Kenyatta University and University of Pittsburgh Atlanta, Georgia, United States
Background/Case Studies: In Sub-Saharan Africa, blood products are frequently unavailable when clinically needed and are rarely stored in advance for anticipated urgent need. WHO estimates that 1-2% of a population represents an estimated annual blood unit need on a national level. But at the local level, data is lacking that reflects context-specific clinical need. Blood transfusion data can guide proactive preparation of blood units and decrease response time to urgent clinical situations. Therefore, blood transfusion practice patterns were studied in a Kenyan health care setting. This study will serve as a first step to determine facility-specific blood product demand with an aim to reduce the mismatch between immediate need and lack of timely availability.
Study
Design/Methods: From Sept 1/21-Feb 28/22, patients admitted to a Level-V referral hospital in Kenya, were reviewed for possible blood product transfusion requirement. Patients were included if they displayed a clinical indication for blood transfusion: diagnosis of hemorrhage, hemorrhage-related diagnosis(GI bleed/trauma with signs of hemorrhage)or an admission hemoglobin of < 10 gm/dl. Data was entered into RedCap and analyzed using Stata 8.1. The study received ethical approval from the Hospital Ethics Committee.
Results/Findings: There were 4,753 patients admitted during the study period. The median age at admission was 38 years, ranging from 1 month to 90 years, 40% male. There were 528 blood products dispatched in the study period, with an average of 106 units per month (range:50-136). The average number of units transfused per patient was 2(range:1-5 units).Of the 4,753 patients admitted, 541 patients (11.4%) met inclusion criteria of a potential indication for blood transfusion. The study cohort had a mortality rate of 5%(29/541). There were 140 patients with an admission diagnosis of hemorrhage(25.9%, 140/541). There were 401 patients who had anemia (Hgb < 10 gm/dl) without a source of hemorrhage. Only 84 patients(84/401,21%) had a hemoglobin of < 7 gm/dl on admission. Of the total study cohort, 227 had a transfusion ordered(227/541,42%), and of these, 184 received at least one unit of blood(184/541, 34%). 43 patients(43/227, 19%) did not receive their ordered transfusion before discharge and 4 of these patients died(4/227, 2%). Conclusions: A quarter of the patients receiving transfusion, required an urgent transfusion for hemorrhage. About half of the patients had a transfusion ordered and one-third received one or more units. About 100 units are transfused monthly and about a third are required urgently. Blood delivery gaps exist between ordering and administering blood products and future studies need to focus on determining and evaluating interventions. Local hospital data can be used to determine realistic blood storage targets and calculation of urgent blood demand, to improve timely blood delivery and ultimately reduce mortality.
Importance of research: Low resource settings need local data to understand their context-specific blood product transfusion patterns and improve their ability to plan and prepare blood products as opposed to the reactionary approach most common presently. This data can be a step toward developing par levels for blood storage in anticipation of urgent blood needs, improving timely administration. This data shows blood delivery system gaps that need addressing to reduce mortality.
