Abstract 1554

The hospital course of preterm infants is often complicated by anemia requiring blood transfusion. Advances in blood banking techniques have shown some promise in reducing blood transfusion and donor number. Objective: study the use of older red blood cells (RBC) in sick newborn infants. Methods: all infants requiring RBC transfusion during a 6 month period after change in blood preservative to Nutricell (AS-3), with a 42 day shelf life (n=47), were compared to those transfused in the previous 6 month period with CPDA-1 preserved RBC, with a shelf-life of less than 7 days (n=63). Each infant receiving AS-3 RBC received blood from the same designated unit until the 42 day expiration. Group demographics were compared by Chi-square or Student's test. Donor and transfusion numbers were compared by Mann-Whitney U test. Forward stepwise logistic regression was also used to construct a predictive model of donor number. All data are expressed as mean ± SD, a p value <.05 was considered significant. Results: infants receiving AS-3 had a reduction in donor number/hospital course (1.4 ± 0.7 vs 3.3 ± 2.6, p<.01) and an increase in the ratio of transfusion/donor number (4.2 ± 2.9 vs. 1.5 ± 0.8, p<.01) compared to the CPDA-1 group. Similarly, when the data were analyzed using only infants <33 wks gestation, infants receiving AS-3 (n=39) had a reduction in blood donor number/hospital course (1.5 ± 0.8 vs 3.5 ± 2.6, p<.01) compared to those receiving CPDA-1 (n=57). Of the infants in the CPDA-1 group, 51% received RBC from multiple blood donors (>1) compared to 28% in the AS-3 group (45% reduction, p=.02). There was no difference in number of RBC transfusions/hospital course between infants receiving CPDA-1 RBC (4.8 ± 4.1) and AS-3 RBC (6.0 ± 6.1, p=.24). There was no statistical difference between groups in birth weight, gestational age, mortality, race, need for mechanical ventilation or incidence of sepsis, BPD, NEC, PDA between the groups. However, infants in the AS-3 group had a shorter average length of hospital stay (42.7 ± 2.9 vs 59.3 ± 33 days, p=.007) compared to those in the CPDA-1 group. Using multivariate modeling, length of hospital stay (15% variability, p<.01) was the variable most associated with donor number. The blood preservative still accounted for 14% (p<.01) of the variability in donor number (overall model R2=.76, p<.01). Conclusions: AS-3 preserved RBC, with a shelf life of 42 days, leads to a significant reduction in blood donor number and a 45% reduction in infants receiving RBC from multiple donors. We speculate that further reduction in blood donor number may be achieved by targeting infants with prolonged hospital stay for additional interventions.