OVERVIEW: What every practitioner needs to know

Are you sure your patient has anemia of prematurity? What are the typical findings for this disease?

Anemia in a preterm infant less than 32 weeks of gestation

Inappropriately low reticulocyte count for severity of anemia

Inappropriately low circulating erythropoietin concentration for the degree of anemia

Continue Reading

Typical physical examination findings


Hemodynamic perturbations (mainly tachycardia)


Poor growth

Respiratory irregularities (e.g., apnea)

Typical laboratory findings

Hematocrit levels:

The normal hematocrit for neonates varies by gestational age, with preterm infants typically having lower hematocrits at birth than do term infants. The mean hematocrit of a 32-week-gestation infant is 50%, whereas infants born at less than 28 weeks’ gestation have a mean hematocrit of 40%. The corresponding hemoglobin levels are on average 3.3 g/dL lower in preterm infants when compared with term infants. In the 3 months after birth, the hematocrit and hemoglobin of all newborns decreases.

In healthy term infants, this decrease is physiologic and they remain asymptomatic. The nadir hematocrit in term infants occurs between 10 and 12 weeks of age and rarely falls to less than 30%, with hemoglobin concentrations of 10-12 g/dL. After 10-12 weeks, the hematocrit and hemoglobin increase slowly to reach adult values by 2 years of age.

In preterm infants, the drop in hematocrit and hemoglobin is more rapid and more profound. The nadir hematocrit occurs between 4 and 6 weeks of age, with hematocrit values of 21% and 28% seen commonly in infants with birth weights less than 1.0 kg and less than 1.5 kg, respectively. This level of anemia is due to a combination of physiologic factors (decrease in erythropoietin [Epo] production) and nonphysiologic factors such as iatrogenic blood loss for laboratory testing and iron deficiency.

Reticulocyte counts:

The reticulocyte count is a means of assessing the erythropoietic activity of the patient’s bone marrow and is typically expressed as a percentage of total red cells. At birth, preterm infants typically have higher absolute reticulocyte counts than term infants (400,000-555,000 versus 200,000-400,000). When the hematocrit is low, the reticulocyte percentage does not accurately reflect reticulocyte production relevant to the degree of anemia, so a correction factor is used. The corrected reticulocyte count, which reflects the reticulocyte response relative to the hematocrit, is the most informative and is calculated by the following formula:

Corrected reticulocyte count = patient’s reticulocyte (%) x (patient’s hematocrit/normal hematocrit (45 is typically used)

What other disease/condition shares some of these symptoms?

Anemia secondary to blood loss

Prenatal: placental abruption, placenta previa, ruptured placental vessels related to vasa previa or velamentous cord insertion, umbilical cord rupture, fetomaternal and fetoplacental bleeding, twin-to-twin transfusion

Perinatal: cephalohematoma, subgaleal hemorrhage, subcapsular hematoma of the liver, adrenal hemorrhage

Neonatal: intracranial hemorrhage, necrotizing enterocolitis, iatrogenic phlebotomy

Anemia secondary to hemolytic causes

Immune mediated: ABO, Rh, and minor blood group incompatibilities, maternal autoimmune hemolytic disease such as lupus

Acquired sepsis (bacterial, viral, fungal), disseminated intravascular coagulation, vitamin E deficiency, iron deficiency

Hereditary red blood cell disorders

Metabolic disorder: glucose-6-phosphate dehydrogenase deficiency

Red blood cell membrane disorder: spherocytosis, eliptocytosis

Hemoglobinopathies: α- and β-thalassemias

Anemia secondary to decreased red blood cell production

Infectious: rubella, parvovirus, malaria

Drug-induced: chloramphenicol

Genetic: Diamond-Blackfan anemia

What caused this disease to develop at this time?

Transitional physiology:

Physiologic changes occur as the fetus transitions from the placenta-dependent, relatively hypoxic, intrauterine environment to the lung-dependent, oxygen-rich environment.

Erythropoietin (Epo), an endogenous glycoprotein, is the primary regulator of erythrocyte production. After birth, Epo production is decreased due to the extrauterine oxygen-rich environment and the transition from liver Epo production to renal production. During this transition, which takes 3 to 4 months past term birth, the body is less sensitive to tissue hypoxia as a stimulus for Epo production. This decrease in Epo production translates into a 20% decrease in erythroid progenitor cells in the marrow.

