Anemia of Prematurity Treatment & Management

Updated: Jan 08, 2016
  • Author: George Cassady, MD; Chief Editor: Ted Rosenkrantz, MD  more...
  • Print

Approach Considerations

Medical treatment options are blood transfusion(s), recombinant erythropoietin (EPO) treatment, and observation.

Go to Anemia, Pediatric Chronic Anemia, Anemia and Thrombocytopenia in Pregnancy, and Emergent Management of Acute Anemia for complete information on these topics.


Observation and Supportive Care

Observation may be the best course of action for infants who are asymptomatic, not acutely ill, and are receiving adequate nutrition. Adequate amounts of vitamin E, vitamin B-12, folate, and iron are important to blunt the expected decline in hemoglobin levels in the premature infant. Periodic measurements of the hematocrit level in infants with AOP are necessary after hospital discharge. Once a steady increase in the hematocrit level has been established, only routine checks are required.


Blood Transfusion Considerations and Concerns

Packed red blood cell (PRBC) transfusions

Packed red blood cell (PRBC) transfusions are the mainstay of therapy for anemia of prematurity (AOP). The frequency of blood transfusion varies with gestational age, degree of illness, and, interestingly, the hospital evaluated. Unfortunately, there is considerable disagreement about the indication, timing, and efficacy of PRBC transfusion.

There is agreement, however, that the decision to give a transfusion should not be made lightly, because significant infectious, hematologic, immunologic, and metabolic complications are possible. Late-onset necrotizing enterocolitis has been reported in stable-growing premature infants electively transfused for AOP. [3, 4] Transfusions also transiently decrease erythropoiesis and EPO levels. There is also agreement that the number of transfusions, as well as the number of donor exposures, should be reduced as much as possible.

Reducing the number of transfusions

Studies from individual centers have documented a marked decrease in the administration of PRBC transfusions in the past decades, even before the use of EPO became more frequent. This decrease in transfusions is almost certainly multifactorial in origin. Adoption of standardized transfusion protocols that take various factors into account, including hemoglobin levels, degree of cardiorespiratory disease, and traditional signs and symptoms of pathologic anemia, are acknowledged as an important factor in this reduction. A restricted transfusion protocol may decrease the number of transfusions while also decreasing the hematocrit at discharge. [5]

A 2011 study evaluated 41 preterm infants with birth weights of 500-1300 g who were enrolled in a clinical trial that compared high and low hematocrit thresholds for transfusion. A rise in systemic oxygen transport and a decrease in lactic acid after transfusion was noted in both groups; however, oxygen consumption did not change significantly in either group. In the low hematocrit group only, cardiac output and fractional oxygen extraction fell after transfusion, which shows that these infants had increased their cardiac output to maintain adequate tissue oxygen delivery in response to anemia. The results demonstrate that infants with low hematocrit thresholds may benefit from transfusion, while transfusion in those with high hematocrit thresholds may provide no acute physiological benefit. [6]

The Premature Infant in Need of Transfusion (PINT) study showed that transfusing infants to maintain higher hemoglobin level (8.5-13.5 g/dL) conferred no benefit in terms of mortality, severe morbidity, or apnea intervention compared with infants transfused to maintain a low hemoglobin levels (7.5-11.5 g/dL). [7]

These findings differ from the Iowa study, which found less parenchymal brain hemorrhage, periventricular leukomalacia, and apnea in infants whose transfusion criteria was not restricted and whose hemoglobin level was higher. Clearly, no universally accepted guidelines for transfusion in infants with AOP are available at this time, and clinicians must determine their individual standardized transfusion practices.

