doi:10.1111/j.1440-1754.2010.01888.

x

ORIGINAL ARTICLE

Has anti-D prophylaxis increased the rate of positive direct antiglobulin test results and can the direct antiglobulin test predict need for phototherapy in Rh/ABO incompatibility?
jpc_1888 40..43

Alexander Dillon,1 Tejasvi Chaudhari,2 Phillip Crispin,1,3 Bruce Shadbolt1,4 and Alison Kent1,2
1 Australian National University Medical School, Canberra and Departments of 2Neonatology and 3Haematology, and 4Clinical Epidemiology Unit, Canberra Hospital, Woden, Australian Capital Territory, Australia

Aim: To determine the impact of Rhesus (Rh) D prophylaxis on positive direct antiglobulin test (DAT) results and ability of a DAT grade to predict an infant’s need for phototherapy. Methods: Laboratory and infant medical records were reviewed for DAT status, DAT grade, interventions for hyperbilirubinemia including phototherapy, blood transfusion, exchange transfusion and intravenous immunoglobulin. Two epochs of DAT results were reviewed, the first in the era prior to Rh D prophylaxis, the second after introduction of standardised Rh D prophylaxis for Rh negative women. Results: A total of 165 DAT-positive infants’ medical records were reviewed. The number of positive DAT results increased from 1.5% to 2.3% (P < 0.0001) following introduction of anti-Rh D prophylaxis, the increase related to an increase in anti-D DATs (7.4% to 32% – P < 0.0001). An infant with a DAT grade of 5–8 was 2.6 times more likely to need phototherapy than an infant with a DAT grade of 2–4 (odds ratio (OR), 2.571; 95% confidence interval (CI), 1.225–5.393; P = 0.08) and an infant with a DAT grade of 10–12 was 4.7 times more likely to need phototherapy than an infant with a DAT grade of 2–4 (OR, 4.724; 95% CI, 1.602–13.926, P = 0.013). Conclusions: Rh D prophylaxis has increased positive DAT results, which may increase the number of unnecessary bilirubin measurements. A low or high DAT grade is strongly predictive of whether an infant does or does not require phototherapy. However, an intermediate DAT requires concomitant bilirubin measurements to determine phototherapy requirements. Key words: direct Coombs’ test; hyperbilirubinemia; jaundice; neonatal; newborn haemolytic disease; Rh prophylaxis.

What is already known on this topic 1 Rh D prophylaxis has become routine practice across Australia. 2 Rh prophylaxis for Rh negative women who have not been sensitised to the Rh D antigen previously can lead to a positive (direct antiglobulin test) DAT if the DAT is performed within approximately six weeks of prophylaxis administration. 3 The DAT has a limited predictive value in diagnosing and predicting the course of neonatal hyperbilirubinemia.

What this paper adds 1 Since its introduction, Rh prophylaxis has led to an increase in the number of positive infant DAT results in the ACT. 2 An infant with a DAT grade of 5 or above has an increased chance of receiving phototherapy for treatment of hyperbilirubinemia compared with an infant with a lower DAT grade. 3 DAT grade continues to have limited predictive value in diagnosing and predicting the course of neonatal hyperbilirubinemia.

Introduction
Hyperbilirubinemia is one of the most common causes for readmission to hospital of newborn infants in the first weeks of life (1–3).1 High bilirubin levels are potentially harmful to the developing central nervous system. One in 10 000 infant’s bilirubin levels may be elevated to the point that they cause acute biliruCorrespondence: Associate Professor Alison Kent, Australian National University Medical School, Department of Neonatology, The Canberra Hospital, PO Box 11, Woden, ACT 2606, Australia. Fax: +61 2 6244 3112; email: alison.kent@act.gov.au Accepted for publication 25 March 2010.

bin encephalopathy in the short term and brain damage (kernictus) in the longer term.2,3 Leading causes of severe neonatal hyperbilirubinemia include ABO (blood groups) incompatibility, Rh incompatibility and G6PD deficiency but may be caused by other conditions.4 Early detection of hyperbilirubinemia allows early initiation of treatment, avoiding potential complications utilising phototherapy, exchange transfusion and/or intravenous immunoglobulin (IVIG). The provision of anti-D immunoprophylaxis to (Rhesus) Rh-negative mothers during pregnancy, a practise recommended as a safe and cost-effective solution,5 may have an effect on both the DAT status/strength of reaction as well as the requirement for phototherapy.6,7

