Abstract
Background and Objectives
Preclinical studies generated the hypothesis that older stored red blood cells (RBCs) can increase transfusion risks. To examine the most updated and complete clinical evidence and compare results between two trial designs, we assessed both observational studies and randomized controlled trials (RCTs) studying the effect of RBC storage age on mortality.
Materials and Methods
Five databases were searched through December 2014 for studies comparing mortality using transfused RBCs having longer and shorter storage times.
Results
Analysis of six RCTs found no significant differences in survival comparing current practice (average storage age of 2 to 3 weeks) to transfusion of 1- to 10-day-old RBCs (OR 0·91, 95% CI 0·77–1·07). RBC storage age was lower in RCTs vs. observational studies (P = 0·01). The 31 observational studies found an increased risk of death (OR 1·13, 95% CI 1·03–1·24) (P = 0·01) with increasing age of RBCs, a different mortality effect than RCTs (P = 0·02).
Conclusion
RCTs established that transfusion of 1- to 10-day-old stored RBCs is not superior to current practice. The apparent discrepancy in mortality between analyses of RCTs and observational studies may in part relate to differences in hypotheses tested and ages of stored RBCs studied. Further trials investigating 1-to 10-day-old stored RBC benefits would seem of lower priority than studies to determine whether 4- to 6-week stored units have safety and efficacy equivalent to the 2- to 3-week-old stored RBCs commonly transfused today.
Keywords: blood safety, clinical trial, quality control, red cell components, transfusion medicine (in general), transfusion therapy
Introduction
Red blood cells (RBCs) stored for transfusion progressively undergo a series of metabolic and rheologic changes referred to as the ‘storage lesion’ [1]. Although some loss of efficacy likely occurs at the end of the 5- to 7-week storage period (depending on the country) [2], a decrease in safety is not expected. However, laboratory [1], preclinical animal [3] and clinical studies [4] suggest that transfusion of RBCs at the end of the storage period may increase morbidity and/or mortality. There is no standard definition for either ‘old’ or ‘fresh’ red cells, and the terms relate to arbitrary short and long refrigerated storage intervals.
In a prior analysis of primarily observational studies, transfusion of older, conventionally issued RBCs was reported to increase mortality [4]. The emphasis on observational studies was necessary because of the paucity of randomized controlled trials (RCTs), accounting only for 126 patients. Since the publication of this analysis in 2012, multiple large RCTs have been completed and published. These RCTs compared current practice, where transfused RBCs are stored on average for 2 to 3 weeks, to very fresh RBCs, transfused within the first 10 days after donation. The number of observational studies evaluating the effect of RBC storage on outcomes has also surged during this time. Observational studies compared two age groups of stored RBCs commonly found within current practice: the shortest to the longest transfused conventionally issued RBCs within usual care. These two types of studies addressed slightly different questions, respectively: whether 1- to 10-day-old stored RBCs should be transfused instead of current practice, and whether the acceptable RBC storage time should be decreased by excluding the oldest units currently licensed for transfusion. Therefore, to assess the most updated clinical evidence examining the effect of storage age of RBCs on mortality, we have reviewed the available data from recent RCTs and observational studies and compared results between the two trial designs.
Materials and methods
Data sources and study selection
Five databases (PubMed, EMBASE, Web of Science, Scopus and Cochrane library) were searched from May 2011 to 15 December 2014 for publications in English comparing mortality in patients transfused with shorter and longer stored RBCs (supplemental methods). Abstracts from transfusion medicine conferences and published systematic reviews were also reviewed to identify studies. Studies were excluded if they did not report mortality data or document the storage age of transfused RBCs. For clinical studies of storage age with outcomes other than mortality, the authors were queried about mortality rates and, if provided, included in the meta-analysis [5, 6].
Our literature search yielded a total of 565 new publications since May 2011. After excluding those articles unrelated to RBC storage and those not in English (n = 509), 39 articles and one conference abstract were evaluated in detail. Studies that did not report mortality or storage age of transfused RBCs adequately were further excluded. Fifteen studies met inclusion criteria (Fig. 1, Tables 1, 2) and were added to the 21 trials from our previous meta-analysis. After completion of our search, another multicentre RCT was published [7] that enrolled more than double the sum of the patients in the previous five RCTs and was therefore included in a number of confirmatory analyses. RCTs and observational studies were analysed separately given their differences in study design, and hypothesis tested.
Fig. 1.
Study selection.
Table 1.
