Rapid Responses to:

PAPERS: Neil McGill, Denise O'Shaughnessy, Ruth Pickering, Mike Herbertson, and Ravi Gill Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial BMJ 2002; 324: 1299 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Cell Savers in Cardiac Surgery

Russell WJ Millner   (1 June 2002)

Blood Use in Cardiac Surgery

Vipin Zamvar, Nicola Payne   (12 June 2002)

Reducing Blood Transfusion in Cardiac Surgery

Nick Stratford   (13 June 2002)

Journal's responsibility is not to exaggerate findings

Alain Vuylsteke, Caroline Gerrard   (20 June 2002)

Cell salvage reduces blood use – but does it do it on its own?

Michael H Cross, Christopher M Munsch (Dept of Cardiothoracic Surgery), Derek R Norfolk (Dept of Haematology).   (1 July 2002)

Untitled

Ivor Cavill, lternatives to blood transfusion   (2 July 2002)

Cell Savers in Cardiac Surgery 1 June 2002
Russell WJ Millner, Consultant Cardiothoracic Surgeon Blackpool Victoria Hospital, FY3 8NR Send response to journal: Re: Cell Savers in Cardiac Surgery	For a number of years we have routinely used cell savers for all cardiac surgery undertaken in this institute. When we looked at our own data we found a roughly similar fall of about 50% in our percentage of patients being transfused. Currently under 20% of my CABG practice will receive any blood products. Interestingly we found a sharp fall also in the use of FFP and Platelets when we started cell saving. I would question why 50 plus patients in this paper received a single unit blood transfusion. Indeed I would ask why anyone should need a single unit blood transfusion. Many units would find an Hb transfusion trigger of 9 as being high. We certainly would use 8 as a trigger level for transfusion, and then would only transfuse in a patient with an established clinical need, such as bleeding in the early post-operative period or symptoms clearly attritable to anaemia in the later post- operative period. (And would then plan to give 2 units.) I was surprised also by the implication that patients who were not being transfused appeared to be receiving other blood products. My understanding was that RBC's were less of a risk than pooled platelets or FFP. Putting these points aside I'm sure any unit that doesn't routinely cell save should use the evidence in this paper to support the lease or purchase of cell savers. It is difficult to see how it is possible to justify not using them. Russell Millner
Blood Use in Cardiac Surgery 12 June 2002
Vipin Zamvar, Consultant Cardiothoracic Surgeon Royal Infirmary of Edinburgh, EH93YW., Nicola Payne Send response to journal: Re: Blood Use in Cardiac Surgery	Dear Sir, We have a few observations regarding McGill et al's [1] study regarding blood use in cardiac surgery. Blood transfusion in cardiac surgery is very common, and there is no doubt, a great potential for decreasing the incidence. Off-Pump Coronary Artery Bypass Graft (OPCAB) surgery is rapidly gaining popularity [2], and has been shown to significantly decrease blood product requirements, in randomised [3], and observational [4] studies. How did the investigators treat patients scheduled for OPCAB? Presumably they were not invited to take part in the study (not mentioned in the exclusion criteria). This paper and the accompanying Science Commentary by Ali Berger are silent on the role of OPCAB. Am important piece of information missing from the paper is the volume of blood scavenged by the cell saver during the course of the operation. In the cell saver group, was a cardiotomy suction used during the course of the bypass? This would result in considerably less volume being scavenged by the cell saver. If it had not been used, the conclusions drawn regarding the benefits of using a cell saver may be erroneous. The benefits could be due to the avoidance of the cardiotomy suction, which is a known cause of hemolysis [5]. Most cardiac units using the cell saver routinely, wash only the blood scavenged by the cell saver (and then only if it is a significant amount), and not the residual in the cardiopulmonary bypass (CPB) circuit. In this study, in the cell saver group, the blood from the CPB circuit was washed. In the control group, however, the residual blood in the bypass circuit was not washed. Typically, the Hb concentration of the pump blood at the end of bypass is between 6 to 8 g/dl. Washing