Open Access Highly Accessed Research article

Phase II trial of standard versus increased transfusion volume in Ugandan children with acute severe anemia

Peter Olupot-Olupot1, Charles Engoru2, Jennifer Thompson3, Julius Nteziyaremye3, Martin Chebet1, Tonny Ssenyondo1, Cornelius M Dambisya1, Vicent Okuuny2, Ronald Wokulira2, Denis Amorut2, Paul Ongodia1, Ayub Mpoya4, Thomas N Williams45, Sophie Uyoga4, Alex Macharia4, Diana M Gibb3, A Sarah Walker3 and Kathryn Maitland45*

Author Affiliations

1 Department of Paediatrics, Mbale Regional Referral Hospital, Pallisa Road Zone, PO Box 921, Mbale, Uganda

2 Department of Paediatrics, Soroti Regional Referral Hospital, PO Box 289, Soroti, Uganda

3 Medical Research Council (MRC) Clinical Trials Unit, Aviation House, 125 Kingsway, London WC2B 6NH, UK

4 Kilifi Clinical Trials Facility, KEMRI-Wellcome Trust Research Programme, PO Box 230, Kilifi, Kenya

5 Wellcome Trust Centre for Clinical Tropical Medicine, Department of Paediatrics, Faculty of Medicine, St Marys Campus, Norfolk Place, Imperial College, London W2 1PG, UK

For all author emails, please log on.

BMC Medicine 2014, 12:67  doi:10.1186/1741-7015-12-67

Published: 25 April 2014



Severe anemia (SA, hemoglobin <6 g/dl) is a leading cause of pediatric hospital admission in Africa, with significant in-hospital mortality. The underlying etiology is often infectious, but specific pathogens are rarely identified. Guidelines developed to encourage rational blood use recommend a standard volume of whole blood (20 ml/kg) for transfusion, but this is commonly associated with a frequent need for repeat transfusion and poor outcome. Evidence is lacking on what hemoglobin threshold criteria for intervention and volume are associated with the optimal survival outcomes.


We evaluated the safety and efficacy of a higher volume of whole blood (30 ml/kg; Tx30: n = 78) against the standard volume (20 ml/kg; Tx20: n = 82) in Ugandan children (median age 36 months (interquartile range (IQR) 13 to 53)) for 24-hour anemia correction (hemoglobin >6 g/dl: primary outcome) and 28-day survival.


Median admission hemoglobin was 4.2 g/dl (IQR 3.1 to 4.9). Initial volume received followed the randomization strategy in 155 (97%) patients. By 24-hours, 70 (90%) children in the Tx30 arm had corrected SA compared to 61 (74%) in the Tx20 arm; cause-specific hazard ratio = 1.54 (95% confidence interval 1.09 to 2.18, P = 0.01). From admission to day 28 there was a greater hemoglobin increase from enrollment in Tx30 (global P <0.0001). Serious adverse events included one non-fatal allergic reaction and one death in the Tx30 arm. There were six deaths in the Tx20 arm (P = 0.12); three deaths were adjudicated as possibly related to transfusion, but none secondary to volume overload.


A higher initial transfusion volume prescribed at hospital admission was safe and resulted in an accelerated hematological recovery in Ugandan children with SA. Future testing in a large, pragmatic clinical trial to establish the effect on short and longer-term survival is warranted.

Please see related commentary article webcite.

Trial registration

ClinicalTrials.Gov identifier: NCT01461590 registered 26 October 2011.

Transfusion; Severe anemia; African children; Clinical trial; Infectious disease