IVIG to Prevent Exchange Transfusion in Infants With Indirect Hyperbilirubinemia
| Citation and Setting . | Methods . | Participants or Inclusion Criteria . | Exclusion Criteria . | Interventions . | Outcomes . | Results . | Risk of Bias . | Notes or Conclusions . |
|---|---|---|---|---|---|---|---|---|
| ABO Isoimmunization | ||||||||
| Miqdad 2004 Riyadh, Saudi Arabia | Randomized controlled trial | 112 newborn infants born in a single-center between 2000-2002 at ≥36 wk’ gestation with ABO incompatibility and a positive direct antiglobulin test (DAT) with serum bilirubin rate of increase ≥0.5 mg/dL/hour. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG or phototherapy alone. | Exchange transfusion (conducted in both groups if the serum bilirubin was ≥20 mg/dL). (Exchange transfusion was also conducted in the phototherapy-only control group [not the IVIG group] if the serum bilirubin rate of increase was ≥0.5 mg/dL/hour). | Exchange transfusion risk in IVIG exposed: 7.1%. Risk in placebo group: 28.6%. Risk ratio: 0.25 (95% CI: 0.09–0.7). Risk difference: −21.4 (95% CI: −35.1 to −7.8). No adverse events noted. | The investigators and clinicians were not blinded to treatment. The control group had an additional trigger for exchange transfusion based on bilirubin rate of increase. This increased the likelihood that infants in the control group might qualify for exchange transfusion. | IVIG treated infants with ABO incompatibility were less likely to receive an exchange transfusion than those in the control group. It should be noted that by design, the control group had an increased opportunity to qualify for an exchange transfusion (bilirubin rate of rise ≥0.5 mg/dL/hour or serum bilirubin ≥20 mg/dL), relative to the IVIG treated group (exchange transfusion only for serum bilirubin ≥20 mg/dL). The number of infants in the control group that were exchanged because of bilirubin rate of increase (≥0.5 mg/dL/hr) rather than a serum bilirubin ≥20 mg/dL was not reported. |
| Rhesus (Rh) Isoimmun-ization | ||||||||
| Rübo 1992 Germany | Randomized controlled trial | 32 term and preterm newborn infants with positive direct antiglobulin tests born at 11 German children’s hospitals between 1989-1990. Treatment began as soon as Rh-isoimmunization was established and consent was obtained. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG, as soon as Rh- isoimmunization was established and informed consent obtained, or to phototherapy alone. | Exchange transfusion (conducted if bilirubin concentrations exceeded the modified curves of Poláceka by more than 2 mg/dL). | Exchange transfusion risk in IVIG exposed: 12.5%. Risk in placebo group: 68.8%. Risk ratio: 0.18 (95% CI: 0.05–0.69). Risk difference: -56.3% (95% CI: -84.2 to -28.4). No adverse events noted. | The investigators and clinicians were not blinded to treatment. Randomization concealment and allocation process not described. Noted that 1 child in each group was excluded for unnamed protocol violations. The study was originally planned for 34 infants. | IVIG treated infants with Rh-isoimmunization were less likely to receive an exchange transfusion than those in the control group. The authors reported that the difference in birth wt or age between groups was not statistically significant. However, there was no report of the number of infants by specific gestational age. Of the 13 infants that received exchange transfusion, 8 were >2.5 kg birth wt term infants and 5 were <2.5 kg preterm infants. |
| Dağoğlu 1995 Istanbul, Turkey | Randomized controlled trial | 41 newborn infants (mean gestational age: 36.1 wk) with Rh- incompatibility and a positive DAT who were admitted to a single-center between August 1992- and June 1994. Mean gestation of treated group: 36.1 wk; SD: 2.3. Mean gestation of untreated group: 36.1 wk; SD: 1.7. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG as soon as possible after birth (usually within 2 h) or to phototherapy alone. Phototherapy was initiated when bilirubin concentrations exceeded the relevant curves of Oski and Naiman.b IVIG treatment was usually administered within 2 h postnatal. | Exchange transfusion (conducted if the increase of total bilirubin exceeded 1 mg/dL/hour or when, during the first 72 h, the total concentration exceeded 20 mg/dL in term infants or 18 mg/dL in infants weighing more than 2000 g). | Exchange transfusion risk in IVIG exposed: 20%. Risk in placebo group: 75%. Risk ratio: 0.27 (95% CI: 0.11–0.66). Risk difference: −55% (95% CI: -80.8 to -29.17). No adverse events noted. | The investigators and clinicians were not blinded to treatment. | IVIG treated infants with Rh- isoimmunization were less likely to receive an exchange transfusion than those in the control group Mothers of enrolled infants received 2–3 in-utero transfusions on average, before delivery for treatment of Rh-mediated hemolysis. |
