[Prophylaxis of nausea and vomiting in the postoperative phase: relative effectiveness of droperidol and metoclopramide].

The aim of the present study was to conduct a meta-analysis of the results from randomized controlled trials investigating the relative efficacy of droperidol versus metoclopramide for the prevention of postoperative nausea and vomiting (PONV). A systematic literature search for randomized controlled trials comparing droperidol and metoclopramide for the prevention of PONV was performed according to the PRISMA recommendations. The incidence of PONV within the early (0-6 h) and cumulative postoperative periods (0-48 h) was collated and the pooled relative risk (RR) with the corresponding 95% confidence interval (CI) was calculated. Results from a subgroup analysis are presented excluding the data of a Japanese group (Fujii et al.) which are given in parentheses. A total of 41 (30) trials with a total number of 3,491 (2,721) patients were included and of these 12 (8) trials with 1,403 (1,083) patients reported data of the early period and 32 (21) studies with 2,656 (1,836) patients comprised data of the cumulative period. A total of 1,797 (1,309) patients were treated with droperidol (0.25-5 mg) and 1,694 (1,412) with metoclopramide (5-50 mg). In the early period the risk for PONV after metoclopramide was 35% (95%-CI: 17-57%) higher than after prophylaxis with droperidol (without Fujii data: 46%; 23-73%). During the cumulative period the risk for PONV after metoclopramide was increased by 20% (95%-CI: 7-37%) compared to droperidol (without Fujii data: 25%; 4-50%). Due to heterogenous dosing of both drugs subgroup analyses with distinct dose intervals were performed with increments of 0.75 mg for droperidol and 7 mg for metoclopramide. Droperidol was superior in 17 (12) out of 19 (14) subgroup analyses. Comparing recommended doses of droperidol (0.75-1.5 mg) with low doses of metoclopramide (7-14 mg) and medium metoclopramide doses (14-21 mg) PONV was increased by 12% (95%-CI: -11% to 42%) and 32% (95%-CI: 4%-66%), respectively when metoclopramide instead of droperidol was used. When higher doses of metoclopramide (>20 mg) were used the superiority of droperidol was less pronounced and did not reach statistical significance due to the limited numbers of trials included in this analysis (3 studies, 662 patients). The risk for PONV after high-dose metoclopramide was increased by 13% (95%-CI: -21% to +61%) for the early period and by 19% (95%-CI: -11% to +57%) for the cumulative observation period. For the prevention of postoperative nausea and vomiting droperidol is significantly superior to metoclopramide doses below 20 mg. There was no obvious positive dose response with respect to increasing doses of metoclopramide. There was also a trend towards higher efficacy of droperidol compared to higher doses of metoclopramide (≥20 mg). However, there were not enough comparative studies to show a statistically significant result in this subgroup analysis. These data support the notion that droperidol in low doses may represent the more effective D(2)-antagonist for a pharmacological armamentarium to cope with PONV.