Epo clearance and volume of distribution is also high in neonates relative to adults, and this likely contributes to low circulating concentrations. The fact that erythrocyte progenitors are quite sensitive to Epo, and endogenous Epo production is low, forms part of the rationale for Epo treatment of anemia of prematurity.

Postnatal physiology:

Endogenous: Preterm and term infants have decreased red blood cell survival compared with adults (70 days compared with 120 days in adults). This decreased survival is thought to contribute to anemia. Red blood cell volume must increase as the baby grows. An estimated 5 x 109red blood cells per day must be produced to maintain a stable hematocrit in relation to growth. This does not include replacement of phlebotomy losses.

Exogenous: Phlebotomy losses for laboratory studies in the extremely preterm infant can range up to several blood volumes, depending on the severity of illness and ease of vascular access. The majority of these losses occur in the first 2 weeks of life, the period in which most blood transfusions occur. Iron deficiency occurs from phlebotomy loss and decreased postnatal intake. The bulk of iron transfer to the fetus occurs in the third trimester. Although most preterm infants are not born iron deficient, they may become deficient if they have inadequate intake over time (2-4 mg/kg/day). Restrictive transfusion guidelines may contribute to iron deficiency in the preterm infant.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

A complete blood cell count will show low hemoglobin and hematocrit values. The reticulocyte count will be low. To maintain a stable hematocrit in a growing preterm infant with minimal phlebotomy losses, a corrected reticulocyte count should be 3% or more.

A blood smear should show normocytic, normochromic red blood cells with no evidence of hemolysis. This is less helpful after a baby has been transfused.

Iron balance should be assessed. The two most commonly used methods include serum ferritin and zinc protoporphyrin to heme ratios (ZnPP/H). Both values change with increasing gestational age, reflecting the increase in iron transfer that occurs over the third trimester.

If a baby is born prematurely, normal iron transfer in the third trimester is interrupted. Therefore, if iron replacement is not given, the baby becomes progressively more iron deficient. The 5th percentile for ferritin levels in term and preterm infants are 40 and 35 g/mL respectively. Ferritin levels of 75 or less have been associated with abnormal neurologic reflexes or ABRs. Supplemental iron should be held if ferritin values are greater than 500 ng/mL.

In the presence of iron deficiency, zinc is incorporated into the protoporphyrin ring instead of Fe iron. If the ZnPP/H ratio rises, iron deficiency is present. For infants of 26 weeks’ gestational age + postnatal age (PMA) or less, ZnPP/H should be less than 155. For infants 27-29 weeks PMA, ZnPP/H should be less than 120; for infants 30 weeks PMA, ZnPP/H should be <95. (Normal adult values range from 30-80).

The Coombs test should be negative.

Would imaging studies be helpful? If so, which ones?

There are no imaging studies that are diagnostic of anemia of prematurity, but imaging can be used to exclude other causes of anemia. If bleeding is suspected, cranial ultrasonography should be performed to look for intracranial hemorrhage. Abdominal ultrasonography may detect intraabdominal blood such as subcapsular hematoma of liver or adrenal hemorrhage.

If hemolysis is suspected in association with a critically ill infant, abdominal radiography should be performed to look for evidence of NEC.

Confirming the diagnosis

Anemia in the preterm infant can be categorized by whether there is blood loss, increased consumption (hemolysis), or decreased production. Anemia of prematurity is a problem of decreased/inadequate production. Once the diagnosis of anemia of prematurity has been made, treatment options include recombinant human Epo treatment to stimulate erythrocyte production or red blood cell transfusions to replace loss.

Epo treatments have been studied in randomized controlled studies for more than 25 years with more than 3000 infants enrolled. Epo doses, dosing intervals, and duration of therapy in these studies have varied widely. Early studies used doses per kilogram that were more comparable to those used in adults. These showed little benefit because newborns have a volume of Epo distribution and clearance rate that is two to four times that measured in adults. Doses of 200 U/kg/day (IV) or 400 U/kg subcutaneously three times a week used with 6-8 mg/kg oral supplemental iron per day or 1 mg/kg iron sucrose IV are generally effective in increasing erythropoiesis.