| Elalfy 2011 Cairo, Egypt | Randomized controlled trial | 90 newborns with Rh- isoimmunization who were born at >38weeks’ gestation with serum bilirubin increasing by 0.5 mg/dL/hour and/or required phototherapy in the first 12 h postnatal, who were not yet eligible for exchange transfusion and who were admitted to a single center between March 2009 and January 2010. | Perinatal asphyxia, neonatal sepsis, or presence of hematomas. | Randomized at 12 h to 500 mg/kg IVIG plus phototherapy (n = 23 assigned; n = 25 treated), 1 g/kg IVIG plus phototherapy (n = 22 assigned; n = 15 treated) or phototherapy only (n = 45 assigned; n = 50 treated). Parents of 5 infants assigned to the intervention group were reassigned to the control group when their parents refused IVIG and 5 babies assigned to the high-dose IVIG group were assigned to the low-dose IVIG group per parental wishes. Parental reasoning for refusing treatment or switching dosage groups was not provided. | Exchange transfusion (conducted per the 2004 AAP exchange transfusion thresholds if the bilirubin was increasing by ≥1 mg/d/hour). | Exchange transfusion risk in IVIG exposed (inclusive of low- and high-dose treatment groups): 5%. Risk in placebo group: 22%. Risk ratio: 0.23 (95% CI: 0.05–0.97). Risk difference: −17% (95% CI: -30.3 to -3.7). The 2 IVIG-treated infants who received an exchange transfusion were both in the low-dose (500 mg/kg) group. Risk difference between high- versus low-dose treatment of preventing an exchange transfusion: -8% (95% CI: -18.6 to 2.63). Note: risk ratio not calculable because of no events in high-dose group. No adverse events noted. | The investigators and clinicians were not blinded to treatment. Parents were able to refuse assigned treatment after study enrollment and receive treatment in a different group. Small sample sizes in the high- and low-dose treatment groups reduced the power for comparison of the effect of low- vs high-dose IVIG dosing for preventing exchange transfusion. | IVIG-treated infants with Rh-incompatibility were less likely to receive an exchange transfusion than those in the control group. It should be noted that the original randomization assignment was not followed for all enrolled infants. An intention to treat analysis was not provided and authors did not respond to emailed queries requesting one. No data on outcomes according to original treatment assignment were provided. None of the mothers of enrolled infants had a history of antenatal IVIG treatment or in utero transfusion. |
| Smits-Wintjens 2011 Leiden, Netherlands | Randomized, double-blinded, placebo-controlled trial | 80 newborn infants born at ≥35 wk’ gestation and admitted to a single center between 2006 and 2010 with maternal and neonatal testing consistent with Rhesus D- or anti C-mediated hemolytic disease of the newborn (HDN). Rhesus HDN defined as (1) a maternal antibody-dependent cellular cytotoxicity test with a>50% result (comparable with an antibody titer of >1:64)c and (2) a direct antiglobulin test with positive results caused by antirhesus D or c antibodies in the fetus/neonate of a rhesus D or c-negative mother. | Infants with perinatal asphyxia (5-min Apgar <3 or cord pH <7.0), hemolytic disease other than rhesus D or c, and rhesus hemolytic disease presenting >4 h after birth. | Randomized to phototherapy and IVIG, 750 mg/kg within 4 h postnatal, or phototherapy plus the same vol of a 5% dextrose placebo. Both groups received intensive white light phototherapy with an intensity of 12 to 20 μW/cm2 per nm given by air shield and Ohmeda lamps in combination with blue-light bilirubin blanket phototherapy (30 μW/cm2 per nm). | Primary: need for exchange transfusion (per the 2004 AAP recommended exchange transfusion levels) and the number of exchange transfusions per infant. Secondary: duration of phototherapy and hospital stay, maximum serum bilirubin concentrations, and need for red-cell transfusion in the first 3 mo postnatal. | Exchange transfusion risk in IVIG exposed infants: 17.1%. Risk in placebo group: 15.4%. Risk ratio: 1.11 (95% CI: 0.41–3.01). Risk difference: 1.69% (95% CI: -14.5% to 17.8%). No differences between groups in secondary outcomes. No adverse events noted. | Double-blinded. Adequate allocation concealment. Inadequate power: confidence interval includes clinically meaningful absolute risk reduction -14%). | No difference in exchange transfusion risk or in measured secondary outcomes were detected between groups randomized to IVIG treatment versus placebo. |
| Santos 2013 Rio de Janeiro, Brazil | Randomized, double-blinded, placebo-controlled trial | 92 newborn infants born at ≥32 wk’ gestation admitted to a single-center between April 2006 and June 2009 with a positive DAT whose mothers where Rh-alloimmunized. Mean gestation of treated group: 36.5 wk; SD: 1.5. Mean gestation of placebo group: 36.1 wk; SD: 1.6. | Hydrops fetalis, cardiac instability, severe asphyxia, other non-Rh red blood cell antibodies, initial bilirubin >4 mg/dL and/or cord hematocrit <30%. | Randomized to phototherapy plus IVIG, 500 mg/kg over 2-h shortly after birth, versus phototherapy plus the same vol of a normal saline placebo. Both groups received high-intensity blue fluorescent phototherapy. | Primary: need for exchange transfusion (at a serum bilirubin concentration of 20 mg/dL or increase ≥0.5 mg/dL/hour). Secondary: peak bilirubin, phototherapy duration, length of stay, drug, or drug administration related adverse effects or events. | Exchange transfusion risk in IVIG-exposed infants: 13%. Risk in placebo group: 15.2%. Risk ratio: 0.857 (95% CI: 0.32–2.36). Risk difference: −2.17% (95% CI: -16.4% to 12.1%). No differences between groups in secondary outcomes. No adverse events noted. | Double-blinded. Adequate allocation concealment. Inadequate power: confidence interval includes clinically meaningful absolute risk reduction -16%). | No difference in exchange transfusion risk or in measured secondary outcomes were detected between groups randomized to IVIG treatment versus placebo. |
