Transfusion of red blood cells stored for shorter versus longer duration for all conditions

Abstract

Background: Red blood cell (RBC) transfusion is a common treatment for anaemia in many conditions. The safety and efficacy of transfusing RBC units that have been stored for different durations before a transfusion is a current concern. The duration of storage for a RBC unit can be up to 42 days. If evidence from randomised controlled trials (RCT) were to indicate that clinical outcomes are affected by storage duration, the implications for inventory management and clinical practice would be significant. Objectives: To assess the effects of using red blood cells (RBCs) stored for a shorter versus a longer duration, or versus RBCs stored for standard practice duration, in people requiring a RBC transfusion. Search methods: We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, CINAHL, PubMed (for epublications), LILACS, Transfusion Evidence Library, Web of Science CPCI-S and four international clinical trial registries on 20 November 2017. Selection criteria: We included RCTs that compared transfusion of RBCs of shorter versus longer storage duration, or versus standard practice storage duration. Data collection and analysis: We used standard Cochrane methods. Main results: We included 22 trials (42,835 participants) in this review. The GRADE quality of evidence ranged from very low to moderate for our primary outcome of in-hospital and short-term mortality reported at different time points. Transfusion of RBCs of shorter versus longer storage duration Eleven trials (2249 participants) compared transfusion of RBCs of shorter versus longer storage duration. Two trials enrolled low birth weight neonates, two enrolled children with severe anaemia secondary to malaria or sickle cell disease, and eight enrolled adults across a range of clinical settings (intensive care, cardiac surgery, major elective surgery, hospitalised in-patients, haematology outpatients). We judged only two trials to be at low risk of bias across all domains; most trials had an unclear risk for multiple domains. Transfusion of RBCs of shorter versus longer storage duration probably leads to little or no difference in mortality at seven-day follow-up (risk ratio (RR) 1.42, 95% confidence interval (CI) 0.66 to 3.06; 1 trial, 3098 participants; moderate quality evidence) or 30-day follow-up (RR 0.85, 95%CI 0.50 to 1.45; 2 trials, 1121 participants; moderate quality evidence) in adults undergoing major elective cardiac or non-cardiac surgery. For neonates, no studies reported on the primary outcome of in-hospital or short-term mortality. At 40 weeks gestational age, the effect of RBCs of shorter versus longer storage duration on the risk of death was uncertain, as the quality of evidence is very low (RR 0.90, 95% CI 0.41 to 1.85; 1 trial, 52 participants). The effect of RBCs of shorter versus longer storage duration on the risk of death in children with severe anaemia was also uncertain within 24 hours of transfusion (RR 1.50, 95% CI 0.43 to 5.25; 2 trials, 364 participants; very low quality evidence), or at 30-day follow-up (RR 1.40, 95% CI 0.45 to 4.31; 1 trial, 290 participants; low quality evidence). Only one trial, in children with severe anaemia (290 participants), reported adverse transfusion reactions. Only one child in each arm experienced an adverse reaction within 24 hours of transfusion. Transfusion of RBCs of shorter versus standard practice storage duration Eleven trials (40,588 participants) compared transfusion of RBCs of shorter versus standard practice storage duration. Three trials enrolled critically ill term neonates; two of these enrolled very low birth weight neonates. There were no trials in children. Eight trials enrolled critically ill and non-critically ill adults, with most being hospitalised. We judged four trials to be at low risk of bias across all domains with the others having an unclear risk of bias across multiple domains. Transfusion of RBCs of shorter versus standard practice storage duration probably leads to little or no difference in adult in-hospital mortality (RR 1.05, 95% CI 0.97 to 1.14; 4 trials, 25,704 participants; moderate quality evidence), ICU mortality (RR 1.06, 95% CI 0.98 to 1.15; 3 trials, 13,066 participants; moderate quality evidence), or 30-day mortality (RR 1.04, 95% CI 0.96 to 1.13; 4 trials, 7510 participants;moderate quality evidence). Two of the three trials that enrolled neonates reported that there were no adverse transfusion reactions. One trial reported an isolated case of cytomegalovirus infection in participants assigned to the standard practice storage duration group. Two trials in critically ill adults reported data on transfusion reactions: one observed no difference in acute transfusion reactions between arms (RR 0.67, 95% CI 0.19 to 2.36, 2413 participants), but the other observed more febrile nonhaemolytic reactions in the shorter storage duration arm (RR 1.48, 95% CI 1.13 to 1.95, 4919 participants). Trial sequential analysis showed that we may now have sufficient evidence to reject a 5% relative risk increase or decrease of death within 30 days when transfusing RBCs of shorter versus longer storage duration across all patient groups. Authors' conclusions: The effect of storage duration on clinically important outcomes has now been investigated in large, high quality RCTs, predominantly in adults. There appears to be no evidence of an effect on mortality that is related to length of storage of transfused RBCs. However, the quality of evidence in neonates and children is low. The current practice in blood banks of using the oldest available RBCs can be continued safely. Additional RCTs are not required, but research using alternative study designs, should focus on particular subgroups (e.g. those requiring multiple RBC units) and on factors affecting RBC quality.