Ambulatory Anesthesiology

Section Editor: Peter S. A. Glass

Intraoperative Esmolol Infusion in the Absence of Opioids Spares Postoperative Fentanyl in Patients Undergoing Ambulatory Laparoscopic Cholecystectomy
Vincent Collard, MD* Giovanni Mistraletti, MD* Ali Taqi, MD Juan Francisco Asenjo, MD* Liane S. Feldman, MD Gerald M. Fried, MD Franco Carli, MD, MPhil*
BACKGROUND: The use of opioids during ambulatory surgery can delay hospital discharge or cause unexpected hospital admission. Preliminary studies using an intraoperative continuous infusion of esmolol in place of an opioid have inconsistently reported a postoperative opioid-sparing effect. In this study, we compared esmolol versus either intermittent fentanyl or continuous remifentanil on postoperative opioid-sparing, side effects, and time of discharge. METHODS: Ninety patients (consisting of three groups) were enrolled in this prospective, randomized, and observer-blinded study. The control group (n 30) received intermittent doses of fentanyl, the esmolol group (n 30) received a continuous infusion of esmolol (515 g kg 1 min 1) and no supplemental opioids during surgery, and the remifentanil group (n 30) received a continuous infusion of remifentanil (0.1 0.5 g kg 1 min 1). General anesthesia was standardized, and adjuvant medications included acetaminophen, ketorolac, local anesthetics in the skin incisions, dexamethasone, and droperidol. Postoperative analgesia included fentanyl. RESULTS: The amount of fentanyl in the postanesthesia care unit was significantly less in the esmolol group, 91.5 42.7 g, compared with the other two groups, remifentanil, 237.8 54.7 g, control, 168.1 96.8 g (P 0.0001). The incidence of nausea was more frequent in the control (66.7%) and remifentanil (67.9%) groups compared with the esmolol group (30%) (P 0.01). The esmolol group reached the White-Song score of 12 of 14 faster than the remifentanil group (P 0.01), and left the hospital 45 60 min earlier (P 0.004). CONCLUSIONS: Intraoperative IV infusion of esmolol contributes to a significant decrease in postoperative administration of fentanyl and ondansetron and facilitates earlier discharge.
(Anesth Analg 2007;105:125562)

dvances in surgical technology, anesthesia, and perioperative care have made it possible to perform laparoscopic cholecystectomy on an outpatient basis with significant reduction in procedural costs and favorable levels of patient satisfaction (1). In spite of the high success rate of ambulatory laparoscopic cholecystectomy, the most common side effects reported with this surgery are postoperative nausea and vomiting (PONV), pain, medical complications, and urinary retention, leading to an unanticipated admission rate of approximately 5% (2).

From the Departments of *Anesthesia and Surgery, SteinbergBernstein Centre for Minimally Invasive Surgery, McGill University Health Centre, Montreal, Quebec, Canada. Accepted for publication July 17, 2007. Dr. G. Mistraletti was awarded a scholarship from the University of Milan and the Societa Italiana di Anestesia, Analgesia, Rianima` zione e Terapia Intensiva. Address correspondence and reprint requests to Franco Carli, Department of Anesthesia, McGill University Health Centre, Room D10.144, Montreal, Quebec, Canada. Address e-mail to franco.carli@mcgill.ca. Copyright 2007 International Anesthesia Research Society
DOI: 10.1213/01.ane.0000282822.07437.02

One of the major challenges of ambulatory anesthesia is the ability to provide effective postoperative analgesia and hemodynamic stability, thus allowing patients to be safely discharged (3). Opioids have been widely used in the ambulatory setting; however, they can be associated with many side effects and delayed discharge (4). In addition, opioids do not offer the quality of postoperative pain relief that one would expect, probably because of the pain specific to laparoscopic cholecystectomy. It has been shown that the acute pain after laparoscopic cholecystectomy has some characteristics not shared by other laparoscopic procedures. Three components have been described: incisional pain as dominant, visceral deep pain, and shoulder referred pain (5). This pain presentation would therefore require that treatment not be limited to one intervention, and a multimodal therapy would likely provide better quality of pain relief, spare opioids, and thus facilitate the recovery process. Although the emphasis has been on postoperative analgesia with multimodal therapy, opioids are predominantly used during the intraoperative period. The use of short-acting opioids such as remifentanil
1255

Vol. 105, No. 5, November 2007

Figure 1. Randomization of the patients.