| Rh- and ABO isoimmunization | ||||||||
| Alpay 1999 Ankara, Turkey | Randomized controlled trial (RCT) | 116 term newborn infants with ABO and/or Rh incompatibility with bilirubin >11.9 mg/dL, positive DAT, and reticulocyte count ≥10% who were admitted to a single-center between March 1992 and November 1996. Mean postnatal age of treated group: 51.5 h; SD: 26.7. Mean postnatal age of untreated group 54.3 h; SD: 30.5. | As per inclusion criteria. | Randomized to conventional phototherapy plus a single 1 g/kg dose of IVIG or to conventional phototherapy alone. IVIG treatment was initiated soon after the inclusion criteria were met. | Exchange transfusion (conducted if serum bilirubin exceeded 16.96 mg/dL and increased ≥1 mg/dL/hour despite phototherapy). Duration of phototherapy. Duration of hospitalization. | Exchange transfusion risk in IVIG exposed: 13.8%. Risk in placebo group: 37.9%. Risk ratio: 0.36 (95% CI: 0.18–0.75). Risk difference: −24.1% (95% CI: -39.5% to -8.82%). Duration of phototherapy and duration of hospitalization were shorter in the IVIG-treated group (P < .05). No adverse events were reported. | The investigators and clinicians were not blinded to treatment. Randomization assignment was done by a neonatologist different from those conducting the study. Randomization concealment and allocation process not otherwise described. | IVIG treated infants with ABO or Rh-incompatibility were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (80%),14% had Rh incompatibility, and 6% had ABO and Rh incompatibility. |
| Tanyer 2001 Ankara, Turkey | Quasi-randomized trial | 61 full-term newborn infants admitted to a single center between January 1996 and November 1998 with Rh-, ABO-, or subgroup- incompatibility, bilirubin concentrations below the exchange transfusion criteria at admission, and a positive DAT. | Contributing risk factors (such as sepsis, drug use by mothers) that could raise bilirubin concentrations. | Assigned by order of admission to phototherapy plus 500 mg/kg multiple-dose IVIG treatment (daily on 3-consecutive days), phototherapy plus single-dose 500 mg/kg IVIG treatment, or phototherapy alone. Initial IVIG doses were administered between 2–4 h of admission. | Exchange transfusion (conducted if bilirubin exceeded limits on included table [20 mg/dL for low-risk infants at >2500 g; 18 mg/dL for low-risk infants at 2000–2499 g]). | Exchange transfusion risk in IVIG exposed (inclusive of single- and multiple-dose treatment groups): 7.5%. Risk in placebo group: 33.3%. Risk ratio: 0.23 (95% CI: 0.06–0.78). Risk difference: -25.8 (95% CI: -47.6% to -4.1). Risk difference between multiple- vs single-dose treatment of preventing an exchange transfusion: -15% (95% CI: -30.7 to 0.65). Note: risk ratio not calculable due to no events in multidose group. No adverse events noted. | The investigators and clinicians were not blinded to treatment. Treatment assignment was per order of admission rather than by a randomization process. Small sample sizes in the multiple- and single-dose treatment groups reduced the power for comparison of the effect of multiple- versus single-dose IVIG dosing for preventing exchange transfusion. | IVIG treated infants with Rh-, ABO-, or subgroup incompatibility were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (55.7%), 29.5% had Rh incompatibility, 3.3% had subgroup incompatibility, and 11.5% had more than 1 incompatibility type. |
| Nasseri 2006 Mashdad, Iran | Randomized controlled trial | 34 newborn infants at a single center between October 2003 and October 2004 who were Born at ≥37 wk’ gestation with Rh- or ABO- incompatibility and had a positive DAT and a serum bilirubin rate of increase >0.5 mg/dL/hour. | Sepsis, glucose-6-phosphate dehydrogenase (G6PD) deficiency, or other illnesses that could result in increased bilirubin concentrations. | Randomized to phototherapy plus 500 mg/kg IVIG within 2 h of admission and then every 12 h for 3 total doses or to phototherapy alone. | Exchange transfusion (conducted if serum bilirubin was ≥20 mg/dL or for a rate of rise >1 mg/dL/hour). | Exchange transfusion risk in IVIG exposed: 17.7%. Risk in placebo group: 64.7%. Risk ratio: 0.27 (95% CI: 0.09–0.8). Risk difference: -47.1 (95% CI: -76.1 to -18.0). No adverse events noted. | The investigators and clinicians were not blinded to treatment. Manuscript describes randomization in multiple places, but also describes the study as a prospective, case-control study in the Methods section. Randomization concealment and allocation process not described. | IVIG treated infants with ABO or Rh-isoimmunization were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (62%). The remaining 38% had Rh-incompatibility. |
| Systematic Reviews | ||||||||
