Early versus late erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants
Abstract
Background
Low plasma levels of erythropoietin (EPO) in preterm infants provide a rationale for the use of EPO to prevent or treat anaemia.
Objectives
To assess the effectiveness and safety of early versus late initiation of EPO in reducing red blood cell (RBC) transfusions in preterm and/or low birth weight (LBW) infants.
Search methods
The standard search of the Cochrane Neonatal Review Group (CNRG) was performed in 2006 and updated in 2009. Updated search in September 2009 as follows: The Cochrane Library, MEDLINE (search via PubMed), CINAHL and EMBASE were searched from 2005 to September 2009. The searches were repeated in March 2012. The Pediatric Academic Societies' Annual meetings were searched electronically from 2000 to 2012 at Abstracts2ViewTM as were clinical trials registries (clinicaltrials.gov; controlled‐trials.com; and who.int/ictrp).
Selection criteria
Randomised or quasi‐randomised controlled trials enrolling preterm or LBW infants less than eight days of age. Intervention: Early initiation of EPO (initiated at less than eight days of age) versus late initiation of EPO (initiated at eight to 28 days of age).
Data collection and analysis
The standard methods of the CNRG were followed. Weighted treatment effects included typical risk ratio (RR), typical risk difference (RD), number needed to treat to benefit (NNTB), number needed to treat to harm (NNTH) and mean difference (MD), all with 95% confidence intervals (CI). A fixed‐effect model was used for meta‐analyses and heterogeneity was evaluated using the I‐squared (I2) test.
Main results
No new trials were identified in March of 2012. Two high quality randomised double‐blind controlled studies enrolling 262 infants were identified. A non‐significant reduction in the 'Use of one or more RBC transfusions' [two studies 262 infants; typical RR 0.91 (95% CI 0.78 to 1.06); typical RD ‐0.07 (95% CI ‐0.18 to 0.04; I2 = 0% for both RR and RD] favouring early EPO was noted. Early EPO administration resulted in a non‐significant reduction in the "number of transfusions per infant" compared with late EPO [typical MD ‐ 0.32 (95% CI ‐0.92 to 0.29)]. There was no significant reduction in total volume of blood transfused per infant or in the number of donors to whom the infant was exposed. Early EPO led to a significant increase in the risk of retinopathy of prematurity (ROP) (all stages) [two studies, 191 infants; typical RR 1.40 (95% CI 1.05 to 1.86); typical RD 0.16 (95% CI 0.03 to 0.29); NNTH 6 (95% CI 3 to 33)]. There was high heterogeneity for this outcome (I2 = 86% for RR and 81% for RD). Both studies (191 infants) reported on ROP stage > 3. No statistically significant increase in risk was noted [typical RR 1.56 (95% CI 0.71 to 3.41); typical RD 0.05 (‐0.04 to 0.14)] There was no heterogeneity for this outcome (0% for both RR and RD). No other important favourable or adverse neonatal outcomes or side effects were reported.
Authors' conclusions
The use of early EPO did not significantly reduce the 'Use of one or more RBC transfusions' or the 'Number of transfusions per infant" compared with late EPO administration. The finding of a statistically significant increased risk of ROP (any grade) and a similar trend for ROP stage > 3 with early EPO treatment is of great concern.
PICOs
Plain language summary
Early versus late erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants
The number of red blood cells falls after birth in preterm infants due to the natural breakdown of erythrocytes and blood letting. Low levels of erythropoietin (EPO), a substance in the blood that stimulates red blood cell production in preterm infants, provide a rationale for the use of EPO to prevent or treat anaemia. A total of 262 infants born preterm have been enrolled in two studies of early versus late administration of EPO to prevent blood transfusions. There were no demonstrable benefits of early versus late administration of EPO with regards to reduction in the use of red blood cell transfusions, number of transfusions, the amount of red cells transfused or number of donor exposures per infant. However, the use of early EPO compared with late EPO administration increases the risk of retinopathy of prematurity, a serious complication in babies born before term. Currently, there is a lack of evidence that either treatment confers any substantial benefits with regard to any donor blood exposure, as many infants enrolled in both studies were exposed to donor blood prior to study entry, and early EPO increases the risk of retinopathy of prematurity. Neither early nor late administration of EPO is recommended.
Authors' conclusions
Background
Description of the condition
The haemoglobin concentration falls to minimal levels of 11 g/dL in term infants by eight to 12 weeks of age and 7.0 to 10.0 g/dL in preterm infants by six weeks of age (Stockman 1978). This process is called physiologic anaemia of infancy (Strauss 1986). In very low birth weight (VLBW) infants, the haematocrit falls to approximately 24% in infants weighing 1.0 to 1.5 kg and to 21% in infants weighing less than 1.0 kg at birth (Stockman 1986). In extremely low birth weight (ELBW) infants, this decline in haematocrit is not "physiologic", as it is associated with clinical findings that prompt red blood cell transfusions. The diagnostic accuracy of different clinical signs and laboratory findings has not been studied. It is still unknown how low haematocrit levels can fall before clinical signs of anaemia of prematurity occur and what is the minimal haematocrit level acceptable in infants requiring supplemental oxygen (Ohls 2002). Nevertheless, "top‐up" transfusions to treat low haemoglobin or low haematocrit levels are frequently used. As many as 80% of VLBW infants and 95% of ELBW infants receive blood transfusions during their hospitalisations (Widness 1996).