has been suggested as the treatment of choice for intense painful stimuli caused by pneumoperitoneum and to ensure rapid recovery without increasing opioid-related side effects (6,7). -Adrenergic receptor antagonists have been used during surgery with the intention to attenuate the stress response and decrease unwanted perioperative hemodynamic changes (8). Esmolol, an ultrashort-acting cardioselective 1-adrenergic receptor antagonist, has been proposed as an alternative to intraoperative use of opioids, and found to facilitate the fast-tracking process and speed home-readiness of patients undergoing outpatient surgery (9 11). The present prospective, randomized study was designed to compare the effect of intraoperative IV esmolol, remifentanil, and fentanyl on postoperative opioid sparing and the incidence of side effects in patients scheduled for laparoscopic cholecystectomy. It was hypothesized that the group receiving intraoperative esmolol infusion would benefit from a significant opioid-sparing effect in the postoperative period.

leaving 90 patients suitable to be enrolled in this investigation (Fig. 1). Patients were instructed before surgery in the use of the verbal rating scale (VRS). They were also informed that they would receive a call 24 h after the surgery and would be asked their VRS score at that time as well as amounts of medications used. Before induction of anesthesia, the patients were randomly assigned using a computer-generated block randomization schedule, to compose three equal groups of 30 patients each: the control group receiving fentanyl as the only intraoperative opioid, the esmolol group receiving a continuous infusion of esmolol in place of an opioid throughout surgery, and the remifentanil group receiving a continuous infusion of remifentanil as the only intraoperative opioid. The study (GEN#04-051) was approved by the McGill University Health Centre Ethics Board, and written consent was obtained from all subjects.

Anesthesia, Analgesia, and Surgical Care

All patients were anesthetized by three anesthesiologists who were instructed to follow the study design, but were not involved in the preoperative and postoperative assessment. On arrival at the operating room, baseline heart rate, arterial blood pressure, and oxygen saturation were measured. Premedication was with IV midazolam 0.03 mg/kg. After administration of fentanyl 1.0 g/kg (control group), esmolol 1.0 mg/kg (esmolol group), or remifentanil (remifentanil group) 1.0 g/kg, general anesthesia was induced with propofol 2.5 mg/kg and tracheal intubation facilitated with rocuronium 0.8 mg/kg. Throughout surgery, the patients received regular fentanyl boluses of 50 g (control group) every 30 min as scheduled or esmolol infusion of 515 g kg 1 min 1 (12) (esmolol group) or remifentanil infusion of 0.1 0.50 g kg 1 min 1 (remifentanil group). Anesthesia was maintained in all groups with desflurane at an end-tidal concentration adjusted between 4% and 8%. This range of concentrations was chosen after a pilot study in which five patients
ANESTHESIA & ANALGESIA

METHODS
Patients
The study was conducted between March and December 2005 at the McGill University Health Centre. One hundred and twenty-two consecutive patients, scheduled for elective outpatient laparoscopic cholecystectomy, were approached by one of the authors (A.T.) in this prospective, randomized study. Exclusion criteria were: age 18 yr or 85 yr, ASA physical status III and more, history of hepatic, renal, or cardiac failure, organ transplant, diabetes, morbid obesity (body mass index 40), chronic use of opioids or -adrenergic receptor antagonists, known asthma or reactive airway disease, severe mental impairment, allergy to local anesthetics, or inability to comprehend pain assessment. Of the 122 patients approached, 21 refused to participate in the study, and 11 did not meet the criteria for inclusion,
1256
Postoperative Opioid-Sparing with Esmolol

