Skip to main content

Key messages

  • Uncertainty remains about the precise timing and basis for transfusion.
  • There are no simple clinical or laboratory indicators that determine the necessity for blood transfusion.
  • The theoretical critical haemoglobin (Hb) or haematocrit (Hct) threshold level required to maintain tissue oxygenation will vary dependent on the form of hypoxaemia.
  • There is no good evidence to support the use of transfusion for weight gain, apnoea requiring ventilation or dependency on supplemental oxygen.

Please note that all guidance is currently under review and some may be out of date. We recommend that you also refer to more contemporaneous evidence in the interim.

Red blood cell (RBC) transfusions in neonates are used to treat hypovolaemic anaemic shock and normovolaemic anaemia of prematurity, secondary to:

  • delayed and reduced RBC production
  • shortened RBC survival
  • rapid postnatal growth
  • iatrogenic loss from frequent blood sampling for laboratory monitoring.

When to transfuse

There are no simple clinical or laboratory indicators that determine the necessity for blood transfusion.

Oxygen delivery

As the consideration of transfusion includes the combination of oxygen carrying and delivery it should be noted that oxygen delivery is primarily determined by:

  • cardiac output
  • haemoglobin concentration
  • haemoglobin oxygen saturation.

Final oxygen delivery is further dependent on:

  • the oxygen diffusion gradient (determined in part by the haemoglobin oxygen affinity)
  • the diffusion distance between the capillary and the cell
  • cellular uptake mechanisms.

The theoretical critical haemoglobin (Hb) or haematocrit (Hct) threshold level required to maintain tissue oxygenation will vary dependent on the form of hypoxaemia.

Transfusion - the research

A summary of recent research includes:

  • Historically, transfusion guidelines have been shown to reduce transfusion rates in very premature newborns without worsening outcome though care should be taken when interpreting these reports.
  • Two recent trials have examined haemoglobin transfusion thresholds in very premature newborns.
  • The randomised PINT (Premature Infants in Need of Transfusion) trial included 451 newborns < 1,000 g birthweight who were less than 48 hours of age. The trial compared two different transfusion algorithms (liberal vs restrictive) based on chronologic age, haemoglobin threshold and need for respiratory support.  The trial showed no difference in hospital mortality or survival with BPD, vision threatening ROP or brain injury on head ultrasound. At two-year follow-up there was no significant difference in the composite outcome of death, cerebral palsy, cognitive delay, hearing loss or blindness.
  • The other large randomised trial included 100 newborns between 500 and 1,300 g birthweight. This trial compared slightly different transfusion algorithms based on haematocrit thresholds that varied by need for respiratory support. These targets were also based on the spectrum of clinical practice at the time. This trial showed no difference to in-hospital survival, ROP, BPD, time in supplemental oxygen, apnoea requiring ventilation or growth.
  • Unfortunately, neither trial provides a definitive answer as to when to transfuse a premature newborn with normovolaemic anaemia of prematurity. Both trials show a statistically non-significant trend towards an advantage in the liberal transfusion arm with no difference in blood donor exposure.

Research summary

In summary:

  • There is no good evidence to support the use of transfusion for weight gain, apnoea requiring ventilation or dependency on supplemental oxygen.
  • Despite these trials there remains uncertainty about the precise timing and basis for transfusion in the context of anaemia of prematurity.

Complications of transfusion

Potential adverse effects of transfusion include:

  • volume overload
  • immunosensitisation
  • graft versus host disease
  • metabolic derangements (hyperkalaemia, hypocalacaemia)
  • transmission of infection (cases per million donations)
    • HIV (0.79)
    • hepatitis C (4.27)
    • hepatitis B (2.71)
  • transfusion related acute lung injury (TRALI)
  • necrotising enterocolitis (NEC) has occasionally occurred post transfusion.

Minimise the risks

You can minimise risks of transfusion if you:

  • Use guidelines for RBC transfusion as these can significantly reduce the number and volume of RBC transfusions. Although based on good quality randomised trials these algorithms were derived from variations in clinical practice.
  • Use Pedi packs to reduce donor exposure.
  • Request red cells less than five days old at cross-match.
  • Commence transfusion within 30 minutes of receiving blood and complete within four hours of ‘spiking’ pack.
  • Use CMV negative blood. The UK Transfusion Services guidelines (2002) recommend using CMV negative transfusions in the first year of life. However, leucocyte depleted (< 5 x 106/unit) products may significantly reduce the risk of CMV transmission. Those at greatest risk of CMV transmission include:
    • fetuses
    • infants < 1,500 g
    • immunodeficient patients
    • stem cell transplant recipients
  • Use leucocyte depleted products as an acceptable alternative if CMV negative products are not available (60-70 per cent of adults are CMV positive),
  • Use irradiated blood. Red cells are irradiated for:
    • exchange transfusion
    • after intrauterine transfusion
    • directed donation (first or second-degree relative) or human leucocyte antigen (HLA) matched donor
    • when infant has suspected or proven immunodeficiency.

