A number of algorithms for the prevention of postoperative nausea and vomiting have been proposed. Their efficacy and efficiency remains unclear.

We assumed that four antiemetic interventions were similarly effective and achieved additive effects when combined. We applied published and hypothetical algorithms for the prevention of postoperative nausea and vomiting to patient populations with different baseline risks. As indicators of efficacy and efficiency we computed for each baseline risk and each algorithm the total number of patients receiving prophylaxis, the total number of administered interventions, the cumulative 24 h incidence of postoperative nausea and vomiting, and an Efficiency Index (i.e. the number of administered interventions divided by the achieved absolute risk reduction). This was done for cohorts of 100 patients.

Ten algorithms were tested in seven populations with different baseline risks. Algorithms were fixed (⩾ 1 intervention given to all patients, independent of baseline risk) or risk-adapted (⩾ 1 intervention administered depending on the presumed baseline risk). Risk-adapted algorithms were escalating (the greater the baseline risk, the more interventions are given) or dichotomous (a fixed number of interventions is given to high-risk patients only). With some algorithms, when applied to selected patient populations, the average postoperative nausea and vomiting incidence could be decreased below 15%; however, none produced consistent postoperative nausea and vomiting incidences below 20% across all populations. With all, the number of administered antiemetic interventions was the major factor for improved efficacy. Depending on the baseline risk, some algorithms offered potential towards improved efficiency.

Despite improved knowledge on risk factors and antiemetic strategies, none of the tested algorithms completely prevents postoperative nausea and vomiting and none is universally applicable. Anesthesiologists should try to identify the most useful antiemetic strategy for a specific setting. That strategy may be prophylactic or therapeutic or a combination of both, and it should consider institutional policies and individual baseline risks.