By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Major trauma increases vulnerability to systemic infections due to poorly defined immunosuppressive mechanisms. It confers no evolutionary advantage. Our objective was to develop better biomarkers of post-traumatic immunosuppression (PTI) and to extend our observation that PTI was reversed by anti-coagulated salvaged blood transfusion, in the knowledge that others have shown that non-anti-coagulated (fibrinolysed) salvaged blood was immunosuppressive.
A prospective non-randomized cohort study of patients undergoing primary total knee arthroplasty included 25 who received salvaged blood transfusions collected post-operatively into acid–citrate–dextrose anti-coagulant (ASBT cohort), and 18 non-transfused patients (NSBT cohort). Biomarkers of sterile trauma included haematological values, damage-associated molecular patterns (DAMPs), cytokines and chemokines. Salvaged blood was analysed within 1 and 6 h after commencing collection. Biomarkers were expressed as fold-changes over preoperative values. Certain biomarkers of sterile trauma were common to all 43 patients, including supranormal levels of: interleukin (IL)-6, IL-1-receptor-antagonist, IL-8, heat shock protein-70 and calgranulin-S100-A8/9. Other proinflammatory biomarkers which were subnormal in NSBT became supranormal in ASBT patients, including IL-1β, IL-2, IL-17A, interferon (IFN)-γ, tumour necrosis factor (TNF)-α and annexin-A2. Furthermore, ASBT exhibited subnormal levels of anti-inflammatory biomarkers: IL-4, IL-5, IL-10 and IL-13. Salvaged blood analyses revealed sustained high levels of IL-9, IL-10 and certain DAMPs, including calgranulin-S100-A8/9, alpha-defensin and heat shock proteins 27, 60 and 70. Active synthesis during salvaged blood collection yielded increasingly elevated levels of annexin-A2, IL-1β, Il-1-receptor-antagonist, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-17A, IFN-γ, TNF-α, transforming growth factor (TGF)-β1, monocyte chemotactic protein-1 and macrophage inflammatory protein-1α. Elevated levels of high-mobility group-box protein-1 decreased. In conclusion, we demonstrated that anti-coagulated salvaged blood reversed PTI, and was attributed to immune stimulants generated during salvaged blood collection.
In 1856 Florence Nightingale drew attention to the ‘utter insignificance’ of risk of dying from battle wounds acquired during the Crimean War compared with the subsequent risk of dying from zymotic (infectious) diseases . Since then, improvements in hospital hygiene have reduced combat deaths from infectious disease, but an equivalent phenomenon persists [2, 3]. Sterile trauma due to major surgery or closed injury increases vulnerability to pneumonia, urinary infection, alimentary infection, sepsis and multi-organ failure, and these outcomes are exacerbated by haemorrhage, transfusion and ageing [4-9].
Vulnerability is attributed to suppression of immunity via multiple pathways, whose natures are only partially understood. Ischaemic and necrotic tissues release anti-microbial, anti-inflammatory and tissue regenerative molecules known as damage-associated molecular patterns (DAMPS) . These trigger a ‘genomic storm’ dominated by increased expression of interleukin (IL)-6, IL-1-receptor antagonist (IL-1RA), IL-8 (IL-8/CXCL8) and macrophage chemotactic protein-1 (MCP-1/CCL-2) ; mediators that mobilize neutrophil, monocyte and mesenchymal stromal cell (MSC) traffic [12-15]. IL-6 acts via both the hypothalamic–pituitary–adrenal axis to increase corticosteroid production and hepatocytes to initiate the acute phase response typified by increased activated protein-C . Meanwhile, IL-1RA neutralizes IL-1β .
Both sterile and infected trauma induce the systemic inflammatory response syndrome (SIRS), originally defined by pyrexia, tachycardia, hyperventilation and neutrophilia, the latter responding poorly to pathogens [18, 19]. Accompanying these changes is a decrease in circulating basophil, eosinophil and natural killer precursors (NKp), which further weakens systemic immunity . Sterile SIRS is clinically indistinguishable from infected SIRS, but differences exist in their gene expression profiles .