Description of the intervention
Erythropoietin (EPO) and iron effectively stimulate erythropoiesis. Plasma erythropoietin levels in neonates are lower than those for older children and adults. Brown and colleagues reported that, between days two and 30 of life, the mean EPO concentration was 10 mLU/mL, as compared to 15 mLU/mL in concurrently studied adults (Brown 1983). A low plasma EPO level is a key reason that nadir haematocrit values of preterm infants are lower than those of term infants (Stockman 1986; Dallman 1981). Low plasma EPO levels provide a rationale for use of EPO in the prevention or treatment of anaemia of prematurity. Studies in newborn monkeys and sheep have demonstrated that neonates have a large volume of distribution and more rapid elimination of EPO, necessitating the use of higher doses than required for adults (Ohls 2000). A systematic review of EPO administration noted a wide range of doses used, from 90 to 1400 IU/kg/week (Kotto‐Kome 2004). Side effects reported in adults include hypertension, bone pain, rash and rarely seizures. Only transient neutropenia has been reported in neonates (Ohls 2000).
How the intervention might work
The primary goal of EPO therapy is to reduce transfusions. Most transfusions are given during the first three to four weeks of life. The larger or stable preterm infants, who respond best to EPO, receive few transfusions. ELBW infants, who are sick and have the greatest need for red blood cell (RBC) transfusions shortly after birth, have not consistently responded to EPO. This suggests that EPO is a more effective erythropoietic stimulator in more mature neonates. ELBW neonates are more likely to need transfusions even if their erythropoiesis is stimulated (Kotto‐Kome 2004). In addition, ELBW neonates have a smaller blood volume and the relatively larger phlebotomy volumes that are required during hospital stay often necessitate "early" transfusions. In contrast "late" transfusions are more often given because of anaemia of prematurity (Garcia 2002). Most preterm infants who require blood transfusions will receive their first transfusion in the first two weeks of life (Zipursky 2000). Reducing the number of RBC transfusions reduces the risk of transmission of viral infections and may reduce costs. Frequent RBC transfusions may be associated with retinopathy of prematurity (ROP) (Hesse 1997) and bronchopulmonary dysplasia (BPD).
Preterm infants need iron for erythropoiesis. As neonatal blood volume expands with rapid growth, infants produce large amounts of haemoglobin. Several studies have observed a decrease in serum ferritin concentration ‐ an indication of iron deficiency (Finch 1982) ‐ during EPO treatment. The use of higher, more effective doses of EPO might be expected to be particularly likely to increase iron demand and the risk of iron deficiency (Genen 2004). Iron supplementation during EPO treatment has been observed to reduce the risk of the development of iron deficiency (Shannon 1995). The range of iron doses used in EPO treated infants is between 1 mg/kg/day to 10 mg/kg/day (Kotto‐Kome 2004).
Why it is important to do this review
The efficacy of EPO in anaemia of prematurity has recently been systematically reviewed (Vamvakas 2001; Garcia 2002; Kotto‐Kome 2004). Vamvakas et al concluded that there is extreme variation in the results, and until this variation is better understood, it is too early to recommend EPO as standard treatment for the anaemia of prematurity (Vamvakas 2001). Garcia et al concluded that administering EPO to VLBW neonates can result in a modest reduction in late erythrocyte transfusions and that this effect is dependent on the dose of EPO used (Garcia 2002). Kotto‐Kome et al concluded that if EPO is begun in the first week of life, a moderate reduction can be expected in the proportion of VLBW neonates transfused. The reduction is less significant for early transfusion than for late transfusion (Kotto‐Kome 2004).
Additional studies of EPO in preterm or LBW infants have been published since the reviews noted above, justifying additional reviews. The cut‐off of less than eight days of age for early and > eight days for late treatment with EPO, although somewhat arbitrary, was chosen based on previously published meta‐analyses (Garcia 2002; Kotto‐Kome 2004).
This review compares early administration of EPO (starting in infants less than eight days of age) versus late administration of EPO (starting > eight days). The main rationale for this review was to evaluate whether early treatment with EPO in preterm infants is more effective than late treatment to decrease exposure to RBC transfusion and the total transfusions required. We performed a systematic review to compare all available studies where EPO was begun during first week of life versus EPO started after the first week of life to assess the effect on any and total number of erythrocyte transfusions.
Objectives
Primary objective:
To assess the effectiveness and safety of early (before eight days after birth) versus late (between eight to 28 days after birth) initiation of EPO in reducing the need for red blood cell transfusions in preterm and/or low birth weight infants.
Subgroup analyses:
We planned subgroup analyses within this review for low (< 500 IU/kg/week) and high (> 500 IU/kg/week) doses of EPO, and low (< 5 mg/kg/day) and high (> 5 mg/kg/day) doses of supplemental iron administered by any route.
Methods
Criteria for considering studies for this review
Types of studies
Randomised or quasi‐randomised controlled trials.
Types of participants
Preterm (< 37 weeks) and/or low birth weight (< 2500 g) neonates less than eight days of age.
Types of interventions
Early initiation of EPO (initiated before eight days of age, using any dose, route or duration) versus late initiation of EPO (initiated between eight to 28 days of age, using any dose, route or duration).
Types of outcome measures
Primary outcomes
-
Use of one or more red blood cell (RBC) transfusions.
