Comparison of three remifentanil dose-finding regimens for coronary artery surgery☆☆☆

Article Outline

• Abstract
• Methods
• Results
  • Induction phase
  • Maintenance phase
  • ICU recovery phase
  • Safety
• Discussion
• Acknowledgements
• References
• Copyright

Abstract 

Objectives: To identify the remifentanil dosing regimen providing safe and optimal anesthetic conditions during coronary artery bypass graft surgery and to evaluate postoperative recovery characteristics. Design: Open-label, randomized, parallel group. Setting: Three centers in the United States. Participants: Seventy-two patients with left ventricular stroke volumes ≥50 mL. Interventions: Patients were randomized to remifentanil doses of 1 μg/kg/min (group 1, n = 23); 2 μg/kg/min (group 2, n = 24), or 3 μg/kg/min (group 3, n = 25). Somatic, sympathetic, and hemodynamic responses indicating inadequate anesthesia were treated with bolus doses of remifentanil, 1 to 2 μg/kg, and infusion rate increases, and, if necessary, isoflurane 0.5% to 1.0% was added as a rescue anesthetic. In the intensive care unit, the remifentanil infusion was reset to 1 μg/kg/min, with midazolam administered for supplemental sedation and morphine for analgesia. Measurements and Main Results: The durations of anesthesia, surgery, and cardiopulmonary bypass were similar for the 3 study groups. In addition, dose of lorazepam premedication, time to loss of consciousness, preoperative left ventricular ejection fraction, age, weight, and sex were similar for the 3 study groups. Remifentanil alone (infusion and boluses) prevented and controlled all responses to stimulation in 44% of group 3, 37% of group 2 and 9% of group 1 patients intraoperatively. Isoflurane (0.5%-1% inspired) rescue was successful in the remaining patients in each group. Hypotension indicating discontinuation of isoflurane and reductions of remifentanil infusion rates occurred in 64% to 75% of all patients. The optimal range of remifentanil infusion was 2 to 4 μg/kg/min with isoflurane to supplement the opioid. Fifty-one patients (71%) met the criteria for extubation within 6 hours postoperatively; because of surgical practice differences, only 30 patients (59%) were actually extubated. Conclusions: After lorazepam premedication, remifentanil infusion (2-4 μg/kg/min) supplemented intermittently with low inspired concentrations of isoflurane provided an effective anesthetic regimen for coronary artery bypass graft surgery. Early extubation times were feasible after remifentanil continuous infusions (1-5 μg/kg/min) used as the primary anesthetic component intraoperatively and for analgesia (≤1 μg/kg/min) in the immediate postoperative setting. Copyright 2003, Elsevier Science (USA). All rights reserved.

Keywords:  coronary artery bypass graft surgery, remifentanil, responses to surgical stimuli, postoperative recovery

Anesthetic techniques for coronary artery bypass graft (CABG) surgery have in the past relied on an opioid administered intravenously in high doses as the primary anesthetic agent.¹, ², ³ Although high doses of opioids can provide intraoperative hemodynamic stability, the use of high-dose opioids can result in prolonged recovery from anesthesia because of sustained respiratory depression resulting from drug accumulation.⁴, ⁵ In recent years, there has been a move away from the use of high doses of opioids toward balanced anesthesia, largely because of increasing evidence that techniques using lower doses of opioids allow faster postoperative recovery, thereby decreasing the requirement for expensive intensive care unit (ICU) resources.⁶, ⁷ There is also evidence that reducing a patient's need for prolonged ventilatory support in the ICU ultimately improves cardiac function and outcome.⁸, ⁹ To facilitate rapid postoperative recovery without compromising hemodynamic stability or myocardial performance, the ideal opioid should be rapidly titratable and should not accumulate in the body with prolonged administration.

Remifentanil hydrochloride is a 4-anilidopiperidine opioid with highly potent opioid μ-receptor agonist effects, similar to those of other fentanyl derivatives.¹⁰, ¹¹, ¹² Remifentanil is characterized by a short elimination half-life of approximately 3 to 10 minutes because of rapid hydrolysis of the methyl ester linkage by nonspecific blood and tissue esterases¹¹, ¹³ to produce a carboxylic acid metabolite that is essentially inactive (1/4,600 as potent as remifentanil in dogs).¹¹, ¹⁴ Unlike other opioids, the duration of action of remifentanil at a given dose does not increase with increasing duration of administration because it does not accumulate in the body.¹³, ¹⁵, ¹⁶ Remifentanil's short half-life, lack of accumulation after prolonged administration, and organ-independent metabolism make it an ideal agent to provide intense analgesia during surgery followed by rapid and predictable postoperative recovery.

The objectives of this study were to identify a remifentanil dosing regimen that will provide optimal intraoperative hemodynamic stability when used for maintenance of anesthesia in patients undergoing CABG and to evaluate the effects of remifentanil anesthesia and a postoperative regimen of remifentanil plus morphine analgesia on recovery times. One important efficacy endpoint in this study was the proportion of patients in each treatment group showing inadequate anesthesia responses to electrocautery during dissection of the internal mammary artery (IMA). Almost continuous electrocautery of the sternal periosteum provides more prolonged stimulation of an intensity similar to that of sternotomy, which is brief. (That is, the strength duration of stimulation by electrocautery exceeds that of a first-time sternotomy.) Sternal cauterization is used both before and after cardiopulmonary bypass (CPB). Patients were carefully assessed for the occurrence of adverse events throughout the perioperative period and up to the time of hospital discharge.