40

Journal of Paediatrics and Child Health 47 (2011) 40–43 © 2010 The Authors Journal of Paediatrics and Child Health © 2010 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

Parity.2) 6 (5) 36 (32) 55 (50) 7 (6) P value <0.5 (1. All DATs were performed in one laboratory. P <0. All infants born to Rh-negative mothers. Cases were excluded if they were born prematurely (less than 37 weeks gestation) or born at another hospital. Chicago.2 Current detection methods for hyperbilirubinemia include clinical examination for jaundice (high sensitivity but low specificity with 60–80% of term or near-term infants developing jaundice with only a small number requiring treatment in the first week of life).5%) infants born in the period between 1 January 2001 and 31 December 2002 were DAT-positive. which only provides current bilirubin levels and cannot accurately predict whether a baby will develop hyperbilirubinemia.f. there were a total of 4739 live births. and 111 (2. the decision based on consideration of postnatal age and jaundice level as per hospital guidelines. d. Phototherapy was commenced in 84/165 (50. DAT grades are not continuous but ordinally assigned numbers.11 while Rh-negative mothers of infants born in the period January 2000 to December 2002 were provided with Rh D prophylactic immunoglobulin for procedures or complications of pregnancy and at birth. = 1.2 The aims of this study are to determine whether anti-D prophylaxis has increased the rate of DAT positive results and whether DAT grade at birth can predict the need for phototherapy. There was a significant rise of DAT-positive results between the two epochs (P < 0. Australia) in both epochs.05 0. A total of 140 (85%) infants had their serum bilirubin levels measured an average of five times. two B Rh D-positive and one O Rh D-positive. The infants had serum bilirubin measurements with a positive DAT test when born to Rh-negative mothers. 8. IL. Infants received phototherapy for a median period of 30 h (SD Ϯ 45).3 (1.21 <0. with other integers used for results that are felt to have reactions in between these results.0001 0. and the Direct Coombs Test (DAT). DAT grade.9 and other clinical techniques including clinical examination and assessment of risk factors including ethnicity.3) 80 (72) 31 (28) 3349 (475) 39. Data were maintained and analysed using SPSS version 16 (SPSS Inc. or when it was clinically indicated in a jaundiced baby. especially in cases of ABO incompatibility8. Methods A retrospective review of laboratory and medical records of DA-positive infants born at the Canberra Hospital (TCH) was performed in two epochs from December 2000–2002 and 2006– 2008.0001) (Table 1). The numbers 3. The DAT is a blood test used to detect in vivo sensitisation of red blood cells.. The DAT has a high false negative rate. and those infants that subsequently developed jaundice after birth received a DAT. 10 and 12 are used routinely. Elution was routinely performed on positive neonatal DATs for antibody identification. Victoria. birthweight. direct measurement of serum bilirubin levels. Mt Waverley. Anti-D prophylaxis Detection of hyperbilirubinemia is complicated by the limited predictive value of current screening measures combined with the frequently unpredictable rate of rise of bilirubin levels. while three of the four (75%) infants 41 Journal of Paediatrics and Child Health 47 (2011) 40–43 © 2010 The Authors Journal of Paediatrics and Child Health © 2010 Paediatrics and Child Health Division (Royal Australasian College of Physicians) . exchange transfusions or blood transfusions. The Australian Capital Territory Human Research Ethics Committee approved the retrospective audit as per National Health and Medical Research Council (NHMRC) guidelines. USA).10 DATs were performed using column agglutination (Diamed. and a logistic regression was used to assess the association between DAT and need for phototherapy. Table 1 Characteristics of DAT-positive infants between epoch 1 and epoch 2 Epoch 1 n = 54 Cases per total births (%) Multips (%) Primips (%) Mean Bwt grams (ϮSD) Mean gestation weeks (ϮSD) Diluent not available (%) Anti-D (%) Anti-A (%) Anti-B (%) 54/3816 (1.5) 44 (82) 10 (28) 3508 (528) 39. birthweight and gestational age were considered as confounders in the model.0001). there were a total of 3816 live births. Four A Rh D-positive infants received IVIG. serum bilirubin levels.0001 formed (including means.9%) of the infants. DAT status. and between January 2007 and December 2008. Three infants received top-up blood transfusions. 5. One O Rh D-positive infant received an exchange transfusion. treatments for hyperbilirubinemia including phototherapy.2) 1 (2) 4 (7) 30 (56) 14 (26) Epoch 2 n = 111 111/4739 (2. Fifty-four (1.4% (4/54) ) than in the first epoch (c2 = 11. standard deviation. using the 0–12 scale with manual interpretation. IVIG. Data collected included gestation.3%) were DAT-positive between 1 January 2006 and 31 December 2007. which may return a positive result in ABO and Rh incompatibility. and blood type can be used as effectively to assess the need for phototherapy. Descriptive statistics were per- Results From 2001 to 2002.19 0. with measurement ceasing when their serum bilirubin levels dropped below phototherapy range and there was no change in the clinical condition. and a positive result in an infant may indicate haemolytic disease of the newborn (HDN).A Dillon et al. Infants born to Rh D-negative mothers and infants that developed jaundice had routine cord blood DAT and blood group performed. Seventeen premature infants with positive DAT results were excluded. 16 of the 36 (44%) infants with anti-D detected in the second epoch received phototherapy. There were significantly more infants with anti-D antibodies in the second epoch (32% (36/111) versus 7.1.2 Previous studies have indicated that the DAT is relatively poor at predicting whether an infant will develop hyperbilirubinemia and has limited utility in predicting the need for phototherapy. percentages and frequencies). Rh-negative mothers of infants born in the period January 2006 to December 2008 were provided with Rh D prophylactic immunoglobulin (provided they had no preformed antibodies) at recommended levels of 625 international units (IU) at 28 and 34 weeks.

