Abstract

Background

Recently, 5-hydroxytryptamine 3 (5-HT3) receptor antagonists have been reported to prevent postanesthetic shivering. This placebo-controlled study was performed to evaluate the efficacy of granisetron, a 5-HT3 antagonist, in comparison with meperidine and tramadol in preventing postanesthetic shivering.

Methods

In this prospective, randomized, double-blind study, 132 ASA I and II patients undergoing elective orthopedic surgery under standardized general anesthesia were included. At the end of surgery, patients were randomly assigned to one of four groups (each group n= 33) using a double-blinded protocol. Group T received 1 mg/kg tramadol, group G rceived 40 μg/kg granisetron (an antiemetic dose), group M received 0.4 mg/kg meperidine, and group P received saline 0.9% as placebo. Shivering was graded according to the following: 0 = no shivering; 1 = piloerection, peripheral vasoconstriction or peripheral cyanosis without other cause; 2 = visible muscular activity confined to one muscle group; 3 = visible muscular activity in more than one muscle group; and 4 = gross muscular activity involving the entire body. The emergence time from anesthesia, defined as the time between withdrawal of isoflurane and tracheal extubation, was documented.

Results

The number of patients with observable shivering was 19 in group P, nine in group G, seven in group T and six in group M. Granisetron significantly reduced the incidence of shivering in comparison with placebo (p= 0.013). Although the frequency of shivering was higher with granisetron in comparison to tramadol and meperidine, it was not statistically significant (p> 0.05). The number of patients with a shivering score of 2, 3 and 4 was significantly higher in group P compared with the other groups (p = 0.001). Both meperidine and tramadol caused a significantly prolonged emergence time (20.58 ± 3.56 and 16.45 ± 4.13 minutes, respectively) as opposed to granisetron (13.58 ± 3.41 minutes) and placebo (12.61 ± 3.31 minutes).

Conclusion

The prophylactic use of granisetron 40 mg/kg is as effective as meperidine (0.4 mg/kg) and tramadol (0.1 mg/kg) in preventing postanesthetic shivering without prolonging the emergence time from anesthesia.

Keywords

5-HT antagonists; general anesthesia; granisetron; meperidine; mu receptors; shiveringtramadol;

---

1. Introduction

Postanesthetic shivering occurs in up to 60% of patients recovering from general anesthesia.1 All general anesthetics markedly impair normal autonomic thermoregulatory control.¹ Postanesthetic shivering is very unpleasant and physiologically stressful to the patient after receiving uneventful modern anesthesia.¹ In addition, postanesthetic shivering is a potentially serious complication because it causes an increase in oxygen consumption of 100−600% disproportionate to intraoperative heat loss, and also increases intraocular and intracranial pressures.^1,2 A variety of physical and pharmacological methods have been tested to reduce perioperative hypothermia and suppress postanesthetic shivering; however, none have gained universal acceptance.³

Various opioid and non-opioid pharmacological treatments for postanesthetic shivering have been used; however, there are some specific side-effects of these agents including hypotension, hypertension, sedation, respiratory depression, nausea and vomit ing.^3,4 Serotonin (5-hydroxytryptamine [5-HT]), a biological amine found in the brain and the spinal cord, has a role in neurotransmission and studies suggest that the serotoninergic system has a role in the control of postanesthetic shivering.^3,5 Re cently, the 5-HT3 antagonists, ondansetron and dolasetron, have been tested for prevention of postanesthetic shiv ering.^5,6 However, results of these studies are controversial.

The aim of this placebo-controlled study was to compare the efficacy of granisetron, a relatively new 5-HT3 antagonist, with those of tramadol and meperidine, two drugs commonly used to treat shivering after general anesthesia.

2. Methods

This prospective, randomized, double-blind, placebocontrolled study was approved by the local ethics committee. Informed consent was obtained from all studied patients. One hundred and thirty-two ASA physical status I or II patients (age 18−65 years, weight 50−70 kg) undergoing elective orthopedic surgery were enrolled in the study. Patients were randomly assigned to one of the following four groups (n= 33 in each group): group T received 1 mg/ kg tramadol;¹ group G received 40 μg/kg granisetron (an antiemetic dose);⁷ group M received 0.4 mg/kg meperidine (this dose is often used to treat shivering);⁸ and group P received 0.9% saline as placebo. Patients who needed vasoconstrictors during surgery, patients with fever (temperature > 37.5ºC), drug or alcohol abuse, muscular disease or Parkinson’s disease, were excluded.