Characteristics of randomized controlled trials included in the meta-analysis
| Age of stored blood compareda |
Actual median age of stored blood |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Author [Reference) |
Years of Enrollment |
Study Type |
Study Population |
No. of Patients in study |
No. of Patients used in Meta- |
Mean volume of RBCs transfused (units of PRBC) |
New blood (days) |
Old blood (days) |
New blood (days) |
Old blood (days) |
Blood storage solutionb |
Leucoreduced | Mortality | Risk of bias |
| Schulman 06/02 | 2000–2001 | RCT | Trauma | 17 | 17 | New = 9·3 (1·9), Old = 10·6 (3·35)e |
<11 | >20 | NR | NR | NR | Yes | Hospital mortality |
Higher |
| Hebert 05/05 | 1999–2001 | RCT | Cardiac surgery |
57 | 57 | New = 3 (2–5), Old = 2 (2–4)c |
<8 | 15–42 | 4d | 19d | CPD-2, AS-3 | Yes | Hospital mortality |
Lower |
| Fernandes da Cunha 12/05 |
2002–2003 | RCT | Premature infant |
52 | 52 | New = 4·2 (3·1), Old = 4·4 (4–0)e |
1·6 | 9 | 1·6 (0·6)e | 9·0 (8·9)e | CPDA-1 | Yes | Hospital mortality |
Higher |
| Fergusson 10/12 | 2006–2011 | RCT | NICU patient |
377 | 377 | New = 5·01 (4·00), Old = 4–94 (3·88)e |
<7 | 2–42 | 5·1 (20)e | 14·6 (8·3)e | CPD, SAGM, AS-3 |
Partially | 90-day mortality |
Lower |
| Steiner 09/14 | 2010–2014 | RCT | Cardiac surgery |
1481 | 1098 | New = 3 (2·6), Old = 4 (2·6)c |
<10 | >21 | 7 (5–9)c | 28 (24·33)c | AS-3 | Yes | 28-day mortality |
Lower |
| Lacroix 03/15 | 2009–2014 | RCT | ICU patient | 2430 | 2430 | New = 4·3 (5·2), old = 4·3 (5·5)h |
<8 | 1–42 | 6·1 (4·9)e | 22·0 (8·4)e | SAGM | Yes | 90-day mortality |
Lower |
NR, not reported; OBR, observational retrospective study; OBP, observational prospective study; RCT, randomized controlled trial; ICU, intensive care unit; PICU, pediatric intensive care unit; PRBC, packed red blood cells.
When multiple ages of stored blood were presented only the extremes were analysed (see statistical method).
The numbers and letters indicate the type of storage solution: C=citrate, A=adenine, P=phosphate, D=dextrose, S=sodium chloride, M=mannitol, G=glucose.
Median (IQR).
Median.
Mean (SD).
Table 2.
Characteristics of observational studies included in the meta-analysis
| Author (Reference) |
Years of enrollment |
Study type |
Study population |
No. of patients in study |
No. of patients used in the meta- analysisa |
Mean volume of RBCs transfused (units of PRBC) |
Age of stored blood compareda |
Actual median age of stored blood |
Blood storage solutionc |
Leucoreduced | Mortality | Risk of bias |
||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| New blood (days) |
Old blood (days) |
New blood (days) |
Old blood (days) |
|||||||||||
| Mynster 07/01 | 1991–1993 | OBP | Colorectal surgery |
740 | 429 | 3g | <21 | >21 | 19 (3–35)b | SAGM | Yes | Long-term mortality |
Higher | |
| Murrell 09/05 | 2001–2002 | OBR | Trauma | 275 | 275 | 3 (2–6)a | NR | NR | NR | NR | NR | Yes | Hospital mortality |
Higher |
| Van de Watering 10/06 |
1993–1999 | OBR | Cardiac surgery | 2732 | 1895 | New = 4 (3–5), Old = 4 (3–5)a |
<18 | >18 | 13 (11–14)a | 24 (21–27)a | SAGM | Yes | 30-day mortality |
Lower |
| Koch 03/08 | 1998–2006 | OBR | Cardiac surgery | 6002 | 6002 | 2 (2–4)a | <14 | >14 | 11 (3–9)a | 20 (17–25)a | NR | NR | Hospital mortality |
Lower |
| Leal-Noval 04/08 |
2004–2006 | OBP | Anemia | 66 | 34 | NR | <10 | >19 | 16 (8–8)h | SAGM | Yes | ICU mortality |
Higher | |
| Weinberg 08/08 | 2000–2007 | OBR | Trauma | 1813 | 1169 | 4·95 (1–74)i | <14 | >14 | NR | NR | NR | Yes | Hospital mortality |
Higher |
| Yap 08/08 | 2001–2007 | OBR | Cardiac surgery | 670 | 670 | 3 (2–5)a | 8 & | 19 & | NR | Partially | Hospital mortality |
Lower | ||
| Weinberg 10/08 | 2000–2007 | OBR | Trauma | 430 | 430 | 3 (1–104)b | <14 | >14 | NR | NR | NR | Yes | Hospital mortality |
Lower |
| Spinel la 09/09 | 2004–2007 | OBR | Trauma | 202 | 176 | New = 9 (6–12·5), old = 9 (6–12)a |
<21 | >21 | 14 (11–17)a | 20·5 (15·5–26)a |
NR | Partially | Hospital mortality |
Higher |
| Van Buskirk 09/09 |
NR | OBR | ICU patient | 298 | 298 | NR | <8 | >14 | NR | NR | NR | NR | Hospital mortality |
Lower |
