Objective: To determine whether preferential X-chromosome inactivation (P-X
CI) relates to idiopathic recurrent pregnancy loss.
Design: A retrospective study.
Setting: Infertility clinics and laboratory.
Patient(s): Women with idiopathic recurrent pregnancy loss (group I), women
who had given birth to children but with no history of spontaneous abortio
n (group II), and women without a history of pregnancy (group III)
Intervention(s): DNA samples from the heterozygotes for the (CAG)n polymorp
hism within the androgen receptor gene were modified with sodium bisulfite,
PCR-amplified with primer pairs for methylated androgen receptor alleles (
M-PCR) and unmethylated alleles (U-PCR), and subjected to electrophoresis.
Main Outcome Measure(s): Band peak patterns and peak area sizes.
Result(s): In group I, 7 (16.7%) of 42 heterozygotes exhibited P-XCI; four
possessed single-peak patterns in the M-PCR and U-PCR products, and three h
ad two-peak patterns in which the peak sizes differed considerably. In grou
p II, 2 (5.6%) of 36 heterozygotes exhibited P-XCI as determined by the two
-peak patterns. In group III, none of the 47 heterozygotes exhibited P-XCI.
Conclusion(s): The incidence of P-XCI was statistically higher in group I t
han in the other groups. As P-XCI characterized by single-peak patterns was
observed only in group I, such patterns, which may result from undiscovere
d cytogenetic or molecular abnormalities of the X-chromosome, likely correl
ate with pregnancy loss. (C) 2001 by American Society for Reproductive Medi
cine.