HOUSTON – Failure to achieve threshold respiratory parameters within the first 48 hours after implementation of high-frequency oscillatory ventilation was linked to higher mortality in patients with severe acute respiratory distress syndrome.
This finding from a retrospective study suggests that the absence of sufficient early improvements in oxygenation in patients with the fulminant lung condition may justify a switch to an alternate ventilation strategy, according to Dr. Samantha Tarras of the University of Michigan Health System in Ann Arbor.
Although high-frequency oscillatory ventilation (HFOV) is indicated as a rescue therapy for patients with severe acute respiratory distress syndrome (ARDS), specific threshold parameters predictive of outcome have not been described, contributing to uncertainty about its optimal application, she said at the annual congress of the Society of Critical Care Medicine.
In a retrospective investigation, Dr. Tarras and her colleagues evaluated the link between threshold oxygenation values and mortality in patients placed on HFOV in the University of Michigan extracorporeal membrane oxygenation (ECMO) referral surgical ICU during 2005-2011. Patients were excluded from analysis if their baseline PaO2/FiO2 (P/F) ratio was 100 or more; if they had ECMO support; or if transition to conventional ventilation, withdrawal of care, or death occurred within the first 48 hours.
Of 112 patients placed on HFOV as part of a standardized ARDS treatment algorithm during the period of study, 58 met entry criteria, according to Dr. Tarras. "Most of the patients were male, young, and critically ill. The median number of days on mechanical ventilation prior to HFOV was 3, and the largest risk factors for ARDS were pneumonia followed by sepsis," she said. The mean P/F ratio at baseline of the patients included in the analysis was 58.4, the mean oxygenation index at baseline was 51.5, and in-hospital mortality was 41.3%, she said.
In univariate analyses, the mortality of patients who failed to reach a threshold P/F ratio of at least 100 within 48 hours was 75%, three times higher than the 24.3% observed in patients who achieved the threshold ratio, Dr. Tarras reported. The sensitivity and specificity of this threshold for predicting survival were 82.4% and 62.5%, respectively, and the positive and negative predictive values were 75.7% and 71.4%, respectively. "Similarly, a significant mortality rate was identified at a threshold oxygenation index of 25 at 48 hours," she said.
Failure to reach the threshold P/F ratio of 100 remained significantly associated with mortality in a multivariate logistic regression model that incorporated patient age, APACHE (Acute Physiology and Chronic Health Evaluation) score, and number of days of mechanical ventilation prior to initiation of HFOV, said Dr. Tarras. "The area under the receiver operator characteristic curve for model performance was 0.774," she noted.
The findings are limited by the lack of information on patients’ cause of death, which could have an effect, Dr. Tarras acknowledged. Even so, "the results tell us that for patients whose oxygenation is not improving after 48 hours of HFOV, clinicians should start thinking about other rescue strategies as well as referral to an ECMO center." Additionally, she said, "this group of patients that is at greater risk of death despite 48 hours [of HFOV] should be targeted in future intervention trials."
Dr. Tarras disclosed having no relevant conflicts of interest.