| Karam 04/10 | 2004–2005 | OBP | PICU patient | 296 | 296 | New = 2·6 (3·6), old = 5·5 (5·7)h |
<14 | >14 | 14c | NR | Partially | 28-day mortality |
Higher | |
| Eikelboom 05/10 | 2002–2006 | OBP | Cardiac disease | 4933 | 4933 | 3 (2–5)a | <10 | 31–42 | 17 (13–22)a | AS-3 | Yes | Hospital mortality |
Lower | |
| Robinson 05/10 | 1999–2005 | OBR | Cardiac surgery | 909 | 712 | 2 (1–21)b | <21 | >21 | 25 (18–34)a | NR | Yes | 30-day mortality |
Higher | |
| Edgren 06/10 | 1995–2002 | OBR | Transfused at least one RBC unit |
387 130 | 387 130 | New = 2 (2–4), old = 2 (2–3)a |
<9 | >30 | 7 (5–8)a | 34 (31–37)a | SAGM | NR | Short-term mortality (1 week) |
Higher |
| Gauvin 09/10 | 2001–2005 | OBP | PICU patient | 455 | 224 | New = 1·2 (0·5), old = 2·3 (5·1)h |
<7 | >21 | 16·7 (10·3)h | NR | Yes | 28-day mortality |
Higher | |
| Weinberg 12/10 | 2000–2009 | OBR | Trauma | 1647 | 1647 | New = 2·9, old = 3·4 |
<14 | >14 | NR | NR | NR | Yes | Hospital mortality |
Higher |
| Van Straten 01/11 |
1998–2007 | OBR | Cardiac surgery | 3,597 | 3,141 | New = 2·6 (1·5), old = 2·4 (1·4)h |
<14 | >14 | NR | NR | SAGM | Yes | 30-day mortality |
Lower |
| Petti la 04/11 | 2008 | OBP | CU patient (55% surgical, 45% non-surgical) |
757 | 379 | 2 (1–4)a | <11 | >28 | 7·7 (2–11)b | 34·4 (28–42)b | NR | Partially | Hospital mortality |
Lower |
| Mc Kenny 05/11 |
2002–2007 | OBR | Cardiac surgery | 1153 | 972 | New = 2·4 (1·5), Old = 4·6 (3·9)h |
<14 | >14 | 10·8 (2·6)h | 25·4 (4·8)h | NR | Yes | 30-day or hospital mortality |
Lower |
| Sanders 11/11 | 2005–2007 | OBP | Cardiac surgery | 176 | 142 | 2 (1–2)a | <14 | >14 | 9 (2–12)a | 20 (16–23)h | NR | Yes | Hospital mortalityd |
Higher |
| Phelan 02/12 | NR | OBP | Trauma | 153 | 153 | 9·9 (9·0)h | <14 | >14 | NR | NR | NR | Yes | Higher | |
| Dunn 04/12 | 2000–2010 | OBR | Orthotopic liver transplantation |
509 | 509 | 10 (5–18)a | 11·8 & | 20 & | NR | NR | 2-y mortality | Lower | ||
| Middelburg 01/13 |
2005–2009 | OBR | Adult patients | 4549 | 947 | New = 2 (2–3). Old = 2 (2 3)a |
<10 | >24 | NR | NR | SAGM | Yes | 1-y mortality | Higher |
| Kadar 06/13 | 2007–2010 | OBR | Hip fracture surgery |
1381 | 663 | New = 1·36, old = 1·33 |
<14 | >14 | 9·1i | 22·3i | NR | NR | 1-year mortality |
Higher |
| Kaukonen 10/13 | 2011 | OBP | ICU patient | 652 | 652 | 3 (2–6)b | 12 (10–13)a | 21 (16, 27)a | SAGM | Yes | Hospital mortality |
Lower | ||
| Cywinski 11/13 | 2001–2011 | OBR | Orthotopic liver transplantation |
820 | 637 | New = 6 (3–9), Old = 6 (3–10)e |
<15 | >15 | 12 (9–13)a | 19 (18–22)a | NR | NR | ICU mortality | Higher |
| Cartotto 03/14 | 2000–2010 | OBR | Burns > 20% TBSA |
127 | 127 | 11 (1–40)b | <28 | >28 | 22·2 (10·3–28)b |
31·5 (28·3–41)b |
AS-3, SAGM |
Yes | ICU mortality | Higher |
| Aubron 06/14 | 2001–2011 | OBR | ICU patient | 8416 | 8416 | 4 (2–7)a | <21 | >21 | 13 (9–17)a | 27 (24–31)a | NR | Partially | Hospital mortality |
Lower |
| Baltsavias 06/14 | 2006–2010 | OBR | Pediatric cardiac surgery |
570 | 570 | New = 1·11, Old = 1·16 |
<7 | >7 | 6 (5–7)a | 14 (11–19)a | SAGM | No | ICU mortality |
Higher |
| Min 10/14 | 2005–2012 | OBR | Cardiac surgery |
1072 | 1072 | 7(8)h | <14 | >14 | 11·7 (5·2)h | CPDA-1 | No | Hospital mortality |
Lower | |
| Heddle 02/15 | 2006–2011 | OBR | Cardiac disease | 4594 | 4594 | 3 (2–5)a | <14 | >35 | 18 (14–24)a,b | AS-3 | Yes | Hospital mortality |
Lower | |
NR not reported; OBR, observational retrospective study; OBP, observational prospective study; RCT, randomized controlled trial; ICU, intensive care unit; PICU, pediatric intensive care unit; PRBC, packed red blood cells.
When multiple ages of stored blood were presented, only the extremes were analysed (see statistical method).
See statistical method for how stored blood groups were determined for comparison. For Pettila, Table 2 data was used. For Heddle, Table 3 results for week 6 vs. weeks 1–2 were used.
The numbers and letters indicate the type of storage solution: C = citrate, A = adenine, P = phosphate, D = dextrose, S = sodium chloride, M = mannitol, G = glucose.
Raw mortality data was obtained directly from the authors.
Median (IQR).