Secondary outcomes
-
the total volume (mL/kg or mL/kg/day) of RBCs transfused per infant;
-
number of transfusions per infant;
-
number of donors to whom the infant was exposed;
-
mortality during initial hospital stay (all causes);
-
retinopathy of prematurity (ROP) (any stage and stage > 3);
-
proven sepsis (clinical symptoms and signs of sepsis and positive blood culture for bacteria or fungi);
-
necrotising enterocolitis (NEC) (Bell's stage II or more);
-
intraventricular haemorrhage (IVH); all grades and severe IVH (grades III and IV);
-
periventricular leukomalacia (PVL); cystic changes in the periventricular areas;
-
length of hospital stay;
-
bronchopulmonary dysplasia (BPD) (supplemental oxygen at 28 days of age or at 36 weeks postmenstrual age and compatible X‐ray);
-
sudden infant death (SID) after discharge;
-
long‐term outcomes assessed at any age beyond one year of age by a validated cognitive, motor, language, or behavioural/school/social interaction/adaptation test;
-
neutropenia;
-
any side effects reported in the trials.
Search methods for identification of studies
Electronic searches
The Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, 2006, Issue 2) was searched to identify relevant randomised and quasi‐randomised controlled trials. MEDLINE was searched for relevant articles published from 1966 to November 2005 using the following MeSH terms or text words: (exp Erythropoietin/OR erythropoietin:.mp. OR rhuepo.mp.) AND (anaemia/OR exp anaemia, neonatal/) AND (blood transfusion/OR blood component transfusion/OR erythrocyte transfusion/) AND (infant, newborn/OR infant, low birth weight/OR infant, very low birth weight/OR infant, premature/OR exp Infant, Premature, Diseases) OR (neonate: OR prematur*: OR newborn:).mp. OR newborn infant [age limit]) AND (clinical trial.pt. OR Randomized Controlled Trials/OR (random: OR rct OR rcts OR blind OR blinded OR placebo:).mp. OR (review.pt. OR review, academic.pt.) AND human. EMBASE from 1980 to November 2005 and CINAHL 1982 to November 2005 using the following MeSH terms or text words: (Erythropoietin/OR erythropoietin: OR epo OR epogen OR epoetin: OR (rhuepo).mp. AND (anaemia/OR exp anaemia, neonatal/) AND (blood transfusion/OR exp blood component transfusion/OR erythrocytes/) AND exp Infant, Premature, Diseases/OR infant, newborn/OR infant, low birth weight/OR infant, very low birth weight/OR infant, premature/OR (neonate: OR newborn: OR prematur*:).mp. OR newborn infant [age limit].
In September 2009, we updated the search as follows: The Cochrane Library, MEDLINE (search via PubMed), CINAHL and EMBASE were searched from 2005 to September 2009.
Search terms: erythropoietin OR rhuepo AND anaemia OR anaemia AND blood transfusion OR blood component transfusion OR erythrocyte transfusion. Limits: human, infant and clinical trial. No language restrictions were applied. Clinicaltrials.gov was also searched with no date restriction. Search terms: Erythropoietin OR rhuepo AND infant.
In March 2012, we updated the searches as per above. In addition we searched clinical trials registries (clinicaltrials.gov; controlled‐trials.com; and who.int/ictrp).
Searching other resources
In addition to the electronic searches, manual searches of bibliographies and personal files were performed. No language restrictions were applied.
Abstracts published from the Pediatric Academic Societies' Meetings and the European Society of Pediatric Research Meetings (published in Pediatric Research) were handsearched from 1980 to April 2006.
In May 2012 abstracts from the Pediatric Academic Societies' Annual Meetings were searched electronically at Abstracts2ViewTM from 2000 to 2012.
Data collection and analysis
The standard review methods of the Cochrane Neonatal Review Group and the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) were used.
Selection of studies
For the original review, both review authors assessed all abstracts and published studies identified as potentially relevant by the literature search for inclusion in the review.
Data extraction and management
For the original review, each review author extracted data separately on a data abstraction form. The information was then compared and differences were resolved by consensus. One review author (AO) entered data into RevMan 5.1 (RevMan 2011) and the other (SA) cross checked the printout against his own data abstraction forms and any errors were corrected. We planned that for any studies identified only as abstracts, we would contact the primary author to obtain further information.
Assessment of risk of bias in included studies
For the original review, the quality of included trials was evaluated independently by the review authors, using the following criteria:
Blinding of randomisation; Blinding of intervention; Blinding of outcome measure assessment; Completeness of follow‐up. There are three potential answers to these questions yes, no, cannot tell. This information was added to the Characteristics of included studies table.
For the update in 2012, the following issues were evaluated and entered into the 'Risk of bias' table.
Selection bias (random sequence generation and allocation concealment).
For each included study, we categorised the risk of selection bias as:
Random sequence generation
Low risk ‐ adequate (any truly random process e.g. random number table; computer random number generator);
High risk ‐ inadequate (any non random process e.g. odd or even date of birth; hospital or clinic record number);
Unclear risk ‐ no or unclear information provided.
Allocation concealment
For each included study, we categorised the risk of bias regarding allocation concealment as:
Low risk ‐ adequate (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
High risk ‐ inadequate (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth);
Unclear risk ‐ no or unclear information provided
Performance bias
For each included study, we categorised the methods used to blind study personnel from knowledge of which intervention a participant received. (As our study population consisted of neonates they would all be blinded to the study intervention).
