The use of a combination of antiemetics with different mechanisms of action is an important option for the treatment of postoperative nausea and vomiting (PONV).1,2 Droperidol, a major tranquilizer with dopamine 2 (D2) receptor antagonist effect, has often been tested in combination with other antiemetics for treating PONV.1–3 However, in December 2001, the Federal Food and Drug Administration issued a “black-box” warning for droperidol due to its potential QT prolongation-related cardiac side effect.4 This announcement led to a marked reduction of droperidol use for PONV and a search for a substitute.
Haloperidol, a major tranquilizer with a D2-receptor antagonist effect, was considered a possible substitute for droperidol.5–10 Haloperidol has been used in palliative care as an antiemetic for nausea and vomiting.7 Recently, haloperidol was further found to have an effect on PONV.8–10 However, most of the previous reports studied haloperidol as a sole drug for PONV, and combinations of haloperidol with other potent antiemetics, e.g., dexamethasone, were rarely evaluated. In addition, the advantage of combining droperidol with dexamethasone for the prevention of PONV has been established.11 It thus would be expected that haloperidol, when combined with dexamethasone, would similarly provide enhanced prophylaxis than haloperidol alone.
The aim of the present study was to evaluate the prophylactic effect of haloperidol plus dexamethasone on PONV. We hypothesized that prophylactic haloperidol plus dexamethasone will provide a greater reduction in the incidence of PONV than either drug used alone, placebo or droperidol.
We conducted a double-blind, randomized, placebo, and positive-control study in patients undergoing laparoscopic-assisted vaginal hysterectomy (LAVH). The study protocol was approved by the IRB of Human Research, Chi-Mei Medical Center, and written informed consent was obtained from all enrolled 400 ASA class I or II women scheduled for LAVH. The exclusion criteria were clinical evidence of a difficult airway, obesity (body mass index >35 kg/m2), pregnancy, psychiatric illness, clinically significant major organ disease, or consumption of a drug with antiemetic properties within 24 h before the study.
In the preoperative holding area, patients were randomly assigned by use of a computer generated random number table into one of five groups (n = 80 for each group): (1) group S, saline; (2) group D, droperidol 1.25 mg; (3) group H, haloperidol 2 mg; (4) group Dx, dexamethasone 5 mg; or (5) group H + Dx, haloperidol 2 mg plus dexamethasone 5 mg. Study medications were prepared by a nurse anesthetist in identical syringes with a total volume of 2 mL (dilution with saline), and given IV 15 min after the induction of anesthesia. Patients and the investigators who collected the postoperative data were blinded to the randomization.
Anesthesia was standardized for all patients. Anesthesia was induced using fentanyl (2 μg/kg; IV), lidocaine (1 mg/kg; IV), and propofol (2 mg/kg; IV). Laryngoscopy and endotracheal intubation was facilitated with rocuronium (1 mg/kg; IV). Anesthesia was maintained using 8%–12% (inspired concentration) desflurane in oxygen. No additional opioids were given during surgery. Ventilation was controlled mechanically, and adjusted to maintain end-tidal CO2 values between 35 and 45 mm Hg throughout the surgery. Additional rocuronium was administered as required. A LAVH was performed under video guidance with three punctures in the abdomen.12 At the end of surgery, residual neuromuscular blockade was antagonized using neostigmine (0.04 mg/kg; IV) mixed with atropine (0.02 mg/kg; IV), and the trachea was extubated after the recovery of spontaneous respiration and consciousness.
During anesthesia, a standard lead II electrocardiogram was recorded at a paper speed of 25 mm/s and amplification of 0.1 mV/mm at the time of drug injection, as well as every 1 min for 10 min after injection. The QT intervals were measured manually from the onset of the QRS complexes to the end of the T wave and corrected for the patient’s heart rate using a previously published formula.13 A corrected QT interval (QTc) was then obtained.
After surgery, patients were observed for 24 h. They were first transferred to the postanesthetic care unit (PACU) where they remained for 2 h. Thereafter, they were moved to a ward. These three variables are (1) the incidence and severity of PONV, (2) the severity of postoperative pain, and (3) the occurrence of postoperative extra-pyramidal side effects (EPS) were monitored every 15 min in the PACU and every 4 h from 8:00 am to 10:00 pm in the ward.
The incidence and severity of PONV were evaluated using these four variables: the incidences of (1) nausea, (2) vomiting, (3) PONV (nausea + vomiting), and (4) rescue antiemetics used. Nausea alone was graded as tolerable or intolerable nausea. Patients who reported intolerable nausea were treated with rescue antiemetic (ondansetron 4 mg; IV). For data collection, the incidences of both tolerable and intolerable nausea were combined and considered as nausea. Vomiting alone, vomiting with nausea, and retching were all considered as vomiting. A rescue antiemetic (ondansetron) was given, and repeated if necessary, if vomiting or intolerable nausea occurred. A complete response was defined as no PONV during the 0–24 h postoperative period. This constituted the primary end-point of the study. Secondary end-points of the study were the side effects of each treatment.