receiving intraoperative esmolol infusion were studied. Here, the electroencephalographic bispectral index (BIS) value of 60 was targeted by adjusting the end-tidal concentration of desflurane between 4% and 8% and minimal alveolar anesthetic concentration (MAC-h) between 0.7 and 1.2. On the basis of these pilot data and a previous publication (9), maintenance of desflurane at this concentration in the esmolol group was considered appropriate to maintain adequate depth of anesthesia ( 60) and prevent intraoperative awareness. In all groups, the desflurane concentration was titrated to maintain systolic blood pressure within 20% of baseline, whereas fentanyl (50 g boluses), and esmolol and remifentanil increments were administered to keep the heart rate within 20% of baseline. The lungs were mechanically ventilated with a mixture of air in oxygen (Fio2 40%) to maintain normocapnia. Supplemental neuromuscular blockade was achieved with rocuronium 0.1 mg/kg after assessment of neuromuscular function with train-of-four. IV normal saline (0.9% NaCl) was administered during surgery at a rate of 6 mL kg 1 h 1, and intraoperative normothermia was maintained with forced air warming blankets positioned over the exposed parts of the body. Soon after induction of anesthesia, acetaminophen 1.3 g was administered per rectum, and dexamethasone 8 mg was given IV. Fifteen minutes before the end of surgery, ketorolac 30 mg and droperidol 0.625 mg were administered IV. If required, intraoperative hypotension (mean arterial blood pressure lower than 60 mm Hg) and bradycardia (heart rate 40 bpm) were to be treated in all groups with a fixed dose of intermittent neosynephrine 40 g or atropine 0.4 mg, respectively. Desflurane was discontinued after the last skin suture, and the remifentanil or esmolol infusions were also stopped at completion of surgery. Residual neuromuscular block was antagonized with neostigmine 0.05 mg/kg and glycopyrrolate 0.01 mg/kg. Patients were tracheally extubated on the operating table and then transferred to the postanesthesia care unit (PACU). All patients were operated by two surgeons highly experienced in laparoscopic cholecystectomy (L.F. and G.F.). After infiltration of lidocaine 2% in the infraumbilical skin, open insertion of a blunt-tipped 12-mm trocar was used to access the peritoneal cavity. Pneumoperitoneum was achieved with carbon dioxide, and intraabdominal pressure was maintained at 12 mm Hg throughout surgery. Three additional 5-mm ports were introduced after local anesthetic infiltration with lidocaine 2%. All patients received a single dose of 5000 U of subcutaneous heparin and wore antiembolic stockings. Patients were positioned in 30 degrees anti-Trendelenburg position and rotated toward the left side to facilitate exposure of the gallbladder. At the end of surgery, patients were returned to a supine position and the carbon dioxide left in the peritoneal
Vol. 105, No. 5, November 2007

cavity was expelled by abdominal compression. Bupivacaine 0.25% with epinephrine 1:200.000, 10 mL, was injected again in the surgical incisions.

Postoperative Care and Evaluations

Patients were transferred to the PACU where the arterial blood pressure, heart rate, respiration, and temperature were monitored by nurses unaware of the study hypothesis. The nursing staff did not interact with the anesthesiologists and used a standardized prescription for all patients. The anesthesia record was not made available to the recovery nurse or to the research fellow (G.M.) to avoid bias. Written orders were prepared on a separate order medication sheet. Fentanyl 25 g IV was prescribed for postoperative pain relief, to be administered every 5 min up to a maximum of 200 g/h by nursing staff only if the VRS for pain (0 10, where 0 no pain and 10 excruciating pain) was more than 3 at rest. The nurses evaluated patients every 5 min or at the patients request. Ondansetron 4 mg IV was prescribed for persistent nausea (lasting 5 min) or vomiting and could be repeated up to three times over a 3-h period if necessary. Recovery status was evaluated on arrival in the recovery room and every 30 min for the first 2 h by the research fellow unaware of the study hypothesis and having no interaction with the nurses or the anesthesiologists. The White-Song scoring system, which has been previously validated for bypassing the PACU and transfer of patients directly from the operating room to the step-down unit, was used (13). The variables evaluated included level of consciousness, physical activity, hemodynamic stability, respiratory stability, oxygen saturation status, postoperative pain assessment, and postoperative emetic symptoms. A minimal score of 12 of 14 points would be required for a patient to be fast-tracked after general anesthesia. In our institution, no step-down unit is available and patients are discharged home directly from the PACU. In this study, the time to achieve the White-Song score of 12 was used as a tool to assess the speed of recovery. Patients were discharged home independently by the nursing staff according to the following institutional standardized criteria used for all outpatient surgery: awake and oriented, stable cardiovascular hemodynamics, stable oxygen saturation 95% on room air, minimal pain (VRS 4 on ambulation), absence of nausea and vomiting, ability to tolerate oral fluids and to void, and to walk unaccompanied. Standardized rescue oral analgesics (acetaminophen 650 mg every 4 h, naproxen 500 mg every 12 h, codeine 30 60 mg every 4 h, oxycodone 5 mg every 4 h, hydromorphone 2 mg every 4 h) to be taken at home were prescribed by the surgical team at discharge. The following data were collected: demographic characteristics of the patients studied, history of smoking, motion sickness and PONV, duration of surgery and anesthesia, amount of fentanyl, esmolol, and remifentanil used during surgery, total amount of IV fluids during surgery, amount of fentanyl used in the
2007 International Anesthesia Research Society

1257

Table 1. Demographic Characteristics and Perioperative Data Control n 30

Male/female Age (yr) Weight (kg) Height (cm) Body mass index (kg/m2) ASA 1/2 History of allergy: n (%) History of smoking: n (%) History of previous PONV: n (%) History of motion sickness: n (%) Duration of surgery (min) Duration of anesthesia (min) Converted to open: n (%) Amount of intraoperative fentanyl ( g) Amount of intraoperative esmolol (mg/kg) Amount of intraoperative remifentanil ( g/kg) 11/19 53.7 17.9 74.1 15.1 168 (157176) 26.7 4.8 12/18 10 (33.3) 5 (16.7) 1 (3.3) 2 (6.7) 69 (5686) 87.5 (70111) 3 (10.0) 200.5 54.9 0 0

