Study was halted prematurely for futility after an intermediate analysis

Postoperative nausea and vomiting (PONV) are frequent after surgery and anaesthesia. Dexamethasone is widely used as antiemetic for the prevention of PONV. Little is known about the efficacy of antiemetic drugs for the treatment of established PONV symptoms. No single randomised trial has been published so far that tests the efficacy of dexamethasone for the treatment of established PONV symptoms. In this trial the investigators want to test the antiemetic efficacy of three different doses of intravenous dexamethasone for the treatment of established PONV symptoms. In adjunct protocols of this study the investigators aim to establish a novel method to quantify the anti-nausea efficacy of an antiemetic drug, to study pharmacogenetics of PONV, and to further our understanding on the smoking status as a predictive factor of PONV.

This study is divided into a main study and three adjunct protocols:

Main study:

Postoperative nausea and vomiting (PONV) are frequent adverse effects of surgery and anesthesia. Dexamethasone is well established for the prophylaxis of PONV, but the efficacy of dexamethasone for the treatment of established PONV symptoms remains unknown. The primary objective of the main study is to test the antiemetic efficacy of dexamethasone for the treatment of established PONV in adults undergoing surgery under general anaesthesia, and to test for dose-responsiveness. The secondary objective is the evaluation of the potential adverse effect profile of dexamethasone, No prophylactic antiemetics are allowed. Premedication, conduct of anesthesia and postoperative analgesia will be at the discretion of the responsible anesthesiologist. Patients who have received their assigned study drug and who continue to experience PONV will receive antiemetic rescue treatment (ondansetron, droperidol). The minimum delay between administration of the study drug and rescue is 60 min (to ensure that dexamethasone has the scope to show antiemetic efficacy). Primary efficacy endpoint: Complete absence of any nausea and/or vomiting (including retching) in a previously nauseated or vomiting patient within 24 hours after administration of the study treatment. Secondary endpoints: Time to treatment failure; quality of sleep during the first postoperative night (numerical rating scale ranging from 0 = no sleep at all to 10 = excellent sleep); blood glucose (in the morning after administration of study drug); any minor or major adverse effects during 24h.

Adjunct protocol 1:

In therapeutic PONV trials, efficacy is usually quantified using a dichotomous outcome, i.e. a previously nauseous or vomiting patient stays totally PONV-free over a given period of time (for instance, 24 hours). This outcome is valid for the endpoint vomiting (including retching): a previously vomiting patient stays completely vomiting-free after treatment. For nausea, the situation is different. A patient may suffer from severe nausea, and, after treatment, the degree of nausea may significantly decrease although nausea may not disappear completely. Nevertheless, such a treatment may be regarded as efficacious. A very similar context can be found in the postoperative pain setting where an efficacious analgesic may decrease the degree of pain to a significant extent although the pain does not disappear completely. In analgesic trials, efficacy is therefore not expressed as a dichotomous outcome (pain - no pain) but as a continuum on a 0 to 100 mm visual analogue scale. Patients with moderate to severe postoperative pain (VAS ≥3/10) would be randomised to receive an experimental or a control (placebo) intervention and relief would be then recorded over time. Outcomes are expressed as maximum pain relief, area under the time-analgesic effect curve for pain intensity (i.e. summed pain intensity difference, or SPID), or summed pain relief (TOTPAR). AIM: To apply the methodology acute pain trials to the measurement of the anti-nausea efficacy of antiemetic drugs. To investigate the variability of responses based on measurement of nausea intensity using a visual analogue scale compared with a 4-point categorical scale. METHODS: Patients of the main study protocol will be asked to rate their degree of "baseline" nausea on a 100 mm visually analogue scale ranging from 0 mm = no nausea at all to 100 mm = worst possible nausea, and on a 4-point categorical scale (none, moderate, severe, intolerable). Subsequently, after having received the assigned study drug, patients will be asked to score their degree of nausea on VAS every 15 minutes until 60 minutes after administration of the study drug (i.e. until they would receive the rescue medication).

Adjunct protocol 2: Pharmacogenetics of PONV.

