Anaesthesia 2014, 69, 1206–1213
doi:10.1111/anae.12751
Original Article Conditioning out-of-date bank-stored red blood cells using a cell-saver auto-transfusion device: effects on numbers of red cells and quality of suspension fluid* M. S. Read,1 P. Coles,2 M. Pomeroy,3 E. Anderson3 and M. I. Aziz4 1 Consultant Anaesthetist, 3 Core Trainee in Paediatrics, University Hospital of Wales, Cardiff, UK 2 Consultant Anaesthetist, Morriston Hospital, ABMU Health Board, Swansea, UK 4 Consultant Anaesthetist, Prince Charles Hospital, Merthyr Tydfil, UK
Summary We investigated the utility of a cell-saver device for processing out-of-date red blood cells, by washing twenty bags of red blood cells that had been stored for between 36 and 55 days. The volume of recovered cells, and the characteristics of the suspension fluid, were measured before and after treatment. The ratio of free haemoglobin to total haemoglobin was up to 0.02 before processing, and up to 0.011 afterwards, changing by between 0.013 and +0.003. This ratio met the current standard for free haemoglobin (less than 0.008 in more than 75% of samples), both before and after processing. Ninety-three percent of red blood cells survived the process. Potassium ion concentration fell from above 15 mmol.l1 in all cases, to a mean of 6.4 mmol.l1 (p < 0.001). The pH rose to a mean value of 6.44 (p = 0.001). Lactate ion concentration fell to a mean value of 14 mmol.l1 (p < 0.001). Sodium ion concentration rose from a mean value of 93 mmol.l1 to a mean value of 140 mmol.l1 (p < 0.001). A useful proportion of outof-date red blood cells remained intact after conditioning using a cell-saver, and the process lowered concentrations of potentially toxic solutes in the fluid in which they were suspended.
.................................................................................................................................................................
Correspondence to: M. S. Read Email: [email protected] *Presented in part at The World Anaesthesia Congress, Cape Town, South Africa, 2008. Accepted: 26 April 2014
Introduction Bank-stored red blood cells (RBCs) are discarded after a prescribed number of days in storage (35 in Britain, 42 in the USA) because of concerns about deterioration and resulting toxicity. Even for in-date RBCs, some authors have expressed concerns about high concentrations of potassium ions in the suspension fluid [1–3], and concentrations of hydrogen ions, lactate ions and free haemoglobin are potentially also of concern as storage time increases. During storage, 1206
RBCs develop a ‘storage lesion’ during storage, the early stages of which include suboptimal oxygen carriage and red cell deformability, and effects on the recipient’s vascular tone and on immune system [4, 5]. In the late stages, changes to the RBCs are evident on microscopy. Ideally, laboratory testing would be able to differentiate between bags of RBCs that are in good condition, transfusion of which will benefit the patient, and those that have deteriorated over time and will therefore do more harm than good. However, © 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
there is no universally accepted test or criterion for this [6]. Cell-savers are used routinely in operations with significant blood loss. Placing RBCs into a cell-saver that is already in use would involve almost no additional effort or cost compared with transfusing them directly, and it may be that this strategy would convert out-of-date RBCs from being unsafe for transfusion to being safe. Doing this in a blood bank would require extra equipment and training, and would introduce further concerns regarding the safe storage of the RBCs after processing. Investigating whether out-of-date RBCs can be made safe for transfusion into patients if they are processed by a cell-saver immediately before transfusion means determining: (1) the fraction of the original RBCs that survive the cell-saver washing process; (2) the toxicity of the resulting suspension fluid; (3) the fraction of transfused RBCs that survive for 24 h after transfusion; (4) physiological, biochemical, immunological or haematological changes in recipients; and (5) clinically relevant outcomes such as mortality, infection rates, etc. This article reports our investigation into the first two of these.
Methods We obtained confirmation from the local research and development committee that this study did not require formal ethical approval. Twenty bags of leucocyte-depleted RBCs suspended in ‘SAG-M’ (adenine, glucose and mannitol in saline) that had been refrigerated and stored for between 36 and 55 days were included in the study. Before being processed, each bag of bank-stored RBC suspension was first gently mixed, and then sampled by collecting 15 ml into a universal container and 1 ml into a heparinised blood gas syringe. The RBC suspension was then transferred from the storage bag into the bowl of a BRAT 2 (Baylar Rapid Autologous Transfusion, model 2) cell-saver system (COBE Cardiovascular Inc., Arvada, CO, USA) through a UHS Maxi Set Blood Infusion Set (Universal Hospital Supplies Ltd, Enfield, UK). These sets contain a 200-lm filter. The recommended volume of wash fluid during cell-saver processing differs according to circumstance. © 2014 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia 2014, 69, 1206–1213
Larger volumes are used in obstetric and orthopaedic surgery than in general or urological surgery. Since we were unable to predict the ideal wash volume in this novel use of the cell-saver, we chose to end the washing process when the emerging fluid appeared clear to the naked eye. This process produced RBCs suspended in wash fluid (0.9% saline). Volumes of RBC suspension put in, wash fluid used and RBC suspension produced were displayed by the cell-saver, and note was taken of these. After gentle mixing, samples of the resulting RBC suspension were again taken, 15 ml in a universal container and 1 ml in a heparinised blood gas syringe. The universal container samples were kept refrigerated before being analysed within three hours. An ABZ 800 Flex blood gas analyser (ABX Horiba Diagnostics, Montpellier, France) was used to measure pH and concentrations of bicarbonate, lactate, potassium and sodium ions. These analyses were done within 15 min of sampling. The haematocrit and the haemoglobin concentration of the RBC suspension in the universal containers was measured using a Pentra 120 blood analyser (ABX Horiba Diagnostics) and a Beckmann DU-640 spectrophotometer (Beckmann Coulter, Brea, CA, USA), respectively. The suspension fluid in the samples was isolated by centrifugation using a Heraeus Labofuge 400-R Centrifuge (Heraeus, Hanau, Germany), and the haemoglobin concentration of this supernatant was measured using the same spectrophotometer. Haemoglobin concentration was calculated from the spectrophotometer readings in the manner described in the extract from Welsh Blood Service Standard Operating Procedure CCQ-099, included below as Appendix 1. The two-way paired Student’s t-test was used to assess the significance of the haemoglobin concentration results. The Wilcoxon signed-rank test was used to analyse the pH, bicarbonate, lactate, potassium and sodium results.
