Scandinavian Journal of Clinical & Laboratory Investigation, 2015; 75: 314–318
ORIGINAL ARTICLE
Autotransfusion of a restricted volume of shed mediastinal blood does not affect the haemostatic capacity in patients following cardiac surgery
MELISSA PEDERSEN1, MICHAEL KREMKE1, ANNE-METTE HVAS2 & HANNE B. RAVN1,3 Department of 1Anaesthesia and Intensive Care, 2Clinical Biochemistry, Aarhus University Hospital, Aarhus N and 3Department of Cardiothoracic Anaesthesia, The Heart Centre, Rigshospitalet, Denmark Abstract The aim was to investigate the haemostatic capacity after autotransfusion of shed mediastinal blood in patients following cardiac surgery. Fifteen cardiac surgery patients with a chest tube drainage ⱖ 300 mL blood within the first 6 hours postoperatively were included. The haemostatic capacity was evaluated using whole blood thromboelastometry (ROTEM®), impedance aggregometry (Multiplate®) and conventional coagulation tests. Measurements were carried out in (1) mediastinal blood, and in blood samples obtained, (2) before autotransfusion, and (3) after autotransfusion of mediastinal blood. In shed mediastinal blood, ROTEM® analyses showed reduced clot firmness in the EXTEM (p ⬍ 0.001), INTEM (p ⬍ 0.001), and FIBTEM assay (p ⫽ 0.002). Platelet function and conventional coagulation parameters were significantly impaired (p ⬍ 0.001). However, ROTEM®, platelet function and conventional coagulation tests remained unchanged after autotransfusion. Conclusion: Shed mediastinal blood has a substantially reduced haemostatic capacity, but autotransfusion of an average of 350 mL did not affect the overall haemostatic capacity. Key Words: Blood coagulation, blood coagulation tests, operative blood salvage, platelet function tests, thoracic surgery, thromboelastography
Introduction Transfusion of blood products after cardiac surgery is associated with increased morbidity and mortality [1]. The risks associated with blood transfusion and the scarcity of resources has led to various strategies to reduce the need for blood transfusions. In this respect, autotransfusion of shed mediastinal blood (SMB) is a possible supplement. Previous studies have demonstrated conflicting results regarding the efficacy of SMB in reducing the need for allogeneic blood transfusion [2–6]. However, there has been concern that the use of SMB may increase postoperative bleeding due to transfusion of fibrin degradation products, tissue-type plasminogen activator, and blood containing a low number of platelets [3,7]. In terms of haemostatic capacity, previous studies have mainly focused on measuring concentrations of various coagulation components [5,8].
The aim of the present study was to evaluate the effect of autotransfusion of SMB on the haemostatic capacity by thromboelastometry, platelet function tests, and conventional coagulation assays. Our hypothesis was that autotransfusion of SMB had a negative effect on the haemostatic capacity in cardiac surgery patients.
Methods Study population Patients who underwent elective cardiac surgery with cardiopulmonary bypass (CPB) and a postoperative stay on the cardiac recovery unit were eligible. Fifteen patients were included on different, randomly chosen days during the period November 2010– December 2011. Patients could be included if chest tube drainage volumes were ⱖ 300 mL within the
Correspondence: Hanne B. Ravn, Prof. DMSc, PhD, Department of Cardiothoracic Anaesthesia, The Heart Centre, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. Tel: ⫹ 45 3545 1710. E-mail: [email protected] (Received 1 June 2014 ; accepted 23 November 2014 ) ISSN 0036-5513 print/ISSN 1502-7686 online © 2015 Informa Healthcare DOI: 10.3109/00365513.2014.992943
Autotransfusion and haemostasis
315
first 6 postoperative hours. Exclusion criteria were: contraindications to autotransfusion (systemic infection, endocarditis, haemolysis or malignant disease), vitamin K-antagonist or antiplatelet treatment later than five days prior to surgery, or hypothermia at the time autotransfusion was initiated (core temperature ⬍ 36°C). Patients in need for surgical re-exploration and patients with excessive postoperative chest drain loss, necessitating immediate transfusion of blood products, were also excluded. Clinical data were drawn from the Western Denmark Heart Registry. A registry comprising detailed clinical and demographic data on cardiac surgery patients during the entire perioperative period. Chest drain loss and the amount of autotransfusion were gathered from the