| Walsh 2009 | Systematic review | Manuscripts on neonatal hyperbilirubinemia and immunoglobulin. Restricted to humans, newborn infants, and English. Included 7 randomized controlled trials and 3 prospective observational studies, 2 retrospective reviews, and a single case series. | As per inclusion criteria. | IVIG administration to patients with isoimmune hemolytic jaundice. | Effective reduction of serum bilirubin concentrations. Avoidance of exchange transfusion. | The authors noted a large degree of heterogeneity between studies. Overall, they concluded that IVIG is a relatively safe and effective means of reducing the need for exchange transfusion. | Search terms specified. Loose inclusion criteria based on broad search. | IVIG is relatively safe and effective means of reducing the need for exchange transfusion. There is a large degree of heterogeneity between the included studies. Note: This review was conducted before the placebo-controlled RCTs by Santos and Smit-Wintjens. |
| Zwiers 2018 | Systematic review | Randomized trials of IVIG for treatment of alloimmune hyperbilirubinemia of the newborn infant. Trials must have used predefined criteria for IVIG use and exchange transfusion. (included 9 RCTs published between 1992 and 2013; 658 total participants). | IVIG treatment of alloimmune thrombocytopenia. Infants in all arms of the included studies received intensive phototherapy. | Primary: need for exchange transfusion. | Overall, found that IVIG reduced exchange transfusion. Risk ratio: 0.35 (95% CI: 0.25–0.49). Risk difference: −0.22 (95% CI: -0.27 to -0.16). Two placebo-controlled RCTs of infants with Rh-incompatibility Risk ratio: 0.98 (95% CI: 0.48–1.98). | Reported that overall RCT evidence was of low quality. Only two RCTs (Santos 2013 and Smits-Wintjens 2011), both on Rh- alloimmunization, used a placebo. | There was a great deal of heterogeneity between studies. Five studies were restricted to Rh-disease, one study was restricted to ABO disease, and three enrolled patients with both ABO and Rh incompatibility. The systematic review conclusions were based primarily on the 2 placebo-controlled trials of infants with Rh-disease (Santos 2013; Smits-Wintjens 2011). The most recent, placebo-controlled trials did not demonstrate a reduced risk for exchange transfusion in Rh-positive infants after IVIG treatment. Overall, there is weak evidence that IVIG may reduce the need for exchange transfusion. Non-Rh mediated forms of hyperbilirubinemia including ABO incompatibility are not well studied. |
| Citation and Setting . | Methods . | Participants or Inclusion Criteria . | Exclusion Criteria . | Interventions . | Outcomes . | Results . | Risk of Bias . | Notes or Conclusions . |
|---|---|---|---|---|---|---|---|---|
| ABO Isoimmunization | ||||||||
| Miqdad 2004 Riyadh, Saudi Arabia | Randomized controlled trial | 112 newborn infants born in a single-center between 2000-2002 at ≥36 wk’ gestation with ABO incompatibility and a positive direct antiglobulin test (DAT) with serum bilirubin rate of increase ≥0.5 mg/dL/hour. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG or phototherapy alone. | Exchange transfusion (conducted in both groups if the serum bilirubin was ≥20 mg/dL). (Exchange transfusion was also conducted in the phototherapy-only control group [not the IVIG group] if the serum bilirubin rate of increase was ≥0.5 mg/dL/hour). | Exchange transfusion risk in IVIG exposed: 7.1%. Risk in placebo group: 28.6%. Risk ratio: 0.25 (95% CI: 0.09–0.7). Risk difference: −21.4 (95% CI: −35.1 to −7.8). No adverse events noted. | The investigators and clinicians were not blinded to treatment. The control group had an additional trigger for exchange transfusion based on bilirubin rate of increase. This increased the likelihood that infants in the control group might qualify for exchange transfusion. | IVIG treated infants with ABO incompatibility were less likely to receive an exchange transfusion than those in the control group. It should be noted that by design, the control group had an increased opportunity to qualify for an exchange transfusion (bilirubin rate of rise ≥0.5 mg/dL/hour or serum bilirubin ≥20 mg/dL), relative to the IVIG treated group (exchange transfusion only for serum bilirubin ≥20 mg/dL). The number of infants in the control group that were exchanged because of bilirubin rate of increase (≥0.5 mg/dL/hr) rather than a serum bilirubin ≥20 mg/dL was not reported. |
| Rhesus (Rh) Isoimmun-ization | ||||||||
| Rübo 1992 Germany | Randomized controlled trial | 32 term and preterm newborn infants with positive direct antiglobulin tests born at 11 German children’s hospitals between 1989-1990. Treatment began as soon as Rh-isoimmunization was established and consent was obtained. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG, as soon as Rh- isoimmunization was established and informed consent obtained, or to phototherapy alone. | Exchange transfusion (conducted if bilirubin concentrations exceeded the modified curves of Poláceka by more than 2 mg/dL). | Exchange transfusion risk in IVIG exposed: 12.5%. Risk in placebo group: 68.8%. Risk ratio: 0.18 (95% CI: 0.05–0.69). Risk difference: -56.3% (95% CI: -84.2 to -28.4). No adverse events noted. | The investigators and clinicians were not blinded to treatment. Randomization concealment and allocation process not described. Noted that 1 child in each group was excluded for unnamed protocol violations. The study was originally planned for 34 infants. | IVIG treated infants with Rh-isoimmunization were less likely to receive an exchange transfusion than those in the control group. The authors reported that the difference in birth wt or age between groups was not statistically significant. However, there was no report of the number of infants by specific gestational age. Of the 13 infants that received exchange transfusion, 8 were >2.5 kg birth wt term infants and 5 were <2.5 kg preterm infants. |