Low risk ‐ adequate for personnel (a placebo that could not be distinguished from the active drug was used in the control group);
High risk ‐ inadequate ‐ personnel aware of group assignment;
Unclear risk ‐ no or unclear information provided.
Detection bias
For each included study, we categorised the methods used to blind outcome assessors from knowledge of which intervention a participant received. (As our study population consisted of neonates they would all be blinded to the study intervention). Blinding was assessed separately for different outcomes or classes of outcomes. We categorised the methods used with regards to detection bias as:
Low risk ‐ adequate; follow‐up was performed with assessors blinded to group assignment;
High risk ‐ inadequate; assessors at follow‐up were aware of group assignment;
Unclear risk ‐ no or unclear information provided
Attrition bias
For each included study and for each outcome, we described the completeness of data including attrition and exclusions from the analysis. We noted whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported or supplied by the trial authors, we re‐included missing data in the analyses. We categorised the methods with respect to the risk attrition bias as:
Low risk ‐ adequate (< 10% missing data);
High risk ‐ inadequate (> 10% missing data);
Unclear risk ‐ no or unclear information provided.
Reporting bias
For each included study, we described how we investigated the risk of selective outcome reporting bias and what we found. We assessed the methods as:
Low risk ‐ adequate (where it is clear that all of the study's pre‐specified outcomes and all expected outcomes of interest to the review have been reported);
High risk ‐ inadequate (where not all the study's pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);
Unclear risk ‐ no or unclear information provided (the study protocol was not available).
Other bias
For each included study, we described any important concerns we had about other possible sources of bias (for example, whether there was a potential source of bias related to the specific study design or whether the trial was stopped early due to some data‐dependent process). We assessed whether each study was free of other problems that could put it at risk of bias as:
Low risk ‐ no concerns of other bias raised;
High risk ‐ concerns raised about multiple looks at the data with the results made known to the investigators, difference in number of patients enrolled in abstract and final publications of the paper;
Unclear ‐ concerns raised about potential sources of bias that could not be verified by contacting the authors.
If needed, we planned to explore the impact of the level of bias through undertaking sensitivity analyses.
For the original review, independent quality assessments were conducted by two review authors (SA, AO), who were not blinded to authors, institution or journal of publication.
The update in 2012 was conducted by one author (AO).
Measures of treatment effect
Statistical analyses were performed using Review Manager software (RevMan 5.1) (RevMan 2011). Categorical data were analysed using risk ratio (RR), risk difference (RD), number needed to treat to benefit (NNTB) or number needed to harm (NNTH) for dichotomous outcomes and mean difference (MD) for continuous data. The 95% Confidence interval (CI) was reported on all estimates.
Assessment of heterogeneity
We examined heterogeneity between trials by inspecting the forest plots and quantifying the impact of heterogeneity using the I2 statistic (Higgins 2003). We used the following criteria for describing the percentages of heterogeneity; < 25% no heterogeneity, > 25% to 49% low heterogeneity, 50% to 74% moderate heterogeneity and > 75% high heterogeneity. If we detected statistical heterogeneity, we planned to explore the possible causes (for example, differences in study quality, participants, intervention regimens, or outcome assessments) using post hoc subgroup analyses if possible.
Data synthesis
The analysis was performed using RevMan5.1 (RevMan 2011). For estimates of typical RR and RD, we used the Mantel‐Haenszel method. For measured quantities, we used the inverse variance method. All meta‐analyses were done using the fixed‐effect model.
Subgroup analysis and investigation of heterogeneity
We planned subgroup analyses within this review for low (< 500 IU/kg/week) and high (> 500 IU/kg/week) doses of EPO, and low (< 5 mg/kg/day) and high (> 5 mg/kg/day) doses of supplemental iron administered by any route.
Results
Description of studies
The update in 2012 did not identify any additional studies. Two studies enrolling 268 infants were identified (Donato 2000; Maier 2002). For details of the studies see table Characteristics of included studies. One non randomised study was excluded (Rudzinska 2002). For transfusion guidelines see Additional table (Table 1 Transfusion guidelines).
| Reference | Indications |
| Indications for transfusion followed slightly modified criteria described in a previous study (Shannon 1995). | |
| Infants on assisted ventilation or > 40% of inspired oxygen were not transfused unless Hct dropped below 0.40. Spontaneously breathing infants were not transfused unless Hct dropped below 0.35 during the first 2 weeks of life, 0.30 during the 3rd to 4th weeks, and 0.25 thereafter. Transfusion was allowed when life threatening anaemia or hypovolaemia was diagnosed by the treating neonatologist, or surgery was planned. Twelve of the 14 centres used satellite packs of the original red cell pack to reduce donor exposure. The amount of packed red cells transfused was not reported. |
CPAP = continuous positive airway pressure
Hct = haematocrit
Donato 2000: This was a blinded multicentre randomised placebo‐controlled study conducted in seven private hospitals in Buenos Aires, Argentina between July 1996 to October 1997.
-
Objective: To evaluate whether early treatment (day two to three of life) with EPO in preterm infants with birthweight < 1250 g 1) decreases the number of transfusions during the first two weeks of life; and 2) is more effective than late treatment (day 15) to decrease the total transfusion requirement.
