Secondary Logo

Journal Logo

Ondansetron Given Before Induction of Anesthesia Reduces Shivering After General Anesthesia

Powell, Robert M., FRCA; Buggy, Donal J., MD, MSc, MRCPI, DME, FFARCSI

doi: 10.1097/00000539-200006000-00032

The neurotransmitter pathways involved in the mechanism of postanesthetic shivering (PAS) are poorly understood. Meperidine, clonidine, and physostigmine are all effective treatments, indicating that opioid, α2-adrenergic, and anticholinergic systems are probably involved. We investigated the effect of ondansetron, a 5-HT3 antagonist used to treat postoperative nausea and vomiting, on intraoperative core and peripheral temperatures and PAS. Eighty-two patients (age, 18–60 yr) undergoing orthopedic, general, or urological surgery were randomized into three groups in this double-blinded, placebo-controlled, study: Group O4 (n = 27) received ondansetron 4 mg IV, Group O8 (n = 27) received ondansetron 8 mg IV, and Group C (n = 28) received saline IV immediately before the anesthetic induction. Core (tympanic) and fingertip temperature (dorsum of middle finger) were recorded. Anesthesia was induced with IV fentanyl 1 μg/kg and propofol 2.0–2.5 mg/kg and maintained with 1 minimum alveolar anesthetic concentration isoflurane in 70% nitrous oxide/oxygen. The occurrence of shivering was documented clinically during recovery by nursing staff, who were unaware of the group assignment. PAS occurred in 16 of 28 (57%) patients in Group C, compared with 9 of 27 (33%) in Group O4 (P = 0.13) and 4 of 27 (15%) patients in Group O8 (P = 0.003). Within each group, core temperature decreased and peripheral temperature increased significantly, but there were no significant differences among the groups at any time interval. We conclude that ondansetron 8 mg IV given during the induction of anesthesia prevents PAS without affecting the core-to-peripheral redistribution of heat during general anesthesia. This suggests that serotonergic pathways have a role in the regulation of PAS.

Implications In a randomized, double-blinded, placebo-controlled, clinical study, ondansetron 8 mg IV, given just before the induction, reduced the incidence of postanesthetic shivering compared with saline. The anticipated core-to-peripheral redistribution of body temperature during general anesthesia was not affected. This implies that ondansetron probably acts by a central inhibitory mechanism, and that 5-hydroxytryptaminergic pathways have a role in regulating postanesthetic shivering.

University Department of Anaesthesia, Leicester University, Leicester, United Kingdom

February 23, 2000.

Address correspondence and reprint requests to Donal J. Buggy, MD, MSc, MRCPI, DME, FFARCI, University Department of Anaesthesia, Leicester General Hospital, Leicester LE5 4PW, UK. Address e-mail to

Presented, in part, at the Anaesthetic Research Society, March 1999, Aberdeen, UK.

Postanesthetic shivering (PAS) occurs frequently postoperatively and may be very distressing for patients (1,2). Although there is general agreement that it is a thermoregulatory phenomenon, i.e., a physiological response to anesthesia-induced core hypothermia (3,4), there is some evidence that it may also have a nonthermoregulatory component (5,6). The neurotransmitter pathways conveying signals from central nervous system control centers, such as the hypothalamus, to skeletal muscle are not clearly understood, but probably involve multiple levels of information integration and numerous neurotransmitters. IV administration of the opioids, meperidine, alfentanil, and nalbuphine (7–9), the nonopioid analgesic tramadol (10), the α2-adrenergic agonist clonidine (11), the respiratory stimulant doxapram (12), and the cholinomimetic agent physostigmine (13) have all reduced the incidence of shivering or suppressed established shivering.

Serotonin, (5-hydroxytryptamine [5-HT]), a biological amine found in the brain and spinal cord, has a role in neurotransmission (14). Indeed, tramadol, which inhibits 5-HT reuptake (10), and ketanserin, a 5-HT2 antagonist, inhibit established PAS (15). In animal models, direct intraventricular injections of 5-HT influence body temperature and shivering (16). All these observations suggest that the serotonergic system has a role in the control of postanesthetic shivering. Ondansetron is a specific 5-HT3 antagonist (17) that may affect perioperative thermoregulation and PAS. Therefore, we evaluated the effect of ondansetron, given just before the induction of anesthesia, on the typical core-to-peripheral temperature redistribution evoked by general anesthesia and on the incidence of PAS, in a randomized, placebo-controlled, double-blinded study.