| Dağoğlu 1995 Istanbul, Turkey | Randomized controlled trial | 41 newborn infants (mean gestational age: 36.1 wk) with Rh- incompatibility and a positive DAT who were admitted to a single-center between August 1992- and June 1994. Mean gestation of treated group: 36.1 wk; SD: 2.3. Mean gestation of untreated group: 36.1 wk; SD: 1.7. | As per inclusion criteria. | Randomized to phototherapy plus 500 mg/kg IVIG as soon as possible after birth (usually within 2 h) or to phototherapy alone. Phototherapy was initiated when bilirubin concentrations exceeded the relevant curves of Oski and Naiman.b IVIG treatment was usually administered within 2 h postnatal. | Exchange transfusion (conducted if the increase of total bilirubin exceeded 1 mg/dL/hour or when, during the first 72 h, the total concentration exceeded 20 mg/dL in term infants or 18 mg/dL in infants weighing more than 2000 g). | Exchange transfusion risk in IVIG exposed: 20%. Risk in placebo group: 75%. Risk ratio: 0.27 (95% CI: 0.11–0.66). Risk difference: −55% (95% CI: -80.8 to -29.17). No adverse events noted. | The investigators and clinicians were not blinded to treatment. | IVIG treated infants with Rh- isoimmunization were less likely to receive an exchange transfusion than those in the control group Mothers of enrolled infants received 2–3 in-utero transfusions on average, before delivery for treatment of Rh-mediated hemolysis. |
| Elalfy 2011 Cairo, Egypt | Randomized controlled trial | 90 newborns with Rh- isoimmunization who were born at >38weeks’ gestation with serum bilirubin increasing by 0.5 mg/dL/hour and/or required phototherapy in the first 12 h postnatal, who were not yet eligible for exchange transfusion and who were admitted to a single center between March 2009 and January 2010. | Perinatal asphyxia, neonatal sepsis, or presence of hematomas. | Randomized at 12 h to 500 mg/kg IVIG plus phototherapy (n = 23 assigned; n = 25 treated), 1 g/kg IVIG plus phototherapy (n = 22 assigned; n = 15 treated) or phototherapy only (n = 45 assigned; n = 50 treated). Parents of 5 infants assigned to the intervention group were reassigned to the control group when their parents refused IVIG and 5 babies assigned to the high-dose IVIG group were assigned to the low-dose IVIG group per parental wishes. Parental reasoning for refusing treatment or switching dosage groups was not provided. | Exchange transfusion (conducted per the 2004 AAP exchange transfusion thresholds if the bilirubin was increasing by ≥1 mg/d/hour). | Exchange transfusion risk in IVIG exposed (inclusive of low- and high-dose treatment groups): 5%. Risk in placebo group: 22%. Risk ratio: 0.23 (95% CI: 0.05–0.97). Risk difference: −17% (95% CI: -30.3 to -3.7). The 2 IVIG-treated infants who received an exchange transfusion were both in the low-dose (500 mg/kg) group. Risk difference between high- versus low-dose treatment of preventing an exchange transfusion: -8% (95% CI: -18.6 to 2.63). Note: risk ratio not calculable because of no events in high-dose group. No adverse events noted. | The investigators and clinicians were not blinded to treatment. Parents were able to refuse assigned treatment after study enrollment and receive treatment in a different group. Small sample sizes in the high- and low-dose treatment groups reduced the power for comparison of the effect of low- vs high-dose IVIG dosing for preventing exchange transfusion. | IVIG-treated infants with Rh-incompatibility were less likely to receive an exchange transfusion than those in the control group. It should be noted that the original randomization assignment was not followed for all enrolled infants. An intention to treat analysis was not provided and authors did not respond to emailed queries requesting one. No data on outcomes according to original treatment assignment were provided. None of the mothers of enrolled infants had a history of antenatal IVIG treatment or in utero transfusion. |
