Hematology: NICU Handbook

John A. Widness, MD
Peer Review Status: Internally Peer Reviewed

Diagnosis of anemia:

Well established, scientifically founded criteria for the diagnosis of anemia in the neonate are not available at present. This void is in part due to the difficulty in studying patients who are unable to communicate how they feel ("symptoms") and the fact that objective clinical "signs" of neonatal anemia are non-specific and thus frequently are indicators of problems other than anemia, e.g., sepsis, apnea, seizures, growth failure ("failure to thrive"), etc. Defining when clinically significant anemia is present is an area of active research.

See UI NICU Guidelines for Administering 15mL/kg Erythrocyte Transfusions to Neonates for our NICU's transfusion guidelines for preterm infants. Infants should not be treated to replace phlebotomy losses alone. Instead, as shown in the Table, a combination of the patient's clinical condition (primarily his/her respiratory status) and the presence of peripherial hematocrit values below levels specified for the various degrees of illness are used.

Treatment and prevention of anemia:

Mimimize phlebotomy losses:

Keep laboratory testing to only those tests which are needed. This is especially true in the first weeks of life when sick infants have the greatest amount of blood drawn due to their often tenuous condition.

Transfusion with packed red blood cells (PRBCs):

Good clinical practice dictates and regulatory agencies advise that chart documentation of the reason that the transfusion is being administered should be recorded. Transfuse to achieve a calculated hematocrit of approximately 45%, or give a maximum volume of 15 mL/kg. As a "rule of thumb," for each 1 mL of PRBC’s transfused (Hct of ≈ 85%)/kg, anticipate a 1% increase in the patient’s hematocrit. Hence for 15 mL of PRBC/kg, a pre-transfusion hct of 32% should rise to approximately 47% when checked several hours after transfusing.

Transfuse using irradiated (only infants with birth weights <1.5 kg) filtered to reduce CMV risk, packed red blood cells (Hct ≈ 85%). The blood bank routinely screens all blood for other viral pathogens including HIV, hepatitis B, hepatitis C, and HTLV I/II. Assuming a packed cell hematocrit of 80-90% and a blood volume of 80 mL/kg:

Erythropoietin (EPO):

Multicenter U.S. and European VLBW clinical trial results indicate that EPO therapy is of marginal clinical benefit as it is currently being administered. Thus, if EPO therapy is considered this needs to be discussed with the staff attending. If EPO is to be used, the dose is 200-300 U s.q./kg/d every other day. Enteral iron intake should be increased to 6 mg/kg/d during EPO treatment.

John A. Widness, MD
Peer Review Status: Internally Peer Reviewed

References:

Aher S, Malwatkar K, Kadam S. Neonatal anemia. Semin Fetal Neonatal Med 2008; 13:239-247.

Bell EF. When to transfuse preterm babies. Arch Dis Child Fetal Neonatal Ed 2008; 93:F469-F473.

Bell EF, Strauss RG, Widness JA, Mahoney LT, Mock DM, Seward VJ, Cress GA, Johnson KJ, Kromer IJ, Zimmerman MB. Randomized trial of liberal versus restrictive guidelines for red blood cell transfusion in preterm infants. Pediatrics 2005; 115:1685-1691.

Kirpalani H, Whyte RK, Andersen C, Asztalos EV, Heddle N, Blajchman MA, Peliowski A, Rios A, LaCorte M, Connelly R, Barrington K, Roberts RS. The Premature Infants in Need of Transfusion (PINT) Study: a randomized, controlled trial of a restrictive (low) versus liberal (high) transfusion threshold for extremely low birth weight infants. J Pediatr 2006; 149:301-307.

John A. Widness, MD
Peer Review Status: Internally Peer Reviewed

Historical perspective & overview

Hemolytic disease of the newborn has become a less and less common condition due largely to improved preventative measures such as the maternal administration of Rh immune globulin during the early 3rd trimester and the immediate postpartum period. With rare exceptions, it is presently possible to prenatally detect all non-ABO affected fetuses by testing for antibodies in maternal blood.

