<p><b>OBJECTIVES</b>To evaluate the three different
methods in
monitoring the
lamivudine-resistant HBV mutants in
lamivudine-treated
patients with
chronic hepatitis B.</p><p><b>
METHODS</b>The
sensitivity and specialty of
melting curve assay and
polymerase chain reaction microplate
nucleotide hybridization-
enzyme linked immunosorbent assay (PCRmnh-
ELISA) were compared with those of mismatch
polymerase chain reaction-
restriction fragment length polymorphism (mPCR-
RFLP) and
sequence analysis, through
detection of HBV YMDD mutants in 44
serums from
chronic hepatitis B patients receiving
lamivudine monotherapy at the
time of viral breakthrough.</p><p><b>RESULTS</b>mPCR-
RFLP assay was more sensitive (10(4) copies/ml) than both PCRmnh-
ELISA (10(5) copies/ml) and
melting curve assay (10(6) copies/ml). 26 YMDD mutants and 18 wild-types were determined by the means of mPCR-
RFLP. Among the 26 mutants, only 16 and 18 mutants were found by
melting curve assay and PCRmnh-
ELISA, respectively. Whereas, out of the 18 wild-types, 2 and 13 mutants were detected by
melting curve assay and PCRmnh-
ELISA, respectively. To confirm the different results determined by the three
methods in 16 samples,
sequence analysis was conducted and showed that the rate of consistency with sequencing was 93.8% by mPCR-
RFLP, 43.8% by
melting curve, and 18.8% by PCRmnh-
ELISA, respectively (chi2=18.7, P<0.01).</p><p><b>CONCLUSIONS</b>The mPCR-
RFLP assay is reliable to monitor HBV YMDD
mutations.
Melting curve assay and PCRmnh-
ELISA should be further improved to increase their
sensitivity and specialty.</p>