Rh disease may get all the headlines, but immunologic reactions to the Kell antigen can be just as lethal to a K1-positive fetus-if not more so.
Understanding what causes Rh disease and developing an effective way to prevent the disorder is one of the genuine success stories of modern medicine. In fact, the administration of Rhesus immune globulin before and after delivery has saved countless lives and led to a major reduction in the incidence of Rhesus alloimmunization in pregnancy. Unfortunately, immune globulins to prevent maternal antibody formation to other red cells antigens have yet to be developed. A case in point is the Kell (K1) antigen, which has become a growing concern to clinicians as it, too, contributes to hemolytic disease of the fetus/newborn (HDFN).
In the mid 1960s, studies indicated that Kell antibodies occurred in 1.6 per 1,000 reproductive-aged women.1 In 1995, that number had doubled to 3.2 cases per 1,000 women-for reasons that are still unclear.2 And while countries like the Netherlands and Australia routinely administer Kell negative blood to female children and women of reproductive age when a transfusion is required, the United States is still a little behind the times. Here red blood cells are routinely cross-matched only for the ABO and RhD blood groups. It's no surprise then that two thirds of American women with Kell antibodies can trace their sensitization to a history of previous blood transfusion.3 In fact, in most large series, antibodies to the Kell antigen account for approximately 10% of fetuses with HDFN who undergo intrauterine transfusions.4
The Kell blood group system consists of two proteins-Kell and XK. The Kell protein is structurally similar to the protein family of zinc neutral endopeptidases, indicating that it probably plays an important role in red cell growth and differentiation. More than 23 different RBC antigens belong to this system. Individual antigens in the system are designated by name, letter abbreviation, or number. Antibodies to at least nine of the antigens have been linked with fetal anemia. The most immunogenic of these is the KEL1 or K1 antigen. Additional antibodies that have been reported to contribute to HDFN include –cellano (k, K2), -Penny (Kpa, K3), -Rautenberg (Kpb, K4), -Peltz (Ku, K5), -Sutter (Jsa, K6), -Matthews (Jsb, K7), -Karhula (Ula, K10), and -K22.5
More than 91% of Caucasians and 98% of African-Americans are Kell (K1) negative; these individuals are designated kk phenotype. Kell-positive individuals are designated either Kk or KK depending on their zygosity. In those with a Kell-positive blood type, 98% of Caucasians and virtually 100% of African-Americans are heterozygous. Serologic testing with anti-K and anti-k reagents can accurately determine zygosity in the case of a K1-positive father. Therefore, if 9% of male partners are Kell positive and the majority of these are heterozygous-which translates into a 50% risk for an affected fetus-the overall risk for an affected fetus in a Kell alloimmunized pregnancy, when the paternal Kell type and zygosity are unknown, is only 4.5%.
Most maternal RBC antibodies, including anti-D, cross the placenta and attach to fetal red cells that exhibit the specific antigen. Once these sensitized cells pass through the fetal spleen, they are filtered out of the circulating RBC pool and broken down by reticuloendothelial cells. The free hemoglobin that is released by this process is then converted to bilirubin. The fetal bone marrow responds to the resulting anemia by increasing the number of early red cell precursors (reticulocytes and erythroblasts) that are released into the fetal circulation.
The Kell antibody seems to cause fetal anemia through a two-prong attack on fetal red cells. Like anti-D, sensitized fetal red cells are sequestered in the fetal spleen. However, an additional mechanism involves suppression of red cell production so that there is an inadequate response to the fetal anemia. Laboratory studies have confirmed a direct effect of anti-Kell antibodies on developing red cells.6 In addition, blood from fetuses in Kell alloimmunized pregnancies contain fewer circulating reticulocytes and lower levels of serum bilirubin, when compared to fetuses with anti-D antibody induced HDFN.7
Because there is a difference between the mechanism of fetal anemia in cases of Kell alloimmunization and the mechanism behind the anemia of Rh disease, experts disagree about what level of maternal antibody should prompt fetal surveillance. A critical titer of 32 (dilution: 1:32) has been suggested for anti-D. Various researchers have advocated that a lower maternal titer be used for anti-Kell antibodies. Some recommend a maternal titer of 2 while others say 8.8,9 One recent series found that a titer of 32 identified all fetuses that were severely anemic in response to Kell antibodies.3 It therefore seems that a maternal indirect Coombs titer of 8 or greater is a reasonable threshold for concern regarding fetal anemia.
In the past, amniotic fluid analysis for bilirubin, an indirect measure of the degree of fetal hemolysis, was used to follow the condition of the fetus once a critical maternal titer had been detected. These data were measured as the ΔOD450 and plotted on the Liley curve, which took into account changes in values based on advancing gestational age. The role of the Kell antibody in the suppression of red cell production has led several experts to express concern regarding the use of the Liley curve to assess the presence of fetal anemia. Cases of low levels of ΔOD450 followed by the rapid development of fetal hydrops have been reported.10,11 This has led some to advocate that a ΔOD450 in the 65th percentile of Zone 2 of the Liley curve should be used as a threshold for instituting ultrasound-directed umbilical cord sampling to determine if the fetus is anemic.8 Others question whether amniocentesis should be used at all in lieu of the predictability of the middle cerebral artery Doppler-which we discuss below.
