NoBlood

Strict red blood cell transfusion guideline reduces the need for transfusions in very-low-birthweigh

Abstract

Strict red blood cell transfusion guideline reduces the need for transfusions in very-low-birthweigh

Publication: Vox Sang 2005;88:107-13.
Monday, February 28th, 2005

Author: Miyashiro AM, Dos Santos N, Guinsburg R, Kopelman BI, de Araujo Peres C, de Lima Taga MF, Shinzato A

Article
Miyashiro AM, Dos Santos N, Guinsburg R, Kopelman BI, de Araujo Peres C, de Lima Taga MF, Shinzato AR, de Paula Fiod Costa H. Strict red blood cell transfusion guideline reduces the need for transfusions in very-low-birthweight infants in the first 4 weeks of life: a multicentre trial. Vox Sang 2005;88:107-13.

Design
Multicenter, prospective cohort trial performed in two phases (“before-and-after� study). Number of patients: 345. Study period: July 1, 2000 – December 15, 2000 (Phase 1) and January 16, 2001 – October 15, 2001 (Phase 2). Setting: Seven public hospital neonatal intensive care units (NICUs) in São Paulo, Brazil.

Overview

Background
Very-low-birthweight (VLBW) infants are among the most heavily transfused patients, especially during the first four weeks of life. Concern over the adverse effects of red blood cell (RBC) transfusions has led to the investigation of various strategies aimed to reduce allogeneic blood exposure, including the decrease of blood removed for laboratory tests, the use of recombinant human erythropoietin (rHuEPO), and the adoption of strict transfusion criteria. Limiting phlebotomy losses is the most effective of these methods, but modern laboratory microtechniques allowing laboratory examinations to be completed with negligible amounts of blood are expensive. rHuEPO administration, besides its high cost, has not had the expected impact on transfusion requirements. In developed countries, the adoption of strict transfusion criteria has been shown to reduce transfusion requirements. However, this method has not been assessed in NICUs units that have had difficulties in controlling blood loss. In this multicenter, prospective cohort study, the authors investigated the effectiveness of the use of a guideline for packed RBC transfusions in reducing transfusions in VLBW infants during their first 28 days of life.

Methods
Neonates with a gestational age of < 37 weeks and a birthweight lower than 1500 g born in seven public hospital NICUs of the city of São Paulo, Brazil, were included in the study. Patients were excluded from the study if they had major congenital anomalies or hemolytic disease of any etiology, underwent total or partial exchange transfusion performed during the hospital stay, or died within the first 24 h after birth.

The study performed in two phases. In Phase 1 (July 1, 2000 – December 15, 2000), the indications for transfusions were heterogeneous and at the discretion of the attending neonatologist, whereas in Phase 2 (January 16, 2001 – October 15, 2002), a written guideline for RBC transfusions adapted from a published guideline (Shannon KM et al., Pediatrics 1995;95:1-8) was posted in the participating NICUs and all attending neonatologists requested to follow the new criteria.

In the two study periods, information related to RBC transfusions during the hospital stay as well as data regarding the demographics and clinical history of the neonates were prospectively collected. Transfusion data included the hematocrit (Hct) and the hemoglobin (Hb) concentration within the first 12 hours of life, at 28 days of life, at 40 weeks of corrected gestational age, and before administration of any RBC transfusion; the volume of blood drawn for laboratory testing within the first 28 days of life; the percentage of neonates receiving at least one RBC transfusion; the number of transfusions and the volume of transfusions during the first 4 weeks of life; and the percentage of transfusions performed on the basis of the indication criteria adopted in Phase 2.

Results
A total of 149 VLBW premature neonates were included in Phase 1 and 196 in Phase 2, corresponding to 97.5% of the population meeting inclusion criteria in the seven neonatal units during the study period. Each participating NICU maintained the same ratio of neonates included in Phases 1 and 2 (p = 0.656). Demographics and clinical problems were similar in both phases. Mean weight at discharge was also similar in both groups.

Blood losses during the first week of life were slightly higher in Phase 1 (12.8 ± 7.1 mL/kg) than in Phase 2 (11.3 ± 5.8 mL/kg) (p = 0.049), whereas there was no significant difference in the subsequent 3 weeks. During the first 28 days of life, the blood volume withdrawn for laboratory tests was similar in both groups.