Back to Article Outline

Methods 

This randomized, open-label, parallel-group study was conducted at 3 centers in the United States. The protocol was approved by an institutional review board at each center, and all patients provided written informed consent before participation. A total of 75 patients were entered into the study, but 3 were removed as a result of protocol violations. Therefore, 72 patients with American Society of Anesthesiologists' status III to IV and a preoperative left ventricular stroke volume (LVSV) ≥50 mL who were scheduled for elective CABG surgery were entered into the study. Female patients were postmenopausal, surgically sterile, or had a negative urine or serum pregnancy test on the day of surgery. Patients were excluded from the study if they had clinically significant cardiac arrhythmia; severe or uncontrolled central nervous system; lung, liver, kidney, or endocrine disease; pulmonary edema or pleural effusion; congestive heart failure; or ejection fraction <35%. Patients were also excluded if they had suffered a myocardial infarction within 3 days of surgery, if they had a history of ethanol or drug abuse, or if they required emergency surgery (<6 hours after failed PTCA) or concurrent cardiac valve repair or replacement. Patients with known hypersensitivity to opioids and those who had received an opioid within the 12 hours preceding the study were also excluded. Eligible patients were randomly assigned to 1 of 3 remifentanil groups, each with a different remifentanil infusion dose for induction of anesthesia: group 1 (1 μg/kg/min), group 2 (2 μg/kg/min), or group 3 (3 μg/kg/min).

All patients continued to receive their routine cardiac medications until the morning of surgery. Patients were premedicated with lorazepam, 40 μg/kg PO 2 hours before surgery. On arrival in the preoperative holding area, patients received morphine, 0.05 mg/kg intravenously (IV), to facilitate intravascular cannulation. Standard leads II and V₅ were placed for continuously monitoring electrocardiogram and heart rate (HR). During the placement of intravascular catheters, 1 or 2 additional doses of lorazepam, 20 μg/kg IV could be given as required for sedation. Under local anesthesia, 2 peripheral venous catheters were inserted for fluid and drug infusions; a radial artery catheter was inserted for intermittent blood sampling and for monitoring arterial blood pressure; and a balloon-tipped thermodilution catheter was inserted for monitoring pulmonary artery pressure, pulmonary artery occlusion pressure (PAOP), and central venous pressure. Cardiac output was measured in triplicate by thermodilution. Oxyhemoglobin saturation was monitored by a pulse oximeter. Respiratory rate and end-tidal carbon dioxide concentration (P_ETCO₂) were monitored with a capnometer. After instrumentation, PAOP was measured, and lactated Ringer's solution was given to increase PAOP to 10 to 15 mmHg. After this fluid adjustment, LVSV was calculated, and patients with LVSV <50 mL were to be excluded from the study (none were).

Before induction of anesthesia, a priming dose of vecuronium (0.01 mg/kg) was given to minimize opioid-induced muscle rigidity, while the patient breathed 100% oxygen for 3 minutes via a mask. Anesthesia was induced with a continuous infusion of remifentanil (initial rates were 1, 2, or 3 μg/kg/min) administered through the side port of the pulmonary artery catheter introducer cannula by an automated infusion pump. The remifentanil dose range selected for this study was based on pharmacokinetic/pharmacodynamic data, indicating that the optimal therapeutic dose would lie in the 1 to 3 μg/kg/min range.¹³, ¹⁴, ¹⁵, ¹⁶ Immediately after loss of consciousness (LOC; defined as lack of response to verbal command and to a noxious stimulus such as pressure on the styloid process), vecuronium, 0.15 mg/kg, was given to facilitate endotracheal intubation. Patients were ventilated with 100% oxygen to maintain P_ETCO₂ between 29 and 36 mmHg. Incremental boluses or an infusion of up to 2 μg/kg/min of vecuronium were given as needed to maintain 1 to 2 twitches to train-of-4 nerve stimulation for the duration of surgery.

During the intraoperative period, patients were continually monitored for hemodynamic changes and sympathetic and somatic responses to surgical stimuli including skin incision, sternotomy, IMA dissection, vascular cannulation and decannulation, and skin closure. Specific signs of light anesthesia were defined as one or more of the following.¹⁷

Hypotension was defined as SBP <15 mmHg below baseline before and after CPB or MAP <50 mmHg for >10 minutes during CPB. Bradycardia was defined as HR <50 beats/min for ≥1 minute.

Leads II and V₅ of a standardized electrocardiogram (1 mm = 0.1 mV) were continually monitored for ST-segment changes, and an ischemic response was defined as a shift from baseline of at least 1.0 mm (0.1 mV) of ST segment elevation or depression lasting ≥1 minute. Other potential signs of ischemia, such as increasing PAOP or regional wall motion abnormalities (in those patients being monitored with transesophageal echocardiography), were also monitored.

The remifentanil infusion rate was maintained at the initial rate unless signs of light anesthesia occurred in response to surgical stimuli. Signs of light anesthesia were treated with bolus doses of remifentanil, 1 or 2 μg/kg, separated by 0.5- to 1-minute intervals. If 2 sequential bolus doses did not control the response, the infusion rate was increased by 1 μg/kg/min every 2 minutes up to a maximum of 2 μg/kg/min above the initial assigned rate. Isoflurane 0.5% to 1.0% was given as a “rescue” anesthetic if the signs of light anesthesia persisted for ≥5 minutes after the remifentanil infusion had been at the maximum rate for that treatment group. Isoflurane administration was maintained until the response was controlled or until the stimulus ended, and then it was discontinued. The remifentanil infusion rate was decreased to the initial rate in decrements of 1 μg/kg/min if hypotension occurred. Tachycardic responses unaccompanied by hypertension or other signs of light anesthesia were to be treated with a β-blocker rather than with incremental increases in the remifentanil infusion rate as tachycardia rarely occurs as the only sign of light anesthesia.¹⁷ (No β-blocker was needed.)

