Article type
Year
Abstract
Background: Haematocrit falls after birth in preterm infants due to physiological factors and frequent blood letting. Low plasma levels of erythropoietin (EPO) in preterm infants provide a rationale for the use of EPO to prevent/treat anemia. The main goal of EPO treatment is to avoid exposure to donor blood.
Objective: To assess the effectiveness/safety of late-EPO administration (mean postnatal age of study population >14 days) to prevent preterm infants from exposure to donor blood.
Methods: Randomized controlled trials comparing late-EPO to placebo/no intervention were identified by searching personal files, MEDLINE, CINAHL, the Cochrane Trial Registry and the reference lists of identified trials in November 2003. No language restrictions were applied. The primary outcome of interest was need for one RBC transfusion (i.e. exposure to one donor). Other outcomes included; mortality, sepsis, necrotizing enterocolitis (NEC), hypertension, and adverse events reported by the authors. Meta-analyses were performed using RevMan 4.2 and reported as relative risk (RR), risk difference (RD), and if statistically significant, number needed to treat (NNT) with their 95% confidence intervals (CI). Heterogeneity and inconsistency (I2) were calculated.
Results: Twelve studies (sample size ranging from 8 to 157) from 8 countries were identified with a total of 569 infants enrolled. Iron and vitamins were supplemented to all infants. The dosing regimen for EPO varied, as did the criteria for RBC transfusion. The risk for exposure to one RBC transfusion (one donor) was reduced with EPO (8 studies, total n=377) RR 0.58 (0.46, 0.72). The test for heterogeneity was statistically significant [(p= 0.01); I2 60%]. RD was -0.25 (95% CI -0.34, -0.16) with no significant heterogeneity [(p=0.19), I230.4%]; NNT 4 (95% CI 3-6). There were no significant differences in the RR for mortality, sepsis, NEC or hypertension. No other serious adverse effects were reported. Conclusions: Administration of late-EPO reduced the risk for exposure to donor blood in preterm infants. Only 4 infants need to be treated with EPO to avoid one infant being exposed to RBC transfusion. The inference for clinical practice is weakened by the fact that most studies did not use strict RBC transfusion criteria and there was statistically significant heterogeneity for some outcomes.
Objective: To assess the effectiveness/safety of late-EPO administration (mean postnatal age of study population >14 days) to prevent preterm infants from exposure to donor blood.
Methods: Randomized controlled trials comparing late-EPO to placebo/no intervention were identified by searching personal files, MEDLINE, CINAHL, the Cochrane Trial Registry and the reference lists of identified trials in November 2003. No language restrictions were applied. The primary outcome of interest was need for one RBC transfusion (i.e. exposure to one donor). Other outcomes included; mortality, sepsis, necrotizing enterocolitis (NEC), hypertension, and adverse events reported by the authors. Meta-analyses were performed using RevMan 4.2 and reported as relative risk (RR), risk difference (RD), and if statistically significant, number needed to treat (NNT) with their 95% confidence intervals (CI). Heterogeneity and inconsistency (I2) were calculated.
Results: Twelve studies (sample size ranging from 8 to 157) from 8 countries were identified with a total of 569 infants enrolled. Iron and vitamins were supplemented to all infants. The dosing regimen for EPO varied, as did the criteria for RBC transfusion. The risk for exposure to one RBC transfusion (one donor) was reduced with EPO (8 studies, total n=377) RR 0.58 (0.46, 0.72). The test for heterogeneity was statistically significant [(p= 0.01); I2 60%]. RD was -0.25 (95% CI -0.34, -0.16) with no significant heterogeneity [(p=0.19), I230.4%]; NNT 4 (95% CI 3-6). There were no significant differences in the RR for mortality, sepsis, NEC or hypertension. No other serious adverse effects were reported. Conclusions: Administration of late-EPO reduced the risk for exposure to donor blood in preterm infants. Only 4 infants need to be treated with EPO to avoid one infant being exposed to RBC transfusion. The inference for clinical practice is weakened by the fact that most studies did not use strict RBC transfusion criteria and there was statistically significant heterogeneity for some outcomes.