A Systematic Literature Review of Packed Red Cell Transfusion Usage in Adult Extracorporeal Membrane Oxygenation
Round 1
Reviewer 1 Report
Hughes et al conducted a systematic review and meta-analysis of the transfusion usage in adult ECMO. They conducted a predefined search and short-listed 54 for their analysis. Substantial number of publications on ECMO did not report sufficient transfusion data. Most of the manuscripts included were single-centered and retrospective, with a great variation in practice and transfusion target amongs centers. The manuscript is well-written and clear. As demonstrated in this review, transfusion is an important area with inherent risks, yet often not reported. The review has helped highlighted a gap in current knowledge of ECMO and should be accepted for publication. It is a shame that finding is severely limited by the lack of data. Some minor suggestions: - please consider converting the table into landscrape, or at least put the heading at the beginning of each page so readers don't have to scroll back to the top all the time. - under Result 3.1, it will be helpful to mentioned the number of manuscripts included in the review after all the exclusion (line 109). - consistent formating, some paragraphs started with indentation, some didn't. e.g. line 105, 129, 137 etc. - some of the numbers in Figure 1 do not add up. e.g. Full text accessed (331) + excluded manuscripts (1187) = 1518, 4 more than records screended (1514). Also all the sub-categories of excluded records do not add up to 1187.Author Response
Dear colleague,
Thank you for your review of our manuscript.
I have made minor changes reflecting all 4 of your suggestions which will appear in the revised submission - I agree with your comments and appreciate your attention to fine details. In particular I have corrected the arithmetic in the CONSORT diagram.
Kind regards,
Tom
Reviewer 2 Report
It was a great honor to review A systematic literature review of packed red cell transfusion usage in adult extracorporeal membrane oxygenation (ECMO) by Hughes et al.
It is of very high quality and should be considered for publication in Membranes.
I have some small comments and questions.
Please remove the word ECMO form the title, abbreviations should me in the text.
in the material and methods sections you mention concurrent use of other support modalities. Why did you exclude these as they are currently used often for LV unloading. Also which modalities were most often used?
Did you observe any difference between the transfusion rate when you looked the kind of pump used?
Otherwise I have no comments and must applaud the authors at their solid piece of work
Author Response
Dear colleague,
Thank you for your review of our manuscript.
I have amended the title in the revised manuscript.
Regarding secondary modalities - IABP use was explicitly allowed and this was commonly seen. 15 studies reported IABP use with a median of 63% in those cohorts; 2 studies reported IABP use but not frequency whilst the remaining studies were either VV-only or IABP use was not mentioned.
One study reported 1/37 patients receving ECMO concurrently with Impella and no other studies noted this. Additionally, I reviewed our exclusions and no studies were excluded because their main cohort used both ECMO and another modality (such as Impella) as an adjunct. However, the exclusion criteria was part of the original design to rule out alternative devices (pECLA/iLA/ECCO2R for VV and VAD/pVAD) which might have a confounding effect on haemorrhage/clotting risk and thus transfusion profile.
In a future cohort I think it would be reasonable to consider explicitly including devices used as an adjunct to optimise ECMO delivery and I would expect in more recent cohorts (our last publication date was 2016 with most cohorts finishing significantly earlier than that) that Impella use would be more frequent and worth noting.
Comparison of pump type and transfusion rate did not show a significant difference (centrifugal 3.03unit/day, all others 2.37unit/day, p=0.15). It was not included initially for brevity and because it is a very lop-sided comparison - 40 of the 54 papers used only centrifugal pumps and half of the remainder did not specify their pump type.
Thank you again for your feedback.
Kind regards,
Tom
