Department of Anesthesiology and Critical Care
School of Medicine
University of Navarra
Despite significant advances and the
introduction of new antiemetic agents, the incidence of postoperative nausea and vomiting (PONV) in high-risk patients is as high as 70%.(1) The overall incidence of PONV during the first 24h after surgery is currently estimated to be 20–30%,(2) with little improvement in recent years. PONV can cause prolonged postanaesthesia care unit stay and unanticipated admissions following ambulatory surgery, which increases medical costs.(3) PONV also affects the degree of patient satisfaction, comfort and quality of life.
Numerous pathophysiological mechanisms are known to cause nausea or vomiting, but their role in PONV is unclear. However, as far as can be ascertained, all the pathways capable of causing emesis converge in the brainstem, with the nucleus of the tractus solitarius (NTS) being the focus of attention. Several stimuli activate a group of cells of the area postrema (AP), termed the chemoreceptor trigger zone for emesis, which activates the vomiting centre, located in the NTS. Other pathways act through direct stimulation of the vomiting centre.
The aetiology of PONV is multifactorial, including the particular patient’s characteristics, factors related to the operation and anaesthesia, and the intensity of pain and its management in the postoperative period. For daily clinical purposes, simplified risk scores(1,4) are easy to handle and show a good correlation between the expected and actual incidence of PONV.(5) The main benefit of risk scores is that prophylaxis may be targeted at patients who actually need it. Recently, an international expert panel tried to frame evidence-based strategies for a rational approach to the prevention and treatment of PONV.(6)
The main patient and surgical factors associated with PONV and that cannot be modified are: female sex; history of being a nonsmoker; history of PONV or motion sickness; and postoperative opioid use and certain types of surgery (such as laparoscopy, laparotomy, plastic surgery, major breast surgery, craniotomy, otolaryngological procedures and strabismus surgery).
Once the high-risk patients have been identified, the problem is the choice of an antiemetic. Antiemetic drugs (antidopaminergic, anticholinergic, anti-histaminic and antiserotonergic) are based upon the receptor types believed to be involved in PONV. Although a lot of drugs have been used in this condition, only a few are now routinely used by anaesthesiologists. Metoclopramide, although used for many years, is no more effective than placebo in the prevention of PONV.(7,8) Droperidol has been extensively used in anaesthesia. In 2000, a meta-analysis showed that droperidol had antinausea and antiemetic effects, although a dose of 1.25mg was needed.(9) Dexamethasone has been used for prophylaxis in PONV for more than 10 years, although the antiemetic action of steroids is unknown. A few years ago, a review of the effectiveness of dexamethasone concluded that it is more efficacious than placebo in adults and in children.(10)
Selective 5-HT(3) receptor antagonists act as highly selective and potent antagonists of 5-HT(3) receptors in the brain (AP, NTS), and also act peripherally in the gastrointestinal tract. Ondansetron was the first 5-HT(3) antagonist available. In 1997, Tramer et al analysed the efficacy, dose–response and safety of ondansetron in the prevention of PONV in all randomised placebo-controlled trials then available,(11) and concluded that if the risk of PONV is very high, 20% of patients will not vomit if they receive an adequate dose of ondansetron. The antinausea effect was less pronounced. Adverse effects include elevated liver enzymes (3%) and headaches (3%). Tramer et al also published a quantitative systematic review of ondansetron in the treatment of established PONV,(12) concluding that further nausea or vomiting could be prevented with ondansetron in 25% of patients who had PONV (number-needed-to-treat: 4).
These authors did not find evidence of a clinically relevant dose–response between 1mg and 8mg. The efficacy and safety of ondansetron, droperidol and dexamethasone are similar.
Recent research involving antiemetics has focused on the development of 5-HT(3) antagonists, such as tropisetron, granisetron, dolasetron and, more recently, ramosetron. Although these new 5-HT(3) antagonists were demonstrated to be more efficacious than placebo in the prophylaxis of PONV, results were similar to those obtained with ondansetron. In addition, these compounds are more expensive than ondansetron.Despite being superior to that of placebo, the efficacy of currently available antiemetics is often poor. There can be many causes for PONV, and antagonising only one type of receptor is not sufficient in many patients. Thus, drug combination could be a solution to control severe, frequent PONV. Drugs with a different mechanism of action should be given, and there is evidence that combinations of antiemetics can act synergistically.