Results The median (IQR [range]) age of the bags of RBC suspensions was 46 (36–48 [36–530]) days, and the volume of saline used for washing was 2000 (1013–2056 [1005–2542]) ml. 1207
Anaesthesia 2014, 69, 1206–1213
Read et al. | Cell saver conditioning of out-of-date RBCs
The volume of the bags was changed little by processing, from a mean of 234 ml to a mean of 241 ml. The mean difference was: +7.2 ml (SD 42 ml, 95% CI 18.6 ml [range 103 to +70 ml]). Total haemoglobin in the bags also changed little, from a mean of 42.1 g to a mean of 39.1 g. The mean difference was 3.0 g (SD 6.8 g, 95% CI 3.0 g [range 23.9 g to +6.3 g]). Because the fraction of haemoglobin that was extracellular was very small compared with the intracellular fraction, calculation of the amount of intracellular haemoglobin in the bags after processing produced very similar results. There was a change from a mean of 41.8 g to a mean of 38.9 g. The mean difference was 3.0 g (SD 6.5 g, 95% CI 2.7 g [range 23.9 g to +6.3 g]). The mean reduction in intracellular haemoglobin in the bags (3 g) was 7% of the mean intracellular haemoglobin in the bags before processing (41.8 g), implying a mean survival of RBCs of 93%. The percentage of bags with free haemoglobin less than 0.8% of total haemoglobin was 85% before processing and 90% after processing. Pre- and post-processing ratios of free to total haemoglobin concentrations, pH and concentrations of
bicarbonate, lactate, potassium and sodium ions are shown in Table 1 and in Figs 1–6. Among the pre-processing samples, some readings were outside the range for which the blood gas analyser could give values. Thus, some pre-processing bicarbonate, lactate,
Figure 1 Ratio of concentrations of extracellular-tototal haemoglobin before and after processing.
Table 1 Haemoglobin (Hb) ratios, bicarbonate (mmol.l1), lactate (mmol.l1) potassium (mmol.l1), pH, and sodium (mmol.l1) in supernatant, before (Pre) and after (Post) processing. (Results rounded to either 2 or 3 significant figures). Extracellular/total Hb ratio
Bicarbonate**
Lactate**
Pre
Post
Change*
Pre
Post
Pre
0.014 0.020 0.006 0.005 0.006 0.007 0.003 0.004 0.008 0.012 0.006 0.008 0.004 0.004 0.002 0.003 0.006 0.001 0.000 0.004 Means: 0.006
0.009 0.006 0.005 0.004 0.004 0.008 0.004 0.005 0.011 0.006 0.005 0.005 0.004 0.002 0.003 0.003 0.004 0.004 0.003 0.004 0.005
0.004 0.013 0.001 0.000 0.002 0.001 0.001 0.001 0.003 0.006 0.001 0.003 0.000 0.002 0.000 0.000 0.002 0.002 0.003 0.000 0.001
6.6 9.8 9.2 8.8 8.1 4.6 5.4 5.4 >
> > > >
26 24 24 24 24 27 27 29 30 30 30 30 30 21 25 30 24 30 30 30 27
Potassium**
pH***
Post
Pre
Post
Pre
Post
Pre
Post
14.7 13.0 11.1 10.8 10.7 8.7 14.0 19.0 14.6 19.0 17.0 14.9 12.3 10.0 17.0 20.0 18.0 8.8 11.9 13.9 14
> > > > > > > > > > > > > > > > > > > >
14.6 5.0 5.7 7.5 8.9 4.9 6.9 9.9 7.4 4.4 4.3 4.2 2.4 4.2 7.1 8.3 9.1 1.2 2.3 9.2 6.4
6.422 6.487 6.502 6.491 6.502 6.336 6.358 6.302 < 6.3 6.410 6.360 6.372 6.335 6.580 6.447 6.426 6.539 < 6.3 6.332 6.393 6.4
6.456 6.512 6.518 6.524 6.535 6.432 6.381 6.341 6.327 6.393 6.383 6.387 6.344 6.566 6.452 6.441 6.525 6.429 6.385 6.460 6.44
84 84 85 84 82 104 100 102 94 96 99 100 105 83 108 104 83 91 98 77 93
135 141 145 143 142 146 144 131 145 145 145 145 147 143 130 127 125 146 144 143 140
15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15
Sodium**
*p = 0.16; **p < 0.001; ***p = 0.001. 1208
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
Anaesthesia 2014, 69, 1206–1213
Figure 2 Bicarbonate ion concentration before and after processing. Figure 4 Potassium ion concentration before and after processing.
Figure 3 Lactate ion concentration before and after processing. potassium and pH readings were recorded as < 4 mmol1, > 30 mmol1, > 15 mmol1 and < 6.3, respectively. For the purposes of comparison and statistical testing, these were replaced with values of exactly 4 mmol1, 30 mmol1, 15 mmol1 and 6.3, respectively, which increased the likelihood of a type-2 error (failing to show a significant difference when in © 2014 The Association of Anaesthetists of Great Britain and Ireland
Figure 5 pH before and after processing. fact there was one). Non-parametric tests of significance were used for these parameters because this assumption would mean that the data were not normally distributed. 1209
Anaesthesia 2014, 69, 1206–1213
Read et al. | Cell saver conditioning of out-of-date RBCs
bags in which it was initially above the median, it fell in eight.
