Ginger Does Not Prevent Postoperative Nausea and Vomiting After Laparoscopic Surgery

Ginger (Zingiber officinale) has traditionally been used in China for gastrointestinal symptoms, including nausea and vomiting. As a part of oral premedication, ginger might be especially valuable, because it is inexpensive, lacks known relevant side effects (e.g., sedation), and has a high patient acceptance, because it is part of complementary medicine and is perceived by the public as “natural” and therefore safe (1). Gingerly, especially 8-gingerol, enhances gastrointestinal transport in an animal model (2). Eight-gingerol is a functional antagonist at ileal 5-hydroxytryptamine receptors (3). The subtype 5-hydroxytryptamine type 3 receptor has an important role in the etiology of postoperative nausea and vomiting (PONV) (4). Ernst and Pittler (5) systematically reviewed trials investigating the antiemetic effect of ginger (Zingiber officinale). They performed a metaanalysis of three available studies investigating the herbal remedy in preventing PONV and further critically reviewed another three trials for nausea and vomiting of other conditions. They concluded that ginger was a promising antiemetic, but the clinical data were insufficient to draw firm conclusions. Thus, they called for further rigorous studies to establish whether ginger is effective for nausea and vomiting. In this context, this randomized, placebo-controlled trial was performed to gain further data on the antiemetic properties of the herbal remedy and to evaluate effects on postoperative patient satisfaction. Methods A total of 184 healthy, consenting women undergoing gynecologic laparoscopies were included in this randomized, placebo-controlled trial that was approved by the local ethics committee. They were randomly allocated to one of the following three treatment groups using computer-generated permuted blocks. This list was not broken before patient recruitment and validation of the data were completed. 1) placebo group: 3 × 2 placebo capsules (2 capsules preoperatively as well as 3 and 6 h postoperatively), 2) G300 group: 3 × 1 verum capsules and 3 × 1 placebo capsules (1 capsule preoperatively as well as 3 and 6 h postoperatively) (= 300 mg of ginger extract); and 3) G600 group: 3 × 2 verum capsules (2 capsules preoperatively as well as 3 and 6 h postoperatively) (= 600 mg of ginger extract). One verum capsule contained 100 mg of standardized extract of the rhizome of Zingiber officinale (drug extract ratio 10–20:1; extraction agent: acetone). The gelatin capsules were aromatized with mint flavor to make the ginger and the identically looking placebo capsule indistinguishable with regard to smell and taste. These efforts were made because ginger can be smelled and tasted even when covered with a cellulose-microcrystalline (6). All patients were orally premedicated with 20 mg of clorazepate. General anesthesia was induced with propofol, fentanyl (2 μg/kg), atracurium (0.3–0.5 mg/kg), and in some patients (n = 33), with shorter surgical procedures with succinylcholine (1–2 mg/kg). Desflurane (3–6 vol%) in nitrous oxide/oxygen (Fio2: 0.3) and additional fentanyl were used for maintenance. Fluid administration was standardized. Neuromuscular blockade was antagonized with pyridostigmine (0.1 mg/kg) and atropine (0.5 mg) when clinically indicated (n = 16). No additional antiemetics were applied. Postoperatively, pain was treated with repeated small doses of an opioid (piritramid 2–3 mg) until the patient had reached a level of no or mild pain. If a patient requested therapy for PONV or experienced nausea for at least 10 min or had ≥2 emetic episodes, rescue therapy for PONV was given (0.75–1 mg of droperidol, followed by 2 mg of tropisetron, if ineffective). Patients were visited 1, 3, 6, and 24 h postoperatively. Patients who were operated on a day-case basis were not discharged before the 6-h observation point and contacted 24 h postoperatively by telephone. Patients were asked to rate nausea and postoperative pain since the last visit using a four-point verbal rating scale (none, mild, moderate, severe) and to quantify the episodes of emetic symptoms (vomiting or retching). Twenty-four hours postoperatively, the Quality of Recovery score (7,8) was assessed during a standardized postoperative interview. Patients were also asked to grade their satisfaction using a six-step grading scale (1 = very good, 2 = good, …, 6 = unsatisfactory). A prospective power analysis assumed an incidence rate of PONV in the placebo group of 0.50 (9) and incidence rates of 0.4 and 0.3 in the G300 and G600 groups, respectively. Three hundred sixty patients provide a 90% chance of