05).7 Given that the study population from both time periods is similar epidemiologically.2% (14/33) with DAT grades of 2–4 received phototherapy. d.006) (Fig.04). and group 3 – DAT grades 10–12) with no loss of variation in the association with phototherapy. P < 0.7 times more likely to need phototherapy than infants with a DAT grade of 2–4.1% (40/74) of the infants with DAT grades of 5–8 received phototherapy and 66. Our with anti-D detected during the first epoch received phototherapy. 2. This may be due to increased numbers of neonates with haemolysis. the infants with a DAT grade of 5–8 had a phototherapy rate of 55. This study also identifies limitations in the predictive value of DAT grades. P = 0.3% of the population between 2006 and 2008. The relationship between DAT grade and need for phototherapy was not significantly different between those infants with anti-D antibodies and those with other antibodies. Prior to the second epoch. the infants with a DAT grade of 5–8 were 2. Of the infants. 2 Treatment with phototherapy and grade of direct antiglobulin test.Anti-D prophylaxis A Dillon et al.122.f. Discussion Our study is the first in Australia to examine the effect of Rh D prophylaxis on infant DAT results. had no anti-D antibodies) at 28 and 34 weeks gestation.4% (51/92) the and infants with a DAT grade 10–12 had a phototherapy rate of 69. The results from the two epochs were combined to determine whether the DAT grade could be used to predict need for phototherapy.013). 54. group 2 – DAT grades 5–8. 1. our hospital introduced anti-Rh D prophylaxis for all Rh-negative women who had not previously developed antibodies to the Rh D antigen (i. Our study has shown an increase in DAT-positive infants from 1.489. Range of DAT Grades 50 40 Number 30 20 Fig. 61. the infants in the anti-D group had a stronger trend for receiving phototherapy. The infants with a DAT grade of 10–12 were 4. 4. = 7. received phototherapy and 75% (6/8) of the infants with DAT grades of 10–12 received phototherapy.5% of the population between 2001 and 2002 to 2. it is likely that Rh prophylaxis of Rh-negative women has been the reason for this concomitant rise in positive anti-D DATs. This study shows that there is a correlation between high DAT grade at birth and future need for phototherapy.7 times more likely to need phototherapy than the infants with a DAT grade of 2–4 (OR. 7. phototherapy may have been administered to neonates with physiologic jaundice who have been interpreted as possibly having HDN due to positive DATs. given that 30% of the infants with DAT grades of 10 to 12 did not receive phototherapy and 32% of the infants with DAT grades of 2 to 4 received phototherapy. Infants with a DAT grade of 5 to 8 are 2.7% (1/13) with anti-D antibodies with DAT grades of 2–4 received phototherapy.724. 4. However.7% (10/15) of the infants with DAT grades of 10–12 received phototherapy. 42.39). gestation and birthweight. The increase in phototherapy is nevertheless an unintended consequence of antenatal prophylaxis. 2). P = 0.e.2–5.6% (16/23) (c2 = 10.926. and that there is no other readily available reason for the large increase in infants with a positive DAT result carrying anti-D antibodies.6 times time more likely to need phototherapy and those with a DAT grade of 10–12 are 4. there was no significant difference in treatment with phototherapy and antigen diluent (c2 P = 0. 95% confidence interval (CI). When combining the two epochs.6% (15/46).602– 13. Other studies have shown that recent anti-D prophylaxis (within six weeks) in pregnant Rh-negative women can result in positive DAT results. The DAT grades were subsequently grouped into three categories (group 1 – DAT grades 2–4. Of those infants with ABO antibodies.6. However. 10 n=12n=32 2 3 4 n=39 5 6 7 n =51 8 9 n =14 10 11 n=9 12 DATgrade Fig. The infants with a DAT grade of 2–4 had a phototherapy rate of 32.6 times more likely to need phototherapy than infants with a DAT grade of 2–4 (odds ratio (OR). other studies have not found increased haemolysis with antenatal anti-D prophylaxis. Alternatively. 95% CI. 1 Range of direct antiglobulin test grades.1% (11/18) of the infants with DAT grades of 5–8 42 Journal of Paediatrics and Child Health 47 (2011) 40–43 © 2010 The Authors Journal of Paediatrics and Child Health © 2010 Paediatrics and Child Health Division (Royal Australasian College of Physicians) .6. P = 0. A positive association between DAT grade and the need for phototherapy was found (c2 = 14. After adjusting for the confounding factors of parity. The range of DAT grades are shown in Figure 1. This study found that this increase was due to an increase in the proportion of DAT-positive infants expressing anti-D antibodies. 1.

Maisels MJ. 24: 391–4. clinical prediction is similarly deficient in predicting the course of HDN because of Rh incompatibility.html [accessed 23 July 2009]. Available from: http://www. Andrew K. 3 Bhutani VK. Vinod P. it is best used in combination with other predictive tools/risk factors and constant monitoring.gov. 7. 1985. which provide a framework for decision-making. Acta Paediatr. Guidelines for blood grouping and antibody screening in the antenatal and perinatal setting [homepage on the Internet]. The Canberra Hospital. 75: 157–64. Deorari AK. The sample sizes allowed statistical confidence in the findings made and the dual epoch study design enabled a comparison of the pre. Jaundice in the newborns. 2006. 14th edn.cs. It is unknown whether the clinicians were aware of the DAT grade and whether this influenced their commencement of phototherapy. which would have helped to ensure more accurate results.12 Previous studies have indicated that while the DAT is less useful in predicting the severity of HDN due to ABO incompatibility than in predicting HDN due to Rh incompatibility. The data in this study are not consistent with current guidelines for blood grouping and antibody screening in the antenatal and perinatal setting.au/pubs/pdf/glinesanti-d. Conclusions Our study demonstrates that routine Rh D prophylaxis for Rh-negative women has increased the number of positive DAT results. 4 Han P. Joseph R.1. Royal Prince Alfred Hospital – Haemolytic Jaundice Protocol [Homepage on the Internet]. 6. Neonatology 2008. Indian J. Anti-D prophylaxis findings are similar to those of previous studies in that while the DAT has some value in predicting the severity of neonatal hyperbilirubinemia and subsequent need for phototherapy.1 The DAT’s poor accuracy in predicting the severity of hyperbilirubinemia via the proxy of phototherapy may be a reflection of current conservative guidelines for the administration of phototherapy. 6 Royal Prince Alfred Hospital (RPA). Pediatr. MD: AABB Press. this weakness is limited by the medical staff’s adherence to existing hospital guidelines for the application of phototherapy. Benitz W. breastfeeding. Clin.1. 43: 63–8. even in the neonate. which may result in an increased number of unnecessary bilirubin levels taken and phototherapy commenced. 11 Chandra A. Screening for neonatal hyperbilirubinemia and ABO alloimmunization at the time of testing for phenylketonuria and congenital hypothyreosis. However. 2004. Graham H. 7 Australian and New Zealand Society of Blood Transfusion Ltd. J. Journal of Paediatrics and Child Health 47 (2011) 40–43 © 2010 The Authors Journal of Paediatrics and Child Health © 2010 Paediatrics and Child Health Division (Royal Australasian College of Physicians) 43 . and the tester is unable to decide for certain which grade should be used. Despite the DAT’s limitations.nba. ethnicity. Available from: http://www.5.pdf [accessed 20 July 2009]. 1998. no other test is currently available to both predict the need for phototherapy or assess the severity of the antibody reaction.gov.2. it has certain fixed values between 0 and 12. 10 American Association of Blood Banks. This study’s chief limitation is its retrospective nature. 2007. Management of jaundice and prevention of severe neonatal hyperbilirubinemia in infants >/= 35 weeks gestation. Ong R. Stark AR. Tan KL.and post-Rh D prophylaxis era. clinical monitoring of serum bilirubin levels and jaundice during inpatient stay and for at least seven to 10 days post-discharge by health-care workers and appropriately educated parents. Belton MH. 5 National Blood Authority.au/rpa/neonatal/ html/newprot/rhesus. 