Chapter 57. Regional Anesthesia & Cardiovascular Disease

The decision to use regional anesthesia can be a complex medical choice. Preexisting medical conditions, type of surgery, anesthetic risks, and patient characteristics all may have a profound impact on anesthetic choice and perioperative management. In patients with cardiovascular disease, regional anesthesia techniques (either alone or in conjunction with general anesthesia) can offer the potential perioperative benefits of stress response attenuation, cardiac sympathectomy, earlier extubation, shorter hospital stay, and intense postoperative analgesia.

Neuraxial anesthesia has been shown to be useful in treating patients with coronary artery disease. This includes treatment of anginal symptoms, primary (or part of a combined) anesthetic for the surgical procedure, and acute postoperative pain management. Its use in the high-risk patient during noncardiac surgery can offer reduced blood loss and need for transfusion and a decreased incidence of thromboembolic events.

Regional anesthetic options are not limited to neuraxial techniques when dealing with patients with cardiovascular disease. Paravertebral block has also been used as an adjunct to general anesthesia in the management of patients undergoing cardiac surgery. In addition, intercostal nerve blockade and parasternal block can be used for postoperative pain relief in patients after cardiac surgery. Cervical plexus block can be used in the anesthetic management of patients undergoing carotid endarterectomy. Lower extremity blocks, such as combined sciatic–femoral nerve block can also be used in high-risk patients in whom even modest alterations in hemodynamics would not be tolerated.

Local anesthetics alone or in combination with narcotics are used for regional anesthesia in patients with cardiovascular disease. Ropivacaine is a commonly used local anesthetic during neuraxial anesthesia for cardiac surgery. Although its pharmacologic properties, including onset of action and duration of action, are essentially the same as those of bupivacaine, ropivacaine possesses less cardiotoxic properties, produces less central nervous system depression, and is associated with less motor blockade. Ropivacaine is often combined with sufentanil when used during epidural administration for cardiac surgery. Intrathecal narcotics in combination with general anesthesia can provide intraoperative and postoperative analgesia. The most common undesirable effects of intrathecal opioids are respiratory depression, nausea and vomiting, pruritus, and urinary retention. However, advances in narcotic formulations may allow for sustained-release delivery and target-specific affinity, which may reduce the potential of common narcotic side effects. Standard local anesthetic preparations are used in performing upper and lower extremity blocks in patients with cardiovascular disease.

The most severe complication from neuraxial anesthesia is epidural hematoma formation. The incidence of hematoma formation ranges from 1:150,000 after epidural instrumentation to 1:220,000 for intrathecal instrumentation. Risk of hematoma formation after either technique is increased if performed before systemic heparinization. Currently, only a minority of practicing anesthesiologists use neuraxial blockade in the management of patients undergoing cardiac surgery. Patients presenting for major noncardiac surgery often receive anticoagulation to prevent venous thrombosis and pulmonary embolism. It appears that anesthesiologists more frequently perform neuraxial blockade on these patients after weighing the risk of hematoma formation against the benefits.

Physiologic Effects on the Cardiovascular System

Thoracic epidural anesthesia (TEA) blocks the cardiac afferent and efferent sympathetic fibers with loss of chronotropic and inotropic drive to the myocardium.1 The level of sympathetic blockade that follows a TEA depends in part on the degree of sympathetic tone before the block. This may explain why different studies report different effects of TEA on the cardiovascular system. Goertz et al.2 used transesophageal echocardiography to assess the effects of TEA on cardiac function in healthy volunteers. TEA was not shown to produce significant changes in systolic or diastolic arterial pressures, heart rate, left ventricular end-systolic and end-diastolic cross-sectional areas and left ventricular wall stress as measured under general anesthesia. However, left ventricular maximum elastance, as a measure of left ventricular contractility, was significantly reduced. This observation led the investigators to conclude that TEA severely alters left ventricular contractility even in healthy subjects without preexisting cardiac disease. In another study of healthy subjects, left ventricular ejection and diastolic filling performance were unchanged, but a decrease in cardiac output and fractional area shortening were observed3 (Table 57–1).

Ottesen4 reported that TEA does not affect oxygen consumption (Vo2) during exercise. However, even at moderate workloads, systemic arterial blood pressures were significantly lower with TEA than during control exercise in healthy subjects. No other changes in systemic or pulmonary circulatory parameters were observed. In another exercise study, Wattwil et al.5 reported that after administration of TEA, heart rate, systolic blood pressure, stroke volume, and cardiac output decreased. They also injected local anesthetic (0.5% bupivacaine) intramuscularly and observed similar cardiovascular effects, leading them to the conclusion that the changes may be due in part to systemic effects.

Several studies have documented the effects of TEA on cardiovascular function in patients with heart disease. In a small study of 10 patients scheduled for thoracotomy, a TEA with a mean analgesic level of C7 to T5 had only minor effects on the cardiovascular system.6 In patients with severe coronary artery disease and unstable angina pectoris, Blomberg et al.7 observed that TEA relieved chest pain. It also significantly decreased heart rate, systolic arterial, and pulmonary arterial and pulmonary capillary wedge pressures without any significant changes in coronary perfusion pressure, cardiac output, stroke volume, and systemic or pulmonary vascular resistances.

Intraoperatively, Reinhart et al.8 observed lower cardiac index and oxygen delivery (Qo2) in patients receiving TEA and general anesthesia than in those receiving general anesthesia alone; Vo2 were similar. They also reported that the oxygen supply-demand ratio (Qo2/Vo2) was less in the TEA group throughout the perioperative period and about 30% below baseline values during early recovery. The authors attributed the reduced adaptation of cardiac output to tissue oxygen needs during TEA to negative inotropic and chronotropic effects of sympathetic blockade. In patients on chronic β-adrenergic blocking medication, TEA has been reported to induce a moderate decrease in mean arterial pressure and coronary perfusion pressure, but without producing clinically significant cardiovascular effects.9

Berendes et al.10 reported that in patients undergoing coronary artery bypass grafting (CABG) with combined TEA and general anesthesia, regional left ventricular function was significantly improved, and cardiac troponin I concentrations were reduced when compared with patients receiving general anesthesia alone (Figure 57–1). In addition, TEA reduced the concentrations of atrial and brain natriuretic peptides. The authors concluded that cardiac sympathectomy by TEA improves regional left ventricular function and reduces postoperative ischemia after CABG. In another study of CABG patients, TEA was associated with better hemodynamic stability before and after cardiopulmonary bypass when compared with general anesthesia.11 In addition, TEA may provide improvement in pulmonary function, possibly because of more profound postoperative analgesia after cardiac surgery.12

Cardiac function was also evaluated in patients with two- or three-vessel coronary artery disease who were treated with β-adrenergic blocking agents. Systolic and diastolic arterial pressures, heart rate, and global and regional ejection fractions using equilibrium radionuclide angiography were measured at rest and during maximal exercise before and after TEA. During TEA exercise, systolic arterial pressure, diastolic arterial pressure, and the rate–pressure product, but not heart rate, were significantly lower when compared with control exercise. The global and anterolateral ejection fractions were significantly higher, and the regional wall motion score was significantly lower during TEA exercise than during control exercise. ST-segment depression was significantly lower during TEA exercise.13 Schmidt et al.14 noted that in patients with coronary artery disease, TEA induced a significant improvement in left ventricular diastolic function, whereas indices of systolic function did not change (Figures 57-2, 57-3, and 57-4).

Patients with cardiac disease are prone to hemodynamic changes during laryngoscopy and intubation, thus placing them at risk for ischemic events. TEA has been associated with stable hemodynamics and preservation of baroreflex sensitivity, suggesting withdrawal of vagal activity. Licker et al.15 reported that patients who received TEA in addition to general anesthesia had smaller increases in mean arterial pressure and heart rate during laryngoscopy and tracheal intubation than those who received general anesthesia only; this would suggest that TEA affords hemodynamic protection during these maneuvers.

Management of Cardiac Disease

TEA has been used in the management of cardiac disease (Figure 57–5). Cardiac sympathetic blockade by TEA dilates stenotic coronary arteries and has been used to control pain in patients with unstable angina.16 The mechanism by which TEA reduces angina during long-term treatment is unclear. It has been reported that the improvement in symptoms probably results partly from an anesthetic effect and does not appear to be related to a change in myocardial blood flow or a reduction in stress-induced ischemia. In addition, new myocardial infarctions are not masked or missed in patients receiving TEA for symptomatic treatment of angina.17 The effects of long-term, home self-treatment with TEA on angina, quality of life, and safety were studied by Richter et al.18 Thirty-seven patients with refractory angina began treatment with TEA, using a subcutaneously tunneled epidural catheter and a bupivacaine solution. All but one of the patients improved symptomatically. The improvement was maintained throughout the treatment period (4 days to 3 years). The authors reported that the frequency of anginal attacks and nitroglycerin intake decreased, whereas the overall self-rated quality of life assessed by visual analog scale (VAS) increased.

Studies have looked at the number of anginal attacks and the severity of myocardial ischemia assessed by 48-hour ambulatory Holter monitoring in patients with severe, refractory unstable angina receiving TEA or standard antianginal therapy. Olausson et al.19 reported that the incidence of myocardial ischemia was lower in the TEA group (22% versus 61%; P < .05). The number of ischemic episodes per patient was 1.0 ± 0.6 in the TEA group and 3.6 ± 0.9 in the control group (P < .05), and the episode duration per patient was 4.1 ± 2.5 minutes and 19.7 ± 6.2 minutes in the TEA and the control groups, respectively (P<.05).

Cardiac Surgery

Coronary Artery Bypass Grafting

As an anesthetic for CABG, TEA offers thoracic and cardiac sympathectomy, attenuation of the stress response, analgesia for surgery, and postoperative analgesia, which may improve outcome after CABG. One review suggests that for patients undergoing CABG surgery, the risk:benefit ratio is in favor of epidural and spinal anesthesia, provided no specific contraindications exist and the guidelines for the use of regional techniques in cardiac surgery are followed.20 Patients managed with regional anesthetic techniques seem to benefit from superior postoperative analgesia, shorter postoperative ventilation, reduced incidence of supraventricular arrhythmias, and lower rates of perioperative myocardial infarction. The results of this particular analysis suggest that for each episode of neurologic complication that might occur, 20 myocardial infarctions and 76 episodes of atrial fibrillation would be prevented. Thus, regional anesthesia and analgesia would be considered an effective strategy that improves perioperative morbidity.

As an adjunct to general anesthesia, TEA can be useful in off-pump CABG. Salvi et al.21 reported on 106 patients receiving TEA combined with sevoflurane general anesthesia. The mean time to extubation was 4.6 ± 2.9 hours. VAS scores for pain during the first 24-hour period were less than 2 in all patients. The average intensive care unit (ICU) stay was 1.5 ± 0.8 days. Incidences of perioperative myocardial infarction, myocardial ischemia, and atrial fibrillation were 2.8%, 7.5%, and 10.6%, respectively. Two patients died—one from multiorgan failure and the other from myocardial infarction. Heart rate, mean arterial pressure, cardiac index, and systemic vascular resistance were not affected by TEA alone. Mean arterial pressure and cardiac index decreased (P < .05) when general anesthesia was induced and remained stable thereafter. Neither heart rate nor systemic vascular resistance changed from baseline during the operation. The authors concluded that thoracic epidural block as an adjunct to general anesthesia is a feasible technique in off-pump CABG. It induces intense postoperative analgesia and does not compromise central hemodynamics.

