Background: Postoperative nausea and vomiting (PONV) is a common occurrence with a reported incidence between 20% and 40%. In this prospective observational study, we sought to determine the incidence of PONV in a South African population, differentiating between black South African (African) and the remainder of the multiethnic South African population (non-African). We attempted to identify individual risk factors for PONV and to test the performance of the Apfel PONV predictive scoring system in our patient population.
Methods: The primary outcome for the study was nausea, vomiting or retching, or the combination of both events within 24 hours of surgery. We collected 800 patients, 400 Africans and 400 non-Africans in each group, over a 4-month period.
Results: There was a statistically significant difference in the incidence of PONV between African and non-African groups (27% vs 45%, P < 0.0001). Stepwise, backward logistic regression analysis identified female sex (odds ratio [OR], 1.9; 95% confidence interval [CI], 1.4-2.6), non-African ethnicity (OR, 2.1; 95% CI, 1.5-2.82), PONV or motion sickness history (OR, 2.6; 95% CI, 1.8-3.7), and the use of postoperative opioids (OR, 1.4; 95% CI, 1-1.9) to be independent predictors of PONV. The area under the receiver operator curve for the Apfel score was 0.62. When modeling the independent risk factors in our population, the combination of non-African ethnicity, female sex, and a history of motion sickness or PONV resulted in a receiver operator curve area of 0.67.
Conclusion: We were able to identify black South African ethnicity as an independent risk factor for decreasing the incidence of PONV. The reason for this observation remains speculative and further investigation is warranted. The inclusion of ethnicity as a risk factor into PONV scoring systems should be explored.