Most recently fetal cordocentesis has been utilized with increasing success to detect and to treat fetal anemia, i.e., with intravascular transfusion, in pregnancies identified prenatally, perinatal mortality and morbidity have been significantly improved. Infants followed by the High Risk Obstetrical service at the University of Iowa, are most commonly born close to term, have no to mild anemia, and are not jaundice in the first 24 hours. The most common neonatal problem today is that of anemia developing following discharge.

Diagnosis

ABO blood group incompatibility:

Since blood type is not routinely tested at birth, the diagnosis is almost always made after it is recognized that the infant is jaundice. It is uncommon for these infants to be significantly anemic and very rare for them to present with hydrops at birth. The diagnosis is made when the infant is A, B or AB and has a positive direct Coombs test and a positive indirect Coombs result for anti-A or anti-B. The mother will lack the A or B antigen which is positive in the indirect Coombs test.

Rh and other "minor" blood group incompatibilities:

Due to maternal screening for this condition, these infants are almost recognized prior to delivery. A positive direct Coombs test on the neonates blood with identification of a specific serum antibody known to be associated with hemolytic disease (some blood group antigens, e.g., Lewis are not) makes the diagnosis.

Management

ABO blood group incompatibility:

Although anemia should be looked for, hyperbilirubinemia is the primary morbidity associated with ABO blood group incompatibility. Management of this condition follows that described elsewhere in this manual (see section on "Management Of Hyperbilirubinemia in the Newborn Period"). The chance for this occurring again in future pregnancies is unpredictable.

Rh and other "minor" blood group incompatibilities

Prior to delivery

• obtain a careful history of past and present obstetrical history and a history of previous neonatal outcomes including
  • Outcome of previous pregnancies, i.e., fetal & neonatal deaths, prematurity, etc.
  • past & present history of in utero erythrocyte transfusion(s)
  • past & present hydrops, and
  • previous neonatal exchange transfusion for hyperbilirubinemia.
• In cases where a severely affected, anemic infant is anticipated (a rarity in recent years), packed type O Rh- blood cross-matched against maternal serum should be available for possible immediate booster transfusion (see Neonatal Blood Bank Procedure Manual).

At delivery

• Severely Affected Infants: Immediately following birth, the severely affected infant may have problems with circulatory and respiratory failure due to intrapartum depression and anemia, not bilirubin toxicity. Fortunately, this is a rare event with present obstetrical management. If present, ascites may create ventilatory embarrassment and paracentesis should be considered. Pulmonary problems similar in infants with neonatal depression and/or RDS may also occur. After initial stabilization, the infant should be transferred to the NICU. In the rare event that severe anemia is thought to be present and the infant's primary problem, a small exchange transfusion with packed red blood cells, 20-40 mL/kg given in the delivery room, may be indicated.
• Mild to Moderately Affected Infants: If the delivery room assessment of infants indicates that the infant is not severely affected but still has some concerning signs, these infants should be transferred to the NICU (or Intermediate Care Nursery if appropriate).
• Infants With No Signs Of Clinical Illness Or Jaundice: These infants may be sent to normal nursery if they meet this nursery's other criteria for admission.
• Cord Blood Laboratory Determinations: Before the umbilical cord blood clots, an immediate blood sample should be drawn with a large gauge needle and syringe from the placental portion of the umbilical cord and placed in an EDTA anticoagulated tube (lavender top) and red top tube. These samples should be sent to the for the hospital laboratory for blood group and direct Coomb's test.

In the nursery:

• Severely Affected Infants: Following transfer to the NICU most severely affected infants warrant having an umbilical or peripheral arterial catheter inserted for monitoring blood pressure, pH and blood gases. As noted above, if severe anemia is present, a small exchange transfusion with packed red cells, 20-40 mL/kg, may be indicated. An infusion of D10W with maintenance electrolytes should be initiated through an arterial line or peripheral IV. A full "two volume" exchange for hyperbilirubinemia should be delayed for several hours until the infant's initial condition has stabilized (see below). If blood is not needed to treat anemia, hypotension may be corrected with Plasmanate®.
• Mild to Moderately Affected Infants: Treatment of less severely affected infants starts with correction of hypotension and acidosis. If clinical condition and gestational age allow, oral feedings should be started in first four hours of life.
• All Affected Infants, i.e., those which are Coombs positive:
  • Laboratory Determinations:
    Bilirubin: The frequency of laboratory determinations will depend on the severity of the hemolytic disease, previous values and therapy. Data available on the cord blood sample will be helpful in anticipating these needs as well. In the first 12-24 hours, severely affected jaundiced and/or anemic infants should be started on phototherapy and have their serum total bilirubin levels be measured every 2-4 hours to establish a trend in its rate of rise. Less severely affected and apparently normal infants may be managed without phototherapy but should have serum bilirubin levels measured every four to six hours for the first 24 hours of life. Measurement of direct bilirubin should be one once, preferably during the first day. Infants found to have an elevated direct bilirubin in cord blood should liver enzyme determinations made and be repeated weekly.
    
    Hemoglobin and hematocrit values should be determined at 8-12 hours of age, before and after each exchange transfusion and daily until stable. Since the severely affected infant often has ß-cell hyperplasia, the infant should be monitored and treated in a similar manner to infants of diabetic mothers. In addition, blood glucose levels should be monitored 1 and 2 hours after each exchange transfusion in which CPDA-1 blood is used.
  • Phototherapy:
    Phototherapy should be initiated within the first 4 hours of life based on the cord bilirubin level and the subsequent rate of rise of the serum bilirubin concentration. This may avoid the need for an exchange transfusion. It is essential that the infant continue to have serum bilirubin levels monitored while under phototherapy.
  • Exchange Transfusions:
    The need for and timing of exchange transfusions should be done in consultation with the attending physician. Criteria for exchange transfusion do not change because of phototherapy. After exchange transfusion, serum bilirubin levels should be measured by the chemical method at 2-4 hours after the exchange, and then every 4-6 hours.
  • Intravenous Immune Globulin Therapy
    Although the mortality rate for exchange transfusion is probably lower than 1%, treatments as effective but less invasive and which have fewer risks would be appealing. One such treatment appears to be evolving.
    Rh antibodies do not fix compliment and do not induce intravascular hemolysis. The mechanism of destruction of antibody-sensitized red blood cells is probably antibody-dependent cellular cytotoxic effects mediated by cells of the RE system. Thus, erythrocyte destruction is similar to destruction of antibody-sensitized platelets in neonatal isoimmune thrombocytopenia. It has been shown in this latter disease that high dose intravenous immune globulin therapy can produce beneficial effects. Accordingly, it seemed plausible that similar therapy might alter the course of bilirubin production and reduce the rate of exchange transfusions in infants with Rh isoimmunization. The results of a recent study that tested this hypothesis concluded that, although the mechanisms was yet unknown, that indeed, high dose intravenous immune globulin therapy (500 mg/kg i.v. over 2-3h as soon as Rh incompatibility is established) did reduce serum bilirubin levels and the need for blood exchange transfusions in children with Rh hemolytic disease (J PEDIATR 1992;121:93-7). The optimum dose of intravenous immune globulin, the most efficacious number of infusions, and the best preparation remain to be determined. Undoubtedly, some of these questions are being answered in trials currently in progress. We would encourage ongoing dialogue of house staff with attendings on the neonatology services to ascertain the current status of this treatment.

At discharge: Post-hospital care plan

• Parents: Parents need to be aware that affected infants who may or may not have been anemic at birth (especially those who received one or more in utero erythrocyte transfusions) are at considerable risk for developing clinically significant anemia during the first 3-4 months of life. Their infants should have weekly hematocrit and reticulocyte counts performed and receive simple packed erythrocyte transfusions (20-25 mL/kg of PRBCs) if clinical symptoms appear if Hb levels fall below 6-7 gm/dL without evidence of a reticulocytosis, i.e., reticulocyte count <1%, or <100,000 per µL. Although infants can become sufficiently anemic to develop congestive heart failure, more often they manifest evidence of poor feeding or lack of activity. Life threatening clinical signs can occur in the presence of superimposed acute illnesses, i.e., viral infections.
• Local Physician: S/he should be contacted and given the same information as the parents along with an offer to provide the opportunity for future telephone consultation with an NICU staff neonatologist (Dr. Widness or Bell are particularly interested in following these infants).