Multiple mechanisms that suppress systemic immunity which, in moments of crisis, would have conferred no evolutionary advantage other than to assist regeneration by avoiding unrestrained activation of tissue destructive inflammatory cascades. Post-traumatic immunosuppression (PTI) could thus have evolved as a non-adaptive consequence of healing .
Knee arthroplasty offered an opportunity to characterize PTI in terms of post-operative depression of NKp cells. This was exacerbated by allogeneic or pre-deposited autologous blood transfusions. However, reinfusion of unwashed anti-coagulated salvaged blood caused NKp cells to rise above preoperative levels, accompanied by enhanced interferon (IFN)-γ synthesis and macrophage activation, suggesting that PTI had been reversed [20, 23]. This was supported by studies from others showing enhanced post-operative immunity following reinfusion of fresh anti-coagulated autologous blood [24, 25].
In contrast, others showed that non-anti-coagulated salvaged blood conferred no change in blood levels of NK cells, albeit identified by the CD56+CD16+ phenotype using flow cytometry and suppression of immune status [26-28]. Our NKp cell subset would have been negative for CD16, and only measurable by limiting dilutions analysis at levels well below detection by routine flow cytometry [29-31].
In this study we have extended our original work by defining PTI in terms of haematological, DAMP, chemokine and cytokine biomarkers. Our results confirmed our observation that anti-coagulated salvaged blood reverses PTI through immune stimulatory properties acquired ex vivo during its collection.
Materials and methods
This prospective, non-randomized, cohort study was approved by the Southmead Hospital Research Ethics Committee on 9 March 2010 (ref. no. 10/H0102/8). Between 2010 and 2011 a total of 43 patients undergoing primary total knee arthroplasty at Avon Orthopaedic Centre, Southmead Hospital, Bristol, were recruited after giving written consent. Exclusion criteria included revision arthroplasty, pre-existing infections, previous blood transfusion, malignant disease, non-quiescent autoimmune disorders and diabetes. Study patients had the following characteristics: 70% were female and 30% male; median age was 74 (56–86) years; primary diagnosis was osteoarthritis (41), quiescent rheumatoid arthritis (one) or quiescent psoriatic arthritis (one) and median hospital stay was 4 (3–7) days. Non-steroidal anti-inflammatory drugs (NSAIDs) were given preoperatively to seven patients. Study size was determined by our previous observation that 39 of 40 patients who received salvaged blood showed evidence of reversal of PTI. Arthroplasty was performed under tourniquet. Assignment to the study cohorts was determined by the volume of blood drained from the wound site within the first 6 h after surgery. If total fluid volume accumulated within 6 h was greater than 175 ml it was reinfused; lesser volumes and any fluid salvaged beyond 6 h was discarded. No study patient received allogeneic or autologous pre-deposited blood transfusions. Those 25 patients who received autologous salvaged blood were termed the ASBT cohort, the median transfused volume at 6 h after surgery being 360 (175–550) ml. For comparison, 18 patients received no salvaged blood transfusions and were termed the NSBT cohort. Blood samples for analyses were obtained pre- and post-operatively (Fig. 1). Similar to our previous study, all patients had a subcutaneous drain inserted into their wound site which was attached to a Dideco-797 Recovery Device (Sorin Group Ltd, Milan, Italy) through which negative pressure was applied .
Fluid was collected via the drain tube under constant negative pressure (–25 mmHg) applied with a rotary pump. A smaller diameter tube was joined by Y connector to the drain at the wound site exit. This second tube delivered anti-coagulant acid–citrate–dextrose (ACD-A) solution (Fenwal Inc., Lake Zurich, IL, USA) from a reservoir at a rate that yielded a final ratio of salvaged blood to ACD-A solution of 12 : 1. ACD-A formulation was: acid citrate monohydrate, 4·0 g; sodium citrate dihydrate, 11·0 g; glucose monohydrate, 12·25 g; total H2O volume, 500 ml; pH 4·7–5·3. After 6 h, or after total volume reached 550 ml, the blood collection bag contents were reinfused via a 40 μm blood transfusion filter (Dideco Micro-40-Goccia; Sorin Group Ltd), without further washing or manipulation.