this blood could have resulted in a slightly higher hematocrit for the control group, and possibly a decreased need for blood transfusion (as the protocol for blood transfusion was strictly the Hb value in the postoperative period). Another observation is regarding the use of logistic regression to adjust for the effect of surgeon. How do the adjusted odds ratios compare with the univariate ones you have shown in the table? In comparing the percentages of patients receiving blood products, what was the need to control for the surgeon, given that there was a protocol in place to guide postoperative blood transfusion. These observations notwithstanding, this paper highlights the tremendous potential that exists of decreasing blood use in routine cardiac surgery. Yours sincerely, Vipin Zamvar, MS, FRCS(CTh). Nicola Payne MPhil. References 1) McGill H, O'Shaughnessey D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: a randomised controlled trial. BMJ 2002;324:1299-303. 2) Mack MJ. Pro: Beating-heart surgery for coronary revascularization: Is it the most important development since the introduction of the heart-lung machine? Ann Thorac Surg 2000;70:1774-8. 3) Angelini GD, Taylor FC, Reeves BC, Ascione R. Early and midterm outcome after off-pump and on-pump surgery in Beating Heart Against Cardioplegic Arrest Studies (BHACAS 1 and 2): a pooled analysis of two randomised controlled trials. Lancet 2002;359:1194-9. 4) Zamvar V, Khan NU, Madhavan A, Kulatilake ENP, Butchart EG. Clinical outcomes in Coronary artery bypass graft surgery: Comparison of the off-pump and on-pump techniques. Heart Surgery Forum 2002;Vol 5: (In Press). 5) Sobel M, Dyke CM. Hemorrhagic and Thrombotic Complications of Cardiac Surgery. In: Baue AE, Geha AS, Hammond GL, Laks H, Naunheim KS, editors. Glenn's Thoracic and Cardiovascular Surgery. Stamford: Appleton and Lange, 1996. pp 1793-1809.
Reducing Blood Transfusion in Cardiac Surgery 13 June 2002
Nick Stratford, Consultant Cardiothoracic Anaesthetist James Cook University Hospital Send response to journal: Re: Reducing Blood Transfusion in Cardiac Surgery	The results suggest five patients would need to be treated with cell salvage to prevent one patient recieving a transfusion of one unit of blood (number needed to treat 5, average number of units transfused in control group 1.07). Unfortunately, this does not yet clinch the economic argument as 5 sets of cell saver disposables still cost more than one unit of blood. However, the scarce resource and risk of infection considerations compel us to minimise perioperative transfusion. A recent audit of our practice demonstrated that we were only giving red cell transfusions to 18% of our routine coronary artery bypass patients. We do not use cell salvage or haemodilution and all were done on bypass. We do however use some of the other strategies suggested by the authors and respondents such as lower haemoglobin targets and anti- fibrinolytics. I'm sure that a balanced approach is the best one to reduce our blood use still further.
Journal's responsibility is not to exaggerate findings 20 June 2002
Alain Vuylsteke, Consultant in Anaesthesia and Intensive Care Papworth Hospital, Caroline Gerrard Send response to journal: Re: Journal's responsibility is not to exaggerate findings	Editor - We wish to congratulate McGill et al for their randomised controlled trial on mechanical methods of reducing transfusion in cardiac surgery1 but we would like to challenge the editor of the BMJ for choosing this paper as illustration of the front cover of the journal, with the added line: 'Cell salvage reduces the need for a transfusion'. We believe the journal has a responsibility not to exaggerate the perceived importance of findings, either in press releases2 or by other means. The authors have themselves highlighted the main limitations of the study in their discussion and we believe that these, added to others, are important safeguards against widespread use of the described methods without further scientific evaluation. While the authors are commenting on the high cost of transfusion, they have not included a cost analysis in their study. We understand the risks associated with transfusion and agree that these in themselves may justify any added cost. However, the introduction of strict transfusion guidelines and monitoring of their use is certainly