References:

Millard DD, Gidding SS, Socol ML, et al. Effects of intravascular, intrauterine transfusions on prenatal and postnatal hemolysis and erythropoiesis in severe fetal isoimmunization. J Pediatr 1990;117:447-454.

Weiner CP, Williamson RA, Wenstrom KD, Sipes S, Grant SS, Widness JA. Management of fetal hemolytic disease by cordocentesis: I. Prediction of fetal anemia. Am J Obstet Gynecol 1991;165:546-553.

John A. Widness, MD
Peer Review Status: Internally Peer Reviewed

Diagnosis:

True whole blood viscosity cannot be routinely measured on blood samples. Viscosity increases with increasing central venous or arterial hematocrit, the diagnosis rests on the presence of polycythemia (hematocrit > 65%) and in the presence of clinical signs consistent with the diagnosis. This condition is almost always found in high risk infants during the first 24 hours of life.

High risk infants:

• SGA infants
• LGA infants
• IDMs
• Delayed cord clamping
• Transfusions, e.g., twin-twin, maternal -> fetal
• Trisomies, e.g., Down Syndrome, 13, & 18

Clinical signs of hyperviscosity:

• Lethargy  
• Hypotonia
• Weak suck
• Difficult to arouse
• Irritable when aroused
• Tremulousness
• Seizures
• Plethora
• Tachypnea or respiratory distress
• Abdominal distention

Screening:

High risk infants who are asymptomatic for hyperviscosity should have a screening capillary hematocrit obtained at 4-6 hours of age. This allows equilibration of postnatal hematocrit following placental transfusion at delivery. Infants with symptoms consistent with hyperviscosity should be tested immediately.

• Capillary hematocrit > 65% at 4-6 hours of life should be followed up immediately with a peripheral venous (or arterial) spun hematocrit.
• Capillary hematocrits > 60% drawn before 4 hours of life should be repeated with a second capillary hematocrit at 4-6 hours of life.

Associated laboratory findings:

• Abnormal chest X-ray: cardiomegaly, increased vascularity, hyperaeration, alveolar infiltrates, pleural effusions.
• Thrombocytopenia
• Hypoglycemia
• Hyperbilirubinemia (not apparent for at least a day or two)

Associated clinical conditions attributable to hyperviscosity:

• Increased pulmonary vascular resistance leading to pulmonary hypertension
• Increased systemic vascular resistance
• Increased myocardial strain
• Hypoxemia
• Pulmonary venous congestion
• Decreased regional blood flow
  • Gut (NEC)
  • Kidney
  • Brain
  • Myocardium (CHF) 
• Thromboses and gangrene
• Increased glucose utilization
• Local consumption of platelets

Treatment:

• Treatment of the asymptomatic infant with a hematocrit between 65-70% is controversial.
• Treatment of the symptomatic infant with partial exchange transfusion.

Estimated blood volume = 80-85 mL/kg; Hct desire = 50-55%

Estimated blood volume = 80-85 mL/kg; Hct desired = 50-55%;
Example: A 3.3 kg infant has a venous hematocrit of 72% and needs a partial exchange transfusion. You would like his post exchange Hct to be 50%:

Avoid push-pull technique through the UV catheter. To do so, remove blood through umbilical venous (or arterial if necessary) catheter while infusing an equal volume of normal saline at a similar rate through a peripheral vein. Many patients may require the same volume of normal saline to be given over an hour a 2nd time for tachycardia while their blood volume equilibrates.

• The push-pull method has been associated with an increased risk of NEC.

References:

Goldberg KE, Wirth FH, Hathaway WE, Guggenheim MA, Murphy JR, Braithwaite WR, and Lubchenco LO. Neonatal hyperviscosity II. effect of partial plasma exchange transfusion. Pediatrics 1982;69:419-425.

Gross GP, Hathaway WE and McGaughey HR. Hyperviscosity in the neonate. J Pediatr 1973;82:1004.

Hein HA and Lathrop SS. Partial exchange transfusion in term, polycythemic neonates: absence of association with severe gastrointestinal injury. Pediatrics 1987;80:75-78.