Median (range).
Median.
Mean (SD).
Mean (range).
Data extraction
Two investigators (K.E.R and I.C.P.) independently reviewed each publication and extracted data on study design, patient characteristics, storage age of transfused RBCs and mortality rates (Tables 1, 2). A third reviewer resolved disagreements (C.N.).
Data synthesis and analysis
The odds ratio of death (OR) following transfusion of old versus fresh stored RBCs was estimated using a random-effects model. Each study was summarized consistent with our previous meta-analysis [4]. We considered statistical significance (P < 0·05) when the OR did not cross 1. When both count data (survivor or non-survivor) and adjusted results [hazard ratio (HR), odds ratio (OR) or relative risk (RR)] were provided, all analyses were performed using both the adjusted data and the counts. Heterogeneity among studies was assessed using the Q statistic and I2 value. See supplemental methods for further details.
Results
Study characteristics
Tables 1, 2 summarize the characteristics of the included studies. Two RCTs [8, 9] and thirteen observational studies [5, 6, 10–20] published between 2011 and 2015 were added to the 21 studies (18 observational studies and 3 RCTs) included in our previous analysis (Fig. 1, Tables 1, 2) [21–38]. One RCT [39] was reported only in abstract form during our study enrolment period, but has subsequently been published in full [9].
Effect of old versus fresh transfused RBCs on mortality
Among the five RCTs [8, 9, 40–42], comparing 1- to 10-day-old stored RBC transfusion to usual care, heterogeneity was minimal (I2 = 0%, P = 0·56) and there was no significant effect on mortality [OR of death = 1·03, 95% confidence interval (CI) = 0·72–1·47, Fig. 2]. Including a large RCT in critically ill patients completed after the enrolment dates for studies in our meta-analysis [7], the effect of 1- to 10-day-old stored RBC transfusion on mortality compared to current practice among the six RCTs was similar, with minimal heterogeneity among the trials (I2 = 0%, P = 0·61), and a non-significant effect on mortality (OR = 0·91, 95% CI = 0·77–1·07). Based on the combined results of six RCTs, including 4031 patients, transfusion of recently donated RBCs offers no survival advantage in comparison with current practice.
Fig. 2.
Mortality in randomized controlled trials (RCTs). Lacroix trial was published after are enrolment period closed and part of a post hoc analysis. The size of the data markers is proportional to the inverse variance of each point estimate.
In contrast, among the 31 observational studies evaluating current practice, there was moderate heterogeneity (I2 = 38·9%, P = 0·02), and overall, the OR of death with transfusion of older compared with fresher conventionally issued RBCs was significantly increased 1·13 (95% CI = 1·03–1·24) (Fig. 3). When the analysis was done using adjusted results instead of counts for those studies in which these data were available, heterogeneity among studies moderately increased (I2 = 59·4%) and the overall effect size was slightly greater (OR = 1·18, 95% CI = 1·01–1·36). We performed multiple subgroup analyses aiming at identifying potential sources of this increased heterogeneity (supplemental results). An overall increase in mortality with longer stored transfused RBCs was consistently found among different surgical patient populations, in both adult and paediatric studies regardless of the volume of RBCs transfused, the size of the study and the date of publication (Supplemental Figures 2–8). The results of these sensitivity analyses were essentially unchanged if counts alone were employed or if adjusted analyses were substituted for counts when available. However, in contrast to the lack of significant mortality differences found in the six RCTs, among the 31 observational studies there was a different effect on mortality (P = 0·01), with a significant increase during current practice in mortality with transfusion of older versus fresher conventionally issued RBC units.
Fig. 3.
Mortality in observational studies. The overall summary of observational studies is provided using both counts for all studies and using adjusted results when available. The summary data show the RCTs had a significantly different effect on mortality than observational studies.
We next examined whether the different study designs employed and data sets obtained were associated with any divergence in the storage age of transfused RBCs. Among observational studies [6, 10–12, 14–18, 23, 24, 29, 30, 35, 38] and RCTs [7–9, 41, 42] reporting the median or mean storage age for the two transfused groups, we found that, although the degree of the differences in the storage age of transfused RBCs between the old and fresh RBC groups was similar for observational studies and RCTs (P = 0·69 for interaction), there was an overall significant downward shift of the median/mean storage age of transfused RBCs from the observational studies to the RCTs (main effect, P = 0·01) (Fig. 4). Thus, the median/mean storage age of transfused RBCs was significantly lower in both arms of the RCTs compared with both arms of the observational studies.
Fig. 4.
Storage age of transfused red blood cells comparing observational studies and RCTs divided into older versus fresher blood groups.
In the observational studies, a baseline difference in severity of illness of patients could underlie the ‘harmful’ effect of older RBCs. In 8 of 31 observational studies, the baseline severity of illness of patients in both study arms was described using a variety of scoring systems (Supplemental Figure 1) [10, 11, 14, 16–18, 25, 30]. In these studies with different severity of illness score balances between older and fresher RBC transfusion, there was no significant difference in treatment effect (P = 0·66 for interaction) and overall, among these eight studies, older RBC transfusion significantly increased the risk of death (OR = 1·46, 95% CI = 1·00–2·14, P = 0·05). Furthermore, five studies had identical baseline severity of illness scores in the two arms and also found an increased risk of death with transfusion of longer stored RBCs (OR = 1·65, 95% CI = 0·96–2·82, P = 0·07). These results are unchanged if we use count data alone for analysis or substitute adjusted values for counts when available. Examining the 10 observational studies with adjusted analysis and a low risk of bias, there was still an overall increase in the OR of death with transfusion of older versus fresher stored RBCs (OR = 1·12, 95% CI = 0·87–1·44) (Supplemental Figure 9). See supplemental results for details on the risk of bias within studies and publication bias.
