Postoperative nausea and vomiting (PONV) remains a significant problem following general anaesthesia. In high-risk populations, such as females undergoing gynaecologic surgery, the incidence of this distressing adverse effect can be as high as 79% . A number of antiemetics have been used to prevent the occurrence of PONV. Droperidol, a butyrophenone, is a cost-effective drug for the prevention of PONV . However, the Food and Drug Administration (FDA)-imposed black box warning due to its potential for QT prolongation has led to a marked reduction in its clinical use, despite the fact that many authors have expressed concern regarding what is widely viewed as an inappropriate action by the FDA [3–5]. Hence, 5-HT3 receptor antagonists, such as ondansetron, have become the first line drug for prevention of PONV in many places .
Haloperidol, a butyrophenone with a high affinity for dopamine D2 receptors similar to droperidol, has been used as an antipsychotic and, for the control of severe agitation in medical and surgical patients. Also, early studies have shown that haloperidol is effective in the setting of PONV [7–9]. Haloperidol has been widely used as an antiemetic despite a lack of recent clinical data on efficacy and adverse effects . A recent systematic review addressed the usefulness of haloperidol as an antiemetic in palliative care patients receiving large doses of morphine . Also, a recent meta-analysis of randomized trials in the set up of PONV, gastroenterology, chemotherapy and radiation therapy suggested that haloperidol is antiemetic at doses much lower than those used to treat psychiatric disorders . For the prevention and treatment of PONV, parenteral single doses of 1–2 mg are efficacious with minimal toxicity . Lately, Parlow and colleagues showed that a single intramuscular (i.m.) dose of haloperidol 1 or 2 mg decreases the incidence of intrathecal morphine-induced PONV . Most of the studies related to the antiemetic effect of haloperidol for PONV after general anaesthesia are old [7–9] and compare its efficacy to placebo groups.
At our institution, droperidol is not available anymore, and ondansetron is the most widely used prophylactic antiemetic. Thus, we designed this prospective, randomized, double-blinded study to compare the efficacy and adverse effect profile of the prophylactic administration of haloperidol 1 mg intravenous (i.v.) and ondansetron 4 mg i.v. vs. placebo, for the prevention of PONV in patients undergoing gynaecological procedures.
After institutional review board approval and written informed consent, 93 female patients ASA Grade I–II, aged 23–64 yr, scheduled for gynaecological surgery under general anaesthesia were enrolled in the study. Exclusion criteria included pregnancy, breast feeding, psychiatric illness, previous intolerance to ondansetron or haloperidol, preexisting nausea or intake of antiemetics 24 h prior to the study, morbid obesity and major organ disease.
Preoperative factors that influence the incidence of PONV, such as history of PONV, motion sickness, or smoking were recorded. Patients were assigned to one of three treatment groups using a computer generated table of random numbers: patients in the haloperidol group received haloperidol 1 mg i.v. following induction of anaesthesia and saline 10 min before the end of surgery; patients in the ondansetron group received saline following induction of anaesthesia and ondansetron 4 mg i.v. 10 min before the end of surgery; patients in the control group received saline following induction of anaesthesia and 10 min before the end of surgery. Study drugs were mixed with 0.9% saline to a final volume of 5 mL by personnel not involved in the study and were administered in a double-blind fashion. In all patients, anaesthesia was induced with i.v. lidocaine 1.5 mg kg−1, fentanyl 2 μg kg−1 and propofol 1.5–2 mg kg−1. Lidocaine was used to blunt the haemodynamic response to laryngoscopy  and to decrease pain following i.v. propofol. Tracheal intubation was facilitated with cisatracurium 0.15 mg kg−1. Anaesthesia was maintained with 67% nitrous oxide in oxygen, 1–1.5% isoflurane and additional boluses of cisatracurium and fentanyl as required. A nasogastric tube was inserted following induction of anaesthesia and suction was applied to empty the stomach. Prior to extubation of the trachea, the stomach was again suctioned via the nasogastric tube, which was then removed. Paracetamol 1 g i.v. was administered half an hour prior to the end of surgery to control postoperative pain. Residual neuromuscular blockade was antagonized with neostigmine 0.05 mg kg−1 and glycopyrrolate 0.01 mg kg−1 in all patients. Emergence time, defined as the time from discontinuation of isoflurane to tracheal extubation, was recorded.
