Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Is intravenous immunoglobulin superior to exchange transfusion in the management of hyperbilirubinaemia in term neonates?

Three Part Question

In [a term infant with immune haemolytic disease of the newborn] does [intravenous immunoglublin therapy] reduce [serum bilirubin (SBR) levels effectively and avoid exchange transfusion]?

Clinical Scenario

A 36-week 3550 g neonate is admitted to the intensive care unit and commenced on intensive phototherapy for known Rhesus haemolytic disease. In spite of intensive phototherapy, the bilirubin level approaches the exchange transfusion threshold by hour 16 of life. The specialist registrar orders a crossmatch of blood and arranges for central line insertion in preparation for an exchange transfusion. The new registrar queries why intravenous immunoglobulin is not being used first in an attempt to avoid exchange transfusion.

Search Strategy

Search date: August 2008. Pubmed and Medline 1951–2008; Embase 1974–2008; Cinahl 1982–2008 using Dialog Datastar.
Cochrane Library (2002). One relevant systematic review.

Search terms: [Neonatal hyperbilirubinaemia or hyperbilirubinaemia AND neonate or hyperbilirubinaemia and newborn] AND [ intravenous immunoglobulin or immunoglobulin]. Limit to newborn, human and English language.

Search Outcome

14 papers

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Alcock et al,
Total number of infants: 189Cochrane systematic review of 3 RCTsExchange transfusion. Also mean number of exchange transfusions per infantExchange transfusion decreased in IVIG group (typical RR 0.28, 95% CI 0.17 to 0.47; NNT 2.7) Mean number exchange transfusions per infant lower in IVIG group (WMD –0.52, 95% CI –0.7 to –0.35)None of the studies assessed long term outcomes
Nasseri et al,
34 infants with Rhesus disease (n = 17) and direct Coombs test (DCT) positive (n = 17)RCT (single centre)Primary outcome: exchange transfusion (subdivided for Rhesus and ABO incompatibility). Also duration of hospitalisation, phototherapy and late anaemiaExchange transfusions: lower in IVIG group, 95% CI –0.46 to –0.41 Hospitalisation (days): mean 6 (SD 1) in IVIG group vs 7.41 (SD 2.09) in controls Phototherapy: mean (days) 4.9 (SD 0.96) in IVIG group vs 6.4 (SD 2) in controls. Late anaemia was more common in the IVIG group: 11.8% vs 0% controls, p = 0.48In ABO haemolytic disease there was no significant difference between IVIG and double surface blue light phototherapy There were no adverse effects during IVIG administration
37 infants with Rhesus disease/DCT positiveRCT (single centre)Total transfusion therapy (not divided into simple and exchange transfusion), duration of hospitalisation and phototherapyExchange transfusions: lower in IVIG group (RR 0.28, 95% CI 0.17 to 0.47) Hospitalisation: mean 8 (SD 3) days in IVIG group vs 14 (SD 10) days in controls Phototherapy: mean 6 (SD 3) days in IVIG group vs 6 (SD 3) days in controlsStated that 40 infants were eligible for inclusion but results only on 37 infants. Randomisation method unclear
Rübo et al,
32 infants with Rhesus incompatibilityRCT (multicentre)Primary outcome was exchange transfusion, also late anaemiaExchange transfusion: 2/16 in IVIG group vs 11/16 controls (RR 0.18, 95% CI 0.05 to 0.69) RCC transfusion for late anaemia: 2/16 in IVIG group vs 0/16 controls (RR 5, 95% CI 0.26 to 96.59)Failed to report on an intention to treat basis for protocol violations No information regarding in utero transfusions included
Tanyer et al,
61 babies who had ABO (n = 34) and Rhesus haemolytic disease (n = 18) or a combination (n = 9)RCT (single centre) Three groups: group I received multiple dose IVIG, group II received single dose IVIG, and group III was not given any IVIGThe effect of multiple dose IVIG therapy vs single dose therapy. Primary outcome: exchange transfusionExchange transfusion rate: group I: 0% group II: 12% group III: 33% Duration of phototherapy was shorter in group I vs groups III (p<0.01) and II vs III (p<0.05)Randomisation by order of admission. No difference in phototherapy duration found between single and multiple dose IVIG
Alpay et al,
116 term infants with ABO (n = 93) or Rhesus incompatibilities (n = 16) or both (n = 9)RCTPrimary outcome was exchange transfusion, duration of phototherapy and duration of hospitalisationExchange transfusion 8/58 in IVIG group vs 22/58 in controls (p<0.001) (RR 0.36, 95% CI 0.18 to 0.75) Hospitalisation: IVIG mean 4.52 (SD 1.45) days vs control 5.5 (SD 1.83) days2 patients in the control group developed hypocalcaemia and hypoglycaemia secondary to exchange transfusion. Follow-up and results data were complete
