Pr. Field et De. Dwyer, DIFFICULTIES WITH THE SEROLOGIC DIAGNOSIS OF INFECTIOUS-MONONUCLEOSIS- A REVIEW OF THE RCPA QUALITY ASSURANCE PROGRAMS, Pathology, 28(3), 1996, pp. 270-276
The Royal College of Pathologists of Australasia's Quality Assurance P
rograms (QAP) on serologic assays for Infectious Mononucleosis (IM) ha
ve identified a number of important problems in laboratory diagnosis o
f this condition. A wide range of assays for the diagnosis of acute Ep
stein-Ban virus (EBV) infection are available, although heterophile an
tibody tests are still the most frequently performed procedures for th
e diagnosis of IM. In 1993, eighty-six (54%) of the 159 participating
laboratories performed only a heterophile test; of these, 71% did not
provide an interpretation of their results and none mentioned the need
for confirmatory EBV-specific antibody testing. This revealed a lack
of appreciation that heterophile tests should only be used for screeni
ng, due to their inferior sensitivity of less than 50% in children and
80-90% in adults and specificity of 95% compared to EBV-specific assa
ys. For these reasons the use of heterophile tests is discouraged. Alt
hough EBV-specific serology has traditionally been done by immunofluor
escence (IF), the use of reliable ELISA methods using purified EBV-ant
igens is increasing. False negative EBV VCA IgM ELISA test results wer
e obtained by laboratories using unpurified or unspecified VCA antigen
s. As only five laboratories used these tests in the 1992 QAP, the tes
ts' true performances could not be properly assessed, suggesting the n
eed for independent studies. Variable results were obtained by laborat
ories using commercial EBNA IgG assays (in addition to EBV IgM) sugges
ting that an assessment of these kits would also be worthwhile. The QA
P reaffirm that reliable EBV IgM detection is the most useful test for
the diagnosis of IM, even where other EBV markers are also used. In t
he 1992 and 1993 QAP, laboratory performance of both heterophile and E
BV-specific tests was better with non-reactive serum specimens (99% co
rrect) than with reactive serum specimens (90% correct).