At the end of surgery (defined as tying the last surgical suture), the vecuronium infusion was discontinued, but the final intraoperative remifentanil infusion rate was maintained until after ICU arrival and completion of measurements of vital signs, at which point the infusion rate was reset to 1 μg/kg/min for all patients and supplemental midazolam (1-2 mg IV bolus) was administered as required for sedation. Eligibility for extubation was initially assessed 2 to 3 hours later. Patients who were normothermic, hemodynamically stable, and without excessive bleeding or significant arrhythmia were judged to be eligible for early extubation.

Patients who were not eligible for early extubation received morphine, 0.1 mg/kg IV, and the remifentanil infusion rate was decreased by 50% of the previous rate every 10 minutes to the minimum rate of 0.1 μg/kg/min. The morphine dose was adjusted as necessary, and the remifentanil infusion was discontinued as soon as morphine analgesia was judged to be adequate.

Patients who were eligible for early extubation were given morphine, 0.1 mg/kg IV, and the remifentanil infusion rate was maintained for an additional hour, after which patients were reassessed for eligibility for early extubation. For eligible patients, the remifentanil infusion rate was decreased to 0.5 μg/kg/min, and mechanical ventilation was decreased progressively as long as SpO₂ remained ≥90% and end-tidal carbon dioxide remained ≤50 mmHg. Subsequent reductions in the remifentanil infusion rate by 50% were performed every 10 minutes until respiration was adequate (P_ETCO₂ <50 mmHg and/or respiratory rate ≥8 breaths/min). The remifentanil infusion could be titrated upward by 50% from the current rate, and additional morphine was administered if a patient became agitated or hypertensive before extubation. Extubation was performed when the particular institutional weaning criteria were satisfied. If the remifentanil infusion was not discontinued before extubation, adequate analgesia was maintained for up to 30 minutes after extubation by titrating the remifentanil infusion while transition to a long-term alternative analgesia regimen was achieved. In all cases, the remifentanil infusion was to be discontinued within 6 hours of ICU entry.

During induction of anesthesia, the time to LOC and the duration and severity of any muscle rigidity were noted for each patient. The severity of muscle rigidity was rated by the anesthesiologist on a scale of 0 to 3, where 0 = no clinical evidence of increased muscle tone, 1 = mildly increased muscle tone without impairment of positive-pressure ventilation, 2 = moderate to marked increase in muscle tone that did not prevent positive-pressure ventilation of the lungs, and 3 = severe muscle rigidity and unable to ventilate with positive pressure. The incidence and type of responses to endotracheal intubation were also recorded. During maintenance of anesthesia, the following were recorded for each patient: the proportion of intraoperative time free from any sign of light anesthesia (defined as the absence of somatic, tachycardic, hypertensive, autonomic, or ischemic responses); the incidence of responses to predefined surgical events and remifentanil infusion rates administered, including the number of boluses given and any infusion rate increases or decreases; and the incidence of volatile anesthetic use and reasons for use.

Hemodynamic values were separately recorded at the following times: (1) before induction of anesthesia (control); (2) at 1, 3, 5, 10, and 15 minutes after starting the remifentanil infusion; (3) at LOC; (4) every 15 minutes during the prebypass and postbypass periods; (5) every 10 to 15 minutes during CPB; and (6) immediately before and at 1, 3, and 5 minutes after tracheal intubation, skin incision, sternotomy, IMA dissection, aortic cannulation, aortic decannulation, and skin closure. If SBP readings in the period after induction (but before endotracheal intubation) were persistently at least 10 mmHg above the lowest preoperative reading, then the higher value was set as the new control value throughout the intraoperative period. The selected control was used to determine the SBP value that was considered to be a hypertensive response to a surgical stimulus. Cardiac index was calculated at each time CO was measured during the normal course of patient management in the perioperative period, and the incidence of cardiac index <2.0 L/min/m² was recorded.

The times from ICU entry to the first remifentanil infusion rate decrease, spontaneous respiration, eye opening, the lifting of any extremity to command, tracheal extubation, ICU discharge, and hospital discharge were recorded.

The incidence of adverse events (defined as any untoward medical event, potentially drug related or not) was recorded throughout the study. Muscle rigidity, nausea, vomiting, blood pressure, HR changes treated with vasodilators or vasoconstrictors, and ischemic episodes requiring pharmacologic intervention were all recorded as adverse events. On the first and fifth postoperative days, patients were interviewed for recall of perioperative events and their assessment of the anesthetic technique. In addition, blood samples for creatine phosphokinase isoenzyme (CK-MB) analysis were collected every 8 hours for the first 48 hours postoperatively. Relative index (RI) expresses the ratio of CK-MB to total CK. Assessing RI in addition to CK-MB is thought to distinguish more specifically those increases caused by cardiac muscle cell death from increases caused by skeletal muscle cell death.

Computer-generated randomization codes were issued at each site and assigned sequentially to each randomized patient who met all the screening criteria. Each treatment number identified the randomly assigned treatment dose level administered to each patient.

Statistical analyses were performed using SAS version 6.08 (SAS Inc., Cary, North Carolina), and p values ≤0.05 were considered statistically significant. The Fisher exact test was used to compare differences between groups in the proportion of patients with muscle rigidity at induction. The Cochran-Mantel-Haenszel test, adjusted for institutional site, was used to compare differences between groups in the proportion of patients given volatile anesthetic at any time for rescue, the proportion of patients with any light anesthesia response, and the proportion of patients with no occurrence of signs of inadequate anesthesia. Logistic regression analysis, adjusted for treatment and institutional site, was used to compare differences between groups in the number of patients exhibiting signs of light anesthesia at the times of the predefined surgical stimuli.

The weighted means of the remifentanil infusion rates for the following intervals were calculated for each treatment group: induction (from the start of remifentanil infusion until after tracheal intubation), prebypass (postintubation until initiation of CPB), bypass (duration of CPB), and postbypass (up to the time of entry into the ICU). The weighted mean remifentanil infusion rate was calculated as the area under the curve over time for each interval divided by the duration of the measurement, assuming a step-wise distribution.