2002. Haemolytic disease due to ABO incompatibility: incidence and value of screening in an Asian population. 70: 217–22. We were reliant upon medical records for evidence of phototherapy and needed to assume that infants who received phototherapy actually required it. 87: 1269–74. This study’s strengths are that it involves two study populations of reasonable size. Readmission for newborn jaundice: the value of the Coomb’s test in predicting the need for phototherapy. Wong HB. Reflecting this. 12 Mishra S. Huntsinger K. Meyer P. rather are included for use where a sample’s characteristics lies between recognised fixed values.A Dillon et al.org.12 These include cord bilirubin levels. Paediatr. Bethesda.6 The results of this study indicate that an infant with a DAT grade of 5 or above is at greater risk of requiring phototherapy than an infant with a DAT grade of between 2 and 4. but the values 1. 2. Acta Paediatr. Pediatr. DAT grades of 5 and above increase the chance of an infant requiring phototherapy. Ryan B. while a DAT grade of less than 8 could be considered to be due to passive immunisation if there is a history of prophylaxis. Prediction of the severity of ABO haemolytic disease of the newborn by cord blood tests. 2008. Burgos A. Johansen KB.nsw. 2 Madan A.anzsbt. Acknowledgements We thank the staff of TCH medical records department for facilitating access to patient medical records. The DAT test itself is constrained in accuracy due to a degree of subjectivity in observer judgement of degree and speed of agglutination. 94: 63–7. both of which are from the same epidemiological backgrounds. 2003. allowing measurement of the effect of Rh D prophylaxis on DAT results. Our study supports the contention that the DAT’s value in predicting an infant’s need for phototherapy is not sufficient to obviate the need for other investigations and clinical monitoring. 1981. The data in this study indicate that infants with DAT grades between 5 and 8 have an increased chance of needing phototherapy. 8 Levine DH. Clinical guidelines for administration of phototherapy have been considered by some to be overly conservative and that phototherapy is frequently administered at levels far below those that would cause kernictus. Clin. 24: 35–8. 1988. A prospective study would have ensured rigid guidelines for administration of phototherapy by medical staff. Sydney: RPA. 4. Antenatal Rh (D) Immunoglobulin (Anti D) Prophylaxis. Pediatr. The Canberra Hospital Maternity Practise Standards. Agarwal R.au/publications/documents/ Antenatal_Guidelines_Mar07. References 1 Meberg A.pdf [accessed 26 July 2009].1 Meberg and Johansen suggest that higher serum bilirubin levels could possibly be accepted prior to intervention. Available from: http://www. Buonocore G. 2003. Technical Manual. These data also imply that a DAT grade cut-off of 8 or higher to be considered immune in origin may be too high. 9 and 11 are not routinely used. ABO/Rh group. Newborn screening for ABO haemolytic disease. Kiruba R. Aust. 9 Whyte J. Guidelines on the prophylactic use of Rh D immunoglobulin (anti-D) in obstetrics [homepage on the Internet].2.7 These guidelines imply that a maternal DAT grade of 8 or greater should be considered immune in origin.