Kessler et al.22 compared TEA alone (ropivacaine plus sufentanil, n = 30), TEA combined with general anesthesia (n = 30), and general anesthesia alone in patients scheduled for off-pump CABG surgery. The general anesthetic consisted of propofol, remifentanil, and cisatracurium. Intraoperative heart rate decreased significantly with TEA alone or in combination with general anesthesia. None of the patients with TEA alone was admitted to the ICU; all were monitored in the intermediate care unit for an average of 6 hours postoperatively and were allowed to eat and drink as desired on admission. Postoperative pain scores were lower in both groups with TEA. There were no differences among groups in overall patient satisfaction. The authors concluded that general anesthesia + TEA appeared to be the most comprehensive anesthetic, allowing for revascularization of any coronary artery and providing good hemodynamic stability and reliable postoperative pain relief.

Stritesky et al.23 reported on 129 patients undergoing cardiac surgery awake with spontaneous ventilation using TEA for anesthesia. Ninety patients underwent on-pump surgery and 39 underwent off-pump surgery. A thoracic epidural block was placed 1 hour before skin incision at the T2-T4 level. Forty-two cases had aortic valve replacement, 32 patients underwent on-pump CABG, 12 had off-pump CABG, 12 had mitral valve replacement, 27 underwent sternal wound reexploration, and 4 had combined procedures. There were 10 conversions to general anesthesia and no deaths. Mean duration of stay in the ICU was 7.2 hours; in-hospital stay 5.1 days. Low cardiac output syndrome, stroke, renal insufficiency, and pulmonary dysfunction were not observed in patients who underwent TEA. Less postoperative pain was also demonstrated.

The authors felt that TEA provided rapid recovery and early outpatient care of patients after cardiac surgery and that TEA would be beneficial for patients with preoperative pulmonary dysfunction.

Several other studies have shown TEA to be effective for CABG surgery in the awake patient24–26 (Figure 57–6). Kessler et al.27 reported on the feasibility and complications of TEA as the sole anesthetic in 20 patients undergoing beating-heart arterial revascularization. Minimally invasive direct coronary CABG (MIDCAB) via partial lower sternotomy was used in 10 patients with single-vessel disease, whereas complete median sternotomy with off-pump coronary artery bypass grafting (OPCAB) was chosen for 10 patients with multivessel disease. An epidural catheter was inserted at the T1-2 or T2-3 level. An epidural infusion of ropivacaine 0.5% and sufentanil 1.66 mcg/mL was started to establish anesthetic levels at C5-6 for OPCAB and at T1-2 for MIDCAB. Nine OPCAB and eight MIDCAB procedures were completed while the patients were awake and spontaneously breathing during the entire procedure. Three patients required intraoperative conversion to GA because of surgical pneumothorax (OPCAB), insufficient anesthesia, or phrenic nerve palsy (both MIDCAB).

In both groups, the heart rate decreased significantly (P < .05) by 10–15% during the procedure. Compared with baseline, mean arterial blood pressure was decreased significantly only during coronary anastomosis. Paco2 increased from 42 ± 2 mm Hg to 46 ± 7 mm Hg (P < .05) throughout the perioperative course during OPCAB, whereas it remained almost unaltered during MIDCAB procedures. All patients rated TEA as “good” or “excellent” and reported a high degree of satisfaction with the procedure.

Anderson et al.28 found similar results when studying the use of TEA for awake cardiac surgery. He reported on a total of 10 operations including 7 MIDCAB, 2 transmyocardial revascularizations (TMR) and 1 MIDCAB/TMR hybrid. The mean preoperative forced expiratory volume for 1 second (FEV1) was 1.9 liters. Significant intraoperative hypoxia or hypercarbia was not seen. One patient required intubation during the procedure for restlessness not associated with hypoxia. Two others required brief periods of assisted ventilation. All procedures were completed without incident. The mean operating time and length of stay were 70 minutes and 4.7 days, respectively. Postoperative pain control and patient satisfaction were excellent.

Although ischemic damage to the myocardium is inevitable during CABG surgery, the extent of the damage may be influenced by the anesthetic technique used. Barrington et al.29 reported on the effect of TEA on the release of troponin I, time to tracheal extubation, and analgesia during elective CABG surgery. One hundred twenty patients were randomized to general anesthesia or general anesthesia plus TEA. The general anesthesia group received fentanyl (7–15 mcg/kg) and a morphine infusion. The TEA group received fentanyl (5–7 mcg/kg) and an epidural infusion of ropivacaine 0.2% and fentanyl 2 mcg/mL until postoperative day 3. Researchers found no differences in troponin I levels between the study groups. The time to tracheal extubation in the TEA group was 15 minutes (range 10–20) compared with 430 minutes (range 284–590) in the general anesthesia group (P < .0001). Analgesia was improved in the TEA group compared with the general anesthesia group. Mean arterial blood pressure and systemic vascular resistance in the ICU were lower in the TEA group. They concluded that TEA for CABG had no effect on troponin release but improved postoperative analgesia and was associated with a reduced time to extubation (Table 57–2; Figure 57–7). Kendall et al.30 conducted a prospective, randomized study to determine the baseline values of troponin T release after off-pump CABG in 30 patients randomly allocated to receive propofol, isoflurane, or isoflurane plus TEA. All other treatments were standardized. They found that mean troponin T levels at 24 hours were not significantly different between the groups (P = .41).

Loick et al.31 investigated the effects of general anesthesia with TEA or with intravenous clonidine on the stress response and incidence of myocardial ischemia in patients undergoing CABG surgery. Seventy patients scheduled for elective CABG surgery received general anesthesia with sufentanil and propofol. In 25 patients, TEA was induced before general anesthesia and continued during the entire study period. Another 24 patients received intravenous clonidine as a bolus of 4 mcg/kg before the induction of general anesthesia. Clonidine was then infused at a rate of 1 mcg/kg/h during surgery and at 0.2–0.5 mcg/kg/h postoperatively. The control group consisted of 21 patients who underwent general anesthesia as performed routinely. Hemodynamics, plasma epinephrine and norepinephrine, cortisol, troponin T, and other cardiac enzymes were measured pre- and postoperatively. Both TEA and clonidine reduced the postoperative heart rate compared with the control group without jeopardizing cardiac output or perfusion pressure. Plasma epinephrine increased perioperatively in all groups but was significantly lower in the TEA group. Neither TEA nor clonidine affected the increase in plasma cortisol. The release of troponin T was attenuated by TEA. New ST-segment elevation or depression occurred in more than 70% of the control patients but only in 40% of the clonidine group and in 50% of the TEA group. The investigators concluded that TEA, but not intravenous clonidine, combined with general anesthesia for CABG demonstrated a beneficial effect on the perioperative stress response and decreased postoperative myocardial ischemia.

Liu et al.32 recently conducted a meta-analysis of 15 trials that studied the effects of perioperative central neuraxial analgesia on outcome after CABG. The total number of patients was 1178. According to the analysis, TEA does not affect the incidences of mortality (0.7% TEA vs 0.3% general anesthesia) or myocardial infarction (2.3% TEA vs 3.4% general anesthesia). However, TEA does significantly reduce the risk of arrhythmias (odds ratio 0.52), pulmonary complications (odds ratio 0.41), and time to tracheal extubation (by 4.5 hours). TEA reduces analog pain scores at rest by 7.8 mm and with activity by 11.6 mm. The authors concluded that there were no differences in the rates of mortality or myocardial infarction after CABG surgery with central neuraxial analgesia. More rapid tracheal extubation, decreased pulmonary complications and cardiac arrhythmias, and reduced pain scores were, however, benefits of TEA.

Fillinger et al.33 conducted a prospective, randomized, nonblinded clinical trial assessing the effects of anesthetic on recovery from cardiac surgery. Sixty patients scheduled for elective cardiac surgery with cardiopulmonary bypass were randomly assigned to 1 of 2 study groups. One group was to receive general anesthesia during surgery and intravenous opioid analgesia after surgery, whereas the second group received TEA combined with general anesthesia during surgery and epidural analgesia for the first 24 postoperative hours. They found no statistically significant differences in time to tracheal extubation, duration of postoperative ICU stay, duration of postoperative hospitalization, pain control, urinary free cortisol, cardiopulmonary complication rate, or total hospital charges.

The most feared complication of TEA is epidural hematoma. Studies have found that, in the setting of cardiac surgery, following a set of standard safety measures averts the occurrence of symptomatic epidural hematomas. These measures consist of preoperative coagulation tests including aPTT, platelet count, and prothrombin time and the cessation of antiplatelet drugs before surgery.34,35 Scott et al.36 conducted a prospective, randomized, controlled study of the incidence of major organ complications in 420 patients undergoing routine CABG with or without TEA. All patients received a standardized general anesthetic. Patients in the TEA group received TEA for 96 hours. Patients in the general anesthesia group received narcotic analgesia for 72 hours. Both groups received supplementary oral analgesia. New supraventricular arrhythmias occurred in 21 of 206 patients (10.2%) in the TEA group compared with 45 of 202 patients (22.3%) in the general anesthesia group (P = .0012). Pulmonary function (maximal inspiratory lung volume) was better in the TEA group (P < .0001). Extubation was achieved earlier (P < .0001) and with significantly fewer lower respiratory tract infections in the TEA group (TEA = 31 of 206, general anesthesia = 59 of 202; P = .0007). Significantly fewer patients had acute confusion (general anesthesia = 11 of 202, TEA = 3 of 206; P = .031) and acute renal failure (general anesthesia = 14 of 202, TEA = 4 of 206; P = .016) in the TEA group. The incidence of stroke was insignificantly less in the TEA group (general anesthesia = 6 of 202, TEA = 2 of 206; P = .17). No neurologic complications were associated with TEA. The researchers concluded that continuous TEA significantly improved the quality of recovery after CABG surgery compared with general anesthesia and conventional narcotic analgesia.

Turfrey et al.37 reported similar results after performing a retrospective analysis of the perioperative course of 218 patients who underwent routine CABG. All patients received a standardized general anesthesia, using target-controlled infusions of alfentanil and propofol. One hundred patients also received TEA with bupivacaine and clonidine, started before surgery and continued for 5 days after surgery. The remaining 118 patients received a target-controlled infusion of alfentanil for analgesia for the first 24 hours after surgery, followed by intravenous patient-controlled morphine analgesia for a further 48 hours. New arrhythmias requiring treatment occurred in 18% of the TEA group of patients compared with 32% of the general anesthesia group (P = .02). There was also a trend toward a reduced incidence of respiratory complications in the TEA group. The time to tracheal extubation was decreased in the TEA group, with 21% of the patients being extubated immediately after surgery compared with 2% in the general anesthesia group (P < .001). No serious neurologic problems resulted from the use of TEA.