| Smits-Wintjens 2011 Leiden, Netherlands | Randomized, double-blinded, placebo-controlled trial | 80 newborn infants born at ≥35 wk’ gestation and admitted to a single center between 2006 and 2010 with maternal and neonatal testing consistent with Rhesus D- or anti C-mediated hemolytic disease of the newborn (HDN). Rhesus HDN defined as (1) a maternal antibody-dependent cellular cytotoxicity test with a>50% result (comparable with an antibody titer of >1:64)c and (2) a direct antiglobulin test with positive results caused by antirhesus D or c antibodies in the fetus/neonate of a rhesus D or c-negative mother. | Infants with perinatal asphyxia (5-min Apgar <3 or cord pH <7.0), hemolytic disease other than rhesus D or c, and rhesus hemolytic disease presenting >4 h after birth. | Randomized to phototherapy and IVIG, 750 mg/kg within 4 h postnatal, or phototherapy plus the same vol of a 5% dextrose placebo. Both groups received intensive white light phototherapy with an intensity of 12 to 20 μW/cm2 per nm given by air shield and Ohmeda lamps in combination with blue-light bilirubin blanket phototherapy (30 μW/cm2 per nm). | Primary: need for exchange transfusion (per the 2004 AAP recommended exchange transfusion levels) and the number of exchange transfusions per infant. Secondary: duration of phototherapy and hospital stay, maximum serum bilirubin concentrations, and need for red-cell transfusion in the first 3 mo postnatal. | Exchange transfusion risk in IVIG exposed infants: 17.1%. Risk in placebo group: 15.4%. Risk ratio: 1.11 (95% CI: 0.41–3.01). Risk difference: 1.69% (95% CI: -14.5% to 17.8%). No differences between groups in secondary outcomes. No adverse events noted. | Double-blinded. Adequate allocation concealment. Inadequate power: confidence interval includes clinically meaningful absolute risk reduction -14%). | No difference in exchange transfusion risk or in measured secondary outcomes were detected between groups randomized to IVIG treatment versus placebo. |
| Santos 2013 Rio de Janeiro, Brazil | Randomized, double-blinded, placebo-controlled trial | 92 newborn infants born at ≥32 wk’ gestation admitted to a single-center between April 2006 and June 2009 with a positive DAT whose mothers where Rh-alloimmunized. Mean gestation of treated group: 36.5 wk; SD: 1.5. Mean gestation of placebo group: 36.1 wk; SD: 1.6. | Hydrops fetalis, cardiac instability, severe asphyxia, other non-Rh red blood cell antibodies, initial bilirubin >4 mg/dL and/or cord hematocrit <30%. | Randomized to phototherapy plus IVIG, 500 mg/kg over 2-h shortly after birth, versus phototherapy plus the same vol of a normal saline placebo. Both groups received high-intensity blue fluorescent phototherapy. | Primary: need for exchange transfusion (at a serum bilirubin concentration of 20 mg/dL or increase ≥0.5 mg/dL/hour). Secondary: peak bilirubin, phototherapy duration, length of stay, drug, or drug administration related adverse effects or events. | Exchange transfusion risk in IVIG-exposed infants: 13%. Risk in placebo group: 15.2%. Risk ratio: 0.857 (95% CI: 0.32–2.36). Risk difference: −2.17% (95% CI: -16.4% to 12.1%). No differences between groups in secondary outcomes. No adverse events noted. | Double-blinded. Adequate allocation concealment. Inadequate power: confidence interval includes clinically meaningful absolute risk reduction -16%). | No difference in exchange transfusion risk or in measured secondary outcomes were detected between groups randomized to IVIG treatment versus placebo. |
| Rh- and ABO isoimmunization | ||||||||
| Alpay 1999 Ankara, Turkey | Randomized controlled trial (RCT) | 116 term newborn infants with ABO and/or Rh incompatibility with bilirubin >11.9 mg/dL, positive DAT, and reticulocyte count ≥10% who were admitted to a single-center between March 1992 and November 1996. Mean postnatal age of treated group: 51.5 h; SD: 26.7. Mean postnatal age of untreated group 54.3 h; SD: 30.5. | As per inclusion criteria. | Randomized to conventional phototherapy plus a single 1 g/kg dose of IVIG or to conventional phototherapy alone. IVIG treatment was initiated soon after the inclusion criteria were met. | Exchange transfusion (conducted if serum bilirubin exceeded 16.96 mg/dL and increased ≥1 mg/dL/hour despite phototherapy). Duration of phototherapy. Duration of hospitalization. | Exchange transfusion risk in IVIG exposed: 13.8%. Risk in placebo group: 37.9%. Risk ratio: 0.36 (95% CI: 0.18–0.75). Risk difference: −24.1% (95% CI: -39.5% to -8.82%). Duration of phototherapy and duration of hospitalization were shorter in the IVIG-treated group (P < .05). No adverse events were reported. | The investigators and clinicians were not blinded to treatment. Randomization assignment was done by a neonatologist different from those conducting the study. Randomization concealment and allocation process not otherwise described. | IVIG treated infants with ABO or Rh-incompatibility were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (80%),14% had Rh incompatibility, and 6% had ABO and Rh incompatibility. |