Mukhopadhya et al,
34 infants with Rhesus haemolytic diseaseProspective studyPeak SBR, phototherapy, exchange transfusions, late anaemia and discharge PCVPeak SBR: IVIG 16.5 (SD 2.96) vs no IVIG 22.7 (SD 8.84) (p = 0.004). Exchange transfusion: IVIG 62.5% vs no IVIG 88.5% (p = 0.014). Multiple exchange transfusions higher in no IVIG group (p = 0.03). Phototherapy duration: IVIG 165 (SD 109) h vs 119 (56) h no IVIG (p = 0.29). Blood transfusions for anaemia higher in IVIG group (37.5%) vs no IVIG group (11.5%), p = 0.126. PCV at discharge similar in both groups
Hammerman et al,
36 DCT positive term infantsProspective study IVIG administered if SBR reached 13 mg/dl (222 µmol/l) at 24 h of age and/or 16 mg/dl (274 µmol/l) at >24 h of ageClassified as IVIG responder (n = 18) if SBR stable or decreased after IVIG. Non-responder (n = 5) if SBR rose in 24 h after IVIG or non-treated (n = 13). Groups compared for exchange transfusion, pretreatment rate of SBR rise and pretreatment carboxyhaemoglobin (COHbc) levels4/5 of non-responders required exchange transfusion vs 0/31 others (p<0.001) 4/5 non-responders had pretreatment SBR rise of >1 mg/dl/h (17 µmol/l/h) vs 1/18 responders and 0/13 non-treated (p<0.001) Pretreatment CoHbc levels highest in non-responders vs othersFollow-up study looked at percentage decrease of SBR and COHbc at 24 h in responding group. Correlation between the two found (p = 0.007)
Walsh et al,
11 infants with Rhesus haemolytic disease/ABO incompatibilityRetrospective reviewExchange transfusion, rate of rise/fall in SBR with IVIG, rate of fall in SBR11/11 patients avoided exchange transfusion. The mean (SD) rate of rise before dose was 7.4 (11.4) µmol/l, and fell after IVIG by 9.3 (8.1) µmol/l (p = 0.001). Decrease in SBR levels before and after treatment from 234 to 219 µmol/l (p = 0.001)
Aggarwal et al,
23 infants with Rhesus haemolytic diseaseRetrospective review All received exchange transfusion within 4–6 h of birth. 13 infants subsequently given 2 doses of IVIG (0.5 g/kg)Subsequent exchange transfusions and phototherapy hoursIVIG group required fewer subsequent exchange transfusions (p<0.05). No infant who received IVIG required more than 2 exchange transfusions (p<0.05). Phototherapy hours reduced in IVIG group (p>0.05)All infants initially received a double volume exchange transfusion prior to randomisation
Ergaz et al,
5 infants with Rhesus haemolytic disease (n = 3) and ABO incompatibility (n = 2)Case seriesCOHbc levels (reflecting haemolysis) in HDN infants vs non-HDN infants. COHbc levels in treated vs untreated HDN infantsCOHbc levels higher in HDN infants vs control Rapid decline (>30%) in 4/5 HDN infants treated with IVIGNo infant required exchange transfusion Supports theory that attenuation of jaundice by IVIG is, at least in part, attributable to a reduction in haemolysis
Miqdad et al,
112 term babies with ABO haemolytic disease with significant hyperbilirubinaemiaRCT (single centre) Group I (n = 56) received phototherapy plus IVIG, group 2 (control group) received phototherapy alonePrimary outcome: exchange transfusionExchange transfusion rate: group I 7%, group II 28.5%. Late anaemia was not of concern in either groupIVIG dose of 0.5 g/kg No adverse effects related to IVIG administration were recorded
Dagoglu et al,
41 infants all with Rhesus incompatibilityBlinded RCT (single centre)Primary outcome was exchange transfusion, also late anaemiaExchange transfusion 4/22 IVIG group vs 15/19 in controls (RR 0.23, 95% CI 0.09 to 0.58) Red cell concentrate (RCC) transfusion for late anaemia: 22/22 under treatment vs 19/19 controls29 preterm, 12 term. Criteria for RCC transfusion not defined Follow-up and results data were complete
Hofstadler et al,
14 patients with Rhesus haemolytic diseaseProspective studyExchange transfusion11/14 patients avoided exchange transfusion


Haemolytic disease of the newborn (HDN) is an isoimmune haemolytic jaundice which prior to modern interventions had a perinatal mortality rate of 50% that has now decreased to 7 per 100 000 births (Clarke). The mainstay of treatment is intensive phototherapy with exchange transfusion recommended if, in spite of phototherapy, the total serum bilirubin (TSB) approaches a level considered to cause a risk of bilirubin encephalopathy (Ip). However, Patra found that 74% of exchange transfusions were associated with an adverse event, most commonly thrombocytopenia (44%), hypocalcaemia (29%) and metabolic acidosis (24%).

