Presented at the Neonatal Society 2002 Autumn Meeting.

Sen S, Ferguson SD (introduced by Cora Doherty)

Neonatal Intensive Care Unit, Royal Gwent Hospital, Newport, S Wales NP20 2UB

Introduction: Very low birthweight (VLBW) babies often receive multiple transfusions during their stay in hospital. Red blood cell transfusion (RBCT) policies have evolved over the last decade primarily because of the erythropoietin studies in which strict and restrictive criteria were used. Many units worldwide have adopted these and similar criteria and have not observed any major short or medium term problems.

Aims: Since there are no uniform criteria, the aim of this survey was to get a baseline data on the practice of RBCT in VLBW babies in Wales, to look at donor exposure rates and to compare them with published studies. The secondary aim was to determine whether RBCTs could be reduced had restrictive criteria been applied.

Methods: Thirteen of 14 SCBU of Wales participated in the survey. All babies born in 2001 with birth weight 1500 g or less and had received RBCT were identified from hospital records. The following data was extracted: (a) basic demographic details; (b) number and volume of each RBCT; (c) age, weight, ventilation and oxygen requirements at each transfusion; (d) reason for transfusion; (f) pre and post RBCT haemoglobin, hematocrit (HCT) and reticulocyte count and (e) number of donors exposed to. Means and standard deviations and Median and interquartile ranges were calculated. The results were compared to published data (1,2). Continuous values were compared with the Student’s t test and proportions by the z-test.

Results: 261 VLBW babies were identified of whom 154 (59%) received RBCT. 132/154 (85%) notes were retrieved of which 16 babies were excluded because of early death. The remaining 116 babies received 531 RBCTs. 36 RBCTs were excluded because they were given for acute blood loss, exchange transfusions or for surgery. The remaining 495 RBCTs analysed. The mean birth weight was 970 g (SD 233) and the mean gestation was 27.2 weeks (SD 2.2). 50% of all the RBCTs were given in the first 2 weeks of life 95% of all were within 55 days. The mean number of RBCTs given was 4.3/ baby {SD 3.3, P<0.001 compared to (1,2)} and the smaller babies received the most RBCTs. 35% babies received 5 or more RBCT. The mean volume of red cells given was 64.3 ml/baby {SD 44, P<0.001 compared to (1), P<0.01 compared to (2)}. The mean pre transfusion HCT was 0.35 (SD 0.05). Babies on higher ventilation support were transfused at higher HCT. The donor exposure rate was 2.9 (SD 2) per baby compared to 2.2 by Widness et al (2). 28% of the RBCTs were of 10ml/kg or less and of these, 40% of resulted in a post-transfusion HCT rise of less than 10%.

Conclusions: This survey shows that the mean number of RBCTs, the volume of blood/baby, pre-transfusion HCT and the donor exposure rates were significantly higher when compared to recently published studies on similar babies. The pre-transfusion HCT was consistently higher in “growing” babies than those recommended (3). If international protocols were followed, 15-22% of all RBCTs could have been avoided. Following this survey a plan is being formulated to reduce the number and volume of RBCT and donor exposure in VLBW infants in Wales.

References
1. Beeram MR, Kuauss DR, Riggs MW. Red blood cell transfusion practices in very low birth weight infants in the 1990s postsurfactant era. J Natl Med Assoc 2001, 93:405-409
2. Widness JA, Seward VJ, Kromer IJ et al. Changing patterns of red blood cell transfusion in very low birth weight infants. J Pediatr 1996, 129:680-687
3. Shannon KM, Keith JF, Mentzer WC et al. Recombinant human erythropoietin stimulates erythropoiesis and reduces erythrocyte transfusions in very low birth weight babies. Pediatrics 1995, 95:1-8.