Discussion
The six RCTs included in this analysis addressed the hypothesis that recently obtained RBCs, transfused 1–10 days after donation, are superior to the 2- to 3-week-old RBCs commonly transfused in standard practice. Ethical considerations prevented the six RCTs from intentionally ‘ageing’ RBCs and selecting the longest possible stored units to be randomized for human use. Therefore, these RCTs focused on evaluating the potential benefit of transfusing very recently donated RBCs, rather than on the potential risks associated with transfusing the longest possible stored units licensed for transfusion. These RCTs are appropriately assigned the highest level of evidence, and their combined analysis confirms that recently donated RBCs provide no survival advantage to current practice. This result was consistent across different clinical populations, including neonatal [8], cardiac surgery [9] and critically ill patients [7].
No similar body of high-quality evidence evaluating the risks of transfusing the oldest stored RBCs exists. The RCTs performed to date cannot exclude the possibility that RBCs stored for 4 to 6 weeks increase the risks of transfusion. The best available evidence evaluating the potential risks associated with transfusing these oldest RBCs is provided by a large number of observational analyses that studied significantly older RBC units than did the RCTs and were designed specifically to examine in usual practice the risk of the oldest donated RBCs. These observational studies suggest an overall increase in the OR of death with transfusion of longer stored conventional issue RBCs.
Evidence from non-randomized studies is not equivalent to evidence from RCTs. Lack of randomization and controls can subject observational studies to both recognized and unrecognized bias [43]. Even large observational studies are useful primarily for generating hypotheses. However, in our analysis, the adverse effect of longer stored RBC transfusion on survival rates was also present among those studies with a lower risk of bias that reported adjusted mortality data based on prognostic factors (Supplemental Figure 9). Even observational studies in which the baseline severity of illness score was identical for patients enrolled in the two arms found a significant increase in mortality risk with older stored transfused RBCs (Supplemental Figure 1). Consistent with observational studies, a large body of animal data [3, 44–46] suggests that transfused RBCs stored for more than 4 weeks can worsen outcomes, particularly in clinical settings such as infection in which the consequences of in vivo haemolysis, with iron and cell-free haemoglobin release, may increase risks.
Despite being considered the highest level of evidence, RCTs are also subject to acknowledged limitations. RCTs represent a snapshot over a short period of time and may not be fully representative of the total target population and of actual usual care practices. Further, most of the RCTs included here had substantial overlap in the ages of RBCs transfused in both groups. Our meta-analysis did not evaluate the effects of different storage ages of transfused RBCs on outcomes other than mortality. However, recently published systematic reviews of exclusively RCTs have evaluated the effect of fresher RBCs compared to current practice on secondary outcomes and found minimal adverse effects [47, 48].
Multiple in vitro experiments [1, 49], studies conducted in human volunteers [50] and animal models [3, 44], have provided evidence that RBCs undergo a series of progressive changes during storage. With increasing storage age, haemolysis occurs both in the storage bag and in vivo after transfusion, with release of iron and cell-free haemoglobin (CFH) [3, 44, 46, 51]. Increases in iron and CFH with transfusion may increase risks by well-defined mechanisms. Transfusion of 42-day-old stored RBCs in human volunteers [50] and animal models [3, 44–46, 51] results in increased plasma levels of iron in the form of non-transferrin-bound iron (NTBI). Increases in plasma NTBI have been shown to worsen outcomes in the presence of established infection [3, 44–46, 51, 52]. Similarly, in populations at risk of infection, increasing volumes of older RBC transfusion appear to be associated with an increased risk of complicated sepsis [53]. Increased CFH levels can also worsen outcomes through the vascular effects of nitric oxide (NO) scavenging [54, 55]. Transfusion-related increases in CFH have been shown in canines to scavenge NO, producing significant vascular effects [3].
Whereas the UK, the Netherlands and several other countries have limited the storage period of RBCs to 35 days or less, longer stored RBCs are still transfused throughout the world. Few published data and statistics address distribution of transfused RBCs by storage age. The mean age of conventionally issued transfused RBCs in the USA is only 17·9 days [6]. However, unpublished data from major blood collectors in the USA indicate that thousands of units of RBCs aged 4 to 6 weeks are released from blood collectors each week, accounting for approximately 8–18% of all the released RBC units (HG Klein, personal communication). Many more fresher units released to hospitals likely reach the last 2 weeks of storage on the hospital shelf prior to transfusion. This is of more than academic interest. While RCTs have successfully addressed whether recently donated RBCs are safer to transfuse than current transfusion practice, the potentially harmful effects of the longest possible stored RBCs are still incompletely understood.