In the postanaesthesia care unit, the following parameters were recorded on arrival and every 30 min, or whenever PONV occurred during a period of 2 h: nausea score, episodes of vomiting, sedation score, visual analogue pain score (VAS) and adverse effects such as dysrhythmias, extrapyramidal signs or headache. Scales used to obtain data included a 5-point descriptor score for nausea using Melzack Overall Nausea Index  (0 = none, 1 = mild, 2 = discomforting, 3 = distressing, 4 = horrible, 5 = worst possible), a 5-point descriptor score for sedation (1 = alert, 2 = occasionally drowsy, 3 = frequently drowsy, 4 = sleepy, easy to arouse, 5 = somnolent, difficult to arouse) and a 10-point pain score (0 = no pain, 10 = worst pain imaginable). Retching was classified as vomiting. Complete responders were those who experienced no nausea, no vomiting and did not need any rescue antiemetics. Electrocardiogram (ECG), blood pressure, heart rate (HR) and oxygen saturation were continuously monitored. As a rescue medication for nausea and/or vomiting, promethazine 12.5 mg i.v. was given if nausea score was more than two and lasted for at least 10 min, if vomiting occurred more than once, or whenever the patient requested it. If nausea or vomiting persisted after the administration of promethazine, ondansetron 4 mg i.v. was administered. Patients were allowed to receive incremental doses of 20 mg of i.v. meperidine or a diclofenac suppository for pain relief. All patients stayed in the postanaesthesia care unit for 2 h and were then discharged to the floor or to the step-down unit. A 12-lead ECG was performed in all patients after arrival to the postanaesthesia care unit. ECGs were recorded at a paper speed of 25 mm s−1 and amplitude of 0.1 mV mm−1. Reading and analysis of ECG tracings were performed after completion of the study by an investigator who was blinded to the antiemetic drug administered. HR was calculated from three R–R intervals preceding the measured QT intervals. The QT intervals were measured manually from the onset of the QRS complexes to the end of the T wave (defined as the intersection of the isoelectric line and the tangent of the maximal downward limb of the T wave). Each registered measurement was an average of two consecutive QT intervals. The QT interval was adjusted for the patient's HR using the formula of Bazett (QTc = QT/R–R0.5). If the T wave amplitude was too low (<0.1 mV), the lead was excluded from the analysis.
In the postanaesthesia care unit, patients were observed by the same nurse who was blinded to study group assignment. After discharge, patients were allowed to receive analgesics and antiemetics as needed according to the surgeon's prescription. Twenty-four hours following discharge from the postanaesthesia care unit, the cumulative incidence of nausea and vomiting, nausea score, number of vomiting episodes, as well as the antiemetic and the analgesic medications used were recorded. A follow-up call was made to the outpatients 24 h after discharge from the postanaesthesia care unit to collect this information.
Continuous data are reported as mean ± standard deviation and are analyzed using analysis of variance (ANOVA). Categorical data are reported as numbers and percentages and are analyzed using χ2 - or Fisher's exact test as appropriate. Non-parametric data such as scores are reported as medians and ranges and are analyzed using Mann-U-Whitney test. The primary end-point of the study was complete absence of nausea and vomiting. Considering from previous studies that the incidence of PONV in the placebo group would be in the range of 55% , 29 subjects per group would be required to yield a 64% reduction in this incidence as reported by Büttner and colleagues  (power 80% and α = 0.05).
Of 93 patients enrolled, four were excluded (3 loss to follow-up, 1 protocol violation). Therefore, data from 89 patients were analyzed in the postanaesthesia care unit: 29 haloperidol, 30 ondansetron and 30 controls (see Fig. 1). In addition, two patients in the haloperidol group and three patients each in the ondansetron and the placebo group were lost to long-term follow-up. Thus, 27 patients in each group were analyzed for 24 h following discharge from the postanaesthesia care unit. There were no differences among the groups with respect to patients' characteristics, type of surgery, duration of anaesthesia, emergence time, average fentanyl consumption and average Lactated Ringer received (Table 1). No difference among the groups was noted with respect to risk factors predisposing to PONV (Table 2). Nausea scores >1 were analyzed as presence of nausea.
Overall time-interval (0–24h)
During the 24-h study period, the incidence of nausea, vomiting, treatment with antiemetics, nausea and/or vomiting and complete responders, as well as the median and the distribution of the highest nausea scores recorded were not significantly different among the three groups (Table 3, Fig. 2).
Early time-interval (0–2h)
During the first 2 h in the postanaesthesia care unit, the incidence of nausea (P = 0.04), vomiting (P = 0.005), treatment with rescue antiemetics (P = 0.01), nausea and/or vomiting (P = 0.04) and complete responders (P = 0.04), as well as the median of the highest nausea score recorded (P = 0.03) were significantly lower in the haloperidol group compared with the placebo group. These values, although lower in the ondansetron group, were not significantly different from the placebo group. Also, the differences between haloperidol and ondansetron did not reach statistical significance (Table 4). Two patients in the placebo group needed two rescue antiemetics in the postanaesthesia care unit (promethazine and ondansetron). All other patients who were treated for PONV in the postanaesthesia care unit received only promethazine.