Esmolol n 30
13/17 47.2 16.7 71.9 14.3 162 (160165) 25.8 3.3 21/9 9 (30.0) 4 (13.3) 3 (10.0) 3 (10.0) 74 (5394) 97.5 (76117) 0 (0) 0 1.3 0.4 0

Remifentanil n 30
3/27 48.7 18.3 68.7 16.3 166.5 (160173) 26.0 6.1 14/16 9 (30.0) 2 (6.7) 1 (3.3) 1 (3.3) 60.5 (5070.5) 80 (7090.5) 2 (6.7) 0 0 7.6 2.8

P
0.012 0.3249 0.4055 0.1773 0.4312 0.183 0.960 0.512 0.441 0.604 0.0742 0.1012 0.238

Values are presented as relative number of patients, mean standard deviation, median (interquartile range), or absolute number (percentage). P values are calculated with Pearson 2 test for categorical variables, ANOVA one-way analysis of variance for parametric normally distributed variables and KruskalWallis ranked sum test for parametric not-normally distributed variables; signicant values are with P 0.017 (Bonferroni correction). Highlighted in bold are the signicant differences among groups. ASA American Society of Anesthesiologists; PONV postoperative nausea and vomiting.

recovery room, time spent in recovery room until discharged home, VRS, incidence of PONV, pruritus, urinary retention, and White-Song score. All patients were phoned at home 24 h after they were discharged from hospital to assess pain intensity (VRS) and presence of PONV at that moment, and the amount of analgesics and antiemetics used in the first postoperative 24 h.

RESULTS
Five patients (three patients in the control group and two in the remifentanil group) were excluded from the analysis because their surgery was converted from laparoscopy to laparotomy and they received either intercostals block, epidural, or wound infiltration with local anesthetics followed by patientcontrolled analgesia morphine on the surgical ward (Fig. 1).

Statistical Analysis
All data are presented in the tables as means sd, median (interquartile range), absolute values (percentage), or relative number of patients, when appropriate. Comparisons for each demographic and clinical variable among the three groups were performed by using ANOVA (with Scheffe test) for normally distrib uted variables, KruskalWallis rank-sum test for VRS score and for not normally distributed variables, and Pearson 2 test for categorical variables. Comparisons between each pairing of groups were performed by using Wilcoxons ranked sum test and Pearson 2 test. The level of significance was set at P 0.017 for all analysis (0.05/3, for the Bonferroni correction). Statistical analysis was performed with the Intercooled Stata 9.0 statistical package (Stata Corporation, College Station, TX). The primary outcome was the amount of fentanyl used for postoperative pain relief in the recovery room. Secondary outcome measures were the incidence of nausea and use of ondansetron in the recovery room, White-Song score, and time spent in the recovery room. Calculations for sample size were based on a study in which women underwent laparoscopic gynecological surgery (9). Thirty subjects in each group were sufficient to detect over 40% reduction in fentanyl requirement in the PACU, with a type-1 error of 0.05 and a power of 80%.
1258
Postoperative Opioid-Sparing with Esmolol

Patients Characteristics and Clinical Data on Preoperative Status and Intraoperative Anesthesia
No difference was observed among the three groups except for the significantly smaller number of male patients in the remifentanil group (Table 1). ASA, medical history, duration of surgery, and anesthesia were similar in all groups. There was no difference in the doses of rocuronium (68 16 mg) and MAC-h of desflurane among the three groups (0.93 0.4 for the control group, 0.89 0.3 for the remifentanil group, and 0.95 0.4 for the esmolol group). Intraoperative mean arterial blood pressure was within 20% of baseline, although there was a tendency for the remifentanil group to be lower than the other two groups. Heart rates were kept within 20% of the baseline in all groups. Although there were occasional mild bradycardia and hypotension, these episodes were of short duration and therefore did not requiring either neosynephrine or atropine.

Postoperative Clinical Data

Three patients (3.5%) were admitted by the surgical team for overnight stay for the following reasons: one patient aged 83 yr, another with persistent hypertension in the recovery room, and the third with a heart rate 100 bpm. VRS at rest and White-Song scores at 1 min and every 30 min for the first 2 postoperative
ANESTHESIA & ANALGESIA

Table 2. Postoperative Clinical Data Control n 27

VRS score 1 min 30 min 1h 1 h 30 min 2h White-Song score 1 min 30 min 1h 1 h 30 min 2h 3 (06) 3.5 (25) 3 (15) 2.5 (24) 2 (23) 12 (1013) 12 (1114) 13 (1214) 13 (1214) 13 (1214)

Esmolol n 30
5 (26) 4.5 (46) 3 (24) 2 (23) 2 (22) 12 (1112) 13 (1213) 13 (1214) 13 (1314) 14 (1314)
0.017 (Bonferroni correction).