Moderately sized RCTs similar to the ones included in our analysis are unlikely to answer whether the longest possible stored RBCs are safe for transfusion. Such trials do not enrol a large enough group of patients transfused with RBCs at the end of the storage period for meaningful comparisons. Future directions should focus on alternative study designs such as multicentre RCTs in which patient enrolment is limited to days when stored RBCs available for transfusion are older than 4 weeks, or large ‘pragmatic’ multicentre RCTs such as the INFORM trial in Canada which has enrolled 31 497 patients across four countries [56]. Lastly, high-quality, prospective, observational studies with a careful adjustment for prognostic factors and an adequate sample size can provide relevant data to address this question. There remains a pressing need to establish the safety of RBCs transfused at the end of the licensed storage period worldwide.
Supplementary Material
Acknowledgments
JS performed statistical analyses; JW performed the literature search; DW and SBS helped in manuscript preparation; KER, IC, HGK and CN participated in study design literature search, data extraction, data analysis/interpretation and manuscript preparation.
Funding
Intramural sources at NIH.
Footnotes
Disclaimer
The opinions expressed here are the authors’ own and do not represent any position or policy of NIH, HHS or US government.
Conflicts of interest
The authors declare no conflict of interests.
Additional Supporting Information may be found in the online version of this article: Data S1. Search Terms.
References
- 1.Orlov D, Karkouti K. The pathophysiology and consequences of red blood cell storage. Anaesthesia. 2015;70(Suppl 1):29–37. doi: 10.1111/anae.12891. e9–12. [DOI] [PubMed] [Google Scholar]
- 2.Flegel WA, Natanson C, Klein HG. Does prolonged storage of red blood cells cause harm? Br J Haematol. 2014;165:3–16. doi: 10.1111/bjh.12747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Solomon SB, Wang D, Sun J, et al. Mortality increases after massive exchange transfusion with older stored blood in canines with experimental pneumonia. Blood. 2013;121:1663–1672. doi: 10.1182/blood-2012-10-462945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang D, Sun J, Solomon SB, et al. Transfusion of older stored blood and risk of death: a meta-analysis. Transfusion. 2012;52:1184–1195. doi: 10.1111/j.1537-2995.2011.03466.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Phelan HA, Eastman AL, Aldy K, et al. Prestorage leukoreduction abrogates the detrimental effect of aging on packed red cells transfused after trauma: a prospective cohort study. Am J Surg. 2012;203:198–204. doi: 10.1016/j.amjsurg.2011.05.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Aubron C, Bailey M, McQuilten Z, et al. Duration of red blood cells storage and outcome in critically ill patients. J Crit Care. 2014;29:476. doi: 10.1016/j.jcrc.2014.01.006. e1–476.e8. [DOI] [PubMed] [Google Scholar]
- 7.Lacroix J, Hebert PC, Fergusson DA, et al. Age of transfused blood in critically Ill adults. N Engl J Med. 2015;372:1410–1418. doi: 10.1056/NEJMoa1500704. [DOI] [PubMed] [Google Scholar]
- 8.Fergusson DA, Hebert P, Hogan DL, et al. Effect of fresh red blood cell transfusions on clinical outcomes in premature, very low-birth-weight infants: the ARIPI randomized trial. JAMA. 2012;308:1443–1451. doi: 10.1001/2012.jama.11953. [DOI] [PubMed] [Google Scholar]
- 9.Steiner ME, Ness PM, Assmann SF, et al. Effects of red-cell storage duration on patients undergoing cardiac surgery. N Engl J Med. 2015;372:1419–1429. doi: 10.1056/NEJMoa1414219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McKenny M, Ryan T, Tate H, et al. Age of transfused blood is not associated with increased postoperative adverse outcome after cardiac surgery. Br J Anaesth. 2011;106:643–649. doi: 10.1093/bja/aer029. [DOI] [PubMed] [Google Scholar]
- 11.Sanders J, Patel S, Cooper J, et al. Red blood cell storage is associated with length of stay and renal complications after cardiac surgery. Transfusion. 2011;51:2286–2294. doi: 10.1111/j.1537-2995.2011.03170.x. [DOI] [PubMed] [Google Scholar]
- 12.Dunn LK, Thiele RH, Ma JZ, et al. Duration of red blood cell storage and outcomes following orthotopic liver transplantation. Liver Transpl. 2012;18:475–481. doi: 10.1002/lt.23379. [DOI] [PubMed] [Google Scholar]
- 13.Middelburg RA, van de Watering LM, Briet E, et al. Storage time of red blood cells and mortality of transfusion recipients. Transfus Med Rev. 2013;27:36–43. doi: 10.1016/j.tmrv.2012.06.002. [DOI] [PubMed] [Google Scholar]
- 14.Kadar A, Chechik O, Katz E, et al. The effects of ‘old’ red blood cell transfusion on mortality and morbidity in elderly patients with hip fractures-a retrospective study. Injury. 2013;44:747–750. doi: 10.1016/j.injury.2013.02.011. [DOI] [PubMed] [Google Scholar]