Epo use in preterm infants has been studied for periods limited to 2 weeks or for more prolonged durations such as from 2 days of age to 35 weeks postmenstrual age. It has been quite safe, with none of the complications seen in adults. There was early concern that early Epo treatment might increase the risk of retinopathy of prematurity, but this has not been borne out.

Results show that Epo clearly and definitively increases erythropoiesis and decreases the volume and number of transfusions to which an infant is exposed. The hematocrit of Epo-treated infants also tends to be about 5 points higher than that in transfused infants. Despite this, the value of this intervention is questioned because depending on the amount of phlebotomy an infant experiences and whether aliquoted blood is used, Epo may not eliminate all transfusions and may not decrease the number of donor exposures.

Two large prospective, randomized controlled trials of restrictive versus liberal transfusion guidelines for preterm infants have been published. No differences in clinical outcomes were evident based on transfusion practices.

The single center study of 100 infants published by Bell et al suggested an increase in intraventricular hemorrhage (IVH) plus periventricular leukomalacia (PVL) with restrictive transfusion practices in a post hoc assessment, but head ultrasonography was not obtained before enrollment, so it is unclear whether this finding was the result of transfusion practices. A follow-up study of this population showed larger brain volumes in the subset of children who were randomized to the restrictive arm of the study.

The PINT study (Kirpalani et al), a multicenter trial of 451 infants, showed no differences in IVH or PVL. No differences in other measures of illness severity were documented by either study, including bronchopulmonary dysplasia (BPD) and length of stay.

No trials comparing transfusions versus no transfusions have been carried out in neonates. In adults, such studies have demonstrated increased risk of multisystem organ failure, infection, immune suppression, and death in the transfused arms. Retrospective studies of transfusion practices in neonates have suggested an increase in BPD, diuretic use, NEC, and death in infants who were transfused more liberally.

If you are able to confirm that the patient has anemia of prematurity, what treatment should be initiated?

Anemia of prematurity occurs gradually and should not require emergent treatment. This is in contrast to anemia from acute hemorrhage, intensive phlebotomy loss, or hemolysis in a critically ill infant, for which emergent transfusions might be indicated. Management should be as follows:

Preventive treatment: practice delayed cord clamping or cord stripping.

Use cord blood for initial laboratory test.

Limit iatrogenic blood loss.

Institute restrictive transfusion practices (see below).

Epo (400 U/kg/dose subcutaneously 3 times a week or 200 U/kg/dose daily intravenously) may be used to stimulate erythropoiesis; darbepoietin alfa 10 µg/kg once a week subcutaneously is an alternative erythropoietic agent that may be used.

Monitor and maintain adequate iron balance. If recombinant Epo or Darbepoetin alfa are used, therapeutic oral iron supplements should be administered (6-8 mg/kg/day). If the patient cannot take oral supplements, 1 mg/kg/day iron dextran or iron sucrose can be given intravenously. Either ZnPP/H or serum ferritin can be used to assess iron balance. This should be done every 2-4 weeks.

Indications for the transfusion of preterm neonates vary by developmental stage and severity of illness. An example of restrictive blood transfusion guidelines that take both factors into consideration are shown below (recommended by authors).

Transfuse 15-20 mL/kg PRBC over 3-4 hours (volume needed will depend on the hematocrit of transfused blood) if hematocrit is less than 35% in the first week of life and infant is unstable (instability is defined as an increased risk for poor oxygen delivery, e.g., prolonged oxygen desaturation episodes or hypotension requiring treatment [pressors, hydrocortisone, boluses of isotonic fluid]); if the hematocrit is less than 28% in the first week of life or the infant is unstable; if the hematocrit is less than 20% and if the infant is older than 1 week and stable.

If restrictive transfusion guidelines are instituted, babies will receive less iron in the form of transfused red blood cells, and may require additional supplementation to maintain iron sufficiency.

More liberal transfusion guidelines are described by Strauss:

Transfuse PRBC over 3-4 hours to maintain hematocrit (words in italics must be defined locally)

Greater than 40% for severe cardiopulmonary disease

Greater than 30% for moderate cardiopulmonary disease (e.g., nasal continuous pulmonary artery pressure [CPAP] or supplemental oxygen)

Greater than 30% for major surgery

Greater than 20%-25% for infants with stable anemia especially if unexplained breathing disorders, tachycardia, or poor growth

Infants should receive irradiated, cytomegalovirus-negative or leukocyte-depleted hemoglobin S-negative, typed and screened packed red blood cells.

