The effect of leukocyte-reduction method on the amount of human cytomegalovirus in blood products: a comparison of apheresis and filtration methods

This study examined the effectiveness of 3 leukocyte-reduction (LR) methods in depleting the residual level of cytomegalovirus (CMV) in blood products measured by quantitative polymerase chain reaction (QA-PCR). At 2 locations over 3 allergy seasons, apheresis platelets and whole blood were collected from 52 healthy CMV seropositive subjects having an elevated titer of CMV DNA (median = 2400 genome equivalents [GE]/mL) resulting in 32 evaluable LR apheresis platelets, 31 filtered platelets from whole blood, and 31 filtered red blood cells (RBCs) from whole blood. Leukoreduction by apheresis and filtration resulted in substantial reduction of detectable CMV DNA Levels with 99.9% of the LR products expected to have less than 500 GE/mL of CMV DNA. No difference was found between methods (P = .52). CMV genomic leukocyte subset localization was determined by QA-PCR of fluorescence-activated cell sorter (FACS)-sorted peripheral blood from 20 seropositive subjects (n = 10 > 100 GE/mL, n = 10 QA-PCR negative). CMV was detected in monocyte (13 of 20) and granulocyte (3 of 20) fractions. Presence of competent virus in QA-PCR positive (> 100 GE/mL) peripheral blood samples was verified with 4 of 19 subjects positive in shell vial assay, and 8 of 18 positive for CMV gene products (messenger RNA). We observed a seasonal DNAemia variation in seropositive subjects. CMV seropositive subjects (n = 45) entered into longitudinal monitoring in March/April 1999 were QA-PCR negative at baseline. Subjects converted to a positive QA-PCR coincident with increased seasonal allergen levels (Norfolk 15 of 18 evaluable in 43.4 +/- 9.48 days; Denver, 16 of 23 evaluable in 96 +/- 26.3 days). These data demonstrate effective reduction of CMV load by LR during periods of DNAemia in CMV seropositive subjects.