Postoperative pain was assessed 24 h after surgery using a 10-cm visual analog scale score (VAS; 0 = no pain to 10 = most severe pain). During the observation period, ketorolac (30 mg; IV) was given if a VAS score of pain exceeded 5 or if a patient complained of pain and requested an analgesic. If pain persisted after two injections of ketorolac (30 mg), meperidine (25 mg; IV) was given instead.
The level of sedation was classified as: 1 = awake; 2 = drowsy, responds to verbal commands; 3 = asleep, responds to touch or pain stimuli; and 4 = does not respond. This was evaluated 30 min after patients arrived in the PACU. All EPS (motor restlessness, acute dystonia, or tardive dyskinesia) during the 24-h observation period were recorded.
Sample size was predetermined using a power analysis based on two assumptions: first, the total incidence of PONV without antiemetic prophylaxis would be 65%,14–16 and, second, a 25% reduction in the total incidence of nausea and vomiting (from 65% to 40%) would be clinically relevant (α = 0.05 [two-sided] and β = 0.2 [power = 80%]). The calculated minimum sample size was 62 patients in each group. A series of one-way analyses of variance was conducted to examine differences in the parametric variables among the five groups. If a significant difference was found, the Bonferroni t-test was used to detect the intergroup differences. The Kruskal-Wallis test was used to determine differences in the nonparametric variables among groups, and the Mann–Whitney rank sum test was then used to detect inter-group differences. Categorical variables were analyzed using a series of 2 × 5 χ2 tests to detect differences among groups, and a 2 × 2 χ2 test or Fisher’s exact test, as appropriate, was then used to detect inter-group differences. All follow-up analyses were corrected for the number of simultaneous comparisons using the Bonferroni adjustment. Statistical significance was set at P < 0.05.
Twenty-eight of the 400 patients enrolled in the study were excluded from the analysis; 16 patients required an abdominal hysterectomy and, in 12 patients, we were unable to collect complete data. The data obtained from the remaining 372 patients were analyzed. Patient characteristics such as age, weight, height, histories of PONV and motion sickness, and durations of surgery and anesthesia were not different among the five groups (Table 1).
When compared with the S group, the D, H, Dx, and H + Dx groups had a lower incidence of PONV (2–24 h and 0–24 h) and higher incidence of complete response (Table 2). The H + Dx group had the lowest incidence of PONV (2–24 and 0–24 h) and the highest incidence of complete response (Table 2). No differences were found among the D, H, and Dx groups (Table 2).
During anesthesia, the QTc interval after the administration of the test medication in all groups was not different from their preinjection values. The mean values of QTc intervals in all groups during a 10-min observation period were below 500 ms (data not shown). No differences among groups were found in QTc intervals during the 10-min observation period.
Within the 24-h observation period, between 77% and 87% of the patients were given ketorolac, and 19% to 26% of the patients were given ketorolac with meperidine as rescue analgesics (Table 3). The mean values of the VAS pain scores in all groups were between 2 and 4 during the 2–24-h postoperative observation period. The intensity of postoperative pain and the use of ketorolac and meperidine were not different among groups. The levels of sedation were between 1 and 2 in all groups 30 min after the patients arrived in the PACU, and we found no differences in sedation levels among groups.
Three patients (two in the droperidol group and one in the haloperidol group) reported EPS (motor restlessness). This symptom was relieved within 15 min after they were given IV diphenhydramine hydrochloride 30 mg. No untoward long-term consequences were observed. No patients in the H + Dx group reported any symptoms of EPS, and no differences were found in the incidence of EPS among the five groups.
We demonstrated that a combination of haloperidol 2 mg plus dexamethasone 5 mg produced a greater reduction in the incidence of PONV than did either drug used alone, without increasing the following perioperative risks: QT prolongation, intensity of postoperative pain, level of sedation, and occurrence of EPS.
PONV is one of the most common and annoying side effects in female patients undergoing laparoscopic surgery.2,16 Risk factors, such as female gender, the laparoscopic technique, the duration of anesthesia, desflurane and fentanyl, the intensity of postoperative pain, and analgesics, may contribute to PONV.2,14,16 In our study design, we minimized differences in all of these factors. All the patients received a LAVH done by the same team of anesthesiologists and surgeons. Furthermore, no differences were found among groups in the use of desflurane and fentanyl, the duration of anesthesia and surgery, the intensity of postoperative pain, and the use of analgesics (Table 1). The incidences of motion sickness and previous PONV were not different (Table 1). Therefore, the differences in the occurrence of PONV among groups can likely be attributed to the differences in the study medications given to the patients in each group.
Haloperidol, a major tranquilizer having a similar mechanism of action as droperidol, was recently found to have a prophylactic effect on PONV.8–10 In the current study, we found that small doses of haloperidol (2 mg) and droperidol (1.25 mg) alone effectively prevented PONV but did not produce a QT prolongation or increase levels of sedation occurrences of EPS.
