Portal de Periódicos da CAPES

Droperidol: past, present and future

2009; Wiley; Volume: 64; Issue: 11 Linguagem: Inglês

10.1111/j.1365-2044.2009.06124.x

1365-2044

J. Robert Sneyd,

Pathogenesis and Treatment of Hiccups

Droperidol, a butyrophenone derivative and dopamine D2 receptor antagonist, was developed in the 1950s as an antipsychotic and was first approved in Denmark in 1963 for use in anaesthesia, postoperative nausea and vomiting (PONV) and psychiatry. It has subsequently been used for neurolept anaesthesia and as an anti-emetic. Droperidol acts centrally at sites where dopamine, norepinephrine and serotonin act and binds to dopamine D2 receptors in the emetic centre in the chemoreceptor trigger zone, thereby blocking one of the pathways leading to nausea and vomiting [1]. Since the late 1970s, low-dose droperidol (0.625–1.25 mg) has been widely used to prevent and treat PONV in adults and children [2, 3]. Droperidol was withdrawn from clinical use in 2001 but has recently been re-licenced in Europe for the prevention and treatment of PONV (0.625–1.25 mg in adults; 20–50 μg.kg−1, maximum 1.25 mg as second line in children over 2 years old and adolescents) and for the prevention of nausea and vomiting during postoperative patient-controlled analgesia (PCA) in adults. It is therefore appropriate to remind ourselves why it disappeared, what has changed since then and where it fits into our clinical practice today. During the 1990s cases of QT prolongation were linked to the use of droperidol. Due to the accumulating evidence of QT prolongation and concerns about the risk of torsades de pointes, the manufacturer Janssen-Cilag Pharmaceuticals stopped marketing the oral dose form in 2001 to prevent its use in chronic psychiatric conditions. The injectable form was withdrawn for commercial reasons rather than safety concerns [4]. At the end of that year, the United States Food and Drug Administration (FDA) issued a ‘black box’ warning concerning droperidol and torsades de pointes [5]. The warning applied to the approved US dose of intravenous (IV) droperidol (Inapsine®: starting dose 2.5 mg). This was followed by a collapse in the use of droperidol in the US despite continued interest from clinicians – in 2007 an independent international consensus panel remarked that ‘If it were not for the black-box warning, droperidol would have been the panel’s overwhelming first choice for PONV prophylaxis’ [6]. In 2001 the FDA warned that ‘In the last year, there have been reports of torsades de pointes within or below the currently labelled dose range. There have also been reports of sudden death or other serious cardiac adverse events.’ This warning has subsequently been dissected [7], criticised [8] and defended [9]. Examination of the FDA database showed ten cases in which serious cardiovascular events were possibly related to the administration of droperidol at doses of 1.25 mg or less but analysis of these cases suggested that there was a lack of evidence of a cause-and-effect relationship [7]. The FDA points out that the black box warning applies only to high-dose droperidol (≥ 2.5 mg) and it has not been asked to review nor has it reviewed low-dose droperidol [10]. Interestingly, calculations extrapolated from in vitro data indicate that low-dose droperidol, even a 0.625 mg dose, may produce blood concentrations which affect cardiac conduction [9]. In contrast, a study in a day surgery setting indicated that the corrected QT (QTc) changes following saline and droperidol 0.625 or 1.25 mg were modest, similar, reversible and of short duration [11]. Furthermore, when QTc changes were compared in randomised studies, both droperidol and ondansetron were associated with modest increases in QTc with no difference [12] or a small but statistically significant difference [13] between the two drugs and the combination of the two drugs produced no additional increase in QTc [14]. When small differences were detected, the QTc interval returned to baseline or below within 10 min of injection [11]. Moreover, the relevance of these small changes in such a surrogate endpoint is unknown. Thus, there are few data to support a link between low-dose droperidol and torsades de pointes. The difficulty of proving a negative relationship between droperidol and QTc prolongation [15] and the generic status of droperidol