Practitioners should be aware of the preservative used for the blood they are transfusing: blood stored in citrate-phosphate-adenine (CPDA) preservative has a hematocrit of approximately 70% (35-day storage permitted), whereas blood preserved in AS-1, AS-3, or AS-5 solutions (42-day storage permitted) has a hematocrit of 55%-60%.

Longer term treatments include daily maintenance iron, vitamin E, vitamin B12, and folate

What are the adverse effects associated with each treatment option?

Adverse effects of blood transfusion include infection from contaminated blood products, fluid overload, electrolyte and calcium disturbances, immune-mediated adverse reactions (e.g., acute hemolytic reaction, febrile nonhemolytic transfusion reaction, graft-versus-host disease, transfusion-related acute lung injury, immune suppression), allergic reactions, and transfusion of other toxic substances contained in blood such as lead, mercury, and plasticizers.

Recently, a concern of transfusion-related NEC has been raised, but this association has not been proven. Iron overload can occur if multiple transfusions are given with blood volumes that are significantly greater than blood loss from phlebotomy. This might occur in the situation of hemolysis or disseminated intravascular coagulation. It is rare for transfusions in preterm infants. One milliliter of CPD-stored blood contains approximately 0.5 mg iron, so a 20 mL/kg transfusion contains 10 mg/kg iron, which is released over a 30-day half-life.

Adverse effects of recombinant human Epo and iron therapy appear to be minimal in preterm infants. Many adverse side effects have been documented in adults, including major venous thrombosis, stroke, polycythemia, hypertension, seizures, immune-mediated anemia, and unexpected death. None of these has been seen in the more than 3000 neonates studied in randomized controlled studies.

One possible adverse effect of early Epo administration (<8 days) unique to preterm infants may be an increased risk of retinopathy of prematurity (ROP), although in the 2014 Cochrane review, this effect was not significant. Additional safety information from a Swiss study of high dose early Epo also showed no increase in adverse side effects, including ROP.

Adverse effects of iron supplementation include feeding intolerance and iron overload if iron balance is not monitored and multiple transfusions occur concurrently. Iron overload may increase the risk of oxidant-mediated tissue injury.

Adverse effects of vitamin E supplementation are not common. Large doses have been associated with an increased incidence of NEC, thought to be due to the hyperosmolarity of the preparation. An increased incidence of sepsis has also been reported and is thought to be secondary to a pharmacologic serum vitamin E–related decrease in oxygen-dependent intracellular killing, resulting in increased susceptibility to infection in preterm infants.

What will you tell the family about prognosis?

Anemia of prematurity is a transient, physiologic process that is normal for preterm infants. Anemia is worsened by problems of prematurity that require blood draws for laboratory monitoring. Because of this, preterm infants are one of the most highly transfused patient populations. Fortunately, the blood supply in the United States and Europe is very safe. Epo is a reasonable treatment alternative that may help avoid transfusions.

As preterm neonates mature, anemia of prematurity resolves. A follow-up hematocrit may be required after discharge, but with adequate iron supplementation, anemia should not persist.

What will you tell the family about risks/benefits of the available treatment options?

The risks of a red blood cell transfusion to treat neonatal anemia are low because of advances in blood banking practices. The risks of blood transfusions that should be discussed include the potential for immune-mediated adverse reactions (e.g., acute hemolytic reaction, febrile nonhemolytic transfusion reaction), allergic reactions, and infectious complications.

Treatment with erythropoietic stimulating agents (Epo or darbepoetin), which have fewer potential adverse side effects than blood transfusion, will decrease the volume and number of transfusions, but may not eliminate blood exposure completely. The down side of Epo treatments are the subcutaneous shots are needed three times a week. Darbepoetin alfa, which is dosed once a week, is an excellent alternative.

What causes this disease and how frequent is it?

Prematurity underlies the development of anemia of prematurity for the reasons described above. Since birth does not accelerate the transition to renal Epo production, the more extreme the prematurity, the longer the lag in Epo production and the more severe the anemia of prematurity is likely to be. Concurrent illnesses worsen the anemia.

There is new evidence that the degree to which infants respond to Epo may be genetically mediated. Thus there may be a group of infants who respond more robustly to Epo than others. Research in this area is ongoing.

What complications might you expect from the disease or treatment of the disease?