Induction of anesthesia was done using thiopental (5 mg/kg) and fentanyl (2 μg/kg). Atracurium (0.5 mg/kg) was given to facilitate orotracheal intubation, and general anesthesia was maintained with 70% nitrous oxide and isoflurane (1.2 ± 0.5% end tidal oxygen). Ventilation was mechanically controlled to maintain an end tidal carbon dioxide tension of 30−35 mmHg. Patients were covered with cotton surgical drapes during anesthesia but were not purposefully warmed. Ambient temperature was kept at 22−24ºC with constant humidity. Using a computerized randomization table, patients were randomly assigned to one of the four groups before induction of anesthesia. All test drugs were given intravenously (i.v.) at the end of surgery according to the double-blinded protocol.

The prophylactic effect on shivering was assessed by one investigator in charge of the postanesthesia care unit, who was not aware of the administered drug. Postanesthetic shivering was graded using a scale similar to that validated by Wrench et al: 0 = no shivering; 1 = piloerection, peripheral vasoconstriction, peripheral cyanosis or a combination with no other cause, but without visible muscular activity; 2 = visible muscular activity confined to one muscle group; 3 = visible muscular activity in more than one muscle group; and 4 = gross muscular activity involving the entire body.⁸ Before surgery (T0), before administration of the test drug (T1), and 10 minutes (T2), 20 minutes (T3), and 30 minutes (T4) after extubation, systolic arterial blood pressure, diastolic arterial blood pressure, mean arterial blood pressure (MAP), heart rate (HR), and tympanic core temperature were measured. All data were measured by Datex Caridocap II (Datex, Division of Instrumentarium Corp., Helsinki, Finland). Emergence time, the time from termination of application of isoflurane to extubation of the trachea, was documented. Furthermore, postanesthesia awareness was scored according to the following scale: 0 = alert; 1 = arousal to voice; 2 = arousal to gentle tactile stimulation; 3 = arousal to vigorous tactile stimulation; and 4 = no awareness on arrival at the recovery room.¹ Postoperative pain relief was offered to patients as needed by giving i.v. bolus doses of methadone (increments of 5 mg). Methadone was chosen instead of meperidine because we did not want to prejudice the effect of the tested meperidine on shivering although it is the first line drug for pain control in our postanesthesia care unit. Postanesthesia recovery time was considered as the time between extubation of the trachea and discharge from the recovery unit with a score > 9, according to the postanesthesia discharge scoring system.⁹

Statistical analysis was carried out using SPSS version 14 (SPSS Inc., Chicago, IL, USA). Differences in age, body weight, time to extubation, recovery time and the effects of treatment in all studied parameters (tympanic temperature, MAP, HR) were tested using one-way ANOVA, and multiple comparisons between pairs was done by the Bonferroni test. The frequency distributions of sex, ASA physical status, methadone administration, shivering and the severity of shivering, degree of sedation and drug side effects were tested statistically using the χ² test. We expected an incidence of shivering in the placebo group of at least 53% and a reduction in the incidence to approximately 17% with an effective medication such as meperidine. Consequently, the present study was powered to detect such a reduction from 53% to 17% incidence with an alpha error of 0.05 (two-sided) and a beta error of 0.2. Based on that assumption, 32 patients were included in each group. All values were expressed with absolute number (n), percentage (%) and mean. A p value less than 0.05 was considered to be significant.

3. Results

The groups did not differ significantly regarding demographic characteristics, ASA physical status, duration of anesthesia, recovery time and perioperative use of analgesics (Table 1). The number of patients with observable shivering was 19 (57.6%) in group P, nine (27.3%) in group G, seven (21.2%) in group T and six (18.2%) in group M. The incidence of shivering was statistically higher in the placebo group compared with other groups (p= 0.002). Gran isetron significantly reduced the incidence of shivering in comparison with placebo (p= 0.013). Although the number with shivering was higher after granisetron in comparison with tramadol and meperidine, it was not statistically significant (p= 0.566 and p= 0.378 respectively, using the χ² test). The number of patients with a shivering grade of 2, 3 or 4 was statistically higher in group P compared with other groups (p= 0.001) (Table 1). There were no significant differences in grades of shivering between granisetron, tramadol and granisetron, meperidine (p= 0.569 and p= 0.382 respectively, using the Mann-Whitney U test). There were no significant differences in tympanic temperature at all time points among the groups (Table 2). Tramadol and meperidine significantly reduced MAP at T3 and T4 in comparison with placebo and granisetron (Table 2). However, there was no patient in any group whose hemodynamics required therapeutic intervention.