Upon arrival in the postanaesthesia care unit, sedation scores were significantly higher in the haloperidol group as compared to the placebo group. At all other time-intervals, as well as after administration of meperidine, sedation scores were comparable among the three groups (Table 5).
In the postanaesthesia care unit, the median VAS pain scores upon arrival and at 30 min were significantly higher in the ondansetron and the placebo groups as compared to the haloperidol group (Table 5). However, the number of patients who received meperidine as well as the average dose of meperidine was similar among the three groups (Table 2).
The QTc interval was measured in all patients upon arrival in the postanaesthesia care unit. The mean QTc interval was not significantly different among the three groups (haloperidol group: 427.16 ± 31.9 ms, ondansetron group: 417.19 ± 23 ms, placebo group: 422.77 ± 18.5 ms, P = 0.3). No patient in any of the groups had a QTc value longer than 470 ms.
Headache was reported in three patients in the ondansetron group. No other relevant adverse effect was noted in the other two groups.
Late time-interval (2–24 h)
Follow-up data during the first postoperative day showed no significant difference among the three groups in the incidence of nausea, vomiting, nausea and/or vomiting, the highest nausea score recorded and the number of vomiting episodes, as well as in the number of patients requiring antiemetics or analgesics and the number of complete responders (Table 6).
Our results show that the prophylactic administration of 1 mg i.v. haloperidol or 4 mg ondansetron did not improve the overall (0–24 h) incidence and severity of PONV vs. placebo in female patients undergoing gynaecological surgery. However, during the early time-interval in the postanaesthesia care unit (0–2 h), haloperidol 1 mg decreased the incidence and severity of PONV, as well as the need for rescue antiemetics as compared to the placebo group. During the remaining 2–24 h, no difference was noted among the three groups.
Our power analysis was based on the findings of a recent meta-analysis by Büttner and colleagues . The authors found a 64% reduction in the incidence of early nausea with haloperidol vs. placebo from 43.8% to 16.4% . Those results are similar to our results during the early time-interval (66% reduction from 40% in the placebo group to 13.7% in the haloperidol group). A 34% reduction was achieved between the ondansetron group and the placebo group during the early time-interval. This reduction is a reasonable estimate of what can be expected from efficacious anti-PONV interventions . However, the lack of statistical significance between ondansetron and placebo may be related to the small number of patients included. In our study, the predisposing factors to PONV were equally distributed among the three groups (Table 2) and were unlikely to explain some of the observed differences.
Droperidol has been considered as the most cost-effective single drug therapy for prevention of PONV . However, since the black box warning of the FDA, droperidol is not available anymore at our institution. In our study, we showed that haloperidol can significantly decrease the incidence and severity of PONV and hence, may be a useful alternative to droperidol. No head to head comparison between droperidol and haloperidol is available in the literature. Thus, future comparative studies are needed.
The plasma half-life of haloperidol is longer than that of droperidol and thus, it has been speculated that the drug probably has a long duration of action even when administered in low doses . Our results did not show any effect of haloperidol beyond 2 h. However, as 8 and 11 out of 30 patients in the ondansetron and the placebo groups received rescue antiemetics during the 0–2 h period vs. only two patients in the haloperidol group, we cannot conclude that haloperidol did not exert any effect during the 2–24 h period, since more patients in the two other groups were protected with the rescue drug they received.
We have chosen an i.v. dose of 1 mg of haloperidol since it has been shown that there is no dose responsiveness within a dose range between 0.5 and 4 mg . Whether increasing the dose to 2 mg might have improved long-term outcome in the haloperidol group needs to be evaluated, and dose-finding studies are clearly warranted. However, any additional benefit has to be weighed against potential adverse effects associated with larger doses.
Concerning the timing of administration of the prophylactic antiemetics, studies have shown that ondansetron was most effective when given at the end of surgery [18,19]. Büttner and colleagues have shown that haloperidol prevented further vomiting at 2–4 h after treatment . Also, Parlow and colleagues administered haloperidol 1 or 2 mg to patients undergoing orthopaedic or endoscopic urologic procedures at the beginning of the surgery to prevent PONV related to intrathecal morphine . Thus, we elected to administer haloperidol immediately following induction of anaesthesia.