Hemodynamic values for each patient group were averaged for the times before each predefined surgical stimulus and for the maximum values observed in the 5-minute interval after each stimulus. Within each treatment group the paired t-test was used to compare the pre-event means to the baseline means and to compare the maximum post-event means to the corresponding pre-event means for intubation, sternotomy, and IMA dissection. The F-test was used to compare differences between groups in changes from pre-event to postevent values for sternotomy and IMA dissection.

Back to Article Outline

Results 

Results for all 72 patients randomized into the protocol are included in the data analyses. Treatment groups were comparable with regard to demographic and surgical background characteristics (Table 1).

Table 1. Patient characteristics and details of surgery

Abbreviation: LVEF, Left ventricular ejection fraction.

Induction phase 

The mean preoperative lorazepam dose used (IV and PO) was similar among the 3 treatment groups (Table 2). The interval between the last lorazepam dose and the initiation of the remifentanil infusion was more than 30 minutes in the majority of patients. The time to LOC and the incidence of light anesthesia responses were not related to the total dose of lorazepam. The median time to LOC ranged from 2.5 to 4.0 minutes (range, 1-8 minutes). Two patients, one in group 1 and one in group 3, required supplemental etomidate as “rescue” medication to induce LOC. The proportion of patients with a hemodynamic response to tracheal intubation (Table 2) appeared to decrease at the higher remifentanil infusion rates, although these differences were not statistically significant (p > 0.33). It is noteworthy that the remifentanil infusion rate had been increased by 70%, 15%, and 13% in groups 1, 2, and 3, respectively, at the time of chest skin incision reflecting the need to control responses to tracheal intubation (and the short time between intubation and incision in which to meet protocol mandates to decrease remifentanil infusion rates).

Table 2. Summary of induction phase results

NOTE. There are no statistically significant differences between the groups for any of the values shown.

Abbreviation: LOC, loss of consciousness.

Maintenance phase 

From skin incision to sternotomy, the remifentanil infusion rates were essentially unchanged (compare last line of Table 2 and first line of Table 3). With sternotomy, 17% to 25% of patients in the 3 groups showed a hypertensive response and treatment of that response is reflected in the higher remifentanil infusion rates at the time of IMA dissection (Table 3). Again, a larger proportional increase in remifentanil infusion rate was required in group 1 with the lowest infusion rate (ie, 28%, 17%, and 11% increases in groups 1, 2, and 3, respectively) And the incidence of isoflurane “rescue” (0.5%-1% inspired) was significantly greater for group 1 compared with groups 2 and 3 (Table 3).

Table 3. Summary of maintenance phase results

Abbreviation: CPB, cardiopulmonary bypass.

It is important to note that high percentages of patients in all 3 groups required remifentanil boluses and increased infusion rates to control hemodynamic responses to surgical stimuli in the pre-CBP period. The percent of the total intraoperative time without any response (Table 3) was virtually the same in all 3 groups, suggesting that the treatment of responses was equally successful in all 3 groups. In fact, no patient received any drug other than remifentanil ± isoflurane to treat any hypertensive or tachycardic responses to surgical stimuli, and the hemodynamic stability overall was similar and satisfactory in all 3 groups of patients (Fig 1).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Mean diastolic arterial pressure and mean heart rates are displayed for each event. Data are means ± standard error of mean. *p < 0.05 versus control for groups 1, 2, and 3. +p < 0.05 versus control for group 1. ^@p < 0.05 versus control for groups 1 and 3. Measurement times: Control (baseline before initiation of remifentanil), Pre-INT (after induction and before intubation), INT (maximum value ≤ 5 minutes after intubation), Pre-STN (presternotomy), STN (maximum value ≤5 minutes after sternotomy), Pre-IMA (pre-IMA dissection), IMA (maximum value ≤5 minutes after IMA dissection), Pre-BYP (prebypass), Pre-DCN (preaortic decannulation), DCN (maximum value ≤5 minutes after aortic decannulation), Pre-SKC (preskin closure), SKC (maximum value ≤5 minutes after skin closure). BP valves pre-BYP (bypass) reflect the use of vasodilators to lower systemic BP just before aortic cannulation. HR values at pre-DCN and later reflect electrical pacing.

By far, the most frequent indication of light anesthesia was a 15-mmHg or greater increase in blood pressure (Table 4).

Table 4. Profile of light anesthesia responses during each intraoperative period

Note. The sums of the signs of light anesthesia exceed the total number of responders because some patients showed more than one sign. All values are number of patients (%).

Abbreviations: CPB, cardiopulmonary bypass; NA, not applicable.

Before CPB, the frequency of these hypertensive responses was greater in group 1 than in groups 2 and 3, which had a similar incidence.

IMA dissection is the most intense and prolonged noxious stimulus in CABG surgery. The percentage of patients showing any response to this phase of surgery was 22% to 26% for all 3 groups of patients, and the mean remifentanil infusion rates before and shortly after IMA dissection (ie, aortic cannulation) ranged from 2.3 to 4.0 μg/kg/min supplemented with isoflurane (Table 3).

Blood pressures (BP) were maintained within the narrow limits prescribed by the experimental protocol for each patient, and there were no clinically important or statistically significant differences among the 3 groups of patients (Fig 1). Blood pressures were generally lower after induction of anesthesia by remifentanil superimposed on lorazepam premedication. They were even lower immediately before CPB when the routine clinical practice is to lower BP deliberately with a vasodilator in anticipation of cannulation of the aorta. It should be noted that when BP increases occurred in response to surgical stimulation, the values rarely exceeded the pre-remifentanil induction values and the duration of hypertensive responses was limited to 5 minutes by protocol-mandated increases in remifentanil ± the addition of isoflurane.

