A colleague in Philadelphia would like to activate a discussion of the paper by Nichols WG et al. from Seattle (Blood online Jan. 16, 2003) on transfusion-transmitted CMV after receipt of leukoreduced blood components. (In essence, this study showed that red cells leukoreduced by filtration were not fully effective in preventing post-transfusion cytomegalovirus infection.) Specifically, have these data led to changes in practice at any institution that previously considered leukocyte-reduced cellular blood products "equivalent" to those made from CMV-seronegative donors. (See earlier discussion on this forum.

The following responses were received.

ADDENDUM June 1, 2003

1.One of the authors of the paper under discussion provided the following insights into the impact of the paper on their local practice (verbatim) “Our overall strategy has not changed; seronegative products remain preferred for this indication in our stem cell transplant recipients. Leukoreduced components, however, are extremely valuable when these components are not available, and are thus acceptable (but not, in our estimation, equivalent). As discussed in our paper, however, we believe that the most effective method for preventing the complications of primary CMV infection in patients at risk is to monitor patients for primary infection and treat them preemptively, given that primary CMV infection still occurs among patients treated with only seronegative components. Many resist this strategy, stating that TT-CMV is not a problem at their center. I'm not so sure. Seronegative recipients of seronegative transplants represent a minority of most center's patients (approximately 20% at our center, substantially lower at others). Given the small number of such patients transplanted (and a ~2% rate of CMV disease among those treated with leukoreduced blood), it may take years for a center to see a single case of fatal CMV disease, which is easily dismissed. The message? If universal leukoreduction is applied in lieu of CMV seronegative products (and patients are not screened and treated preemptively), the rate of disease is likely to increase among patients at high risk (a population that potentially include neonates or patients with AIDS). Our study is by no means definitive, but should be regarded as the proverbial canary in the coalmine.

ADDENDUM June 2, 2003

2. Neil Blumberg MD, Professor of Pathology & Laboratory Medicine and Director, Transfusion Medicine/Blood Bank University of Rochester Medical Center (attribution used with permission) reports the following opinion (verbatim): "We continue to use filtration leukoreduction (universal in our case) as our only means of CMV risk reduction, and have done so for several years. We've used it in solid organ transplants for about 13 years with good results. We also give only ABO-identical platelets to our stem cell transplant patients, which may reduce the risk of CMV reactivation; in my opinion this has a much great impact on mortality and morbidity than does CMV transmission in the era of ganciclovir, etc. We've seen perhaps 2 or 3 cases of CMV disease in the last thousand transplants, all in patients with CMV-positive stem cell donors or who were CMV-positive themselves prior to transplant.

My take on the data is that there is extensive evidence that leukoreduction and CMV serotesting used independently lower the risk of infection to about 1%, give or take. The new data are from a cohort study, so it is entirely possible that changes in infectivity through conventional transmission means (i.e., not transfusion) account for the small, but statistically significant differences seen in the two cohorts. In other words, CMV is a very infectious virus, which explains the 50-80% typical prevalence of CMV positivity in the adult population. Small changes in prevalence due to mini-epidemics in the general population could easily account for a 1-2% increase in seropositivity of transplant recipients that would be coincident with but not due to transfusion. Infections might occur environmentally through asymptomatic but recently infected staff, visitors, etc. I would point out that the cohorts also differ in that apheresis, process leukoreduced transfusions increased during the same time as the increase in infection rate. To my knowledge no one has ever shown the equivalence of filtration leukoreduced and process leukoreduced platelets in CMV risk reduction. The original studies were done exclusively with filtration leukoreduced products (which is also what we use exclusively). Finally, as the authors state, no one became ill as they were monitoring patients and instituted anti-CMV therapy in those who became infected. This is a common approach in most transplant centers, at least with high-risk patients such as allo-transplant recipients. Thus on balance, I feel quite comfortable with continuing to use filtration leukoreduction as our exclusive risk reduction technique for CMV, given our track record and the overall data in the literature."

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator