Early Human Development (2008) 84, 493498 a v a i l a b l e a t w w w. s c i e n c e d i r e c t .

c o m

w w w. e l s e v i e r. c o m / l o c a t e / e a r l h u m d e v

BEST PRACTICE GUIDELINE ARTICLE

Management of anemia in the newborn

Naomi L.C. Luban
Departments of Pediatrics and Pathology, The George Washington University School of Medicine, United States Laboratory Medicine and Pathology, Transfusion Medicine/The Edward J. Miller Blood Donor Center, Children's National Medical Center, Washington, DC, United States

KEYWORDS
Neonate; Premature; Blood transfusion

Abstract Red blood cell (RBC) transfusions are administered to neonates and premature infants using poorly defined indications that may result in unintentional adverse consequences. Blood products are often manipulated to limit potential adverse events, and meet the unique needs of neonates with specific diagnoses. Selection of RBCs for small volume (520 mL/kg) transfusions and for massive transfusion, defined as extracorporeal bypass and exchange transfusions, are of particular concern to neonatologists. Mechanisms and therapeutic treatments to avoid transfusion are another area of significant investigation. RBCs collected in anticoagulantadditive solutions and administered in small aliquots to neonates over the shelf life of the product can decrease donor exposure and has supplanted the use of fresh RBCs where each transfusion resulted in a donor exposure. The safety of this practice has been documented and procedures established to aid transfusion services in ensuring that these products are available. Less well established are the indications for transfusion in this population; hemoglobin or hematocrit alone are insufficient indications unless clinical criteria (e.g. oxygen desaturation, apnea and bradycardia, poor weight gain) also augment the justification to transfuse. Comorbidities increase oxygen consumption demands in these infants and include bronchopulmonary dysplasia, rapid growth and cardiac dysfunction. Noninvasive methods or assays have been developed to measure tissue oxygenation; however, a true measure of peripheral oxygen offloading is needed to improve transfusion practice and determine the value of recombinant products that stimulate erythropoiesis. The development of such noninvasive methods is especially important since randomized, controlled clinical trials to support specific practices are often lacking, due at least in part, to the difficulty of performing such studies in tiny infants. 2008 Elsevier Ireland Ltd. All rights reserved.

Contents 1. 2. 3. Red blood cells for small volume transfusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494 Iatrogenic anemia of infancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 Criteria and guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495

Department of Laboratory Medicine, Children's National Medical Center, 111 Michigan Avenue, N.W., Washington, DC 20010, United States. Tel.: +1 202 884 5292; fax: +1 202 884 2007. E-mail address: nluban@cnmc.org. 0378-3782/$ - see front matter 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2008.06.007

494 Measurements to define the need to transfuse. In-line devices . . . . . . . . . . . . . . . . . . Autologous transfusion . . . . . . . . . . . . . 6.1. Cord clamping . . . . . . . . . . . . . . 7. Cord blood collection . . . . . . . . . . . . . . Key guidelines . . . . . . . . . . . . . . . . . . . . . Research directions . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . 4. 5. 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

N.L.C. Luban . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 496 496 496 497 497 497 497

1. Red blood cells for small volume transfusion

Most premature infants of 2427 weeks gestation require transfusions of red blood cells (RBCs) during their neonatal course [1]. Most of these transfusions will be administered in the first postnatal month [2]. Restrictive guidelines have been developed which have decreased donor exposure and transfusion number, but several factors continue to contribute to the need to transfuse. These include iatrogenic anemia, oxidative hemolysis secondary to sepsis, and rapid growth with concomitant protein and iron deficiency. The selection of the transfusion product continues to be controversial. Issues include 1) intraerythrocyte, 2,3DPG and subsequent oxygen offloading capacity which decrease during storage; 2) potassium content which increases during storage; 3) solute load from the anticoagulant solutions which might result in osmotic diuresis with subsequent alteration of cerebral microcirculation and result in intracranial (periventricular) hemorrhage; 4) transfusionassociated viral diseases and graft-versus-host-disease resulting from passive transfer of viral-infected white blood cells (monocytes) and engraftable lymphocytes, respectively. Several clinical articles and reviews have addressed the transfusion of RBCs packaged in small volumes in different anticoagulant-additive solutions (herein called a-RBCs) to reduce donor exposure [312]. The solutions have in common the use of mannitol, glucose, sodium chloride, phosphate and other additives (Table 1). To ensure that the hematocrit is more comparable to a standard packed RBCs, and to increase the red cell mass of the product, a-RBCs can be concentrated by either centrifugation [13] or inverted

