Abstract | OBJECTIVE: DESIGN: A derivation cohort from three acute respiratory distress syndrome studies was used to estimate the 7-day change in oxygenation index. Receiver operating characteristic curves were used to calculate optimal thresholds and predictability of the change in oxygenation index for 28-day mortality and ventilator-free days. The thresholds were then validated in two cohorts. Then, for each individual acute respiratory distress syndrome study, the threshold 7-day oxygenation index change was tested as an outcome measure and compared with mortality and ventilator-free days as reported in the original study. SETTING: Medical ICUs. PATIENTS: INTERVENTIONS: Various. MEASUREMENTS AND MAIN RESULTS: Change in oxygenation index, 28-day mortality, and ventilator-free days. In the derivation cohort, the mean 7-day oxygenation index improved by 4.2 (± 11.7) in 28-day survivors compared with an increase of 2.4 (± 11.6) in 28-day nonsurvivors (p < 0.001). The mean 7-day oxygenation index decreased by 5.9 (± 8.4) in patients with more than 14 ventilator-free days, compared with a decrease of 1.9 (± 12.4) among those with less than 14 ventilator-free days (p = 0.001). The optimal 7-day oxygenation index threshold for predicting mortality was an increase of 1.71 and for predicting less than 14 ventilator-free days, a decrease of 2.34. When used as a surrogate endpoint, the optimal 7-day oxygenation index change closely approximated mortality and ventilator-free day outcomes in three Acute Respiratory Distress Syndrome Network studies used for the derivation cohort and a distinct study used for validation. The change in oxygenation index was a poor predictor of individual patient outcome. CONCLUSIONS: Failure to meet a threshold improvement in the oxygenation index over the first 7 days of therapy can be used to identify therapies unlikely to succeed in subsequent trials powered for mortality and ventilator-free days. By reducing trial time and costs, use of the 7-day oxygenation index change as an intermediate endpoint could increase the number of clinical trials of promising therapies for acute respiratory distress syndrome and reduce the number of large-scale trials of therapies unlikely to be of benefit.
|
Authors | Leonard Go, G R Scott Budinger, Mary J Kwasny, Jie Peng, Jean-Marie Forel, Laurent Papazian, Manu Jain |
Journal | Critical care medicine
(Crit Care Med)
Vol. 44
Issue 1
Pg. e40-4
(Jan 2016)
ISSN: 1530-0293 [Electronic] United States |
PMID | 26427588
(Publication Type: Journal Article, Research Support, N.I.H., Extramural, Research Support, U.S. Gov't, Non-P.H.S.)
|
Chemical References |
|
Topics |
- Blood Gas Analysis
- Humans
- Oxygen
(blood)
- Predictive Value of Tests
- Randomized Controlled Trials as Topic
- Respiratory Distress Syndrome
(blood, therapy)
- Retrospective Studies
- Treatment Failure
- Ventilators, Mechanical
|