Yasser A. Sowb, Ph.D.*, Robert G. Loeb, M.D.^#
* Simulation Center for Crisis Management Training in Health Care, Veterans Affairs Palo Alto Health Care System & Stanford University School of Medicine, and ^# Department of Anesthesiology, University of Arizona

Background: Management of patient's ventilation is a fundamental yet demanding anesthesia task where humans commonly err. Of 2000 intraoperative critical incidents reported in the Australian Incident Monitoring Study, 16 percent involved problems with ventilation (Ventilation-Related Events or VRE). Monitoring equipment and their displays extend anesthesiologists’ resources during VRE but at the expense of additional cognitive demands. Cognitive Task Analysis (CTA) techniques were used to model clinicians’ management of patient’s ventilation during anesthesia and to map the demands of VRE. CTA techniques were also used to identify VRE which were then incorporated in a clinical scenario and simulated using the MedSim-Eagle patient simulator. The response of eight experienced clinicians was captured on videotape and analyzed to investigate the effectiveness of medical equipment in supporting clinical decision-making during VRE.

Methods: The study was approved by the University's HSRC. CTA techniques were used to identify two VREs related to clinicians’ management of patient’s minute ventilation during general anesthesia. The VREs were incorporated in a clinical scenario and simulated using the MedSim-Eagle patient simulator. Below is a description of the scenario.

The sequence of events during the scenario was as the following. Surgical incision stimulates a sudden onset of bronchospasm and peak airway pressure goes from 18 cm H2O to 35 cm H2O. The bronchospasm event continues until the subject administers three different therapies. In response to effective treatment, the bronchospasm event is cleared and peak airway pressure returns to normal. Five minutes following resolution of bronchospasm there is a gradual onset of unilateral tension pneumothorax over 5 minutes. Peak airway pressure increases from 18 cm H2O to 50 cm H2O and blood pressure falls from 105 to 93 mm Hg mean arterial pressure.

Results: Table 1 shows clinical data that aided subjects in detecting bronchospasm and pneumothorax. During bronchospasm, positive inspiratory pressure (PIP) measurement aided four (50%) of the subjects, ETCO2 measurement aided two subjects while heart rate (HR) and auscultated lung sounds (AUSC) aided one subject, apiece. During pneumothorax, patient’s blood oxygen saturation (SAT) aided 4 subjects, PIP measurement aided 3 subjects, and measurement of ETCO2 aided one subject. All subjects detected wheezing during bronchospasm. All subjects detected high positive inspiratory pressure and six subjects detected reduced breath sounds on the right when auscultating the lungs during pneumothorax.

Conclusion: The PIP and SAT measurements were effective in aiding clinicians’ detection of bronchospasm and pneumothorax, respectively. Higher PIP alarm settings and more frequent assessment of the PIP measurement could have improved clinicians’ detection during both bronchospasm and pneumothorax. The audible tone of the pulse oximeter helped direct subjects’ attention to pneumothorax. The stethoscope aided the subjects in identifying bronchospasm and in detecting abnormal breath sounds on the right during pneumothorax. The PIP measurement and stethoscope did not effectively aid the subjects in identifying pneumothorax. All eight subjects mistakenly treated pneumothorax as bronchospasm and repeatedly administered bronchodilators to the patient. These results will be used to improve intraoperative equipment design in order to better aid clinicians’ management of VRE.