Valvular Surgery

Although most studies on the use of TEA in cardiac surgery have focused on CABG, a few investigations and case reports describe its use in valvular surgery. Hemmerling et al.38 have reported on the feasibility and hemodynamic stability of immediate extubation after simple or combined aortic valve surgery using TEA and general anesthesia. Thirty patients with an ejection fraction of more than 30% undergoing aortic valve surgery were studied prospectively. After initiation of TEA, general anesthesia was induced with fentanyl 2–4 mcg/kg, propofol 1–2 mg/kg, and endotracheal intubation was facilitated by rocuronium. Anesthesia was maintained with sevoflurane titrated according to bispectral index (target = 50). Perioperative analgesia was provided by TEA (bupivacaine 0.125% at 6–14 mL/h). Patients underwent simple aortic valve surgery (n = 17) or combined aortic valve surgery (n = 13) with additional CABG (n = 8), replacement of the ascending aorta (Bentall procedure; n = 4), and repair of open foramen ovale (n = 1). All 30 patients were extubated within 15 minutes after surgery at 36.5°C. There was no need for reintubation. Pain scores were low immediately after surgery and at 6, 24, and 48 hours after surgery. During and up to 6 hours after surgery, there was no significant hemodynamic change due to TEA. Fifteen of the 30 patients needed temporary pacemaker activation. There were no complications related to TEA. The authors concluded that immediate extubation is feasible after aortic valve surgery with TEA and maintenance of hemodynamic stability throughout surgery.

Klokocovnik et al.39 describe a patient who underwent aortic valve replacement through a ministernotomy while awake under TEA. The procedure was not converted to general anesthesia or to a conventional operation and was performed without adverse incidents. The patient was discharged from the hospital on the second postoperative day. There were no complications within 30 days after surgery. Kozian et al.40 report on a tricuspid valve replacement without any adverse events using TEA and balanced general anesthesia.

Congenital Surgery

The effects of combined TEA and general anesthesia on hemodynamic and respiratory variables have been reported in children undergoing cardiac heart surgery. Slin'ko et al.41 report that in 55 patients age 1–14 years, TEA was used in combination with oxygen-air-halothane anesthesia. In one group, lidocaine and fentanyl were used for TEA, and in another clonidine and lidocaine. In a control group, standard intravenous fentanyl-diazepam anesthesia was combined with oxygen-air-halothane anesthesia. In the clonidine-lidocaine group, the endocrine stress response was decreased in comparison with other groups, even without narcotics. Hemodynamics remained stable even in patients with NYHA (New York Heart Association) class III–IV heart disease. These same investigators also report that TEA has been found to be safe and effective for postoperative pain in children after heart surgery.42 In one study, 40 children received epidural analgesia after open-heart surgery. Lidocaine was injected in a dose of 1.5–2 mg/kg every 1.5–2 hours. Controls (n = 16) received intravenous fentanyl + diazepam analgesia. Respiratory response and awakening were significantly earlier (P < .001) in the epidural group. Cooperation with nurses was much better in this group, too. No side effects were observed in the epidural group.

Peterson et al.43 report on a retrospective study describing the results of the use of regional anesthesia in 220 pediatric cardiac operations. They indicate that tracheal extubation in the operating room could be achieved for 89% of the patients. Ninety-five percent of the patients had pain scores less than or equal to 4.0 at all intervals postoperatively. Adverse effects of regional anesthesia included emesis, pruritus, urinary retention, postoperative transient paresthesia, and respiratory depression (Table 57–3). The incidence of peridural hematoma was zero. The rate of adverse effects was lower using the TEA approach compared with various caudal, lumbar epidural, and spinal approaches. Hospital stay was not prolonged because of regional anesthetic complications. In this study, regional anesthesia was found to be safe and effective in the management of pediatric patients undergoing cardiac surgery.

Hammer et al.44 evaluated whether spinal anesthesia or TEA in combination with general anesthesia was associated with circulatory stability, satisfactory postoperative sedation/analgesia, and a low incidence of adverse effects. They found no significant differences in the incidence of clinically significant changes in vital signs, oxygen desaturation, hypercarbia, or vomiting when comparing TEA with spinal anesthesia for children undergoing cardiac surgery.

Noncardiac Surgery in Patients with Cardiovascular Disease

When TEA is used for major vascuar surgery, it is reported to provide more hemodynamic stability and better pain control than general anesthesia or monitored anesthesia care with local anesthesia.45 TEA has been found safe and effective for endovascular aortic aneurysm repair, especially in patients with severe coexisting diseases. In addition, TEA minimizes sedation and postoperative analgesic requirements, decreases cardiopulmonary complications, and decreases overall hospital stay, thereby reducing cost.

Bonnet et al.46 investigated the hemodynamic consequences of abdominal aortic surgery with infrarenal cross-clamping in 21 patients randomized to one of two groups. In group I (n = 11), neuroleptanesthesia was used, whereas group II (n = 10) received TEA at the T8-T9 level. In all patients, hemodynamic measurements were performed using pulmonary artery catheters. The use of TEA was characterized by greater hemodynamic stability during surgery. Patients in the neuroleptanesthesia group experienced significant lability of blood pressure, heart rate, and cardiac index.

Her et al.47 compared intraoperative hemodynamic variables and postoperative morbidity between a group of patients undergoing abdominal aortic surgery—one with combined TEA and general anesthesia (n = 30) and one with general anesthesia alone (n = 19). Patients in the combined group were given epidural bupivacaine intraoperatively and epidural morphine postoperatively. After cross-clamping of the aorta, cardiac index and pulmonary capillary wedge pressure did not change in the combined group, whereas cardiac index decreased (mean change, 0.30 L/min/m2; P = .006) and pulmonary capillary wedge pressure increased (mean change 1 mm Hg; P = .007) in the group with general anesthesia alone. After unclamping of the aorta, cardiac index increased in both groups (mean change, 0.26 L/min/m2, P = .002 and 0.30 L/min/m2 P = .001, respectively). Postoperatively, the necessity for ventilatory support and the incidence of respiratory failure were lower in the combined group than in the general anesthesia alone group (P =.0002 and P= .018, respectively). In addition, vasodilator therapy was required less frequently in the combined group (P = .002). Duration of ICU stay was shorter in the combined group (2.7 days vs 3.8 days; P = .003). The authors concluded that for infrarenal abdominal aortic surgery, combined TEA and general anesthesia is associated with more stable intraoperative hemodynamics and significantly less postoperative morbidity than general anesthesia alone.

Different results, however, were obtained by Garnett et al.,48 who studied the incidence of perioperative myocardial ischemia in patients undergoing elective aortic surgery. Patients were randomly assigned to one of two groups. One group (n = 48) received combined general anesthesia and epidural anesthesia and postoperative epidural analgesia for 48 hours; the other group (n = 51) received general anesthesia followed by postoperative intravenous analgesia. The authors reported that myocardial ischemia was common because it occurred in 55% of patients. In the hospital, preoperative ischemia was uncommon (combined = 8; general anesthesia = 3). Ischemic events were common intraoperatively (combined = 25; general anesthesia = 18), with mesenteric traction producing the largest number of events (combined = 11; general anesthesia = 11). Postoperative ischemia was most common on the day of surgery. Termination of epidural analgesia produced rebound ischemia (60 events in 9 patients). The authors concluded that combined general anesthesia and epidural anesthesia and postoperative epidural analgesia do not reduce the incidence of myocardial ischemia or morbidity compared with general anesthesia and postoperative intravenous analgesia.

Norris et al.49 studied patient outcomes in 168 patients undergoing surgery of the abdominal aorta using different types of anesthesia. Patients were randomly assigned to receive either TEA combined with general anesthesia or general anesthesia alone intraoperatively and either intravenous or epidural patient-controlled analgesia (PCA) postoperatively (four treatment groups). PCA was continued for at least 72 hours. Length of stay and direct medical costs for patients surviving to discharge were similar among the four treatment groups. Postoperative outcomes were also similar among the groups with respect to death, myocardial infarction, myocardial ischemia, reoperation, pneumonia, and renal failure. Postoperative pain scores were the same for the four groups, but epidural PCA was associated with a significantly shorter time to extubation (P = .002). Times to ICU discharge, ward admission, first bowel sounds, first flatus, tolerating clear liquids, tolerating regular diet, and independent ambulation were also equivalent for the four groups. The authors concluded that in patients undergoing surgery of the abdominal aorta, TEA combined with general anesthesia and followed by either intravenous or epidural PCA, offers no major advantage or disadvantage when compared with general anesthesia alone followed by either intravenous or epidural PCA.

Davies et al.50 prospectively studied intraoperative hemodynamics and outcomes in 50 patients undergoing elective abdominal aortic surgery who were randomized to receive either combined epidural (T9-T10 level) and general anesthesia and postoperative epidural analgesia or general anesthesia and postoperative intravenous morphine infusion. The use of intraoperative vasopressors was significantly higher in the combined group (P < .01), but the use of intravenous glyceryl trinitrate was significantly lower (P < .01). No significant difference was found between groups in regard to blood loss, volume replacement, and the number of patients requiring postoperative ventilation. Two patients in the combined group died postoperatively compared with one in the general anesthesia group (nonsignificant). There was no significant difference between groups in the total number or type of postoperative complications. The authors concluded that combined epidural anesthesia with general anesthesia altered intraoperative cardiovascular management but did not affect postoperative outcome.

Gelman et al.51 studied the effect of TEA on the cardiovascular function of morbidly obese patients undergoing gastric bypass surgery. Patients were given general anesthesia alone or a combination of TEA and general anesthesia. Circulatory function was measured and calculated using radial artery cannulation and pulmonary artery catheterization with pulmonary artery thermodilution catheters. During surgery, the TEA group demonstrated greater decreases in cardiac index, left and right ventricular stroke work, systolic blood pressure–heart rate product, arteriovenous oxygen content difference, oxygen consumption, and intrapulmonary shunt compared with the general anesthesia group. Postoperatively, epidural analgesia was associated with decreases in left ventricular stroke work, systolic pressure–heart rate product, arteriovenous oxygen content differences, and oxygen consumption compared with values observed when patients experienced pain. Morphine given for relief of postoperative pain was not associated with significant changes in cardiovascular function. The authors concluded that continuous TEA used for upper abdominal surgery in morbidly obese patients benefits intraoperative cardiovascular function, as reflected by a decrease in left ventricular stroke work, and postoperatively by relief of pain.

Acute Pain Management

In patients with cardiac disease undergoing cardiac surgery, good pain management is an important goal to improve outcomes and reduce postoperative complications. The value of TEA in this setting has been studied extensively. Royse et al.52 studied 80 patients who were randomized to TEA or intravenous morphine analgesia for postoperative pain control after CABG with cardiopulmonary bypass. A thoracic epidural catheter was inserted the night before surgery at either the T1-T2 or T2-T3 level. Eight milliliters of 0.5% ropivacaine with 20 mcg fentanyl was administered before induction of anesthesia. Ropivacaine 0.2% with 2 mcg/mL fentanyl was then infused at a rate of 5–14 mL/h to attain a sensory block of T1 to T10. Pain was measured using a VAS scale from 0 to 10. Psychological morbidity, intraoperative hemodynamics, ventricular function, lung function, and physiotherapy cooperation were also assessed. On the third postoperative day, TEA and morphine were stopped, and only oral medications were used. Acetaminophen, indomethacin, and tramadol were allowed as supplemental analgesics in both groups. Pain scores were significantly less with TEA on postoperative days 1 and 2 at rest and with coughing. When TEA and morphine were stopped on day 3, there were no significant differences. Secondary endpoints of postoperative depression and posttraumatic stress subscales of the Minnesota Multiphasic Personality Inventory were lower with TEA. In addition, extubation occurred earlier with TEA (2.6 vs 5.4 hours; P < .001). TEA showed improved physiotherapy cooperation (P < .001), arterial oxygen tension (P = .041), and peak expiratory flow rate (P = .001). Mean arterial pressure was lower with TEA (P = .036); otherwise, no differences were found in intraoperative hemodynamics or ventricular function.

