Wound Complications with Dexamethasone for Postoperative Nausea and Vomiting Prophylaxis

Dexamethasone and 5-hydroxytryptamine-3 receptor antagonists are probably the most frequently used drugs for postoperative nausea and vomiting (PONV) prophylaxis in modern anesthetic practice. The number needed to treat to prevent PONV for ondansetron has been found to be 6 to 7, whereas that for dexamethasone is 4.1,2 With better efficacy, the relatively low cost of dexamethasone renders it a more desirable option,3 except for the potential side effects that can be associated with steroid use. In this issue of Anesthesia & Analgesia, Bolac et al.4 present a single-center, retrospective observational study that attempts to assess the effect of an intraoperative single dose of IV dexamethasone (4–12 mg) on postoperative wound complications in 431 female patients undergoing gyne-oncological surgery for endometrial cancer. The primary outcome was a composite of postoperative wound complications (surgical site infection, wound cellulitis, wound separation and fascial dehiscence). The authors also reported other outcomes, such as time to complete wound healing in patients who had wound complications. The results of the study suggest that there is no statistically significant increase in the incidence of wound complications or time to complete wound healing in patients who received dexamethasone for PONV prophylaxis. This study has strengths, such as the reported homogeneity of the study population. There are also a number of weaknesses that may limit the validity and generalizability of the results. First, similar to previous studies on the same topic,5,6 the authors chose a retrospective design, possibly due to time and cost constraints. Interestingly, although these previous studies5,6 were from the same institution, they had different results, likely due to their different retrospective study designs. There are always inherent questions about the presence of unbalanced confounding covariates, which could influence the results of retrospective analyses. For example, the authors failed to consider some potentially important covariates such as preoperative infectious processes, immune competency, poor nutritional status, and others.7–9 Similarly, 6 surgeons performed the procedures in the study population, but surgeon-specific factors were not considered in the analysis. Second, a nonsignificant absolute decrease of 6.3% in wound complications was found in the dexamethasone group. With a sample size of 431 patients, the study only had sufficient power to detect a 30% or greater increase in the primary end point. We estimate that with a baseline complication rate of 34% (as reported by the authors), a sample size of >2000 patients would be needed to detect a 6% difference in primary outcome between the groups. What are the potential noninfectious complications associated with steroid use in the context of PONV prophylaxis? Suppression of cortisol secretion has been associated with antiemetic doses of dexamethasone (8 mg) for up to 24 hours,10 but the clinical relevance of this effect is uncertain.11 Gastrointestinal complications including perforation and bleeding have been reported with dexamethasone use but only at high doses (16–100 mg) and after prolonged administration (24 days).12 Similarly, avascular necrosis of the femoral head is associated with dexamethasone use for prolonged periods (2–78 months).13 Evidence for other dexamethasone-induced complications, such as perineal pain, and resistance to muscle relaxants after chronic therapy, have been reported.14,15 Finally, glucocorticoid administration is associated with neuromuscular weakness in critically ill patients.16 What about infectious and wound healing complications? To begin to answer this question, let us consider the biologic plausibility. Although there is evidence that a single dose of dexamethasone given as PONV prophylaxis can cause mild hyperglycemia for 24 hours,10 there is no evidence that this transient effect would be sufficient to promote wound complications.17 Animal studies18,19 suggest that a single dose of dexamethasone in very high doses (1 mg/kg) can decrease plasma cytokine levels (interleukins 1, 3, and 6), collagen formation, epithelialization, and fibroblast content in wounds; thus, there is reason for concern in nonhuman models. A recently published multicenter prospective randomized placebo-controlled trial (Dexamethasone for Cardiac Surgery [DECS] trial) studying the effect of a single intraoperative high dose of dexamethasone (1 mg/kg) on outcomes including death, stroke, myocardial infarction, renal, and respiratory failure, suggested that the incidence of postoperative wound infection was not significantly increased.17 Interestingly, although the postoperative glucose levels were elevated in the dexamethasone group, the incidence of many postoperative infections, including pneumonia and catheter-related infection, was lower in this group. Similarly, short duration glucocorticoid therapy has not been associated with infectious complications in other vulnerable populations. Methylprednisolone prophylaxis for posttracheal extubation stridor in critically ill patients was found to be efficacious and without infectious complications.20 A 3-week course of high-dose glucocorticoid (methylprednisolone) therapy in late phase lung injury (acute respiratory distress syndrome) patients also failed to show any difference in infectious complications.16 However, high doses (1 mg/kg) and chronic long-term (>21 days) administration of glucocorticoids does increase infectious risks and inhibit wound healing.19,21 Unfortunately, to our knowledge, prospective comparative data do not exist in many surgical populations, such as in neurosurgery, where perioperative glucocorticoid therapy is used routinely. The point at which the risk/benefit ratio for infectious and wound healing complications with short-term glucocorticoid therapy shifts from benefit to harm remains unknown. However, the available literature suggests that this point is well beyond a single dose of intraoperative dexamethasone for PONV prophylaxis, and the findings of the current report confirm this. Therefore, it is our opinion that anesthesiologists can safely continue to use 4 to 8 mg dexamethasone for the prevention of PONV. High-level evidence from a large multicenter randomized trial that assesses surgical wound complications as a primary outcome would be extremely useful. Such a study (NCT00273377) has been completed recently, and its results should be reported in the near future.22 DISCLOSURES Name: Shariq Ali Khan, MD, FRCA. Contribution: This author helped prepare the manuscript. Attestation: Shariq Ali Khan approved the final manuscript and is the archival author. Conflicts of Interest: The author has no conflicts of interest to declare. Name: David L. McDonagh, MD. Contribution: This author helped prepare the manuscript. Attestation: David L. Mcdonagh approved the final manuscript. Conflicts of Interest: The author is the paid consultant of Cephalogics Corp LLC. Not related to PONV. Name: Tong J. Gan, MD, MHS, FRCA. Contribution: This author helped prepare the manuscript. Attestation: Tong J. Gan approved the final manuscript. Conflicts of Interest: The author received research grant support and/or honoraria from AcelRx, Baxter, Fresenius, Hospira, Merck, Pacira. This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).