Blood sample collection and assay
The time-points for collection and processing of blood samples are summarized in Fig. 1. Venous blood samples were collected on two occasions into cell preparation tubes containing sodium citrate (CPT Vacutainer; Becton Dickenson Biosciences, San Jose, CA, USA), the first, immediately before anaesthesia (preoperative sample) and the second between 2 and 6 days after surgery on the day of discharge from hospital (post-operative sample). With the exception of one patient sampled on day 2 and another on day 6, all others were bled between days 3 and 5 (mean 3·5). Two autologous salvaged blood samples were taken: the first within 1 h of surgery, hereafter termed ‘wound site blood’ (WSB), and the second prior to reinfusion at approximately 6 h after surgery, hereafter termed ‘transfused salvaged blood’ (TSB). Within 20–30 min of sampling all bloods were separated into plasma and peripheral blood mononuclear cells (PBMC). Plasma and PBMC were aliquoted and stored at −80°C. Biomarkers were assayed by enzyme-linked immunosorbent assay (ELISA) and flow cytometric bead array (Supporting information, Fig. S1). An automated robotic liquid handling system (JANUS Automated Workstation; Perkin Elmer, Akron, OH, USA) was used to assay cytokines and chemokines by ELISA, thereby minimizing interassay variations and human error. Details of the biomarkers and their properties and numbers of patients assayed for each are summarized in Table 1.
Inhibits inflammatory cascades ; induces IL-9 production by Tcells 
Readouts from assays were corrected for dilution factors, and expressed as fold-changes over preoperative blood levels. Thus, for each patient, fold-changes were calculated for biomarkers in post-operative blood, WSB and TSB samples. Pooled results were illustrated as median interquartile range (IQR) in box-and-whisker plots on a log10 scale. Greater than 1 and less than 1 indicated supra- and subnormal values, respectively, compared to baseline preoperative levels. Missing values disqualified individual fold-calculations from inclusion in IQRs, with no attempt being made to insert estimates. Missing fold-values for each biomarker are listed in Tables 1 and 2. Individual variations attributed to genetic polymorphisms  were highlighted by plotting post-operative against preoperative values for each patient on a log10/log10 scale (Supporting information, Fig. S2a–d). Similar individual variations were apparent in salvaged blood values. Thus fold-values corrected for variations solely attributable to individual baselines. Analyses were by GraphPad Prism software version 5.0.1. Non-parametric tests were used, including Wilcoxon's test (paired) for preoperative versus post-operative values in NSBT and ASBT; Mann–Whitney U-test (unpaired) for fold-changes in NSBT versus in ASBT; Wilcoxon's test (paired) for comparing WSB and TSB levels with preoperative values and between themselves; and Mann–Whitney U-test (unpaired) to compare fold differences in cytokine production between NSBT and ASBT PBMCs.
There were no differences between the NSBT and ASBT cohorts for age, gender and length of hospital stay (Table 2). For all biomarkers analysed no differences were found that were attributable to gender, diagnosis, NSAIDs and age. No significant correlations were found between fold-values and post-operative day of blood sampling (Supporting information, Table S4c). Individual preoperative biomarker (i.e. baseline) levels differed by several orders of magnitude between patients, but trends in post-operative blood values were consistent across groups when values were expressed as fold-changes over preoperative levels (Supporting information, Fig. S2). Thus, major changes were revealed between NSBT and ASBT cohorts (Figs 2 and 3) and between WSB and TSB samples (Figs 4 and 5). In the ASBT cohort no significant correlations were observed between fold-values and volume of blood salvaged (Supporting information, Table S4a,b). There were no post-operative infections or wound-healing problems in any patient, and no significant differences were observed between ASBT and NSBT cohorts in other clinical outcome measures.