Extubation time was significantly longer in patients treated with meperidine (20.58 ± 3.56 minutes) and tramadol (16.45 ± 4.13 minutes) compared with patients who received placebo (12.61 ± 3.31 minutes), whereas in the granisetron-treated patients (13.58 ± 3.41 minutes), it was similar to the placebo group (Table 1). Sedation scores on arrival at the recovery room were not significantly different among groups (Table 3). The frequency distribution of complications was higher in the meperidine and tramadol groups in comparison with the granisetron and placebo groups (Table 4).

4. Discussion

During the postoperative period, shivering is a common and undesirable complication of general anesthesia. It is defined as readily detectable fasciculation of the face, jaw, head, trunk or extremities lasting longer than 15 seconds and occurs in up to 65% of patients recovering from general anesthesia.^2,3

This study was designed to evaluate the efficacy of granisetron in preventing postanesthetic shivering in comparison with meperidine and tramadol, two well known drugs for prevention and treatment of postanesthetic shivering after a uniform anesthetic regimen. In our study, 19 of 33 patients (57.6%) in the placebo group shivered postoperatively. This incidence is in accord with those of previous studies.^1,4,10,11 Granisetron reduced the inci dence and severity of shivering significantly as did meperidine and tramadol, in comparison with the placebo group. Meperidine and tramadol caused some side-effects such as respiratory depression, hypotension, nausea and vomiting. How ever, there was no respiratory depression, nausea and vomiting in the granisetron group. Granisetron did not alter the time of extubation, and the sedation it caused was not clinically or statistically significant. There were no hemodynamic alterations during the stay in the recovery room.

Recently, 5-HT3 antagonists such as ondansetron and dolasetron, two well known drugs for postoperative nausea and vomiting, have been sug gested to prevent postanesthetic shivering.⁴ However, the results of the relevant studies are controversial.^5,6,12

Powell and Buggy investigated the effect of ondansetron on intraoperative core and peripheral temperatures and postanesthetic shivering.⁵ They reported that, after general anesthesia, the incidence of shivering was 57%, 33% and 15% in patients of the control, ondansetron 4 mg and ondansetron 8 mg groups, respectively. They suggested that 5-HT3 inhibition did have specific antishivering activity and could produce a dose-dependent reduction in shivering, which was statistically significant at larger doses. In another study, Bock et al reported that dolasetron 1 mg/kg decreased the incidence of shivering from 62% to 27%.¹²

Piper et al studied the effect of dolasetron on postanesthetic shivering.⁶ The medication was given after induction of anesthesia. They found that the incidence and severity of postanesthetic shivering was similar with dolasetron 12.5 mg i.v. and placebo. This discrepancy between Piper et al’s study and those of Powell and Buggy, Bock et al and ours might be related to the time that the drug was administered and also the dose of drug used.^4,5 Granisetron may be superior to ondansetron and dolasetron because of its antishivering activity in addition to its antiemetic effect.

The usefulness of opioids in the treatment and prevention of shivering is limited because of the risk of respiratory depression, sedation, nausea and vomiting, especially when repeated doses are necessary.^3,4 In our study, the time to extubation was significantly longer and side-effects such as respiratory depression, hypotension, nausea and vomiting were statistically significantly higher in the meperidine and tramadol groups compared with the placebo and granisetron groups. In our study, granisetron had no effect on the cardiovascular system. Powell and Buggy’s study also revealed the same result with ondansetron.⁵ Previous studies have confirmed the efficacy of granisetron in the prevention of postoperative nausea and vomiting.^7,13 Granisetron also proves useful in the prevention of shivering without causing respiratory and cardiovascular side effects and prolongation of recovery time.

As shivering is a response to hypothermia, body temperature should be normally maintained within limits of 36.5−37.5ºC.¹⁴ For these reasons, in our study, all patients were covered with cotton surgical drapes, the temperature of the operating room was actively maintained between 22ºC and 24ºC and infusions of cold crystalloid solutions were avoided. Also, in the recovery room, the temperature was maintained around 25ºC with constant humidity and again cold crystalloid solutions were not used for infusion.

Limitations of this study include lack of previous studies to indicate the dose of granisetron to control postanesthesia shivering and lack of a dose response study for tramadol to determine its lowest effective dose.

In conclusion, we found that granisetron could effectively prevent postanesthetic shivering as did meperidine and tramadol when administered at the time of wound closure. However, granisetron seemed to produce fewer side effects compared with the other two drugs.

Acknowledgments

Financial support for this study was provided by the Research Department of Isfahan University of Medical Sciences, Isfahan, Iran.