Liem et al.53 studied the effects of intraoperative and postoperative epidural pain management during and after CABG on the recovery time, postoperative pulmonary and cardiac parameters, VAS scores, and sedation scores. They compared the findings with those of patients anesthetized with general anesthesia whose postoperative pain was relieved with intermittent intravenous administration of nicomorphine. Fifty-four patients were studied postoperatively after uncomplicated CABG surgery. In the TEA group (n = 27), intraoperative analgesia was based on TEA in combination with general anesthesia. In the general anesthesia group (n = 27), intravenous anesthesia with high-dose sufentanil and midazolam was used. Postoperative pain management in the general anesthesia group consisted of intermittent intravenous administration of nicomorphine 0.1 mg/kg every 6 hours, whereas in the TEA group patients received a continuous high TEA with 0.125% bupivacaine plus sufentanil. Patients in the TEA group awakened earlier (148 minutes vs 335 minutes), resumed spontaneous respiration earlier (326 minutes vs 982 minutes), and were extubated earlier (463 minutes vs 1140 minutes). VAS scores, sedation scores, and postoperative Pao2 were significantly (P ≤ .01) better in the TEA group. The incidence of tachycardia (15 vs 2 patients) and postoperative myocardial ischemia (12 vs 4 patients) was higher in the general anesthesia group. The authors concluded that intraoperative and postoperative pain treatment with epidurally administered bupivacaine plus sufentanil improved the recovery time, as well as pulmonary and cardiac outcomes after CABG, when compared with intravenous postoperative pain treatment after intraoperative general anesthesia with sufentanil and midazolam.

Hemmerling et al.54 studied 100 consecutive patients undergoing OPCAB surgery to examine the feasibility of immediate extubation after using opioid-based analgesia or TEA and compare postoperative analgesia between continuous TEA versus PCA. Perioperative analgesia was provided by TEA (n = 63) using bupivacaine 0.125% at a continuous rate of 8–14 mL/h and repetitive boluses of bupivacaine 0.25% during surgery. In the other group (n = 37), perioperative analgesia was achieved by intravenous fentanyl boluses (up to 15 mcg/kg) and remifentanil 0.1–0.2 mcg/kg/min, followed by morphine PCA after surgery. Ninety-five patients were extubated within 25 minutes after surgery (TEA n = 62; PCA n = 33). Five patients were not extubated immediately because their core temperature was lower than 35°C. One patient was reintubated because of agitation (TEA group); one was reintubated because of severe pain and morphine-induced respiratory depression (PCA group). Pain scores were generally low after surgery, with pain scores in the TEA group being significantly lower immediately, at 6 hours, 24 hours, and 48 hours after surgery (P < .05) (Figure 57–8). The authors concluded that immediate extubation is possible after OPCAB surgery using either opioid-based analgesia or TEA, but TEA provides significantly lower pain scores after surgery compared with morphine PCA.

However, when Bois et al.55 studied 124 patients to assess the role of postoperative analgesia on myocardial ischemia after aortic surgery using intravenous PCA or TEA, different results were obtained. In the PCA group, a bolus of morphine, 0.05 mg/kg, was given, followed by 0.02 mg/kg of morphine on demand every 10 minutes. Bupivacaine 0.125% and fentanyl 10 mcg/mL were used in the TEA group. Analgesics were titrated to maintain a VAS score ≤ 3. The overall incidence of myocardial ischemia was 18.4–18.2% for TEA and 18.6% for PCA (P = not significant). There were no differences between the groups in the total duration of ischemia per patient (22.2 ± 119.8 minutes for TEA and 20.5 ± 99 minutes for PCA) and the number of episodes per patient (0.69 ± 2.1 for TEA and 1.2 ± 4.9 for PCA). Twenty-three patients had an adverse cardiac outcome, but there were no differences between the groups. Although the postoperative pain control was superior with TEA, its use did not result in a lower incidence of early myocardial ischemia when compared with intravenous PCA with morphine.

In patients who underwent CABG, a comparative audit of the use of TEA versus intravenous opioids for postoperative pain control showed no significant differences in the frequency or intensity of persistent pain (defined as pain still present 2 or more months after surgery).56 Similarly in patients undergoing cardiac valve replacement, TEA was shown to provide excellent analgesia in the peri- and postoperative period, but did not offer a protective effect on chronic poststernotomy pain.57

Physiologic Effects on the Cardiovascular System

The influence of lumbar epidural anesthesia (LEA) without cardiac sympathectomy on global and regional left ventricular function was investigated before surgery in healthy subjects and in patients suffering from stable mild effort-related angina.58 In both groups, epidural blockade was performed with 10 mL of 0.5% bupivacaine. Radionuclide angiography was used to determine cardiac output, left ventricular ejection fraction, and end-systolic and end-diastolic volumes and to analyze left ventricular wall motion. Throughout the procedure, patients with a history of angina exhibited neither chest pain nor ECG evidence of myocardial ischemia. At control, left ventricular ejection fraction and systolic pressure-volume ratio were lower in the patients with angina. These patients also had evidence of regional left ventricular dysfunction. Epidural blockade without volume loading resulted in slight improvements in left ventricular ejection fraction and regional function. Such changes were not observed in normal patients. After volume loading, the improvements in ventricular function subsided. These observations led the authors to conclude that LEA may improve global and regional ventricular function in patients with angina provided that volume loading is limited.

Another study reported that LEA enhances cardiac vagal tone mainly through a decrease in venous return.59 In hypertensive patients, LEA has been shown to cause decreases in mean arterial pressure, with associated decreases in systemic vascular resistance and cardiac output.60

Noncardiac Surgery

Perioperative myocardial ischemia occurs frequently in elderly patients undergoing hip fracture surgery. Matot et al.61 prospectively studied 68 patients with known coronary artery disease or those at high risk for coronary artery disease undergoing hip fracture surgery. On admission to the emergency room, patients were assigned to receive a usual care analgesic regimen (intramuscular meperidine control group, n = 34) or continuous epidural infusion of local anesthetic and opioid (epidural group, n = 34). Preoperative adverse cardiac events were significantly more prevalent in the control group compared with the epidural group (7 of 34 vs 0 of 34; P = .01). Adverse cardiac events included fatal myocardial infarction in three patients, fatal congestive heart failure in one, nonfatal congestive heart failure in one, and new-onset atrial fibrillation in two (Table 57–4). The incidence of intraoperative and postoperative adverse cardiac events was similar for the two groups. The significant difference between groups in the incidence of preoperative cardiac events prompted interruption of the study after the planned interim analysis. The authors concluded that compared with conventional analgesia, early administration of continuous epidural analgesia is associated with a lower incidence of preoperative adverse cardiac events in elderly patients with hip fracture who have or are at risk for coronary artery disease.

Scheinin et al.62 also looked at the impact of analgesic management on the incidence and severity of cardiac ischemia in at-risk patients. Seventy-seven elderly patients undergoing surgical treatment of traumatic hip fractures were randomized to conventional analgesic regimen (intramuscular oxycodone, OPI group) or continuous epidural infusion of bupivacaine/fentanyl (LEA group). The analgesic regimens were started preoperatively. Patients were operated under spinal anesthesia, and the treatments were continued 3 days postoperatively. The ECG was continuously recorded by Holter monitor. Any ST-segment depression of ≥0.1 mV or elevation of ≥0.2 mV lasting 1 minute or more was considered an ischemic episode. Nocturnal arterial oxygen saturation (Sao2) was recorded perioperatively, and subjective pain was assessed every morning using a VAS scale. Fifty-nine (OPI 30; LEA 29) patients were evaluated for efficacy. Thirteen patients (43%) in the OPI and 12 patients (41%) in the LEA group had ischemic episodes; the difference was not significant. However, significantly more patients in the OPI group had ischemic episodes during the surgery (8 vs 0 in the LEA group; P = .005). The median total ischemic burden (ie, the integral of ST change over time) in patients with ischemic episodes was 10 times larger in the OPI group (340 mm × minutes) compared with the LEA group (30 mm × minutes) (P = .002) (Table 57–5). There were no significant differences between the groups in average heart rates or in heart rates at the start of ischemic episodes or in maximal heart rates during the attacks. Average nocturnal Sao2 was similar in the two groups, and there were no differences in the number of hypoxemic (Sao2 < 90%) episodes. Preoperatively, no differences were reported of subjective pain, but postoperative and average perioperative VAS scores were almost 40% lower in the LEA group (P = .006). Perioperative myocardial infarctions were not detected. The authors concluded that a continuous epidural bupivacaine/fentanyl analgesic regimen that is started preoperatively reduces perioperative pain and the amount of myocardial ischemia in elderly patients with hip fracture.

Perler et al.63 evaluated the records of 78 consecutive patients undergoing elective femoropopliteal or femorotibial bypass grafts to assess the impact of the anesthetic management on graft patency. They were randomized to receive general anesthesia and postoperative PCA (general anesthesia, n = 41), or epidural anesthesia and postoperative continuous epidural analgesia (LEA, n = 37). Graft occlusion occurred in 11 cases within the first 7 postoperative days, including 9 general anesthesia and 2 LEA patients (P < .05). There were two femoropopliteal occlusions—both in patients who had received general anesthesia. Nine femorotibial occlusions were observed; 7 were in patients receiving general anesthesia and 2 in LEA patients. Graft occlusion occurred in 11 of the 64 limb salvage cases, including 9 general anesthesia and 2 LEA cases (P < .05). Seven of 55 greater saphenous vein grafts occluded; 6 were in patients who had general anesthesia versus 1 LEA patient (P < .05).

Another study compared the effects of general anesthesia with postoperative PCA and LEA with epidural PCA on hemostatic function and postoperative arterial thrombotic complications in patients undergoing elective lower extremity vascular reconstruction.64 The findings indicate that LEA may decrease the risk of arterial thrombotic complications in patients undergoing lower extremity revascularization.

Bode et al.65 studied the effects of general anesthesia versus LEA on cardiac outcome in patients undergoing peripheral vascular surgery who had a high likelihood of associated coronary artery disease. Four hundred twenty-three patients were randomly assigned to receive general anesthesia (n = 138), LEA (n = 149), or spinal anesthesia (n = 136) for femoral-to-distal artery bypass surgery. All patients were monitored with radial and pulmonary arterial catheters. Postoperatively, patients were in a monitored setting for 48–72 hours and had daily ECGs for 4–5 days and creatine phosphokinase/isoenzymes every 8 hours for 3 days, then daily for 4 days. The recorded cardiac outcomes were myocardial infarction, angina, and congestive heart failure. The authors found that cardiovascular morbidity and overall mortality were not significantly different among groups when analyzed by either intention to treat or type of anesthesia received. In the intention-to-treat analysis, the incidences of cardiac events or death for the general anesthesia, spinal, and LEA groups were 16.7%, 21.3%, and 15.4%, respectively. The absolute risk difference observed between general anesthesia and all regional anesthesia groups for cardiac event or death was –1.6% (95% confidence interval –9.2%, 6.1%). The authors concluded that the choice of anesthesia does not significantly influence cardiac morbidity and overall mortality in patients undergoing peripheral vascular surgery.