has stymied progress in the discussion. QTc changes are ubiquitous in anaesthesia and have recently been described following induction of anaesthesia with propofol [16], etomidate [17] or sevoflurane [18] when combined with laryngoscopy; however, this effect is attenuated by fentanyl and remifentanil [18, 19]. This dynamic background further complicates interpretation of QTc changes attributed to droperidol as perioperative use is invariably associated with the co-administration of anaesthetic agents and narcotics. In a Cochrane review of drugs for prevention of PONV, the authors calculated that compared to placebo, droperidol at any dose reduced the relative risk (RR) of PONV by 38% (RR 0.62, 95% confidence interval 0.58–0.67) [20]. In a general surgical population, droperidol 0.625 mg administered 30 min before the end of anaesthesia reduced the incidence of PONV to 7%, compared with 41% for placebo [21]. The number needed to treat (NNT) for droperidol to prevent PONV has been estimated at 7–8 for doses between 1 and 1.25 mg [22]. Ondansetron 4 mg is reported to have a NNT of 6–7 [6]. These numbers concur with the IMPACT study which indicated that a single anti-emetic intervention in a patient with an 80% risk of PONV at baseline will reduce the risk to 59%, translating to a NNT of approximately 5 to prevent nausea and vomiting in one patient [23]. However, for a patient with a baseline risk of 10%, the absolute risk reduction is about 3%, which corresponds to a NNT of around 40. In the IMPACT study a factorial design randomly assigned 4123 patients at risk of PONV to receive: 4 mg of ondansetron or no ondansetron; 4 mg of dexamethasone or no dexamethasone; 1.25 mg of droperidol or no droperidol; propofol or a volatile anaesthetic; nitrogen or nitrous oxide; and remifentanil or fentanyl. Ondansetron, dexamethasone and droperidol each reduced the relative risk of PONV by about 26% [23]. However a superior antinausea effect for droperidol 1.25 mg in comparison with ondansetron 4 mg has been reported [24] and noted in guidelines [6, 20]. IMPACT reported that the addition of second and third anti-emetics resulted in substantial additional reductions in PONV, from 37% with one anti-emetic to 28% with two and 22% with three anti-emetics [23]. The combination of dexamethasone and droperidol reduced the incidence of PONV by about half compared with 15% for the single agents [23]. Efficacy, cost and the potential for side effects are rational criteria to consider when making this choice. As the IMPACT study reported that single agent dexamethasone, droperidol and ondansetron had equivalent efficacy [23], and these drugs are all relatively inexpensive for hospitals to purchase (< £2 per single dose), then practical considerations and the drugs’ individual side-effect profiles should be considered. Unlike dexamethasone, which should be administered at the induction of anaesthesia [25], droperidol should be administered 30 min before the end of surgery [26]. Ondansetron should be given at the end of anaesthesia or as required for established PONV [27]. Ondansetron is widely available on wards and in the postoperative recovery area and frequently administered by nursing staff. In contrast, if previous experience is repeated, droperidol, like dexamethasone, will typically be given in the operating room by the anaesthetist and even if prescribed for postoperative use would not usually be administered by nursing staff unfamiliar with its use. Both the IMPACT study [23] and the Cochrane review [20] suggest that the use of multiple anti-emetics of different pharmacological classes gives a cumulative benefit. A practical scheme for patients at risk of PONV might therefore comprise dexamethasone or droperidol (medium-risk patients) or dexamethasone and droperidol (high-risk patients) given during anaesthesia with ondansetron reserved for rescue therapy either in the recovery area or on the general ward [28]. Dexamethasone is efficacious for preventing PONV, but is ineffective once nausea and/or vomiting is established due to its slow mode of action [29]. However, although a low dose, typically 4 mg, of dexamethasone appears to be well-tolerated [30], its systemic effects require further evaluation and care is required when considering use in patients with diabetes [31]. Moreover, dexamethasone is not approved for