Immunological evaluation of patients with recurrent abortion

In a prospective study, we performed immunological tests in patients with recurrent abortion. Nine couples with two or more fetal losses of no apparent cause were selected as the patient group, and nine volunteer couples with at least two children and without a history of abortion were used as controls. The frequency of major histocompatibility complex (human leukocyte antigens, HLA) antigen sharing was determined by serological methods. antipaternal antibodies by microlymphocytotoxicity, lymphocyte phenotypes (CD4, CD8, CD19, CD16, CD56 and HLA-DR positive cells) by flow cytometry and natural killer (NK) cytotoxicity by Cr-51 release. NK activity was correlated to the degree of HLA-C sharing and to the percentage of CD16+ and CD56+ cells and to progesterone levels measured by radioimmunoassay. No difference in class I or class 11 HLA antigen sharing was detected between couples with and without recurrent abortion. Antipaternal antibodies were not found in the serum of any woman of the study. A higher absolute number of CD8+ cells (P = 0.01) and a trend to increased CD19+ cells (P = 0.05) were observed among patients. NK activity did not differ between the two groups when expressed as specific cytotoxicity and it was reduced among patients with recurrent abortion when expressed as lytic units/10(7) cells (P = 0.04). There was correlation between NK activity and the percentage of CD16+ and CD56+ cells but not with progesterone levels in patients with recurrent abortion. Our data suggest that an increased NK activity may not play a role in the occurrence of repeated abortion. On the other hand, an increase in circulating CD8+ T cells was observed in patients suggesting that antifetal cytotoxicity in recurrent abortion may be mediated by T cells and not by NK cells. (C) 2002 Elsevier Science Ireland Ltd. All rights reserved.