In the ambulatory surgery setting, the incidence of post-operative nausea and vomiting (PONV) ranges from 30 percent to 50 percent, making it one of the most common complications. Besides prolonging a patient's recovery time, PONV that occurs post-discharge forces the patient to treat the PONV without medical supervision.

Much has been written about the high cost of the newer antiemetic agents. At my own institution, the cost of a 4mg vial of ondansetron (Zofran) is $16 more than a 5mg vial of droperidol (Inapsine). However, it's important to look at the big picture. Non-drug costs (such as unanticipated admissions) associated with PONV can be several times higher than even the most expensive antiemetic agent. The meds are easily cost-justified. With that in mind, here's what you need to know about selecting them.

Understanding PONV
Nausea and vomiting are separate entities. Nausea isn't always followed by vomiting and can be as distressing to patients, or even more so. This concept is important because certain antiemetics are more effective for preventing nausea while others are more effective for preventing vomiting.1

There are several receptors, mainly found in the brain, that result in nausea and/or vomiting when activated. These are dopamine type 2 (D2), muscarinic cholinergic type 1 (M1), histamine type 1 (H1), and serotonin type 3 (5HT3). Most antiemetics produce their effects by blocking one or more of these receptors. Why is this important? If you're going to use multiple antiemetics to try to prevent PONV, you want to make sure they work at different receptors to get maximum effect. Further, if the patient receives a prophylactic antiemetic agent but still experiences PONV, you'd use an agent with a different mechanism of action (meaning it acts at a different receptor) to treat it.

Assessing risk factors
PONV is multifactorial. Risk factors can be broadly classified into four general categories: patient-specific, surgical, anesthetic and post-operative. The questions that need to be asked are whether they're all equally important and if not, which place the patient at greatest risk. Taking this one step further, how many risk factors need to be present to be predictive of PONV?

Simplified risk scores, in particular those suggested by Apfel and Koivuranta, effectively predict the likelihood of a patient developing PONV (see "PONV Predictors"). Other factors considered include use of perioperative opioids, duration of anesthesia (there's greater risk with longer duration), middle ear surgery, laparoscopic cholecystectomy and GYN laparoscopy.

It would be a waste of resources to give all patients a prophylactic antiemetic agent, and it would expose them to potential adverse effects. The goal is to have a systematic process to identify patients who would benefit from a prophylactic antiemetic agent and assist in appropriate agent selection (see "Creating a Regimen" on page 17.)

Theory to therapy
So how is a patient's PONV risk tied to antiemetic therapy? Using Apfel's risk model, for example, no prophylactic antiemetic agent would be administered to mild-risk patients. Moderate- to high-risk patients would receive one- or two-agent prophylaxis with an agent with a different mechanism of action if nausea and/or vomiting occurs post-operatively. Very high-risk patients would receive three-agent prophylaxis (see "Antiemetic Therapy" on page 16).

Once an anesthesia provider assesses at least moderate PONV risk, it's important to incorporate appropriate prophylaxis into the patient's anesthetic plan. The patient should receive one or more agents to block one or more PONV-causing receptors.

A consensus panel recently recommended that droperidol, dexamethasone and/or a serotonin antagonist used alone or in combination, depending on PONV risk, be considered first-line agents for prevention of PONV.1 An important point to remember is that administering a prophylactic antiemetic agent only reduces the incidence of PONV by approximately 20 percent. Or put another way, the agent needs to be administered to five patients with at least a moderate risk for PONV to prevent one patient from developing PONV who would have otherwise experienced it.

In addition to antiemetics, there are other prophylaxis strategies anesthesia providers may consider. Making sure the patient is adequately hydrated has been demonstrated to help reduce the likelihood of post-op nausea. Anesthesia providers have used high inspired-oxygen concentration (80 percent) to try to reduce PONV, with mixed results reported. To minimize pain and opioid use, both of which can cause PONV, regional anesthesia may be used if feasible. If regional anesthesia isn't possible, total intravenous anesthesia using propofol may be indicated, depending on a patient's risk for PONV. Propofol has a lower incidence of PONV compared to volatile inhalational agents. However, when propofol is used to just induce general anesthesia, its contribution to reducing the likelihood that a patient will develop nausea and vomiting postoperatively is relatively short-lived (providing benefit for roughly 30 minutes after induction). Finally, the anesthesia provider may decide to use ketorolac (Toradol) or local anesthetics, if appropriate, for pain management to reduce the need for post-op opioids.

