Rapid Responses to:

PAPERS: Fiona A M Regan, Patricia Hewitt, John A J Barbara, and Marcela Contreras Prospective investigation of transfusion transmitted infection in recipients of over 20 000 units of blood BMJ 2000; 320: 403-406 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Transfusion risk underestimated.

Chris Cowie   (2 March 2000)

Systematic underestimation of transfusion risks?

Gregor Caspari   (8 March 2000)

SCREENING FOR HTLV-I - WHY WAIT ?

Tony Pagliuca   (27 March 2000)

Transfusion risk underestimated. 2 March 2000
Chris Cowie, Epidemiology and Public Health University of Newcastle Upon Tyne Send response to journal: Re: Transfusion risk underestimated.	Editor, Regan estimates the risk of Transfusion Transmitted Infection (TTI). We are concerned about the lack of follow up of those patients (30%) who were excluded due to illness. This group would surely be at a greater risk of having a TTI than healthy individuals. The study's inability to include these patients reduced its power and may have biased the results. Analysis techniques were changed during the course of the study, introducing another source of bias. The possible effects of this change were not addressed in the discussion and inconsistencies in follow up time were not explained. The study criteria excluded those under 18 but the age range was stated as 16 - 96. This raises the issue of informed consent, which was not addressed in the paper. It is likely that the methods used resulted in an underestimation of the risk of TTI. This undermines the reassurance given by the authors on the safety of transfusions. C.J.A.Cowie, R.M.Dixon, S.J.Gonsalves, J.H.Lloyd, S.J.Parkin. Stage 3 Medical Students.
Systematic underestimation of transfusion risks? 8 March 2000
Gregor Caspari, research fellow Institute of Clinical Chemistry, Justus Liebig University, Gaffkystr. 11, 35392 Giessen, Germany Send response to journal: Re: Systematic underestimation of transfusion risks?	There are three different ways to estimate the residual risk of infections transmitted by blood transfusion: (1) follow-up of blood recipients, (2) screening donated blood with more sensitive tests than routinely used, e.g. nucleic acid amplification tests, and (3) by estimating the number of undetectably infectious units from seroconversions of repeat donors. In their prospective follow-up of 5,579 out of 9,220 patients Regan et al. did not detect a single transfusion-transmitted infection, but in view of estimated risks of lower than 1 in 100,000 for HBV and HCV transmission and lower than 1 in 1 million for HIV transmission such a result was not unexpected. The estimate would change dramatically if only one of the 657 participants not followed-up because of death or of the 2,734 participants not followed-up because they were too ill to participate died or was too ill because of transfusion-transmitted infection. Unfortunately, all methods estimating the number of undetectably infectious units on the basis of repeat donor's seroconversion underly the same sort of possible bias: If the donor is too ill to come back for a further donation (or is diagnosed with clinically apparent infection and told to stop donating blood) seroconversion is not detected by the blood center and the residual risk systematically underestimated. Even estimations on the basis of PCR results tend to underestimate the risk as Schüttler et al. could recently show (Lancet, Jan 1, 2000) that also PCR-negative blood preparations are able to transmit HCV infection. One can only estimate a correction factor for this possible bias if one assumes that donors have no preference to donate immediately after infection when PCR is still negative. In conclusion, all methods for estimating the residual risk of blood transfusion with respect to infection contain a possible bias which may result in underestimation of the true risk.
SCREENING FOR HTLV-I - WHY WAIT ? 27 March 2000
Tony Pagliuca Send response to journal: Re: SCREENING FOR HTLV-I - WHY WAIT ?	EDITOR - Two recent articles1,2 on the risk and prevalence of HTLV-I provide differing viewpoints regarding the importance of this virus to the community. Data from the transfusion service suggest that with a maximum incidence of 1:19344 donations3 this virus does not, with its low transmission risk, pose a threat to the blood supply. The study by Regan et al1 is underpowered to detect a real risk of HTLV-I as only 5579 patients were followed and only 21923 units of blood were followed up. Furthermore this paper probably underestimates the risk which is underscored by the ante-natal screening paper by Donati et al2. In this retrospective screen of 8656 women they found a positive result in 0.39%. When one looked at Black African and Black Caribbean women this figure ranged from 0.7 - 1.51%. These women risk transmitting the virus to both their partners (sexual) and their children (breastfeeding)4. This patient group form a substantial percentage of the local south London population and have in the past been specifically targeted to become blood and bone marrow transplant donors. The risk is likely to increase unless policy changes are made. The majority of prevalence data relates to major inner city areas with the belief that this is not a problem outside these areas. Modern lifestyles require mobility and patients with HTLV-I related disease, ie adult T-cell leukaemia/lymphoma and tropical spastic paraparesis, are starting to present in non-urban areas as the case below exemplifies. Case History A 33 year old women was admitted to Aberdeen Royal Infirmary with a one month history of abdominal pain and increasing abdominal distension. She had lost a half a stone in weight and suffered drenching night sweats. On examination she had widespread 1-3 cm lymphadenopathy in all peripheral lymph node groups, bilateral pleural effusions and 8cm of hepatomegaly. Investigations revealed HTLV I positivity and histology confirmed ATLL of lymphoma subtype. This lady was born in Jamaica and moved to Scotland in 1991 after marrying a local man. She is currently undergoing treatment. Family testing has not been carried out. Regan et al summarise by saying "current risks of known transfusion transmitted infections in the United Kingdom are very small, though the possibility of new unrecognised agents always remain." We would suggest that there are well known risks from a known virus which our European and N American colleagues are already testing for. Universal leucodepletion of the blood supply is costing tens of millions to theoretically reduce the risk of nvCJD. Let's screen for what we already know about and attempt to obviate the risk to the blood supply completely. Tony Pagliuca consultant haematologist Dominic Culligan consultant haematologist Henry Watson consultant haematologist Departments of Haematological Medicine, King's College Hospital, London, SE5 8RX and Aberdeen Royal Infirmary, Aberdeen, AB25 2ZD References 1. Regan FAM, Hewitt P, Barbara JA, Contreras M et al. Prospective investigation of transfusion transmitted infection in recipients of over 20000 units of blood. BMJ 2000;320:403-6. 2. Donati M, Seyedzadeh H, Leung T, Blott M, Zuckerman M. Prevalence of antibody to human T cell leukaemia/lymphoma virus in women attending antenatal clinic in southeast London; retrospective study. BMJ 2000;320:92 -3. 3. Brennan M, Runganga J, Barbara JAJ, Contreras M Tedder RS, Garson J et al. Prevalence of antibodies to human T cell leukaemia/lymphoma virus in blood donors in north London. BMJ 1993;307:1235-9 4. Pagliuca A, Pawson R, Mufti GJ. HTLV-I screening in Britain. BMJ Editorial comment The patient has given her written informed consent to publication of her case.