Article Text
Abstract
The direct antiglobulin test (DAT) detects the presence of immunoglobulin, complement or both bound to the red blood cell membrane. The test, historically called the ‘Coombs test’, was first described in 1945 by Cambridge immunologist Robin Coombs. Suspected haemolytic disease of the newborn, due to either Rhesus disease or ABO incompatibility, is one of most common reasons for requesting a DAT in newborns. In this article, we discuss the physiological background and technological background of the DAT. We also provide a clinical framework for a rational approach to the use and interpretation of the DAT in newborns.
- Neonatology
- Haematology
- Jaundice
Statistics from Altmetric.com
Introduction
The direct antiglobulin test (DAT) detects the presence of immunoglobulin, complement or both bound to the red blood cell (RBC) membrane. The test, historically called the ‘Coombs test’, was first described in 1945 by Cambridge immunologist Robin Coombs. Common indications for requesting a DAT are outlined in box 1. Suspected haemolytic disease of the newborn (HDN) due to either Rhesus (Rh) disease (or other red cell alloantibodies or maternally derived autoantibodies) or ABO incompatibility is one of most common reasons for requesting a DAT in newborns.
When should the direct antiglobulin test (DAT) be requested?
To investigate the aetiology of haemolysis
It is appropriate to request a DAT in a newborn infant with anaemia, hyperbilirubinaemia, high reticulocyte count and high lactate dehydrogenase.
What is the DAT commonly used for?
To assist in determining if haemolysis is immune mediated, such as in the following scenarios:
Haemolytic disease of the fetus and newborn
Autoimmune haemolytic anaemia
Sensitisation of red cells due to drugs
Investigation of haemolysis (eg, haemolytic transfusion reactions, ABO-incompatible platelet transfusions)
HDN occurs when maternal IgG antibodies to the fetal RBC antigens cross the placenta and bind to the fetal RBC antigens leading to haemolysis. The majority of cases of Rh disease are due to Rh antibodies, primarily anti-D, anti-c and anti-E.1 HDN due to ABO incompatibility is more common than Rh disease but is clinically less severe due to weaker expression of ABO antigens at birth and the presence of A and B antigens in the plasma of the majority of non-O infants (termed ‘secretors’). Fifteen to twenty-five per cent of all maternal/infant pairs are ABO incompatible but only approximately 1% of these women will have high-titre IgG antibodies.2 Of infants born of these ‘incompatible’ pairs, 33% of these infants will have a positive DAT but very few will go on to develop HDN with its incidence only about 0.02–1.4% for all births,3 highlighting the limited role the DAT has in identifying infants at risk of HDN.
However, it is not an uncommon clinical practice to diagnose infants with HDN without evidence of haemolysis first being established with over-reliance on a positive DAT (box 2). In our clinical practice, we have observed cases when levels of bilirubin reach exchange level, despite intensive phototherapy; intravenous immunoglobulin (IVIG) therapy is considered as therapy to mitigate haemolysis, without first seeking biochemical evidence of haemolysis. With the evidence base for this treatment in HDN limited,4 it should not be used in infants with an ABO set-up alone, without evidence of haemolysis. The DAT should not be relied upon alone and additional investigations are required to support the diagnosis of haemolysis. IVIG is not a benign treatment and treated infants are placed at risk of transfusion complications, most notably an increased risk of haemolysis as IVIG contains both anti-A and anti-B.
Suggested criteria for the diagnosis of immune haemolytic disease of the newborn
Blood group incompatibility
Mother with known red cell alloimmunisation (eg, anti-c, D, K1)
A, B or AB blood type in an infant born to a group O mother
Demonstration of haemolysis
Elevated reticulocyte count OR
Microspherocytes or spherocytes (due to partial membrane loss) on peripheral blood film OR
Progressive anaemia on serial full blood counts OR
Biochemical markers consistent with haemolysis: hyperbilirubinaemia and elevated lactate dehydrogenase OR
End-tidal carbon monoxide concentration15 if available
Demonstration of antibody-mediated red cell binding
By a positive DAT*
AND
Exclusion of other causes of hyperbilirubinaemia
*If DAT is negative by tube but haemolysis is highly suspected, more sensitive methodologies should be used to confirm an immunological cause of haemolysis, for example, gel or flow cytometric detection of IgG.
Requesting a DAT and cord blood grouping are generally recommended for infants born to women with potentially significant antibodies5 ,6 (box 3). Although some clinical guidelines also suggest that if maternal blood group is O+, it is an option to test the cord blood for the infant's blood group and DAT,5 this testing is not required if there is adequate surveillance for hyperbilirubinaemia, risk assessment prior to discharge and appropriate follow-up.5 Despite this, some institutions routinely perform cord-blood typing and DATs on all infants born to group O women, even though it is a cost-ineffective strategy3 and not supported by published clinical guidelines.6
Indications for requesting a direct antiglobulin test (DAT) in a well, term newborn infant or on cord blood
Suspected haemolytic disease of the newborn (HDN) due to evidence of haemolysis on laboratory testing (even if maternal antibody screen is negative*)
Infants born to mothers with known potentially significant antibodies
N.B. Hyperbilirubinaemia alone is not an indication to requesting a DAT in the absence of anaemia, reticulocytosis, high lactate dehydrogenase or spherocytosis on blood film examination.
*HDN caused by an antibody directed against a low-incidence antigen, for example, anti-Wra not present on routine reagent screening cells.
In this article, we discuss the physiological and technological background of the DAT. We also provide a clinical framework for a rationale approach to the use and interpretation of the DAT in newborns. Please see online supplementary appendix 1 for our search strategy.
Physiological background
Coombs, Mourant and Race developed the antiglobulin test in 1945 when they used it to detect non-agglutinating RBC antibodies (indirect antiglobulin test (IAT) or ‘antibody screen’) or sensitised RBCs (direct antiglobulin test).7 The DAT is currently used to demonstrate antibody or complement coating of RBCs. Washed RBCs from a patient or blood donor are mixed with antiglobulin (AHG) and anti-complement (anti-C3d), then centrifuged and observed for agglutination.
A positive DAT demonstrates in vivo sensitisation of RBCs with IgG and/or complement components. In newborn infants, it indicates the presence of maternal IgG on the infant's RBCs. If maternal serum contains an IgG class immunoglobulin directed against a fetal RBC antigen, transplacental passage of this antibody will result in RBC antibody coating and a positive neonatal DAT.8 A key point to highlight is that a positive DAT does not mean haemolysis is present and in the vast majority of cases, there is no biochemical evidence of haemolysis when further investigations are performed.
Technological background
The DAT is performed with polyspecific antiglobulin reagent containing antibodies to IgG and the C3d component of complement (figure 1). Cord blood or a sample taken directly from a newborn is tested with monospecific anti-IgG reagent, as this is the only coating protein expected in the newborn infant. Testing with complement proteins is not necessary in suspected HDN as the protein sensitising the infant's RBCs is maternal IgG.
Several methods for performing the DAT exist, including the conventional test tube (CTT) (figure 2) method and the more sensitive gel microcolumn (figure 3). In the CTT method, the patient's RBCs are washed with saline to remove unbound immunoglobulin and complement, anti-human globulin (AHG) is added and the RBC suspension centrifuged. The RBC pellet is dislodged using gentle tapping and examined for agglutination.9 Agglutination is graded on a scale from 0 indicating no agglutination to 4+ indicating solid agglutination. In the gel microcolumn method, RBCs are filtered through a gelatinous matrix mixed with AHG reagents. The gel traps the agglutinated RBCs and non-agglutinated RBCs pass through the column.9
Historically, the DAT was uniformly performed in a test tube and agglutination was observed under an inverted light microscope by the technologist. More recently, blood transfusion laboratories are migrating towards performing the testing in a more sensitive platform using gel technology to minimise the frequency of false-negative results. The causes of false-positive and false-negative DATs are listed in box 4.
