Introduction: The optimal combination of intraoperative anesthesia and postoperative analgesia for patients undergoing major vascular surgery remains controversial. Recently there has been considerable interest in the utilization of regional anesthetic and analgesic techniques to improve outcome and reduce cost and LOS. This prospective, double-blind, randomized clinical trial was conducted to examine the impact of alternate combinations of intraoperative anesthesia and postoperative analgesia on LOS after aortic surgery. Methods: Following institutional approval and written informed consent, 160 consecutive patients presenting for abdominal aortic reconstruction were studied. All patients were monitored with arterial and pulmonary artery catheters and received a T10-T12 epidural catheter. After epidural placement was confirmed by a test dose, patients were randomized to one of four possible treatment combinations: 1) epidural supplemented general anesthesia/epidural PCA (EPI/EPI); 2) epidural supplemented general anesthesia/IV PCA (EPI/IV); 3) general anesthesia/IV PCA (GA/IV); and 4) general anesthesia/epidural PCA (GA/EPI). Each patient received simultaneous IV and epidural boluses/infusions, one of which contained active drug and the other saline according to randomized treatment assignment. All aspects of perioperative anesthetic and medical management were rigorously controlled by protocol. Postoperative PCA was continued for 72 hours and managed by an acute pain service. LOS was calculated using date of surgery and date of discharge. Patients who did not survive to discharge were excluded from analysis. Data were analyzed using linear regression of ranks of LOS and are reported as mean +/- SD and [median]. Results: The results are shown in Table 1. Differences in LOS between the four treatment groups were not statistically significant. Nine patients did not survive to discharge.Table 1: LOS by treatment assignmentDiscussion: This is the first double-blind, randomized clinical trial comparing alternate combinations of intraoperative anesthesia and postoperative analgesia in vascular surgery patients. Unique features of this trial include standardization of clinical care throughout the perioperative period, inclusion of all four possible treatment combinations, and patient and treating physicians blinded to treatment assignment. In this high-risk surgical population, we were unable to demonstrate a reduction in the LOS. Supported by NIH grant GM-38177-06.

Study Objective: To examine the degree of success at maintaining patients randomized to epidural or general anesthesia for peripheral vascular surgery within predetermined blood pressure (BP) and heart rate (HR) limits. To investigate associations between such hemodynamic control and intraoperative myocardial ischemia and postoperative major cardiac morbidity.Design: Prospective randomized clinical trial.Setting: University-affiliated hospital.Patients: 100 patients undergoing elective lower extremity revascularization for atherosclerotic peripheral vascular disease.Interventions: Blood pressure and HR limits were determined prior to randomization. Hemodynamic monitoring and management of anesthesia was 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 BP and HR data were analyzed by investigators masked to the type of anesthesia given.Measurements and Main Results: A greater percentage of patients randomized to general anesthesia had intraoperative BPs more above their limit (95% vs. 72%, p = 0.002) and/or more rapid changes in HR (75% vs. 48%, p = 0.008) or BP (100% vs. 93%, p = 0.04) than those randomized to epidural anesthesia. Intraoperative ischemia and major cardiac morbidity were similar in the two anesthesia groups. Patients experiencing intraoperative ischemia, regardless of anesthetic type, more frequently had BPs greater than 10% above their upper limit (90% vs. 60%, p = 0.04) and/or more rapid HR changes (90% vs. 58%, p = 0.03) compared with patients without ischemia. These vital sign abnormalities, however, were not necessarily temporally related to the ischemic episodes. Patients experiencing subsequent major cardiac morbidity were not different from other patients with respect to excursions out of BP or HR limits.Conclusions: Prevention of elevated intraoperative BP and/or rapid changes in BP or HR may be more successful with epidural than with general anesthesia. Such 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. (C) 1996 by Elsevier Science Inc.