- 15.Kaukonen KM, Vaara ST, Pettila V, et al. Age of red blood cells and outcome in acute kidney injury. Crit Care. 2013;17:R222. doi: 10.1186/cc13045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cywinski JB, You J, Argalious M, et al. Transfusion of older red blood cells is associated with decreased graft survival after orthotopic liver transplantation. Liver Transpl. 2013;19:1181–1188. doi: 10.1002/lt.23695. [DOI] [PubMed] [Google Scholar]
- 17.Cartotto R, Yeo C, Camacho F, et al. Does the storage age of transfused blood affect outcome in burn patients? J Burn Care Res. 2014;35:186–197. doi: 10.1097/BCR.0b013e318295755d. [DOI] [PubMed] [Google Scholar]
- 18.Baltsavias I, Faraoni D, Willems A, et al. Blood storage duration and morbidity and mortality in children undergoing cardiac surgery. A retrospective analysis. Eur J Anaesthesiol. 2014;31:310–316. doi: 10.1097/EJA.0000000000000024. [DOI] [PubMed] [Google Scholar]
- 19.Min JJ, Bae JY, Kim TK, et al. Association between red blood cell storage duration and clinical outcome in patients undergoing off-pump coronary artery bypass surgery: a retrospective study. BMC Anesthesiol. 2014;14:95. doi: 10.1186/1471-2253-14-95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Heddle NM, Eikelboom J, Liu Y, et al. Exploratory studies on the age of transfused blood and in-hospital mortality in patients with cardiovascular diagnoses. Transfusion. 2015;55:364–372. doi: 10.1111/trf.12861. [DOI] [PubMed] [Google Scholar]
- 21.Mynster T, Nielsen HJ. Danish RCCSG: storage time of transfused blood and disease recurrence after colorectal cancer surgery. Dis Colon Rectum. 2001;44:955–964. doi: 10.1007/BF02235483. [DOI] [PubMed] [Google Scholar]
- 22.Murrell Z, Haukoos JS, Putnam B, et al. The effect of older blood on mortality, need for ICU care, and the length of ICU stay after major trauma. Am Surg. 2005;71:781–785. doi: 10.1177/000313480507100918. [DOI] [PubMed] [Google Scholar]
- 23.van de Watering L, Lorinser J, Versteegh M, et al. Effects of storage time of red blood cell transfusions on the prognosis of coronary artery bypass graft patients. Transfusion. 2006;46:1712–1718. doi: 10.1111/j.1537-2995.2006.00958.x. [DOI] [PubMed] [Google Scholar]
- 24.Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358:1229–1239. doi: 10.1056/NEJMoa070403. [DOI] [PubMed] [Google Scholar]
- 25.Leal-Noval SR, Munoz-Gomez M, Arellano-Orden V, et al. Impact of age of transfused blood on cerebral oxygenation in male patients with severe traumatic brain injury. Crit Care Med. 2008;36:1290–1296. doi: 10.1097/CCM.0b013e3181692dfc. [DOI] [PubMed] [Google Scholar]
- 26.Weinberg JA, McGwin G, Jr, Griffin RL, et al. Age of transfused blood: an independent predictor of mortality despite universal leukoreduction. J Trauma. 2008;65:279–282. doi: 10.1097/TA.0b013e31817c9687. discussion 82–4. [DOI] [PubMed] [Google Scholar]
- 27.Weinberg JA, McGwin G, Jr, Marques MB, et al. Transfusions in the less severely injured: does age of transfused blood affect outcomes? J Trauma. 2008;65:794–798. doi: 10.1097/TA.0b013e318184aa11. [DOI] [PubMed] [Google Scholar]
- 28.Weinberg JA, McGwin G, Jr, Vandromme MJ, et al. Duration of red cell storage influences mortality after trauma. J Trauma. 2010;69:1427–1431. doi: 10.1097/TA.0b013e3181fa0019. discussion 31–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Yap CH, Lau L, Krishnaswamy M, et al. Age of transfused red cells and early outcomes after cardiac surgery. Ann Thorac Surg. 2008;86:554–559. doi: 10.1016/j.athoracsur.2008.04.040. [DOI] [PubMed] [Google Scholar]
- 30.Spinella PC, Carroll CL, Staff I, et al. Duration of red blood cell storage is associated with increased incidence of deep vein thrombosis and in hospital mortality in patients with traumatic injuries. Crit Care. 2009;13:R151. doi: 10.1186/cc8050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.van Buskirk CM, Kashyap R, Thakur SJ, et al. Red blood cell storage has no impact on clinical outcome in critically ill patients. Transfusion. 2009;49(Suppl):10A. [Google Scholar]
- 32.Karam O, Tucci M, Bateman ST, et al. Association between length of storage of red blood cell units and outcome of critically ill children: a prospective observational study. Crit Care. 2010;14:R57. doi: 10.1186/cc8953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Eikelboom JW, Cook RJ, Liu Y, et al. Duration of red cell storage before transfusion and in-hospital mortality. Am Heart J. 2010;159:737–743. doi: 10.1016/j.ahj.2009.12.045. e1. [DOI] [PubMed] [Google Scholar]
- 34.Robinson SD, Janssen C, Fretz EB, et al. Red blood cell storage duration and mortality in patients undergoing percutaneous coronary intervention. Am Heart J. 2010;159:876–881. doi: 10.1016/j.ahj.2010.02.018. [DOI] [PubMed] [Google Scholar]
- 35.Edgren G, Kamper-Jorgensen M, Eloranta S, et al. Duration of red blood cell storage and survival of transfused patients (CME) Transfusion. 2010;50:1185–1195. doi: 10.1111/j.1537-2995.2010.02583.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Gauvin F, Spinella PC, Lacroix J, et al. Association between length of storage of transfused red blood cells and multiple organ dysfunction syndrome in pediatric intensive care patients. Transfusion. 2010;50:1902–1913. doi: 10.1111/j.1537-2995.2010.02661.x. [DOI] [PubMed] [Google Scholar]