Intravenous immunoglobulin (IVIG) has been used sporadically in the treatment of HDN to avoid exchange transfusion since the early 1990s (Sato). The exact mechanism of action is unknown but it is thought to inhibit haemolysis by blocking antibody receptors on red blood cells: IVIG occupies the Fc receptor sites thus competing with anti-D sensitised neonatal erythrocytes and preventing further haemolysis (Urbaniak). The 2004 American Academy of Pediatrics (AAP) guidelines recommend that in isoimmune haemolytic disease IVIG (0.5–1 g/kg over 2 h) should be administered if the total serum bilirubin (TSB) is rising despite intensive phototherapy or the TSB level is within 34–51 µmol/l (2–3 mg/dl) of the exchange level (Gottstein). If necessary, this dose can be repeated in 12 h.

The use of IVIG to avoid exchange transfusion in HDN has been explored in seven randomised controlled trials (RCT) and a Cochrane review for three of these studies (Alcock ). All RCTs showed a statistically significant reduction in the need for exchange transfusion with the use of IVIG, although Nasseri et al did find that when analysed separately, patients with ABO haemolytic disease showed no significant difference between IVIG and double surface blue light phototherapy. In Voto et al, however, the randomisation method is unclear, and in Rübo et al there was failure to report on an intention to treat basis. Only one study (Tanyer) looked at single versus multiple dose IVIG, and found that multiple dose IVIG resulted in a greater percentage reduction in the need for exchange transfusion.

IVIG is considered to be a reasonably safe product and is currently licensed and recommended for paediatric use in cases of fetal hydrops, primary immunodeficiencies, Kawasaki disease, alloimmune thrombocytopenia, autoimmune thrombocytopenic purpura and Gillian-Barre and in adult HIV related/associated thrombocytopenia. However, as a pooled blood product, the risk, albeit small, of transfusion transmitted disease remains despite viral partitioning and inactivation process (Yap). Serious side effects are very rare but include hypersensitivity and anaphylaxis. None of the studies of IVIG use for HDN reported any serious side effects. Only the study by Alpay et al reported any adverse effects, both of which were mild (one patient with hypocalcaemia and one with hypoglycaemia). None of these studies assessed long term outcome.

Two prospective studies again showed a reduction in the need for exchange transfusion (Mukhopadhyay, Hammerman). However, the criteria for administering IVIG in Hammerman et al resulted in it being administered at a lower SBR than in other studies. In Walsh et al (n = 11) there was a significant fall in SBR after IVIG administration and no exchange transfusions were required. Aggarwal et al (n = 26) used a cohort of patients all of whom received exchange transfusions within 6 h of birth and showed that IVIG administration reduced the need for subsequent exchange transfusions. The case series by Ergaz et al found a reduction in carboxyhaemoglobin levels (>30%) with the use of IVIG and exchange transfusion was avoided. This supports the theory that attenuation of jaundice by IVIG is, at least in part, attributable to a reduction in haemolysis. Additional studies also did not reveal any side effects of therapy with IVIG (Aylin, Sebag, Mundy, Merchant).

The NHS recently produced clinical guidelines for the use of immunoglobulin due to limited supply. The report suggests that IVIG may be used in selected cases of HDN with worsening hyperbilirubinaemia, terming this a grade B recommendation, level III evidence. The report references two studies (Ip, Alcock), one of which is a meta-analysis of RCTs (grade A, level 1a evidence). The other is the report of the AAP expert advisory group giving the number needed to treat with IVIG to prevent one exchange transfusion as 2.7. Therefore, the evidence presented by the NHS supports the selected use of IVIG in cases of HDN with worsening hyperbilirubinaemia.

Editor Comment

COHbc, carboxyhaemoglobin; DCT, direct Coombs test; HDN, haemolytic disease of the newborn; IVIG, intravenous immunoglobulin; NNT, numbers needed to treat; PCV, packed cell volume; RCC, red cell concentrate; RCT, randomised controlled trial; RR, relative risk; SBR, serum bilirubin; WMD, weighted mean difference.

Clinical Bottom Line

Intravenous immunoglobulin (IVIG) is a relatively safe and effective means of reducing the need for exchange transfusion in haemolytic disease of the newborn. (Grade A)

The American Academy of Pediatrics 2004 guidelines recommend administration of IVIG in isoimmune haemolytic disease if the total serum bilirubin (TSB) is rising despite intensive phototherapy or the TSB level is within 34–51 µmol/l (2–3 mg/dl) of the exchange level.