OBJECTIVE:To determine whether catecholamine and cortisol secretory responses to surgery contribute to postoperative complications.DESIGN:Prospective, randomized, case series.SETTING:A university hospital operating suite and surgical intensive care unit.PATIENTS:Sixty patients undergoing lower extremity vascular surgery.INTERVENTIONS:Patients were randomized to receive either epidural anesthesia/epidural opiate analgesia (regional anesthesia) or general anesthesia/intravenous patient-controlled analgesia (general anesthesia).MEASUREMENTS AND MAIN RESULTS:Anesthesia was managed according to a prospectively designed protocol. Hemodynamic parameters and plasma catecholamine concentrations were determined at specific intraoperative and postoperative time points. Intraoperative and postoperative urine samples were collected and analyzed for free cortisol concentrations. Outcomes evaluated were cardiac (nonfatal myocardial infarction and cardiac death) and surgical (graft occlusion). Mean arterial pressure during emergence from anesthesia and in the early postoperative period correlated positively with plasma norepinephrine concentration (p < .01). In addition, plasma catecholamine concentrations were higher in patients with postoperative hypertension. Plasma norepinephrine concentrations at the time of emergence from anesthesia and postoperatively were also higher in patients requiring repeat surgery for graft revision, thrombectomy, or amputation (p < .05). Multivariate analysis indicated that the norepinephrine concentration at the time of emergence, but not type of anesthesia, correlated with reoperation for graft occlusion, suggesting that the previously reported beneficial effect of regional anesthesia may be due to modulation of the stress response. Myocardial infarction or cardiac death occurred in three patients. These patients had markedly increased catecholamine concentrations.CONCLUSIONS:The catecholamine response to lower extremity vascular surgery contributes to the development of postoperative hypertension and may also be important in the development of thrombotic complications.

Objective: To determine whether catecholamine and cortisol secretory responses to surgery contribute to postoperative complications.Design: Prospective, randomized, case series.Setting: A university hospital operating suite and surgical intensive care unit.Patients: Sixty patients undergoing lower extremity vascular surgery.Interventions: Patients were randomized to receive either epidural anesthesia/epidural opiate analgesia (regional anesthesia) or general anesthesia/intravenous patient-controlled analgesia (general anesthesia).Measurements and Main Results: Anesthesia was managed according to a prospectively designed protocol. Hemodynamic parameters and plasma catecholamine concentrations were determined at specific intraoperative and postoperative time points. Intraoperative and postoperative urine samples were collected and analyzed for free cortisol concentrations. Outcomes evaluated were cardiac (nonfatal myocardial infarction and cardiac death) and surgical (graft occlusion). Mean arterial pressure during emergence from anesthesia and in the early postoperative period correlated positively with plasma norepinephrine concentration (p < .01). In addition, plasma catecholamine concentrations were higher in patients with postoperative hypertension. Plasma norepinephrine concentrations at the time of emergence from anesthesia and postoperatively were also higher in patients requiring repeat surgery for graft revision, thrombectomy, or amputation (p < .05). Multivariate analysis indicated that the norepinephrine concentration at the time of emergence, but not type of anesthesia, correlated with reoperation for graft occlusion, suggesting that the previously reported beneficial effect of regional anesthesia may be due to modulation of the stress response. Myocardial infarction or cardiac death occurred in three patients. These patients had markedly increased catecholamine concentrations.Conclusions: The catecholamine response to lower extremity vascular surgery contributes to the development of postoperative hypertension and may also he important in the development of thrombotic complications.