| Tanyer 2001 Ankara, Turkey | Quasi-randomized trial | 61 full-term newborn infants admitted to a single center between January 1996 and November 1998 with Rh-, ABO-, or subgroup- incompatibility, bilirubin concentrations below the exchange transfusion criteria at admission, and a positive DAT. | Contributing risk factors (such as sepsis, drug use by mothers) that could raise bilirubin concentrations. | Assigned by order of admission to phototherapy plus 500 mg/kg multiple-dose IVIG treatment (daily on 3-consecutive days), phototherapy plus single-dose 500 mg/kg IVIG treatment, or phototherapy alone. Initial IVIG doses were administered between 2–4 h of admission. | Exchange transfusion (conducted if bilirubin exceeded limits on included table [20 mg/dL for low-risk infants at >2500 g; 18 mg/dL for low-risk infants at 2000–2499 g]). | Exchange transfusion risk in IVIG exposed (inclusive of single- and multiple-dose treatment groups): 7.5%. Risk in placebo group: 33.3%. Risk ratio: 0.23 (95% CI: 0.06–0.78). Risk difference: -25.8 (95% CI: -47.6% to -4.1). Risk difference between multiple- vs single-dose treatment of preventing an exchange transfusion: -15% (95% CI: -30.7 to 0.65). Note: risk ratio not calculable due to no events in multidose group. No adverse events noted. | The investigators and clinicians were not blinded to treatment. Treatment assignment was per order of admission rather than by a randomization process. Small sample sizes in the multiple- and single-dose treatment groups reduced the power for comparison of the effect of multiple- versus single-dose IVIG dosing for preventing exchange transfusion. | IVIG treated infants with Rh-, ABO-, or subgroup incompatibility were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (55.7%), 29.5% had Rh incompatibility, 3.3% had subgroup incompatibility, and 11.5% had more than 1 incompatibility type. |
| Nasseri 2006 Mashdad, Iran | Randomized controlled trial | 34 newborn infants at a single center between October 2003 and October 2004 who were Born at ≥37 wk’ gestation with Rh- or ABO- incompatibility and had a positive DAT and a serum bilirubin rate of increase >0.5 mg/dL/hour. | Sepsis, glucose-6-phosphate dehydrogenase (G6PD) deficiency, or other illnesses that could result in increased bilirubin concentrations. | Randomized to phototherapy plus 500 mg/kg IVIG within 2 h of admission and then every 12 h for 3 total doses or to phototherapy alone. | Exchange transfusion (conducted if serum bilirubin was ≥20 mg/dL or for a rate of rise >1 mg/dL/hour). | Exchange transfusion risk in IVIG exposed: 17.7%. Risk in placebo group: 64.7%. Risk ratio: 0.27 (95% CI: 0.09–0.8). Risk difference: -47.1 (95% CI: -76.1 to -18.0). No adverse events noted. | The investigators and clinicians were not blinded to treatment. Manuscript describes randomization in multiple places, but also describes the study as a prospective, case-control study in the Methods section. Randomization concealment and allocation process not described. | IVIG treated infants with ABO or Rh-isoimmunization were less likely to receive an exchange transfusion than those in the control group. The majority of study subjects had ABO incompatibility (62%). The remaining 38% had Rh-incompatibility. |
| Systematic Reviews | ||||||||
| Walsh 2009 | Systematic review | Manuscripts on neonatal hyperbilirubinemia and immunoglobulin. Restricted to humans, newborn infants, and English. Included 7 randomized controlled trials and 3 prospective observational studies, 2 retrospective reviews, and a single case series. | As per inclusion criteria. | IVIG administration to patients with isoimmune hemolytic jaundice. | Effective reduction of serum bilirubin concentrations. Avoidance of exchange transfusion. | The authors noted a large degree of heterogeneity between studies. Overall, they concluded that IVIG is a relatively safe and effective means of reducing the need for exchange transfusion. | Search terms specified. Loose inclusion criteria based on broad search. | IVIG is relatively safe and effective means of reducing the need for exchange transfusion. There is a large degree of heterogeneity between the included studies. Note: This review was conducted before the placebo-controlled RCTs by Santos and Smit-Wintjens. |
| Zwiers 2018 | Systematic review | Randomized trials of IVIG for treatment of alloimmune hyperbilirubinemia of the newborn infant. Trials must have used predefined criteria for IVIG use and exchange transfusion. (included 9 RCTs published between 1992 and 2013; 658 total participants). | IVIG treatment of alloimmune thrombocytopenia. Infants in all arms of the included studies received intensive phototherapy. | Primary: need for exchange transfusion. | Overall, found that IVIG reduced exchange transfusion. Risk ratio: 0.35 (95% CI: 0.25–0.49). Risk difference: −0.22 (95% CI: -0.27 to -0.16). Two placebo-controlled RCTs of infants with Rh-incompatibility Risk ratio: 0.98 (95% CI: 0.48–1.98). | Reported that overall RCT evidence was of low quality. Only two RCTs (Santos 2013 and Smits-Wintjens 2011), both on Rh- alloimmunization, used a placebo. | There was a great deal of heterogeneity between studies. Five studies were restricted to Rh-disease, one study was restricted to ABO disease, and three enrolled patients with both ABO and Rh incompatibility. The systematic review conclusions were based primarily on the 2 placebo-controlled trials of infants with Rh-disease (Santos 2013; Smits-Wintjens 2011). The most recent, placebo-controlled trials did not demonstrate a reduced risk for exchange transfusion in Rh-positive infants after IVIG treatment. Overall, there is weak evidence that IVIG may reduce the need for exchange transfusion. Non-Rh mediated forms of hyperbilirubinemia including ABO incompatibility are not well studied. |
Question: What is the effectiveness of IVIG for prevention of exchange transfusion in infants with indirect hyperbilirubinemia?