The 24-h follow-up has shown that more than 25% of patients in the three groups still suffered from PONV. Therefore, a more effective antiemetic therapy is needed. There is increasing evidence that a multimodal approach may lead to an improved outcome as there is no single intervention that is effective in reducing the incidence of PONV to a satisfying level . In addition, new promising drugs are now available, such as aprepitant. This drug belongs to a new class of antiemetics that exerts its effect by antagonizing the tachykinin substance P at the NK1 receptors in the central nervous system. NK1 receptor antagonists have a broad spectrum antiemetic potential that has proven to be highly effective in the control of both chemotherapy induced and postoperative emesis where multifactorial stimuli for nausea and vomiting are present [20,21].
The antihistaminic promethazine was elected as the first line drug for treatment of PONV following failure of prophylaxis. Data suggest that an antiemetic acting at a receptor different from the site of action of the drug used for PONV prophylaxis should be considered for the treatment of established PONV following failure of prophylaxis .
White and colleagues  demonstrated a similar QTc prolongation following induction of anaesthesia in patients receiving either droperidol for the prophylaxis of PONV or saline. Also, Charbit and colleagues showed that droperidol and ondansetron used to treat PONV induced similar clinically relevant QTc interval prolongation, with maximum QT interval lengthening observed at 2 and 3 min after droperidol and ondansetron respectively . Concerning haloperidol, observational studies have suggested that high-dose haloperidol may cause QT prolongation in critically ill patients in the intensive care unit (ICU) [24,25], or postoperatively . In our report, the QTc interval was recorded in all patients upon arrival in the postanaesthesia care unit and the mean was not significantly different among the three groups. We did not obtain a preoperative ECG for the majority of our patients. However, according to sex-related thresholds, no prolonged QTc interval (QTc > 470 ms in female ) was found immediately after surgery in any patient in the three groups. The 12-lead ECG was not recorded within minutes following the administration of the prophylactic antiemetic therapy to detect the maximum lengthening in the QTc interval, but lead II ECG was continuously monitored in all patients intraoperatively and during the first 2 h postoperatively and did not show any significant dysrhythmias in any of the patients. However, failure to show an effect of haloperidol on the ECG in a group of 30 patients can only permit to conclude that the 95% confidence interval (CI) for the occurrence of these changes is between 0% and 10% .
There is some evidence that haloperidol even at low antiemetic doses may cause sedation and drowsiness, and in rare instances extrapyramidal symptoms . In our report, haloperidol has resulted in higher sedation scores upon arrival in the postanaesthesia care unit than the placebo group. However, this sedation was short lived, since at all the following time-intervals in the postanaesthesia care unit, sedation scores were comparable among the three groups. No extrapyramidal reactions, such as twitching, dystonia, akathisia or rigour were reported in any of our patients. Büttner and colleagues reported that these adverse effects occurred in one patient who had received 4 mg i.v. haloperidol out of 806 patients exposed to 0.25–5 mg haloperidol . In our study, 3 of 30 patients in the ondansetron group developed headache. In a previous meta-analysis, Tramèr and colleagues showed that the number needed to harm for headache with a single dose ondansetron of 4 mg was 36 .
Pain VAS scores were lower in the haloperidol group upon arrival in the postanaesthesia care unit and at 30 min as compared to the other two groups. This finding may be explained by the fact that patients who had received haloperidol were more sedated during this time-interval. Also, animal models have shown that pretreatment with droperidol can produce a leftward shift in the analgesic dose–response curve for fentanyl and thus lower its 50% effective dose . This may apply to haloperidol that is pharmacologically similar to droperidol. In addition, it has been previously demonstrated that ondansetron may reduce the efficacy of analgesics following surgery by interfering with serotoninergic pathways .
In conclusion, the prophylactic administration of 1 mg i.v. haloperidol or 4 mg ondansetron in female patients undergoing gynaecological surgery did not improve the overall 24-h incidence of PONV compared with placebo. However, haloperidol 1 mg proved to be an effective antiemetic for the early prevention of nausea and vomiting, while the reduction achieved with ondansetron during the early phase did not reach statistical significance. Comparative studies with a larger number of patients are needed, as well as dose-finding studies.
This project was funded by the Department.