Although several studies have demonstrated a reduced incidence of postoperative deep venous thrombosis among patients who receive regional anesthesia, the influence of anesthetic method on early arterial bypass graft patency has not been well studied. The records of 78 consecutive patients undergoing elective femoro-popliteal (FP) or femoro-tibial (FT) bypass grafts, and who were randomized to receive general anesthesia and postoperative patient-controlled intravenous narcotic analgesia (GEN, n = 41), or epidural anesthesia and postoperative continuous epidural analgesia (EPI, n = 37), were retrospectively reviewed. The two groups were evenly matched with respect to demographic characteristics, risk factors, and vascular variables. There was one death in each group, yielding an operative mortality of 2.6 per cent, and leaving 76 patients available for further analysis. Graft occlusion occurred in 11 (14.5%) cases within the first 7 postoperative days, including 9 (22.5%) GEN and 2 (5.6%) EPI patients (P < 0.05). There were two (4.4%) FP occlusions, including two (8.7%) GEN and 0(0%) EPI cases; there were nine FT occlusions, including seven (41.2%) GEN and two (14.3%) EPI cases. Graft occlusion occurred in 11 (17.1%) of the 64 limb salvage cases, including nine (27.3%) GEN and two (6.5%) EPI cases (P < 0.05), and in seven (12.7%) of 55 greater saphenous vein grafts, including six (22.2%) GEN and 1 (3.6%) EPI cases (P < 0.05). By multivariate analysis, FT grafts, preoperative plasminogen activator inhibitor-1 (PAI-1) levels, and GEN were predictive of early graft occlusion (P < 0.05). Furthermore, the levels of circulating PAI-1 were higher 24 hours postoperatively among patients in the GEN group (P < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Glavan, N J B.A.; Christopherson, R M.D.,Ph.D.; Norris, E J M.D.; Frank, S M M.D.; Gottlieb, S O M.D.; Beattie, C Ph.D.,M.D. Author Information

Hypothermia occurs commonly during surgery and can be associated with increased metabolic demands during rewarming in the postoperative period. Although cardiac complications remain the leading cause of morbidity after anesthesia and surgery, the relationship between unintentional hypothermia and myocardial ischemia during the perioperative period has not been studied.One hundred patients undergoing lower extremity vascular reconstruction received continuous Holter monitoring throughout the first 24 h postoperatively. Myocardial ischemia was determined by a cardiologist masked to clinical variables. The patient's sublingual temperature on arrival at the intensive care unit immediately after the surgical procedure was used to divide the patients into two groups: hypothermic (temperature, < 35 degrees C; n = 33) and normothermic (temperature, > or = 35 degrees C; n = 67). The relationship between intentional hypothermia and myocardial ischemia occurring during the first postoperative day was evaluated by univariate and multivariate analyses.A greater percentage of patients had electrocardiographic changes consistent with myocardial ischemia in the hypothermic group (36%, 12 of 33) compared with those in the normothermic group (13%, 9 of 67, P = 0.008). Preoperative risk factors for perioperative cardiac morbidity were similar between the two groups, except for patient age. The mean age was 70 +/- 2 yr and 62 +/- 1 yr in the hypothermic and normothermic groups, respectively (P = 0.001). When subgroup and multivariate analyses were used to adjust for differences in age, temperature remained an independent predictor of ischemia (odds ratio, 1.82 per degree Celsius; 95% confidence interval, 1.09-3.02). The incidence of postoperative angina was greater in the hypothermic group (18%, 6 of 33) than in the normothermic group (1.5%, 1 of 67, P = 0.002). The incidence of PaO2 < 80 mmHg in the arterial blood was greater in the hypothermic group (52%, 17 of 33) than in the normothermic group (30%, 20 of 67, P = 0.03).Unintentional hypothermia is associated with myocardial ischemia, angina, and PaO2 < 80 mmHg during the early postoperative period in patients undergoing lower extremity vascular surgery.

