• H A Schwid and D O'Donnell.
• Department of Anesthesiology, University of Washington, Seattle.
• Anesthesiology. 1992 Apr 1;76(4):495-501.

AbstractHuman error is believed to contribute to the majority of negative anesthesia outcomes. Because retrospective analysis of critical incidents has several shortcomings and prospective studies are limited by the low frequency of critical incidents, an anesthesia simulator was used to evaluate the management of simulated emergency situations by ten anesthesia residents, ten faculty anesthesiologists, and ten anesthesiologists in private practice in order to identify specific patterns of errors in diagnosis and treatment. The simulator is a computer program that presents the patient, monitors, and management choices in a graphical display on an IBM or compatible personal computer. Many errors were observed in the management of these emergency situations, and even anesthesiologists with years of experience made serious errors. Although all experienced anesthesiologists correctly diagnosed simulated esophageal intubation, two residents misinterpreted the lack of end-tidal carbon dioxide. Only 40% of subjects correctly diagnosed simulated anaphylactic reaction; 27% adequately treated simulated myocardial ischemia; and 30% managed a simulated cardiac arrest according to Advanced Cardiac Life Support (ACLS) guidelines. Problems with continuous infusions of vasoactive agents were common. Fixation errors or failure to revise a plan in the presence of inconsistent cues were made by 63% of subjects. The subjects that gathered more information during simulated anaphylaxis made the correct diagnosis more often and made fewer treatment errors. The time since the last ACLS training was found to be an important predictor of correct management of simulated cardiac arrest.(ABSTRACT TRUNCATED AT 250 WORDS)

16 keywords...

hide…