-
Population: Infants with a birthweight < 1250 g and a gestational age < 32 weeks. Infants were excluded if they had major congenital malformations, chromosomal anomalies, haemolytic and/or haemorrhagic disease, intrauterine infections, systemic hypertension, and neutropenia (absolute neutrophil count < 1.5x109/L). No infant was excluded based on severity of illness.
-
Intervention: In the early EPO group, infants received EPO 1250 IU/kg/week (high dose) intravenously (iv) from day two to day 14 of life; in the late EPO group, infants received placebo during this time period. Subsequently, all infants received EPO 750 IU/kg/week (high dose) subcutaneously (sc) for six additional weeks. All infants were given oral iron 6 mg/kg/day as ferrous sulphate (high dose), starting as soon as enteral feedings were initiated and continuing during the entire treatment period as well as folic acid.
-
Outcomes assessed: Number of transfusions during the first two weeks of life and the total transfusion requirement per infant. Other outcomes included use of one ore more RBC transfusions, mortality during hospital stay, intraventricular haemorrhage (grade III/IV), weight gain, ROP (during the study period and among infants examined during the first year of life) and sudden infant death syndrome (SIDS).
Maier 2002: This was a blinded multicentre randomised placebo‐controlled study conducted in 14 centres in four European countries (Belgium, France, Germany, Switzerland) between May 1998 to June 1999. An early EPO, a late EPO and a control group were studied. The early EPO and late EPO groups were eligible for inclusion in this review.
-
Objective: To investigate whether EPO reduces the need for transfusion in ELBW infants and to determine the optimal time for treatment.
-
Population: Infants with a birthweight between 500 to 999 g were eligible. Infants were excluded prior to randomisation if they had cyanotic heart disease, major congenital malformation requiring surgery, mother had received an investigational drug during pregnancy, gestational age > 30 completed weeks or lack of consent.
-
Intervention: In the early EPO group infants received 750 IU/kg/week of EPO (high dose) from the first week of life for nine weeks. In the late EPO group, infants received the same treatment three weeks later (high dose). Treatment in both groups continued until days 65 to 68 of life. EPO was given iv in both groups as long as the infant had an iv line in place and sc thereafter. The late EPO group (until EPO was given) received sham‐injections. Enteral iron 3 mg/kg was given to all infants from days three to five (low dose) and was increased at days 12 to 14 to 6 mg/kg/day (high dose) and to 9 mg/kg/day at days 24 to 26 of life (high dose).
-
Outcomes assessed: The primary outcome was no transfusion and haematocrit levels never below 30%. Other outcomes included use of one ore more RBC transfusions, mean number of transfusions per infant, volume of red cells transfused (ml/kg/day), mortality during hospital stay, NEC, IVH (grade III/IV), PVL, ROP, BPD, length of hospital stay).
Risk of bias in included studies
Neither of the two trials provided information on how the random sequence was generated. Both included studies (Donato 2000; Maier 2002) were randomised double‐blind controlled studies with concealed allocation. Donato et al. (Donato 2000) report that infants were assigned to one of the two groups at birth through a central randomisation process. Placebo and EPO were indistinguishable before and after reconstitution. Parents, investigators, and nurses were unaware of each patient's treatment group. Maier et al. (Maier 2002) concealed the allocation by means of numbered sealed envelopes. To assure blinding and to avoid placebo injections, sham injections were given to the late EPO group prior to starting treatment with EPO. In both studies, outcomes were assessed by individuals unaware of treatment assignment. Sample size calculations were performed in both studies. Six infants with significant protocol violations were excluded from the study by Donato et al (Donato 2000). The group assignment was not reported, but it is likely that the excluded infants were equally distributed between the groups as 57 infants remained in each group. In the study by Maier et al (Maier 2002), no infant was withdrawn from the early EPO and late EPO groups. Follow‐up was complete for the study by Maier (Maier 2002), but not in the study by Donato (Donato 2000). (See Table Characteristics of included studies). Both studies were industry funded (Donato 2000; Maier 2002). We obtained unpublished data from the authors of both studies.
Effects of interventions
Early (zero to seven days) versus late (eight to 28 days) initiation of EPO (Comparison 01):
PRIMARY OUTCOME:
Outcome 1,1: Use of one or more red blood cell transfusions (Figure 1)
Forest plot of comparison: 1 Early (0‐7 days) vs. late (8‐28 days) initiation of EPO, outcome: 1.1 Use of one or more red blood cell transfusions.
Both studies reporting on 262 infants assessed the use of one or more red blood cell transfusions. Neither found a significant effect. The meta‐analysis did not find a significant effect [typical risk ratio (RR) 0.91 (95% confidence interval (CI) 0.78 to 1.06); typical risk difference (RD) ‐0.07 (95% CI ‐0.18 to 0.04)]. This result was consistent across the two studies.There was no heterogeneity for RR or RD (0%).
SECONDARY OUTCOMES:
Outcome 1.2.1 and 1.2.2: The total volume (mL/kg) of red blood cells transfused per infant
Donato et al (Donato 2000) reported on this outcome in 144 infants. They did not find a significant effect [mean difference (MD) 0.30 mL/kg (95% CI ‐13.04 to 13.64)].
We obtained unpublished data from Maier et al. (Maier 2002). They reported on the volume of red blood cells transfused in 148 infants in mL/kg/day. They did not find a significant effect [MD ‐0.80 mL/kg/day (95% CI ‐1.88 to 0.28)]. Test for heterogeneity not applicable.