Cohen et al.66 investigated postoperative cardiac outcomes in patients with congestive heart failure undergoing femoral-to-distal artery bypass surgery, comparing LEA with general anesthesia. One hundred and six patients with prior or persistent congestive heart failure undergoing femoral-to-distal artery bypass surgery were randomized to general anesthesia (n = 29) or regional anesthesia (LEA n = 42; or spinal anesthesia, n = 3). The primary endpoint was death or adverse cardiac events (myocardial infarction, unstable angina, or congestive heart failure). The authors found no statistically significant difference among groups in the incidence of combined cardiac events, death, myocardial infarction, death or myocardial infarction combined, unstable angina, or congestive heart failure.

To determine whether tight hemodynamic control would decrease intraoperative myocardial ischemia and major postoperative cardiac morbidity, Christopherson et al.67 maintained blood pressure and heart rate within predetermined limits in patients randomized to LEA or general anesthesia for peripheral vascular surgery. One hundred patients undergoing elective lower extremity revascularization for atherosclerotic peripheral vascular disease were randomized to receive either LEA or general anesthesia. Blood pressure and heart rate limits were determined before randomization. Hemodynamic monitoring and management of anesthesia were standardized. Myocardial ischemia and major cardiac morbidity were diagnosed by a “blinded” cardiologist, based on continuous ambulatory ECG monitoring, cardiac enzymes, and 12-lead ECGs. Intraoperative blood pressure and heart rate were analyzed by investigators masked to the type of anesthesia given. The authors found that in a greater percentage of the patients randomized to general anesthesia the intraoperative blood pressures were above the predefined limits when compared with the LEA patients. They also noted that the patients had more rapid changes in heart rate and blood pressure. Intraoperative ischemia and major cardiac morbidity were similar in the two groups. Patients experiencing intraoperative ischemia, regardless of anesthetic type, more frequently had blood pressures more than 10% above their upper limit and/or more rapid heart rate changes compared with patients without ischemia. The investigators concluded that prevention of elevated intraoperative blood pressure and/or rapid changes in blood pressure or heart rate may be more successful with LEA than with general anesthesia. Vital sign abnormalities may occur more frequently in patients who have had intraoperative ischemia or are at risk for having it later in the procedure.

Another study compared outcomes between LEA and general anesthesia in a group of patients with a high risk for cardiac and other morbidity who were undergoing elective vascular reconstruction of the lower extremities.68 The results showed comparable rates of cardiac and most other morbidity between the two groups, but LEA was associated with a lower incidence of reoperation for inadequate tissue perfusion.

Acute Pain Management

The data that support the use of LEA for postoperative pain management in patients with cardiac disease are limited. Mehta et al.69 compared the effect of buprenorphine through the lumbar and thoracic epidural routes for postoperative analgesia after CABG. Forty patients with normal left ventricular ejection fraction scheduled for CABG were randomly divided into two groups: a TEA group (n = 19) and a LEA group (n = 20). For postoperative pain relief, they received epidural 0.15 mg of buprenorphine at the first demand for pain relief after extubation. A top-up dose of 0.15 mg of buprenorphine was administered in patients whose VAS score was higher than 3 at 1 hour after the first dose. Subsequent breakthrough pain was treated with ketorolac 30 mg intramuscularly. Pain, assessed by VAS scores on a 0–10 scale, respiratory rate, FEV1, forced vital capacity, mean arterial blood pressure, cardiac index, Pao2, and Paco2 were measured at frequent intervals. Side effects of epidural opioids were noted. Both groups were comparable in demographic characteristics, had similar VAS scores from 1 to 24 hours postoperatively, required similar amounts of intramuscular ketorolac for break-through pain, and had comparable pulmonary functions and side effects. The authors concluded that buprenorphine given by the lumbar epidural route after CABG compares favorably in terms of quality of analgesia and incidence of side effects with the same drug given via the thoracic epidural route.

Shayevitz et al.70 compared the effects of LEA compared with intravenous opioids for postoperative pain control in children who had undergone cardiac surgery for palliation or repair of congenital heart disease. In the LEA patients (n = 27), epidural catheters were placed after anesthetic induction, but before anticoagulation. A bolus of 50 mcg/kg of preservative-free morphine sulfate was administered through the catheter, followed by a continuous infusion at 3–4 mcg/kg/h for 22–102 (median, 46) hours postoperatively. The intravenous opioid patients received 50 mcg/kg of intravenous fentanyl before incision, followed by a continuous infusion at 0.3 mcg/kg/min. The fentanyl infusion rate was decreased to 0.1 mcg/kg/min postoperatively and maintained for 24 hours. Times to tracheal extubation, transfer from the ICU, and resumption of regular diet were significantly shorter in LEA patients. LEA and intravenous opioid patients received similar amounts of fentanyl during surgery. However, during the postoperative recovery, LEA patients who were extubated late received significantly less supplemental opioid medication than intravenous opioid patients extubated late during the first 5 postoperative days. No complications related to dural puncture, bleeding into the epidural space, or respiratory depression were encountered. Pruritus and nausea/vomiting were the most commonly reported morbidities in both groups. Fifty-six percent of LEA patients and 41% of intravenous opioid patients reported pruritus (P = .4). There was no significant difference in the incidence of nausea and vomiting between the groups (34% vs 30%, respectively). The authors found improved outcomes only in LEA patients extubated late compared with intravenous opioid patients.

Adult Cardiac Surgery

Although some anesthesiologists perform spinal anesthesia in adult patients undergoing cardiac surgery, its use is probably limited by the concern for postheparinization lumbar hematoma. In one survey of 892 cardiac anesthesiologists, only 68 (7.6%) reported using of spinal techniques.71

The hemodynamic response to lumbar spinal anesthesia using hyperbaric bupivacaine or lidocaine with morphine has been evaluated in cardiac surgical patients by Kowalewski et al.72 These investigators observed that induction of general anesthesia produced a decrease in mean arterial pressure. The addition of spinal anesthesia produced a decrease in heart rate. Heart rate and mean arterial pressure did not change with sternotomy. In almost all patients, the arterial blood pressure needed to be supported with phenylephrine at some time during the operation. High-dose intrathecal bupivacaine, when combined with general anesthesia has been shown to result in less β-receptor dysfunction and a lower stress response (serum epinephrine, norepinephrine, and cortisol concentrations) during the CABG procedure than with general anesthesia alone73 (Figure 57–9). However, other studies have reported that although large-dose intrathecal morphine initiates reliable postoperative analgesia, it does not reliably attenuate the stress response during and after cardiac surgery.74,75

Vanstrum et al.76 assessed the use of intrathecal morphine compared with placebo during cardiac surgery on hemodynamics and postoperative pain to determine whether intrathecal morphine is effective in decreasing analgesic and antihypertensive drug requirements after CABG in a prospective, randomized, double-blind study. Approximately 30 minutes before induction of anesthesia with intravenous sufentanil and diazepam and 2 hours before heparinization, one group of patients (n = 16) was given intrathecal morphine 0.5 mg, whereas the control group (n = 14) was given placebo intrathecal injections through 22- or 25-gauge lumbar puncture needles. Intraoperatively, there were no differences in the number of patients requiring vasodilator drugs or those requiring volatile agent titration. During the postoperative period, the treated group required significantly less (P < .05) intravenous morphine compared with the placebo group during the first 24 hours (1.8 ± 0.7 vs 5.4 ± 1.5 mg) and 30 hours (2.4 + 0.8 vs 8.3 ± 1.9 mg). The treated group also required significantly less sodium nitroprusside in the first 24 hours (58.1 ± 29.0 vs 89.1 ± 18.4 mg). There were no differences in pain scores, and the only complications (itching, nausea, and vomiting) were infrequent. It was concluded that an intrathecal dose of 0.5 mg morphine is efficacious in reducing analgesic and antihypertensive drug requirements after CABG.

Bettex et al.77 compared the effects of combined intrathecal morphine and sufentanil with low-dose intravenous sufentanil during propofol anesthesia for fast-track cardiac surgery. Twenty-four consecutive patients with normal cardiopulmonary function who were scheduled for elective cardiac surgery were randomized to receive either a continuous intravenous infusion of sufentanil 0.9–1.8 mcg/kg/min (n = 13) or a single lumbar intrathecal dose of sufentanil 50 mcg and morphine 500 mcg (n = 11). Intrathecal sufentanil and morphine allowed a shorter duration of intubation (104 ± 56.5 vs 213 ± 104 min; P = .01), reduced the need for postoperative analgesia with nicomorphine (equipotent to morphine) (0.7 ± 0.4 vs 1.2 ± 0.4 mg/h; P = .008), and improved postoperative maximal inspiratory capacity (53.4 ± 16.1 vs 38.4 ± 12.5% of the norm; P = .05). The authors concluded that in low-risk patients undergoing CABG or valvular surgery, combined intrathecal sufentanil and morphine with a target-controlled infusion of propofol satisfies the goals of fast-track cardiac surgery. Shorter intubation times and superior postoperative analgesia have also been reported with the intrathecal use of remifentanil, morphine, and clonidine after cardiac surgery.78–80

Shroff et al.81 studied 21 ASA physical status III and IV men scheduled for elective CABG. Patients were randomized to receive intrathecal 10 mcg/kg morphine and 25 mcg fentanyl preoperatively (n = 12) or no intrathecal opioid (n = 9). The latter group received 25–50 mcg/kg fentanyl and 0.05 to 0.1 mg/kg midazolam intraoperatively, whereas the intrathecal opioid group received intravenous fentanyl and midazolam only as needed. Both groups were administered intravenous morphine and midazolam postoperatively as needed. For the first 24 hours postoperatively, pain levels (0 = none to 10 = most severe) and sedation levels (1 = none to 5 = unconscious) were measured hourly. The time to extubation and discharge from the ICU was recorded. ECG evidence of myocardial ischemia was noted. Pain scores were low for both groups (1.5), but the intrathecal opioid subjects exhibited less sedation than the high-dose fentanyl subjects. Extubation time was 12 hours shorter in the intrathecal opioid group (2.9 ± 5.3 vs 14.7 ± 6.8 hours, P = .001). The five subjects with a 1-day ICU stay were all in the intrathecal opioid group (P = .04). The incidence of myocardial ischemia did not differ between the two groups. The authors concluded that intrathecal opioids can facilitate early extubation and discharge from the ICU without compromising analgesia or increasing myocardial ischemia.

Mehta et al.82 studied the effects of preoperatively administered intrathecal morphine sulfate on extubation time and postoperative pulmonary function and postoperative analgesia after OPCAB. One hundred adult patients scheduled for elective primary OPCAB were randomized to preoperative administration of 8 mcg/kg intrathecal morphine sulfate (group 1) with a 25-gauge spinal needle or to receive sterile normal saline placebo subcutaneously (group 2). Anesthetic induction and maintenance were standardized to facilitate early tracheal extubation; patients were extubated in the ICU. Postoperative times to extubation were 9.47 ± 3.83 hours in group 1 versus 11.25 ± 3.94 hours in group 2 (P = .025). Postextubation bedside spirometric lung volumes, expressed as percentage of preoperative lung volume, showed significant differences in group 1 compared with group 2 in forced vital capacity (39.66% ± 15.42% vs 31.85% ± 11.65%; P = .016), FEV1 (44.8% ± 16.18% vs 35.97% ± 13.32%; P = .013), maximum voluntary ventilation (39.40% ± 13.57% vs 33.11% ± 14.80%; P = .056), and expiratory flow rate (47.76% ± 24.61% vs 37.37% ± 4.33%; P = .031). There was no mortality or neurologic complication in either group. They concluded that intrathecal morphine provided superior quality of analgesia, which translated into better maintenance of postoperative lung volume as determined by spirometry.