Table 2. Patient characteristics
NSBT and ASBT combined mean-fold change over preop.
Median age in years (range)
Gender (male : female)
Hospital stay median days (range)
Haematological biomarkers: mean ± s.d. (missing values)
*P ≤ 0·05 **P ≤ 0·001 ***P ≤ 0·0001. NSBT = non-salvaged blood transfused; ASBT = autologous salvaged blood transfused; s.d. = standard deviation; RBC = red blood celLS; WBC = white blood cells; s.d. = standard deviation.
Biomarkers of sterile trauma (BST) were divisible into two groups, the first of which were common to all patients, and the second of which were altered significantly after anti-coagulated salvaged blood reinfusion. All patients in the NSBT and ASBT cohorts showed similar post-operative trends in the haematological biomarkers (an exception being NKp cell levels published elsewhere ). These included above-normal levels of neutrophils and monocytes and subnormal levels of lymphocytes, eosinophils, basophils, platelets and red blood cells (Table 2). All patients exhibited increases in the DAMPs, heat shock protein (HSP)-70 and S100A8/9 (Fig. 2a). All patients exhibited increases in the cytokines IL-6 and IL-1RA and in the chemokine IL-8/CXCL8 (Fig. 2b). No changes were detected in post-operative levels of HSP-27, HSP-60, transforming growth factor (TGF)-β1, macrophage inflammatory protein (MIP)-1α/CCL3 and IL-12p70 (Fig. 2a,b). Combined mean fold-changes for all these biomarkers are documented in Fig. 2, and separate means for NSBT and ASBT cohorts are shown in Supporting information, Table S1.
Salvaged blood sensitive biomarkers of sterile trauma
Certain BST that were either supranormal, normal or subnormal in the NSBT cohort were changed significantly in the ASBT cohort. These included the DAMP–annexin-A2, which was normal in the NSBT but above normal in the ASBT cohort (Fig. 3a). The key anti-inflammatory cytokines, IL-4, IL-10 and IL-13, were normal in the NSBT cohort and IL-5 was above normal, whereas in the ASBT cohort they were all subnormal (Fig. 3b). The key proinflammatory cytokines, IL-1β, IL-2, IL-17A, IFN-γ and TNF-α, were subnormal in the NSBT cohort, but markedly supranormal in the ASBT cohort (Fig. 3c). Mean fold-changes [± standard deviations (s.d.)] are summarized separately for NSBT and ASBT cohorts in Fig. 3 (see also Supporting information, Table S2).
Characteristics of salvaged blood
Elevated levels of 22 of 24 biomarkers were observed in salvaged blood when expressed as fold-changes over preoperative levels. Exceptions were IL-13 and IL-5, which remained equivalent to normal preoperative levels (Fig 4). Comparisons between WSB and TSB values allowed biomarkers of the salvaged blood transfusion effect to be divided into those that remained unchanged, termed ‘stable biomarkers’, and those whose levels increased during the collection period, termed ‘dynamic biomarkers’.
Stable biomarkers were assumed to have been synthesized continuously by necrotic tissue within the wound site rather than in the collection bag. These were exemplified by massive and sustained fold-increases in S100-A8/9, α-defensin, HSP-27, HSP-60 and HSP-70 (Fig. 4a), and modest increases in IL-9 (Fig 4b). Mean fold-changes for stable biomarkers are documented in Fig. 4, and separately for WSB and TSB samples in Supporting information, Table S3.
Dynamic biomarkers were assumed to have increased through continuous synthesis within the bag during collection. These were illustrated by comparing fold-increases in TSB with fold-increases in WSB. Thus significant elevation was observed in annexin-A2, which increased further, and HMGB-1, which was initially elevated, decreased (Fig. 5a). The proinflammatory cytokines, IL-1β, IL-2, IL-6, IL-12p70, IL-17A, IFN-γ and TNF-α, were significantly above normal in WSB and increased further in TSB (Fig. 5b). Similarly, the anti-inflammatory cytokines, IL-4, IL-10, IL-1RA and TGF-β1, were above normal in WSB and increased further in TSB (Fig. 5c). The chemokines IL-8/CXCL8, MCP-1/CCL2 and MIP-1α/CCL3 were above normal in the WSB and also increased substantially in TSB (Fig. 5d). Mean fold-changes (± s.d.) are summarized separately for WSB and TSB in Fig. 5 (see also Supporting information, Table S3).