a step that should precede the introduction of new equipment. We have been reinforcing the transfusion guidelines for the last 2 years in our institution and are continuously monitoring blood transfusion within our hospital. Interestingly, when looking at a very similar patient population as the one described by McGill et al. (Table 1), but lacking the research methodology and statistics, we have achieved a similar decrease in use of all blood products (Table 2), mainly by insuring that blood is only transfused when required and blood products given when need is documented. Despite our program, blood products are still administered too often outside the hospital guidelines and we feel that we could decrease their use even further before introducing new expensive equipment in our routine practice. The cover of the BMJ may only reinforce incorrect practices by encouraging widespread use of techniques rather than common sense. While we agree that the combination of acute normovolaemic haemodilution to cell saving did not confer any benefit in this study, the authors have not explored acute normovolaemic haemodilution as such and therefore no conclusion can be made with certainty about the absence of benefit of acute normovolaemic haemodilution itself. In the author's own words, we can say that this study adds to many others that are inconclusive because of the quality of the evidence. We feel therefore that the cover of the BMJ, while attractive, may have mislead readers and therefore question the rationale behind such a decision. 1. McGill N, O'Shaughnessy D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ 2002; 324:1299-303 2. Woloshin S, Schwartz LM. Translating research into news. JAMA 2002; 287:21:2856-8. Alain Vuylsteke, MD. Consultant in Cardiothoracic Anaesthesia and Intensive Care Papworth Hospital NHS Trust, Cambridge CB3 8RE, UK. alain@vuylsteke.net Caroline Gerrard Surgical Transfusion Coordinator Anaesthetic Research Unit, Papworth Hospital NHS Trust, Table 1 Characteristics of patients (n=255 in each survey). Each survey includes consecutive patients who had elective first time coronary artery bypass graft surgery in our institution during the time period indicated. Values are numbers (%) of patients, unless stated otherwise Initial Survey (06/04/00 - 11/08/00) Latest Survey (12/11/01 - 05/06/02) Men 212 (83%) 208 (82%) Age (years): Mean (SD) 64.6 (8.9) 65.3 (9.1) Range 36 - 83 37 - 83 Weight (Kg) Mean (SD) 81.3 (14) 82.0 (14.4) Range 44 - 122 52 - 119 Parsonnet score† Mean (SD) 7.1 (5.9) 7.9 (5.9) Range 0 - 36 0 - 26 Aspirin taken in 72 hours before surgery 14 (5.5%) 24 (9.4%) † 0 = low risk, 25 = high risk Table 2 Blood products given to patients (n=255 in each group) during their perioperative course. Each survey includes consecutive patients who had elective first time coronary artery bypass graft surgery in our institution during the time period indicated Blood product received Initial Survey (06/04/00 - 11/08/00) Latest Survey (12/11/01 - 05/06/02) Allogeneic blood: No of patients 102 94 Mean (SD) No of units received per patient 1.27 (3.18) 1.05 (2.04) Range 0 - 43 0 - 15 Fresh frozen plasma: No of patients 26 10 Mean (SD) No of units received per patient 0.22 (0.73) 0.10 (0.57) Range 0 - 6 0 - 6 Platelets: No of patients 26 10 Mean (SD) No of units received per patient 0.16 (0.52) 0.04 (0.22) Range 0 - 3 0 - 2 No of patients given any blood product 107 95 References 3. McGill N, O'Shaughnessy D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ 2002; 324:1299-303 4. Woloshin S, Schwartz LM. Translating research
Cell salvage reduces blood use – but does it do it on its own? 1 July 2002
Michael H Cross, Consultant Anaesthetist The General Infirmary at Leeds, Great George Street, Leeds, LS1 3EX, Christopher M Munsch (Dept of Cardiothoracic Surgery), Derek R Norfolk (Dept of Haematology). Send response to journal: Re: Cell salvage reduces blood use – but does it do it on its own?	McGill demonstrated in a prospective randomised study that cell salvage could reduce the average blood transfusion during first time coronary artery surgery by 0.39 units (1). Although this was of statistical significance there are those who might question the clinical or financial significance of this finding. It is interesting to note that the effectiveness of cell salvage appears to be decreasing as time goes by. In 1987 cell salvage was shown to save an average of 1.9 units per patient (2), by 1990 it was 0.8 units (3), 197 mls (approx. 