Send to

Choose Destination
Clin Microbiol Infect. 2001 Apr;7(4):179-86.

Molecular-based strategies for assessment of CMV infection and disease in immunosuppressed transplant recipients.

Author information

Department of Virology, Cardiff Public Health Laboratory, University Hospital of Wales, Cardiff, CF14 4XN, UK.



Molecular assays are now considered to be the "gold standard" for assessment of human cytomegalovirus (CMV) infection and disease in those at risk from severe associated clinical manifestations. There is, however little consistency in the methods used in different centres. This study was undertaken to compare different qualitative molecular-based approaches for assessment of CMV activation from latency in samples from immunosuppressed transplant recipients.


Nucleic acid amplification techniques based on the polymerase chain reaction (PCR) and nucleic acid sequence-based amplification (NASBA) were undertaken for the assessment of CMV replication and associated disease in immunosuppressed transplant recipients. Samples from 32 transplant recipients were tested during this study using three molecular-based strategies: (1) detection of CMV DNA in whole blood extracts (positive after a single round of PCR considered "high-level" positive, N = 55); (2) detection of cell-free CMV DNA in plasma (two methods, N = 55 for each); and (3) detection of late pp67 CMV mRNA after NASBA (N = 51). Results using a commercial pp65 antigenemia assay were available for comparison from 40 samples.


Seven samples were positive for CMV by all methods and 36 were negative by all methods undertaken. The other 12 samples gave discordant results using different molecular methods. The correlation between whole blood "high-level" PCR, NASBA for pp67 mRNA and antigenemia results was generally good. Results presented show that plasma PCR results do not always correlate with methods utilizing whole blood as the substrate and that inhibitors in these samples could be problematic. Whole blood PCR gave more positive results than the other assays but use of a nested assay on whole blood or plasma led to detection of CMV in individuals who had no other indicators of virus replication and who did not develop associated disease (low specificity). Although the number of confirmed CMV disease episodes was low in this study, the problems of low positive predictive value for sensitive, qualitative PCR assays was clearly demonstrated.


Assays based on qualitative detection of viral nucleic acid may provide information useful for management of CMV but caution is necessary when making comparisons between results using different molecular strategies. It remains to be proven in large, comparative clinical studies in which the approach and method give the best balance between sensitivity, specificity and clinical relevance for different patient groups.

[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center