We analysed the haematological, immunological and molecular biomarker profiles of patients undergoing total knee arthroplasty , which offered a clinical model for studying PTI. Previous reports were either multi-centre studies of patients following diverse incidents of trauma  or studies using alternative blood collection procedures [26-28, 33-35]. Major features of PTI included subnormal levels of the proinflammatory cytokines IL-1β, IL-2, IL-17A, IFN-γ and TNF-α, a picture that was reversed dramatically by salvaged blood. Lesser features of PTI included normal levels of anti-inflammatory cytokines IL-4, IL-5, IL-10 and IL-13, but these were reduced to subnormal levels by salvaged blood. Other biomarkers of sterile trauma, including supranormal levels of HSP-70, S100-A8/A9, IL-6, IL-1RA and IL-8, remained unchanged after salvaged blood transfusion.
The size and non-randomized prospective cohort design of this study was dictated by our previous assumption that reinfusion of unwashed salvaged blood was beneficial, and that to deny this benefit by randomizing patients in advance to a non-transfusion group would have been unethical. Notwithstanding the blood processing device manufacturers' advocacy of the need to wash salvaged blood cells before reinfusion to remove inflammatory and immunomodulatory contaminants [33, 36], unwashed salvaged blood conservation is widely practised in arthroplasty, albeit with fibrinolysed blood. This study was not designed to evaluate different salvaging procedures, effects of blood loss or duration of surgery on immune status. Theoretically, randomization would have avoided potential bias due to the extent of post-operative bleeding, but in the event no correlation was observed between biomarker profiles and salvaged blood volume (Supporting information, Table S4a,b). Wide individual variations in preoperative biomarker levels (Supporting information, Fig. S2a–d) were attributable to gene promoter polymorphisms, but we were unable to show that these differences contributed to PTI, as trends expressed as fold-changes were consistent within each patient . Lack of correction for individual variations in studies by others may have reduced sensitivity of analyses [26-28].
Constituents of salvaged blood that were assumed to have originated from surgically induced necrotic and ischaemic tissues were called stable biomarkers, and these included S100 A8/A9, α-defensin, HSP-27, HSP-70, HSP-60 and IL-9. Other constituents, called dynamic biomarkers, increased during collection due to ex-vivo synthesis, and included all three chemokines, all proinflammatory cytokines and the four anti-inflammatory cytokines, IL-4, IL-10, TGF-β and IL-1RA. Some of these constituents may have triggered the systemic response to trauma. Thus, increased DAMPs and chemokines would have accounted for the neutrophilia, monocytosis and pyrexia encompassed within the SIRS tautology; however, our overall findings were more consistent with an anti-inflammatory status throughout [12, 18, 38-40].
Our data could be interpreted according to the following sequence of events. At the wound site, necrotic and ischaemic tissues secreted DAMPs, such as α-defensin, which induced viable cells to synthesize IL-6, IL-8/CXCL8, IL-10, MCP-1/CCL2 and MIP-1α/CCL3 . Simultaneously, HMGB-1 enhanced IL-10 synthesis and recruited immature MSC to the wound site, suppressing major proinflammatory pathways en passant [13, 15, 41]. High levels of calgranulin-S100A8/A9 acted as chemoattractant for leucocytes and as an important regulator of inflammation, whose synthesis is enhanced by IL-17A . The heat shock proteins, HSP-27, HSP-60 and HSP-70, interacted with dendritic cells, monocytes and MSC to induce IL-10 synthesis [42, 43]; furthermore, activation of the HSP gene (HSF1) would have repressed transcription of IL-1β in the wound site . In short, we propose that sterile trauma induced a DAMP-driven anti-inflammatory milieu, reinforced by IL-6, IL-9, IL-10, TGF-β and IL-1RA, whose primary function was to suppress activation of tissue damaging inflammatory cascades, while simultaneously facilitating tissue regeneration by inducing the chemokines IL-8/CXCL8, MCP-1/CCL2 and MIP-1α/CCL3. This sterile trauma induced anti-inflammatory/pro-regenerative paradigm was supported by other endotoxin-free in-vitro models of sterile inflammation [45, 46].
