Eugene M. Katzin

The role of iso-immunization in the pathogenesis of erythroblastosis fetalis☆

STUDIES on the cause of intra-group transfusion accidents associated with pregnancy have established the importance of the concept of iso-immunization of the mother by blood factors in the fetus transmitted from the father.1–5 More recently it was found that the same theory of iso-immunization may serve as the basis for a theory on the pathogenesis of erythroblastosis fetalis, the well-described familial hemolytic disease of the newborn.3, 4, 6

Isoimmunization in Pregnancy and the Varieties of Isoagglutinins Observed.

The production of immune isoagglutinins following repeated transfusions, although of common occurrence in some animal species, is very rare in man. The first instance of this sort in man was described by Landsteiner, Levine, and Janes. 1 Levine and Stetson 2 described a case in which an intra-group ag-glutinin was responsible for severe post-transfusion symptoms, with anuria after the first transfusion. This occurred in a woman who, after retaining a dead fetus for a period of about 2 months, finally delivered a macerated fetus. In view of the activity of the agglutinin at 37° C and because of its gradual disappearance, it was believed that the antibody developed as a result of immunization, and it was suggested that the products of the retained dead fetus served as the antigenic stimulus. This suggestion seemed plausible because the cells of the husband, of the same blood group, (who was the donor) were sensitive to the action of the agglutinin. Presumably, the fetus inherited from the father a dominant agglutinogen which was absent in the tissues of the mother, who could thus be immunized. Our own recent experience and other cases from the literature 3 indicate that the isoimmunization-hypothesis may hold also in certain cases of pregnancy with complications other than the retention of the dead fetus. In each of the cases cited, atypical agglutinins could be demonstrated in the absence of a history of repeated blood transfusions. In 5 cases 4 there was a history of varying degrees of toxic symptoms during the pregnancy (in 2 of which macerated fetuses were delivered). In one of these cases, the patient had some fever, and rupture of the membranes occurred a few days before delivery, which was normal but followed by a post-partum endometritis. In 3 patients 5 the agglutinins were observed following abortion, one of them of the septic variety. One patient, whose eighth pregnancy resulted in premature separation of the placenta, had stillbirths in the 4 preceding pregnancies. 6 In 2 of the remaining 3 cases there were histories of repeated miscarriages but the present pregnancies, apparently normal, had to be terminated by Caesarian section.† In one case the pregnancy was normal except for morning vomiting during the last six weeks; due to lack of progress in labor the delivery was by Caesarian section. 7

Temporary Agglutinability of Red Blood Cells

As a result of studies on atypical isoagglutination it was concluded that the atypical feature is an unusual isoagglutinin acting at room temperature (20°-25°) on cells of the same blood group, or on cells of group O. These atypical agglutinins which occur in about 3% of normal individuals, possess a distinct specificity, the great majority describing the subgroups A 1 and A 2 , or the agglutinable substance known as P. Landsteiner and Levine 1 emphasized that the atypical property is a peculiarity, not of the blood cell, which could be readily grouped, but of the agglutinin in the serum. Consequently these atypical reactions do not interfere with the scheme of the 4 blood groups. In the blood to be reported, however, the results indicate that in this instance (a patient of group O) the exceptional property is the agglutinability of the cell which could be demonstrated by its sensitivity to an agglutinin present in about 15% of the sera, irrespective of the blood group. This peculiar property of the cell could be demonstrated during the illness of the patient, but not after recovery. From this it was concluded that the agglutinable property was induced by the specific nature of the illness (soluble products of pneumococcus type I) or by the therapeutic substances employed. Patient A. F., age 4, developed measles on March 17, 1938, and on the following day was admitted to the Willard Parker Hospital of New York City. The patient also showed signs of otitis media, bronchopneumonia, and later also fluid in the right pleural cavity from which pneumococcus type I was cultured. The same organism was also recovered from a pharyngeal swab and from blood cultures.

Atypical Warm Isoagglutinins

Zacho 1 reported an instance of premature separation of the placenta with a transfusion-accident, attributable to an atypical isoag-glutinin active on the donors cells. This agglutinin had the unique property of greater activity at 37° C than at lower temperature. Although Zachos case was the first and only published instance of this sort, it is likely that hitherto antibodies of this character may have been overlooked. In support of this, case reports are presented of 5 patients whose blood recently investigated were found to contain such agglutinins. A brief discussion of the properties and the origin of these agglutinins forms the basis of this communication. One of these cases was observed in a patient† (B.M.) transfused 6 times because of bleeding from a duodenal ulcer. Apparently, this agglutinin was induced as a result of the antigenic stimulus of repeated transfusions. No such history could be elicited in the remaining 4 instances, nor in the case of Zacho, all of which were observed in women suffering from a variety of complications of pregnancy. In one patient (G.B.) who gave a history of 2 miscarriages, the present pregnancy was complicated by death of the fetus during labor. A second patient (H.H.) had 3 consecutive miscarriages and in the present pregnancy delivery was by Caesarian section because of uterine inertia. A third patient (D.D.) who suffered from persistent vomiting from the fourth to the seventh months had fever and premature rupture of the membranes a few days before delivery which was followed by a parametritis. The fourth patient (A.H.) died of an induced septic abortion. (Zachos patient gave a history of 4 consecutive stillbirths preceding the last pregnancy.) The correlation of atypical agglutinins in general with certain complications of pregnancy has been mentioned elsewhere. 2