Table 1 Formulation of anticoagulant-additive solutions in blood collection sets Constituent Volume (mL) Hematocrit (%) Sodium chloride (mg) Dextrose (mg) Adenine (mg) Mannitol (mg) Trisodium citrate (mg) Citric acid (mg) Sodium phosphate (monobasic) (mg) CPDA-1 63 7080 None 2000 17.3 None 1660 206 140 AS-1 100 5560 900 2200 27 750 None None None AS-3 100 5560 410 1100 30 None 588 42 276 AS-5 100 5560 877 900 30 525 None None None

gravity sedimentation [14]. We evaluated the potential toxicities arising from the use of three then common anticoagulant solutions in the settings of massive or large volume transfusion (i.e., exchange transfusion, cardiopulmonary bypass pump prime, extracorporeal membrane oxygenation) and small volume transfusions [15]. We estimated that concentrations of the solutes might reach dangerous or toxic levels in the setting of massive transfusion while the quantity of additives in small volume transfusions are unlikely to result in either acute or cumulative toxicity over time; solutes could be further reduced by hard packing a-RBCs to a hematocrit of 80%. Our theoretical calculations have been supported by more than nine infant studies (Table 2). While each study differs in the age of the product used, volume per kilogram (kg) transfused, type of anti-coagulant-additive solution, clinical and laboratory parameters measured, none demonstrated adverse metabolic consequences that might be expected from the solute loads. In a two arm randomized study of infants weighing 0.6 to 1.3 kg, 31 infants received CPDA-1 RBCs stored up to 7 days, while 30 received either AS-1 (19) or AS-3 RBCs (11) stored up to 42 days. Approximately half of the AS-1 and AS-3 units were greater than 15 days of age at transfusion. Changes in pH, glucose, lactate, calcium, sodium and potassium were minimal. In an expanded open safety study, 33 infants weighing 0.6 to 1.25 kg received 120 transfusions of AS-3 RBCs, of which 42 were older than 21 days at time of transfusion [4]. No clinical adverse consequences or significant differences in chemical analyses were observed. Mangel et al. [10] reported on their experiences with AS-3 units stored for up to 35 days in 56 infants who received 263 transfusions. Donor exposures were minimal (mean of 1.7) despite 4.7 transfusion episodes per infant. This study did not evaluate pre- and post-blood chemistries (other than hematocrit) and the indications for transfusion were not detailed. Of note are the volume of RBCs in mL/kg transfused (7 mL/kg), which is much lower than U.S. and UK practice and the lower hematocrit (5560%), which likely increased the donor exposure number when compared to U.S. and UK practice. Van Stratten et al. [16] performed a study of different design utilizing a bag system with SAG-M. Blood units were split using a sterile docking device into 4 equal volumes in small bags integrally attached to the main bag. These units were arbitrarily held for 21 days. Ninety-six preterm infants were classified as high or low risk based on the expected need for transfusion. Mean donor exposure was 1.1 with 3.2 2.1 transfusion/infant in the high and 1.1 with 0.4 transfusions/infant in the low risk groups, with limited wastage. This study was not designed to evaluate

Management of anemia in the newborn

Table 2 Quantity (total mg/kg) of additives infused during a transfusion of 15 mL per kg of AS-1 or AS-3 RBCs at Hct of 60% Additive Sodium chloride Dextrose Adenine Citrate Phosphate Mannitol AS-1 42 129 0.6 9.8 2.0 33 AS-3 7.5 23 0.6 12.6 5.6 0 Toxic dose 137 mg/kg/day 240 mg/kg/hr 15 mg/kg/dose 180 mg/kg/hr N 60 mg/kg/day 360 mg/kg/day

495

3. Criteria and guidelines

Transfusion practices can be guided by the use of guidelines which should be evidence-based, weigh benefits and risks and ensure that improved patient outcome is the primary end point. Optimally, well powered, randomized clinical trials, blinded and/or placebo controlled, would inform practice. In fact, such rigor is rare outside of large network trials. There are, however, at least two sets of transfusion guidelines that have been established based on consensus. These include those prepared by members of the Pediatric Hemotherapy Committee of the American Association of Blood Banks [22] and by the UK's National Blood Service [23]. These guidelines are often abstracted from clinical trials whose focus were neither reduction in transfusion nor morbidity and mortality (and so may not accurately reflect clinical practice in a rapidly changing area of neonatal medicine). Liberal versus restrictive transfusion strategies have been tested in several adult studies, in a few pediatric ICU and two recent NICU studies. In an effort to guide transfusion practices, Bell et al. [24] conducted a randomized single institution study comparing two sets of guidelines utilizing a threshold hematocrit below which a RBC transfusion was administered. Their hypothesis was that infants on the restrictive arm would require fewer transfusions with no more adverse outcomes than the liberally transfused infants. Among the 100 infants with birth weights of 500 to 1300 g in the study, the liberal group received more RBC transfusions (5.2 + 4.5) than the restrictive (3.3 + 2.9) with no difference in donor exposure (2.8 + 2.5 versus 2.2 + 2.0). The restrictive group had more brain hemorrhage, periventricular leukomalacia and apnea, ascribed to lower arterial oxygen and compensatory increase in cerebral blood flow. However, these findings were not repeated in the PINT study [25], a randomized trial of similar design of over 200 infants highlighting the need for further large, multi-institution randomized clinical trials in this area.

From Luban et al. [15], 11

safety or efficacy. In all published studies to date (Table 3), exposure of one to two donors is common with variable transfusion number and volume, despite the use of single unit assignment.

2. Iatrogenic anemia of infancy

Despite limiting donor exposures and transfusion episodes, premature infants still require transfusions of RBCs for iatrogenic loss [17] and for cardiorespiratory instability. A few studies [1720] have addressed blood sampling as a cause of iatrogenic anemia. Iatrogenic blood loss in 99 extremely premature infants was studied in relationship to severity of disease and gestational age. Sampling in mL/kg body weight exhibited a wide range (0.9 to 39 mL in the first week of life) and was correlated directly with volume transfused (mean 33.3 mL/kg) over a 4 week period [20]. Since some NICUs now use indwelling catheters more often than heel picks to obtain blood for testing and point of care testing (POCT) has dramatically decreased iatrogenic blood loss, the relevance of this study should be questioned. In a retrospective chart review, the effect of introduction of POCT on RBC transfusion frequency in the first 2 weeks of life was reviewed. There were 46 infants less than 1000 g pre POCT who received 5.7 + 3.7 transfusions and 78.4 + 51.6 mL/ kg transfusion volume versus 3.1 + 2.07 and 44.4 + 32.9 mL/kg in 34 infants in post POCT period [21]. These studies highlight the need to analyze the necessity of every blood draw, support the use of noninvasive monitoring methods and develop measures of true tissue oxygenation as a guide to the need for transfusion.

