The impact of postoperative nausea and vomiting prophylaxis with dexamethasone on postoperative wound complications in patients undergoing laparotomy for endometrial cancer

Abstract

Background: Dexamethasone is widely used for postoperative nausea and vomiting (PONV) prophylaxis. However, there are limited data on the risk of wound complications associated with single-dose dexamethasone use for this purpose. We performed this retrospective study to determine whether intraoperative dexamethasone for PONV prevention increases the risk or severity of postoperative wound complications.

Methods: Women who underwent laparotomy for endometrial cancer between 2002 and 2007 were identified from a tumor registry. Perioperative records were reviewed to determine dexamethasone administration. Medical records were reviewed to identify wound complications including cellulitis, superficial surgical site infection, wound separation, and fascial dehiscence. Wound care needs and time to complete wound healing were compared based on dexamethasone exposure. The rate of wound complications was also compared based on dexamethasone dose. Baseline characteristics and perioperative details were evaluated for independent associations with wound complications. Logistic regression analyses were performed to predict the occurrence of wound complications.

Results: Four hundred thirty-one patients met inclusion criteria; 192 (44.6%) received dexamethasone (4-12 mg) and 31.1% developed a wound complication. In unadjusted analysis, there was no difference in the risk of developing a wound complication based on dexamethasone exposure; 53 of 192 patients (27.6%) who received dexamethasone developed a wound complication, compared with 81 of 239 (33.9%) who did not receive dexamethasone: odds ratio (OR) (95% confidence interval [CI]) = 0.74 (0.49, 1.13), P = 0.16. There was no difference in the distribution of wound complication types based on receipt of dexamethasone (P = 0.71), or in the incidence of wound complications based on the dose of dexamethasone (P = 0.48). Of patients who developed a wound complication, there was no difference in the need for IV antibiotics, vacuum-assisted wound closure, or in the rate of fascial dehiscence based on dexamethasone exposure. The time to complete wound healing was not different between the 2 cohorts (P = 0.48). In univariate analysis, higher body mass index (BMI), higher estimated blood loss, smoking, and longer duration of surgery were predictors of wound complications. Smoking (OR [95% CI]: 2.0 [1.3, 3.2], P = 0.003) and BMI (OR [95% CI]: 1.2 [1.1, 1.3], P = 0.0003) were the only significant predictors of wound complications in the multivariate model, whereas dexamethasone remained a nonsignificant predictor (OR [95% CI]: 0.7 [0.5, 1.1], P = 0.12).

Conclusion: Intraoperative dexamethasone for PONV prophylaxis does not seem to increase the rate or severity of postoperative wound complications in women undergoing laparotomy for endometrial cancer. BMI and smoking were significant predictors of wound complications in this patient population.