Use red cells within 24 hours

Red cells should be used within 24 hours of irradiation. It is the responsibility of the doctor ordering the transfusion to request irradiated red cells. There is no evidence that transfusion with directed donor blood is safer for neonates.

Repeated cross-matching is not necessary

The Australian and New Zealand Society of Blood Transfusion (ASBT) protocol avoids the need for repeated cross matching prior to transfusion in the first four months of life (development of antibodies to red cell antigens is very uncommon in the first four months of life).

Eligibility for ASBT

Once an infant is on the ASBT protocol no further samples are required for pre-transfusion testing and blood may be ordered by telephoning the blood bank. Infants are eligible for ASBT if they are:

  • less than four months old
  • have pre-transfusion ABO and Rh (D) group performed
  • are DAT negative
  • have no atypical red cell antibodies
  • have one continuous admission (babies that are discharged and readmitted must requalify for ASBT).


Investigation required include:

  • haemoglobin (Hb)
  • haematocrit (Hct)
  • reticulocyte count
  • ABO and Rhesus grouping of baby and mother.

Management of normovolaemic anaemia of prematurity

Consideration should be given to following the liberal algorithm of the PINT study. Issues to note:

  • This is an operational guideline only.
  • Consideration should be given to non-algorithm compliant transfusion in the event of poor circulation, shock (of any form), surgery and coagulopathy.

PINT liberal haemoglobin threshold algorithm

Note that there is a difference in threshold dependent on the site of collection (central [arterial/ venous] vs capillary). See Table 1.
Respiratory support was defined as:

  • need for mechanical ventilation by any means (including CPAP)
  • headbox oxygen > 40 per cent
  • intranasal oxygen > 0.25 L/min.
   If respiratory support is required If respiratory support  is NOT required 


Capillary haemoglobin Central haemoglobin Capillary haemoglobin Central haemoglobin
Days 0-7 135 g/L or less 122 g/L or less 120 g/L or less 109 g/L or less
Days 8-14 120 g/L or less 109 g/L or less 109 g/L or less 90 g/L or less
Days 15 to discharge 100 g/L or less 90 g/L or less 85 g/L or less 77 g/L or less

Table 1

Transfusion issues to note

Others points to note about transfusion of neonates:

  • Do not transfuse to replace blood removed for laboratory testing.
  • Parents must be informed of the need to transfuse.
  • The exact volume of transfusion is not precisely known. A volume of 20 mL/kg of packed RBC over four hours would reflect current practice.
  • Consideration may be given to Frusemide 1 mg/kg mid-way through transfusion in infants with critical circulatory status.
  • Feeds do not need to be suspended routinely during transfusion.

Acute blood loss

In cases of acute blood loss:

  • Transfuse with whole blood or packed RBC 20 mL/kg or estimated volume of blood loss titrated to infants response to transfusion.
  • This should be given over 30 minutes.
  • In emergencies, non-cross matched O Rhesus negative blood may be transfused.

Areas of uncertainty in clinical practice

Areas of uncertainty include:

  • The use of recombinant erythropoietin remains controversial.
  • Refer to the Cochrane review on the topic for thorough discussion.
  • Erythropoietin and iron/folate therapy may be considered in exceptional circumstances such as religious opposition to transfusion.

More information



  • Bell EF, et al. Randomized trial of liberal versus restiricive guidelines for red blood cell transfusions in preterm infants. Pediatrics 2005; 115: 1685-1691.
  • Ramasethu J, Luban LC. Red blood cell transfusions in the newborn. Semin Neonatol 1999;4:5-16
  • Fetus and Newborn Committee, Canadian Paediatric Society. Guidelines for transfusion of erythrocytes to neonates and premature infants. Can Med Assoc J 1992;147:1781-6
  • Franz AR. Pohlandt F. Red blood cell transfusions in very and extremely low birth weight infants under restrictive guidelines: is exogenous erythropoietin necessary? Arch Dis Child Fetal Neonatal Ed 2001; 84: F96-100
  • Hume H. Red blood cell transfusions for preterm infants: the role of evidence-based medicaine. Semin Perinatol 1997; 21: 8-19
  • MaierRF et al Changing practices of red blood cell transfusions in infants with birth weights less than 1000g. J Pediatr. 2000 Feb; 136(2): 220-4
  • Kirpalani H et al The premature Infants in Need of transfusion (PINT) study: a randomized, controllde trial of restricitve (low) versus liberal (high) transfusion threshold for extremely low birth weight infants. JPediatr. 2006 Sep; 149(3): 301-307.
  • Paul D, Leef K, Locke R, Stefano J. Transfusion Volume in Infants with very Low Birth Weight: A Randominzed Trial of 10 versus 20mL/kg J Pediatr Hematol Oncol 2002; 24(1) 43-46

Other reading

  • Ramasethu J, Luban LC. Red blood cell transfusions in the newborn. Semin Neonatol 1999;4:5-16

Get in touch

Centre of Clinical Excellence - Women and Children
Safer Care Victoria

Version history

First published: August 2013

Last reviewed: October 2018

Review by: TBC

Uncontrolled when downloaded

Page last updated: 17 Feb 2021

Was this content helpful to you?