Cytomegalovirus (CMV) may be transmitted by transfusion of whole blood and cellular components processed according to standard processing procedures. A need exists to develop new procedures to remove CMV and other leukocyte-borne viruses from donor blood. Ten patients (AIDS/bone marrow transplants) who were CMV antigenemic (virus subsequently confirmed by isolation), donated 50 mL of venous blood within 24 to 72 hours of the initial antigen detection. Twenty-five-milliliter aliquots of each specimen were passed through Purecell Neo Neonatal Leukocyte Reduction Filters (Pall, East Hills, NY). The remaining 25-mL nonfiltered aliquots, as well as the blood filtrates, were subjected to infectivity endpoint determinations. The Purecell Neo filter effected a 3 to 4 log10 leukocyte reduction. CMV input titers ranged from less than 10 to 7.3 x 10(1) median tissue culture infectious dose (TCID50) per milliliter. CMV was not isolated from any postfiltration effluent (i.e., leukocytes, erythrocytes, or plasma). CMV DNA was not detected by nested polymerase chain reaction in 8 of 10 postfiltrate blood specimens. The Purecell Neo filter was efficacious in eliminating or significantly reducing viral (CMV) load in venous blood.

ABSTRACT With the availability of anticytomegalovirus (CMV) therapeutic agents, rapid detection of CMV is important in the care and management of the immunosuppressed patient. The PrimeCapture CMV DNA Detection Plate System (PC-PCR) was evaluated for the detection of CMV in blood and cerebrospinal fluid (CSF). The resolution of discordant results was performed by consensus testing utilizing a combination of conventional cell culture (TC-CPE), the CMV-antigenemia (CMV-Ag) assay, one or more in-house CMV nested PCR assays, and/or patient evaluation and follow-up. Of 51 blood specimens from 34 patients, 23 (45%) were identified as true positives. PC-PCR was significantly more sensitive than the CMV-Ag assay, TC-CPE, or a combination of both tests. The sensitivities, specificities, positive predictive values (PPV), and negative predictive values (NPV) for PC-PCR, the CMV-Ag assay, TC-CPE, and a combination of CMV-Ag and TC-CPE were 78, 75, 72, 81%; 46, 100, 100, 70%; 39, 100, 100, 67%; and 58, 100, 100, 73%, respectively. CMV was not detected or isolated in CSF, resulting in a combined PC-PCR sensitivity, specificity, PPV, and NPV of 77, 90, 68, and 93%, respectively. Among those laboratorians considering the incorporation of molecular CMV diagnostics into their clinical microbiology or virology laboratories, the CMV PC-PCR offers a relatively simple-to-perform and sensitive assay system.

One-hundred twenty AIDS patients, tested by an optimized cytomegalovirus antigenemia (CMV-Ag) assay, were followed over a 6-month period in an attempt to correlate viral (CMV) load with severe CMV organ-specific disease. Among patients available for follow-up, CMV organ-specific disease was present in seven of eight (88%) with pp65 antigen levels ≥50 cells per 4 × 105 polymorphonuclear leukocytes. One-hundred six patients of 107 patients with levels <50 pp65 reactive cells did not develop CMV organ-specific disease within our 6-month follow-up period. The CMV-Ag assay, as standardized in our clinical setting, served as a marker and predictor of CMV organ-specific disease in our AIDS patient population.

Genetic variation in glycoprotein B (gB) may play a role in human cytomegalovirus (HCMV) pathogenesis. Using restriction endonuclease digestion and DNA sequencing, a unique gB genotype was identified in eight HCMV strains isolated from five patients with the acquired immune deficiency syndrome. Nucleic acid homology to the four previously described gB genotypes ranged from 79 to 91% for the two major variable regions of gB. Studies of the role of gB in HCMV pathogenesis should recognize the existence of live gB genotypes.

Background: The quantitative cytomegalovirus-antigenemia (CMV-Ag) assay is an important technology in the regimen of tests utilized in the care and management of acquired immunodeficiency syndrome (AIDS) and other immunocompromised patient groups. Performance of this assay is contingent upon the appropriate processing of the polymorphonuclear leukocyte (PMNL) compartment of the peripheral blood. However, a cell input standard in the performance of the CMV-Ag assay, has not been established. Interpretive differences between laboratories utilizing the CMV-Ag assay may reflect this lack of test uniformity.Objectives: To determine the effect of different PMNL concentrations on the quantitation of CMV in peripheral blood. The leukocyte concentration resulting in optimal rates of viral detection, will be compared with the shell vial assay-indirect immunofluorescent assay (SVA-IFA), and conventional tube culture (TC-CPE).Study design: A total of 74 freshly collected blood specimens were tested by the CMV-Ag assay, using cytospin preparations consisting of 2 x 10(5), 4 x 10(5) and in preliminary experiments, 8 x 10(5) PMNLs/slide. Data obtained from these studies were compared to SVA-IFA and TC-CPE. Viral load was monitored among 11 symptomatic patients through sequential testing of these patients at the start of ganciclovir (GCV), foscarnet (PFA), or combination drug therapy.Results: Among 74 blood specimens tested by the CMV-Ag assay, cytospin preparations consisting of 4 x 10(5) compared with 2 x 10(5) PMNLs/slide, affected a mean positive cell increase of 215% (P=0.03). PMNL slide preparations consisting 8 x 10(5) cells produced background levels which prevented accurate reading of slides. The CMV-Ag assay was more sensitive than the SVA-IFA, but equivalent to TC-CPE. Among 11 patients started on drug therapy, viral load was markedly reduced in 8 within 2-3 weeks; three patients (2 deceased within 3 weeks after receiving therapy), showed no decrease in viral load. One patient was identified as harboring a PFA resistant strain.Conclusions: A PMNL concentration of 4 x 10(5) cells facilitated the reading of CMV-Ag assay slide preparations. The modified CMV-Ag assay furthermore, is applicable in the monitoring of viral load for the tracking of susceptible or resistant CMV strains. (C) 1997 Elsevier Science B.V.

To determine if cytomegalovirus (CMV) retinitis occurs more frequently in patients infected with certain strains CMV isolates from the blood of 44 patients with advanced human immunodeficiency virus disease were grouped by the DNA sequence or the restriction endonuclease digest pattern of a portion of the glycoprotein B (gB) gene. Forty-two patients (95%) were followed clinically until the development of CMV retinitis or death. Fourteen (78%; 95% confidence interval, 7%-39%) of 26 with isolates belonging to other gB groups developed CMV retinitis (P = .002). Viremia caused by gB group 2 CMV strains is associated with higher risk of CMV retinitis than viremia due to other CMV gB groups. The association of CMV gB gene with retinitis suggests this gene, or one linked to it, is a virulence factor for CMV strains causing infection in AIDS patients.