Validation of Risk Scores For Postoperative Nausea and Vomiting in Patients Undergoing General Anaesthesia For Elective Surgery at the University Teaching Hospital in Lusaka Zambia

Abstract

Background: Postoperative nausea and vomiting (PONV) is a common complication of general anaesthesia. There are documented risk factors associated with PONV which include female gender, non-smoker, and history of motion sickness, use of opioids and surgery duration of more than 60 minutes. Other risk factors are types of surgery such as Ear-nose-throat (ENT), gynaecological and laparotomy or laparoscopic surgeries. There are scores which have been created and are used to predict which patients are at high risk of PONV. Different risk scores are universally used to predict PONV in patients scheduled for surgery, but their validation has shown different outcomes in different settings and different ethnic populations. The objective of this study was to validate the commonly used risk score, (Apfel and Koivuranta risk scores) in the patients at the University Teaching Hospital in Lusaka, Zambia. Another objective was to determine the incidence of PONV at University Teaching Hospital. Method: This was a prospective observational cohort study conducted at the University Teaching Hospital in Lusaka Zambia. 246 patients were sampled and comprising both male and female. Two patients had missing data hence were omitted from analysis. The age range was between 18 and 80 years old. Patients were recruited the day before surgery and were seen and interviewed at least six hours post-operatively using a structured questionnaire (Appendix A). Any episode of nausea or vomiting was taken as PONV. The data was analysed using SPSS version 22. And discrimination was done using the receiver operating characteristic curve (ROC). Results: The overall incidence of PONV was found to be 25.4%. The receiver operating curve for Apfel for nausea was 0.63 which was lower than that in the original study which was 0.75. For Koivuranta the ROC area for nausea was 0.62 and for vomiting was 0.64. In the original study this was 0.70 and 0.72 respectively. (A value of 1 means perfect prediction and a value of 0.5 means no predictive values). The sensitivity was seen to be decreasing as the number of risk factors increased. Conclusion: The PONV risk scoring systems do not accurately predict which patients are at high risk of PONV in the population studied. Also the incidence of PONV is not as high as in the derivative population. Hence there is need to develop a new or modified score which will suit our environment.