Patient-controlled analgesia (PCA) with parenteral opioids is now accepted as an effective method for treating post-operative pain. Unfortunately, opioid administration by PCA is associated with a high incidence of post-operative nausea and vomiting (PONV). Rates of PONV of up to 95% have been reported [1-9]; PONV is especially troublesome after gynaecological surgery. In extreme cases, opioid-induced PONV may even be perceived as a failure of PCA as an analgesic technique.
Anti-emetics for the alleviation of PONV associated with PCA opioid may be administered by a variety of routes; for example, intramuscularly, transdermally and intravenously (either alone or as opioid-anti-emetic combinations). The addition of an anti-emetic drug to the PCA-morphine syringe allows the simultaneous administration of both drugs. Several studies have reported that the addition of droperidol to PCA morphine is associated with a reduction in the incidence of PONV [3-5,7,8,10]. However, the incremental dose of droperidol in the PCA bolus in each of these studies has varied greatly between 50 [3,10], 100 [4,7], 125  and 167 μg . Furthermore, in all these studies, an initial loading bolus dose of droperidol (6.75 , 18 , 20 [3,7] or 40 μg kg−1[4,5]) was administered as part of the anaesthetic technique. The studies of Williams et al. and Kaufmann et al. reported the highest doses of droperidol administered, and these authors reported significant increases in patient sedation, anxiety and other side effects. Two studies have specifically investigated the effect of combining droperidol as an initial dose with PCA bolus doses on PONV associated with PCA morphine [7,9]. Barrow et al. compared placebo with droperidol (1.25 mg on induction of anaesthesia) either with or without PCA bolus doses of droperidol 100 μg . These authors reported a reduction in PONV and demands for antiemetic, but no significant increase in sedation associated with the use of either droperidol regimen. In contrast, Gan et al. have compared with placebo the effectiveness of droperidol in an initial dose of 18 μg kg−1, as PCA bolus doses of 160 μg or as a combination of both . These authors advise against administering droperidol in an initial dose together with PCA bolus dosing since combined administration was associated with a greater degree of sedation but no further reduction in PONV.
The wide variation in incremental dose of droperidol in the PCA bolus would seem to suggest that the ideal bolus dose of droperidol has yet to be determined. This study was designed to find the optimal dose of droperidol to be added to the PCA morphine infusate in female patients undergoing gynaecological surgery.
Patient selection and anaesthetic management
Local ethics committee approval was obtained for this study. Healthy, ASA I/II female patients, aged 18-65 years, who were scheduled to undergo elective gynaecological surgery under general anaesthesia were recruited into a double blind, randomized trial of four different doses of droperidol added to the PCA infusate. Patients who had received pre-operative opioid or anti-emetic therapy, who had a history of Parkinsonism, or who had a known sensitivity to phenothiazines or butyrophenones were excluded. Prior to surgery, the use of PCA and the aims of the study were explained and written informed consent obtained. All patients were premedicated with oral temazepam 10-30 mg. A standardized anaesthetic technique was used. Anaesthesia was induced intravenously with thiopentone 3-5 mg kg−1, followed by morphine 0.1-0.2 mg kg−1. Anaesthesia was maintained with isoflurane (<2%) in nitrous oxide and oxygen. Vecuronium 0.1 mg kg−1 was given to facilitate ventilation of the patients' lungs, with further increments as necessary. Droperidol and other anti-emetics were not given pre- or per-operatively.