Mean HRs were maintained within 20% of baseline values before bypass (range, 57-73 beats/min) but increased after bypass (range, 89-96 beats/min) because cardiac pacing was usually used during the postbypass period. Forty-two patients (50%) had at least 1 period of cardiac pacing during the post-bypass period, with paced rates in the 82 to 90 beats/min range for 32 patients, and in the range of 91 to 100 beats/min for 10 patients.

ICU recovery phase 

The median time for the following milestones were 21 minutes, open eyes to command; 23 minutes, lift an extremity; and 35 minutes, first spontaneous breath. The mean ± SD dose of morphine used during the first 6 hours in the ICU was 0.21 ± 0.08 mg/kg.

A total of 30 patients (42%) were extubated within 6 hours (median, 4 hours; range, 1-6 hours) after entry into the ICU, although 71% were judged to be eligible for extubation. The proportion of patients who were eligible for and who were actually extubated within 6 hours varied by center (Fig 2).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  The proportion of patients who met initial criteria for early tracheal extubation compared with those who actually underwent extubation within 6 hours after surgey described by investigator-site. Standard ICU practices that precluded “early” extubation as defined in the protocol, resulted in no patients being extubated in less than 6 hours by the investigator at site #3.

Surgical ICU practices in one institution prevented early extubation despite protocol approval by the institutional review board of that institution. The times to ICU and to hospital discharge did not differ because institutional policies and procedures regarding fast-tracking were not yet fully implemented.

Safety 

The most commonly reported adverse events across all study phases combined related to the cardiovascular system. The most common adverse event was protocol-defined hypertension in response to stimulation (reported in 96%, 92%, and 92% of patients in groups 1, 2, and 3, respectively). The most frequent adverse events considered to be related to remifentanil during induction were expected opioid effects: skeletal muscle rigidity (48%, 36%, and 62% in groups 1, 2, and 3, respectively) and hypotension (16%, 28%, and 19% in groups 1, 2, and 3, respectively). Hypotension (36%, 12%, and 23%) was the most commonly reported adverse event during maintenance of anesthesia, usually occurring after the response to noxious stimulation was controlled by increased remifentanil doses and the administration of isoflurane. No patient experienced significant oxyhemoglobin desaturation or adverse sequelae because of muscle rigidity. No patient recalled intraoperative events, nor did any patient spontaneously report intraoperative awareness.

The 1 death reported during the study was not considered to be related to the anesthetic technique. This 77-year-old man with a history of coronary artery disease and hypertension was randomized to receive an initial remifentanil infusion rate of 2 μg/kg/min during surgery. Surgery was complicated by an extended period of time on cardiac bypass support (3.5 hours, aorta cross-clamped 1.2 hours) and poor quality grafts. The patient was extubated the morning after surgery and scheduled for discharge from the ICU that afternoon. During preparation for transfer the patient became hypertensive, which resulted in aortic rupture (24 hours after discontinuation of remifentanil) and cardiac arrest from exsanguination.

One patient had ventricular failure requiring the use of an intra-aortic balloon pump postoperatively, and 2 patients (1 patient from group 2 and 1 from group 3) had myocardial infarctions (CK-MB ≥75 IU/L or ≥30 ng/mL with an RI >4.0 and new Q waves). Both of these patients were having repeat CABG, and one had intraoperative signs of ischemia. Overall, 10 patients in group 1 and 7 patients each in groups 2 and 3 had intraoperative signs of ischemia. None of these events were judged by the investigators to be related to the anesthetic technique.

Back to Article Outline

Discussion 

The principal objective of this study was to determine the optimal remifentanil dose range, providing safe and effective control of hemodynamic and somatic responses to surgical stimuli, in patients undergoing CABG surgery. This study shows that remifentanil is well tolerated and can be administered safely as the primary anesthetic agent in doses up to 5.9 μg/kg/min in patients undergoing CABG surgery. This study indicates that a 2 to 4 μg/kg/min infusion rate of remifentanil will substantially suppress hemodynamic and somatic responses to surgical stimuli during cardiac surgery after lorazepam premedication. However, additional bolus doses of remifentanil and a volatile anesthetic for rescue were required in the majority of patients to maintain intraoperative hemodynamic values within a narrow range (eg, BP ± 15 mmHg). The results indicate that a potent, short-acting opioid like remifentanil can be used in high doses for cardiac surgical patients without the penalty of drug accumulation and still allow for early extubation. Although high doses of remifentanil were continued throughout surgery and into the ICU, the majority of eligible patients were extubated (or could have been extubated except for institutional restrictions) within 6 hours after ICU entry.

A primary opioid anesthetic protocol was used in this study to evaluate the maximum efficacy of remifentanil as an anesthetic component for cardiac surgery requiring CPB. The maximum remifentanil infusion rates allowed (3-5 μg/kg/min) were up to 5 times the recommended anesthetic dose for general surgery with propofol¹⁸ or isoflurane. Fentanyl concentrations in the range of 10 to 15 ng/mL of plasma produce the maximal reduction in isoflurane concentration requirements during CABG, and these concentrations are maintained by fentanyl infusion rates of 0.15 to 0.3 μg/kg/min.²¹, ²² Given approximately equal anesthetic potencies of remifentanil and fentanyl,², ¹³, ¹⁹ these infusion rates of remifentanil exceed those typically used for fentanyl by 10- to 16-fold.²⁰, ²¹