- 37.van Straten AH, Soliman Hamad MA, van Zundert AA, et al. Effect of duration of red blood cell storage on early and late mortality after coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2011;141:231–237. doi: 10.1016/j.jtcvs.2010.02.059. [DOI] [PubMed] [Google Scholar]
- 38.Pettila V, Westbrook AJ, Nichol AD, et al. Age of red blood cells and mortality in the critically ill. Crit Care. 2011;15:R116. doi: 10.1186/cc10142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Steiner ME, Triulzi DJ, Assmann SF, et al. Randomized trial results: red cell storage age is not associated with a significant difference in multiple-organ dysfunction score or mortality in transfused cardiac surgery patients. Transfusion. 2014;54(Suppl):15A. [Google Scholar]
- 40.Schulman CI, Nathe K, Brown M, et al. Impact of age of transfused blood in the trauma patient. J Trauma. 2002;52:1224–1225. doi: 10.1097/00005373-200206000-00036. [DOI] [PubMed] [Google Scholar]
- 41.Hebert PC, Chin-Yee I, Fergusson D, et al. A pilot trial evaluating the clinical effects of prolonged storage of red cells. Anest Analg. 2005;100:1433–1438. doi: 10.1213/01.ANE.0000148690.48803.27. table of contents. [DOI] [PubMed] [Google Scholar]
- 42.Fernandes da Cunha DH, Nunes Dos Santos AM, Kopelman BI, et al. Transfusions of CPDA-1 red blood cells stored for up to 28 days decrease donor exposures in very low-birth-weight premature infants. Transfus Med. 2005;15:467–473. doi: 10.1111/j.1365-3148.2005.00624.x. [DOI] [PubMed] [Google Scholar]
- 43.Grimes DA, Schulz KF. Bias and causal associations in observational research. Lancet. 2002;359:248–252. doi: 10.1016/S0140-6736(02)07451-2. [DOI] [PubMed] [Google Scholar]
- 44.Hod EA, Zhang N, Sokol SA, et al. Transfusion of red blood cells after prolonged storage produces harmful effects that are mediated by iron and inflammation. Blood. 2010;115:4284–4292. doi: 10.1182/blood-2009-10-245001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Prestia K, Bandyopadhyay S, Slate A, et al. Transfusion of stored blood impairs host defenses against Gram-negative pathogens in mice. Transfusion. 2014;54:2842–2851. doi: 10.1111/trf.12712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Wang D, Cortes-Puch I, Sun J, et al. Transfusion of older stored blood worsens outcomes in canines depending on the presence and severity of pneumonia. Transfusion. 2014;54:1712–1724. doi: 10.1111/trf.12607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Brunskill SJ, Wilkinson KL, Doree C, et al. Transfusion of fresher versus older red blood cells for all conditions. Cochrane Database Syst Rev. 2015;5:CD010801. doi: 10.1002/14651858.CD010801.pub2. [DOI] [PubMed] [Google Scholar]
- 48.Alexander PE, Barty R, Fei Y, et al. Transfusion of fresher versus older red blood cells in hospitalized patients: a systematic review and meta-analysis. Blood. 2015 doi: 10.1182/blood-2015-09-670950. pii: blood-2015-09-670950. [DOI] [PubMed] [Google Scholar]
- 49.Bennett-Guerrero E, Veldman TH, Doctor A, et al. Evolution of adverse changes in stored RBCs. Proc Natl Acad Sci USA. 2007;104:17063–17068. doi: 10.1073/pnas.0708160104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Hod EA, Brittenham GM, Billote GB, et al. Transfusion of human volunteers with older, stored red blood cells produces extravascular hemolysis and circulating non-transferrin-bound iron. Blood. 2011;118:6675–6682. doi: 10.1182/blood-2011-08-371849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Cortes-Puch I, Wang D, Sun J, et al. Washing older blood units before transfusion reduces plasma iron and improves outcomes in experimental canine pneumonia. Blood. 2014;123:1403–1411. doi: 10.1182/blood-2013-11-539353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Ang O, Gungor M, Aricioglu F, et al. The effect of parenteral iron administration on the development of Staphy-lococcus aureus-induced experimental pyelonephritis in rats. Int J Exp Pathol. 1990;71:507–511. [PMC free article] [PubMed] [Google Scholar]
- 53.Hassan M, Pham TN, Cuschieri J, et al. The association between the transfusion of older blood and outcomes after trauma. Shock. 2011;35:3–8. doi: 10.1097/SHK.0b013e3181e76274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Donadee C, Raat NJ, Kanias T, et al. Nitric oxide scavenging by red blood cell microparticles and cell-free hemoglobin as a mechanism for the red cell storage lesion. Circulation. 2011;124:465–476. doi: 10.1161/CIRCULATIONAHA.110.008698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Liu C, Liu X, Janes J, et al. Mechanism of faster NO scavenging by older stored red blood cells. Redox Biol. 2014;2:211–219. doi: 10.1016/j.redox.2013.12.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Blajchman MA, Carson JL, Eikelboom JW, et al. The role of comparative effectiveness research in transfusion medicine clinical trials: proceedings of a National Heart, Lung, and Blood Institute workshop. Transfusion. 2012;52:1363–1378. doi: 10.1111/j.1537-2995.2012.03640.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.