Post-operatively, analgesia was provided by a Bard PCA pump set to deliver a 1-mL bolus of morphine 1 mg mL−1 with a 5-min lockout period. No background infusion was used. Patients were allocated according to a computer-generated randomization sequence into one of four treatment groups: (1) droperidol 0.05 mg mL−1 with each morphine dose (i.e. 0.05 mg bolus); (2) droperidol 0.10 mg mL−1 with each morphine dose (i.e. 0.10 mg bolus); (3) droperidol 0.15 mg mL−1 with each morphine dose (i.e. 0.15 mg bolus); and (4) droperidol 0.20 mg mL−1 with each morphine dose (i.e. 0.20 mg bolus). The droperidol/morphine syringe preparations were performed by an independent third party who was not involved in any of the subsequent assessments. An initial bolus loading dose of droperidol was not administered. Rescue medication of intramuscular metoclopramide 10 mg was administered for breakthrough PONV. The PCA method was commenced in the recovery room. The severity of emetic symptoms and degree of sedation were recorded by the ward nursing staff every 4 h using a five-point verbal rating scale. For the purposes of the present study, the highest score occurring at 12 and 24 h was recorded. At the end of 24 h, the requirement for rescue anti-emetics, occurrence of neuroleptic side effects and patient satisfaction with therapy were recorded.
A statistical analysis of demographic data, intra-operative morphine dose, post-operative PCA morphine consumption and total dose of droperidol was carried out using the analysis of variance (ANOVA) test. The incidence of nausea, vomiting and demand for rescue anti-emetic were compared using the χ2 test with Yates correction factor. Sedation scores were analysed using the Wilcoxon matched-pair signed rank test for within-group differences at 12 and 24 h and the Wilcoxon rank sum test for group differences at each time point. A value of P<0.05 was considered to be statistically significant.
Eighty patients were entered into the study, but nine patients were excluded for a variety of reasons (inadequate follow-up, protocol violations, problems using the PCA pump and self-withdrawal from the study). Thus, 71 patients completed the study: 17 in group 1; 18 in group 2; 20 in group 3; and 16 in group 4. There were no significant differences between the groups with respect to age, weight or intra-operative morphine dose (Table 1). The PCA morphine consumption was initially high in all groups post-operatively. However, total PCA morphine consumption over the first 24 h post-operatively was similar in all groups, so that, during this period, the overall dose of droperidol received in each group increased in proportion to concentration (Table 1).
Post-operatve nausea and vomiting, and anti-emetic needs
The incidence of emetic symptoms, requests for rescue anti-emetic medication and numbers of doses administered during the first 24 h are outlined in Table 2. The number of symptom-free patients increased as the droperidol dose increased. Results for the pooled data indicated that there was a statistically significant inverse association between the total dose of droperidol received and the severity of PONV (Kendall rank correlation 1.97; P<0.05). However, although nausea and vomiting scores tended to decrease as droperidol dose increased, differences between individual groups did not reach statistical significance. The number of patients requesting rescue anti-emetic was highest in group 1, but was similar in the other three groups. Furthermore, the number of doses of anti-emetic received was greatest in group 1. Nevertheless, there were no significant difference between any of the groups at 24 h (Table 2).
There were no major complications attributable to droperidol. Sedation scores were significantly lower in all groups at 24 h compared with 12 h (P<0.01; Table 3). There was no difference in mean sedation scores between the groups at 12 h, but at 24 h, patients in group 1 and patients in group 2 were significantly less sedated than patients in group 4 (P<0.05). Complaints of anxiety and other minor complications such as blurred vision and tremor tended to be greater with the higher doses of droperidol (Table 4). However, these differences did not reach statistical significance. The PCA was electively discontinued within 24 h in three patients in group 2 and three patients in group 3, but in no case could discontinuation be attributable to the side effects of droperidol.
