• Daiane Spitz de Souza, Amine Farias Costa, and Gabriela Villaça Chaves.
• Department of Nutrition and Dietetics, State University of Rio de Janeiro, Rio de Janeiro, Brazil.
• Support Care Cancer. 2016 Nov 1; 24 (11): 4661-7.

PurposeTo evaluate the predictors of postoperative nausea and vomiting (PONV) in women with gynecologic tumor.MethodsThe analysis was based on prospectively collected data of 82 adult patients with gynecologic tumor, who were submitted to open surgical treatment and undergoing general anesthesia. The predictors included were age ≥50 years, non-smoker, use of postoperative opioids, mechanical bowel preparation, intraoperative intravenous hydration (IH) ≥10 mL/kg/h, and IH in the immediate postoperative, first and second postoperative days (PO1 and PO2) ≥30 mL/kg. A score with predictor variables was built. A multiple logistic regression was fitted. To estimate the discriminating power of the chosen model, a receiver operating characteristic (ROC) curve was plotted and the area under the ROC curve (AUC) was calculated. Statistical significance was set at p value <0.05 and the confidence interval at 95 %.ResultsThe incidence (%) of nausea, vomiting and both, in the general population, was 36.6, 28.1, 22.0, respectively. The highest incidences of PONV were found in non-smokers and in patients who received >30 mL/kg of IH in the PO2. The results of the adjusted model showed an increased risk of PONV for each 1-point increase in the score punctuation. The relative risk was higher than 2.0 for vomiting in all period and in the PO1. The ROC curve showed great discrimination of postoperative nausea and vomiting from the proposed score (AUC >0.75).ConclusionsThe study population was at high risk of PONV. Therefore, institutional guidelines abolishing modificable variables following temporal evaluation of the effectiveness should be undertaken.

20 keywords...

hide…