Oski FA, Naiman JL, Stockman III JA., and Pearson HA. Polycythemia and hyperviscosity in the neonatal period. In: M. Markowitz, ed. Hematologic Problems in the Newborn: Volume IV. Major Problems in Clinical Pediatrics (3rd ed.). Philadelphia: PA: Saunders, 1982:87-96.

John A. Widness, MD
Peer Review Status: Internally Peer Reviewed

Chromosome analysis

To obviate the need for doing an unnecessary bone marrow analysis, it is essential to check that a prenatal chromosome analysis via CVS, amniocentesis, or cordocentesis has not previously been done. Infants with multiple anomalies may have been previously evaluated by Ob-Gyn and thus tested.

Situations in which procedure may be considered:

Although a diagnosis can be made in a relatively short time (6-30 hours vs. 48-96 hours for peripheral blood samples), the use of bone marrow aspirations should be reserved for the diagnosis of conditions which are considered to be incompatible with life such as Trisomy 13, 18, or Triploidy (rarely). Accordingly, a patient with probable Down Syndrome would normally not be a candidate. In considering bone marrow aspiration, the technical of doing so must be considered. These include:

1. The failure rate for chromosome results for bone marrow sample is high (8-10%) relative to that of peripheral blood (<1%)
2. Chromosome morphology is generally poor, not allowing for identification of small, chromosomal rearrangements
3. Even in successful cases, the yield of analyzable cells is small increasing the risk of missing mosaicism.

While the results are pending, most situations are ones in which non-surgical life support therapies should be offered after discussion with the medical staff and the family.

Who to contact:

Weekdays before 3 pm: Prior to bone marrow aspiration, contact the Cytogenetics Laboratory staff must be called (6-3877) so that the necessary preparation for processing a specimen can be made. If possible, schedule the aspiration prior to 3 pm on weekdays since specimen processing takes approximately 2 hours.

To obtain a bone marrow specimen, the pediatric hematologist may be consulted (beeper 339-9023) and the Bone Marrow Laboratory notified (6-2543) if a smear of the marrow is desired. Since it is the nursery's responsibility to provide the necessary equipment, Central Sterilizing Service must be contacted to provide a "bone marrow aspirate set-up tray". Delivery takes appropriately one half hour.

Evenings or weekends: Prior to bone marrow aspiration contact the staff geneticist on call (131-1681), and the cytogenetics lab person on call (call the Cytogenetics Lab at 6-3877 for the recorded message of the current beeper number, or call the long-distance beeper: 1-800-202-8098). Do not perform the bone marrow aspiration until someone from the Cytogenetics Laboratory has been contacted since results cannot be obtained from the specimen if the sample is not processed within 1-2 hours.

Procedure:

Obtain 1-3 cc of bone marrow sample in a green-top heparinized tube. Also obtain a blood sample (2-3 cc) in a green-top tube for a chromosome analysis in case the marrow specimen fails to yield adequate results. If there is enough volume from a marrow sample, the specimen can be set up for both 2 and 24 hour processing. In small aspirate samples, only the 24 hour culture will be set up. Preliminary results from banded chromosomes will be available within six to thirty hours. The final report will be provided in 5-10 days.

Thrombocytopenia

Bone marrow aspiration for neonatal thrombocytopenia (platelet count < 50,000/mm3) is usually not necessary. However, in difficult or persistent cases, it may be helpful to discuss the evaluation and treatment of this condition with a member of the pediatric hematology division.

References

1. Christensen R D, Rothstein G, Anstall H B, Bybee B Granulocyte transfusions in neonates with bacterial infection, neutropenia, and depletion of mature marrow neutrophils. Pediatrics, 1982;70:1-6.
2. Cordle D G, Strauss R G (1993). Guidelines for Neonatal Transfusion Therapy - 1993. Iowa City:UIHC Elmer L. DeGowin Memorial Blood Center, University of Iowa Hospitals & Clinics, 1993. (reprinted in this book, pp. __ - ___).
3. Strauss R G. Current status of granulocyte transfusions to treat neonatal sepsis. J Clin Apheresis 1989;5:25-29.