4. Measurements to define the need to transfuse

The observation that infants may be asymptomatic with low hemoglobin concentrations while others are symptomatic with similar or higher hemoglobin concentrations supports the concept that hemoglobin alone is an inadequate measure of the need to transfuse. The ability to reliably and non-

Table 3 Reference

Small-volume RBC transfusions given as stored RBCs to limit donor exposure without causing apparent adverse effects Solution CPDA-1 CPDA-1 NR AS-1 AS-3 SAG-M SAG-M AS-3 CPDA-1 AS-1 Storage 35 days 35 days 35 days 42 days 42 days 35 days 35 days 21 days 28 days 42 days Dose 15 mL/kg 13 mL/kg 15 mL/kg 15 mL/kg 15 mL/kg 15 mL/kg 15 mL/kg 7 mL/kg 15 mL/kg 15 mL/kg Hct (%) 75 6875 NR 85 85 NR NR 5560 NR 60 Transfusions 5.6 6.0 5.6 3.5 3.6 3.2 0.4 4.7 4.4 6.7 Donors 2.1 2.0 4.9 1.2 1.3 1.1 high risk 1.1 low risk 1.7 1.6 1.8

Liu Lee Wood Strauss Strauss van Straaten Mangel da Cunha Jain
NR = not recorded.

496 invasively measure tissue hypoxia could provide improved methods to evaluate different interventions and confirm the clinical significance of measured hemoglobin. Compensatory responses to decreased tissue oxygen concentration include increases in heart rate, cardiac output and cerebral blood flow. Fractional oxygen extraction (FOE) increases to maintain oxygen consumption. There is a progressive shift to anaerobic respiration which results in an increase in lactic acid. Several authors [26,27] have shown post-transfusion decrease in lactic acid and hypothesize that this reflects improved oxygen delivery. However, there are the technical and biological variables that affect the FOE measurement, which include fasting, cold, increased activity, hemolysis and venous occlusion, especially when the specimen is obtained from a peripheral vein. In a study by Fey and Losa [28] that attempted to control these variables, capillary whole blood lactic acid was analyzed pre- and 48 h post-transfusion in 18 premature infants. Lactic acid measurements did not correlate with either respiratory rate or bradycardiac episodes when regression analysis was performed [28]. The high coefficient of variation of 19.8% of repeated measures of lactic acid likely contributed to the poor correlation. Some have suggested that elevated lactic acid results from the catecholamine surge associated with sepsis or injury. In this circumstance, post-transfusion lactate might fall as a result of decrease in catecholamine response rather than from improvement in anaerobic metabolism. Utilizing a different approach, Wardle et al. present support for the use of near infrared spectroscopy (NIRS) measurement of FOE [29,30]. In a pilot study, 37 infants were randomized to one of two groups: the decision to transfuse was based on peripheral FOE or based on the hemoglobin concentration. The NIRS group was determined by an FOE of greater than 0.47 while the conventional group was transfused based on clinical need. Of the 56 transfusions given to the NIRS group, 33 (59%) were given because of clinical concerns. The investigators concluded that either their FOE criteria were insensitive or clinical indications complicated the study design. In another pilot study using a technique similar to NIRS, absolute concentrations of tissue hemoglobin, including oxygen bound and oxygen free hemoglobin was measured in 10 very low birth weight infants using diffuse optical spectroscopy (DOS) [31]. This methodology measured tissue oxygen concentrations non-invasively in muscle pre- and post-transfusion. Increases in tissue oxygenation were noted post-transfusion and correlated with mean hemoglobin increases for all 10 infants. Mock et al. [32] correlated hematocrit to RBC volume in 26 premature infants of birth weight less than 1300 g, studied on 43 occasions using a non-radioactive biotinylated RBC labeling flow cytometric method. Their goal was to develop an accurate RBC volume measurement and establish the relationship between circulating red cell volume and hematocrit to assess whether previously reported poor correlations between hematocrit and RBC volume was due to artifact or unique preterm physiology. They hypothesized that if RBC volume and hematocrit correlated, then hematocrit could be used as a definitive test for transfusion need. Despite good correlation between the two assays (r = 0.907), 95% confidence limits for predicting the circulatory RBC volume ranges were so broad as to make the RBC volume measurement of questionable clinical significance

N.L.C. Luban and usefulness. Other measurements like paired pre- and post-transfusion measurements of oxygen consumption (VO2 ), mixed venous oxygen saturation (M VO 2 ) and hemoglobin-oxygen dissociation curve alone or in combination would theoretically demonstrate post-transfusion improvement in oxygen delivery. Any method must be available real time or its usefulness will be limited.