Remifentanil n 28
2.5 (05.5) 4 (25.5) 3 (14) 2 (14) 2 (23) 11 (1013) 12 (1113) 13 (1214) 13 (12.514) 14 (1314)

P
0.6830 0.3590 0.9491 0.5980 0.8801 0.6488 0.5681 0.8919 0.8009 0.3474

Values are presented as median (interquartile range). P values are calculated with KruskalWallis ranked sum test; signicant values are with P VRS Verbal Rating Scale for pain.

Table 3. Analysis of Signicant Variables Among the Three Groups P Control n 27

Amount of fentanyl used ( g) Nausea in recovery room: n (%) Use of ondansetron: n (%) No. of patients requiring ondansetron(0/4/8 mg) No. of patients with White-Song score 12 at 1st/30th/60th/ 90th min or more Time from arrival to the PACU to discharge home (min) 168.1 96.8 (155)

Esmolol n 30
91.5 42.7 (100)

Remifentanil n 28
237.8 54.7 (238)

Cont Cont 3 versus versus groups Esm Remi

0.0001 0.004 0.001 0.0003 0.0409 0.0010 0.006 0.002 0.0002 0.3563 0.0036 0.925 0.702 0.8146 0.0963

Esm versus Remi

0.0001 0.004 0.001 0.0001 0.0060

18 (66.7) 18 (66.7) 9/5/13 16/6/0/5

9 (30.0) 7 (23.3) 23/6/1 21/4/3/2

19 (67.9) 20 (71.4) 8/9/11 9/8/6/5

180 (130210)

120 (100150)

162.5 (110220)

0.0033

0.0006

0.3900

0.0367

Values are presented as mean standard deviation (median), absolute number (percentage), relative number of patients, or median (interquartile range). P values are calculated with ANOVA one-way analysis of variance with Scheff test for the parametric normally distributed variables, Pearson 2 test for categorical variables, KruskalWallis ranked sum test for comparisons among groups for the parametric not-normally distributed variables and Wilcoxon test between groups for the parametric not-normally distributed variables. Regarding the comparison among the three groups, the Bonferroni correction was used, with the signicant value set at P 0.017. Highlighted in bold are the signicant differences among or between groups. PACU Postanesthesia care unit.

hours were not different among the three groups (Table 2). Most of the patients received fentanyl for pain relief in the PACU; however, the amount of fentanyl was significantly lower in the esmolol group compared with the other two groups, the largest being in the remifentanil group (Table 3). The dose of fentanyl administered in the PACU also showed a significant difference among the three groups when compared in pairs. The number of patients who received ondansetron, and the amount of ondansetron administered for persistent nausea in the PACU were significantly lower in the esmolol group compared with the other two groups. The incidence of nausea was significantly less frequent in the esmolol group compared with the other two groups, whereas there was no difference among the three groups with regard to incidence of
Vol. 105, No. 5, November 2007

vomiting (n 1 in the remifentanil group), itching (n 1 in the remifentanil group), and urinary retention (n 1 in the control group). There was no significant difference in nausea or use of ondansetron when comparing the control group versus the remifentanil group. There were significantly more patients with a White-Song score 12 at 1 min after arrival to the PACU in the esmolol group compared with remifentanil group (Fig. 2). The time interval between arrival in the PACU and discharge home was significantly shorter in the esmolol group. In contrast, there was no difference between the control and the remifentanil groups.

Postoperative Day 1 Clinical Data

The VRS at 24 h after discharge and the incidence of PONV the day after surgery were similar in the three
2007 International Anesthesia Research Society

1259

Figure 2. Cumulative proportion of hospital stay. This figure represents the cumulative proportion of hospital stay, until discharge home (KruskalWallis test: P 0.0033).

