OBJECTIVE. Direct antibody titer–positive, blood group A or B neonates who are born to group O mothers may be at risk for hemolysis and hyperbilirubinemia. Immunoglobulin G1 and immunoglobulin G3 subclasses are associated with increased hemolysis relative to immunoglobulin G2 and immunoglobulin G4. We investigated whether identification of immunoglobulin G subclass 1 or 3 may be predictive of hemolysis and hyperbilirubinemia.
METHODS. Direct antibody titer-positive, blood group A and B neonates born to group O mothers were tested for the presence of immunoglobulin G subclasses 1 and 3 in umbilical cord blood by using a commercially available gel testing technology. By inference, neonates in whom neither immunoglobulin G1 nor immunoglobulin G3 were detected were designated immunoglobulin G2 and/or 4. Mandatory plasma total bilirubin was measured at discharge, and additional measurements performed as clinically indicated. Hyperbilirubinemia was defined as any plasma total bilirubin value >95th percentile for hour of life. Blood carboxyhemoglobin and total hemoglobin concentrations were also measured on the predischarge sample. Measured carboxyhemoglobin, expressed as percentage of total hemoglobin, was corrected for ambient carbon monoxide to derive “corrected carboxyhemoglobin,” a sensitive index of heme catabolism. The corrected carboxyhemoglobin/total hemoglobin ratio was calculated to correct for any differences in total hemoglobin mass between groups.
RESULTS. Eighty-two infants were studied, 18 of whom were designated as immunoglobulin G1, 0 as immunoglobulin G3, and 64 as immunoglobulin G2 and/or 4. The incidence of plasma total bilirubin >95th percentile was similar between the subgroupings. Corrected carboxyhemoglobin values and corrected carboxyhemoglobin/total hemoglobin ratio were also similar between the subgroupings.
CONCLUSIONS. Immunoglobulin G1 was found in 22% of direct antibody titer–positive, group A and B neonates who were born to group O mothers, whereas immunoglobulin G3 was rare. Hemolysis and hyperbilirubinemia could not be predicted by this gel technique that enabled identification of these immunoglobulin G subclasses.
Dear Sir,
We read Dr Sarici's response to our recent paper, "Failure to predict hemolysis and hyperbilirubinemia by IgG subclass in blood group A or B infants born to group O mothers" (1)with great interest and welcome this opportunity to clarify some of the points raised.
With regard to the definition of ABO hemolytic disease, the articles we cited by Sgro et al (2) and Salaas and Mazzi (3) were chosen because they had a high rate of ABO heterospecificity among babies with serum total bilirubin (STB) >25 mg/dL or requiring exchange transfusion, respectively. Surely these criteria should, in the current era, be regarded as extreme and fitting Dr Sarici's description "…the extreme end of ABO heterospecificity"? Unless there were additional etiologic factors, it is most likely that, in the presence of ABO heterospecifity, this condition was the cause of the extreme hyperbilirubinemia, confirming the true diagnosis of this condition.
We used the standard bilirubin nomogram of Bhutani et al (4) in order to make our study useful to clinicians accustomed to using these graphs. These were constructed from STB results obtained from many babies and are therefore widely representative. It stands to reason that in ABO heterospecific, direct antiglobulin test (DAT) positive infants, who have a higher rate of hemolysis than their DAT negative peers, that the upper percentile will be higher than that of pooled infants, in whom only a minority can be expected to have been DAT positive. However, a higher upper percentile does not lessen the danger of bilirubin crossing the blood-brain barrier and infiltrating neural cells. Raising the upper limit of normal to a level of bilirubin higher than the Bhutani et al (4) 95th percentile may be instrumental in introducing complacency to pediatricians' interpretation of STB results, and lowering the degree of physicians' concern when evaluating ABO heterospecific newborns with STB levels bordering on the criteria for commencing phototherapy or performing exchange transfusion. We also wish to point out that the Subcommittee on Hyperbilirubinemia of the American Academy of Pediatrics in its 2004 guidelines recommends using the standard hour of life specific bilirubin nomogram, and does not suggest using a specific nomogram for differing or individual etiologies of hyperbilirubinemia.
With regard to the suggestion that we should have studied a group of ABO heterospecific but DAT negative newborns alongside our study group, we point out that the study was specifically designed to evaluate DAT positive neonates. This is the neonatal population at high risk for increased hemolysis, hyperbilirubinemia and bilirubin induced neurological damage, while DAT negative babies are only rarely jeopardized by severe hyperbilirubinemia.
We did not fail to notice the preponderance of blood group A neonates among those with the IgG1 subclass. Indeed, the carboxyhemoglobin concentrations as well as the incidence of hyperbilirubinemia was lower in the blood group A infants than in blood group B, confirming the lack of effect of IgG1 subclass. We did not show these data as we are currently preparing a report of an expanded database comparing hemolysis and hyperbilirubinemia between blood group A and group B babies, both born to group O mothers.
Michael Kaplan, MB ChB, Cathy Hammerman, MD, Department of Neonatology, Shaare Zedek Medical Center; Faculty of Medicine of the Hebrew University, Jerusalem, Israel.
David K Stevenson, MD, Department of Pediatrics, Stanford University School of Medicine, Stanford, CA.