A combination of haloperidol and dexamethasone produced a greater reduction in the incidence of PONV than either drug used alone, without increasing the perioperative risks. This might have been because when these two antiemetics with different mechanisms of action are combined, only their antiemetic effects are increased.1,3 Haloperidol, like droperidol, is an antiemetic through its antagonism of the D2-receptors in the chemoreceptor trigger zone of the medulla, and dexamethasone is an antiemetic because it activates the glucocorticoid receptors in the bilateral nucleus tractus solitarii of the medulla.7,17
Because dexamethasone has a slow onset of antiemetic action (about 2 h), it is commonly given at the induction of anesthesia.18 To standardize the time of drug administration and simplify the blinding process, all medications tested were given at 15 min after the induction of anesthesia.
A limitation of our study was that our results are not sensitive enough to detect the anti-nausea effect of the antiemetics tested because we did not use a VAS score, a more sensitive method, to evaluate the severity of nausea.
In conclusion, this study demonstrated that haloperidol alone offers no significant advantages over droperidol or dexamethasone for routine antiemetic prophylaxis. The use of a combination (i.e., haloperidol + dexamethasone) is better than any single drug alone, without increasing adverse perioperative outcomes.
1. Tramer MR. A rational approach to the control of postoperative nausea and vomiting: evidence from systematic reviews. Part II. Recommendations for prevention and treatment, and research agenda. Acta Anaesthesiol Scand 2001;45:14–19
2. Eberhart LH, Mauch M, Morin AM, Wulf H, Geldner G. Impact of a multimodal anti-emetic prophylaxis on patient satisfaction in high-risk patients for postoperative nausea and vomiting. Anaesthesia 2002;57:1022–7
3. Chan MT, Choi KC, Gin T, Chui PT, Short TG, Yuen PM, Poon AH, Apfel CC, Gan TJ. The additive interactions between ondansetron and droperidol for preventing postoperative nausea and vomiting. Anesth Analg 2006;103:1155–62
4. White PF, Song D, Abrao J, Klein KW, Navarette B. Effect of low-dose droperidol on the QT interval during and after general anesthesia: a placebo-controlled study. Anesthesiology 2005; 102:1101–5
5. Ayd FJ Jr. Haloperidol: twenty years’ clinical experience. J Clin Psychiatry 1978;39:807–14
6. Vella-Brincat J, Macleod AD. Haloperidol in palliative care. Palliat Med 2004;18:195–201
7. Critchley P, Plach N, Grantham M, Marshall D, Taniguchi A, Latimer E, Jadad AR. Efficacy of haloperidol in the treatment of nausea and vomiting in the palliative patient: a systematic review. J Pain Symptom Manage 2001;22:631–4
8. Aouad MT, Siddik-Sayyid SM, Taha SK, Azar MS, Nasr VG, Hakki MA, Zoorob DG, Baraka AS. Haloperidol v.s. ondansetron for the prevention of postoperative nausea and vomiting following gynaecological surgery. Eur J Anaesthesiol 2007;24: 171–8
9. Parlow JL, Costache I, Avery N, Turner K. Single-dose haloperidol for the prophylaxis of postoperative nausea and vomiting after intrathecal morphine. Anesth Analg 2004;98:1072–6
10. Buttner M, Walder B, von Elm E, Tramer MR. Is low-dose haloperidol a useful antiemetic? A meta-analysis of published and unpublished randomized trials. Anesthesiology 2004;101: 1454–63
11. Pueyo FJ, Lopez-Olaondo L, Sanchez-Ledesma MJ, Ortega A, Carrascosa F. Cost-effectiveness of three combinations of antiemetics in the prevention of postoperative nausea and vomiting. Br J Anaesth 2003;91:589–92
12. Tsai EM, Chen HS, Long CY, Yang CH, Hsu SC, Wu CH, Lee JN. Laparoscopically assisted vaginal hysterectomy versus total abdominal hysterectomy: a study of 100 cases on light-endorsed transvaginal section. Gynecol Obstet Invest 2003;55:105–9
13. Wisely NA, Shipton EA. Long QT syndrome and anaesthesia. Eur J Anaesthesiol 2002;19:853–9
14. Apfel CC, Laara E, Koivuranta M, Greim CA, Roewer N. A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers. Anesthesiology 1999;91:693–700
15. Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997;52:443–9
16. Gan TJ. Risk factors for postoperative nausea and vomiting. Anesth Analg 2006;102:1884–98
17. Ho CM, Ho ST, Wang JJ, Tsai SK, Chai CY. Dexamethasone has a central antiemetic mechanism in decerebrated cats. Anesth Analg 2004;99:734–9
18. Wang JJ, Ho ST, Tzeng JI, Tang CS. The effect of timing of dexamethasone administration on its efficacy as a prophylactic antiemetic for postoperative nausea and vomiting. Anesth Analg 2000;91:136–9