Liberal or Restrictive Transfusion after Cardiac Surgery

Saturday, March 14, 2015

Submitted by

Source

Source Name: New England Journal of Medicine

Source URL: http://www.nejm.org/doi/full/10.1056/NEJMoa1403612?query=featured_home

Author(s)

Gavin J. Murphy, Katie Pike, Chris A. Rogers, Sarah Wordsworth, Elizabeth A. Stokes, Gianni D. Angelini, and Barnaby C. Reeves, for the TITRe2 Investigators

This is a very provocative British study that examined whether a more liberal (Hgb < 9 g/dL) versus a more restrictive (Hgb < 7.5 g/dL) transfusion threshold after cardiac surgery is associated with higher morbidity and costs. The authors conducted a multicenter, prospective randomized trail by randomizing patients who had a Hgb < 9 g/dL after heart surgery to one of the two groups. Over 2,000 patients were enrolled in the study. Interestingly, the group with the more liberal transfusion protocol had no worse outcomes and no higher costs than the restrictive group. Moreover - even more provocatively - the secondary outcome of 90-day all-cause mortality was 64% more likely to occur in the restrictive transfusion threshold group.

Comments

Submitted by Abdul Quadeer on Fri, 2015-03-20 07:09

There is a problem with the definition of liberal group. It is difficult to understand <9g/dl (7.5 to 9g/dl) is liberal group.

Submitted by Kyriacos Kolocassides on Fri, 2015-03-20 07:18

This is a very strange study, as it does not answer that much. 1) The usual target in the UK and other paces has been Hb 8g/dL. With this threshold we have been transfusing less than 10% of our patients, the only patients really needing a transfusion being small women less than 55kg. In this study there was a 54% transfusion with the 7.5g/dL threshold! We would not need to transfuse almost anyone with that limit! 2) We have always excluded the over 80s. We already know that they do better with Hb>10. This is becoming more of an issue as the proportion of over 80s is increasing. This study does not have such a group. 3) The results really show that limiting transfusion is not deleterious. So, why transfuse? 4) Blood transfusion is often accompanied with other product transfusions. This needs to be detailed. 5) I would rather see a study trying to explain how some centres manage with very low transfusions and other centres do not. I suspect local practices have a bigger role than medicine.

Submitted by Kyriacos Kolocassides on Fri, 2015-03-20 07:22

PS, the statistical analysis that I would like to see done is one on the basis of the "exit" Hb.

Submitted by Mohammad Mohammad on Sat, 2015-03-21 03:03

i agree with hb about 9 there is good for all ( wound.lung.saturation.activity.........)

Submitted by Sarkis Ejbeh on Thu, 2015-03-26 00:06

A level of 9 hb is mostly considered as the threshold to transfuse... Some cases should be payed a special attention as per ex: low EF,low SaO2,low BP,where this category blood transfusion is more indicated for even an acceptable Hb..