Discussion
Figure 6 Sodium ion concentration before and after processing. Figure 7 demonstrates that there was no relationship between the concentration of free haemoglobin (before processing) and the storage time of the RBCs. The effect of the cell-saver on the ratio of extracellular-to-total haemoglobin concentration is shown in Fig. 8. The mean change of this ratio was 0.0012. Of the ten bags in which the ratio was initially below the median, it rose in seven. Of the ten
Guidelines for permitted levels of haemolysis for bankstored RBCs in Britain (Appendix 2) stipulate that in more than 75% of bags of RBCs, fewer than 0.8% of the red cells should have undergone haemolysis [7]. In the context of RBCs that have been processed with a cell-saver, the number of cells that have lysed is not of interest: the washing process is able to remove the resulting debris. What is still of interest is the ratio of extracellular haemoglobin (the unwanted fraction) to total haemoglobin. The standard, expressed above as less than 0.8% of RBCs, can equally be expressed as a ratio of extracellular-to-total haemoglobin of < 0.008. Guidelines from the US Food and Drugs Administration (Appendix 3) stipulate that survival of RBCs at 24 h after transfusion should be greater than 75% [8]. In one study in 2008, RBCs 25–35 days old only just met this standard, and younger RBCs performed much better [9]. Most of the removal of what the authors of this article termed ‘removal-prone’ RBCs occurred in the first hour after transfusion. It has been suggested by many authors that RBCs in the first part of the maximum permitted storage time are safer than older RBCs, implying that the current maximum permitted storage time may be too long. This has been refuted recently in an observational study in adults [10] and a randomised controlled trial in infants [11]. Three further randomised controlled trials investigating this question are currently under way: the
Figure 7 Extracellular/total haemoglobin ratio in bags of RBCs (before processing) according to number of days of storage. 1210
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
Anaesthesia 2014, 69, 1206–1213
Figure 8 Extracellular/total haemoglobin ratio before (■) and after (□) processing, ranked in order of increasing pre-processing ratio.
Age of Blood Evaluation (ABLE) trial in the resuscitation of critically ill patients; the Red Cell Storage Duration (RECESS) study; and the Red Cell Storage Duration and Outcomes in Cardiac Surgery study. There are thus three types of evidence indicating the relationship between the age of bags of stored RBCs and their toxicity: (1) laboratory studies, with an uncertain relationship between laboratory measurements and clinical outcomes; (2) studies of physiological, biochemical, immunological or haematological changes in transfusion recipients, including the survival of RBCs after transfusion, again with an uncertain relationship between observed variables and clinical outcomes; and (3) epidemiological evidence that looks specifically at clinical outcomes, without concerning itself with mechanisms of toxicity. Current practice standards are based on some of this evidence, but none of it definitively indicates a safe maximum storage time. Rather than simply transfusing stored RBCs, some workers have tried processing (referring to it as ‘rejuvenating’) them before transfusion. In a study that involved washing 35–39 day-old RBCs with a specific mixture designed to replace lost substances and functions, the 24-hour post-transfusion survival value was 80% [12]. Others have processed in-date RBCs with a cellsaver and demonstrated a reduction in the toxicity of the suspension fluid [1–3]. However, this has been © 2014 The Association of Anaesthetists of Great Britain and Ireland
shown to induce some haemolysis, both at the time of processing and (when stored in vitro) for some hours afterwards [13]. A study using RBCs up to 49 days old also found that processing with a cell-saver reduced the toxicity of the suspension fluid, and that some haemolysis occurred, but the number of bags of RBCs was smaller than in the present study, and the authors did not measure the fraction of RBCs that survived [14]. The propensity for RBCs to haemolyse may be assessed using the mechanical fragility index, the haemolysis rate caused by a standardised mechanical stress applied to RBCs. Compared with fresh blood, this is raised in in-date RBCs [15] and in fresh blood processed by a cell-saver [16]. It is higher in in-date RBCs that have been processed by a cell-saver than in unprocessed, in-date RBCs [17]. The analysis of our results of the pH, bicarbonate, lactate and potassium concentrations included a risk of a type-2 error. Despite this fact, the differences were statistically significant for all four parameters, as well as for the sodium results. Therefore, it is reasonable to conclude that after processing with the cellsaver, reductions were seen in concentrations of potassium and lactate ions; sodium ion concentration rose from sub-physiological levels to physiological ones. Bicarbonate ion concentration fell, but pH rose slightly, in keeping with the fall in lactate ion concentration. 1211
Anaesthesia 2014, 69, 1206–1213
The concentration of free haemoglobin was somewhat high in a number of bags, but within the standard for in-date stored RBCs. Processing did not affect this proportion. There was a trend for the free haemoglobin concentration to rise in bags in which the initial concentration was low, and for it to fall when the initial concentration was high. The fact that it rose in some bags suggests that haemolysis occurred during processing, and that the haemoglobin released by this mechanism had been imperfectly removed. It may be that the final free haemoglobin concentration was determined more by the equilibrium of haemolysis and haemoglobin removal than by the initial free haemoglobin concentration. This would account for the ‘reversion towards the mean’ pattern we observed in changes in free haemoglobin concentration during processing. The small changes in volume of bags and the haemoglobin concentration within them imply that this haemolysis made little difference to the total number of RBCs. During the study, it was noted that fluid volumes measured by the cell-saver could be slight overestimates. This was because the rotary pump within the cell-saver sometimes made a small but variable number of extra turns after fluid had been replaced by air in the tubing at the point of the air sensor. The cell-saver is in daily use at our institution, and clinical decisions are made on the basis of its measurements of fluid volumes. We consider these measurement errors to be small. It may explain the fact that in some cases there appeared to be a small increase in the amount of haemoglobin in the bag. Recent studies (published or ongoing) address whether the maximum permitted storage time is appropriate or whether it is too long. This study has shown that RBCs that have been stored for longer than the currently accepted maximum time can be processed by a cell-saver autotransfusion device to improve the safety profile of the fluid in which they are suspended. This process causes haemolysis, but the number of RBCs lost through this is small. It also washes out free haemoglobin. The balance of these two effects produces an acceptably low concentration of free haemoglobin in the suspension fluid. We surmise that washing the RBCs for longer, with a larger volume of wash fluid, may have further 1212
Read et al. | Cell saver conditioning of out-of-date RBCs
reduced the concentrations of toxic chemicals, may have caused fewer RBCs to survive the process, and may have caused more haemolysis, with an unpredictable effect on the free haemoglobin concentration in the suspension fluid. Further work is warranted to investigate the characteristics of the RBCs that remain after processing. We postulate that the removal-prone RBCs would be lysed by the cell-saver, selecting out robust RBCs that remained intact.