5. In-line devices
If methods are insufficient to guide transfusions, combining methods to decrease iatrogenic loss and improve erythropoiesis might prove to be advantageous to the infant. Since blood gas analyses are a primary source of iatrogenic loss, in-line devices that permit continuous pH, pO2, and pCO2 measurements should theoretically decrease venous blood loss. Using 229 paired samples, an in-line intra-umbilical artery device versus specimens collected by phlebotomy were compared in 16 neonates monitored over 37 days. Blood volume loss and hemolysis were assessed as well as bias, precision and correlation of the in-line devices readout compared to reference methods. A dramatic difference in blood volume loss was observed: 24 7 L in those infants with in-line monitors versus 250 L for reference methods of commonly performed laboratory tests. The devices were not without problems which included large flush volumes, software issues, cost and practicality [33]. Using an in-line, ex vivo bedside monitor that withdraws blood through an umbilical artery catheter and returns it to the patient, Madan et al. [21] noted 25% less cumulative phlebotomy loss in the first two weeks of life in the monitored (n = 46) versus control group (n = 47). Based on previous studies, there was an anticipated 35% decrease in phlebotomy losses, which was only achieved in week one of the two-week analysis [21]. Because the analyte panel was limited, additional phlebotomy was still required. In future studies, technical issues and cost will need to be addressed as well as impact on mortality and morbidity, for which this study was underpowered. Time and technology will ultimately yield improvements with such devices.

6. Autologous transfusion
6.1. Cord clamping
True autologous transfusion in an infant can occur by delaying cord clamping or by collecting, storing, and re-infusing cord blood as a blood product. In a randomized study, 39 infants of b 33 weeks gestation were randomized to a 20 s (n = 20) or 45 s (n = 19) delay in cord clamping with oxytoxin administration after delivery of the first shoulder in mothers delivering by Caesarian section. By day 42 of life, 16 of 20 infants in the immediate and 9 of 19 in the delayed clamp group had been transfused. This resulted in 2.4 transfusions in the immediate versus 1.2 transfusions in the delayed group (p b 0.05) despite equivalent iatrogenic blood loss [34]. This supports the early work of Kinmond who demonstrated less hypovolemia and fewer days of both oxygen dependency and transfusion need in premature infants delivered vaginally [35], but refutes the work of others who showed no advantage of late clamping to the infant. McDonnell and Henderson-Smart [36] measuring circulating red cell volume using biotinylated RBCs

Management of anemia in the newborn could provide the physiological basis of delay of cord clamping. In another study, infants born following delayed clamp had greater circulatory red cell volumes (42.1 mL/kg) as compared to immediate (36.8 mL/kg). Clinical benefit was not measured in this study [37]. Most recently, blood volume measurements using either a dilution of fetal hemoglobin or biotin-labeled autologous red blood cell method were performed in 46 preterm infants, 24 to 32 weeks gestation, randomized to either an early or delayed clamping. This study confirmed that infants born with delayed clamp time had a greater blood volume (74.4 mL/kg) than early clamp group (62.7 mL/kg) for both vaginal and cesarean deliveries. Of interest no differences were noted for hematocrit measured at 4 h [38]. The potential advantages and disadvantages of clamping have been debated and are different for term versus preterm infants. In countries with limited access to safe blood for transfusion, the value of late cord clamping might be of significant value. A meta-analysis of early versus late cord clamping including 7 studies of 297 preterm infants, defined as less than 37 weeks gestation has been published recently. In this analysis, Rabe demonstrated that delayed cord clamping (30 to 120 s) was associated with fewer transfusions for anemia (RR 2.01) or lower blood pressure (RR 2.58) and less intraventricular hemorrhage (RR 1.74) [39]. This meta-analysis supports the need to implement changes in clinical obstetrical practice.