Another study suggests that intrathecal morphine provides better analgesia after cardiac surgery than conventional intravenous analgesia and that a lower dose (1 mg) is associated with less postoperative depression as assessed by Paco2 measurements.83 In a different study, the use of intrathecal opioids did not, however, facilitate earlier tracheal extubation or improve intraoperative hemodynamic stability compared with intravenous sufentanil and desflurane alone for fast-track cardiac anesthesia.84 Latham et al.84 prospectively studied 40 patients undergoing elective primary CABG, aortic valve replacement, or mitral valve replacement. After a standardized anesthetic induction, anesthesia was maintained with a remifentanil infusion, 0.1 mcg/kg/min and desflurane 3–10%, inspired (group I, n = 20) or a sufentanil infusion 0.3 mcg/kg/h and desflurane 3–10%, inspired (group II, n = 20). Patients receiving remifentanil were administered 8 mcg/kg of intrathecal morphine for postoperative analgesia. Both anesthetic regimens provided comparable intraoperative hemodynamic stability and similar recovery profiles, with extubation times of 5.1 ± 4.3 hours (group I) and 5.8 ± 6.7 hours (group II). The authors concluded that the use of remifentanil in combination with intrathecal morphine did not facilitate earlier tracheal extubation or improve intraoperative hemodynamic stability compared with sufentanil alone for fast-track cardiac anesthesia.

Alhashemi et al.85 studied the effects of low doses of intrathecal morphine on extubation time and on postoperative analgesic requirements after CABG. Fifty patients scheduled for elective primary CABG were randomized to receive placebo, 250 mcg, or 500 mcg of intrathecal morphine preoperatively. Intraoperative fentanyl and midazolam were limited to 15 mcg/kg and 20 mcg/kg intravenously. Patients were extubated in the ICU. There were no significant differences in extubation times for the three groups. Postoperative morphine requirements in the 250 mcg and 500 mcg groups were 13.6 ± 7.8 mg and 11.7 ± 7.4 mg, compared with 21.3 ± 6.2 mg in the placebo group (P = .001). The requirements for postoperative midazolam, nitroglycerin, and sodium nitroprusside were also similar among the three groups. The authors concluded that despite decreased postoperative morphine requirements, intrathecal morphine administration did not have a clinically relevant effect on extubation time after CABG surgery.

Chaney et al.86 found that the mean time from ICU arrival to extubation was significantly prolonged in patients who received intrathecal morphine when compared with patients who received intrathecal placebo. Extubation was substantially delayed because of prolonged ventilatory depression in some patients who had received intrathecal morphine. Although postoperative intravenous morphine use was less in patients who received intrathecal morphine than in those receiving placebo, the difference between the groups was not significant.

The hemodynamic effects of spinal anesthesia in combination with general anesthesia were studied prospectively in infants and children undergoing open-heart surgery by Finkel et al.87 After premedication with midazolam and an inhalation induction with sevoflurane, 30 patients, 7 months to 13 years of age, who were undergoing open-heart surgery, received intrathecal injections with 0.5% tetracaine D10 mixed with morphine. The spinal blocks were placed at the L2-3 or L3-4 interspace, and cephalad spread was promoted by positioning the patients in 30 degrees of Trendelenburg position for a minimum of 10 minutes. Maintenance of anesthesia was with isoflurane 0.2–0.5% in 70% nitrous oxide to maintain heart rate and blood pressure within 20% of postinduction baseline values. Hemodynamic values were recorded at predetermined timed intervals and intraoperative events up to and including aortic cannulation. Patients were divided into four age groups (<1 years, 1–3 years, 4–6 years, and >7 years).

The authors reported that hemodynamic stability was demonstrated in all four age groups. Statistically significant slowing of the heart rate did occur in the groups older than 1 year at 25 minutes, although clinically significant bradycardia requiring treatment never occurred. Hypotension did occur during specific surgical manipulations but recovered spontaneously. Atropine, fluid boluses, and vasopressors were never used. At the conclusion of surgery, all patients met extubation criteria and could move all four extremities.

Pirat et al.88 prospectively studied the cardiovascular and neurohumoral responses during intrathecal or intravenous fentanyl anesthesia for pediatric cardiac surgery. Thirty children 6 months to 6 years old were anesthetized with an intravenous fentanyl bolus of 10 mcg/kg. This was followed by a fentanyl infusion of 10 mcg/kg/h in the intravenous group (n = 10), 2 mcg/kg of intrathecal fentanyl in the intrathecal group (n = 10), or a combination of intravenous and intrathecal fentanyl in the combined group (n = 10). The findings in all three groups were statistically similar except for higher blood glucose levels during cardiopulmonary bypass in the intrathecal group compared with the intravenous group (P < .004). The combined intravenous–intrathecal group was the only group in whom the increases in heart rate and mean arterial blood pressure from presurgery to poststernotomy were not significant. The 24-hour urinary cortisol excretion rates were lowest in the combined group. A single intrathecal injection of fentanyl 2 mcg/kg offered no advantage over systemic fentanyl with regard to hemodynamic stability or suppression of the stress response. The combination of these two regimens may provide better hemodynamic stability during the pre-bypass period and may be associated with a decreased 24-hour urinary cortisol excretion rate.

To analyze the feasibility of immediate extubation, Figueira Moure et al.89 studied 29 ASA I-II pediatric patients undergoing surgery to correct simple heart defects. These patients were compared with a control group of 23 patients who had not received caudal morphine or had been selected for early extubation. Anesthesia was provided with sevoflurane, midazolam, rocuronium, fentanyl (maximum dose 10 mcg/kg), and a bolus of caudal morphine (50–60 mcg/kg) after anesthetic induction. All patients were extubated satisfactorily in the operating room. None required reintubation or reoperation. Postoperative pain was controlled with metamizol alone for 79.3% of patients. No episodes of respiratory depression or neurologic complications were observed. Pediatric ICU and hospital stays were significantly shorter in the study group than in the control group.

Williams and Abajian90 reported on the use of spinal anesthesia in high-risk neonates for the repair of patent ductus arteriosus. Spinal anesthesia with tetracaine was used as an alternative to general anesthesia in a series of 15 consecutive patients. The average dose of tetracaine was 2.4 mg/kg. Two patients early in the series had an inadequate level and received supplemental isoflurane. The remainder of the patients received either no or minimal supplementation to the basic technique. The cardiovascular status of the group was very stable with minimal changes in blood pressure, and recovery was rapid. The three patients who had not been intubated before surgery were extubated soon after surgical repair was completed. No complications of the technique were noted.

Noncardiac Surgery

Juelsgaard et al.91 studied 43 patients with coronary artery disease divided into three groups. Group A was allocated to receive either incremental spinal anesthesia (bupivacaine 0.5% plain); group B was allocated to receive single-dose spinal anesthesia (2.5 mL of bupivacaine 0.5% plain); and group C received general anesthesia (fentanyl, thiopentone, atracurium, enflurane, N2O/O2) for hip surgery. The number of patients who developed ST-segment depression was not significantly different among the three groups, but the number of ST-segment depression episodes differed. In group A, a total of 7 ST-segment depressions occurred in the observation period compared with 125 in group B and 16 in group C (P < .05). Intraoperative ST-segment depression occurred only in group B. Hypotensive events also varied among the groups: There were three events in group A, 33 in group B, and 40 in group C (P < .002). Altogether, 56% of hypotensive patients developed ST-segment depression compared with 10% of normotensive patients (P < .003). In group A, 1.6 mL of 0.5% bupivacaine were used as opposed to the fixed dose of 2.5 mL in group B (P < .001). In the first postoperative week, mortality rate was higher in group B (P < .05), but after 1 month there was no significant difference in mortality rates among the three groups. The incidence of hypotension and myocardial ischemia was lowest in the group receiving incremental spinal anesthesia.

Olofsson et al.92 reported that for the repair of hip fractures in aged patients, a reduced dose of hyperbaric bupivacaine (7.5 mg) in combination with sufentanil (5 mcg) provides reliable spinal anesthesia with few episodes of hypotension and little need for vasopressor support of the blood pressure. Ben-David et al.93 studied 20 patients 70 years or older undergoing surgical repair of hip fracture who were randomized into two groups of 10 patients each. Group A received a spinal anesthetic of bupivacaine 4 mg plus fentanyl 20 mcg, and group B received 10 mg bupivacaine. Hypotension was defined as a systolic pressure of less than 90 mm Hg or a 25% decrease in mean arterial pressure from baseline. Hypotension was treated with intravenous ephedrine boluses of 5–10 mg to a maximum of 50 mg and thereafter with phenylephrine boluses of 100–200 mcg. All patients had satisfactory anesthesia. One of 10 patients in group A required ephedrine as a single dose of 5 mg. Nine of 10 patients in group B required vasopressor support of blood pressure. Group B patients required an average of 35 mg ephedrine, and 2 patients required phenylephrine. The lowest recorded systolic, diastolic, and mean blood pressures as fractions of the baseline pressures were, respectively, 81%, 84%, and 85% for group A compared with 64%, 69%, and 64% for group B. The authors concluded that a “minidose” of 4 mg bupivacaine in combination with 20 mcg fentanyl provides adequate spinal anesthesia for surgical repair of hip fractures in the elderly.

In a retrospective study to determine whether spinal anesthesia can be safely used for patients undergoing endoluminal abdominal aortic aneurysm repair, Huang94 concluded that spinal anesthesia was safe. However, the major disadvantage of spinal anesthesia was its limited duration of anesthesia because 25% of patients had to be converted to general anesthesia owing to the extended length of surgery and 8% owing to anxiety. Continuous spinal anesthesia is a better choice for such procedures because it allows continuous intrathecal analgesia in the postoperative period and provides effective pain relief that is reflected by the favorable surgical outcome.95,96

Fleron et al.97 studied 217 patients scheduled to undergo abdominal aortic surgery who were randomly allocated to receive either general anesthesia alone (control) or general anesthesia combined with intrathecal narcotics (1 mcg/kg sufentanil with 8 mcg/kg preservative-free morphine injected at the L4-5 interspace). The administration of intrathecal opioid provided more intense analgesia than PCA during the first 24 hours postoperatively (P < .05). There was no difference between the groups for the incidence of combined major cardiovascular, respiratory, and renal complications or mortality rate. The incidence of myocardial damage or infarction, as evidenced by abnormal plasma concentrations of troponin I, did not differ between the two groups. The authors concluded that intense postoperative analgesia via intrathecal opioids does not alter the combined incidence of major cardiovascular, respiratory, and renal complications compared with general anesthesia alone.

Backlund et al.98 randomized 40 elderly patients (>65 years) to either spinal or general anesthesia, who were scheduled to undergo hip arthroplasty or peripheral vascular surgery and were at high risk for postoperative myocardial ischemia. Ambulatory ECG recordings until the third postoperative morning, a daily 12-lead ECG, and serum creatine kinase and troponin concentrations were obtained. The number of ischemic episodes, total duration of ischemia, and ischemic minutes per hour were noted for each patient perioperatively. Sixteen of the patients (40%) had postoperative myocardial ischemia. An intraoperative increase in the plasma concentration of norepinephrine but not epinephrine was detected in the patients who later developed postoperative myocardial ischemia. There were no differences between the groups, and the authors concluded that the type of anesthesia had no effect on the incidence of myocardial ischemia during or after surgery.