patient data management system (Picis Care Suite, Picis, Wakefield, MA, USA). The primary end-point was change in clotting time (CT) in the EXTEM-assay after mediastinal blood had been autotransfused back to the patient. On the basis of data from our laboratory obtained from patients undergoing heart surgery (non-published data), the mean CT was estimated to be 65 sec with a standard deviation of 10 sec. We chose the minimum relevant difference (MIREDIF) to be 9 sec. With a significance level at 5% (2α) and a test power at 90% (1β) we had to include 13 patients. We chose to include 15 patients to be sure to obtain a complete data set. We had to restrict the actual amount of SMB to less than 500 mL, as a greater amount of autotransfused fluid within 30 min might have compromised the patient’s haemodynamics. The study was approved by the Danish Data Protection Agency (Journal Number: 2007-58-0010). We received a confirmation from The Regional Committee on Biomedical Research Ethics, that the present study did not require informed patient consent due to its purely descriptive character, without therapeutic implications and patient discomfort. Patients were treated according to the institution’s standard care and results obtained during the study period did not influence the patient treatment
the right jugular vein. Haemodynamic monitoring was extended to include a pulmonary artery catheter and/or insertion of a transesophageal echo-probe at the discretion of the anaesthetist in charge. Antifibrinolytic therapy was given to all patients, who received 4 g of intravenous tranexamic acid divided into separate doses during surgery. Cardiac surgery was carried out with CPB in normo- or mild hypothermia.The heart-lung machine (Maquet, Hirrlingen, Germany) consisted of either a Quadrox oxygenator with a VHK 2001 venous reservoir (Maquet, Hirrlingen, Germany) or a PrimO2X oxygenator with a Synthesis R-venous reservoir (Sorin Group, Milano, Italy). The choice was made at the discretion of the perfusionist. The priming fluid was 1650 mL Ringer’s Lactate with 5000 IU heparin. The myocardium was protected by intermittent antegrade crystalloid cardioplegia. Prior to CPB, the patients were heparinised with unfractioned heparin (300 IU/kg) to ensure an activated clotting time (ACT) ⬎ 400 sec. At the end of CPB, the effect of heparin was reversed with one mg protamine sulphate per 100 IU heparin. If the aPTT was higher than aPTTpreoperative ⫹ 5 sec, additional protamine sulphate was administered. Before chest closure, between one and three chest tubes were placed in the mediastinum and the pleural cavities. Postoperatively, the cardiotomy reservoir was used for the collection of shed mediastinal blood.
Anaesthesia and surgery
Blood sampling and laboratory investigations
All patients followed the standard procedure with respect to anaesthesia, cardiac surgery and postoperative care. Preoperatively, a catheter was inserted in the right radial artery for blood sampling and blood pressure monitoring. The activated partial thromboplastin time (aPTT) (APTT Kaolin, Stago Diagnostica, Asnieres, France) was determined before the patient received any medication. Anaesthesia was induced and maintained by sufentanil and propofol. Insertion of an endotracheal tube was performed after muscle relaxation with intravenous rocuronium. All patients had a central venous catheter inserted in
Blood samples were obtained three times: (i) from the patient immediately before autotransfusion, (ii) from the SMB, and (iii) from the patient within 5 min after completion of autotransfusion. Blood samples before and after autotransfusion were drawn from the radial artery catheter. Blood samples from SMB were obtained via the infusion tubes during autotransfusion. The first 8 mL of blood in each sample were discarded. For thromboelastometry (ROTEM®, Pentapharm GmbH, Munich, Germany) analyses, tubes containing 3.2% sodium citrate (Terumo, Lueven,
The cardiac recovery section When chest drain loss reached a minimum of 300 mL within 6 h postoperatively, the venous reservoir was connected to infusion tubes with a 40 μm filter (Sorin Group, Milano, Italy or Pall Corp., NY, USA) primed with 100 mL saline. Reinfusion of SMB was initiated by an infusion pump (Baxter Healthcare, Deerfield, USA) with an infusion rate at 1000 mL/hour in the central venous catheter. None of the patients received intravenous infusions potentially affecting coagulation (i.e. colloids, blood products or haemostatic drugs) during autotransfusion.