Common PONV agents
Here's an introduction to six common PONV drug therapies:

• Dexamethasone (Deca-dron). Dexamethasone has comparable efficacy to serotonin antagonists and droperidol. The most common dose used for PONV prevention is 8mg, although an optimal dose has not been determined. It should be administered immediately before anesthesia induction. There appear to be no adverse effects from a one-time PONV prophylaxis dose. In addition, dexamethasone effectively prevents opioid-induced nausea and vomiting and may provide additional benefits (reducing post-op fatigue, pain, opioid requirements) compared to the other antiemetic agents. Because of the time it takes to produce its effects, however, the drug doesn't play a major role in PONV treatment.
• Serotonin antagonists. Three serotonin antagonists are approved for PONV prophylaxis and treatment: ondansetron (Zofran), dolasetron (Anzemet) and granisetron (Kytril). These agents, which come in both injectable and oral forms, are better against preventing vomiting than nausea. There appears to be no difference in the efficacy and safety profiles of the various serotonin antagonists for PONV prevention. Therefore, agent selection may be based on cost and other contract terms. The drugs should be administered at the end of surgery for optimal efficacy. The most common adverse effects of these agents are headache, constipation, and liver enzyme elevation. An 8mg ondansetron dose provides no additional benefit over a 4mg dose, while granisetron's FDA approved dose of 1mg appears to be greater than necessary; an adult dose of about 0.3mg was found to be effective in preventing PONV. Since granisetron comes in a 1mg vial, issues of waste and/or pharmacy technician (or other healthcare provider) time in preparing appropriate doses need to be considered.
• Transdermal scopolamine (Transderm-Scop). Scopola-mine is the most effective agent for preventing and treating motion sickness. Because of this, patients with a history of motion sickness may obtain the greatest PONV prevention benefit when choosing this option. The drug can also prevent and treat opioid-induced nausea and vomiting. The patch has a four-hour action onset, so place it far enough in advance to ensure it has time to start working. Failure to correctly apply the patch can also limit effectiveness. Visual disturbances and dry mouth are the most common adverse effects.
• Prochlorperazine (Compazine). Prochlorper-azine has been shown to be as or more effective than ondansetron in several studies. Adverse effects can include extrapyramidal symptoms such as restlessness and muscle spasm (dystonia).
• Droperidol (Inapsine). Droperidol has a greater anti-nausea effect than anti-vomiting. A 1mg to 1.25mg dose provides optimal anti-vomiting efficacy. Adverse effects can include sedation, hypotension, dizziness, restlessness and severe muscle spasm (dystonia). At the doses used for PONV prevention and treatment, its adverse-effect profile is similar to placebo. In adults, droperidol is as effective as ondansetron for preventing PONV. It's most effective when given at the end of surgery.

For many years, droperidol was the foundation for any PONV protocol. It's still widely considered the gold standard for efficacy comparisons. Droperidol was also very inexpensive. In December 2001, the FDA issued a "black box" warning for droperidol in response to patient deaths associated with cardiac rhythm abnormalities. This is the most severe warning the FDA issues. The warning states that patients administered the drug should have a 12-lead ECG before and for two hours to three hours after droperidol treatment to monitor for arrhythmias.

While droperidol use isn't prohibited, the black box warning makes it difficult to use as a first-line antiemetic agent. It has gone from a low-cost, first-line treatment to a high-cost (when monitoring is used), second- or third-line treatment.

• Metoclopramide (Reglan). As an antiemetic agent, metoclopramide gets mixed reviews, especially with the commonly used 10mg dose. It needs to be administered near the end of surgery for maximum results. The agent may be useful in patients where gastric stasis is an issue. Common adverse effects include sedation and dizziness.

The bottom line
As long as your facility uses effective tools for assessing PONV risk and developing clinical prophylaxis and treatment pathways, antiemetics are money well spent. It's preferable to treating complications, delaying discharge or reducing patient satisfaction.

References
1. Gan TJ, Meyer T, Apfel CC et al. Consensus guidelines for managing postoperative nausea and vomiting. Anesth Analg 2003; 97:62-71.