PONV. Opioid PCA is highly emetogenic and two-thirds of patients may suffer an emetic event without treatment [32]. In contrast to other anti-emetics, droperidol is licensed for addition to PCA [26]. A systematic review of strategies for patients receiving PCA found that although there was no clear dose–response relationship, droperidol reduced nausea and vomiting by over 50% compared with placebo. Evidence for efficacy of other anti-emetics was lacking. [32]. A subsequent dose-finding study compared droperidol doses of 0, 5, 15 or 50 μg.mg−1 of morphine PCA and concluded that droperidol 5 μg.mg−1 of morphine is not anti-emetic, antipruritic or sedative; droperidol 15 μg.mg−1 of morphine has some anti-emetic efficacy, reduces the incidence of pruritus, and is not sedative; droperidol 50 μg.mg−1 of morphine is anti-emetic, has a similar reduction on the incidence of pruritus as 15 μg, and is sedative [33]. Thus, the optimum scheme for adding droperidol to morphine PCA is between 0.75 and 2.5 mg droperidol per 50 mg morphine in a 50 ml morphine syringe, with scope for further research to determine the optimum dosage within this range. In addition, when droperidol is used in this combination to prevent nausea and vomiting in PCA, it has been associated with reduced use of morphine. In a randomised, controlled study of patients undergoing hysterectomy, the addition of droperidol (50 μg) to morphine (1 mg) PCA significantly reduced morphine consumption vs control (34 mg vs 55 mg) during the first 72 h after surgery, significantly decreased pain scores during movement at 48 and 72 h postoperatively and pain scores at rest at 72 h after surgery [34]. The additive analgesic effect of droperidol is poorly understood, multimodal and complex [34]. The droperidol-morphine-saline solution appears to be stable for up to 14 days [35], although for microbiological reasons this solution should be used immediately [26]. In contemporary practice the addition of an anti-emetic to morphine PCA is not routine. It remains to be seen what impact droperidol will have on the technique, given that it is efficacious and appears to be well-tolerated, despite possible concerns about sedation. Clinical trial data and pooled results from meta-analyses demonstrate that IV droperidol is well-tolerated [21, 22, 24, 33]. Large doses may produce extrapyramidal symptoms, however the relevance of these phenomena to low-dose administration is uncertain [22]. In general, randomised studies report a similar incidence of side effects following either droperidol or ondansetron administration [24, 36]. Concern is sometimes expressed regarding the risk of agitation with droperidol usage. However, Fortney et al. did not find an increased risk of agitation or anxiety with droperidol 1.25 mg vs ondansetron [24]. In addition, Tramèr observed that the incidence of minor effects including anxiety, restlessness and agitation was only increased with respect to placebo at droperidol doses > 4 mg per day [32]. The optimal approach to the management of PONV is multimodal [6]. A suitable integrated strategy includes perioperative fluid administration, IV anaesthesia, avoidance of nitrous oxide and one or more prophylactic anti-emetics given (on the basis of Apfel scoring) to those patients most likely to benefit. Clinicians should consider low-dose droperidol monotherapy as an effective treatment option for patients at a moderate risk of PONV and in combination with complementary antiemetics in patients at moderate/high risk of PONV. If droperidol is combined with dexamethasone, which has a long plasma half life [37], the 5HT3 antagonist may be reserved for rescue therapy [28]. If there are any concerns about sedation with droperidol, the 0.625 mg dose could be administered, given its efficacy. Droperidol’s efficacy is well documented in PCA and unlike other anti-emetics, its use is approved in this indication [26], therefore, clinicians should also consider droperidol for this procedure. Professor Sneyd has no financial interest in ProStrakan or any other manufacturer or supplier of droperidol and has not received any financial support or reimbursement in association with this editorial. ProStrakan were given the opportunity to comment on the editorial at draft stage. ProStrakan are a supporter of the Association of Anaesthetists, Overseas Anaesthesia Fund.