Complications from anemia of prematurity:

Poor growth, apnea, cardiovascular instability if severe

Complications from treatment of anemia of prematurity:

Red blood cell transfusions:

Acute complications:

Possible adverse reactions to a blood transfusion, which include infection from contaminated blood products, fluid overload, electrolyte and calcium disturbances, immune-mediated adverse reactions (e.g., acute hemolytic reaction, febrile nonhemolytic transfusion reaction, graft-versus-host disease, transfusion-related acute lung injury, immune suppression), allergic reactions, and transfusion of other toxic substances contained in blood such as lead, mercury, and plasticizers.

Long-term complications:

Red blood cell transfusions have been associated with increased risk of bronchopulmonary dysplasia, NEC, and diuretic use. Bronchopulmonary dysplasia and NEC are both associated with increased risk of poor neurodevelopmental outcome (mental retardation, cerebral palsy, deafness, blindness). Diuretic use is also associated with adverse effects such as deafness, calcium wasting and osteopenia.

Red blood cell transfusions may adversely affect the long-term outcome of premature infants as indicated by reduced brain volumes on magnetic resonance imaging studies at 12 years of age for neonates who received transfusions using liberal transfusion guidelines.

Treatment with erythropoietin stimulating agents (ESAs):

Acute complications:

In adults, polycythemia, rash, seizures, hypertension, and stroke are known complications. None of these adverse effects have been reported in Epo-treated neonates.

Long-term complications:

In adults, shortened time to death, myocardial infarction, congestive heart failure, and progression of tumors have been identified with long-term therapy. None of these adverse effects has been reported in Epo-treated neonates. In addition, no prospective studies of Epo treatment of neonates have reported group differences in the incidence of neonatal morbidities, including intraventricular hemorrhage, ROP, NEC, chronic lung disease, or late-onset sepsis.

Although not proven in human trials, there are animal and preliminary human data to suggest that ESA treatment is neuroprotective at high doses. Studies are ongoing to test this in humans.

How can anemia of prematurity be prevented?

Measures that decrease phlebotomy losses are important. These may include cord blood sampling for immediate postnatal laboratory tests (e.g., type and cross-match), microsampling, batching and judicious use of laboratory tests, use of point-of-care laboratory testing devices, and prompt removal of central arterial and venous catheters providing easy access to blood.

Parenteral iron has a small effect on enhancing erythropoiesis in preterm infants treated with recombinant human Epo, but there is no evidence that iron treatment is beneficial in the absence of iron deficiency.

Inadequate protein intake may contribute to anemia of prematurity. The normal postnatal decrease in hemoglobin can be improved by 1-1.5 g/dL in preterm very low–birth-weight infants given daily protein intakes of 3.5-3.6 g/kg compared with infants receiving intakes of 1.8 to 1.9 g/kg daily. Once endogenous Epo production is adequate, iron, folate, vitamin E, and vitamin B12 must be sufficient to support erythropoiesis

What is the evidence?

Early Epo reduces the risk of infants needing one or more RBC transfusions [typical relative risk (RR); 0.80 (95% confidence interval [CI] 0.75-0.86); 16 studies, 1825 infants].

Level of evidence: randomized or quasi-randomized controlled trials of early (<8 days of age) initiation of EPO treatment versus placebo or no intervention in preterm and/or low-birth-weight neonates. Reference: Ohlsson A, Aher SM. 2014. Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants. Cochrane Database Syst Rev 2014;4:CD004863.

Vitamin E supplementation increases hemoglobin concentration significantly by a small amount (weighted mean difference, 0.46; CI, 0.24-0.69).

Level of evidence: randomized clinical trials. Reference: Brion LP, Bell EF, Raghuveer TS. Vitamin E supplementation for prevention of morbidity and mortality in preterm infants. Cochrane Database Syst Rev 2003; 4:CD003665.

Neonatal blood transfusion guidelines for anemia in preterm infants are based on expert opinion.

Evidence-based optimal pre-transfusion blood hemoglobin/hematocrit levels remain undefined in preterm infants.

The benefit of red blood cell transfusions on improving respiratory irregularities is not well supported by evidence.

The effect of red blood cell transfusions on improving weight gain in stable preterm infants is not well supported by evidence.

There is insufficient evidence to determine whether or not the hemodynamic response to anemia is of clinical significance.

