Refractory Hypoxemia and Use of Rescue Strategies. A U.S. National Survey of Adult Intensivists.

RATIONALE: The management of severe and refractory hypoxemia in critically ill adult patients is practice based. Variability across individual practitioners and institutions is not well documented.

OBJECTIVES: To conduct a nationwide survey of critical care physicians in the United States regarding accepted definitions and management strategies for severe and refractory hypoxemia.

METHODS: A web-based survey was distributed to a stratified random sample of adult intensivists listed in the American Medical Association Physician Masterfile. The survey was generated by using a mixed-methods approach.

MEASUREMENTS AND MAIN RESULTS: In the survey, 4,865 e-mails were sent and 791 (16.3%) were opened. Among those who opened the e-mail message, 50% (n = 396) responded, representing 8.1% of total surveys sent. Seventy-two percent stated that their institutions lacked a protocol for identification and management of severe or refractory hypoxemia in the setting of acute respiratory failure. While the majority of respondents used low-Vt ventilation (81%), high positive end-expiratory pressure (86%), recruitment maneuvers (89%), and either bolus or infusion neuromuscular blockade (94%), there was marked variability in the use of specific rescue strategies as tier 1 or 2 interventions: prone position (27.8% vs. 47.8%, respectively), extracorporeal membrane oxygenation (2.3% vs. 51.2%, respectively), airway pressure release ventilation (49% vs. 34.5%, respectively), inhaled vasodilators (30.1% vs. 40%, respectively), and high-frequency oscillatory ventilation (7.8% vs. 40%, respectively). The variability was partly explained by providers' expertise with particular rescue strategies (77.7%), advance directives (70.1%), the training of allied health staff (62.3%), and institutional availability (53.8%).

CONCLUSIONS: U.S. adult critical care physicians predominantly employ lung-protective ventilation for severe hypoxemia. A wide variation in other rescue strategies is noted, which is partly explained by user expertise and availability. Less than 30% institutions have formal protocols for management of refractory hypoxemia.