Proton pump inhibitors (PPI) have been shown to improve symptoms in patients with functional dyspepsia . Since the introduction of PPIs for various indications, we have made several observations. First, cancer patients with nausea and vomiting felt better when treated with PPIs. Second, critically ill patients who received PPIs for prevention of stress related ulcer disease reported less general gastrointestinal symptoms. And finally, some patients with refractory postoperative nausea and vomiting (PONV) had a better outcome when treated with a PPI. There are, however, no prospective studies testing the effect of PPIs on nausea and vomiting.
On average 30% of all patients who undergo surgery still have PONV . Predictive factors for PONV have been described; these include female gender, no smoking, postoperative use of opioids, and a history of previous PONV or of motion sickness [3–5]. Ondansetron, dexamethasone and droperidol reduce the incidence of PONV by about 25%, propofol by 20% and an anaesthetic without nitrous oxide by 12% (both of these agents 26%) . It has been suggested that patients at high risk of PONV should receive multimodal prophylaxis to reduce the incidence of PONV . So far there is no single antiemetic drug available that could prevent PONV by more than 30%.
We performed a randomized, double-blind, placebo controlled trial to test the hypothesis that PPIs might be useful in the prevention of PONV in patients at high risk of PONV undergoing elective surgery.
The study protocol was approved by the local Ethics Committee (Ärztekammer Niedersachsen, Hannover, Germany). One hundred consecutive patients with high risk for PONV (Apfel-score III and IV) , undergoing elective surgery under general anaesthesia were enrolled. Inclusion criteria were age 18–80 yr, ASA I–III, no gastrointestinal disease, and no gastrointestinal surgery except for simple cholecystectomy. Exclusion criteria were treatment with H2 receptor inhibitor, PPI, or antiemetic drugs, vomiting before anaesthesia, and severe postoperative complication (for instance, myocardial infarction).
The experimental group received oral 3 × 40 mg esomeprazole (n = 45), the control group matching placebo tablets (n = 48) at 6.00p.m. the day before surgery, 2 h preoperatively and 24 h after the second dose. Active and placebo tablets were identical in colour and shape. Total intravenous (i.v.) anaesthesia with propofol and remifentanil without nitrous oxide (FiO2 0.5) was used throughout. Postoperatively, patients were interviewed by nurses and anaesthesiologist using a standardized questionnaire  at discharge from the post-anaesthesia care unit (PACU) and 6 h and 24 h after admission to the PACU. The severity of nausea was scored on a 0–100 numerical scale ranging from 0 = no nausea to 100 = nausea as bad as it can possibly be. χ2-test and U-test were used for statistical analysis.
Two patients from the placebo group and five from the esomeprazole group were excluded from data analysis; five patients were lost to follow up, in one there was violation of the placebo code and one had chest pain postoperatively. Patient and anaesthesia characteristics are shown in Tables 1 and 2. Except for one man, all study participants were women. There were no significant differences between the study groups.
There was no difference between the two groups in the number of patients who had no PONV symptoms at all during the 24 h observation period (Table 3). There was no difference either in the number of patients who had only nausea symptoms or who vomited without having nausea. Nausea scores at discharge from the PACU, and at 6 h and 24 h after admission to the PACU were similar between groups. Only about 25% of the patients in both groups did not experience any nausea or vomiting (Table 2). Rescue medication (primarily with ondansetron, and in case of persisting PONV with dexamethasone) was administered to 46.7% of patients treated with esomeprazole and to 52.1% of control patients (P > 0.05).
PPIs have been used preoperatively for reduction of gastric acidity and volume and to decrease the risk of aspiration . These substances were also found to reduce symptoms of functional dyspepsia in a subset of patients . Such reports and in addition clinical observations of reduced gastrointestinal symptoms in critically ill patients who were treated with PPIs, led to the prophylactic use of PPIs in patients undergoing elective surgery in some institutions. The effect of PPIs on PONV however has not been evaluated in prospective studies. We designed this randomized, double-blind, placebo controlled study to test the hypothesis that the PPI esomeprazole may have a beneficial effect in patients at high risk of PONV.
Esomeprazole did not reduce the incidence of PONV. This suggests that our initial clinical observations were most likely biased. One limitation of our study is the predominance of women (98.9%) due to the higher risk for PONV of females (female gender being one of the defined risk factors). In addition, 43% of all surgeries in this study were gynaecological interventions. Thus, extrapolation of these results for male patients should be performed with some caution, although there is no evidence for a gender specific response to PONV prophylaxis.
It may be considered unethical to leave patients with high risk for PONV without any prophylaxis. This issue was discussed with the local Ethics Committee. Considering the limited efficacy of single drug PONV prevention and taking into account our preset expectation that esomeprazole may show some efficacy in the prevention of PONV, we decided to design this study with placebo-controls. Due to the ethical concerns, all patients received total i.v. anaesthesia that on its own may reduce the incidence of PONV . In addition, the number of patients was limited to 100. Despite all limitations associated with this study, the presence of a placebo group ensures a proper interpretation of the results.
The observed high frequency of PONV in high-risk patients calls for further research to diminish this burden. A prophylactic multimodal approach including antiemetic therapy with ondansetron, dehydrobenzperidol and dexamethasone in combination with total i.v. anaesthesia  may be useful.
We express our gratitude to all participating physicians and Mr Heinz Geerlings for help with statistical analysis. Funding: AstraZeneca, Wedel, Germany provided the placebo tablets.
No competing interests to declare.
Chritian Weilbach designed and analysed the study, wrote the manuscript and is fully responsible for the scientific content. Katharina Kähler, Ullrich Thissen and Niels Rahe-Meyer were responsible for patient monitoring and data management. Sigfried Piepenbrock was involved in study design and monitoring.
1. Bolling-Sternevald E, Lauritsen K, Aalykke C et al
. Effect of profound acid suppression in functional dyspepsia: a double-blind, randomized, placebo-controlled trial. Scand J Gastroenterol
2. Apfel CC, Korttila K, Abdalla M et al
. A factorial trial of six interventions for the prevention of postoperative nausea and vomiting
. N Engl J Med
3. Sinclair DR, Chung F, Mezei G. Can postoperative nausea and vomiting
be predicted? Anesthesiology
4. Apfel CC, Roewer N. Risk assessment of postoperative nausea and vomiting
. Int Anesthesiol Clin
5. Apfel CC, Kranke P, Eberhart LH, Roos A, Roewer N. Comparison of predictive models for postoperative nausea and vomiting
. Br J Anaesth
6. Apfel CC, Roewer N, Korttila K. How to study postoperative nausea and vomiting
. Acta Anaesthesiol Scand
7. Levack ID, Bowie RA, Braid DP et al
. Comparison of the effect of two dose schedules of oral omeprazole with oral ranitidine on gastric aspirate pH and volume in patients undergoing elective surgery. Br J Anaesth