Pathogenesis of Erythroblastosis Fetalis: Absence of the Rh Factor from Saliva

The pathogenesis of erythroblastosis fetalis has been ascribed to the isoimmunization of the mother by the Rh, or, more rarely, by other factors in the blood of the fetus. 1 , 2 According to this concept maternal agglutinins produced in response to this stimulus after passing the placenta, act on the susceptible blood of the fetus, and thus produce the various syndromes known as erythroblastosis fetalis. This theory does not differ in principle from the older concept of “heterospecific pregnancy” 3 based on the 4 blood groups. Thus, if the property A or B in the fetus is not present in the mother, the maternal agglutinin by its action on the fetal blood was thought to induce icturus gravis, (one of the manifestations of erythroblastosis fetalis). 4 Recently it was shown by Jonsson, 5 and confirmed by Levine, 6 that in the example cited the mothers normal agglutinins may be specifically increased in titer as a result of isoimmunization. Nevertheless, the theory of heterospecific pregnancy was abandoned because of lack of evidence. 7 Furthermore, the demonstration that the factors A and B are present, also, in tissue cells and body fluids indicate that maternal isoantibodies are specifically inactivated after passing the placenta. However, this wide distribution of the A and B blood factors applies only to about 80% of individuals (secretors) and in the remainder (non-secretors) the A and B substances are presumably limited to red blood cells. 8 , 9 , 10 These facts suggested experiments to determine whether the Rh factor is also present in tissue cells or in body fluids. The tests were carried out with saliva because of its high concentration of the A and B substances.

Hemagglutinins of Pneumococcic Antisera.

For a number of years it has been known that some pneumococcic strains possess the Forssman antigen and hence produce in the blood of immunized rabbits hemolysins for sheep blood and agglutinins for blood of Group A. According to a recent report 1 this antigen is present in all strains of pneumococci except those of Types IV, VIB, XI, and XXXI. Prominence was given to these facts only recently on account of the current increased use of therapeutic sera derived from rabbits. We were led to examine the hemagglutinins and hemolysins of all antipneumococcic therapeutic sera available (rabbit and horse), especially in view of recent reports which indicate that horse sera containing strong-titered agglutinins for human blood were responsible for occasional fatal acute hemolytic reactions. 2 , 3 , cf. 4 These accidents were due to injection of horse serum Type XIV. Our own experiments are based on tests done with therapeutic sera administered to patients at the Harlem Hospital. The sera were tested in dilution of 1:50 for hemolysins of sheep blood and for agglutination of human blood of Groups O, A2. A1, and B. Hemolysins for sheep blood were tested by adding 1 drop of 50% suspension of washed sheep-cells to a mixture of 0.5 cc of serum-dilution and 0.5 cc guinea-pig complement 1:10. Tests for agglutination were made by mixing in small tubes 2 drops of a 2% washed blood suspension and 2 or 3 drops of serum dilution; readings were made after the tests stood for one hour at room temperature. Our results indicate that antipneumococcic rabbit sera of numerous types contain hemolysins for sheep blood and agglutinins with specific action on human blood of Group A.

A Simple Method for the Concentration of Blood Plasma and Serum

In the course of the preparation of plasma for human use it was observed that on the thawing of frozen plasma, separation into a number of layers takes place; the lowermost deep-yellow layer is rich in proteins, while the uppermost, nearly colorless layer, contains only a small amount of protein. On repeated freezing and thawing, the separation becomes more distinct. A more rapid and sharp separation of plasma into layers is obtained by centrifuging the frozen material. It is interesting that Bujwid 1 used the freezing method for the concentration of diphtheria and tetanus anti-toxin in 1897. His results were confirmed by Ernst, Coolidge and Cook. 2 Rossi 3 used the freezing and centrifuging procedures for fractionating plasma and other colloidal solutions, and Hati 4 used this technic for the concentration of complement, opsonins and anti-sera. In view of the above observations, the question arose as to whether these procedures might be employed for the concentration of plasma without excessive loss of solids. The following is a representative fractionation and analysis of plasma: 100 cc were placed in a separatory funnel and frozen at—10°C; on partial thawing at 4°C, a deep yellow fluid was obtained with a colorless ice block floating on the surface. The freezing and thawing process was twice repeated. During the thawing which followed the third freezing, the melting fluid was collected directly into graduated cylinders. Three fractions were obtained in this manner. The first fraction of 25 cc was deep-yellow; the second was 17 cc, and the third, derived from the colorless ice block, was 58 cc. Quantitative determinations of proteins, albumin, chlorides, cholesterol and isoagglutinins were performed on the 3 fractions and on a sample of the original plasma. The results are presented in Table I.