Nevertheless amniocentesis still has a role in determining the fetal Kell blood type if the paternal phenotype indicates a heterozygous state. DNA analysis of amniotic fluid can be undertaken to determine if the fetus carries the Kell gene.12 If found to be Kell negative, the fetus is not at risk for HDFN due to the maternal antibodies, and the pregnancy can be considered routine.
Measurement of the peak systolic velocity in the fetal middle cerebral artery (MCA) using Doppler ultrasound has now gained widespread acceptance for the detection of fetal anemia. Values of greater than 1.5 multiples of the median (MoM) have a sensitivity of 88% and a specificity of 82%, superior to the use of serial amniocentesis for analysis of ΔOD450.13 Advantages are related to the noninvasive nature of this modality-no risk for premature rupture of membranes or enhanced maternal sensitization. Although most data published to date have studied the MCA Doppler in Rh pregnancies, several series have found it to be equally effective in detecting anemia in pregnancies complicated by Kell alloimmunization.14,15
Once we know that a fetus is at risk for anemia based on the MCA Doppler, cordocentesis with Kell negative blood readied for intrauterine transfusion should be undertaken. Data from one large European referral center has found a lower rate of perinatal survival in HDFN due to Kell antibody as compared to Rhesus antibody, 58% versus 89%, respectively.16
Once a maternal antibody is detected, clinicians should request an indirect Coombs titer. Paternal testing should be undertaken to see if the patient's partner carries the Kell gene. In the majority of cases, the patient's antibody will have been the result of a previous blood transfusion and her partner will be Kell negative. In these situations, as long as paternity is assured, you have no reason to do further testing because the fetus will not be at risk for HDFN. If paternal testing returns positive, request zygosity from the blood bank; in most cases, you will find a heterozygous state (phenotype: Kk).
Maternal titers should be repeated every month until approximately 28 weeks' gestation, at which point the testing interval should be increased to every 2 weeks. If a value of 8 or greater in conjunction with a heterozygous paternal phenotype is found, you'll want to do an amniocentesis to determine the fetal Kell type. A Kell negative fetus means no further testing is needed. A Kell positive fetus (or a rare homozygous paternal phenotype) requires referral to a perinatal center for serial MCA Dopplers every 1 to 2 weeks. A value of more than 1.5 MoM indicates the need for cordocentesis and probably intrauterine blood transfusion, as well.
In Kell-alloimmunized women with a previous history of an affected fetus or infant, early referral to a perinatal center is warranted. The maternal titer is not useful in these pregnancies as fetal disease usually occurs earlier in gestation than in the previous pregnancy. Serial MCA Dopplers can be initiated as early as 18 weeks' gestation once a positive fetal Kell type has been confirmed by amniocentesis. Intrauterine transfusion can usually be successfully accomplished by 20 weeks.
Although Kell alloimmunization occurs less frequently than Rhesus disease, more severe fetal anemia can be expected due to this antibody's unique ability to suppress fetal erythropoiesis. Early referral to a perinatal center should be considered because perinatal survival declines even when advanced fetal surveillance and therapeutic strategies are employed.
1. Queenan JT, Smith BD, Haber JM, et al. Irregular antibodies in the obstetric patient. Obstet Gynecol. 1969;34:767-771.
2. Geifman-Holtzman O, Wojtowycz M, Kosmas E, et al. Female alloimmunization with antibodies known to cause hemolytic disease. Obstet Gynecol. 1997;89:272-275.
3. McKenna DS, Nagaraja HN, O'Shaughnessy R. Management of pregnancies complicated by anti-Kell isoimmunization. Obstet Gynecol. 1999;93:667-673.
4. van Kamp IL, Klumper FJ, Bakkum RS, et al. The severity of immune fetal hydrops is predictive of fetal outcome after intrauterine treatment. Am J Obstet Gynecol. 2001;185:668-673.
5. Daniels G. Blood group antibodies in haemolytic disease of the fetus and newborn. In: Hadley A, Soothill P, eds. Alloimmune Disorders in Pregnancy Anaemia, Thrombocytopenia, and Neutropenia in the Fetus and Newborn. Cambridge: Cambridge University Press, 2002:21-40.
6. Vaughan JI, Manning M, Warwick RM, et al. Inhibition of erythroid progenitor cells by anti-Kell antibodies in fetal alloimmune anemia. N Engl J Med. 1998;338:798-803.
7. Weiner CP, Widness JA. Decreased fetal erythropoiesis and hemolysis in Kell hemolytic anemia. Am J Obstet Gynecol. 1996;174:547-551.
8. Bowman JM, Pollock JM, Manning FA, et al. Maternal Kell blood group alloimmunization. Obstet Gynecol. 1992;79:239-244.
9. van Wamelen DJ, Klumper FJ, de Haas M, et al. Obstetric history and antibody titer in estimating severity of Kell alloimmunization in pregnancy. Obstet Gynecol. 2007;109:1093-1098.