Maternal Kell blood group alloimmunization.

BACKGROUND: Two recent paper have provided conflicting views regarding the severity of Kell hemolytic disease of the newborn.
METHODS: We reviewed our experience during 1944-1990 with pregnant Kell-alloimmunized Manitoban women and similar women referred from outside of Manitoba.
RESULTS: Between 1944-1990, 311 Kell-immunized Manitoban women had 459 pregnancies, of which 63 ended in abortion or stillbirth unrelated to anti-Kell. Of the infants born, 376 were unaffected and 20 were affected. Twelve did not require treatment; two needed phototherapy, one required a simple transfusion, and one an exchange transfusion. One died of kernicterus and three were hydropic and died; all four deaths occurred between 1948-1954. Fourteen Kell-immunized women with 16 pregnancies were referred from outside Manitoba. Eleven had a history of Kell hydropic fetuses and ten had hydropic fetuses at referral. Five of the hydropic fetuses survived and five died. Five women had Kell-negative infants correctly predicted by amniocentesis (two) and by fetal blood sampling (three). Serial amniotic fluid delta OD 450 readings were 83-89% accurate in predicting the presence and severity of Kell hemolytic disease. Life-threatening inaccuracies occurred, primarily in the early and middle second trimester.
CONCLUSIONS: Kell hemolytic disease, although rare, may be as severe as Rh(D) hemolytic disease when it does occur. When there is a history of hydrops or the father is Kell-positive and the maternal anti-Kell indirect antiglobulin titer is 8 or greater, amniocentesis should be performed at 16-20 weeks' gestation. Fetal blood sampling followed by fetal intravascular transfusion is indicated if delta OD 450 readings approach the 65% level in modified zone 2 of Liley or if amniocentesis is precluded because of an anterior placenta and there is a history of hydrops or ultrasound evidence of fetal hemolytic disease.