Population: Neonates born at ≥35 weeks’ gestation with isoimmunization and hyperbilirubinemia who are at risk of exchange transfusion
Intervention: Treatment with IVIG
Comparator: Nontreatment with IVIG
Outcome: Need for exchange transfusion
Included: randomized controlled trials, quasi-randomized trials, and systematic reviews that include randomized trials
Polacek K. Die fruhzeitige Indikationstellung zur Austausch-transfusion bei hamolytischen Neugeborenerkrankungen. Monatsschr Kinderheilkd. 1963;111:6-10; and Polacek K. Das universale Diagramm zur Behandlung der Hyperbilirubinamie der Neugerborenen. Padiatrische Praxis. 1984;29:1-3.
Oski FA, Naiman JL: Erythroblastosis fetalis. In: Oski FA, Naiman JL, eds. Hematologic Problems in the Newborn. Philadelphia: WB Saunders Company;1982:283-346.
Oepkes D, van Kamp IL, Simon MJ, Mesman J, Overbeeke MA, Kanhai HH. Clinical value of an antibody-dependent cell-mediated cytotoxicity assay in the management of Rh D alloimmunization. Am J Obstet Gynecol. 2001;184(5):1015-1020.
Table 2 References:
1. Alpay F, Sarici SU, Okutan V, et al. High-dose intravenous immunoglobulin therapy in neonatal immune hemolytic jaundice. Acta Paediatr. 1999 Feb;88(2):216–219
2. Dağoğlu T, Ovali F, Samanci N, Bengisu E. High-dose intravenous immunoglobulin therapy for rhesus hemolytic disease. J Int Med Res. 1995 Jul–Aug;23(4):264–271
3. Elalfy MS, Elbarbary NS, Abaza HW. Early intravenous immunoglobin (two-dose regimen) in the management of severe Rh hemolytic disease of newborn—a prospective randomized controlled trial. Eur J Pediatr. 2011 Apr;170(4):461–467
4. Miqdad AM, Abdelbasit OB, Shaheed MM, et al. Intravenous immunoglobulin G (IVIG) therapy for significant hyperbilirubinemia in ABO hemolytic disease of the newborn. J Matern Fetal Neonatal Med. 2004 Sep;16(3):163–166
5. Nasseri F, Mamouri GA, Babaei H. Intravenous immunoglobulin in ABO and Rh hemolytic diseases of newborn. Saudi Med J. 2006 Dec;27(12):1827–1830
6. Rübo J, Albrecht K, Lasch P, et al. High-dose intravenous immune globulin therapy for hyperbilirubinemia caused by Rh hemolytic disease. J Pediatr. 1992 Jul;121(1):93–97
7. Santos MC, Sá C, Gomes SC Jr, et al. The efficacy of the use of intravenous human immunoglobulin in Brazilian newborns with rhesus hemolytic disease: a randomized double-blind trial. Transfusion. 2013 Apr;53(4):777–782
8. Smits-Wintjens VE, Walther FJ, Rath ME, et al. Intravenous immunoglobulin in neonates with rhesus hemolytic disease: a randomized controlled trial. Pediatrics. 2011 Apr;127(4):680–686
9. Tanyer G, Siklar Z, Dallar Y, et al. Multiple dose IVIG treatment in neonatal immune hemolytic jaundice. J Trop Pediatr. 2001 Feb;47(1):50–53
10. Walsh S, Molloy EJ. Toward evidence based medicine for pediatricians. Is intravenous immunoglobulin superior to exchange transfusion in the management of hyperbilirubinaemia in term neonates? Arch Dis Child. 2009 Sep;94(9):739–741
11. Zwiers C, Scheffer-Rath ME, Lopriore E, et al. Immunoglobulin for alloimmune hemolytic disease in neonates. Cochrane Database Syst Rev. 2018 Mar 18;3:CD003313