The purpose of this clinical trial was to compare the effects of different anesthetic and analgesic regimens on hemostatic function and postoperative arterial thrombotic complications.Ninety-five patients scheduled for elective lower extremity vascular reconstruction were randomized to receive either epidural anesthesia followed by epidural fentanyl (RA) or general anesthesia followed by intravenous morphine (GA). Intraoperative and postoperative care were controlled by protocol using predetermined limits for heart rate, blood pressure, and other monitoring criteria. Data collection included serial physical examinations, electrocardiograms, and cardiac isoenzymes to detect arterial thrombosis (defined as unstable angina, myocardial infarction, or vascular graft occlusion requiring reoperation). Fibrinogen, plasminogen activator inhibitor-1 (PAI-1), and D-dimer levels were measured preoperatively and at 24 and 72 h postoperatively.Preoperative fibrinogen levels were similar in both groups, remained unchanged after 24 h, and increased equally (45%) in the first 72 h postoperatively. PAI-1 levels in the GA group increased from 13.6 +/- 2.1 activity units (AU)/ml to 20.2 +/- 2.6 AU/ml at 24 h and returned to baseline at 72 h. In contrast, PAI-1 levels in the RA group remained unchanged over time. Twenty-two of 95 patients (23%) had postoperative arterial thrombosis, 17 of whom had received GA and 5 of whom, RA. Preoperative PAI-1 levels were higher in patients who developed postoperative arterial thrombosis (20.5 +/- 3.6 AU/ml vs. 11.2 +/- 1.4 AU/ml). Multiple logistic regression analysis indicated that GA and preoperative PAI-1 levels were predictive of postoperative arterial thrombotic complications.Impaired fibrinolysis may be related causally to postoperative arterial thrombosis. Because RA combined with epidural fentanyl analgesia appears to prevent postoperative inhibition of fibrinolysis, this form of perioperative management may decrease the risk of arterial thrombotic complications in patients undergoing lower extremity revascularization.

BACKGROUND:Perioperative morbidity may be modifiable in high risk patients by the anesthesiologist's choice of either regional or general anesthesia. This clinical trial compared outcomes between epidural (EA) and general (GA) anesthesia/analgesia regimens in a group of patients at high risk for cardiac and other morbidity who were undergoing similarly stressful surgical procedures.METHODS:One hundred patients scheduled for elective vascular reconstruction of the lower extremities were randomized to receive either EA for surgery followed by epidural analgesia, or GA for surgery followed by intravenous patient-controlled analgesia. Hemodynamic monitoring, blood pressure, and heart rate limits were determined prior to randomization. Management of anesthesia in the immediate postoperative period was standardized. The data collected included continuous electrocardiographic monitoring from the day before surgery through the third postoperative day, serial electrocardiograms, and cardiac enzymes. Cardiac ischemia, myocardial infarction, unstable angina, and cardiac death were identified by a cardiologist blinded to the type of anesthesia received. Other major morbidity was determined at the time of hospital discharge and at 1 and 6 months after surgery.RESULTS:Eleven patients who received GA required regrafting or an embolectomy during their hospital stay, compared with two patients who received EA. This association of GA with reoperation remained significant after adjustment for baseline differences. Cardiac outcomes were similar in the two groups with respect to perioperative death (1 EA and 1 GA), death within 6 months (4 EA and 3 GA), nonfatal myocardial infarction within 7 days (2 EA and 2 GA), unstable angina (0 EA and 2 GA), and myocardial ischemia following randomization (17 EA and 23 GA). Rates of major infections in the two groups (1 EA and 2 GA), renal failure (3 EA and 3 GA), and pulmonary complications (3 EA and 7 GA) also were similar.CONCLUSIONS:Carefully conducted epidural and general anesthesia appear to be associated with comparable rates of cardiac and most other morbidity in patients undergoing lower extremity vascular surgery. However, compared with general anesthesia, epidural anesthesia is associated with a lower incidence of reoperation for inadequate tissue perfusion and, therefore, may be advantageous for this surgical population.