Table 4. Postoperative Day 1 Clinical Data Control n 27

VRS score at 24 h from hospital discharge PONV next day: n (%) Tylenol next day: n (%) Tylenol next day (mg) Naproxen next day: n (%) Naproxen next day (mg) Morphine equivalent next day: n (%) Morphine equivalent next day (mg) 3 (24) 6 (22.2) 16 (59.3) 987.5 (6501300) 8 (29.6) 875 (5001000) 17 (63.0) 60 (3080)

Esmolol n 30
2 (13) 4 (16.0) 16 (53.3) 812.5 (5751300) 9 (30.0) 1000 (5001000) 21 (70.0) 40 (2063)

Remifentanil n 28
3 (1.754) 5 (17.9) 15 (53.6) 1150 (6502275) 7 (25.0) 750 (5001000) 17 (60.7) 50 (3680)

P
0.2790 0.589 0.990 0.4881 0.878 0.7399 0.519 0.4106
0.017

Values are presented as median (interquartile range), or absolute number (percentage). P values are calculated with Pearson 2 test for categorical variables, and KruskalWallis ranked sum test for parametric not-normally distributed variables; signicant values are with P (Bonferroni correction). VRS Verbal Rating Scale for pain; PONV Postoperative nausea and vomiting.

groups (Table 4). The number of patients and the amount of acetaminophen, naproxen, and opioids consumed postdischarge were similar in the three groups.

DISCUSSION
The results of this study indicate that intraoperative continuous infusion of esmolol with no supplemental opioids contributes to a significant fentanyl-sparing effect during the immediate postoperative period. In addition, the incidence of nausea and the amount of ondansetron administered during the recovery period was significantly less, and the recovery period was accelerated, with earlier hospital discharge. Cholecystectomy is one of the most commonly performed abdominal operations, and the laparoscopic approach has become the standard of care in Canada (2). To provide the best outcomes with the least cost, many centers currently attempt to manage these patients as outpatients or day surgery cases. Although the minimally invasive approach allows for rapid discharge home, the recovery process can be further accelerated by
1260
Postoperative Opioid-Sparing with Esmolol

modifications in other aspects of perioperative care. The introduction of rapid and short-acting opioids, anesthetics, and neuromuscular blockers have facilitated the implementation of fast-track surgery (3), but side effects associated with opioids, such as PONV, can still occur. In fact, together with pain, medical comorbidity and PONV are the most common causes for unexpected hospital admission (2,4). Anesthetic and analgesic techniques that avoid or minimize opioid use may therefore be beneficial. In this study, three anesthetic techniques were compared, two based on opioids (intermittent injection of fentanyl and continuous infusion of remifentanil), and the third group receiving a non-opioid regimen consisting of esmolol. In addition, all three groups received a multimodal analgesia regimen consisting of a single dose of ketorolac and acetaminophen and infiltration of local anesthetics in the surgical incisions. Although the standard anesthetic technique for laparoscopic cholecystectomy with intermittent doses of fentanyl is widely used in clinical practice, there has been some interest in the use of continuous infusion of the short-acting opioid
ANESTHESIA & ANALGESIA

remifentanil to attenuate the hemodynamic responses elicited by intense nociceptive stimuli associated with the establishment of pneumoperitoneum (6,7). Remifentanil is a potent opioid with a fast onset and a short duration of action, allowing effective attenuation of the painful stimuli and the unwanted cardiovascular response (6). In addition, the anesthetic-sparing effect of remifentanil has been shown to facilitate a more rapid emergence after laparoscopic cholecystectomy (7). The third group in the present study received a continuous infusion of esmolol, with no supplemental opioids, with the intention to obtund the exaggerated sympathetic responses caused by pneumoperitoneum and maintain the heart rate within 15%20% of the baseline value. Although 90% of the patients all groups required fentanyl for postoperative pain relief, the lowest amount was administered to the esmolol group whereas the largest was to the remifentanil group. The amount of fentanyl required during the recovery period in the remifentanil group was two and a half times more than the esmolol group and also more than the control group. Development of acute opioid tolerance after large doses of remifentanil has been reported in volunteers and in surgical patients in the postoperative period (14,15). However, it is not known to what extent the induced tolerance is related to the intraoperative dose used (12). The number of female patients in the remifentanil group was, unexpectedly, significantly more than in the other two groups. It has been consistently shown that women seem to experience greater intensity of pain and receive more pain relieving medication than males (16,17). This might explain why this group received more postoperative fentanyl than the esmolol group. However, the control group, which included the same number of male and females as the esmolol group, received a significantly larger amount of postoperative fentanyl than the latter group, implying that the use of pain relieving medication was not gender related. This observation is supported by a factored ANOVA test where R2 values for gender and groupgender interaction were less than the group value, implying that the group was the strongest predictor of fentanyl use (0.078 only gender, 0.451 only group, and 0.504 their interaction). The opioid-sparing effect of esmolol has been previously reported (9 11), but this is the first study in which esmolol replaced the opioid intraoperatively and was associated with a significant decrease in the amount of postoperative fentanyl used. Coloma et al. (9) infused esmolol during gynecologic laparoscopic surgery and reported a lower incidence of nausea, but greater postoperative use of hydrocodone. In contrast, White et al. (10) using the same gynecologic surgical model, administered 1.5 g/kg of fentanyl at induction, followed by an infusion of esmolol, and found that this group required less postoperative opioids. Chia et al. (11) administered 3 g/kg of fentanyl followed by an infusion of esmolol in women undergoing total abdominal hysterectomy, and found that
Vol. 105, No. 5, November 2007