The causes of false-positive and false-negative results (DAT)
False positive
Wharton's jelly in cord blood specimens
Spontaneous red blood cell (RBC) agglutination
Technical
Poor washing technique
Improper agitation of specimen during reaction strength determination (test tube method)
Over centrifugation
Clotted specimen
False negative
Inadequate amount of IgG molecules to detect agglutination
Technical
Poor centrifugation
Poor washing technique
Inappropriate concentrated RBC suspension
Improper agitation of specimen during reaction strength determination
Inactive anti-human globulin (AHG) reagent
Delay in testing
Failure to add or delayed addition of AHG reagent
Indications and limitations
The indications for requesting a DAT in a newborn are outlined in box 3. Performing a DAT prior to establishing whether haemolysis is occurring is of limited value and is actively discouraged by our group.
There are a number of limitations of the DAT test. A positive DAT has a poor predictive value for ABO HDN with some authors reporting that only 23% of newborns with a positive DAT on neonatal screening developed hyperbilirubinaemia.10 A positive DAT does not rule in ABO HDN and a negative DAT does not rule out ABO HDN. Additionally, a negative DAT does not exclude a non-immune haemolytic aetiology underlying clinically significant hyperbilirubinaemia. Differential diagnoses include G6PD deficiency, sepsis, hereditary spherocytosis and pyruvate kinase deficiency in a newborn with clinically significant hyperbilirubinaemia and a negative DAT. A cause other than isoimmunisation should be sought for ABO-incompatible, DAT-negative newborns with significant jaundice or increased bilirubin production.11
Requirements for specimens, collection and handling
An EDTA blood collection tube is needed for the DAT as it prevents in vitro fixation of complement by chelating the calcium needed for C1 activation. If a clotted specimen is used, a false-positive result may be obtained due to non-specific binding of complement to the RBCs in vitro. When using cord blood for testing, contamination with Wharton's jelly may cause non-specific agglutination and a false-positive result.9
Clinical questions
Can we use the test to diagnose ABO HDN?
In a term newborn with significant hyperbilirubinaemia and blood group A or B (patient) and a maternal group O blood type, does a DAT (test) in the infant accurately diagnose ABO HDN (outcome)?
A positive DAT does not necessarily indicate that haemolysis is occurring and a workup for haemolysis is required (box 2) before a diagnosis of ABO HDN can be made. A negative DAT does not rule out ABO HDN and does not exclude a non-immune haemolytic aetiology underlying clinically significant hyperbilirubinaemia. Differential diagnoses include G6PD deficiency, sepsis, hereditary spherocytosis and pyruvate kinase deficiency in a newborn with clinically significant hyperbilirubinaemia and a negative DAT.
Can we use the test to identify for infants at risk of developing ABO HDN?
In a newborn infant with a group O mother (patient), should a DAT (test) be routinely requested to identify if the infant is at risk of ABO HDN (outcome)?
As a screening test, the DAT has been identified as having a poor positive predictive value (PPV) for identifying newborns at risk of clinically significant hyperbilirubinaemia.12 ,13 Studies10 ,14 ,15 outlined in table 1 found that a positive DAT had a PPV of 12–53% when used to identify infants at risk of developing clinically significant hyperbilirubinaemia and/or haemolysis. Clinical monitoring of infants for jaundice in the first week of life rather than reliance on a screening test with a poor PPV is likely a more effective management approach.
Another practice undertaken by some centres is selective cord testing (ABO typing and DAT) of infants born to blood group O+ mothers rather than routine testing of all newborn infants. This has been shown by several studies to reduce costs without increasing the risk of clinically significant hyperbilirubinaemia.3 ,16 ,17 This practice is not supported by transfusion expert groups, including the AABB.6 Any routine screening for ABO/D typing and DAT in newborn infants, aside from infants of Rh-negative mothers, is advised against. They recommend clinical monitoring of infants for jaundice in the first week of life as suggested by the American Academy of Pediatrics guidelines for hyperbilirubinaemia5 rather than reliance on a screening test with a poor PPV. There are other well-established adjunct clinical screens to identify infants at risk of hyperbilirubinaemia,18 including transcutaneous bilirubin measurements prior to discharge and follow-up within 48 h postdischarge by a healthcare provider.