Acknowledgements We gratefully acknowledge the help of Christopher Lee, Biomedical Scientist, University Hospital of Wales, for invaluable help in measurement of light absorption for haemoglobin concentration.
Competing interest No external funding and no competing interests declared.
References 1. Knichwitz G, Zahl M, Van Aken H, Semjonow A, Booke M. Intraoperative washing of long-stored packed red blood cells by using an autotransfusion device prevents hyperkalemia. Anesthesia and Analgesia 2002; 95: 324–5. 2. de Vroege R., Wildevuur WR, Muradin JAG, Graves D, van Oeveren W. Washing of stored red blood cells by an autotransfusion device before transfusion. Vox Sanguinis 2007; 92: 130–5. 3. Westphal-Varghese B, Erren M, Westphal M, et al. Processing of stored packed red blood cells using autotransfusion devices decreases potassium and microaggregates: a prospective, randomized, single-blinded in vitro study. Transfusion Medicine 2007; 17: 89–95. 4. Roback JD. Transfusion medicine I: adverse complications of stored blood: vascular effects of the red blood cell storage lesion. Hematology. American Society of Hematology Education Book 2011; 1: 475–9. 5. Yazdanbakhsh K, Bao W, Zhong H. Transfusion medicine I: adverse complications of stored blood: immunoregulatory effects of stored red blood cells. Hematology. American Society of Hematology Education Book 2011; 1: 466–9. 6. D’Alessandro A, Liumbruno G, Grazzini G, Zolla L. Red blood cell storage: the story so far. Blood Transfusion 2010; 8: 82–8. 7. James V. Chapter 8. Specifications for blood components. In: James V, ed. Guidelines for the Blood Transfusion Services in the United Kingdom, 7th edn. Norwich: Her Majesty’s Stationery Office, 2005: 64. 8. Dumont LJ, AuBuchon JP, for the Biomedical Excellence for Safer Transfusion (BEST) Collaborative. Evaluation of proposed FDA criteria for the evaluation of radiolabeled red cell recovery trials. Transfusion 2008; 48: 1053–60. 9. Luten M, Roerdinkholder-Stoelwinder B, Schaap NP, de Grip WJ, Bos HJ, Bosman GJ. Survival of red blood cells © 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
10.
11.
12.
13.
14.
15.
16.
17.
after transfusion: a comparison between red cells concentrates of different storage periods. Transfusion 2008; 48: 1478–85. McKenny M, Ryan T, Tate H, Graham B, Young VK, Dowd N. Age of transfused blood is not associated with increased postoperative adverse outcome after cardiac surgery. British Journal of Anaesthesia 2011; 106: 643–9. bert P, Hogan DL, et al. Effect of fresh red Fergusson DA, He blood cell transfusions on clinical outcomes in premature, very low-birth-weight infants: the ARIPI randomized trial. Journal of the American Medical Association 2012; 308: 1443–51. Valeri CR, Gray AD, Cassidy GP, Riordan W, Pivacek LE. The 24hour posttransfusion survival, oxygen transport function, and residual hemolysis of human outdated-rejuvenated red cell concentrates after washing and storage at 4 degrees C for 24 to 72 hours. Transfusion 1984; 24: 323–6. O’Leary MF, Szklarski P, Klein TM, Young PP. Hemolysis of red blood cells after cell washing with different automated technologies: clinical implications in a neonatal cardiac surgery population. Transfusion 2011; 51: 955–60. Smith T, Riley W, Fitzgerald D. In vitro comparison of two different methods of cell washing. Perfusion 2013; 28: 34–7. Raval JS, Waters JH, Seltsam A, et al. The use of the mechanical fragility test in evaluating sublethal RBC injury during storage. Vox Sanguinis 2010; 99: 325–31. Ley JT, Yazer MH, Waters JH. Hemolysis and red blood cell mechanical fragility in shed blood after total knee arthroplasty. Transfusion 2012; 52: 34–8. Harm SK, Raval JS, Cramer J, Waters JH, Yazer MH. Haemolysis and sublethal injury of RBCs after routine blood bank manipulations. Transfusion Medicine 2012; 22: 181–5.
Appendix 1 Calculations of free haemoglobin in the supernatant and of ‘percent haemolysis’ specified in the Welsh Blood Service standard operating procedures.
Anaesthesia 2014, 69, 1206–1213
where Fr-Hb is the haemoglobin concentration of the supernatant, and 576 nm-A, 623 nm-A and 700 nm-A are absorbance measurements at light wavelengths of 576, 623 and 700 nm, respectively. The dilution factor is used to produce a useful range of absorbances measurable at the above wavelengths. % Haemolysis ¼
ðð100 % HCTÞ supernatant HbÞ Total Hb
where HCT is the haematocrit.
Appendix 2 British guidelines for permitted levels of haemolysis for bank-stored RBCs: “Haemolysis measurements on red cell components are performed at the end of the component shelf life. Due to intermittent availability of outdated red cell components, each primary process should be validated to give haemolysis of < 0.8% of the red cell mass at the end of component shelf life in > 75% of components with a minimum of 20 components tested” [7].
Appendix 3 Levels of RBC survival expected by the US Food and Drugs Administration: “Radiolabelling studies should be performed in at least two separate centres (laboratories) with a total of 20–24 healthy donors. The mean recovery at 24 hours for each unit should be > 75% with SD < 9%; and the one sided 95% lower confidence limit for the population proportion of successes > 70%” [8].