Similarly, Windsor et al.99 studied the incidence of myocardial ischemia in patients undergoing transurethral resection of the prostate who had either spinal or general anesthesia. Ninety-four patients undergoing transurethral resection of the prostate underwent Holter ECG monitoring pre- and postoperatively. There was no difference in the incidence of silent myocardial ischemia between the spinal (n = 60) and the general anesthesia (n = 34) groups. Others have reported similar findings.100 A number of case reports have been published on the usefulness of spinal anesthesia in patients with a variety of cardiac diseases. Velickovic et al.101,102 successfully used continuous spinal anesthesia for two patients with recurrent peripartum cardiomyopathy presenting in congestive heart failure for emergent cesarean section delivery. The spinal anesthesia not only provided adequate anesthesia but also markedly reduced the patients' symptoms. Others have reported similar success with spinal anesthesia in obstetric patients with hypertrophic obstructive cardiomyopathy,103 severe pulmonary stenosis,104 and coronary artery disease.105

Acute Pain Management

Zarate et al.106 compared remifentanil combined with intrathecal morphine with sufentanil for postoperative pain control in patients receiving desflurane anesthesia for elective CABG or valve replacement. Before entering the operating room, patients in the remifentanil group (n = 20) received morphine 8 mcg/kg intrathecally. Anesthesia was induced using a standardized anesthetic technique in all patients. In the remifentanil group, anesthesia was maintained with remifentanil, 0.1 mcg/kg/min in combination with desflurane 3–10%. In the sufentanil group (n = 20), patients received sufentanil 0.3 mcg/kg/h and desflurane 3–10%. There were no differences between the two groups with respect to time from arrival in the ICU to tracheal extubation (5.1 ± 4.3 hours vs 5.8 ± 6.7 hours for remifentanil and sufentanil groups, respectively). After extubation, patients in the remifentanil group had significantly lower VAS scores, reduced PCA requirements, and greater satisfaction with their perioperative pain management compared with patients in the sufentanil group. The researchers concluded that remifentanil combined with intrathecal morphine provided superior pain control after cardiac surgery compared with a sufentanil-based general anesthesia technique.

Boulanger et al.107assessed whether intrathecal analgesia facilitates early extubation and provides superior pain control after cardiac surgery compared with PCA or nurse-administered subcutaneous injections. Sixty-two patients undergoing elective cardiac surgery participated in this prospective, randomized, partly blinded study. Perioperative care was standardized, and patients were assigned to receive intrathecal morphine (ITM group) followed by PCA, intrathecal placebo (ITP group) followed by PCA, or subcutaneous injections of morphine every 4 hours as needed (SC group). Rating scales and questionnaires were used to assess clinical outcomes. The authors found that the use of intrathecal morphine did not favor earlier extubation, and there was even a tendency for longer extubation times in the intrathecal morphine group compared with the intrathecal placebo and subcutaneous groups. Pain scores, adverse effects, postoperative recovery, and patient satisfaction were comparable in the three groups. The report concluded that intrathecal morphine is not indicated in patients undergoing cardiac surgery if early extubation is planned.

Suominen et al.108 assessed the analgesic effect and safety of intrathecal morphine in postoperative pain control in children after heart surgery with a sternotomy incision. Eighty children, 3–55 kg, who were undergoing elective cardiac surgery with opioid-based anesthesia were randomly divided into a treatment group who received 20 mcg/kg intrathecal morphine at induction of anesthesia and a control group. Thirty-five patients were enrolled in the intrathecal morphinegroup and 36 in the control group. The mean time for the first intravenous morphine dose from intrathecal morphine administration or equivalent time zero in the control group was significantly longer (P = .003) in the intrathecal morphine group compared with the control group (12.3 vs 8.7 hours). Time from pediatric ICU admission to the start of intravenous morphine was also significantly longer (P = .01) in the intrathecal morphine group (6.0 vs 3.4 hours). The total intravenous morphine consumption over the mean 19 postoperative hours was significantly lower (P = .03) in the intrathecal morphine group. However, the use of intrathecal morphine did not result in earlier extubation or earlier discharge from the pediatric ICU. Of the 35 patients who received intrathecal morphine at induction of anesthesia, 20% (n = 7) did not require any additional morphine in the pediatric ICU compared with 3 of 36 control group patients. This did not reach statistical significance. The incidence of adverse events was low in both groups.

Physiologic Effects

Saito et al.109 studied the effects of paravertebral blocks in healthy volunteers. Twenty-two milliliters of 1% lidocaine were injected at the T11 level into the ventral area of the right-sided paravertebral space in 16 volunteers. A week later, the procedure was repeated with a saline injection. Unilateral analgesia (with no contralateral element) was induced in every subject injected with lidocaine; there was no block after saline injection. Loss of pinprick sensation was observed within 10 minutes after injection and involved a mean of 12 (range 8–13) dermatomes. A sympathetic block was indicated by cutaneous temperature increase within at least six dermatomes (Figure 57–10). Increase of arterial blood pressure was obtained in all volunteers with no change in pulse rate. No side effects or complications occurred. Epidural spread of the local anesthetic was unlikely because of the absence of contralateral cutaneous analgesia and temperature increase.

Cheema et al.110 studied six patients undergoing paravertebral block for chronically painful conditions of the chest wall, who were thermographically imaged so that the extent of cutaneous vasodilatation—hence sympathetic blockade—could be correlated with the distribution of the somatic block. All blocks were performed by a single experienced operator with a single percutaneous entry, using 15 mL of 0.5% bupivacaine at a mean level of T9-10 (range T7-T8, T10-11). Correct needle placement was confirmed radiologically. There was a mean distribution of the somatic block over five dermatomes (range 1–8), as evidenced by the loss of pinprick sensation; the upper and lower limits of blockade were T6 and L3, respectively. The mean distribution of the sympathetic blockade was eight dermatomes (range 6–10), as evidenced by ipsilateral skin warming with upper and lower limits of T5 and L3, respectively. No bilateral spread was observed. No significant postural changes in blood pressures were seen, although there was a small but significant decrease in supine heart rate (P = .05).

There is anecdotal evidence that paravertebral blocks can be beneficial in patients with ischemic heart disease and can relieve angina. Ho et al.111 reported on the intraoperative resolution of ST-segment depression after a right thoracic paravertebral block (Figure 57–11).

Cardiac Surgery

Canto et al.112 prospectively studied the feasibility and efficacy of bilateral continuous paravertebral block combined with general anesthesia in cardiac surgery. One hundred and-eleven elective patients had two paravertebral catheters inserted: one on either side of the midline within 2.5 cm of the spinous process of the third or fourth thoracic vertebrae. A mixture of ropivacaine and fentanyl was infused through the catheters during and after surgery. The technique was associated with good hemodynamic stability, good postoperative analgesia, and short times to tracheal extubation with few significant complications.

Exadaktylos et al.113 evaluated the preoperative application of intercostal nerve blockade (ICB), combined with general anesthesia for peri- and postoperative pain control. They also assessed its efficacy for early extubation. Nine consecutive patients undergoing MIDCAB surgery were evaluated. Preoperative ipsilateral ICB intercostal nerve block was used in all patients. After induction, isoflurane (0.4–0.8%) and nitrous oxide in combination with the preoperative nerve blockade provided sufficient anesthesia throughout the procedure. Only 2 of 9 patients required additional small doses of narcotics. All patients could be safely extubated within 15 minutes of skin suture. Postoperative discomfort and pain were minimal.

McDonald et al.114 investigated the effects of a parasternal block on postoperative analgesia, respiratory function, and extubation times in 20 patients having cardiac surgery via median sternotomy. Seventeen patients completed the study. A desflurane-based, small-dose opioid anesthetic was used. Before sternal wire placement, the surgeons performed the parasternal block as well as local anesthetic infiltration of the sternotomy and mediastinal tube sites with either 54 mL of saline placebo or 54 mL of 0.25% levobupivacaine with 1:400,000 epinephrine. Effects on pain and respiratory function were studied over 24 hours. Patients in the levobupivacaine group used significantly less morphine in the first 4 hours after surgery (20.8 ± 6.2 mg vs 33.2 ± 10.9 mg in the placebo group; P =.013). They also had better oxygenation at the time of extubation. Four of 9 in the placebo group needed rescue pain medication, versus none of 8 in the levobupivacaine group (P =.08). Peak serum levobupivacaine concentrations were below potentially toxic levels in all patients (0.64 ± 0.43 mcg/mL; range 0.24–1.64 mcg/mL). They concluded that parasternal block and local anesthetic infiltration of the sternotomy wound and mediastinal tube sites with levobupivacaine can be a useful analgesic adjunct for patients who are expected to undergo early tracheal extubation after cardiac surgery.

Noncardiac Surgery

Ohkado et al.115 describe the case of a 74-year-old woman with aortitis syndrome who was scheduled for mastectomy. She also had severe peripheral vascular disease. Anesthesia consisted of a combination of TEA and intercostal blocks, which were performed with bupivacaine 15 mg (0.5%, 3 mL) at the T3 and T4 levels. During the surgery, the patient remained hemodynamically stable, and no neurologic symptoms were observed.

Another report describes the case of a 72-year-old woman with a significant medical history of hypertrophic obstructive cardiomyopathy requiring breast cancer surgery.116 Paravertebral blocks were performed at T1-T6 with 5 mL of 0.5% ropivacaine and epinephrine 1:400,000 injected at each level. Intraoperatively, the patient required no other medication for analgesia and was comfortable and conversant during the 2-hour procedure. She remained pain-free after the operation and did not require any opioid medication until the following day.

Pain Management

Behnke et al.117 evaluated PCA and intercostal block for postoperative pain relief in patients undergoing MIDCAB procedures. Forty-three patients were included in the study. Anesthesia was induced and maintained in a standardized total intravenous manner with propofol, remifentanil, cisatracurium; glyceroltrinitrate, clonidine, and esmolol were given as needed. After revascularization, patients were randomly assigned to one of two groups receiving either 7.5 mg piritramid intravenously before extubation and continuing a PCA with a 2-mg bolus and a 10-minute lockout, or an intercostal block with 20-mL ropivacaine 1% (4 × 5 mL). In addition, all patients received 1 g of paracetamol rectally before induction of anesthesia and 1 g of metamizol intravenously at the end of surgery. A rescue medication of 3.75 mg piritramid intravenously was allowed. Pain scores, Aldrete scores (AS 0–12), and oxygen saturation were obtained at 1, 4, and 8 hours after extubation. The intercostal block group showed a significantly greater pain reduction in the first (5.8 ± 1.8 vs 7.3 ± 1.9; P < .02) and fourth hours (3.6 ± 1.3 vs 4.6 ± 1.4; P < .02), respectively. Transfer to an intermediate care ward 1 hour after extubation was achieved more often in the intercostal block group (ICB 9.6 ± 1.5 vs PCA 8.9 ± 1.2; P < .05). There was no difference between the groups with respect to oxygen saturation. The additional piritramid demand was 9.3 mg in the intercostal block group and 5 mg in the PCA group in the first 8 hours postoperatively. The authors concluded that compared with PCA, intercostal block gives better pain relief in the early postoperative phase after MIDCAB procedures.

