The discriminating power of a risk score for postoperative vomiting in adults undergoing various types of surgery

Background : Recently, we have demonstrated that the probability of postoperative vomiting (PV) following ENT surgery with inhalational anaesthetics can be predicted using a risk score. This score is based on gender, age, smoking status, history of motion sickness or postoperative nausea and vomiting and the duration of anaesthesia. Therefore, it is of interest whether this score is also accurate in predicting PV in patients undergoing different types of surgery.
Methods : Inpatients scheduled for bone, vascular, general or eye surgery were included in a prospective survey for PV over 24 h. Data of 1091 patients were analyzed, of which 542 were used for the validation of the previously constructed risk score (Score I). The data of the remaining 549 patients were used to evaluate the risk factors that contribute to PV in this setting and to develop a new score (Score II). The discriminating power of both scores to predict PV was tested in the validation set (n=542) and compared by calculating the area under the receiver operating characteristic (ROC) curves.
Results : The area under the ROC curve of Score I was 0.77 (SD 0.024). Risk factors for PV in the evaluation set were female gender, young age, history of motion sickness or postoperative nausea and vomiting and the type of surgery. The area under the curve of Score II was 0.75 (SD 0.026) and was not significantly different from Score I ( P =0.57).
Conclusion : Score I was accurate in predicting PV in patients after most types of surgery with volatile anaesthetics, which suggests that this score might be useful for other centres as well.