morphine consumption for the first 3 postoperative days was significantly reduced. In the present study, esmolol was administered as a sole drug replacing an opioid. Antiinflammatory drugs and infiltration of local anesthetics in the surgical incision were used in all three groups, as there is enough evidence that optimization of pain relief is achieved through multimodal intervention. The benefits of -adrenergic antagonists in patients with cardiovascular diseases are well established (8). Stressful interventions that cause intense activation of the sympathetic nervous system can result in cardiovascular consequences. Thus, -adrenergic antagonists can be targeted to attenuate this unwanted response (18). Functional magnetic resonance imaging studies in humans have also demonstrated hippocampal activation during emotional distress (19), fear (20), and anxiety (21). This hippocampal activation results, in part, from the activity of stress-related factors, including norepinephrine, which enhances the excitability of hippocampal principal neurons via an interaction with hippocampal n-methyl-d-aspartate subtype glutamate receptors (22). The hippocampus plays a role in nociception, a role predicated, at least in part, on n-methyl-d-aspartate receptors. It then seems possible that the activation of hippocampal -adrenergic receptors might play a role in nociceptive processes. If so, then blockade of these receptors should blunt the contribution of such -adrenergic activation to the nociceptive process, thereby resulting in the attenuation of perceived pain intensity. Autoradiographic studies in the rat hippocampus have reported both 1 and 2 adrenergic receptor expression in the rat hippocampus (23). Esmolol is a moderate lipophilic drug with receptor activity and could be involved in the modulation of central adrenergic activity (24), although some data seem to dispute whether it crosses the blood brain barrier. Another possible mechanism for the decreased postoperative requirement of fentanyl has been proposed on the assumption that blockers have been shown to decrease their own metabolism and that of other drugs, probably by reducing liver blood flow (25). One study showed that propanolol, by inducing a decrease in cardiac output and altering the distribution of organ blood flow, might have altered fentanyls pharmacokinetics (26), thereby reducing the requirement for fentanyl. There are some limitations with the design of the study. First, the three anesthesiologists who administered the anesthetics might have been biased in the administration of the study drugs. Although this is a well accepted criticism, the anesthesiologists did not assess patients either before or after surgery, and therefore could not influence postoperative assessment and administration of fentanyl. Second, it is argued that the use of a BIS monitor would have ensured that the three groups had similar depth of anesthesia. Intraoperative opioids and blockers have not been shown to substantially influence BIS (27,28),
2007 International Anesthesia Research Society

1261

and one would assume that by receiving a similar amount of desflurane (average MAC was 9.2 4.2) the three groups would not be dissimilar. The esmolol group also had the lowest incidence of nausea and received the least amount of ondansetron in the recovery room, and this might have been directly related to the small dose of fentanyl used during the postoperative period or the decreased total dose of fentanyl received during the perioperative period. Another direct benefit of the opioid-sparing effect of esmolol was the shorter time to reach a White-Song score 12 when compared with the other two groups. The White-Song scoring system determines when patients can be safely discharged from the PACU. A minimal score of 12 of 14 indicates readiness as it takes into consideration pain, emetic symptoms, assessment of consciousness, physical activity, and hemodynamic and respiratory stability. In conclusion, intraoperative esmolol used within a multimodal intervention strategy was effective in sparing postoperative opioids, therefore minimizing side effects and allowing patients to be discharged home earlier. REFERENCES
1. Richardson WS, Fuhrman GS, Burch E, Bolton JS, Bowen JC. Outpatient laparoscopic cholecystectomy. Outcomes of 847 planned procedures. Surg Endosc 2001;15:1935 2. Lau H, Broks DC. Contemporary outcomes of ambulatory laparoscopic cholecystectomy in a major teaching hospital. World J Surg 2002;26:111721 3. Avramov MN, White PF. Use of alfentanil and propofol for outpatient monitored anesthesia care: determining the optimal dosing regimen. Anesth Analg 1997;85:566 72 4. Zhao SZ, Chung F, Hanna DB, Raymunso AL, Cheung RY, Chen C. Dose-response relationship between opioid use and adverse effects after ambulatory surgery. J Pain Symptom Manage 2004;28:35 46 5. Bisgaart T. Analgesic treatment after laparoscopic cholecystectomy. Anesthesiology 2006;104:835 47 6. Song D, White PF. Remifentanil as an adjuvant during Desflurane anesthesia facilitates early recovery after ambulatory surgery. J Clin Anesth 1999;11:364 7 7. Song D, Whitten CW, White PF. Remifentanil infusion facilitates early recovery for obese outpatients undergoing laparoscopic cholecystectomy. Anesth Analg 2000;90:11113 8. Davidson EM, Szmuk P, Doursout MF, Chelly J. Antinociceptive and cardiovascular properties of esmolol following formalin injection in rats. Can J Anaesth 2001;48:59 64 9. Coloma M, Chiu JW, White PF, Armbruster SC. The use of esmolol as an alternative to remifentanil during desflurane anesthesia for fast track outpatient gynecologic laparoscopic surgery. Anesth Analg 2001;92:3527