The variability among patients in the ability of remifentanil to suppress responses to surgical noxious stimulation is typical for opioids, as are the differences in the doses required to suppress responses to stimuli of different intensities.¹⁷, ²² In the authors' experience with different anesthetics, there is an incrementally increasing dose requirement as cardiac operations progress from skin incision to sternotomy to IMA dissection. The last has the greatest intensity-duration product exceeding that of sternotomy, which is very intense but also very brief compared with sustained electrocauterization of the periosteum and other tissues of the IMA bed. Remifentanil infusion rates of 2 to 4 μg/kg/min combined with low inspired concentration of isoflurane (0.5%-1%) appear to be optimal in terms of limiting the number of patients (22%-26% in all 3 groups) responding to IMA dissection. This pattern of responsiveness to combinations of fentanyl-type opioids and isoflurane is evident in the observations of Thomson et al.²²

The similarities in the overall observations (ie, intraoperative time without responses, hemodynamic trends) among the 3 groups of patients in this study are to be expected given the dynamic design of the experimental protocol. That is, the anesthesiologist caring for the patient monitored and promptly treated any and all responses (somatic, autonomic, and hemodynamic as defined in the protocol) to noxious stimulation. The treatment regimen defined in the protocol required increases in the remifentanil infusion rates, and the brief intervals between successive surgical stimuli did not allow time to reduce the remifentanil infusion rates and to reassess patient responsiveness. Hence, there was a persistent gradation of remifentanil infusion rates among the 3 groups. Nevertheless, it seems clear that infusion rates less than 2 μg/kg/min were associated with more episodes of responses and required more frequent rescue maneuvers than with infusion rates in the range of 2 to 4 μg/kg/min. Infusion rates above 4 μg/kg/min were rarely indicated.

The mean hemodynamic values observed intraoperatively compare favorably with those reported in the literature for patients undergoing cardiac surgery.²³, ²⁴, ²⁵ IMA dissection was an important efficacy endpoint because the electrocautery involved in this process is an intense and sustained noxious stimulus during CABG surgery, both before and after CPB. Remifentanil infusion rates were highest during IMA dissection, and isoflurane was most often administered to control hemodynamic responses during this event. The results of this study show that remifentanil reached a typical opioid plateau effect.²⁶ The ceiling dose for remifentanil appears to be between 2 and 4 μg/kg/min because higher infusion rates did not significantly decrease the need for isoflurane rescue. Significantly fewer patients in the remifentanil 2 μg/kg/min (63%) and 3 μg/kg/min (56%) groups required isoflurane rescue compared with the 1 μg/kg/min group (91%). Overall, two thirds of the patients in the 2 and 3 μg/kg/min groups (67% and 68%) received at least 1 period of supplemental volatile anesthetic to treat or prevent hypertension during maintenance of anesthesia. These requirements for supplementation of remifentanil are compatible with previous studies that have shown that very high doses of fentanyl-type opioids do not predictably blunt all hemodynamic and somatic responses to surgical stimuli in all patients undergoing CABG surgery.²⁰, ²¹, ²², ²⁶, ²⁷, ²⁸, ²⁹

The results of the present study indicate that, unlike other high-dose opioid regimens, extubation in less than 6 hours after surgery can be achieved with a high-dose opioid regimen using remifentanil, which has a predictably rapid offset of action. To implement significant changes in patient care (eg, early extubation), the “culture” of the institution has to change along with the drugs and techniques that facilitate the changes (Fig 2). With high-dose fentanyl (95-162 μg/kg), reported extubation times ranged from approximately 17 to 29 hours, and with high-dose sufentanil (18-20 μg/kg),²⁸, ³⁰ extubation times ranged from approximately 9 to 23 hours.²⁹, ³⁰, ³¹, ³², ³³ A study by Cheng et al³⁴ showed that extubation could be achieved in a median time of 4 hours using an anesthetic regimen limiting the fentanyl dose to a maximum of 15 μg/kg and supplementing it with propofol, 2 to 6 μg/kg/min, and isoflurane. In the same study, a high-dose fentanyl regimen (50 μg/kg/min) prolonged extubation time and other recovery times. At a time when the emphasis is on rapid recovery from cardiac surgery to reduce costs, these results indicate that remifentanil should be a useful agent in the armamentarium of anesthetic drugs available for patients undergoing CABG surgery, providing intense analgesia intraoperatively and during the early postoperative period, while still allowing early extubation. The cost-effectiveness of a high-dose remifentanil anesthetic regimen remains to be determined.

Although the timing of extubation after cardiac surgery is dependent on a number of factors,³¹, ³⁵ including differences in institutional practices, a high proportion of patients were successfully extubated within 6 hours after CABG surgery in this study using a high-dose remifentanil infusion regimen.

Adverse events noted in this study, including skeletal muscle rigidity and hypotension, were expected opioid effects. Muscle rigidity on induction is expected with the use of opioids and at the infusion rates used in this study.², ³ The incidence and severity of muscle rigidity on induction increased with increasing remifentanil infusion rate. For this reason, opioids, including remifentanil, used at high doses, may be unsuitable as sole agents for induction of anesthesia except in critically unstable patients. A more appropriate and widely accepted regimen would incorporate the administration of remifentanil either simultaneously or after administration of a hypnotic agent for inducing LOC. No patient experienced any recall of intraoperative events during this study, which included lorazepam premedication.

The 1 death reported in this study was caused by events unrelated to remifentanil administration (an aortic tear the day after surgery). This patient had many of the factors that have been identified with increased risk in cardiac surgery, including age >75 years, 2 previous incidences of MI, unstable angina, obesity, extended bypass time, poor graft quality, and ventricular pacing.

No correlation between intraoperative ischemia and perioperative myocardial infarction (MI) was observed: 33% of patients had some signs of perioperative ischemia, but just one of these patients had a postoperative MI. The other patient with postoperative MI did not have perioperative ischemia. These results are consistent with those reported by Slogoff and Keats³⁶ for patients anesthetized with sufentanil or volatile anesthetics.