Back to Top | Article Outline


After institutional ethics committee approval and informed consent, 87 patients (ASA physical status I or II), scheduled for elective minor general surgery (i.e., herniorrhaphy, hemorrhoidectomy, varicose veins), urological, or knee arthroscopic surgery, were enrolled into the study. Exclusion criteria were a pyrexial illness, allergy to ondansetron, a surgery anticipated to be longer than 90 min or to require mechanical ventilation, age less than 18 or more than 60 yr, or if use of vasoconstrictors or vasodilators was planned. No patient received premedication.

Patients were randomized, by using a sealed envelope technique, to one of three groups before the induction of anesthesia: Group O4 received ondansetron 4 mg IV, Group O8 received ondansetron 8 mg IV, and Group C received IV saline as a negative control. These trial preparations were prepared freshly by operating department technologists who were not involved in the study and who recorded the group randomization separately, such that the anesthesiologist recording the data and caring for the patient was unaware of what the preparation contained. All patients received a 4-mL injection, the volume of ondansetron being made up with normal saline to 4 mL when necessary. This injection of trial medication was given immediately after placement of the IV cannula and 3–5 min before the induction of anesthesia. Baseline core and fingertip temperatures were recorded by using a tympanic thermometer. Core temperature was recorded at the tympanic membrane and after switching the thermometer to skin temperature mode, fingertip temperature was recorded at the dorsum of the middle finger, on the contralateral hand to the IV infusion. These recordings were repeated at 10-min intervals during surgery and in recovery. Baseline mean arterial pressure (MAP) was also recorded before the induction as the mean of two recordings at 3-min intervals by using a noninvasive oscillotonometer (Dinamap Cardiocap II, Helsinki, Finland) and repeated at 10-min intervals thereafter.

General anesthesia was induced with fentanyl 1 μg/kg followed by propofol 2.0–2.5 mg/kg. After placement of a laryngeal mask airway, anesthesia was maintained with 1 minimum alveolar anesthetic concentration isoflurane in 70% nitrous oxide/oxygen. Patients had standard cotton surgical drapes and were not actively heated. Ambient temperature was maintained at 20°–22°C with constant humidity.

PAS was documented visually by three specified members of the recovery room staff, who were unaware of the study group allocation. They were also carefully briefed on the definition of shivering used in the study. Shivering was defined as readily detectable fasciculations or tremors of the face, trunk, or limbs of a minimum of 15-s duration.

As stated, previous studies have found an incidence of PAS of the order of 40%–65%. We anticipated an incidence of 45% in the control group and took a difference of 40% in incidence of shivering between control and treated groups as being clinically meaningful. Hence, we prospectively calculated that 29 patients were required in each group for a Type I error of 0.05 and a Type II error of 0.2. Continuous variables, including demographic data and temperature values over time within groups were analyzed by using repeated measures analysis of variance. One-way analysis of variance was used to analyze differences among the groups. Incidence of shivering was analyzed by using χ2 test with Yates’ correction. Data were expressed as mean ± sd, with P < 0.05 being considered significant.

Back to Top | Article Outline


One patient from the O4 and one from the O8 group were withdrawn because of technical difficulties with the thermometer. Another was withdrawn from the control group and O8 group as a result of a requirement for intraoperative opioids. One patient was also withdrawn from the O4 group because of an unexpected surgical requirement to turn the patient prone, necessitating endotracheal intubation. Hence, 27 patients from O4 and O8 and 28 patients from C completed the study.

Demographic data, sex, and duration of anesthesia were similar among the groups (Table 1). PAS was significantly reduced in patients receiving O8 compared with saline control (15% vs 57%, respectively, P = 0.003), with patients receiving O4 having an intermediate incidence (33%), not significantly different from either of the other groups, Table 2. MAP and heart rate (HR) are shown in Table 3. This predictably indicates a significant reduction in MAP and HR compared with baseline after the induction of anesthesia, with MAP values remaining lower than baseline for the duration of the administration of anesthesia, but HR values tending to recover to preinduction values once surgery had commenced. However, there were no significant differences among the groups at any time. MAP and HR values in recovery room also did not indicate any significant differences among the groups (Table 3).

Table 1

Table 1

Table 2

Table 2

Table 3

Table 3

Figures 1 and 2 show core and fingertip temperature changes with the duration of the administration of anesthesia, respectively. Although core temperature decreased and fingertip temperature increased significantly in all groups with respect to baseline, there were no significant temperature differences among the groups at any time interval. visual analog scale pain scores were low in all groups at 15 min after arousal, with the control, O4, and O8 groups having mean (sd) visual analog scale scores of 2.1 (1.0), 2.3 (1.1), and 1.9 (1.2), respectively.