References
1. Kaplan M, Na'amad M, Kenan A, et al. Failure to predict hemolysis and hyperbilirubinemia by IgG subclass in blood group A or B infants born to group O mothers. Pediatrics. 2009;123(1). Available at: www.pediatrics.org/cgi/content/full/123/1/e132
2. Sgro M, Campbell D, Shah V. Incidence and causes of severe neonatal hyperbilirubinemia in Canada. CMAJ. 2006;175:587-590.
3. Salas AA, Mazzi E. Exchange transfusion in infants with extreme hyperbilirubinemia: an experience from a developing country. Acta Paediatr. 2008;97:754-758
4. Bhutani VK, Johnson L, Sivieri EM. Predictive ability of a predischarge hour-specific serum bilirubin for subsequent significant hyperbilirubinemia in healthy term and near-term newborns.Pediatrics. 1999;103:6-14.
Conflict of Interest:
None declared
Dear Sir,
In the last issue of the journal, Kaplan et al investigate whether identification of immunoglobulin G subclass 1 or 3 may be predictive of neonatal hemolysis and hyperbilirubinemia and report their experience about the relationship between immunoglobulin subclasses and heme catabolism (corrected carboxyhemoglobin and corrected carboxyhemoglobin/total hemoglobin ratio) in 82 newborns studied prospectively (1). They add valuable information to our knowledge about the roles of immunoglobulin G subclasses in ABO incompatibility and present the first data about the indices of heme catabolism in significant hyperbilirubinemia of ABO heterospecificity. However I have some criticisms regarding the design, results and conclusions of the study, and would like to make some contribution related to the topic.
The rates of ABO incompatibility (21%) and of those with a direct antiglobulin test in these (15%) during the study period are in accordance with those reported in the literature (2-4), however, the definition of ABO hemolytic disease, which is the extreme end of ABO heterospecificity (some of its criteria are early onset hyperbilirubinemia, a positive direct antiglobulin test, a mild to moderate anemia, reticulocytosis and a weak or no response to phototherapy), is not always considered in the studies giving rise to a high false positive rate of diagnosis of true ABO incompatibility as in those referenced also by the authors (5,6).
In such a study one would expect from the authors the use of a nomogram constituted for newborns with ABO heterospecificity (4) instead of using a standard one, and indeed when we compare the first plasma total bilirubin concentrations performed in the study (9.6±2.2 mg/dl at 19.0±11.0 hours) wee see that those equally match the >90th percentile of newborns with ABO incompatibility (a mean value of 9.2 mg/dL at 18th hours) (4). And vitally, if there is such a sensitivity of the nomogram in determining the cases who have or who will have significant hyperbilirubinemia -Current evidence-based data support it, indeed (4,7,8) - then, we should not hesitate to use it in necessary conditions when needed, while searching for more sensitive alternatives.
An important conclusion of this study, remaining open to discussion, would have been the preponderance of blood group A in newborns with IgG1 subclass when compared to that of blood group B in those with assumed IgG2/4 subclass (17 of 18 with blood group A in the former versus 45 of 64 in the latter; p=0.028, Fisher`s chi-square test) if it had not been overlooked by the authors.
Finally, I think it would be absolutely necessary to compare the immunoglobulin G subclasses and the incidence of significant hyperbilirubinemia in newborns with both a positive DAT and ABO heterospecificity (37/240) with those of newborns with a negative DAT and ABO heterospecificity (203/240) at least selected midway through the study before reaching a conclusion that "hemolysis and hyperbilirubinemia could not be predicted by this gel technique that enabled identification of these immunoglobulin G subclasses."
References
1. Kaplan M, Na'amad M, Kenan A, et al. Failure to predict hemolysis and hyperbilirubinemia by IgG subclass in blood group A or B infants born to group O mothers. Pediatrics. 2009;123(1). Available at: www.pediatrics.org/cgi/content/full/123/1/e132
2. Zipursky A, Bowman JM. Isoimmune hemolytic disease. In: Nathan DG, Oski FA, eds. Hematology of Infancy and Childhood. 4th ed. Philadelphia, PA: WB Saunders; 1993:44-74
3. Ozolek JA, Watchko JF, Mimouni F. Prevalence and lack of clinical significance of blood group incompatibility in mothers with blood type A or B. J Pediatr. 1994;125(1):87-91
4. Sarici SU, Yurdakok M, Serdar MA, et al. An early (sixth-hour) serum bilirubin measurement is useful in predicting the development of significant hyperbilirubinemia and severe ABO hemolytic disease in a selective high-risk population of newborns with ABO incompatibility. Pediatrics. 2002;109(4). Available at: www.pediatrics.org/cgi/content/full/109/4/e53
5. Sgro M, Campbell D, Shah V. Incidence and causes of severe neonatal hyperbilirubinemia in Canada. CMAJ. 2006;175(6):587-590
6. Salas AA, Mazzi E. Exchange transfusion in infants with extreme hyperbilirubinemia: an experience from a developing country. Acta Paediatr. 2008;97(6):754-758
7. Bhutani VK, Johnson L, Sivieri EM. Predictive ability of a predischarge hour-specific serum bilirubin for subsequent significant hyperbilirubinemia in healthy term and near-term newborns. Pediatrics. 1999;103(1):6-14
8. Sarici SU, Serdar MA, Korkmaz A, et al. Incidence, course, and prediction of hyperbilirubinemia in near-term and term newborns. Pediatrics. 2004;113(4):775-780
S.Umit Sarici, MD Division of Neonatology Department of Pediatrics Gulhane Military Medical Academy Etlik-06018, Ankara, Turkey s.umitsarici@tr.net
Conflict of Interest:
None declared