detecting a significant effect (α = 0.05) using the Cochran Armitage test. Because of the uncertainty with respect to the efficacy and dose responsiveness, an adaptive interim analysis (10) was prospectively planned after half of the foreseen number of patients (n = 180). This design allows stopping the study in case of a significant reduction of the PONV incidence for the verum groups (adapted P ≤ 0.0131) or if there is no realistic chance to detect a significant and relevant difference if the study is continued (P ≥ 0.30). For 0.0131 <P < 0.30, the study could be completed with the attempted number of patients. Additional descriptive analyses were performed for the incidences of nausea, vomiting, and need for antiemetic rescue treatment using 95% confidence intervals. Results A total of 202 consecutive female patients undergoing laparoscopy were screened. Eighteen patients were not suitable (body mass index >35, intellectual or linguistic deficits, previous intake of antiemetics, refusal to participate). The remaining 184 patients gave their informed consent but 4 women were not randomized (cancellation of surgery). Thus, 180 patients were finally randomized. Five of these were subsequently excluded for surgical reasons. No patient was excluded postoperatively or was lost to follow-up. The patients were comparable with respect to their demographic data, type and duration of surgery, the anesthesia technique, and dosage of drugs applied. A simplified risk score (11) successfully validated in our institution (12) was used to demonstrate an identical mean baseline risk for all 3 groups (47%–49%; see Table 1).Table 1: Demographic Data, Information on Anesthesia and Surgery, and Risk Factors for Postoperative Nausea and Vomiting (PONV) of the 175 Patients Included in the Intention-to-Treat AnalysisIn the interim analysis (n = 180 patients), the observed incidence of PONV was 49% (95% confidence interval: 36%–63%) in the placebo group, 58% (95% confidence interval: 44%–70%) in the G300 group, and 53% (39%–66%) in the G600 group (P = 0.69). Thus, the study was stopped according to the prospectively defined protocol. There was also no difference with regard to nausea, vomiting, and the need for antiemetic rescue treatment among the three groups at any of the four postoperative observation points (Table 2). Adverse events were infrequent for all treatment groups (placebo group = 11, G300 group = 8, G600 group = 13) and organ classes, respectively.Table 2: Incidences of Nausea, Vomiting, Need for Antiemetic Rescue Treatment and Postoperative Nausea and Vomiting (PONV)Discussion In a systematic review, Ernst and Pittler (5) concluded that ginger is a promising antiemetic herbal remedy, but the clinical data were insufficient to draw firm conclusions. Thus, they called for further rigorous studies to establish whether ginger is effective for clinical nausea and vomiting. In our placebo-controlled, double-blinded study, we failed to demonstrate that ginger possesses antiemetic activity. Gynecologic laparoscopies were associated with a 48% incidence of PONV in the placebo group, very similar to that reported previously from our institution in an identical setting (9). Premedication with 100 or 200 mg of the extract of the ginger root and subsequently administering repetitive doses of 100 or 200 mg, respectively, did not decrease this incidence. This initial preoperative dose (approximately 1–2 g of the crude drug) is comparable with the doses used in previous studies on ginger observing an antiemetic effect [1 g: Bone et al. (13) and Phillips et al. (14)] and also in “negative” studies [0.5–1 g: Arfeen et al. (6) and Visalyaputra et al. (15)]. Furthermore, the total dose as applied in the G600 group is larger than the maximal dose of ginger recommended according to German monographs on ginger. 1 The administration of the drug was repeated postoperatively, because it is known that some major pharmacologically active compounds of ginger have a short half-life (16). General postoperative recovery was measured by using the score published by Myles et al. (7). It is one of the very few postoperative questionnaires carefully developed according to psychological standards and validated in a large number of patients (8). In addition, patients were asked to rate satisfaction with anesthesia care using a simple six-step grading system. Both methods had a fairly good correlation (ρ = −0.3;P = 0.0004 with the Spearman rank correlation test) but none showed any significant difference among the three groups (see Table 2). We thank Dr. Willmar Schwabe Pharmaceuticals, Karlsruhe, Germany for providing a grant and the preparation of the study medication.