Known pleiotrophy of certain cytokines in salvaged blood, such as IL-17A, IFN-γ and TNF-α, could have facilitated regenerative rather than proinflammatory processes, as they would have facilitated synthesis of the chemokines MCP-1/CCL2, MIP-1α/CCL3 and IL-8/CXCL8 . Moreover, IL-9 and IL-10 would have facilitated wound healing by suppressing autoinflammatory reactions and excessive scarring [48-50].
Dynamic biomarkers in salvaged blood included IL-6, which may function either as a pro- or anti-inflammatory cytokine. When bound to its soluble receptor (sIL-6R), the IL-6/sIL-6R complex triggers proinflammatory cascades via a trans-signalling pathway involving ubiquitous gp130 receptors, but when IL-6 binds directly to its receptor (IL-6R) on cell membranes of hypothalamic–pituitary and hepatic cells it triggers anti-inflammatory cascades . Subsequent studies of sIL-6R levels suggested that IL-6 in post-operative venous blood was mostly unbound, but this observation is awaiting further biochemical confirmation.
During 6 h of collection, cells within salvaged blood were subjected to stress by a combination of negative pressure (–25 mmHg), Ca2+ and Mg2+ depletion, plasma dilution, reduced temperature, surface contact with synthetic materials, hypoxia and proteolysis . Simple ex-vivo dilution of whole blood triggers synthesis of inflammatory mediators, including IL-1β and TNF-α . Ca2+ and Mg2+ depletion enhances production of IL-6, IL-1β, IL-12p70 and other proinflammatory cytokines [53, 54]. In combination, these stresses facilitated synthesis of dynamic biomarkers as exemplified by rapidly increasing IL-6 (from ×193 to ×1559-fold), IL-8/CXCL8 (from ×10 to ×122-fold), MCP-1/CCL2 (from ×6 to ×27-fold) and MIP-1α/CCL3 (from ×6 to ×9-fold). Simulation experiments using normal preoperative venous blood freshly drawn into ACD and incubated under similar conditions in the laboratory for 5 h suggested that normal blood so treated showed increased synthesis of all proinflammatory markers and selected DAMPs (HSP-60, HSP-70 and annexin-A2), with no increase in anti-inflammatory markers or chemokines (Supporting information, Fig. S3). Our observation that IL-1β was synthesized during collection of salvaged blood is consistent with the work of others . IL-1β is a key initiator of several inflammatory cascades, and would have helped to tip the balance from an anti-inflammatory to a proinflammatory cocktail . Both known and unknown molecular structures within this cocktail are currently under investigation using proteomic and functional assays.
Regarding the target sites for these key intermediaries, we showed that PBMC collected post-operatively from NSBT patients and cultured with a typical bacterial endotoxin, namely lipopolysaccharide (LPS), responded by synthesizing anti-inflammatory cytokines IL-10, TGF-β and IL-1RA. In contrast, PBMC from ASBT patients produced no increase in these anti-inflammatory cytokines, implying that PTI was due in part to impaired functional capacity of immune cells against bacteria, and that anti-coagulated salvaged blood transfusion abrogated this effect (Supporting information, Fig. S4). This finding contrasted with observations by others (vide infra) .
