Background: Glucocorticoids are increasingly used perioperatively, principally to prevent nausea and vomiting. Safety concerns focus on the potential for hyperglycemia and increased infection. The authors hypothesized that glucocorticoids predispose to such adverse outcomes in a dose-dependent fashion after elective noncardiac surgery.
Methods: The authors conducted a systematic literature search of the major medical databases from their inception to April 2016. Randomized glucocorticoid trials in adults specifically reporting on a safety outcome were included and meta-analyzed with Peto odds ratio method or the quality effects model. Subanalyses were performed according to a dexamethasone dose equivalent of low (less than 8 mg), medium (8 to 16 mg), and high (more than 16 mg). The primary endpoints of any wound infection and peak perioperative glucose concentrations were subject to meta-regression.
Results: Fifty-six trials from 18 countries were identified, predominantly assessing dexamethasone. Glucocorticoids did not impact on any wound infection (odds ratio, 0.8; 95% CI, 0.6 to 1.2) but did result in a clinically unimportant increase in peak perioperative glucose concentration (weighted mean difference, 20.0 mg/dl; CI, 11.4 to 28.6; P < 0.001 or 1.1 mM; CI, 0.6 to 1.6). Glucocorticoids reduced peak postoperative C-reactive protein concentrations (weighted mean difference, -22.1 mg/l; CI, -31.7 to -12.5; P < 0.001), but other adverse outcomes and length of stay were unchanged. No dose-effect relationships were apparent.
Conclusions: The evidence at present does not highlight any safety concerns with respect to the use of perioperative glucocorticoids and subsequent infection, hyperglycemia, or other adverse outcomes. Nevertheless, collated trials lacked sufficient surveillance and power to detect clinically important differences in complications such as wound infection.