Knowledge about the mechanisms underlying inter-individual differences in PONV is needed to better identify patients who are at risk of PONV. Some of the variability seen with nausea and vomiting in cancer or migraine without aura may be associated with genetic variation either in genes related to the mechanisms of nausea/vomiting or in genes related to opioid action. In cancer patients receiving opioids, the inter-individual differences in nausea and vomiting were associated with genetic markers represented by single nucleotide polymorphisms of candidates in the opioid- or nausea/vomiting signalling pathways that modulate neurotransmission by metabolising the catecholamine dopamine. A possible role has been suggested for catechol-O-methyl-transferase (COMT) which is a key modulator in the metabolism of endogenous monoaminergic neurotransmitters, and hence of opioidergic neurotransmitters. The COMT enzyme possesses a frequent non-synonymous polymorphism that encodes for the substitution of valine (Val) by methionine (Met) at codon 158 (Val158Met). The SNP rs4680 (c.472G > A) is a missense mutation leading to a four-fold reduction of the COMT enzyme. Individuals with the Val/Val genotype have the highest activity of COMT, those with the Met/Met genotype have the lowest activity of COMT, and heterozygous individuals are intermediate.11,12 Pathways involving specific neurotransmitters and their receptors are likely to be involved in PONV, for instance, dopamine, acetylcholine, histamine, cannabinoids, substance P, and 5-hydroxytryptamine. The role of dopamine is pivotal in nausea and vomiting; dopamine D2 receptor blockade, in the area postrema, has an antiemetic effect. In patients receiving COMT inhibitors, the enhanced dopaminergic activity leads dose-dependently to frequent and severe nausea and vomiting. AIM: To compare COMT genotype of patients with PONV to PONV-free controls and to assess the possible relationship between genetic factors and dexamethasone response. A greater proportion of COMT-deficient allele is expected in PONV patients as compared to PONV-free controls. METHODS: The study population consists in all patients with PONV that are included in the main study (n=560). The control group will be 560 consenting patients but who do not develop PONV. COMT genotyping: Genomic DNA will be extracted from 200 µl of whole blood using the QIAamp DNA blood mini kit (QIAGEN, Hombrechtikon, Switzerland). The COMT Val158Met polymorphism (rs4680) will be genotyped using a commercially available TaqMan® SNP genotyping assay (C_25746809_50, Applied Biosystems, Warrington, UK). Twenty-five µl reactions will be performed using 50 ng of template DNA, pre-designed PCR primers and TaqMan® MGB probes (FAMTM and VIC® dye-labelled) according to the manufacturer's instructions in TaqMan® Universal PCR Master Mix using an iCycler iQ detection system (Bio-Rad, Hercules, CA). Analyses: The investigators will analyse allele and genotype frequencies, Hardyweinberg equilibrium, linkage disequilibrium and haplotypes. The individual SNPs and haplotypes association with of occurrence of PONV will be categorized and will be analyzed by Fisher exact test or Chi-square test and regression analysis (logistic, univariate and multivariate). Gene-gene interactions relation with that of PONV will be assessed by multivariate regression analysis. Analyses will be performed using Haploview software v1. (Broad Institute, Cambridge, USA) and SPSS (version 19; SPSS Inc, Chicag0, IL).

Adjunct protocol 3: Role of non-smoking status as a predictive factor of PONV.

BACKGROUND: Non-smoking status has consistently been reported to be an independent predictive risk factor of PONV. At first view, this empirical observation is interesting. However, there are two unresolved issues. Firstly, it remains unclear, why non-smoking patients are at a increased risk of PONV, and consequently, smoking patients are "protected". For instance, the investigators failed to show any benefit of (short-term) perioperative transcutaneous nicotine administration for the prevention of PONV. Secondly, no PONV risk score has shown satisfactory sensitivity and specificity, largely limiting their clinical applicability. It has been argued that one weakness of these scores was that non-smoking status was ill defined and was treated in a purely dichotomous fashion. For instance, it remains unclear from what length of abstinence the formerly PONV-protective effect will be lost in an ex-smokers. Also, it remains unknown whether "heavy" smokers" are more protected. Finally, there are various ways to consume tobacco; tobacco snuffing was also shown to decrease the risk of PONV. AIMS: 1. To test whether the risk of PONV differs between non-, ex- and current smokers. 2. To test whether there is a dose-effect relationship between the amount of tobacco smoking, and the risk of PONV. 3. To test in ex-smokers for a relationship between the duration of smoking cessation and the risk of PONV. METHODS: Patients who participate in the main study will be asked, pre-operatively, to fill in a questionnaire regarding smoking habits. Patients will be classified into non-smokers (exposed or not to passive smoking), ex-smokers (stop >1 month; exposed or not to passive smoking; substituted), and smokers. Variables related to PONV and to smoking (potential confounders) will be recorded.

Inclusion Criteria:

• Adults (age ≥18 years), male or female.
• American Society of Anesthesiology (ASA) status I to III.
• Able to read and understand the information sheet
• Subjects who have signed and dated the consent form.
• Scheduled for elective surgery.
• If the patient is female and of childbearing potential, she must have a negative pregnancy test (serum hCG or urine dipstick).

Exclusion criteria:

• A history of allergy or hypersensitivity to dexamethasone or any component of its formulation.
• Hepatic dysfunction* (i.e bilirubin <1.5 upper limit normal (ULN), alanine aminotransferase (ALT) <2.5 x ULN, aspartate aminotransferase (AST) <2.5 x ULN).
• Renal insufficiency* (i.e. creatinine <1.5 x ULN, creatinine clearance <30ml min-1).
• Pregnant, or intending to become pregnant, women.
• Breastfeeding women.
• Patient having used any investigational drug within 30 days of screening.
• Patient having participated in any clinical trial within 30 days.
• Patients with active GI ulcer.
• Patients needing prolonged postoperative intubation.
• Patients needing a gastric tube postoperatively.
• Patients receiving antiemetic drugs (butyrophenones, 5-HT3 receptor antagonists, dexamethasone).
• Patients taking drugs that interfere with platelet aggregation (for instance, aspirine or clopidogrel) within seven days preoperatively.
• Patients with overt psychosis or taking antipsychotic treatment (for instance, anti-dopaminergic drugs).
• Patients taking drugs with known emetogenic potency (for instance, L-Dopa, COMT inhibitors).
• Specific types of surgery: tonsillectomy (increased risk of postoperative bleeding), interventions that require strict prevention of postoperative vomiting.
• Systemic infections (bacterial, fungal, malaria, viral, tuberculosis).
• Local infections (for instance, ocular herpes simplex).