In summary, a high-dose remifentanil regimen was shown to provide anesthetic conditions similar to those reported using other high-dose opioid anesthetic regimens in cardiac surgical patients. It is notable that the use of remifentanil allowed for earlier extubation than conventional high-dose opioid regimens would generally allow.²⁰, ³⁰, ³¹, ³², ³³ Further large studies are required to assess the safety, efficacy, and cost-effectiveness of a high-dose remifentanil regimen in comparison with other, commonly used opioid regimens for cardiac surgery. However, recently published clinical experience would indicate that remifentanil has been successfully used for cardiac surgery and allowed fast-tracking of eligible cardiac surgical patients.³⁷, ³⁸, ³⁹, ⁴⁰, ⁴¹

Back to Article Outline

Acknowledgements 

The authors thank Amy Peng (GlaxoWellcome) for her contributions and Beth Sherrill (GlaxoWellcome) for statistical analyses and review. Thomas McSweeney, Marshall Naden, and Dianne Lee are thanked for their help in preparation of this manuscript.

Back to Article Outline

References 

1. Bovill JG, Warren PJ, Schuller JL, et al.  Comparison of fentanyl, sufentanil, and alfentanil anesthesia in patients undergoing valvular heart surgery. Anesth Analg. 1984;63:1081–1086
   • View In Article
   • MEDLINE
2. Hall RI. The pharmacokinetic behavior of opioids administered during cardiac surgery. Can J Anaesth. 1991;38:747–756
   • View In Article
   • MEDLINE
   • CrossRef
3. Bovill JG, Sebel PS, Stanley TH. Opioid analgesics in anesthesia: with special reference to their use in cardiovascular anesthesia. Anesthesiology. 1984;61:731–755
   • View In Article
   • MEDLINE
   • CrossRef
4. Cartwright P, Prys-Roberts C, Gill K, et al.  Ventilatory depression related to plasma fentanyl concentrations during and after anesthesia in humans. Anesth Analg. 1983;62:966–974
   • View In Article
   • MEDLINE
5. Shafer SL, Varvel JR. Pharmacokinetics, pharmacodynamics, and rational opioid selection. Anesthesiology. 1991;74:53–63
   • View In Article
   • MEDLINE
   • CrossRef
6. Cheng DC. Pro: early extubation after cardiac surgery decreases intensive care unit stay and cost. J Cardiothorac Vasc Anesth. 1995;9:460–464
   • View In Article
   • Full-Text PDF (532 KB)
   • CrossRef
7. Karski JM. Practical aspects of early extubation in cardiac surgery. J Cardiothorac Vasc Anesth. 1995;9(suppl 1):30–33
   • View In Article
   • MEDLINE
8. Foster GH, Conway WA, Pamulkov N, et al.  Early extubation after coronary artery bypass: Brief report. Crit Care Med. 1984;12:994–996
   • View In Article
   • MEDLINE
   • CrossRef
9. Gall SA. Olsen CO, Reves JG, et al: Beneficial effects of endotracheal extubation on ventricular performance. J Thorac Cardiovasc Surg. 1988;95:819–827
   • View In Article
   • MEDLINE
10. James MK, Feldman PL, Schuster SV, et al.  Opioid receptor activity of GI87084B, a novel ultra-short-acting analgesic, in isolated tissues. J Pharmacol Exp Ther. 1991;259:712–718
    • View In Article
    • MEDLINE
11. Feldman PL, James MK, Brackeen MF, et al.  Design, synthesis and pharmacological evaluation of ultrashort- to long-acting opioid analgesics. J Med Chem. 1991;34:2202–2208
    • View In Article
    • MEDLINE
    • CrossRef
12. Michelsen LG, Salmenpera M, Hug CC, et al.  Anesthetic potency of remifentanil in dogs. Anesthesiology. 1996;84:865–872
    • View In Article
    • MEDLINE
    • CrossRef
13. Glass PSA, Hardman D, Kamiyama Y, et al.  Preliminary pharmacokinetics and pharmacodynamics of an ultra-short-acting opioid: remifentanil (GI87084B). Anesth Analg. 1993;77:1031–1040
    • View In Article
    • MEDLINE
14. Cunningham FE, Hoke JF, Muir KT, et al.  Pharmacokinetic/pharmacodynamic evaluation of remifentanil, GR90291, and alfentanil. Anesthesiology. 1995;83:376A
    • View In Article
15. Westmoreland CL, Hoke JF, Sebel PS, et al.  Pharmacokinetics of remifentanil (GI87084B) and its major metabolite (GI90291) in patients undergoing elective inpatient surgery. Anesthesiology. 1993;79:893–903
    • View In Article
    • MEDLINE
    • CrossRef
16. Egan TD, Lemmens HJM, Fiset P, et al.  The pharmacokinetics of the new short-acting opioid remifentanil (GI87084B) in healthy adult male volunteers. Anesthesiology. 1993;79:881–892
    • View In Article
    • MEDLINE
    • CrossRef
17. Ausems ME, Vuyk J, Hug CC, Stanski DR. Comparison of a computer-assisted infusion versus intermittent bolus administration of alfentanil as a supplement to nitrous oxide for lower abdominal surgery. Anesthesiology. 1988;68:851–861
    • View In Article
    • MEDLINE
    • CrossRef
18. Hogue CW, Bowdle TA, O'Leary CE, et al.  A multicenter evaluation of total intravenous anesthesia with remifentanil and propofol for elective inpatient surgery. Anesth Analg. 1996;83:279–285
    • View In Article
    • MEDLINE
    • CrossRef
19. Sebel PS, Hoke JF, Westmoreland C, et al.  Histamine concentrations and hemodynamic responses after remifentanil. Anesth Analg. 1995;80:990–993
    • View In Article
    • MEDLINE
    • CrossRef
