patients with cardiovascular disease. Shivering is another phenomenon that may occur as a result of hypothermia or emergence from volatile anesthetics. Although the mechanism is unknown, low doses of intravenous meperidine may decrease or eliminate shivering.

The optimal perioperative care of patients with cardiovascular disease is the joint responsibility of anesthesiologists, surgeons, and cardiologist/internists. Any anesthetic agent or technique has the potential for producing adverse effects, and the margin of safety is reduced in patients with cardiovascular disease. It is the anesthesiologist’s role to acquire accurate and relevant information from the preoperative evaluation, to apply appropriate monitoring technology, to select an anesthetic technique that is suited to the planned procedure and the condition of the patient, and to manage hemodynamic alterations and analgesic requirements in the perioperative period. As cardiovascular disease continues to become more prevalent in the surgical population and preoperative testing and intraoperative monitoring become more sophisticated, the need for effective communication between the specialties of cardiology and anesthesiology will become even more important.