Fr-Hb ðmg.100 ml1 Þ ¼ð115 576 nm-AÞ ð102 623 nm-AÞ ð39:1 700 nm-AÞ dilution factor
© 2014 The Association of Anaesthetists of Great Britain and Ireland
1213
doi:10.1111/anae.12751
Original Article Conditioning out-of-date bank-stored red blood cells using a cell-saver auto-transfusion device: effects on numbers of red cells and quality of suspension fluid* M. S. Read,1 P. Coles,2 M. Pomeroy,3 E. Anderson3 and M. I. Aziz4 1 Consultant Anaesthetist, 3 Core Trainee in Paediatrics, University Hospital of Wales, Cardiff, UK 2 Consultant Anaesthetist, Morriston Hospital, ABMU Health Board, Swansea, UK 4 Consultant Anaesthetist, Prince Charles Hospital, Merthyr Tydfil, UK
Summary We investigated the utility of a cell-saver device for processing out-of-date red blood cells, by washing twenty bags of red blood cells that had been stored for between 36 and 55 days. The volume of recovered cells, and the characteristics of the suspension fluid, were measured before and after treatment. The ratio of free haemoglobin to total haemoglobin was up to 0.02 before processing, and up to 0.011 afterwards, changing by between 0.013 and +0.003. This ratio met the current standard for free haemoglobin (less than 0.008 in more than 75% of samples), both before and after processing. Ninety-three percent of red blood cells survived the process. Potassium ion concentration fell from above 15 mmol.l1 in all cases, to a mean of 6.4 mmol.l1 (p < 0.001). The pH rose to a mean value of 6.44 (p = 0.001). Lactate ion concentration fell to a mean value of 14 mmol.l1 (p < 0.001). Sodium ion concentration rose from a mean value of 93 mmol.l1 to a mean value of 140 mmol.l1 (p < 0.001). A useful proportion of outof-date red blood cells remained intact after conditioning using a cell-saver, and the process lowered concentrations of potentially toxic solutes in the fluid in which they were suspended.
.................................................................................................................................................................
Correspondence to: M. S. Read Email: [email protected] *Presented in part at The World Anaesthesia Congress, Cape Town, South Africa, 2008. Accepted: 26 April 2014
Introduction Bank-stored red blood cells (RBCs) are discarded after a prescribed number of days in storage (35 in Britain, 42 in the USA) because of concerns about deterioration and resulting toxicity. Even for in-date RBCs, some authors have expressed concerns about high concentrations of potassium ions in the suspension fluid [1–3], and concentrations of hydrogen ions, lactate ions and free haemoglobin are potentially also of concern as storage time increases. During storage, 1206
RBCs develop a ‘storage lesion’ during storage, the early stages of which include suboptimal oxygen carriage and red cell deformability, and effects on the recipient’s vascular tone and on immune system [4, 5]. In the late stages, changes to the RBCs are evident on microscopy. Ideally, laboratory testing would be able to differentiate between bags of RBCs that are in good condition, transfusion of which will benefit the patient, and those that have deteriorated over time and will therefore do more harm than good. However, © 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
there is no universally accepted test or criterion for this [6]. Cell-savers are used routinely in operations with significant blood loss. Placing RBCs into a cell-saver that is already in use would involve almost no additional effort or cost compared with transfusing them directly, and it may be that this strategy would convert out-of-date RBCs from being unsafe for transfusion to being safe. Doing this in a blood bank would require extra equipment and training, and would introduce further concerns regarding the safe storage of the RBCs after processing. Investigating whether out-of-date RBCs can be made safe for transfusion into patients if they are processed by a cell-saver immediately before transfusion means determining: (1) the fraction of the original RBCs that survive the cell-saver washing process; (2) the toxicity of the resulting suspension fluid; (3) the fraction of transfused RBCs that survive for 24 h after transfusion; (4) physiological, biochemical, immunological or haematological changes in recipients; and (5) clinically relevant outcomes such as mortality, infection rates, etc. This article reports our investigation into the first two of these.
Methods We obtained confirmation from the local research and development committee that this study did not require formal ethical approval. Twenty bags of leucocyte-depleted RBCs suspended in ‘SAG-M’ (adenine, glucose and mannitol in saline) that had been refrigerated and stored for between 36 and 55 days were included in the study. Before being processed, each bag of bank-stored RBC suspension was first gently mixed, and then sampled by collecting 15 ml into a universal container and 1 ml into a heparinised blood gas syringe. The RBC suspension was then transferred from the storage bag into the bowl of a BRAT 2 (Baylar Rapid Autologous Transfusion, model 2) cell-saver system (COBE Cardiovascular Inc., Arvada, CO, USA) through a UHS Maxi Set Blood Infusion Set (Universal Hospital Supplies Ltd, Enfield, UK). These sets contain a 200-lm filter. The recommended volume of wash fluid during cell-saver processing differs according to circumstance. © 2014 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia 2014, 69, 1206–1213
Larger volumes are used in obstetric and orthopaedic surgery than in general or urological surgery. Since we were unable to predict the ideal wash volume in this novel use of the cell-saver, we chose to end the washing process when the emerging fluid appeared clear to the naked eye. This process produced RBCs suspended in wash fluid (0.9% saline). Volumes of RBC suspension put in, wash fluid used and RBC suspension produced were displayed by the cell-saver, and note was taken of these. After gentle mixing, samples of the resulting RBC suspension were again taken, 15 ml in a universal container and 1 ml in a heparinised blood gas syringe. The universal container samples were kept refrigerated before being analysed within three hours. An ABZ 800 Flex blood gas analyser (ABX Horiba Diagnostics, Montpellier, France) was used to measure pH and concentrations of bicarbonate, lactate, potassium and sodium ions. These analyses were done within 15 min of sampling. The haematocrit and the haemoglobin concentration of the RBC suspension in the universal containers was measured using a Pentra 120 blood analyser (ABX Horiba Diagnostics) and a Beckmann DU-640 spectrophotometer (Beckmann Coulter, Brea, CA, USA), respectively. The suspension fluid in the samples was isolated by centrifugation using a Heraeus Labofuge 400-R Centrifuge (Heraeus, Hanau, Germany), and the haemoglobin concentration of this supernatant was measured using the same spectrophotometer. Haemoglobin concentration was calculated from the spectrophotometer readings in the manner described in the extract from Welsh Blood Service Standard Operating Procedure CCQ-099, included below as Appendix 1. The two-way paired Student’s t-test was used to assess the significance of the haemoglobin concentration results. The Wilcoxon signed-rank test was used to analyse the pH, bicarbonate, lactate, potassium and sodium results.