Outcome 1.3: Number of red blood cell transfusions per infant
Both studies reporting on 262 infants assessed the number of red blood cell transfusions per infant. Neither found a significant effect. The meta‐analysis did not find a significant effect [typical MD ‐0.32, 95% CI ‐0.92 to 0.29)]. This result was consistent across studies. There was no heterogeneity for this outcome (0%).
Outcome 1.4: Number of donors to whom the infant was exposed
Maier (Maier 2002) reported on this outcome in 148 infants. They did not find a significant effect [MD ‐0.20; (95% CI ‐0.67 to 0.27)]. Test for heterogeneity not applicable.
Outcome 1.5: Mortality during initial hospital stay (all causes)
Both studies reporting on 262 infants assessed mortality during initial hospital stay. Neither found a significant effect. The meta‐analysis did not find a significant effect [typical RR 0.76 (95% CI 0.39 to 1.51); RD ‐ 0.03, (95% CI ‐0.11 to 0.05)]. This result was consistent across studies. There was no heterogeneity for RR or RD (0%).
Outcome 1.6: Retinopathy of prematurity (ROP) (all stages) (Figure 2)
Forest plot of comparison: 1 Early (0‐7 days) vs. late (8‐28 days) initiation of EPO, outcome: 1.6 Retinopathy of prematurity (all stages).
We obtained unpublished data from both lead authors of the studies included in this review for this outcome. Both studies assessed ROP in a total of 191 infants. Donato et al reported on the rate of ROP in infants examined during the first year of life. Maier et al reported on the worst stage of ROP during the study. There was a significant increase in the incidence of ROP (all stages) in the study by Donato et al (Donato 2000), but not in the study by Maier et al (Maier 2002). The meta‐analysis found a significant effect [typical RR 1.40 (95% CI 1.05 to 1.86); typical RD 0.16 (95% CI 0.03 to 0.29); number needed to harm (NNTH); 6 (95% CI 3 to 33)].There was high heterogeneity for this outcome (RR P = 0.007; I2 = 86%; RD P = 0.02; I2 = 81%).
Outcome 1.7: Retinopathy of prematurity (stage > 3)
We obtained unpublished data from both lead authors of the studies included in this review for this outcome. Both studies assessed this outcome in a total of 191 infants. Donato et al reported on the rate of ROP in infants examined during the first year of life. Maier et al reported on the worst stage of ROP during the study. Neither found a significant effect. The meta‐analysis did not find a significant effect [typical RR 1.56 (95% CI 0.71 to 3.41); typical RD 0.05 (95% CI ‐0.04 to 0.14). This result was consistent across studies.There was no heterogeneity for RR or RD (0%).
Proven sepsis (Clinical symptoms and signs of sepsis and positive blood culture) (No outcome table) .
No data for this outcome were reported
Outcome 1.8 Necrotising enterocolitis (NEC) (Bell's stage II or more)
Maier (Maier 2002) reported on this outcome in 148 infants. The study did not find a significant effect [RR 1.00 (95% CI 0.37 to 2.71); RD 0.00 (95% CI ‐0.09 to 0.09)].
Outcome 1.9.1: Intraventricular haemorrhage (IVH); grades III and IV
Both studies (n = 262) reported on the incidence of IVH grade III and IV. Neither found a significant effect. The meta‐analysis did not find a significant effect [typical RR 1.33 (95% CI 0.84 to 2.13); typical RD 0.06 (95% CI ‐0.04 to 0.16)]. There was no heterogeneity for RR or RD (0%). The results were consistent across studies.
Outcome 1.10: Periventricular leukomalacia (PVL); cystic changes in the periventricular areas
Maier (Maier 2002) reported on PVL in 148 infants. The study did not find a significant effect for RR [RR 0.09 (95% CI 0.01, 1.62)]; but for RD [RD ‐0.07 (95% CI ‐0.13 to ‐0.01)]; number needed to benefit (NNTB) = 14 (95% CI 8 to 100). Test for heterogeneity not applicable.
Length of hospital stay (No outcome table)
Maier (Maier 2002) reported on the length of hospital stay. The median (quartiles) number of days in hospital was 87 days (73 to 107) in the early EPO group and 90 days (68 to 110) in the late group. There was no statistically significant difference (P = 0.94) across three groups including a control group (87 days; 69 to 108).
Outcome 1.11: Bronchopulmonary dysplasia (BPD) (supplementary oxygen at 28 days of age or at 36 weeks postmenstrual age and compatible X‐ray)
Maier (Maier 2002) reported on the need for oxygen at 36 weeks postmenstrual age in 148 infants. The study did not find a significant effect [RR 0.90 (95% CI 0.53 to 1.54); RD ‐0.03 (95% ‐0.17 to 0.12). Test for heterogeneity not applicable.
Outcome 1.12: Sudden infant death after discharge
Donato (Donato 2000) reported no deaths in either group after a follow‐up period of nine to 24 months after discharge.
Long‐term outcomes assessed at any age beyond one year of age by a validated cognitive, motor, language, or behavioural/school/social interaction/adaptation test (No outcome table).
No data for these outcomes were reported.
Neutropenia (No outcome table)
Donato (Donato 2000) reported that the incidence of neutropenia was similar in the two groups, but did not provide any data. Maier (Maier 2002) reported that neutrophil counts did not differ between groups, but did not provide any data.