Should we request the DAT on all critically ill newborns?
Should all infants admitted to the neonatal intensive care unit (patient) routinely have a DAT (test) requested in conjunction with a blood group to identify those at risk of HDN (outcome)? What about infants undergoing routine newborn care?
Requesting a blood type on either cord or infant blood should be restricted to infants in newborn intensive care units that may require transfusion. Other appropriate indications for a blood group include well term newborns with clinically significant hyperbilirubinaemia and/or a positive maternal antibody screening or unavailable maternal blood testing.16
A cord Rh group should be requested for all infants of Rh-negative mothers to determine if the mother requires postpartum Rh immunoglobulin.6 In addition, when Rh D-negative women receive anti-D immunoglobulin up to 15% of their newborn infants will have a positive DAT due to passive transfer of the prophylactic anti-D during the third trimester.19 These antibodies do not cause haemolysis in the fetus or newborn and do not require any further investigation.
Should we request the DAT if the maternal antibody screen is negative and alloimmunisation is suspected?
In a term newborn with clinically significant hyperbilirubinaemia, laboratory signs of active haemolysis and a negative maternal antibody screen (patient), should a DAT (test) be requested as part of the workup for possible HDN (outcome)?
The maternal antibody screen, an IAT, can be negative in the setting of alloimmunisation. The IAT is only 97% sensitive and antibody detection is critical in identifying women at risk of having future pregnancies affected by haemolytic disease of the fetus.20 Requesting the DAT in this situation is critical, as the DAT will be positive and will prompt the blood bank to perform additional studies to identify the antibody specificity (eg, RBC eluate and testing for low-incidence antibodies).
Topics for further research
See box 5.
Topics for further research
What has been the impact of gel technology on the frequency of positive direct antiglobulin tests (DATs) in non-group O newborn infants of group O mothers? What has been the impact on the sensitivity and specificity of the DAT with this new technology?
Can DAT testing be safely restricted to newborn infants with only particular haematologic abnormalities (eg, an elevated reticulocyte count and biochemical markers (eg, elevated lactate dehydrogenase))?
Clinical bottom lines
A positive direct antiglobulin test (DAT) does not rule in a diagnosis of ABO haemolytic disease of the newborn (HDN).
If HDN is suspected, additional investigations beyond the DAT, such as serial full blood examinations, biochemical tests for haemolysis and a peripheral blood film should be requested to confirm haemolysis is occurring.
A negative DAT does not rule out a diagnosis of ABO HDN or another underlying haemolytic process as false negatives may occur (see box 4). If DAT is negative by tube but haemolysis is highly suspected, more sensitive methodologies should be used to confirm an immunological cause of haemolysis, for example, gel or flow cytometric detection of IgG.
DATs should not be routinely requested on all cord bloods or on all newborn infants. A cord Rhesus (Rh) group should be requested on all infants whose mothers are known to be Rh negative.
A DAT should be requested in an infant when there is a high index of clinical suspicion of HDN, irrespective of the maternal blood group and antibody screen result.
Answers to questions from the quiz on page 184
Answers A, C, D and E are all true. Klinefelter typically results in tall stature
Answers A and B are true.
The other conditions may result in short stature, but are not licensed indications of growth hormone.
Answers A, B and D are true.
A growth hormone stimulation test is not usually part of first line investigations but may be considered if indicated by initial investigation results. A bone age tends to be of little value in this age group. An isolated growth hormone level is not of use—IGF-1 is used as a marker of growth hormone-stimulated somatic growth.
Answers B and D are true. Bone age represents growth potential, that is, a younger bone age reflects greater growth potential. Growth can be reduced in any chronic illness, and affected patients will typically have a delayed bone age.
Answers A, B, C and E are all true.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Footnotes
Contributors AK, MA, LL and JC all contributed to the analysis and evaluation of the literature, drafting of the manuscript and approved the final version of the manuscript.
Competing interests None.
Provenance and peer review Commissioned; externally peer reviewed.