497 Encourage delayed cord clamping to increase newborn red cell mass/volume.

Research directions
Perform well powered studies of restrictive vs. liberal transfusion criteria. Develop real time noninvasive measures of tissue oxygen delivery. Develop in-line and reflectometric devices to diminish iatrogenic blood loss. Establish systems to select, screen and collect blood from a single donor that can be processed and packaged to further limit donor exposure.

References
[1] Maier RF, Sonntag J, Walka MM. Changing practices of red blood cell transfusions in infants with birth weights less than 1000 g. J Pediatr 2000;136:2204. [2] Widness JA, Seward VJ, Kromer IJ, Burmeister LF, Bell EF, Strauss RG. Changing patterns of red blood cell transfusion in very low birth weight infants. J Pediatr 1996;129:6807. [3] Strauss RG, Burmeister LF, Johnson K. AS-1 red blood cells for neonatal transfusions: a randomized trial assessing donor exposure and safety. Transfusion 1996;36:8738. [4] Strauss RG, Burmeister LF, Johnson K. Feasibility and safety of AS-3 red blood cells for neonatal transfusions. J Pediatr 2000;136:2159. [5] Liu EA, Mannino FL, Lane TA. Prospective, randomized trial of the safety and efficacy of a limited donor exposure transfusion program for premature neonates. J Pediatr 1994;125:926. [6] Lee DA, Slagel TA, Jackson TM, Evans CS. Reducing blood donor exposures in low birth weight infants by the use of older, unwashed packed red blood cells. J Pediatr 1995;126:2806. [7] Wood A, Wilson N, Skacel P. Reducing donor exposure in preterm infants requiring multiple blood transfusions. Arch Dis Child Fetal Neonatal Ed 1995;72:F2933. [8] Goodstein MH, Locke RG, Wlodarczyk D. Comparison of two preservation solutions for erythrocyte transfusions in newborn infants. J Pediatr 1993;123:7838. [9] Strauss RG. Data-driven blood banking practices for neonatal RBC transfusions. Transfusion Dec 2000;40:152840. [10] Mangel J, Goldman M, Garcia C, Spurll G. Reduction of donor exposures in premature infants by the use of designated adenine-saline preserved split red blood cell packs. J Perinatol Sep 2001;21(6):3637. [11] Fernandes da Cunha DH, Nunes Dos Santos AM, Kopelman BI, et al. Transfusions of CPDA-1 red blood cells stored for up to 28 days decrease donor exposures in very low-birth-weight premature infants. Transfus Med 2005;15:46773. [12] Jain R, Jarosz C. Safety and efficacy of AS-1 red blood cell use in neonates. Transfus Apher Sci 2001;24:1115. [13] Strauss RG, Villhauer PJ, Cordle DG. A method to collect, store and issue multiple aliquots of packed blood cells for neonatal transfusion. Vox Sang 1995;68:7781. [14] Sherwood WC, Donato T, Clapper C. The concentration of AS-1 RBCs after inverted gravity sedimentation for neonatal transfusions. Transfusion 2000;40:6189. [15] Luban NLC, Strauss RG, Hume HA. Commentary on the safety of red cells preserved in extended-storage media for neonatal transfusions. Transfusion 1991;31:22935. [16] van Straaten HL, de Wildt-Eggen J, Huisveld IA. Evaluation of a strategy to limit blood donor exposure in high risk premature newborns based on clinical estimation of transfusion need. J Perinat Med 2000;28(2):1228.

7. Cord blood collection

Collection of autologous cord blood for storage and reinfusion has been studied infrequently. In one study, there were 44 collections into a closed blood collection system with CPD from infants with a mean gestational age of 32 4 weeks while the placenta was still in utero. A mean of forty-four 3.4 mL of cord blood was collected which, at a transfusion volume of 1015 mL/kg, could provide one or 2 transfusions. Bacterial contamination rate was high (15.8%) [40]. While autologous collections can be transfused without adverse consequences [41], smaller collection volumes with poor product quality in infants weighing less than 1000 g raise issues as to whether this process, while feasible, is worthwhile. In another study of infants with congenital anomalies diagnosed prenatally, 26 infants had blood collected and transfused during the first 3 days post delivery. No bacterial contamination was noted. The unusual success in collection and transfusion was ascribed to patient selection, meticulous preparation, and the short storage time [42]. This level of success has been repeated in a recent study of 52 newborns where autologous blood was held for up to 23 days; 40% of infants between 1000 and 2500 g were supported using only autologous blood in this study [43].