Any side effects reported in the trials (No outcome table)
Donato (Donato 2000) stated "No clinical adverse effect attributable to EPO, oral iron, or folic acid administration was observed".
Outcome 1.13: Weight gain during the study period (This outcome was not included in the protocol for this review)
In the study by Donato (Donato 2000), the MD for weight gain during the entire study period was 6.0 g (95% CI ‐137.37 to 149.37) comparing the early EPO group to the late EPO group. In the study by Maier (Maier 2002), the median weight gain was 890 g in the early EPO group and 872 g in the late EPO group during the first nine weeks of life. Quartiles were not provided.
Outcome 1.14: Thrombocytosis (This outcome was not included in the protocol for this review)
Donato (Donato 2000) reported on thrombocytosis (platelet count > 500 x 109/L) in 114 infants. The study did not show a significant effect [RR 1.06 (95% CI 0.61 to 1.84); RD 0.02 (95% CI ‐0.15 to 0.19)].
In the study by Maier (Maier 2002) the median increase in platelet count during the study was 118 x 109/L in the early EPO group, 155x109/L in the late EPO group and 186 x 109/L in the control group. A P value was not reported.
Subgroup analyses
We planned subgroup analyses within this review for low (< 500 IU/kg/week) and high (> 500 IU/kg/week) doses of EPO, and low (< 5 mg/kg/day) and high (> 5 mg/kg/day) doses of supplemental iron administered by any route. However, both included studies used high dose of EPO and high dose of iron, and therefore, subgroup analyses were not performed.
Discussion
The main objective of this review was to assess whether early versus late treatment with EPO would reduce the need for one or more red blood cell transfusions. Two high quality placebo‐controlled, multicentre trials conducted in Argentina and in four European countries were identified for this review. These two studies included a total of 262 preterm infants with very low/extremely low birth weight, who were enrolled at approximately three days of age. The dose of EPO varied from 750 to 1250 IU/kg/week (high dose). All infants received supplemental iron (high dose). Both studies used well defined, although not identical, criteria for red blood cell transfusions (see Additional table; Table 1 Transfusion guidelines). In the study by Maier et al (Maier 2002), more than 30% of the enrolled infants had received red blood transfusions prior to study entry. In the study by Donato et al (Donato 2000), 14% had received red blood cell transfusions prior to enrolment.
Early treatment with EPO did not significantly reduce the risk for an infant of receiving a red blood cell transfusion compared with late treatment. There was no statistically significant heterogeneity for this or any of the other effectiveness outcomes of interest, justifying the combination of the results from the two studies when possible. There was no significant reduction in the number of transfusions per infant nor in the number of donors to whom the infant was exposed.
The results for other secondary outcomes included in this review reached statistical significance only for periventricular leukomalacia (PVL) and retinopathy of prematurity (ROP). PVL was reported in one study (Maier 2002). There was a statistically significant reduction for risk difference (RD) but not for risk ratio (RR) (there was no case of PVL in the early EPO group). We consider this finding of borderline statistical significance.
A total of 191 infants of 268 infants enrolled were assessed for ROP (all stages, and stage > 3). The lack of outcome ascertainment in a large number of infants is of concern. There was a statistically significant increased risk of ROP (any stage reported). The typical RR was 1.45 (95% CI 1.08 to 1.95), the typical RD was 0.18 (95% CI 0.05 to 0.31) and the number needed to harm (NNTH) was six (95% CI 3 to 20). There was statistically significant (high) heterogeneity for this outcome. The heterogeneity is likely at least in part due to differences in the rates of ROP and the different times in the life of the infants when ROP was assessed. There were no striking differences in the dose of EPO or iron used in the two studies. The transfusion guidelines were most strict in the study by Donato et al (Donato 2000). It is possible that there were differences in many aspects of care between Argentina and the four countries enrolling infants in Europe. These differences may have contributed to the differences in rates of ROP. The outcome of ROP (stage > 3) was assessed in the same population as all stages of ROP. The meta‐analysis did not find a significant effect [typical RR 1.56 (95% CI 0.71 to 3.41); typical RD 0.05 (95% CI ‐0.04 to 0.14)]. This result was consistent across studies. There was no heterogeneity for this outcome. The increased risk of ROP is of concern. The results are similar to the Cochrane review of "Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants" (Ohlsson 2006) (updated in 2012). No other important neonatal adverse outcomes or side effects were noted.
The results of this systematic review should be considered in conjunction with our other two Cochrane reviews (both updated in 2012) that have been conducted "Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants" (Ohlsson 2006) and "Late erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants" (Aher 2006a). In our review of "Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants" (Ohlsson 2006), we noted an increased risk of ROP. Based on current evidence, an increased risk of developing ROP with early EPO cannot be excluded. For detailed discussion of the potential association with ROP and EPO, please refer to the Cochrane review of "Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants" (Ohlsson 2006) that was updated in 2012.
Other systematic reviews of EPO do not address the issue of early versus late treatment. The trials included in this review (Donato 2000; Maier 2002) were not included in the meta‐analyses by Vamvakas et al (Vamvakas 2001) and Garcia et al (Garcia 2002). Kotto‐Kome et al (Kotto‐Kome 2004) included both trials in their meta‐analysis of the effect of early EPO on early and late erythrocyte transfusions, but did not compare the effectiveness of early versus late EPO administration. None of these meta‐analyses included analyses on the incidence of ROP.