Results The median (IQR [range]) age of the bags of RBC suspensions was 46 (36–48 [36–530]) days, and the volume of saline used for washing was 2000 (1013–2056 [1005–2542]) ml. 1207
Anaesthesia 2014, 69, 1206–1213
Read et al. | Cell saver conditioning of out-of-date RBCs
The volume of the bags was changed little by processing, from a mean of 234 ml to a mean of 241 ml. The mean difference was: +7.2 ml (SD 42 ml, 95% CI 18.6 ml [range 103 to +70 ml]). Total haemoglobin in the bags also changed little, from a mean of 42.1 g to a mean of 39.1 g. The mean difference was 3.0 g (SD 6.8 g, 95% CI 3.0 g [range 23.9 g to +6.3 g]). Because the fraction of haemoglobin that was extracellular was very small compared with the intracellular fraction, calculation of the amount of intracellular haemoglobin in the bags after processing produced very similar results. There was a change from a mean of 41.8 g to a mean of 38.9 g. The mean difference was 3.0 g (SD 6.5 g, 95% CI 2.7 g [range 23.9 g to +6.3 g]). The mean reduction in intracellular haemoglobin in the bags (3 g) was 7% of the mean intracellular haemoglobin in the bags before processing (41.8 g), implying a mean survival of RBCs of 93%. The percentage of bags with free haemoglobin less than 0.8% of total haemoglobin was 85% before processing and 90% after processing. Pre- and post-processing ratios of free to total haemoglobin concentrations, pH and concentrations of
bicarbonate, lactate, potassium and sodium ions are shown in Table 1 and in Figs 1–6. Among the pre-processing samples, some readings were outside the range for which the blood gas analyser could give values. Thus, some pre-processing bicarbonate, lactate,
Figure 1 Ratio of concentrations of extracellular-tototal haemoglobin before and after processing.
Table 1 Haemoglobin (Hb) ratios, bicarbonate (mmol.l1), lactate (mmol.l1) potassium (mmol.l1), pH, and sodium (mmol.l1) in supernatant, before (Pre) and after (Post) processing. (Results rounded to either 2 or 3 significant figures). Extracellular/total Hb ratio
Bicarbonate**
Lactate**
Pre
Post
Change*
Pre
Post
Pre
0.014 0.020 0.006 0.005 0.006 0.007 0.003 0.004 0.008 0.012 0.006 0.008 0.004 0.004 0.002 0.003 0.006 0.001 0.000 0.004 Means: 0.006
0.009 0.006 0.005 0.004 0.004 0.008 0.004 0.005 0.011 0.006 0.005 0.005 0.004 0.002 0.003 0.003 0.004 0.004 0.003 0.004 0.005
0.004 0.013 0.001 0.000 0.002 0.001 0.001 0.001 0.003 0.006 0.001 0.003 0.000 0.002 0.000 0.000 0.002 0.002 0.003 0.000 0.001
6.6 9.8 9.2 8.8 8.1 4.6 5.4 5.4 >
> > > >
26 24 24 24 24 27 27 29 30 30 30 30 30 21 25 30 24 30 30 30 27
Potassium**
pH***
Post
Pre
Post
Pre
Post
Pre
Post
14.7 13.0 11.1 10.8 10.7 8.7 14.0 19.0 14.6 19.0 17.0 14.9 12.3 10.0 17.0 20.0 18.0 8.8 11.9 13.9 14
> > > > > > > > > > > > > > > > > > > >
14.6 5.0 5.7 7.5 8.9 4.9 6.9 9.9 7.4 4.4 4.3 4.2 2.4 4.2 7.1 8.3 9.1 1.2 2.3 9.2 6.4
6.422 6.487 6.502 6.491 6.502 6.336 6.358 6.302 < 6.3 6.410 6.360 6.372 6.335 6.580 6.447 6.426 6.539 < 6.3 6.332 6.393 6.4
6.456 6.512 6.518 6.524 6.535 6.432 6.381 6.341 6.327 6.393 6.383 6.387 6.344 6.566 6.452 6.441 6.525 6.429 6.385 6.460 6.44
84 84 85 84 82 104 100 102 94 96 99 100 105 83 108 104 83 91 98 77 93
135 141 145 143 142 146 144 131 145 145 145 145 147 143 130 127 125 146 144 143 140
15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15
Sodium**
*p = 0.16; **p < 0.001; ***p = 0.001. 1208
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
Anaesthesia 2014, 69, 1206–1213
Figure 2 Bicarbonate ion concentration before and after processing. Figure 4 Potassium ion concentration before and after processing.
Figure 3 Lactate ion concentration before and after processing. potassium and pH readings were recorded as < 4 mmol1, > 30 mmol1, > 15 mmol1 and < 6.3, respectively. For the purposes of comparison and statistical testing, these were replaced with values of exactly 4 mmol1, 30 mmol1, 15 mmol1 and 6.3, respectively, which increased the likelihood of a type-2 error (failing to show a significant difference when in © 2014 The Association of Anaesthetists of Great Britain and Ireland
Figure 5 pH before and after processing. fact there was one). Non-parametric tests of significance were used for these parameters because this assumption would mean that the data were not normally distributed. 1209
Anaesthesia 2014, 69, 1206–1213
Read et al. | Cell saver conditioning of out-of-date RBCs
bags in which it was initially above the median, it fell in eight.