The combination of droperidol with ondansetron(13) and granisetron(14) has been shown to be more effective in preventing PONV in patients at high risk of suffering PONV than each of these drugs alone. The synergistic antiemetic action of dexamethasone and ondansetron(15) or granisetron(16) has been demonstrated in similar patient groups. Eberhart and colleagues carried out a meta-analysis of 26 randomised, controlled studies in 2,561 patients, and determined that dexamethasone increased the efficacy of the partner antiemetic.(17)
Numerous studies have determined the effectiveness of 5-HT(3) receptor antagonists when used in combination with other antiemetics. Recently, a meta-analysis of randomised controlled trials showed that the efficacy of 5-HT(3) receptor antagonists combined with droperidol for PONV prophylaxis was similar to their combination with dexamethasone.(19)
The combination of droperidol and dexa-methasone has led to controversial results. In a small study carried out in women undergoing major gynaecological surgery with intravenous (IV) patient-controlled analgesia (PCA) with morphine in the postoperative period (with a baseline risk >80%), this combination was found to be worse than ondansetron combined with either droperidol or dexamethasone.(19) However, in a recent study carried out in patients with a risk of PONV >40%, Apfel and colleagues found that the effectiveness of the three combinations was similar.(20)
Combinations are more effective than a single antiemetic, but patients at very high risk may still suffer from PONV. Thus, potent new antiemetics are required. Neurokine-1 (NK(1)) antagonists are being used with good results in highly emetogenic chemotherapy. These drugs could provide a major advance in the management of PONV.(21,22)
High concentrations of inspired oxygen in the intraoperative and immediate postoperative periods are effective in abdominal surgery,(23,24) but not in surgery of the thyroid gland. Opioid antagonists such as naloxone(25) and nalmefene(26) are effective in the prophylaxis of PONV at low doses.
Until new drugs are available, we must act on the factors related to anaesthesia that may modify PONV, as well as on postoperative pain and its management. Several authors recommend a multimodal anti-PONV approach in patients at high risk of suffering PONV. Habib and colleagues demonstrated that total IV anaesthesia (TIVA) with propofol was as effective as droperidol plus ondansetron, and that TIVA with propofol combined with droperidol plus ondansetron was more effective than each antiemetic intervention alone.(27) Apfel et al carried out a study on 5,199 patients and evaluated six prophylactic interventions. (20) These authors showed that ondansetron, dexamethasone and droperidol reduced the risk of PONV by 26%. Propofol reduced the risk by 19%, and nitrogen by 12%; the risk reduction with these two agents (TIVA) was thus similar to that observed with each of the antiemetics.
Many interventions can modify the incidence of PONV, (28–35) and these effects should be analysed before a decision is made.
Despite the introduction of new antiemetic agents, short-acting anaesthetics and minimally invasive surgical techniques, the incidence of PONV remains unchanged. Prophylactic use of antiemetics has become the standard approach to minimising emetic symptoms after surgery in patients at moderate risk of suffering PONV. A combination of antiemetics with different mechanisms of action has additive effects on the incidence of PONV and is of interest in patients at high risk of PONV.
Other actions, such as the use of TIVA with propofol, are effective in lowering the incidence of PONV and must be used in patients with very high risk of suffering PONV.
However, the efficacy of the different antiemetic strategies depends on each patient’s overall risk of PONV. The safest and less expensive antiemetic intervention should be used first.
Finally, it should be noted that “rescue” treatments are ineffective when the same drug has already been used prophylactically.
- Anesthesiology 1999;91:693-700.
- Anesth Analg 1994;78:7-16.
- JAMA 1989;262:3008-10.
- Anaesthesia 1997;52:443-9.
- Br J Anaesth 2002;88:234-40.
- Anesth Analg 2003;97:62-71.
- Br J Anaesth 1992;69 Suppl 1:46S-59S.
- Br J Anaesth 1999;83:761-71.
- Can J Anaesth 2000;47:486-93.
- Anesth Analg 2000;90:186-94.
- Anesthesiology 1997;87:1277-89.
- BMJ 1997;314:1088-92.
- Anesth Analg 1996;83:117-22.
- Can J Anaesth 1998;45:541-4.
- Br J Anaesth 1996;76:835-40.
- Anesth Analg 1997;85:913-7.
- Anaesthetist 2000;49:713-20.
- Can J Anaesth 2004;51:311-9.
- Anesth Analg 2002;95:1590-5.
- N Engl J Med 2004;350:2441-51.
- Br J Anaesth 1999;82:274-6.
- Anesthesiology 2000;93:931-7.
- Anesth Analg 2001;92:112-7.
- Anesthesiology 1999;91:1246-52.
- Anesthesiology 1997;87:1075-81.
- Anesthesiology 1999;90:1007-11.
- Anesth Analg 2004;99:77-81.
- Br J Anaesth 1984; 56:19-27.
- Br J Anaesth 1999;82:379-86.
- Physiology of spinal anesthesia. Baltimore: Williams & Wilkins; 1981. p. 258-60.
- Anesthesiology 1992;77:162-84.
- Can Anaesth Soc J 1984;31:178-87.
- Can J Anaesth 1995;42:201-3.
- Anesthesiology 1992;77:915-9.
- Anesth Analg 1995;80:682-6.