Key guidelines
Utilize transfusion guidelines based on published recommendations, updated regularly based on randomized clinical trials. Assign one or two infants to a single RBC unit for small volume transfusions to limit donor exposure without concern for age/anticoagulant. r-EPO should be limited to selected premature infants until studies better inform its use [4452].

498
[17] Nexo E, Christensen NC, Olesen H. Volume of blood removed for analytic purposes during hospitalization of low-birthweight infants. Clin Chem 1981;5:75961. [18] Ringer SA, Richardson DK, Sacher RA. Variations in transfusion practice in neonatal intensive care. Pediatrics 1998;101:194200. [19] Obladen M, Sachsenweger M, Stahnke M. Blood sampling in very low birthweight infants receiving different levels of intensive care. Eur J Pediatr 1988;147:399404. [20] Madsen LP, Rasmussen MK, Bjerregaard LL. Impact of blood sampling in very preterm infants. Scand J Clin Lab Invest 2000;60:12532. [21] Madan A, Kumar R, Adams MM, Benitz WE, Goaghan SM, Widness A. Reduction in red blood cell transfusions using a bedside analyzer in extremely low birth weight infants. J Perinatol 2005;25:215. [22] Roseff SD, Luban NL, Manno CS. Guidelines for assessing appropriateness of pediatric transfusion. Transfusion 2002;42:1398413. [23] Gibson BE, Todd A, Boulton F, Roberts I, Pamphilon D, Rodeck C, et al. Transfusion guidelines for neonates and older children. Br J Haematol 2004;124:43353. [24] Bell EF, Strauss RG, Widness JA, Mahoney LT, Mock DM, et al. Randomized trial of liberal versus restrictive guidelines for red blood cell transfusion in preterm infants. Pediatrics 2005;115:186591. [25] Kirpalani H, Whyte RK, Andersen C, Asztalos EV, Heddle N, Blajchman MA, et al. 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;149:3017. [26] Soubasi V, Kremenopoulos G, Tsantali C. Use of erythropoietin and its effects on blood lactate and 2,3-Diphosphoglycerate in premature neonates. Biol Neonate 2000;78:2817. [27] Moller JC, Schwarz U, Schaible TF. Do cardiac output and serum lactate levels indicate blood transfusion requirements in anemia of prematurity. Intensive Care Med 1996;22:4726. [28] Frey B, Losa M. The value of capillary whole blood lactate for blood transfusion requirements in anemia of prematurity. Intensive Care Med 2001;27:2227. [29] Wardle SP, Weindling M. Peripheral fractional oxygen extraction and other measures of tissue oxygenation to guide blood transfusions in preterm infants. Semin Perinat 2001;25:604. [30] Wardle SP, Garr R, Yoxall CW, Weindling AM. A pilot randomised controlled trial of peripheral fractional oxygen extraction to guide blood transfusions in preterm infants. Arch Dis Child Neonatal Ed 2002;86(1):F227. [31] Cerussi A, Van Woerkom R, Waffarn F, Tromberg B. Noninvasive monitoring of red blood cell transfusion in very low birthweight infants using diffuse optical spectroscopy. J Biomed Opt 2005;10:051401. [32] Mock DM, Bell EF, Lankford GL. Hematocrit correlates well with circulating red blood cell volume in very low birth weight infants. Pediatr Res 2001;50:52531. [33] Widness JA, Kulhavy JC, Johnson KJ. Clinical performance of an inline point-of-care monitor in neonates. Pediatr 2000;106:497504. [34] Rabe H, Wacker A, Hulskamp Georg. A randomised controlled trial of delayed cord clamping in very low birth weight preterm infants. Eur J Pediatr 2000;159:7757. [35] Kinmond S, Aitchison TC, Holland BM. Umbilical cord clamping and preterm infants: a randomised trial. BMJ 1993;306:1725. [36] McDonnell M, Henderson-Smart DJ. Delayed umbilical cord clamping in preterm infants: a feasibility study. J Paediatr Child Health 1997;33:30810.