Recommended references on anemia of prematurity

Aher, SM, Ohlsson, A. “Late erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants”. Cochrane Database Syst Rev. vol. 4. 2014. pp. CD004868

Bell, EF, Strauss, RG, Widness, JA. “Randomized trial of liberal versus restrictive guidelines for red blood cell transfusion in preterm infants”. Pediatrics. vol. 115. 2005. pp. 1685-91.

Brion, LP, Bell, EF, Raghuveer, TS. “Vitamin E supplementation for prevention of morbidity and mortality in preterm infants”. Cochrane Database Syst Rev. vol. 4. 2003. pp. CD003665

Fauchere, JC, Koller, BM, Tschopp, A. “Safety of early high-dose recombinant erythropoietin for neuroprotection in very preterm infants”. J Pediatr. vol. 167. 2015. pp. 52-7.

Garcia, MG, Hutson, AD, Christensen, RD. “Effect of recombinant erythropoietin on "late" transfusions in the neonatal intensive care unit: a meta-analysis”. J Perinatol. vol. 22. 2002. pp. 108-11.

Haiden, N, Schwindt, J, Cardona, F. “Effects of a combined therapy of erythropoietin, iron, folate, and vitamin B12 on the transfusion requirements of extremely low birth weight infants”. Pediatrics. vol. 118. 2006. pp. 2004-13.

Kirpalani, H, Whyte, RK, Andersen, C. “The Premature Infants in Need of Transfusion (PINT) study: a randomized, controlled trail of restrictive (low) versus liberal (high) transfusion threshold for extremely low birth weigh infants”. J Pediatr. vol. 149. 2006. pp. 301-7.

Neelakantan, S, Widness, JA, Schmidt, RL. “Erythropoietin pharmacokinetic/pharmacodynamic analysis suggests higher doses in treating neonatal anemia”. Pediatr Int. vol. 51. 2009. pp. 25-32.

Nopoulos, PC, Conrad, AL, Bell, EF. “2011 Long-term outcome of brain structure in premature infants: effects of liberal vs restricted red blood cell transfusions”. Arch Pediatr Adolesc Med. vol. 165. 2011. pp. 443-50.

Ohls, RK, Christensen, RD, Kamath-Rayne, BD, Rosenberg, A, Wiedmeier, SE. “A randomized, masked, placebo-controlled study of darbepoetin alfa in preterm infants”. Pediatrics. vol. 132. 2013. pp. e119-27.

Ohls, RK, Ehrenkranz, RA, Wright, LL. “Effects of early erythropoietin therapy on the transfusion requirements of preterm infants below 1250 grams birth weight: a multicenter, randomized, controlled trial”. Pediatrics. vol. 108. 2001. pp. 934-42.

Ohls, RK, Kamath-Rayne, BD, Christensen, RD. “Cognitive outcomes of preterm infants randomized to darbepoetin, erythropoietin, or placebo”. Pediatrics. vol. 133. 2014. pp. 1023-30.

Ohlsson, A, Aher, SM. “Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants”. Cochrane Database Syst Rev. vol. 4. 2014. pp. CD004863

Valieva, OA, Strandjord, TP, Mayock, DE. “Effects of transfusions in extremely low birth weight infants: a retrospective study”. J Pediatr. vol. 155. 2009. pp. 331-37.

Widness, JA, Madan, A, Grindeanu, LA. “Reduction in red blood cell transfusions among preterm infants: results of a randomized trial with an in-line blood gas and chemistry monitor”. Pediatrics. vol. 15. 2005. pp. 1299-306.

Ongoing controversies regarding etiology, diagnosis, treatment

Optimal blood hemoglobin and hematocrit levels in preterm infants remain undefined.

The risk to benefit ratio of blood transfusions for preterm infants requires definition. Using current guidelines (either restrictive or liberal) for nonemergent transfusions, there is no clear evidence of benefit in terms of growth, apnea, short or long-term respiratory parameters, or length of hospital stay. There are also inadequate data on the long-term effects of transfusions on neurodevelopment.

The optimal Epo preparation, dose, dosing interval, and duration of treatment have not been defined.

Delayed umbilical cord clamping at the time of delivery may help to prevent the anemia associated with prematurity. Both short, and long-term outcome assessments are needed.

Transfusion of autologous blood harvested at delivery needs to be studied.