BACKGROUND:Surgical trauma elicits diffuse changes in hormonal secretion and autonomic nervous system activity. Despite studies demonstrating modulation of the stress response by different anesthetic/analgesic regimens, little is known regarding the determinants of catecholamine and cortisol responses to surgery.METHODS:Plasma catecholamines and cortisol secretion data were obtained from 60 patients undergoing lower extremity revascularization. Patients were randomized to receive either general anesthesia combined with patient-controlled intravenous morphine (GA) or epidural anesthesia combined with epidural fentanyl analgesia (RA). All aspects of intra- and postoperative clinical care were defined by written protocol. Plasma catecholamines were measured before induction, intraoperatively, and for the first 18 h postoperatively (by HPLC). Urine cortisol was measured intra- and postoperatively using RIA. Data were evaluated using univariate and multivariate analyses to evaluate demographic and perioperative variables as determinants of stress hormone secretion.RESULTS:Plasma catecholamines increased during skin closure in the GA group, and remained higher relative to the RA group in the postoperative period. Multivariate analysis indicated that age and anesthetic regimen predicted increases in catecholamines during skin closure (P < 0.005), although duration of surgery, blood loss, and body temperature were not correlated. Early postoperative norepinephrine concentrations were correlated with pain score and duration of surgery (P < 0.004), but not with anesthetic management, blood loss, or body temperature. All postoperative norepinephrine levels were highly correlated (r = 0.7) with norepinephrine levels during skin closure. Cortisol excretion was higher postoperatively than intraoperatively. No patient or perioperative variable predicted cortisol excretion, and cortisol excretion was not correlated with catecholamine levels at any time.CONCLUSIONS:These data indicate that patient factors, such as age and inherent sympathetic responsivity, are important determinants of the catecholamine response to surgery. Modulation of the norepinephrine response by regional anesthesia/analgesia appears to be related, in part, to superior analgesia. The lack of correlation between catecholamine and cortisol secretion indicates that the stress response may consist of discrete systems responding to different stimuli.

Background: The purpose of this clinical trial was to compare the effects of different anesthetic and analgesic regimens on hemostatic function and postoperative arterial thrombotic complications.Methods: Ninety-five patients scheduled for elective lower extremity vascular reconstruction were randomized to receive either epidural anesthesia followed by epidural fentanyl (RA) or general anesthesia followed by intravenous morphine (GA). intraoperative and postoperative care were controlled by protocol using predetermined limits for heart rate, blood pressure, and other monitoring criteria. Data collection included serial physical examinations, electrocardiograms, and cardiac isoenzymes to detect arterial thrombosis (defined as unstable angina, myocardial infarction, or vascular graft occlusion requiring reoperation). Fibrinogen, plasminogen activator inhibitor-1 (PAI-1), and D-dimer levels were measured preoperatively and at 24 and 72 h postoperatively.Results: Preoperative fibrinogen levels were similar in both groups, remained unchanged after 24 h, and increased equally (45%) in the first 72 h postoperatively. PAI-1 levels in the GA group increased from 13.6 +/- 2.1 activity units (AU)/ml to 20.2 +/- 2.6 AU/ml at 24 h and returned to baseline at 72 h. In contrast, PAI-1 levels in the RA group remained unchanged over time. Twenty-two of 95 patients (23%) had postoperative arterial thrombosis, 17 of whom had received GA and 5 of whom, RA. Preoperative PAI-1 levels were higher in patients who developed postoperative arterial thrombosis (20.5 +/- 3.6 AU/ml vs. 11.2 +/- 1.4 AU/ml). Multiple logistic regression analysis indicated that GA and preoperative PAI-1 levels were predictive of postoperative arterial thrombotic complications.Conclusions: impaired fibrinolysis may be related causally to postoperative arterial thrombosis. Because RA combined with epidural fentanyl analgesia appears to prevent postoperative inhibition of fibrinolysis, this form of perioperative management may decrease the risk of arterial thrombotic complications in patients undergoing lower extremity revascularization.