316
M. Pedersen et al.
Belgium) were used. Four standard assays (the EXTEM®, INTEM®, FIBTEM® and the HEPTEM®) were performed. The INTEM® assay assesses whole blood coagulation after activation via the contact phase by addition of phospholipid, ellagic acid and calcium. INTEM® is sensitive to factor deficiencies of the intrinsic system and heparinization. In the EXTEM® assay initiation of coagulation is achieved by the means of thromboplastin (tissue factor) and calcium to evaluate the external pathways of whole blood coagulation. The FIBTEM® assay consist of activation with thromboplastin. To block platelet function cytochalasin D is added. Thus, the achieved whole blood coagulation profile reflects fibrinogen levels and the ability of fibrinogen to polymerize. The HEPTEM® assay is comparable to the INTEM®, but as heparinase is added, the eventual heparin effect will not be reflected in the whole blood coagulation profile, and comparison of the INTEM® and HEPTEM® reveals whether the patient is heparinized. From the ROTEM® we obtained dynamic parameters of clot initiation (clotting time: CT, s) and clot propagation, such as the time until maximum velocity of clot propagation (t, MaxVel, s). Whole blood clot strength was assessed by evaluation of the maximum clot firmness (MCF, mm ⫻ 100). By the addition of heparinase, the HEPTEM® assay revealed residual heparin if INTEM® clotting time was extended by more than 25% compared to the HEPTEM® measurement. Blood samples for analysis of platelet function were obtained using tubes containing hirudin (MP0600, Dynabyte Medical). Platelet function was assessed by employing multiple electrode platelet aggregometry by the Multiplate® analyzer (Roche Diagnostics, Mannheim, Germany). Analyses were performed using the following agonists: adenosine diphosphate (ADP) 5 μM and 10 μM (Fluka, SigmaAldrich, Denmark), and collagen 1 μg/mL and 2 μg/mL (Horm, Medinor, Nycomed, Austria). Platelet function was expressed as area under the aggregation curve (AUC, AU*min). All measurements were performed within 30–120 min after blood sampling. Blood samples for conventional coagulation analyses were obtained using tubes containing 3.2% sodium citrate (Terumo, Lueven, Belgium). APTT (APTT Kaolin, Stago Diagnostica, Asnieres, France), fibrinogen (functional), thrombin time and prothrombin time (neoplastin, Stago Diagnostica) were determined on the STA-R analyser (Stago Diagnostica, Asnieres, France). In case the thrombin time exceeded 21 sec, the STA-R analyser added protamine sulphate to reverse any potential effect of heparin. The ratio between the thrombin time and this analysis was calculated to confirm the presence of heparin. Factor XIII:Clot and factor VIII:Clot (Hart Biologicals, UK) were also determined (ACL TOP, Instrumentation Laboratory, Barcelona, Spain). For platelet count analysis (Sysmex, 2100-XE, Kobe, Japan), blood was
collected in 3.0 mL tubes containing EDTA (Terumo, Leuven, Belgium). Haemoglobin levels, haematocrit, pH, potassium, base excess, calcium (pH ⫽ 7.4) and creatinine were measured with bedside analysers (ABL7000; Radiometer, Copenhagen, Denmark). Statistics Normality was tested by performing histograms and Q-Q-plot of the parameters. As data were not normally distributed, non-parametric statistical analyses were carried out. Descriptive data are presented as median with and interquartile range. Wilcoxon matched-pairs signed-ranks test were performed to compare values at different time points. P-values less than 0.05 were considered significant. All statistical analyses were performed using GraphPad Prism (GraphPad Software, Inc., CA, USA).
Results Characteristics of the 15 patients and information related to the cardiac surgery are shown in Table I. Seven patients underwent coronary artery bypass grafting (CABG), five patients an aortic valve replacement and the remaining patients a combined valve and CABG procedure (n ⫽ 3). The median amount of SMB autotransfused was 350 mL, which was collected during approximately 4 h. Table II shows the ROTEM® results. Compared to the patient samples, the haemostatic capacity was reduced in the SMB as indicated by an increased clotting time and increased time to maximum velocity in the INTEM assay. Furthermore, clot strength was decreased, indicated by significantly reduced MCF in the EXTEM, INTEM and the FIBTEM. Noteworthy, the FIBTEM assay demonstrated a 50% reduction in MCF. The patients’ haemostatic capacity after autotransfusion of SMB were unchanged in all ROTEM® assays compared to baseline measurements (Table II). Heparin was found in excess in the SMB, as judged by HEPTEM, aPTT, and thrombin time (Tables II and IV). When comparing patient samples after Table I. Characteristics of patients (n ⫽ 15) and surgery-related information. Variables
Median (IQR)
Men; n ⫽ 12 (80%) Weight (kg) Age (years) Surgery duration (min) ECC duration (min) Time from surgery until autotransfusion (min) Autotransfusion amount (mL) Shed blood 4 h after autotransfusion (mL)
80 (71; 93) 65 (55; 73) 191 (168; 240) 99 (77; 129) 262 (220; 364) 350 (300; 400) 145 (90; 225)
IQR, interquartile range; ECC, extra corporal circulation.