Figure 1

Figure 1

Figure 2

Figure 2

Back to Top | Article Outline


Prevention and treatment of PAS is an important aspect of patient care, as it may be associated with a number of deleterious sequelae, including sympathoadrenal stimulation, increased oxygen consumption, and carbon dioxide production. In the present study, the incidence of PAS was 57% in the saline group compared with 15% in the group receiving the larger dose of ondansetron (8 mg), with an intermediate incidence of shivering (33%) observed in patients receiving the smaller dose (4 mg). We decided not to grade shivering into “mild” and “severe” categories as previous authors have done and then include the aggregate number of patients who shivered in key statistical comparisons (5,8,9,12,13,15,18). Moreover, we felt that distinguishing between “mild,” “moderate,” and “severe” shivering on the basis of clinical observation alone is highly subjective and of limited relevance.

Specific inhibition of the 5-HT3 system, therefore, produced a dose-dependent reduction in shivering, being statistically significant at the larger dose. Perhaps 5-HT3 inhibition has a specific antishivering effect, but given the variety of neurotransmitter systems known to be also involved in regulating shivering, an inhibitory effect at the 5-HT3 receptor probably results from a generalized thermoregulatory inhibition at the level of the hypothalamus, where the bulk of thermoregulatory control occurs (1,19). This explanation is supported by our data on perioperative temperature. The anticipated core-to-peripheral redistribution of body temperature after the administration of general anesthesia is characterized by an approximate 1°C decrease in core temperature within the first 20–30 minutes after the induction, followed by an increase in fingertip temperature, measured at the skin (20,21). This was unchanged in our patients who received ondansetron. Thus, the PAS effect of ondansetron is independent of intraoperative core hypothermia, suggesting that it inhibits thermoregulatory responses by a central mechanism. The extent of core hypothermia seems to be related to the degree of vasodilation induced during anesthesia administration, because a recent study demonstrated greater core hypothermia after propofol induction compared with sevoflurane induction, even when anesthesia was maintained with sevoflurane (21). Ondansetron is notable for its lack of hemodynamic side effects (22); hence, its lack of effect on redistribution hypothermia is unsurprising. Ondansetron did not alter the hemodynamic profile of either group intraoperatively, and although our study did not specifically examine postoperative hemodynamic profiles, it is highly unlikely that this is central to its antishivering effect.

In contrast to some other drugs used to treat PAS, we found that ondansetron has innocuous effects on the cardiovascular system. Clonidine may be associated with significant hypotension and sedation (15), but this has not been consistently shown, with other studies indicating no significant hypotensive or sedative effect (11,13). Tramadol, a nonopioid analgesic which inhibits 5-HT re-uptake and enhances synaptosomal concentration of 5-HT and noradrenaline, also inhibits shivering after the administration of general anesthesia (10). It also reduces the sweating, vasoconstriction, and shivering thresholds (23). Doxapram, a cerebral stimulant, is also effective in suppressing PAS, but is associated with significant hemodynamic effects (24). Physostigmine increases heart rate and blood pressure (13,18), which may be detrimental to myocardial oxygen demand in some patients with coronary artery insufficiency. It is also commonly associated with postoperative nausea and vomiting (13). In contrast, ondansetron effectively relieves postoperative nausea and vomiting.

Although meperidine in the doses effective in treating postanesthetic shivering (0.33–0.4 mg/kg) rarely produces untoward cardiovascular effects (7–9,12,13), it may potentially cause respiratory depression, especially if it or other opioids have been given intraoperatively. However, ondansetron is more expensive than any of the other drugs effective against PAS.

A limitation of our protocol may be that we did not include a positive control group using an established agent such as meperidine. Precisely because meperidine’s dose and efficacy are established, we believed it would be more useful to study two clinically relevant doses of ondansetron, which ultimately confirmed an important dose-response effect. Although our objective in this study was primarily to investigate the mechanism of PAS by evaluating the importance of 5-HT pathways, ondansetron could very plausibly be an attractive and alternative preventive treatment for PAS, especially as it has powerful antiemetic effects (17,22) and a favorable cardiorespiratory side effect profile. The efficacy of ondansetron in treating established PAS and shivering after epidural anesthesia administration also warrants investigation.

Although infrared aural canal thermometry was a convenient and minimally invasive method of estimating core body temperature, using a thermistor in an established core-temperature monitoring site (e.g., the nasopharynx) may have yielded more accurate data. Nonetheless, the core temperature values we observed were similar to those reported in all previous studies of perioperative core temperature.

In conclusion, we have demonstrated that ondansetron 8 mg given before the induction of anesthesia, reduces the incidence of PAS in adults, without effecting the core-to-peripheral redistribution of temperature normally observed during the administration of general anesthesia. This implies that 5-HT3 pathways are also involved in the multifactorial control of PAS.