  1. Clarke CA, Mollison PL, Whitfield AG. Deaths from rhesus haemolytic disease in England and Wales in 1982 and 1983. BMJ 1985;291:17–19.
  2. Ip S, Chung M, Kulig J, et al. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297–316.
  3. Patra K. . Adverse events associated with neonatal exchange transfusion in the 1990s Pediatr 2004;144:626–31.
  4. Sato K, Hara T, Kondo T, et al. High-dose intravenous gammaglobulin therapy for neonatal immune haemolytic jaundice due to blood group incompatibility. Acta Paediatr Scand 1991;80:163–6.
  5. Urbaniak SJ. ADCC (K-cell) lysis of human erythrocytes sensitized with rhesus alloantibodies. II. Investigation into the mechanism of lysis. Br J Haematol 1979;42:315–25.
  6. Gottstein R, Cooke RWI. Systematic review of intravenous immunoglobulin in haemolytic disease of the newborn. Arch Dis Child Fetal Neonatal Ed 2003;88:F6–F10.
  7. Alcock GS, Liley H. Immunoglobulin infusion for isoimmune haemolytic jaundice in neonates. Cochrane Database Syst Rev 2002;(3):CD003313.
  8. Nasseri R, Mamouri GA, Babaei H. Intravenous immunoglobulin in ABO and Rh hemolytic diseases of newborn. Saudi Med J 2006;27:1827–30.
  9. Voto LS, Sexer H, Ferreiro G, et al. Neonatal administration of high-dose intravenous immunoglobulin in rhesus hemolytic disease. J Perinat Med 1995;23:443–51.
  10. Rübo J, Albrecht K, Lasch P, et al. High-dose intravenous immune globulin therapy for hyperbilirubinemia caused by Rh hemolytic disease. J Pediatr 1992;121:93–7.
  11. Tanyer G, Siklar Z, Dallar Y, et al. Multiple dose IVIG treatment in neonatal immune hemolytic jaundice. Trop Pediatr 2001;47:50–3.
  12. Yap PL. Intravenous immunoglobulin and hepatitis C virus: an overview of transmission episodes with emphasis on manufacturing data. Clin Ther 1996;18:43–58.
  13. Alpay F, Sarici SÜ, Okutan V, et al. High-dose intravenous immunoglobulin therapy in neonatal immune haemolytic jaundice. Acta Paediatr 1999;88:216–19.
  14. Mukhopadhyay K, Murki S, Narang A, et al. Intravenous immunoglobulins in rhesus hemolytic disease. Indian J Pediatr 2003;70:697–9.
  15. Hammerman C, Vreman HJ, Kaplan M, et al. Intravenous immune globulin in neonatal immune hemolytic disease: does it reduce hemolysis? Acta Paediatr 1996;85:1351–3.
  16. Walsh SA, Yao N, El-Khuffash A, et al. Efficacy of intravenous immunoglobulin in the management of haemolytic disease of the newborn. Ir Med J 2008;101:46–8.
  17. Aggarwal R, Seth R, Paul VK, et al. High dose intravenous immunoglobulin therapy in the treatment of rhesus hemolytic disease. J Trop Pediatr 2002;48:116–17.
  18. Ergaz Z, Gross D, Bar-Oz B, et al. Carboxyhemoglobin levels in neonatal immune hemolytic jaundice treated with intravenous gammaglobulin. Vox Sang 1995;69:95–9.
  19. Aylin T, Berkan G. Role of IVIG in preventing exchange transfusions in Rh hemolytic disease. Indian Pediatr 2005;42:77–9.
  20. Sebag O, Sirvent N, Dupuy M, et al. Severe newborn icterus due to Rhesus incompatibility. Treatment with intravenous immunoglobulins. Arch Pediatr 1999;6:345.
  21. Mundy CA. Intravenous immunoglobulin in the management of hemolytic disease of the newborn. Neonatal Netw 2005;24:17–24.
  22. Merchant RH, Pradeep S. Intravenous immunoglobulin therapy for hyperbilirubinemia caused by Rhesus hemolytic disease. Indian Pediatr 1994;31:1269–71.
  23. NHS. Clinical guidelines for immunoglobulin use . 2nd edn. London: Department of Health, June 2008. = 289918 (accessed 26 June 2009).
  24. Miqdad AM, Abdelbasit OB, Shaheed MM. Intravenous immunoglobulin G (IVIG) therapy for significant hyperbilirubinemia in ABO hemolytic disease of the newborn. J Matern Fetal Neonatal Med 2004;16:163–6.
  25. Dagoglu T, Ovali F, Samaci N, et al. High-dose intravenous immunoglobulin therapy for rhesus haemolytic disease. J Int Med Res 1995;23:264–71.
  26. Hofstadler G, Hohenauer L. High dosage immunoglobulin therapy in Rhesus incompatibility. Klin Padiatr 1995;207:103–5.