14th February 2015: Effect of Erythropoietin and Transfusion Threshold on Neurological Recovery After Traumatic Brain Injury

Presented by Dr Arun Ramachandran

The most important paper on TBI research this decade?


We have discussed transfusion thresholds in a variety of patient groups, for example SAH patients, trauma patients, and the wider group of critically ill patients. The paper this week is likely  be seen to be one of the most important studies in the management of haemoglobin levels in TBI patients, it seeks to answer two questions: firstly does erythropoietin (Epo) improve outcome in TBI and secondly does a transfusion threshold of 10 g/dL vs 7 g/dL improve outcome (correspondence).

It is a complicated paper and the study is perhaps overstretched in some respects, not least the examination of epo. This is usually thought of as a drug mediating erythrocyte production following release by the kidneys, and has been investigated in critical illness in that context. It has, however, been shown to have a wide range of effects and acts as a signalling molecule in a variety of environments including acute encephalopathy and small studies have demonstrated neuroprotective effects in, for example, TBI. The evidence for benefit is fairly mixed and there are no large RCTs supporting its use (the ANZICS Epo-TBI trail is underway). Even so, the authors have combined this study with a study on correction of anaemia, in itself a question which arguably is too complex be answered in a RCT format.

The decisions around transfusion continue to vex intensive care clinicians. The question, simply put, does the risk of transfusing anaemic critically ill patients outweigh the improvements in oxygen delivery to damaged tissues. Superficially the literature is clear, there is no benefit in critically ill patients to transfusing above 7 g/dl. Inevitably, however, there is more to this story and some background reading can be found in previous JC posts, linked above. Much more usefully, however, our colleagues on NCCU and friends in Vancouver published this recent study (open access) which examines the relationship between two novel indices of haemoglobin level and outcome following TBI, the discussion explores some of the debate in the literature and also outlines why isolated measures of Hb may not be sufficient to make judgements about risk of benefit vs harm. Saying that, this recent study could well add weight to their argument.

Whilst, in medicine, we often are prone to think about benefit and its causes we should remember that the mechanisms of harm are also not entirely elucidated, we frequently talk about TRALI, fluid overload, immune suppression and infection (see links above) but much of the data supporting these concerns is historical and does not reflect current transfusion technology and practice. Arguably we need data about the mechanisms of harm as much as we need the data about outcome benefit and this recent paper from London continues to develop one of the potential mechanisms by which transfusion worsens outcome.

This is a potentially practice changing publication … what effect will it have at Cambridge?

JAMA Clinical Evidence Synopsis for Epo in critically ill patients. Author interview from the JAMA article page:

 Intensive Blog journal club.

Information about the ANZICS Epo-TBI study.

Nature Reviews Neurology review on progress in TBI research by Menon and Maas.

The TRISS study, which we will discuss at a forthcoming journal club, and an editorial on this by Paul Herbert of the TRICC study.