A major component of the organ injury mediated by toxic oxidants, such as seen following reperfusion of the ischemic liver, is due to the peroxidation of polyunsaturated fatty acids, especially of cell membranes. We utilized the measurement of exhaled breath ethane, a metabolic product unique to oxidant-mediated lipid peroxidation, as a noninvasive indicator of this process in swine liver subjected to warm ischemia/reperfusion. Under rigorously controlled anesthesia conditions, pig livers were subjected to 2 h of warm total ischemia, followed by reperfusion in situ. Expired air was collected and its ethane content quantitated by a novel gas chromatographic technique. The time course of breath ethane generation correlated closely with the appearance of hepatocellular injury as measured by impairment of Factor VII generation and other measures of liver integrity. Moreover, the administration of the specific superoxide free radical scavenger, superoxide dismutase (SOD), significantly attenuated both the elaboration of ethane and the hepatocellular injury. These findings not only provide confirmation of the previously reported link between hepatocellular injury by free radicals generated at reperfusion, but also establish the use of expired breath ethane analysis as a sensitive, specific, and noninvasive indicator of the injury process in real time.

To elucidate the multifactorial nature of perioperative changes in body temperature, the influence of several clinical variables, including anesthetic technique, ambient operating room temperature, and age, were evaluated. Perioperative oral sublingual temperatures were measured in 97 patients undergoing lower extremity vascular surgery randomized to receive either general (GA) or epidural (EA) anesthesia. Surgery and anesthesia were performed in operating rooms (OR) with a relatively warm mean ambient temperature (24.5 +/- 0.4 degrees C) (GA, n = 30; EA, n = 33) or relatively cold mean ambient temperature (21.3 +/- 0.3 degrees C) (GA, n = 21; EA, n = 13). Patients were 35-94 yr old, with a mean age of 64.5 +/- 1.1 yr. A regression analysis was performed to determine the variables that correlated with intraoperative decrease in temperature and postoperative rewarming rate. The major correlates of greater intraoperative decrease in temperature were 1) GA (P = 0.003); 2) cold ambient OR temperature (P = 0.07); and 3) advancing patient age (P = 0.03). There was significant interaction between ambient OR temperature and type of anesthesia (P = 0.03): there was a greater intraoperative decrease in temperature with GA compared to EA in a cold OR but a similar decrease with GA and EA in a warm OR. The data also suggest an interaction between type of anesthesia and patient age (P = 0.06), showing a greater decrease in temperature with GA compared to EA in the younger patients, but a similar decrease between GA and EA in older patients.(ABSTRACT TRUNCATED AT 250 WORDS)

To identify a relationship between atherosclerotic vascular disease and differences in blood pressure between the right and left arms, blood pressure differences between arms were measured in patients with peripheral vascular disease (PVD, n = 58), in patients with coronary artery disease (CAD, n = 38), and in patients with no evidence of atherosclerotic disease, who served as a control group (n = 38). The incidence and magnitude of right and left arm pressure difference determined by the oscillometric technique were compared between the patient groups. The incidence of systolic pressure difference greater than or equal to 20 mmHg between arms in patients with PVD (21%) was greater than that in either those with CAD (3%) (P less than or equal to 0.05) or control subjects (0%) (P less than 0.01). The incidence of systolic pressure difference greater than or equal to 45 mmHg between arms in patients with PVD (10%) was greater than that in either those with CAD (0%) (P less than 0.05) or control subjects (0%) (P less than 0.05). Patients with PVD also had a greater incidence of right and left arm difference than did those with CAD or controls for mean and diastolic blood pressures. Of all patients with a systolic difference greater than 10 mmHg, neither the right nor the left arm blood pressure was consistently higher: 21 of 35 (60%) had a higher pressure in the right arm, and 14 of 35 (40%) had a higher pressure in the left arm (P = 0.33). Gender, diabetes, hypertension, smoking, and age were not associated with a difference in blood pressure between the right and left arms.(ABSTRACT TRUNCATED AT 250 WORDS)