There are a number of risk factors which predispose a patient to PONV. Identified risks include female gender, nonsmoking status (1.5–2.5 times risk), age (5 % incidence in infants, 25 % in children <5 years old, 40–50 % in 5–15 year olds, and 20–40 % in adults), history of motion sickness, history of PONV, obesity, anxiety, history of migraines, duration of surgery, use of narcotics during surgery and/or postoperatively, intraoperative use of volatile anesthetics and/or nitrous oxide, decreased administration of fluids during surgery, hypotension, and use of anticholinesterases (>2.5 mg neostigmine). Surgeries that predispose patients to PONV include laparoscopic procedures (especially gynecological), intra-abdominal surgeries, head/neck surgeries, strabismus repair, neurosurgeries, shoulder procedures, dental surgeries, and varicose vein stripping.

Recently, models have been developed that incorporate these risk factors to predict the likelihood of a patient experiencing PONV. Among such models, the Apfel score is the best known and most widely used. This model assigns one point for each of the following patient characteristics: female gender, nonsmoker, history of motion sickness or PONV, and plan to use postoperative opioids. When 0, 1, 2, 3, or 4 of the risk factors are present for a given patient, his/her risk of experiencing PONV is approximately 10 %, 20 %, 40 %, 60 %, or 80 %, respectively. The Koivuranta score and the Eberhart score (specifically for pediatric patients) have also been developed as alternative risk models. All three models have gained popularity because they are very simple and easy to apply—they all use only 4–5 risk factors to calculate a prediction score, with each factor worth exactly one point.

The Apfel, Koivuranta, and Eberhart scores, among other risk models, were developed to minimize unnecessary administration of prophylactic antiemetics. Like all medications, antiemetics have costs, side effects, and variable efficacies. Proponents of the Apfel score and other risk prediction models argue these tools are useful for determining which patients should versus should not be given antiemetic prophylaxis. The Apfel score is often applied in conjunction with recommendations from the “Society for Ambulatory Anesthesia Guidelines for the Management of Postoperative Nausea and Vomiting,” and the “Consensus Guidelines for Managing Postoperative Nausea and Vomiting.” These documents recommend not providing prophylaxis to low-risk patients, giving one or two antiemetics to moderate-risk patients, and administering two or more antiemetics from different classes to those patients at highest risk of PONV. Propofol can also be used throughout the case for total intravenous anesthesia (TIVA), and this allows for one less antiemetic to be given. When applied to help guide decision making regarding prophylaxis, Apfel’s model has been shown to be efficacious and efficient in decreasing PONV in adults.

Despite this evidence, many clinicians support a more liberal use of antiemetics for preventing PONV. They strongly emphasize the low cost, benign side effects, and impressive efficacy of the available antiemetic drugs. Given the essential need to avoid PONV, it is recommended that all surgical patients be treated with at least two antiemetics. The risk factors should certainly be considered and applied to help guide decision making, but physicians should not be restricted to risk models (with inherent limitations) to predetermine their actions.