1. Apfel CC, Laara E, Koivuranta M, Greim CA, Roewer N. A simplified risk score for predicting postoperative nausea and vomiting
2. White PF, Song D, Abrao J, Klein KW, Navarette B. Effect of low dose droperidol on the QT interval during and after general anesthesia. Anesthesiology
3. Habib AS, Gan TJ. Food and drug administration black box warning on the perioperative use of droperidol: a review of the cases. Anesth Analg
4. White PF. Droperidol: a cost-effective antiemetic for over thirty years [Editorial]. Anesth Analg
5. Scuderi PE. You [still] can't disapprove the existence of 1dragons [Editorial]. Anesthesiology
6. Gan TJ, Meyer T, Apfel CC et al
. Consensus guidelines for managing postoperative nausea and vomiting
. Anesth Analg
7. Baron MD, Libonati M, Cohen PJ. The use of haloperidol for treatment of postoperative nausea and vomiting
: a double-blind placebo-controlled trial. Anesthesiology
8. Dyrberg V. Haloperidol [Serenase] in the prevention of postoperative nausea and vomiting
. Acta Anaesthesiol Scand
9. Tornetta FJ. Double-blind evaluation of haloperidol for antiemetic activity. Anesth Analg
10. Büttner M, Walder B, Von Elm E, Tramèr M. Is low dose haloperidol a useful antiemetic? A meta-analysis of published and unpublished randomized trials. Anesthesiology
11. Critchley P, Plach N, Grantham M et al
. Efficacy of haloperidol in the treatment of nausea and vomiting in the palliative patient: a systematic review. J Pain Symptom Manage
12. Parlow JL, Ioana C, Avery N, Turner K. Single-dose haloperidol for the prophylaxis of postoperative nausea and vomiting
after intrathecal morphine. Anesth Analg
13. Taha S, Siddik-Sayyid S, Alameddine M et al
. Propofol is superior to thiopental for intubation without muscle relaxants. Can J Anaesth
14. Melzack R. Measurement of nausea. J Pain Symptom Manage
15. Fortney JT, Gan TJ, Graczyk S et al
. A comparison of the efficacy, safety, and patient satisfaction of ondansetron versus droperidol as antiemetics for elective outpatient surgical procedures. Anesth Analg
16. Apfel CC, Korttila K, Abdalla M et al
. A factorial trial of six interventions for prevention of postoperative nausea and vomiting
. N Engl J Med
17. Eberhart LHJ, Mauch M, Morin AM, Wulf H, Geldner G. Impact of a multimodal antiemetic prophylaxis on patient satisfaction in high-risk patients for postoperative nausea and vomiting
18. Sun R, Klein KW, White PF. The effect of timing of ondansetron administration in outpatients undergoing otolaryngologic surgery. Anesth Analg
19. Graczyk SG, McKenzie R, Kallar S et al
. Intravenous dolasetron for the prevention of postoperative nausea and vomiting
after outpatient laparoscopic gynecologic surgery. Anesth Analg
20. Diemunsch P, Grelot L. Potential of substance P antagonists as antiemetics. Drugs
21. Diemunsch P, Schoeffler P, Bryssine B et al
. Antiemetic activity of NK1
receptor antagonist GR205171 in the treatment of established postoperative nausea and vomiting
after major gynaecological surgery. Br J Anaesth
22. Habib AS, Gan TJ. The effectiveness of rescue antiemetics after failure of prophylaxis with ondansetron or droperidol: a preliminary report. J Clin Anesth
23. Charbit B, Albaladejo P, Funck-Brentano C, Legrand M, Samain E, Marty J. Prolongation of QTc interval after postoperative nausea and vomiting
treatment by droperidol or ondansetron. Anesthesiology
24. Lawrence KR, Nasraway SA. Conduction disturbances associated with administration of butyrophenone antipsychotics in the critically ill: a review of the literature. Pharmacotherapy
25. Sharma ND, Rosman HS, Pdhi ID, Tisdale JE. Torsades de pointes associated with intravenous haloperidol in critically ill patients. Am J Cardiol
26. Perrault LP, Denault AY, Carrier M, Cartier R, Belisle S. Torsades de pointes secondary to intravenous haloperidol after coronary bypass grafting surgery. Can J Anaesth
27. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything alright? JAMA
28. Tramèr MR, Reynolds DJM, Moore RA, Mc Quay HJ. Efficacy, dose–response, and safety of ondansetron in prevention of postoperative nausea and vomiting
: a qualitative systematic review of randomized placebo-controlled trials. Anesthesiology
29. Statile L, Puig MM, Warner W, Bansinath M, Lovitz M, Turndorf H. Droperidol enhances fentanyl and sufentanil, but not morphine, analgesia. Gen Pharmacol
30. Arcioni R, Della Rocca M, Romano S, Romano R, Pietropaoli P, Gasparetto A. Ondansetron inhibits the analgesic effects of tramadol: a possible 5-HT(3) spinal receptor involvement in acute pain in humans. Anesth Analg
Keywords:© 2007 European Society of Anaesthesiology
ANAESTHESIA GENERAL, side-effects; POSTOPERATIVE NAUSEA AND VOMITING; ANTIEMETICS, SURGERY GYNAECOLOGICAL