Detailed analysis of peripheral blood natural killer (NK) cells in women with recurrent miscarriage

Increased peripheral blood natural killer (NK) cell activity has been associated with unexplained reproductive failure including recurrent (three or more) miscarriages (RM). Studies have reported abnormalities in both numbers (absolute and proportion) and activation. This study assessed numerous NK cell parameters to determine which (if any) are altered in women with RM compared with controls, which parameter best differentiated women with RM from controls, and what NK levels should be considered high. Luteal-phase blood samples from women with RM (n = 104) and controls (n = 33) were analysed by four-colour flow cytometry. NK cells were analysed as a percentage of lymphocytes, total NK concentration, CD56(Dim) subtype concentration and percentage, activated CD69(+)CD56(Dim) subtype concentration and percentage and CD56(+Bright):CD56(+Dim) subtype ratio. Women with RM were analysed in two subgroups: those positive in >= 1 RM screening tests (karyotype, uterine, antiphospholipid syndrome, thrombophilia) (n = 48) and those who had negative screening tests (n = 56). Women with RM had significantly higher NK percentage (P < 0.001), and significantly lower CD56(+Bright):CD56(+Dim) ratio (P < 0.05) than controls. NK percentage was the only significantly higher variable in the RM screening test negative subgroup (P < 0.01). A ROC analysis (AUC = 0.71) found that an NK percentage > 18% was highly specific for women with RM (97.0%), and defined 12.5% of women with RM as having high NK percentage, compared with 2.9% of controls. Women with RM have altered peripheral blood NK parameters. NK cells as a percentage of lymphocytes best discriminated RM and control populations. Women with RM and high NK levels may have an immunological disorder.