20. Wynands JE, Townsend GE, Wong P. Blood pressure response and plasma fentanyl concentrations during high- and very high-dose fentanyl anesthesia for coronary artery surgery. Anesth Analg. 1983;62:661–665
    • View In Article
    • MEDLINE
21. Hall RI, Moldenhauer CC, Hug CC. Fentanyl plasma concentrations maintained by a simple infusion scheme in patients undergoing cardiac surgery. Anesth Analg. 1993;76:957–963
    • View In Article
    • MEDLINE
22. Thomson IR, Henderson BT, Singh K, Hudson RJ. Concentration-response relationships for fentanyl and sufentanil in patients undergoing coronary artery bypass grafting. Anesthesiology. 1998;89:852–861
    • View In Article
    • MEDLINE
    • CrossRef
23. Edde RR. Hemodynamic changes prior to and after sternotomy in patients anesthetized with high-dose fentanyl. Anesthesiology. 1981;55:444–446
    • View In Article
    • MEDLINE
    • CrossRef
24. Sebel PS, Bovill JG, Schellekens APM, Hawker CD. Hormonal effects of high-dose fentanyl anesthesia: A study in patients undergoing cardiac surgery. Br J Anaesth. 1981;53:941–948
    • View In Article
    • MEDLINE
25. Parsons RS, Jones RM, Wrigley SR, et al.  Comparison of desflurane and fentanyl-based anaesthetic techniques for coronary artery bypass surgery. Br J Anaesth. 1994;72:430–438
    • View In Article
    • MEDLINE
26. Hug CC. Does opioid “anesthesia” exist?. Anesthesiology. 1990;73:1–4
    • View In Article
    • MEDLINE
    • CrossRef
27. Philbin DM, Rosow CE, Schneider RC, et al.  Fentanyl and sufentanil anesthesia revisited: how much is enough?. Anesthesiology. 1990;73:5–11
    • View In Article
    • MEDLINE
    • CrossRef
28. Stanley TH, Berman L, Green O, Robertson D. Plasma catecholamines and cortisol responses to fentanyl-oxygen anesthesia for coronary artery operations. Anesthesiology. 1980;53:250–253
    • View In Article
    • MEDLINE
    • CrossRef
29. de Lange S, Stanley TH, Boscoe MJ, et al.  Catecholamine and cortisol responses to sufentanil-O₂ and alfentanil-O₂ anesthesia during coronary artery surgery. Can Anaesth Soc J. 1983;30:248–254
    • View In Article
    • MEDLINE
    • CrossRef
30. Howie MB, McSweeney TD, Lingam RP, Maschke SP. A comparison of fentanyl-O₂ and sufentanil-O₂ for cardiac anesthesia. Anesth Analg. 1985;64:877–887
    • View In Article
    • MEDLINE
31. Sanford TJ, Smith NT, Dec-Siver H, Harrison WK. A comparison of morphine, fentanyl, and sufentanil anesthesia for cardiac surgery: Induction, emergence, and extubation. Anesth Analg. 1986;65:259–266
    • View In Article
    • MEDLINE
32. Howie MB, Smith DF, Reilley TE, et al.  Postoperative course after sufentanil or fentanyl anesthesia for coronary artery surgery. J Cardiothorac Vasc Anesth. 1991;5:485–489
    • View In Article
    • MEDLINE
    • CrossRef
33. Mora CT, Dudek C, Torjman MC, White PF. The effects of anesthetic technique on the hemodynamic and recovery profile in coronary revascularization patients. Anesth Analg. 1995;81:900–910
    • View In Article
    • MEDLINE
    • CrossRef
34. Cheng DC, Karski J, Peniston C, et al.  Morbidity outcome in early versus conventional tracheal extubation after coronary artery bypass grafting: A prospective randomized controlled trial. J Thorac Cardiovasc Surg. 1996;112:755–764
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (907 KB)
    • CrossRef
35. Quasha AL, Loeber W, Freeley TW, et al.  Postoperative respiratory care: A controlled trial of early and late extubation following coronary artery bypass grafting. Anesthesiology. 1980;52:135–141
    • View In Article
    • MEDLINE
    • CrossRef
36. Slogoff S, Keats AS. Randomized trial of primary anesthetic agents on outcome of coronary artery bypass operations. Anesthesiology. 1989;70:178–188
    • View In Article
37. Guarracino F, Bossi A, Tedesco M, et al.  Remifentanil/propofol anaesthesia for myocardial revascularisation: Our preliminary experience. Br J Anaesth. 1998;80(suppl 2):2
    • View In Article
38. Lehmann A, Boldt J, Zeitler C, et al.  Total intravenous anaesthesia with remifentanil and propofol for implantation of cardioverter-defribrillators in patients with severely reduced left ventricular function. J Cardiothorac Vasc Anaesth. 1999;13:15–19
    • View In Article
39. Olivier P, Sirieix D, Dassier P, et al.  Continuous infusion of remifentanil and target-controlled infusion of propofol for patients undergoing cardiac surgery: A new approach for scheduled early extubation. J Cardiothorac Vasc Anesth. 2000;14:29–35
    • View In Article
    • Abstract
    • Full-Text PDF (661 KB)
    • CrossRef
40. Howie MB, Cheng D, Newman MF, et al.  A randomised double-blind multicenter comparison of remifentanil versus fentanyl when combined with isoflurane/propofol for early extubation in coronary artery bypass graft surgery. Anesth Analg. 2001;92:1084–1093
    • View In Article
    • MEDLINE
41. Cheng DCH, Newman MF, Duke P, et al.  The efficacy and resource utilization of remifentanil and fentanyl in fast-track coronary artery bypass surgery: A prospective randomised, double-blinded controlled, multi-center trial. Anesth Analg. 2001;92:1094–1102
    • View In Article
    • MEDLINE