Utility of cocaine drug screens to predict safe delivery of general anesthesia for elective surgical patients.

Cocaine has been associated with acute hemodynamic changes, causing anesthesia providers to be concerned about adverse hemodynamic events during general anesthesia. We sought to determine if there were differences in the prevalence of adverse hemodynamic events, and if hemodynamic instability could be predicted in cocaine-positive patients undergoing general anesthesia for elective surgery. A retrospective cohort study was conducted in 300 (150 cocaine-positive, 150 cocaine-negative) consecutive adults with similar general anesthesia plans who were hemodynamically normal at baseline. Subjects were excluded if they were not alert at baseline, or if they required more than 1 surgical procedure. Slightly more than 50% of subjects were female, but cocaine-positive subjects were significantly more likely to be male (chi2 = 5.9; P = .02). Baseline systolic pressure (P = .001; mean difference, 6.5 mm Hg; 95% confidence interval [CI], 2.7-70.2), mean arterial pressure (P = .04; mean difference, 2.9 mm Hg; 95% CI, 1.0-5.7), and heart rate (P = .02; mean difference, 3.3/min; 95% CI, 0.46-6.2) were significantly higher, but not clinically important in the cocaine-positive cohort. Our study demonstrates that use of drug screen results alone is insufficient to predict the safe administration of general anesthesia in patients undergoing elective surgeries.