Extracorporeal Lung Assist Device in Acute Lung Impairment

Overview

The purpose of this study is to compare the effect of interventional Lung Assist iLA activve to standard therapy in mechanically ventilated patients with severe acute lung impairment. Hypothesis: iLA(active) reduces the incidence of an increase in SOFA-Score of ≥3 points (or death) within 28 days compared to standard treatment.

Background: Mechanical ventilation in patients with acute lung impairment further injures the lungs by inspiration forces and inflammatory response. Large efforts have been invested in reducing ventilator-associated lung damage by lower tidal volumes. However, benefits are limited by potential harms of permissive hypercapnia. Therefore, extracorporeal membrane oxygenation (ECMO) and CO2-removal have been studied for more than 40 years. However, ECMO remained restricted to few specialized centres capable to provide extensive resources. Transfer of patients implicates loss of time and risks of transportation. Therefore, less invasive devices have been developed, including "pump-less "extracorporeal lung assist" (pECLA) and pump-driven ECLA (e.g. iLA activve). Despite pilot trials supporting feasibility, safety and efficient oxygenation and decarboxylation by pump-driven ECLA, there are no randomized controlled trials (RCT) proving a benefit regarding long-term endpoints. Objectives: Therefore, the aim of this multicentre RCT is to compare the outcome of 150 patients with early (after ≤96h of mechanical ventilation) acute lung impairment treated by pump-driven ECLA with iLA activve with a blood flow of at least 1L/min vs. 150 controls with standard intensive care including low tidal volume ventilation. Main inclusion and exclusion criteria: While most trials on ECMO and (p)ECLA included patients in a rescue scenario with severe and persisting ARDS, earlier inclusion also implicates modified inclusion criteria: A cumulative Murray score of ≥6 points without radiological points is the most important inclusion criterion. At least four points must result from pO2/FiO2 (mandatory pO2/FiO2\<300mmHg) and PEEP criteria of the Murray score. In order to provide sufficient time for conservative attempts to optimize ventilation, inclusion criteria can be fulfilled for a maximum of 48h before inclusion as long as the patient can be included within a maximum of 96h of mechanical ventilation. Primary efficacy endpoint: Incidence of an increase in SOFA-Score ≥3 points or death within 28 days. Statistical analyses: Generalized linear mixed model (logit link function) will be used to compare the primary efficacy endpoint, the proportion of patients with an increase in SOFA of ≥3 points or death within 28d, between the two groups. In this analysis the random factor variable study centre and anticoagulation therapy will be considered as adjustment variables. The test of group effect estimated by the multivariable mixed logistic model will be conducted at a two-sided 0.05 level of significance. The primary efficacy analysis will be based on the intention-to-treat population. Missing values of SOFA score will be replaced by last-value carry forward approach. Survival status of lost to follow-up patients will be replaced conservatively: missing survival status will be replaced by attribute "death" for patients in the verum arm and replaced by attribute "alive" for patients in the standard treatment arm. A supportive complete case and per-protocol analysis will be conducted for purpose of sensitivity analysis of the primary endpoint. Further sensitivity analyses will be provided to evaluate robustness of results in regard to unexpected circumstances (e.g. impact of 'cross-over' patients who are not treated as randomized but are required to be analyzed as randomized (ITT-principle)). Secondary endpoints will be analyzed in an exploratory manner. Chi-Square test or Fisher-exact test will be used to compare categorical data. For comparisons of continuous data between groups non-parametric tests (Kruskal-Wallis test, Mann-Whitney-U test) will be performed. 95% confidence intervals will be calculated for relevant measurements. SAS software (version 4.9 or future follow-up version).

Conditions

Interventions

Eligibility

Locations (8)

Outcomes

Central Contacts