Few mostly small, studies have investigated the distribution of HLA class I
I antigens among women with unexplained recurrent miscarriage. Although som
e studies have reported statistically significant associations between this
syndrome and certain HLA-DR antigens-especially the -DR1 and -DR3 antigens
-other studies have been unable to demonstrate such associations. For the p
resent meta-analysis, 18 cross-sectional or case-control studies (published
or unpublished) reporting on frequencies of HLA-DR1 and -DR3 antigens amon
g Caucasian women with unexplained repeated miscarriage were identified by
searching literature databases (MEDLINE and EMBASE), reading the references
of identified studies, and by contacting researchers within the field, The
studies comprised a total of 1508 patients. The methodological quality of
most of the studies was low, especially because of small numbers of patient
s and because patients with only two miscarriages were included in many stu
dies; this is defined as repeated miscarriage. The odds ratios of repeated
miscarriage for the HLA-DR1 and -DR3 antigens were calculated for the indiv
idual studies and subsequently the pooled adds ratios for the studies were
calculated, The combined odds ratio for HLA-DR1 was 1.29 [95% confidence in
terval (CI) = 1.05-1.58] (17 studies) which is statistically significant (P
< 0.05), The combined odds ratio for HLA-DR3 was 1.00 (95% CI 0.80-1.24) (
18 studies), which is not significant, The results of the meta-analysis sug
gest that the HLA-DR antigen DR1 is associated with an increased susceptibi
lity to unexplained repeated miscarriage.