Discussion
Figure 6 Sodium ion concentration before and after processing. Figure 7 demonstrates that there was no relationship between the concentration of free haemoglobin (before processing) and the storage time of the RBCs. The effect of the cell-saver on the ratio of extracellular-to-total haemoglobin concentration is shown in Fig. 8. The mean change of this ratio was 0.0012. Of the ten bags in which the ratio was initially below the median, it rose in seven. Of the ten
Guidelines for permitted levels of haemolysis for bankstored RBCs in Britain (Appendix 2) stipulate that in more than 75% of bags of RBCs, fewer than 0.8% of the red cells should have undergone haemolysis [7]. In the context of RBCs that have been processed with a cell-saver, the number of cells that have lysed is not of interest: the washing process is able to remove the resulting debris. What is still of interest is the ratio of extracellular haemoglobin (the unwanted fraction) to total haemoglobin. The standard, expressed above as less than 0.8% of RBCs, can equally be expressed as a ratio of extracellular-to-total haemoglobin of < 0.008. Guidelines from the US Food and Drugs Administration (Appendix 3) stipulate that survival of RBCs at 24 h after transfusion should be greater than 75% [8]. In one study in 2008, RBCs 25–35 days old only just met this standard, and younger RBCs performed much better [9]. Most of the removal of what the authors of this article termed ‘removal-prone’ RBCs occurred in the first hour after transfusion. It has been suggested by many authors that RBCs in the first part of the maximum permitted storage time are safer than older RBCs, implying that the current maximum permitted storage time may be too long. This has been refuted recently in an observational study in adults [10] and a randomised controlled trial in infants [11]. Three further randomised controlled trials investigating this question are currently under way: the
Figure 7 Extracellular/total haemoglobin ratio in bags of RBCs (before processing) according to number of days of storage. 1210
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
Anaesthesia 2014, 69, 1206–1213
Figure 8 Extracellular/total haemoglobin ratio before (■) and after (□) processing, ranked in order of increasing pre-processing ratio.
Age of Blood Evaluation (ABLE) trial in the resuscitation of critically ill patients; the Red Cell Storage Duration (RECESS) study; and the Red Cell Storage Duration and Outcomes in Cardiac Surgery study. There are thus three types of evidence indicating the relationship between the age of bags of stored RBCs and their toxicity: (1) laboratory studies, with an uncertain relationship between laboratory measurements and clinical outcomes; (2) studies of physiological, biochemical, immunological or haematological changes in transfusion recipients, including the survival of RBCs after transfusion, again with an uncertain relationship between observed variables and clinical outcomes; and (3) epidemiological evidence that looks specifically at clinical outcomes, without concerning itself with mechanisms of toxicity. Current practice standards are based on some of this evidence, but none of it definitively indicates a safe maximum storage time. Rather than simply transfusing stored RBCs, some workers have tried processing (referring to it as ‘rejuvenating’) them before transfusion. In a study that involved washing 35–39 day-old RBCs with a specific mixture designed to replace lost substances and functions, the 24-hour post-transfusion survival value was 80% [12]. Others have processed in-date RBCs with a cellsaver and demonstrated a reduction in the toxicity of the suspension fluid [1–3]. However, this has been © 2014 The Association of Anaesthetists of Great Britain and Ireland
shown to induce some haemolysis, both at the time of processing and (when stored in vitro) for some hours afterwards [13]. A study using RBCs up to 49 days old also found that processing with a cell-saver reduced the toxicity of the suspension fluid, and that some haemolysis occurred, but the number of bags of RBCs was smaller than in the present study, and the authors did not measure the fraction of RBCs that survived [14]. The propensity for RBCs to haemolyse may be assessed using the mechanical fragility index, the haemolysis rate caused by a standardised mechanical stress applied to RBCs. Compared with fresh blood, this is raised in in-date RBCs [15] and in fresh blood processed by a cell-saver [16]. It is higher in in-date RBCs that have been processed by a cell-saver than in unprocessed, in-date RBCs [17]. The analysis of our results of the pH, bicarbonate, lactate and potassium concentrations included a risk of a type-2 error. Despite this fact, the differences were statistically significant for all four parameters, as well as for the sodium results. Therefore, it is reasonable to conclude that after processing with the cellsaver, reductions were seen in concentrations of potassium and lactate ions; sodium ion concentration rose from sub-physiological levels to physiological ones. Bicarbonate ion concentration fell, but pH rose slightly, in keeping with the fall in lactate ion concentration. 1211
Anaesthesia 2014, 69, 1206–1213
The concentration of free haemoglobin was somewhat high in a number of bags, but within the standard for in-date stored RBCs. Processing did not affect this proportion. There was a trend for the free haemoglobin concentration to rise in bags in which the initial concentration was low, and for it to fall when the initial concentration was high. The fact that it rose in some bags suggests that haemolysis occurred during processing, and that the haemoglobin released by this mechanism had been imperfectly removed. It may be that the final free haemoglobin concentration was determined more by the equilibrium of haemolysis and haemoglobin removal than by the initial free haemoglobin concentration. This would account for the ‘reversion towards the mean’ pattern we observed in changes in free haemoglobin concentration during processing. The small changes in volume of bags and the haemoglobin concentration within them imply that this haemolysis made little difference to the total number of RBCs. During the study, it was noted that fluid volumes measured by the cell-saver could be slight overestimates. This was because the rotary pump within the cell-saver sometimes made a small but variable number of extra turns after fluid had been replaced by air in the tubing at the point of the air sensor. The cell-saver is in daily use at our institution, and clinical decisions are made on the basis of its measurements of fluid volumes. We consider these measurement errors to be small. It may explain the fact that in some cases there appeared to be a small increase in the amount of haemoglobin in the bag. Recent studies (published or ongoing) address whether the maximum permitted storage time is appropriate or whether it is too long. This study has shown that RBCs that have been stored for longer than the currently accepted maximum time can be processed by a cell-saver autotransfusion device to improve the safety profile of the fluid in which they are suspended. This process causes haemolysis, but the number of RBCs lost through this is small. It also washes out free haemoglobin. The balance of these two effects produces an acceptably low concentration of free haemoglobin in the suspension fluid. We surmise that washing the RBCs for longer, with a larger volume of wash fluid, may have further 1212
Read et al. | Cell saver conditioning of out-of-date RBCs
reduced the concentrations of toxic chemicals, may have caused fewer RBCs to survive the process, and may have caused more haemolysis, with an unpredictable effect on the free haemoglobin concentration in the suspension fluid. Further work is warranted to investigate the characteristics of the RBCs that remain after processing. We postulate that the removal-prone RBCs would be lysed by the cell-saver, selecting out robust RBCs that remained intact.