We acknowledge the enthusiastic cooperation of our recovery room nursing staff.

Back to Top | Article Outline


1. Sessler DI. Temperature monitoring. In: Miller RD, ed. Anesthesia. New York: Churchill Livingstone, 1994: 1363–82.
2. Buggy D, Hughes N. Pre-emptive use of the space blanket reduces shivering after general anaesthesia. Br J Anaesth 1994; 72:393–6.
3. Sessler DI. Perianesthetic thermoregulation and heat balance in humans. FASEB J 1993; 7:638–44.
4. Sessler DI, Rubinstein EH, Moayeri A. Physiologic responses to mild perianesthetic hypothermia in humans. Anesthesiology 1991; 75:594–610.
5. Horn EP, Sessler DI, Standl T, et al. Non-thermoregulatory shivering in patients recovering from isoflurane or desflurane anesthesia. Anesthesiology 1998; 89:878–86.
6. Soliman MG, Gillies DM. Muscular hyperactivity after general anaesthesia. Can Anaesth Soc J 1972; 19:529–35.
7. Kurz M, Belani KG, Sessler DI, Lanson MD. Naloxone, meperidine, and shivering. Anesthesiology 1993; 79:1193–201.
8. Wrench IJ, Cavill G, Ward JE, Crossley AWA. Comparison between alfentanil, pethidine and placebo in the treatment of postanaesthetic shivering. Br J Anaesth 1997; 79:541–2.
9. Wang JJ, Ho ST, Chun S, Liu YC. A comparison among nalbuphine, meperidine and placebo for treating postanesthetic shivering. Anesth Analg 1999; 88:686–9.
10. DeWitte J, Deloof T, DeVeylder J, Housmans PR. Tramadol in the treatment of postanesthetic shivering. Acta Anaesthesiol Scand 1997; 41:506–10.
11. Buggy D, Higgins P, Moran C, et al. Clonidine at induction reduces shivering after general anaesthesia. Can J Anaesth 1997; 44:263–7.
12. Singh P, Dimitriou V, Mahajan RP, Crossley AW. Double-blind comparison between doxapram and pethidine in the treatment of postanaesthetic shivering. Br J Anaesth 1993; 71:685–8.
13. Horn EP, Standl T, Sessler DI, et al. Physostigmine prevents postanesthetic shivering as does meperidine or clonidine. Anesthesiology 1998; 88:108–13.
14. Hindle AT. Recent developments in the physiology and pharmacology of 5-hydroxytryptamine. Br J Anaesth 1994; 73:395–407.
15. Joris J, Banache M, Bonnet F, et al. Clonidine and ketanserin both are effective treatment for postanesthetic shivering. Anesthesiology 1993; 79:532–9.
16. Dawson NJ, Malcolm JL. Initiation and inhibition of shivering in the rat: interaction between peripheral and central factors. Clin Exp Pharmacol Physiol 1982; 9:89–93.
17. Gardner C, Perren M. Inhibition of anaesthetic-induced emesis by a NK1 or 5-HT3 receptor antagonist in the house musk shrew, Suncus murinus. Neuropharmacol 1998; 37:1643–4.
18. Horn EP, Werner C, Sessler DI, et al. Late intraoperative clonidine administration prevents postanesthetic shivering after total intravenous or volatile anesthesia. Anesth Analg 1997; 84:613–7.
19. Guyton AC. Body temperature, temperature regulation and fever. In: Guyton AC, ed. Textbook of medical physiology. 6th ed. Philadelphia: WB Saunders, 1991: 886–98.
20. Matsukawa T, Sessler DI, Sessler AM, et al. Heat flow and distribution during induction of general anesthesia. Anesthesiology 1995; 82:662–73.
21. Ikeda T, Sessler DI, Kikura M, et al. Less core hypothermia when anesthesia is induced with inhaled sevoflurane than with intravenous propofol. Anesth Analg 1999; 88:921–4.
22. Diemunsch P, Conseiller C, Clyti N, Mamet JP. Ondansetron compared with metoclopramide in the treatment of established postoperative nausea and vomiting: the French Ondansetron Study Group. Br J Anaesth 1997; 79:322–6.
23. De Witte JL, Kim J-S, Sessler DI, et al. Tramadol reduces the sweating, vasoconstriction and shivering thresholds. Anesth Analg 1998; 87:173–9.
24. Sharma V, Fry ENS. Doxapram after general anaesthesia: its role in stopping shivering during recovery. Anaesthesia 1991; 46:460–1.
© 2000 International Anesthesia Research Society