There is general agreement that the incidence of PONV associated with PCA opioids is unacceptably high. The majority of studies would seem to consider a target rate of PONV of 15-25% to be both acceptable and achievable, although at least one review  has advocated a target rate of zero. The authors of this review point out that, since the aetiology of PONV is multifactorial, the use of combinations of two or even three agents in low doses may be appropriate, particularly as this would reduce the incidence of side effects. A variety of therapeutic manoeuvres, including the addition of an anti-emetic to the PCA infusate, have been devised in an attempt to reduce the incidence of PONV. Currently, there is considerable interest in the use of droperidol added to PCA opioid infusions. Droperidol is an effective anti-emetic with a side effect profile which appears to be dose-related . The combination of droperidol and morphine diluted in 0.9% saline has been documented as being physically and chemically stable in plastic syringes . The addition of a low dose of droperidol to the PCA infusate is a logical first step in a combined approach, with addition of a second anti-emetic agent of a different chemical group as 'rescue' medication as the next step. This has been the standard approach in the majority of studies of PCA droperidol, although the present authors are aware of one study that has successfully employed a triple-therapy approach combining PCA opioid/promethazine with transdermal hyoscine and intravenous droperidol . However, the present authors are unaware of any other study that has been designed specifically to discover that dose of droperidol which when added to the PCA infusate gives the optimal balance between anti-emetic efficacy and side effect profile.
In the present study, the overall droperidol consumption in 24 h, the associated rate of PONV and the requirement for 'rescue' anti-emetic in the four groups were comparable with previously published data. In addition, the present authors found that a total dose of droperidol of between 2.7 and 5.2 mg in 24 h was associated with an acceptable rate (as previously defined) of PONV; this value is similar to that reported by Barrow et al. and Kaufman et al.. As already stated, most previous studies have administered a bolus dose of droperidol either at induction of anaesthesia or on completion of surgery as well as post-operatively, with the result that patients often have received large doses of droperidol. Consequently, the incidence of side effects has been quite high. One author has pointed out that droperidol administered as a bolus on induction of anaesthesia is less effective as a post-operative anti-emetic than droperidol given towards the end of surgery, but that both may still be associated with an increased incidence of side effects . Even moderately low dose droperidol may result in unpleasant side effects as well as increasing sedation [12,14]. This observation has been confirmed by two recent studies of low dose bolus droperidol 0.5 mg which have reported an increased incidence of restlessness and anxiety [15,16]. Conversely, Morrow & Milligan have observed that the use of a lower dose of droperidol in morphine PCA (equivalent to the dose received by patients in group 1 in this study) does not reduce the anti-emetic effectiveness of droperidol . However, these authors also administered an intramuscular bolus dose of droperidol 2.5 mg. Thus, it is likely that their overall droperidol dose was equivalent to the dose received by patients in group 3 in the present study. Therefore, it is of interest that the incidence of PONV in their droperidol-treated morphine PCA patients was 35%, which is similar to the incidence of PONV in patients in group 3 in this study. Morrow & Milligan's observations would also appear to support the contention of Gan et al. that it may be inadvisable to combine an initial loading dose of droperidol with PCA bolus dosing. In the present study, a PCA bolus dose of droperidol 0.20 mg mL−1 was associated with a high incidence of anxiety and other side effects. Indeed, even in the lowest dose group in the present study, one in four patients experienced such side effects, the incidence in this group being comparable with that reported by Watkin et al.. Finally, Barrow et al. and Sharma & Davies  found no significant difference in the degree of sedation associated with increasing doses of droperidol, whilst Roberts et al. found a significant increase in sedation associated with PCA droperidol. The present authors would agree with Roberts et al. since they too found that increasing the total amount of PCA droperidol was associated with increasing sedation.
It has been suggested that the assessment of anti-nausea prophylaxis is reliable only until rescue therapy is administered [1,7]. Bodner & White noted that the use of rescue anti-emetic therapy had an effect on subsequent visual analogue scores which complicated their interpretation . Likewise, Larijani and colleagues attributed the similarity in PONV scores between ondansetron and placebo treatment groups to the greater use of 'rescue' therapy in the placebo group . An aggregate symptom-therapy score that takes into account not only the incidence and severity of PONV, but also the amount of rescue therapy, may actually improve the ability to discriminate between treatment groups . The validity of such an approach has yet to be confirmed. According to the present study, a PCA bolus dose of droperidol of 0.05 mg mL−1 would seem to be ineffective as an anti-emetic. Conversely, a PCA bolus dose of droperidol of 0.20 mg mL−1 is associated with an unacceptable incidence of side effects. Indeed, this dose of droperidol would seem to offer only minimal advantages over a dose of 0.15 mg mL−1. Consequently, the optimal dose of droperidol would seem to be in the region of 0.10-0.15 mg mL−1. However, applying a symptom-therapy scoring approach indicates that a PCA bolus dose of droperidol of 0.10 mg mL−1 is superior in terms of its efficacy/side effect profile than a dose of 0.15 mg mL−1.
