Martin R Tramèr
Division of Anaesthesiology
Geneva University Hospitals
The importance of postoperative nausea and vomiting (PONV) is generally underestimated because the condition is self-limiting, never becomes chronic and is almost never fatal. However, the impact of PONV on healthcare costs is not negligible. Ten percent of the population undergo a general anaesthetic every year,(1) approximately 30% of whom suffer from PONV.(2) About 1% of patients undergoing ambulatory surgery are admitted overnight because of uncontrolled PONV.(2)
When PONV occurs, the anaesthetist is usually blamed, despite evidence that PONV results from factors that are related not only to anaesthesia but also to surgery and to the patients themselves. Surgical patients would prefer to suffer pain rather than have to be afflicted by PONV,(3) and would be willing to pay considerable amounts of money for an effective antiemetic drug.(4)
A major obstacle to the development of an effective treatment is the lack of a valid animal model for PONV. The extrapolation of data from animal studies on highly emetogenic chemotherapy to PONV is of limited value. Thus, anaesthetists have to rely on the results of clinical trials. Data on an almost infinite number of potentially useful antiemetic interventions have been published during the last 40 years. Despite this large body of literature, fundamental data on dose-responsiveness or side-effect profiles have remained unclear for most antiemetics, and there is no consensus on the gold standard for treatment. Many trials include a small number of patients, with some being of doubtful validity. As a consequence, anaesthetists have been using antiemetics inconsistently.
The good news is that significant progress has been achieved in recent years in the field of PONV control. The literature published on PONV has been systematically reviewed, critically appraised and quantitatively synthesised(2) and, at present, the relative efficacy and harm of most antiemetic interventions is fully understood. Droperidol, a butyrophenone derivative that was withdrawn in some countries for apparent safety reasons,(5) has a pronounced antinausea effect at doses that are so incredibly low that the occurrence of any relevant adverse effect becomes highly unlikely.(6) Ondansetron, which was thought to represent the first universally effective antiemetic for PONV, was shown to have a limited effect on nausea.(7) Metoclopramide, one of the most popular anti-emetics for decades, showed no worthwhile efficacy.(8) Perhaps the most important information from these systematic reviews is that none of the drugs tested can be regarded as a gold standard, and none is efficient enough to be used on its own. At best, these drugs achieve a number-needed-to- treat to prevent PONV of about 5, compared with placebo, in high-risk patients.(2) Thus, to prevent PONV in one patient, five patients would need to receive the antiemetic prophylactically. However, systematic reviews also confirmed improved efficacy with “balanced antiemesis”, as, for example, with the combination of a 5-HT(3) receptor antagonist with droperidol or with dexamethasone, respectively.(9)
Improvements in anaesthetic techniques
As a further important step towards improved control of PONV symptoms, anaesthetists have become increasingly aware of the emetogenic potential of some anaesthetic techniques. For a patient who must not vomit (he/she may have wired jaws after maxillofacial surgery), the anaesthetist may choose the intravenous anaesthetic propofol (which is less emetogenic than volatile anaesthetics) and may avoid drugs that are known to increase the risk of PONV, such as nitrous oxide, physostigmine and opioids.(10) Some of these modalities may be of limited or short-lived efficacy; in addition, they are not systematically feasible in all circumstances. However, a combination of such simple measures clearly decreases the risk of PONV.(11)
Modern anaesthesia benefits from this new knowledge; indeed, a multimodal PONV approach has recently been recommended.(12) Thus, for optimal PONV prophylaxis in high-risk patients, anaesthetists can use a low-emetogenic anaesthesia technique and administer an antiemetic cocktail before emergence. The cocktail can consist of a 5-HT(3)-receptor antagonist plus a dopamine antagonist (droperidol) plus dexamethasone.
Treatment versus prophylaxis
Unfortunately, compared with the large amount of data published on PONV prophylaxis, relatively little is known on the treatment of established PONV symptoms. Interestingly, for patients with PONV symptoms, rescue treatment with the expensive 5-HT(3)-receptor antagonists is efficacious at much lower doses than those required for successful prophylaxis.(13) For example, in the case of ondansetron, the recommended prophylactic dose is 4mg, whereas 1mg is as efficacious as 4mg or even 8mg for the treatment of established symptoms. Although the mechanisms behind this improved therapeutic efficacy with low-dose 5-HT(3)-receptor antagonists remain unknown, it is an argument in favour of therapy rather than regular prophylaxis of PONV symptoms. In the case of antiemetics that do not belong to this drug class, there is a lack of evidence regarding therapeutic efficacy. Published and valid randomised trials are rare, small or, for some drugs, completely lacking.
Opioid-induced nausea and vomiting
A further important clinical issue relates to opioid-induced nausea and vomiting. The typical target population for this condition are patients undergoing major surgery and receiving morphine with a patient-controlled analgesia (PCA) device postoperatively. These patients are at a high risk of nausea, and some want the opioid analgesia to be stopped.
Droperidol has shown consistent antiemetic efficacy in randomised controlled trials.(14) A randomised dose-finding study suggested that 2.5mg of droperidol added to 100mg of morphine provided satisfactory protection against morphine-induced nausea without increasing the risk of sedation or other droperidol-related adverse effects.(15) The stability of a droperidol–morphine solution has been tested for a period of two weeks.(16)
Despite considerable progress in PONV pharma-cology in recent years, research is ongoing for an optimal control of PONV symptoms. Some of the drugs that have been used for decades, often without a strong evidence base, are finally undergoing critical appraisal,(17) and new compounds that block different receptor systems of the emesis pathways in animal models are undergoing clinical studies.(18)
- Clergue F, Auroy Y, Pequignot F, et al. Anesthesiology 1999;91:1509-20.
- Tramèr MR. Acta Anaesthesiol Scand 2001;45:4-13.
- VanWijk MG, Smalhout B. Anaesthesia 1990;45:679-82.
- Gan T, Sloan F, Dear G, et al. Anesth Analg 2001;92:393-400.
- Gan TJ, White PF, Scuderi PE, et al. Anesthesiology 2002;97:287.
- Henzi I, Sonderegger J, Tramèr MR. Can J Anesth 2000;47:537-51.
- Tramèr MR, Reynolds DJ, Moore RA, McQuay HJ. Anesthesiology 1997;87:1277-89.
- Henzi I, Walder B, Tramèr MR. Br J Anaesth 1999;83:761-71.
- Habib AS, El-Moalem HE, Gan TJ. Can J Anaesth 2004;51:311-9.
- Tramèr MR. Acta Anaesthesiol Scand 2001;14-9.
- Habib AS, White WD, Eubanks S, et al. Anesth Analg 2004;99:77-81.
- Gan TJ, Meyer T, Apfel C, et al. Anesth Analg 2003;97:62-71.
- Kazemi-Kjellberg F, Henzi I, Tramèr MR. BMC Anesthesiol 2001;1:2.
- Tramèr MR, Walder B. Anesth Analg 1999;88:1354-61.
- Culebras X, Corpataux JB, Gaggero G, Tramèr MR. Anesth Analg 2003;97:816-21.
- Williams OA, Middleton M, Henderson P, Reilly CS. Hosp Pharm Pract 1992;597-9.
- Büttner M, Walder B, Von Elm E, Tramèr MR. Anesthesiology In press.
- Gardner C, Perren M. Neuropharmacology 1998;37:1643-4.