Intensivists' opinion and self-reported practice of oxygen therapy

Intensivists frequently prescribe oxygen therapy for critically ill patients, however little is known about how intensivists manage oxygen therapy, or what factors influence their decisions. We surveyed intensivists listed on the Australian and New Zealand Intensive Care Society Clinical Trials Group database to investigate how intensivists report their approach to the monitoring, prescription and management of risks associated with oxygen therapy. The response rate was 60.4% (99/164 intensivists). Overall 81 (83.5%) respondents practised in metropolitan units and 50 (50.5%) had > or =14 years of intensive care unit specialty practice. All respondents reported using pulse oximetry and > 93% reported having access to a blood gas machine within their intensive care unit. Sixty-one percent of respondents (60/98) reported assessing other indices of tissue oxygenation (pH, lactate, MvO2). Twelve respondents (12.8%) believed that oxygen toxicity was a greater threat to lung injury than barotrauma when commencing mechanical ventilation. A significantly (P = 0.016) greater proportion of regional (5/16) than metropolitan (7/70) respondents were concerned that a high FiO2 is a greater threat to the lungs than barotrauma. For a ventilated acute respiratory distress syndrome patient, 36.8% (36/98 respondents) would not allow an SaO2 of < 85% for < or =15 minutes, and 27.6% (27/96 respondents) would not allow an SaO2 < 90% for > 24 hours. Respondents with < or = 14 years of specialty practice were more likely to specify the oxygen delivery device to be used (P = 0.014). Recognising the factors that currently influence oxygen administration decisions is a necessary prelude to the potential conduct of interventional studies, as well as for the development of better guidance for oxygen therapy in critical care.