Acknowledgements We gratefully acknowledge the help of Christopher Lee, Biomedical Scientist, University Hospital of Wales, for invaluable help in measurement of light absorption for haemoglobin concentration.
Competing interest No external funding and no competing interests declared.
References 1. Knichwitz G, Zahl M, Van Aken H, Semjonow A, Booke M. Intraoperative washing of long-stored packed red blood cells by using an autotransfusion device prevents hyperkalemia. Anesthesia and Analgesia 2002; 95: 324–5. 2. de Vroege R., Wildevuur WR, Muradin JAG, Graves D, van Oeveren W. Washing of stored red blood cells by an autotransfusion device before transfusion. Vox Sanguinis 2007; 92: 130–5. 3. Westphal-Varghese B, Erren M, Westphal M, et al. Processing of stored packed red blood cells using autotransfusion devices decreases potassium and microaggregates: a prospective, randomized, single-blinded in vitro study. Transfusion Medicine 2007; 17: 89–95. 4. Roback JD. Transfusion medicine I: adverse complications of stored blood: vascular effects of the red blood cell storage lesion. Hematology. American Society of Hematology Education Book 2011; 1: 475–9. 5. Yazdanbakhsh K, Bao W, Zhong H. Transfusion medicine I: adverse complications of stored blood: immunoregulatory effects of stored red blood cells. Hematology. American Society of Hematology Education Book 2011; 1: 466–9. 6. D’Alessandro A, Liumbruno G, Grazzini G, Zolla L. Red blood cell storage: the story so far. Blood Transfusion 2010; 8: 82–8. 7. James V. Chapter 8. Specifications for blood components. In: James V, ed. Guidelines for the Blood Transfusion Services in the United Kingdom, 7th edn. Norwich: Her Majesty’s Stationery Office, 2005: 64. 8. Dumont LJ, AuBuchon JP, for the Biomedical Excellence for Safer Transfusion (BEST) Collaborative. Evaluation of proposed FDA criteria for the evaluation of radiolabeled red cell recovery trials. Transfusion 2008; 48: 1053–60. 9. Luten M, Roerdinkholder-Stoelwinder B, Schaap NP, de Grip WJ, Bos HJ, Bosman GJ. Survival of red blood cells © 2014 The Association of Anaesthetists of Great Britain and Ireland
Read et al. | Cell saver conditioning of out-of-date RBCs
10.
11.
12.
13.
14.
15.
16.
17.
after transfusion: a comparison between red cells concentrates of different storage periods. Transfusion 2008; 48: 1478–85. McKenny M, Ryan T, Tate H, Graham B, Young VK, Dowd N. Age of transfused blood is not associated with increased postoperative adverse outcome after cardiac surgery. British Journal of Anaesthesia 2011; 106: 643–9. bert P, Hogan DL, et al. Effect of fresh red Fergusson DA, He blood cell transfusions on clinical outcomes in premature, very low-birth-weight infants: the ARIPI randomized trial. Journal of the American Medical Association 2012; 308: 1443–51. Valeri CR, Gray AD, Cassidy GP, Riordan W, Pivacek LE. The 24hour posttransfusion survival, oxygen transport function, and residual hemolysis of human outdated-rejuvenated red cell concentrates after washing and storage at 4 degrees C for 24 to 72 hours. Transfusion 1984; 24: 323–6. O’Leary MF, Szklarski P, Klein TM, Young PP. Hemolysis of red blood cells after cell washing with different automated technologies: clinical implications in a neonatal cardiac surgery population. Transfusion 2011; 51: 955–60. Smith T, Riley W, Fitzgerald D. In vitro comparison of two different methods of cell washing. Perfusion 2013; 28: 34–7. Raval JS, Waters JH, Seltsam A, et al. The use of the mechanical fragility test in evaluating sublethal RBC injury during storage. Vox Sanguinis 2010; 99: 325–31. Ley JT, Yazer MH, Waters JH. Hemolysis and red blood cell mechanical fragility in shed blood after total knee arthroplasty. Transfusion 2012; 52: 34–8. Harm SK, Raval JS, Cramer J, Waters JH, Yazer MH. Haemolysis and sublethal injury of RBCs after routine blood bank manipulations. Transfusion Medicine 2012; 22: 181–5.
Appendix 1 Calculations of free haemoglobin in the supernatant and of ‘percent haemolysis’ specified in the Welsh Blood Service standard operating procedures.
Anaesthesia 2014, 69, 1206–1213
where Fr-Hb is the haemoglobin concentration of the supernatant, and 576 nm-A, 623 nm-A and 700 nm-A are absorbance measurements at light wavelengths of 576, 623 and 700 nm, respectively. The dilution factor is used to produce a useful range of absorbances measurable at the above wavelengths. % Haemolysis ¼
ðð100 % HCTÞ supernatant HbÞ Total Hb
where HCT is the haematocrit.
Appendix 2 British guidelines for permitted levels of haemolysis for bank-stored RBCs: “Haemolysis measurements on red cell components are performed at the end of the component shelf life. Due to intermittent availability of outdated red cell components, each primary process should be validated to give haemolysis of < 0.8% of the red cell mass at the end of component shelf life in > 75% of components with a minimum of 20 components tested” [7].
Appendix 3 Levels of RBC survival expected by the US Food and Drugs Administration: “Radiolabelling studies should be performed in at least two separate centres (laboratories) with a total of 20–24 healthy donors. The mean recovery at 24 hours for each unit should be > 75% with SD < 9%; and the one sided 95% lower confidence limit for the population proportion of successes > 70%” [8].
Fr-Hb ðmg.100 ml1 Þ ¼ð115 576 nm-AÞ ð102 623 nm-AÞ ð39:1 700 nm-AÞ dilution factor
© 2014 The Association of Anaesthetists of Great Britain and Ireland
1213