As an example, note the Children's Hospital of Wisconsin Transfusion Committee guidelines for consideration:

  • An infant with a hemoglobin (Hb) level of less than 8 g/dL may be transfused at the discretion of the attending physician
  • A stable infant with an Hb level of 8-10 g/dL without clinical symptoms or other exceptions listed below may be transfused
  • An infant with an Hb level of 11-13 g/dL without a supplemental oxygen or continuous positive airway pressure (CPAP) requirement, apnea/bradycardia, significant tachycardia or tachypnea, or other exceptions listed below should not be transfused
  • An infant with an Hb level of more than 13 g/dL without an oxygen requirement of more than 40% by hood, CPAP, or ventilator; hypotension that requires pressor medication; major surgery; or other exceptions listed below should not be transfused
  • An infant with an Hb level of more than 15 g/dL without cyanotic heart disease, extracorporeal membrane oxygenation (ECMO) therapy, regional oxygen saturations less than 50%, or hypotension that requires pressor medications should not be transfused
  • An infant with a history of massive blood loss may be transfused at the discretion of the attending physician

It is of obvious importance to discuss with the family their child’s need for transfusion and to obtain consent before the transfusion. It is also important to discuss with parents the normal course of anemia, the criteria for and risks associated with transfusions, and the advantages and disadvantages of erythropoietin (EPO) administration. Also necessary is consideration of the family's religious beliefs regarding transfusions.

Reducing the number of donor exposures

Reducing the number of donor exposures is also important. Preservatives and additive systems allow blood to be stored safely for as long as 35-42 days. This can be accomplished by using PRBCs stored in preservatives (eg, citrate-phosphate-dextrose-adenine [CPDA-1]) and additive systems (eg, Adsol). Infants may be assigned a specific unit of blood, which may suffice for treatment during their entire hospitalization and thus limit exposure to the single donor of that unit. Concerns that stored blood might increase serum potassium levels are unfounded if the transfused volume is low.


Potential complications of transfusion include the following:

  • Infection (eg, hepatitis, cytomegalovirus [CMV], human immunodeficiency virus [HIV], syphilis)
  • Fluid overload and electrolyte imbalances
  • Exposure to plasticizers
  • Hemolysis
  • Posttransfusion graft versus host disease

An important tool in reducing at least one of these transfusion risks is to use all available screening techniques for infection. The risk of cytomegalovirus (CMV) transmission can be dramatically reduced by use of CMV-safe blood. This can be accomplished by using CMV serology–negative cells, along with blood processed through leukocyte-reduction filters or inverted spin technique. These methods also reduce other WBC-associated infectious agents (eg, Epstein-Barr virus, retroviruses, Yersinia enterocolitica) by yielding a leukocyte-poor suspension of PRBCs. The American Red Cross now provides exclusively leukocyte-reduced blood to hospitals in the United States.


Clinical trials designed to determine the efficacy of blood transfusions in relieving symptoms ascribed to anemia of prematurity (AOP) have produced conflicting results. [8] Improved growth has been an inconsistent finding. While some studies have demonstrated a decrease in apneic episodes after blood transfusion, others have found similar results with simple crystalloid volume expansion.

Subjective improvement in activity, decreased lethargy, and improved feeding have been described in some studies. Blood transfusions have been documented to decrease lactic acid levels in otherwise healthy preterm infants who are anemic. Blood transfusions have reduced tachycardia in anemic infants who are transfused.

Some medical professionals have suggested using lactate levels as an aid in determining the need for transfusion.


Recombinant Erythropoietin Treatment

Multiple investigations have established that premature infants respond to exogenously administered recombinant human EPO and supplemental iron with a brisk reticulocytosis. Subcutaneous administration of EPO may be preferred, as intravenous administration has increased urinary losses. Although EPO cannot prevent early transfusions, modest decreases in the frequency of late PRBC transfusions have been documented. Additional iron supplementation is necessary during exogenous EPO treatment.

Trials have evaluated the impact of EPO treatment in populations of the most immature neonates. These studies likewise have demonstrated that infants with very low birth weight (VLBW) are capable of responding to EPO with a reticulocytosis.

Studies and a Cochrane Neonatal Systemic review suggest an association between exogenous EPO administration and retinopathy of prematurity. [9, 10, 11]

Yasmeen et al studied 60 preterm low birth weight infants and concluded that short-term recombinant human erythropoietin with iron and folic acid was effective in preventing anemia of prematurity. [12]

EPO with iron does not adversely affect growth or developmental outcomes, but the impact on the number of transfusions a premature infant receives ranges from nonexistent to small.

At this time, no agreement regarding the safety, timing, dosing, route, or duration of therapy has been established. In short, the cost-benefit ratio for EPO has yet to be clearly established, and this medication is not universally accepted as a standard therapy for an infant with AOP.