High-frequency oscillation is used for adults with acute lung injury/acute respiratory distress syndrome. Given the uncertain benefits, our objective was to describe contemporary patient selection, high-frequency oscillation utilization, and outcomes.
Ten academic centers (Ontario, Canada; January 1, 2005–January 31, 2007).
We included 190 consecutive adults treated with high-frequency oscillation and retrospectively abstracted data on patient demographics, gas exchange, hemodynamics, settings during conventional ventilation and high-frequency oscillation, adjunctive therapies, and outcomes. We used logistic regression to explore associations with oxygenation response and hemodynamic and ventilatory complications (2 hrs after high-frequency oscillation initiation) and mortality. Continuous data are reported as mean (sd) or median (quartile 1, quartile 3).
Patients (60.0% male; mean age, 52; sd 17) had predominantly acute lung injury/acute respiratory distress syndrome (89.8%) and were severely ill (mean Acute Physiology and Chronic Health Evaluation II score, 28; sd, 9) and hypoxemic (mean Pao2/Fio2, 80; sd, 42). High-frequency oscillation was started a median of 2 (1, 8) days after intubation and continued for a median of 3 (2, 7) days. During high-frequency oscillation, adjunctive treatments (neuromuscular blockade, 75.6%; recruitment maneuvers, 49.5%; nitric oxide, 34.0%; steroids, 63.1%) and new barotrauma or chest tube placement (23.5%) were common. Ten patients (5.3%) had technical complications; five required a new ventilator. Reasons for stopping high-frequency oscillation included death or withdrawal of life support (39.1%), significant improvement in respiratory failure (37.6%), and inadequate improvement (23.3%). One hundred and twenty-eight of 189 patients died in hospital. Most (62.5%) had a positive oxygenation response after 2 hrs of high-frequency oscillation (mean absolute increase in Pao2/Fio2, 22 [sd, 54]; 95% confidence interval of mean 14–31). A minority had lower Pao2/Fio2 (32.5%) or a hemodynamic (27.5%) or ventilatory (30.5%) complication. Older age, Acute Physiology and Chronic Health Evaluation II score, and Paco2 before high-frequency oscillation were independently associated with mortality.
Most patients treated with high-frequency oscillation have acute respiratory distress syndrome and severe hypoxemia that modestly improves 2 hrs after high-frequency oscillation application. However, oxygenation worsens in some patients, complications are common, and mortality is high.
From the Department of Critical Care Medicine and the Sunnybrook Research Institute (NKJA), Sunnybrook Health Sciences Centre, Toronto; Interdepartmental Division of Critical Care (NKJA, SM, NDF, JTG, JOF) and Department of Medicine, Division of Respirology (KC), University of Toronto, Toronto; Department of Critical Care (AB), Hamilton Health Sciences, Hamilton; Centre de Recherche Clinique Étienne Le Bel (FL), Université de Sherbrooke, Sherbrooke; Department of Medicine, Division of Respirology, University Health Network (NDF, JTG) and Mount Sinai Hospital (SM, NDF), Toronto; Department of Clinical Epidemiology and Biostatistics (QZ, LH, MOM, DJC), McMaster University, Hamilton; Critical Care and Medicine Departments and the Li Ka Shing Knowledge Institute (JOF), St. Michael's Hospital, Toronto; The Ottawa Hospital Research Institute (IW), Ottawa; Department of Medicine (AF, MOM, DJC), McMaster University, Hamilton, Canada.
Supported, in part, by the Ontario Thoracic Society (Hamilton chapter). Drs. Lamontagne and Meade are supported by a RCT Mentoring Award from the Canadian Institutes of Health Research (CIHR), Dr. Ferguson is supported by a New Investigator Award from CIHR. Dr. Friedrich is supported by a Clinician-Scientist Award from CIHR. Dr. Cook is a Research Chair of CIHR.
Preliminary results of the study have been presented in abstract form (Adhikari NK, Ferguson ND, Mehta S, Freitag A, Friedrich JO, Granton JT, Zhou Q, Stewart TE, Meade MO: Current high-frequency oscillation (HFO) utilization in Ontario. Am J Respir Crit Care Med 2009; 179:A3072).
Dr. Ferguson and Dr. Meade are principal investigators, and Dr. Adhikari, Dr. Cook, Dr. Friedrich, Dr. Granton, and Dr. Mehta are coinvestigators for the Canadian Institutes of Health Research (CIHR)-funded OSCILLation for ARDS Treated Early (OSCILLATE) Trial, with in kind support from Care Fusion (oscillators on loan to some study centers). The remaining authors have not disclosed any potential conflicts of interest.
For information regarding this article, E-mail: email@example.com