Article Text

  1. H. Hakim1,
  2. C. Gibson2,
  3. J. Pan3,
  4. K. Srivastava3,
  5. Z. Gu2,
  6. M. J. Bankowski4,
  7. R. T. Hayden2
  1. 1Department of Infectious Diseases
  2. 2Department of Pathology
  3. 3Biostatistics, St. Jude Children's Research Hospital, Memphis, TN
  4. 4Microbiology Department, Diagnostic Laboratory Services, Inc., Honolulu, HI


Background Infections with Epstein-Barr virus (EBV) have been associated with a broad spectrum of disease in both immunocompetent and immunocompromised patients. Although quantification of EBV nucleic acid in the peripheral blood has been demonstrated to be useful in diagnosis and patient care, the best sample type and reporting format for such testing remain uncertain.

Methods Using quantitative real-time PCR (QRT-PCR), we evaluated EBV viral loads in whole blood (WB), peripheral blood mononuclear cells (PBMCs), and plasma in 249 samples from 122 patients received for clinical testing from September 2002 to September 2005. Samples were tested blindly and in duplicate. In WB and PBMCs, results were quantified both in viral copies/mL and in a normalized ratio to input genomic DNA (copies/μg). A single-copy human gene (RNase P) was quantified and used to normalize viral quantity to genomic DNA in each sample. Only one random sample per patient was included in the pairwise analysis; 122 pairs of WB and PBMCs, 79 pairs of WB and plasma, and 79 pairs of PBMCs and plasma samples were used for analysis. In patients with three or more samples, trending of quantitative values seen over time was compared among the different sample types.

Results The sensitivity of QRT-PCR using WB and PBMCs did not differ significantly (p = .3323), and both were more sensitive than plasma (p < .0001). EBV viral load results of all sample type pairs were significantly (p < .05) correlated by regression analysis, with the correlation strongest for paired WB and PBMC samples (r2 .8360) when reported in copies/mL. Regression analysis also showed strong correlation between the two reporting units copies/mL and copies/μg DNA for both WB and PBMCs (r2 > .91). When reported in copies/mL, EBV viral loads detected using WB, PBMCs, and plasma trended very closely in patients with multiple samples, particularly for WB and PBMCs. Similarly, within-patient trends for both reporting units copies/mL and copies/μg DNA tracked very closely.

Conclusion WB and PBMCs show comparable sensitivity and close correlation of quantitative results when assayed for EBV by QRT-PCR. However, WB may have advantages as a sample type when compared with PBMCs since it is readily available, requires a smaller volume of blood drawn from the patient, and requires fewer processing steps. The tight correlation of copies/mL and copies/μg DNA suggests that normalization to cell number or genomic DNA may not be necessary.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.