Early treatment with EPO did not confer any major benefits compared with late EPO treatment. The major obstacle to reduce any donor exposure during the hospital stay is that as many as one third of these infants will require a red blood cell transfusion prior to the initiation of treatment with EPO. In the study by Maier et al (Maier 2002), 12 of the 14 centres used satellite packs of the original red cell pack to reduce donor exposure in both groups. The use of satellite packs and conservative transfusion guidelines reduces the exposure to multiple donors during the hospitalisation for preterm infants.
It is unlikely that long‐acting EPO [Aranesp (Darbepopoietin alfa, Amgen)] would confer any benefits with regards to any donor exposure, but could reduce the number of injections the infant would require (Ohls 2004). To date, only a dose finding trial has been published (Warwood 2005). The authors concluded, based on pharmacodynamic and pharmacokinetic findings, that darbepoetin dosing in neonates would require a higher unit dose/kg and a shorter dosing interval than are generally used for anaemic adults (Warwood 2005).
Administration of EPO could potentially have a neuroprotective effect in preterm infants, especially in infants with perinatal asphyxia (Dame 2001; Juul 2002). This aspect of EPO use will be systematically reviewed separately (Yu 2010).
Forest plot of comparison: 1 Early (0‐7 days) vs. late (8‐28 days) initiation of EPO, outcome: 1.1 Use of one or more red blood cell transfusions.
Forest plot of comparison: 1 Early (0‐7 days) vs. late (8‐28 days) initiation of EPO, outcome: 1.6 Retinopathy of prematurity (all stages).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 1 Use of one or more red blood cell transfusions.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 2 Total volume of red blood cells transfused per infant.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 3 Number of red blood cell transfusions per infant.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 4 Number of donors the infant was exposed to.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 5 Mortality during initial hospital stay (all causes).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 6 Retinopathy of prematurity (all stages).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 7 Retinopathy of prematurity (stage >/= 3).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 8 NEC.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 9 IVH.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 10 PVL.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 11 BPD (oxygen at 36 weeks).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 12 Sudden infant death after discharge.
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 13 Weight gain (grams) during the study period (from entry to exit from study).
Comparison 1 Early (0‐7 days) versus late (8‐28 days) initiation of EPO, Outcome 14 Thrombocytosis (platelet count > 500 x 10 9/L).
| Reference | Indications |
| Indications for transfusion followed slightly modified criteria described in a previous study (Shannon 1995). | |
| Infants on assisted ventilation or > 40% of inspired oxygen were not transfused unless Hct dropped below 0.40. Spontaneously breathing infants were not transfused unless Hct dropped below 0.35 during the first 2 weeks of life, 0.30 during the 3rd to 4th weeks, and 0.25 thereafter. Transfusion was allowed when life threatening anaemia or hypovolaemia was diagnosed by the treating neonatologist, or surgery was planned. Twelve of the 14 centres used satellite packs of the original red cell pack to reduce donor exposure. The amount of packed red cells transfused was not reported. | |
| CPAP = continuous positive airway pressure | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Use of one or more red blood cell transfusions Show forest plot | 2 | 262 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.91 [0.78, 1.06] |
| 2 Total volume of red blood cells transfused per infant Show forest plot | 2 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.1 Total volume of red blood cells transfused (mL/kg) | 1 | 114 | Mean Difference (IV, Fixed, 95% CI) | 0.30 [‐13.04, 13.64] |
| 2.2 Total volume of red cells transfused (mL/kg/day) | 1 | 148 | Mean Difference (IV, Fixed, 95% CI) | ‐0.80 [‐1.88, 0.28] |
| 3 Number of red blood cell transfusions per infant Show forest plot | 2 | 262 | Mean Difference (IV, Fixed, 95% CI) | ‐0.32 [‐0.92, 0.29] |
| 4 Number of donors the infant was exposed to Show forest plot | 1 | 148 | Mean Difference (IV, Fixed, 95% CI) | ‐0.20 [‐0.67, 0.27] |
| 5 Mortality during initial hospital stay (all causes) Show forest plot | 2 | 262 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.76 [0.39, 1.51] |
| 6 Retinopathy of prematurity (all stages) Show forest plot | 2 | 191 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.40 [1.05, 1.86] |
| 7 Retinopathy of prematurity (stage >/= 3) Show forest plot | 2 | 191 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.56 [0.71, 3.41] |
| 8 NEC Show forest plot | 1 | 148 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.0 [0.37, 2.71] |
| 9 IVH Show forest plot | 2 | 262 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.33 [0.84, 2.13] |
| 9.1 Grade III/IV | 2 | 262 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.33 [0.84, 2.13] |
| 10 PVL Show forest plot | 1 | 148 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.09 [0.01, 1.62] |
| 11 BPD (oxygen at 36 weeks) Show forest plot | 1 | 148 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.90 [0.53, 1.54] |
| 12 Sudden infant death after discharge Show forest plot | 1 | 114 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 13 Weight gain (grams) during the study period (from entry to exit from study) Show forest plot | 1 | 114 | Mean Difference (IV, Fixed, 95% CI) | 6.00 [‐137.37, 149.37] |
| 14 Thrombocytosis (platelet count > 500 x 10 9/L) Show forest plot | 1 | 114 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.06 [0.61, 1.84] |