Dowling et al.118 evaluated whether a continuous regional infusion of a local anesthetic delivered to the operative site would result in decreased levels of postoperative pain and narcotic requirements for patients who undergo a standard median sternotomy for cardiac surgery. Patients undergoing elective CABG alone or in combination with laser transmyocardial revascularization received bilateral intercostal block with either ropivacaine or saline. At wound closure, two catheters with multiple side openings were inserted percutaneously and placed directly over the sternum. The same agent was then administered as a continuous regional infusion for 48 hours. The total amount of narcotic analgesia required by the ropivacaine group was significantly less than for the control group (47.3 vs 78.7 mg, respectively; P = .038). The ropivacaine group required less narcotics on postoperative day 2 as well (15.5 vs 29.4 mg; P = .025). The mean overall pain scores for the ropivacaine group were significantly less than the mean overall scores for the normal saline group (1.6 vs 2.6, respectively; P = .005). Patients receiving ropivacaine had a mean length of stay of 5.2 days compared with 8.2 days for patients in the normal saline group (P = .001). Even after excluding outliers (length of stay = 39 days), the normal saline group mean length of stay was still longer at 6.3 days (P < .01). No difference in pulmonary function was seen. The authors concluded that continuous delivery of local anesthetics significantly improved postoperative pain control while decreasing the amount of narcotic analgesia required. There was also a significant decrease in hospital length of stay, which is likely to result in significant cost reductions.

Dhole et al.119 compared continuous TEA and paravertebral block for postoperative analgesia in patients undergoing MIDCAB surgery. They evaluated the quality of analgesia, complications, compliance to chest physiotherapy, hemodynamics, and respiratory effects. Patients in the TEA group had an epidural catheter inserted at the T4-5 interspace, whereas patients in the paravertebral block group had a catheter inserted in the paravertebral space on the left side at the T4-5 level. There was no statistically significant difference in VAS scores and requirement of supplemental analgesia between the two groups (Table 57–6). Cardiac index at 4 hours and 6 hours was significantly higher in the TEA group. Patients in the paravertebral block group had significantly lower respiratory rates at 8, 10, and 12 hours. All other parameters were comparable. The authors concluded that paravertebral block is as effective as TEA for postoperative analgesia after MIDCAB.

Fiorani et al.120 compared the influence of anesthetic technique on perioperative complications in patients undergoing carotid endarterectomy in a retrospective study of 1020 consecutive patients; 337 patients (33%) were treated under general anesthesia and 683 (67%) under cervical block. The overall perioperative stroke rate was 1.9%, the death–stroke rate was 0.7%, and the cardiac complication rate was 0.8%. The perioperative stroke rate was higher in the general anesthesia group than in the cervical block group (3.2% vs 1.3%, P = .01). Cardiac complication rates were similar in the two groups. A carotid artery shunt was used in 75 patients (22%) receiving general anesthesia and in 92 patients (13%) receiving a cervical block (P = .0004). The causes of stroke in the cervical block group were intraoperative embolism (4 cases, 26%), perioperative thromboembolism (7 cases, 58%) and clamping ischemia (1 case, 16%). Mechanisms causing stroke in the general anesthesia group remained unidentified or uncertain. The authors concluded that cervical block anesthesia yields better perioperative results than general anesthesia, probably because it allows more reliable cerebral monitoring. It thereby reduces or even eliminates perioperative strokes related to clamping ischemia. It also facilitates detection of the mechanism underlying intraoperative strokes, allowing surgical techniques and intraoperative management to be modified accordingly. Finally, cervical block anesthesia significantly reduces the need for internal carotid artery shunting.

Sternbach et al.121 studied the differences in perioperative hemodynamics and associated outcomes in patients who underwent carotid endarterectomy with regional and general anesthesia. Carotid endarterectomy was performed in 527 patients (226 with cervical block and 324 with general anesthesia). The patients who underwent the operation under general anesthesia had significantly greater intraoperative and postoperative hemodynamic variability and received more vasoactive medications during surgery (87% vs 51%; P < .001) and in the recovery room (36% vs 21%; P = .0009). Major postoperative blood pressure derangements were more common in the general anesthesia group (18% vs 10%; P < .05). Patients who underwent the operation under general anesthesia more frequently needed ICU admission (16% vs 7%; P = .01) and had more frequent delays in discharge (20% vs 11%; P = .008) with a postoperative length of stay of 2.1 days vs 1.6 days for the regional group (P = .01). Although no difference was seen in neurologic morbidity rates between groups (combined major stroke/death rate, 1.8%), the major cardiac morbidity rate was noted to be lower in the cervical block group (1% vs 4%; P = .05). The total in-room time was shorter in the cervical block group (108 vs 122 minutes; P < .001). The authors concluded that carotid endarterectomy performed under cervical block is associated with significantly less perioperative hemodynamic instability than when performed under general anesthesia. This results in fewer major adverse cardiac events. Ultimately, the use of critical care resources is decreased, as is the hospital length of stay.

Bowyer et al.122 studied the outcomes of 489 patients undergoing 500 carotid endarterectomy, comparing general anesthesia with cervical block anesthesia. Overall, the perioperative mortality was 0.8%. Compared with the regional anesthesia group, the general anesthesia group had greater overall morbidity (23.3 vs 13.6%; P < .008) and more frequent use of postoperative vasoactive drugs to control blood pressure (40.4 vs 26.1%; P < .001). Furthermore, anesthesia time, operative time, and frequency of shunt use were significantly greater in the general anesthesia group (P < .03).

Prough et al.123 report on 185 carotid endarterectomies that were performed on 153 patients under regional anesthesia. Of these patients, 38 (25 %) had suffered a previous myocardial infarction, 63 (41 %) had documented coronary artery disease, and 115 (75 %) had hypertension. Anesthesia was provided by a superficial cervical plexus block. Monitoring consisted of measurement of direct arterial pressure and continuous display of the ECG. Oxygen was administered by nasal cannula throughout the procedure. Mean arterial pressure was elevated, when necessary, by infusion of phenylephrine. No patient in this study suffered an acute myocardial infarction. The only cardiac complications consisted of eight episodes of non–life-threatening arrhythmias. The authors concluded that regional anesthesia for carotid endarterectomy is associated with a low risk of perioperative myocardial infarction.

Davies et al.124 studied 128 patients having carotid endarterectomy under superficial and deep cervical plexus blocks to determine the incidence of intraoperative and postoperative neurologic and cardiovascular complications. Twenty-seven patients who had intraoperative neurologic changes after carotid artery clamping responded to shunt insertion. Six patients had transient neurologic changes after the operation but no permanent neurologic complications. Tachycardia (55%) and hypertension (67%) were the most common intraoperative cardiovascular changes, and one patient had clinical postoperative myocardial infarction. Ninety-two percent of patients who could be adequately assessed preferred to have the same method of anesthesia for a future carotid endarterectomy. The authors concluded that carotid endarterectomy under superficial and deep cervical plexus blocks was associated with a high patient acceptance, low neurologic complication rate, and an acceptable rate of cardiac complications.

Ombrellaro et al.125 investigated the effects of anesthetic technique on cardiac morbidity after carotid artery surgery. Two-hundred-sixty-six consecutive carotid endarterectomies were performed under local/regional (n = 140) or general anesthesia (n = 126). The effects of anesthetic technique on postoperative adverse cardiac events were assessed retrospectively. Forty-seven adverse cardiac events (4 myocardial infarction, 9 congestive heart failure, 7 angina, and 27 new ventricular arrhythmias) occurred postoperatively in 38 patients (14.3%). There were no deaths. The relative risks of general anesthesia for arrhythmias, myocardial infarction, angina, congestive heart failure, and total adverse cardiac events were 2.22, 0.37, 0.83, 1.38, and 1.5, respectively. The only statistically significant difference between the two groups was an increased risk of postoperative arrhythmias after general anesthesia (P < .03). The authors concluded that major cardiac morbidity after carotid endarterectomy is independent of anesthetic technique.

Fanelli et al.126 assessed cardiovascular function in 20 ASA I–II patients scheduled for elective orthopedic surgery with tourniquet. The principal objective of the study was to compare the hemodynamic changes induced by unilateral spinal anesthesia and combined sciatic–femoral nerve block. Spinal anesthesia was obtained by 8 mg of hyperbaric bupivacaine 0.5% (group S, n = 10). Combined sciatic–femoral nerve block was obtained by 7 mg/kg of mepivacaine 2% (group NB, n = 10). The duration of surgery and acceptability of anesthetic techniques were similar in the two groups. In eight patients of group S, spinal block was restricted to the operated side (pinprick test and Bromage scale), whereas the other two patients developed bilateral spinal block after being turned supine. The NB group showed no hemodynamic changes during the study, whereas patients in group S showed a small but significant decrease of mean arterial blood pressure (P < .002 vs baseline and P < .04 vs NB), cardiac index (P <. 01 vs baseline and P< .01 vs NB), and stroke volume index (P < .01 vs baseline and P ≤ .01 vs NB). They concluded that both sciatic–femoral and unilateral spinal blockade provide adequate anesthesia for unilateral leg surgery with tourniquet. The sciatic–femoral technique affects cardiovascular performance less than unilateral spinal.

Peripheral nerve blocks are associated with minimal hemodynamic disturbances. They are perhaps ideal for high-risk surgical patients who cannot tolerate the adverse consequences of even the slightest alteration of hemodynamics. Several case reports illustrate these benefits. Chia et al.127 presented the practical benefits of a combined sciatic–femoral nerve block on a 56-year-old man with severe sepsis and recent myocardial infarction requiring an urgent above-knee amputation. Ho and Karmakar128 report the use of a combined paravertebral lumbar plexus and parasacral sciatic nerve block for reduction of a hip fracture in an elderly patient with severe aortic stenosis. Tanaka and Negoro129 describe the use of a psoas compartment block in a 72-year-old woman with severe heart failure due to rheumatoid myocarditis who required an open reduction of a left femoral neck (trochanteric) fracture. With the patient in the lateral position with the fractured side up, they performed the block at L3-4 using a 22-gauge Tuohy needle to inject 10 mL of normal saline and 20 mL of 2% mepivacaine. No complications were reported. Rizzo et al.130 used regional anesthesia to anesthetize a 32-year-old man suffering from Eisenmenger's syndrome with left-type-only ventricle, who needed an extirpation of meniscus by arthroscopic surgery. They used sciatic, femoral, and lateral cutaneous of thigh nerve blocks with ropivacaine without complications.

Numerous studies have been undertaken to determine the effects of regional anesthesia on the cardiovascular system during cardiac and noncardiac surgery in the high-risk patient. Epidural and intrathecal anesthesia have been shown to provide stress response attenuation, cardiac sympathectomy, improved cardiac performance, and intense postoperative analgesia for cardiac and noncardiac surgery. Conflicting data remain regarding whether neuraxial anesthesia techniques are associated with earlier extubation times and shorter ICU and hospital stays. In addition, few studies have used clinical outcomes as primary endpoints, although the benefits reported may translate into improved patient outcome. Other regional anesthesia techniques such as paravertebral block, intercostal nerve block, parasternal block, cervical plexus block, and combined sciatic–femoral nerve block all have been successfully used in the anesthetic management of patients with cardiovascular disease. In addition, the application of regional anesthesia, especially neuraxial techniques, in anticoagulated patients remains an area of controversy. However, the use of a set of standard safety measures has been shown to avert the occurrence of symptomatic hematomas. Well-designed randomized controlled studies with large patient numbers may help resolve some of the controversies surrounding regional anesthesia's impact on outcome after cardiac and noncardiac surgery in the high-risk patient. Despite the controversies, regional anesthetic techniques have gained popularity in both the anesthesiology and surgical communities and are increasingly considered valuable resources in managing patients with cardiovascular disease.