N.L.C. Luban
[37] Strauss RG, Mock DM, Johnson K, Mock NI, Cress G, Knosp L. Circulating RBC volume, measured with biotinylated RBCs is superior to the Hct to document the hematologic effects of delayed versus immediate umbilical cord clamping in preterm neonates. Transfusion 2003;43:116872. [38] Aladangady N, McHugh S, Aitchison TC, Wardrop CA, Holland BM. Infants' blood volume in a controlled trial of placental transfusion at preterm delivery. Pediatrics 2006;117:938. [39] Rabe H, Reynolds G, Diaz-Rossello J. Early versus delayed umbilical cord clamping in preterm infants. Cochrane Database Syst Rev 2004;4:CD003248. [40] Eichler H, Schaible T, Richter E. Cord blood as a source of autologous RBCs for transfusion to preterm infants. Transfusion 2000;40:11117. [41] Surbek DV, Glanzmann R, Senn HP. Can cord blood be used for autologous transfusion in preterm neonates. Eur J Pediatr Oct 2000;159:7901. [42] Imura K, Kawahara H, Kitayama Y. Usefulness of cord-blood harvesting for autologous transfusion in surgical newborns with antenatal diagnosis of congenital anomalies. J Pediatr Surg 2001;36:8514. [43] Brune T, Garritsen H, Hentschel R. Efficacy, recovery and safety of RBC from autologous placental blood: clinical experience in 52 newborns. Transfusion 2003;43:12106. [44] Vamvakas EC, Strauss RG. Meta-analysis of controlled clinical trials studying the efficacy of recombinant human erythropoietin in reducing blood transfusions in the anemia of prematurity. Transfusion 2001;41:40615. [45] Ohls RK, Harcum J, Schibler KR, Christensen RD. The effect of erythropoietin on the transfusion requirements of preterm infants weighing 750 grams or less: a randomized, double-blind, placebo-controlled study. J Pediatr 1997;131(5):6615. [46] Ohls RK, Ehrenkranz RA, Wright LL, Lemons JA, Korones SB, Stoll BJ, et al. A multicenter randomized double-masked placebo-controlled trial of early erythropoietin and iron administration to preterm infants. Pediatrics 2001;108:93442. [47] Maier RF, Obladen M, Mller-Hansen I, Kattner E, Merz U, Arlettaz R, et al. Early treatment with erythropoietin b ameliorates anemia and reduces transfusion requirements in infants with birth weights below 1000 g. J Pediatr 2002;141:815. [48] Pollak A, Hayde M, Hayn M, Herkner K, Lombard KA, Lubec G, et al. Effect of intravenous iron supplementation on erythropoiesis in erythropoietin-treated premature infants. Pediatrics 2001;107:7885. [49] Akisu M, Tuzun S, Arslanoglu S, Yalaz M, Kultursay N. Effect of recombinant human erythropoietin administration on lipid peroxidation and antioxidant enzymes(s) activities in preterm infants. Acta Med Okayama 2001;55:35762. [50] Romagnoli C, Zecca E, Gallini F, Girlando P, Zuppa AA. Do recombinant human erythropoietin and iron supplementation increase the risk of retinopathy of prematurity. Eur J Pediatr 2000;159:62734. [51] Demers EF, McPherson RJ, Juul SE. Erythropoietin protects dopaminergic neurons and improves neurobehavioral outcomes in juvenile rats after neonatal hypoxia-ischemia. Pediatr Res 2005;58:297301. [52] Franz AR, Pohlandt F. Red blood cell transfusions in very and extremely low birthweight infants under restrictive transfusion guidelines: is exogenous erythropoietin necessary. Arch Dis Child Fetal Neonatal Ed 2001;84:F96F100.