Mean Hct at birth was similar in both groups, but mean Hct at 28 days of life was higher in Phase 1 than in Phase 2 (31.0 ± 5.6% vs. 29.5 ± 5.5%; p = 0.033), and the difference was maintained at 40 weeks of corrected gestational age (30.1 ± 4.6% vs. 28.2 ± 5.3%; p = 0.049). Reticulocyte levels of neonates still hospitalized at 28 days of age were significantly higher in Phase 2 (2.4 ± 2.5% vs. 1.7 ± 1.7%; p = 0•038).

Pretransfusion Hct values were similar in both periods (29.3 ± 5.1% in Phase 1 vs. 28.5 ± 4.4% in Phase 2; p = 0.066). During Phase 2, 88% of the 274 transfusions administered complied with the guideline, whereas of the 282 transfusions administered in Phase 1, 179 (63.5%) would have been in accordance with the guideline. In Phase 2, no neonate was transfused with a Hct > 40%, compared with 6 in Phase 1.

In Phase 1, 102 of 149 neonates (68.5%) received 282 RBC transfusions, compared with 274 transfusions in 117 of 196 (59.7%) infants in Phase 2, corresponding to a 12.8% decrease (p < 0.04 [not provided by the authors]) in the percentage of transfused patients after the guideline was adopted. There was a significant difference between the two periods in the average number of RBC transfusions per patient (1.9 ± 2.0 in Phase 1 and 1.4 ± 1.6 in Phase 2; p = 0.014).

The volume of red blood cells administered during each transfusion episode ranged from 9 to 15 mL/kg, according to the attending physician’s decision. In Phase 2, a significant decrease in the volume of RBCs transfused per neonate was observed compared to Phase 1 (17.7 ± 20.9 mL/kg vs. 24.1 ± 24.3 mL/kg; p = 0.013); this difference was explained by the reduction in number of transfusions.

Multivariate analysis showed that the number of RBC transfusions was significantly associated with the following variables: phase of the study, gestational age, presence of respiratory distress syndrome, and volume of blood loss during the first 28 days of life. Infants enrolled in Phase 2 received 0.3344 (95% confidence interval, -0.5795 to -0.0437) fewer transfusions than those born in Phase 1.

Conclusion
The authors conclude that the adoption of a strict guideline for RBC transfusions is effective in reducing transfusion requirements in VLBW infants without threatening their clinical course during the hospital stay.

Key Points
• Very-low-birthweight infants are among the most heavily transfused patients, especially during the first four weeks of life.
• Concern over the adverse effects of red blood cell transfusions has led to the investigation of various strategies aimed to reduce allogeneic blood exposure, including the decrease of blood removed for laboratory tests, the use of recombinant human erythropoietin, and the adoption of strict transfusion criteria.
• In this multicenter, prospective cohort study of 345 very-low-birthweight infants performed in two phases, the authors found that the adoption of a published transfusion guideline (Shannon KM et al., Pediatrics 1995;95:1-8) significantly reduced the number of red blood cell transfusion administered and the proportion of patients transfused.
• During Phase 2, 88% of the transfusions administered complied with the published guideline, whereas 63.7% of the transfusions administered in Phase 1 would have been in accordance with the guideline. In Phase 2, no neonate was transfused with a Hct > 40%, compared with 6 in Phase 1.


Limitations
• Nonrandomized study.
• Although the authors state that the restrictive transfusion policy reduced transfusion requirements without threatening the patients’ clinical course, the study was not powered to assess mortality and morbidity.


SABM Rating:***

In Brief
In this multicenter, prospective cohort study performed in two phases, the authors evaluated the effectiveness of the use of a published transfusion guideline (Shannon KM et al., Pediatrics 1995;95:1-8), compared to a situation where the indications for transfusions were heterogeneous and at the discretion of the attending neonatologist, in reducing allogeneic blood exposure in infants with a gestational age of < 37 weeks and a birthweight lower than 1500 g in the first 28 days of life. The number of red blood cell transfusions administered per patient was reduced by 0.5 (1.9 ± 2.0 in Phase 1 and 1.4 ± 1.6 in Phase 2; p = 0.014) and the proportion of patients transfused decreased from 68.5% to 59.7% after implementation of the guideline, corresponding to a 12.8% decrease in the percentage of transfused patients. Transfusions administered to patients with a hematocrit greater than 40% were eliminated. Reticulocyte counts were significantly higher in Phase 2, suggesting that a restrictive transfusion policy stimulates erythropoiesis. This study confirms that the adoption of evidence-based transfusion guidelines is an effective and inexpensive method to reduce inappropriate transfusions and the associated risks.

__________________
Mr. Jan B. Wade
Blood Management Consultant
Enhance Outcomes - Control Cost
For Information Call - 360 296-1807
Email