In this present study a distinction emerged between common BST and salvaged blood sensitive BST. Whereas common BST were observed in all patients irrespective of reinfusion, salvaged blood sensitive BST were profoundly altered after reinfusion. An unresolved paradox is that certain anti-inflammatory biomarkers, namely HSP-70, S100-A8/9, IL-6 and IL-1RA, co-existed in ASBT patients with the proinflammatory cytokines IL-1β, IL-2, IL-17A, IFN-γ and TNF-α. In this context, the pro/anti-inflammatory paradigm needs to be nuanced to accommodate these complexities. One hypothesis is that salvage blood sensitive BST reflect evolutionarily redundant changes that account for PTI that are amenable to therapeutic intervention without jeopardizing tissue regeneration.
In apparent contrast to our findings, others investigated effects of non-anti-coagulated salvaged blood and demonstrated either no change in immune status or immunosuppression [26-28, 35]. No differences in post-operative NK cell levels were observed, (vide supra) and endotoxin challenge studies implied an immunosuppressive response, typified by increased IL-10 and decreased TNF-α. These differences need to be resolved by future randomized studies to compare the two methods of salvaging blood.
Notwithstanding these differences, we clearly observed enhanced immune status as exemplified by supranormal levels of functionally active NKp cells, increased IFN-γ synthesis and increased activation of macrophages (Supporting information, Fig. S5) . In this study, these changes are shown to be accompanied by increases in the proinflammatory biomarkers annexin-A2, IL-1β, IL-2, IL-17A, IFN-γ and TNF-α, and decreases in the anti-inflammatory biomarkers IL-4, IL-5, IL-10 and IL-13. Future manipulation of immune status using purified derivatives of anti-coagulated salvaged blood may reduce vulnerability to infection after major surgery or injury, but fine-tuning of inflammatory cascades could also risk triggering inflammatory diseases. These and other speculations remain to be evaluated in future clinical follow-up studies.
We thank Professor Matthew Griffin MBBCh, MD, REMEDI, NUI Galway, for valuable comments and discussions. We thank Professor Corrado Santocanale PhD, DSc, National Centre of Biomedical Engineering Science (NCBES), Galway, for help with Perkin Elmer Automated Liquid Handling. We thank Mr Claas Baustian BSc, MSc and Dr Shirley Hanley BSc, MSc, PhD, Flow Cytometry Core Facility, NUI Galway for training in the use of Accuri-C6 flow cytometer. We acknowledge the following funding sources: Institutes of Technology Ireland – TSR Award (TSR/2008/TL13) provided a scholarship for N. I. to complete the award of MSc by Research; Irish Research Council – EMBARK Award (RS/2011/223) provided a scholarship to support N. I.'s PhD studies; Programme for Research in Third level Institutions-5 – Advancing Medicines Programme funded the core facility at NUIG for automatic liquid handling technology and support; Science Foundation Ireland (SFI09/SRC/B1794) provided additional resources for laboratory assays; and Bristol Orthopedic Trust funded resources for carrying out clinical protocols and sample collection. Finally, this work was supported by Science foundation Ireland under grant numbers SFI PI 06/1N.1/B652 (M. D. G.) and SFI09/B1794 (R. C. and M. D. G.) and by a Science Foundation Ireland Stoke's Professorship SFI07/SK/B1233b (R. C.)
B. B. was the principle investigator and originator of this study. B. B., N. I., G. B., A. B. and M. W. designed the study. M. W., G. B., A. B. and S. M. organized patient recruitment and administered the surgical and post-operative treatment protocol. N. I. collected, processed and preserved all blood samples and conducted all laboratory assays. E. O. C. and N. I. developed and optimized automated ELISA assays. B. B., M. H., J. T. and R. C. supervised and guided laboratory studies. J. H was responsible for statistical analysis of data. B. B., R. C. and N. I. interpreted data. B. B., N. I. and R. C. wrote the manuscript and all authors saw and approved the final version of the manuscript. These results form part of NI's submission for awards of MSc and PhD. B. B. and M. H. jointly supervised N. I. for award of MSc by Research at IT Tralee in 2011. B. B. and R. C. are currently jointly supervising N. I. for the award of PhD at NUI Galway.
There were no competing interests by any of the co-authors.