Aspects on extracorporeal life support for severe acute respiratory failure with special reference to the influenza A/H1N1 2009 pandemic

Abstract

Extracorporeal membrane oxygenation provides pulmonary and circulatory support. First abandoned from the majority of the intensive care community because of disappointing results in several trials (with questionable methods), it experienced a renaissance during the H1N1 influenza pandemic in 2009/2010. Today, it is a widely accepted therapy option for the sickest respiratory failure patients, with survival rates between 50% and 80%, depending on the mode and the center where it is performed. But there is still a lack of high quality randomized trials to answer the question whether ECMO is superior to conventional mechanical ventilation in severe respiratory failure. But such trials are very difficult to perform mainly due to ethical reasons because a randomization to death is not acceptable for the majority of health care professionals. The question we have to ask ourselves today is not whether ECMO should be performed but how it should be performed, and here there is still a lot of work to do.

In this thesis, several aspects on extracorporeal life support for severe respiratory failure were investigated. Paper I describes the treatment strategies and short-term outcome of 13 patients with refractory severe respiratory failure due to infection with influenza A H1N1 2009 at the ECMO Department at the Karolinska University Hospital. All patients survived, and 12 were still alive 3 months after discharge from ECMO. In paper II the ECMOnet score is presented. It was developed by the Italian ECMOnet, and the 13 patients treated due to H1N1 infection in paper I were included in an external validation group for this score. It has a high accuracy for the prediction of mortality risk in the patients treated with venovenous ECMO for H1N1 respiratory failure. The probability of correctly classifying patients with this score was 75%, where a score of 4.5 was the most appropriate cutoff for prediction of mortality risk. In the external validation group, the score had a good capacity to distinguish survivors from non-survivors. Paper III is a neurocognitive long-term follow-up study of seven of the patients presented in paper I. The studied showed that despite prolonged episodes of hypoxemia, cognitive functioning was normal in all patients and that there were no hypoxic cerebral lesions. Severe respiratory failure is a hyperinflammatory condition, and neutrophil granulocytes play a key role in its development and progress. In paper IV, we explored the hypothesis that the change in proportions of mature and immature neutrophils could be used as a prognostic parameter during the course of ECMO treatment, but due to a low number of included patients, no strong conclusion could be drawn. The systematic review presented in paper V is a result of question that arose in paper III, namely whether hypoxemia during the course of acute respiratory failure or ECMO treatment per se is associated with short- and long-term cognitive dysfunction in survivors.

There are no high quality studies addressing this question, and it is therefore still not clear whether there is a causal relationship between hypoxemia and cognitive impairment. New studies are needed to investigate this important question because it is evident that different treatment strategies of acute respiratory failure have an impact on survival.