The present authors conclude that the optimal balance between anti-emetic efficacy with an acceptable incidence of side effects in female patients undergoing gynaecological surgery who have not previously received an anti-emetic is provided by a PCA bolus dose of droperidol of 0.10 mg mL−1.
1 Silverman DG, Freilich J, Sevarino FB, Paige D, Preble L, O'Connor TZ. Influence of promethazine on symptom-therapy scores for nausea during patient-controlled analgesia with morphine. Anesth Analg
2 Semple P, Madej TH, Wheatley IJ, Jackson JB, Stevens J. Transdermal hyoscine with patient-controlled analgesia. Anaesthesia
3 Sharma SK, Davies MW. Patient-controlled analgesia with a mixture of morphine and droperidol. Br J Anaesth
4 Williams OA, Clarke FL, Harris RW, Smith P, Peacock JE. Addition of droperidol to patient-controlled analgesia: effect on nausea and vomiting. Anaesthesia
5 Kaufman MA, Rosow C, Schneiper P, Schneider M. Prophylactic anti-emetic therapy with patient-controlled analgesia: a double-blind, placebo-controlled comparison of droperidol, metoclopramide and tropisetron. Anesth Analg
6 Cran TJ, Collis R, Hetreed M. Double-blind comparison of ondansetron, droperidol and saline in the prevention of postoperative nausea and vomiting. Br J Anaesth
7 Barrow PM, Hughes DG, Redfern N, Urie J. Influence of droperidol on nausea and vomiting during patient-controlled analgesia. Br J Anaesth
8 Roberts CJ, Millar JM, Goat VA. The antiemetic effectiveness of droperidol during morphine patient-controlled analgesia. Anaesthesia
9 Gan TJ, Alexander R, Fennelly M. Comparison of different methods of administering droperidol in patient-controlled analgesia in the prevention of postoperative nausea and vomiting. Anesth Analg
10 Walder AD, Aitkenhead AR. A comparison of droperidol and cyclizine in the prevention of postoperative nausea and vomiting associated with patient-controlled analgesia. Anaesthesia
11 Matson A, Palazzo M. Postoperative nausea and vomiting. In: Adams AP, Cashman JN, eds. Recent Advances in Anaesthesia and Analgesia 19.
Edinburgh: Churchill Livingstone, 1995: 107-126.
12 Rowbotham DL. Current management of postoperative nausea and vomiting. Br J Anaesth
13 Williams OA, Middleton M, Henderson P, Reilly CS. Stability of morphine and droperidol, separately and combined, for use as an infusion. Hospital Pharmacy Practice
14 Melnick BM. Extra-pyramidal reactions to low dose droperidol. Anesthesiology
15 Foster PN, Stickle BR, Dale M, Laurence AS. Akathisia after low-dose droperidol. Br J Anaesth
16 Watkin S, England AJ, Balcina A, Yentis SM, Raine G, Feldman SA. Minor morbidity after day-case laparoscopy and the influence of droperidol 0.5 mg: a month study. Br J Anaesth
17 Morrow BC, Milligan KR. How much droperidol in combination with PCA morphine? Anaesthesia
18 Bodner M, White PF. Antiemetic efficacy of ondansetron after outpatient laparoscopy. Anesth Analg
19 Larijani GE, Gratz I, Afshar M, Minassian S. Treatment of postoperative nausea and vomiting with ondansetron: a randomized, double-blind comparison with placebo. Anesth Analg