0.6 units) in 1993 (4), and in 2002 only 0.39 units are saved. Indeed, it would not be surprising if next year a well conducted study was published showing no demonstrable benefit from cell salvage, and cell salvage may be discarded as worthless technology. The problem for investigators is that cell salvage is just one of a variety of techniques used to reduce blood loss and blood usage during cardiac surgery. For example it is possible that McGill might have failed to demonstrate a benefit from cell salvage if antifibrinolytics had been used in all patients rather than only 40% or if the transfusion trigger had been set at 25% rather than 27%. So should cell salvage be used during cardiac surgery? The answer, in our opinion, is almost certainly yes. However cell salvage must be seen as just one part of an integrated, multidisciplinary approach to blood conservation. We must identify all the technical, pharmacological and clinical blood conservation methods available, optimise each method and then build them into a comprehensive integrated approach to blood conservation. It is this comprehensive approach which has been shown to be of most benefit (5), even though the efficacy of individual elements may be difficult to prove. 1 McGill N, O’Shaughnessy D, Pickering R, Herbertson M, Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ 2002; 324:1299-303. 2 Breyer RH, Engelman RM, Rousou JA, Lemeshow S. Blood conservation for myocardial revascularisation. Is it cost effective. JTCVS 1987; 93: 512- 22. 3 Hall RI, Schweiger IM, Finlayson DC. The benefit of the hemonetic cell saver apparatus during cardiac surgery. Can J Anaes 1990; 37: 618-23. 4 Laub GW, Dharan M, Riebman JB, Chen C et al. The impact of intraoperative autotransfusion on cardiac surgery. A prospective randomised controlled trial. Chest 1993; 104: 686-9. 5 Helm RE, Rosengart TK, Gomez M, Klemperer JD. Comprehensive multimodal blood conservation: 100 consecutive CABG operations without transfusion. Ann Thorac Surg 1998; 65: 125-36.
Untitled 2 July 2002
Ivor Cavill, Senior Lecturer in Haematology Univ of Wales College of Medicine, Heath Park, Cardiff, lternatives to blood transfusion Send response to journal: Re: this article	Dear Editor, McGill et al1 and Berger2 make a good case for reducing dependence on blood transfusion in surgery. The recent Chief Medical Officers' Conference on Better Blood Transfusion II was also directed towards that end and in addition pointed to the likely implosion of the donor population and blood scarcity in the near future. But why is the focus still on trying to improve blood transfusion rather than the patient's ability to make their own blood? Why has the last twelve years of experience in renal medicine not got through to the collective consciousness in making the point that there are alternatives to transfusion that are safer and more effective. The use of recombinant human erthropoietin in the UK now lags embarrassingly behind the rest of Europe. The reason for this may be partly the magnificent job that the British Blood Transfusion Services have done in maintaining the noble altuism of the donors, but it is also because blood as a therapy is perceived to be safe, readily available and free. None of these is entirely true, least of all the cost of blood but this latter perception makes alternatives seem unduly costly. Until the funding for transfusion is placed in the hands of the users and until savings made at this point can be traded against the cost of alternatives, patients will continue to be unnecessarily exposed to allogeneic transfusions. Yours sincerely Dr Ivor Cavill Senior Lecturer in Haematology, University of Wales College of Medicine, Heath Park, Cardiff. References: 1. McGill N, O'Shaughnessy D, Pickering R, Herbertson M and Gill R. Mechanical methods of reducing blood transfusion in cardiac surgery: randomised controlled trial. BMJ 2002; 324, 1299-1302 (1 June) 2. Berger A. Science commentary; why is it important to reduce the need for blood transfusion and how can it be done? BMJ 2002; 324, 1302- 1303 (1 June) 3. Bexon M (2002) European blood use survey. Personal Communication. I declare a competing interest: advocacy of alternatives to transfusion.