10. White PF, Wang B, Tang J, Wender RH, Naruse R, Sloninsky A. The effect of intraoperative use of esmolol and nicardipine on recovery after ambulatory surgery. Anesth Analg 2003; 97:1633 8 11. Chia YY, Chan MH, Ko NH, Liu K. Role of beta-blockade in anaesthesia and postoperative pain management after hysterectomy. Br J Anaesth 2004;93:799 805 12. Cortinez LI, Brandes V, Munoz HR, Guerrero ME, Mur M. No clinical evidence of acute opioid tolerance after remifentanilbased anaesthesia. Br J Anaesth 2001;87:866 9 13. White P, Song D. New criteria for fast-tracking after outpatient anesthesia: a comparison with the Modified Aldretes scoring system. Anesth Analg 1999;88:1069 72 14. Guignard B, Bossard A, Coste C, Sessler D, Lebrault C, Alfonsi P, Fletcher D, Chauvin M. Acute opioid tolerance: intraoperative remifentanil increases postoperative pain and morphine requirement. Anesthesiology 2000;93:409 17 15. Vinik H, Kissin I. Rapid development of tolerance to analgesia during remifentanil infusion in humans. Anesth Analg 1998;86: 130711 16. Logan DE, Rose JB. Gender differences in post-operative pain and patient controlled analgesia use among adolescent surgical patients. Pain 2004;109:4817 17. Taenzer AH, Clark C, Curry CS. Gender affects report of pain and function after arthroscopic anterior cruciate ligament reconstruction. Anesthesiology 2000;93:670 5 18. Koivusalo AM, Scheinin M, Tikkanen I, Yli-Suomu T, Ristkari S, Laakso J, Lindgren L. Effects of esmolol on haemodynamic response to CO2 pneumoperitoneum for laparoscopic surgery. Acta Anaesthesiol Scand 1998;42:510 7 19. Sinha R, Lacadie C, Skudlarski P, Wexler BE. Neural circuits underlying emotional distress in humans. Ann N Y Acad Sci 2004;1032:254 7 20. Surguladze SA, Brammer MJ, Young AW, Andrew C, Travis MJ, William SC, Phillips ML. A preferential increase in the extrastriate response to signals of danger. Neuroimage 2003; 19:131728 21. Ploghaus A, Narain C, Beckmann CF, Clare S, Bantick S, Wise R, Matthews PM, Rawlins JN, Tracey I. Exacerbation of pain by anxiety is associated with activity in a hippocampal network. J Neurosci 2001;15:21:9896 903 22. Sarvey JM, Burgard EC, Decker G. Long-term potentiation: studies in the hippocampal slide. J Neurosci Methods 1989; 28:109 24 23. Booze RM, Crisostomo EA, Davis JN. Beta-adrenergic receptors in the hippocampal and retrohippocampal regions of rats and guinea pigs: autoradiographic and immunohistochemical studies. Synapse 1993;13:206 14 24. Gorczynski RI, Shaffer JE, Lee RJ. Pharmacology of ASL-8052, a novel beta-adrenergic receptor antagonist with ultrashort duration of action. J Cardiovasc Pharmacol 1983;5:668 77 25. Wood AJ, Feely J. Pharmacokinetic drug interactions with propanolol. Clin Pharmacokinet 1983;8:253 62 26. Avram MJ, Krejcie TC, Henthorn TK, Niemann CU. etaadrenergic blockade affects initial drug distribution due to decreased cardiac output and altered blood flow distribution. JPET 2004;311:61724 27. Kalkman CJ, Drummond J. Monitors of depth of anesthesia, quo vadis? Anesthesiology 2002;96:784 7 28. Yang H. Are -blockers anesthetics? Can J Anaesth 2003; 50:62730

1262

Postoperative Opioid-Sparing with Esmolol

ANESTHESIA & ANALGESIA