Autotransfusion and haemostasis
317
Table II. ROTEM® parameters before and after autotransfusion, and from shed mediastinal blood.
ROTEM®
variables
Shed mediastinal blood
Before autotransfusion Median (IQR)
After autotransfusion
p-value*
Median (IQR)
Median (IQR)
p-value*
EXTEM® Clotting time (s) Max clot firmness (mm) Time to max velocity (s) INTEM® Clotting time (s) Max clot firmness (mm) Time to max velocity (s) FIBTEM® Max clot firmness (mm) HEPTEM® INTEM® CT/HEPTEM® CT
64 (58; 90) 59 (57; 61) 143 (65; 154)
75 (69; 128) 47 (36; 49) 156 (122; 185)
0.07 ⬍ 0.001 0.13
69 (64; 74) 58 (57; 60) 141 (73; 162)
0.88 0.15 0.88
183 (171; 200) 59 (57; 63) 212 (193; 244)
314 (273; 820) 46 (28; 50) 354 (266; 520)
⬍ 0.001 ⬍ 0.001 0.02
194 (180; 203) 59 (57; 63) 225 (210; 236)
0.44 0.34 0.81
12 (8; 15)
6 (3; 6)
1.08 (1.02; 1.13)
0.002 ⬍ 0.001
1.9 (1.5; 4.4)
12 (9; 14)
0.38
1.1 (1.0; 1.1)
0.29
IQR, interquartile range; *compared to before autotransfusion.
SMB demonstrated a significantly decreased haemostatic capacity judged from both ROTEM® analysis and platelet function testing. Furthermore, the FIBTEM assay suggested impaired polymerization and a decreased fibrinogen level. Despite this and slight, but not clinical relevant changes in a few of the standard coagulation tests, autotransfusion of SMB did not alter the whole blood coagulation profile in patients. This may be due to the limited volume of SMB used for autotransfusion in the present study. Vertrees et al. observed impaired coagulation after autotransfusion of 582 mL SMB, indicated by prolonged clot formation time, prolonged clotting time, and decreased maximum clot firmness [9]. Likewise, Martin and co-workers infused an average volume of 722 mL SMB, which was associated with increased postoperative bleeding indicated by an increased drain loss in patients receiving autotransfusion compared to the control group [10]. In contrast, three other studies demonstrated that patients receiving SMB (367–508 mL) were less likely to be transfused with allogeneic red blood cells and had reduced postoperative drain losses than control groups [8,11,12]. Hence, moderate volumes of SMB appear not to harm the haemostatic response in the patients and may prevent the need for allogeneic red blood cell transfusions. However, there is no doubt that administration of large amounts of autotransfusion could potentially lead to a vicious circle with dilution coagulopathy and subsequently more bleeding. At what point the disadvantages outweigh the benefit cannot be concluded from the present study.
autotransfusion of SMB with baseline measurements, we found that this heparin excess only caused a minor increase in thrombin time (p ⬍ 0.05), whereas aPTT and HEPTEM remained unchanged (Table IV). As shown in Table III, platelet function was significantly reduced in SMB compared to platelet function. Similar to ROTEM® analyses, the platelet function remained unchanged in patients after autotransfusion (Table III). The platelet count, fibrinogen, prothrombin time and factor XIII:Clot were significantly reduced in SMB, and aPTT and thrombin time were prolonged (p ⬍ 0.001; Table IV). The platelet count was significantly decreased, thrombin and prothrombin time as well as factor VIII:Clot were significantly increased after autotransfusion of SMB. SMB contained significantly lower levels of haemoglobin and a higher potassium concentration (p ⬍ 0.01). Despite these changes, both the haemoglobin and potassium concentration were unchanged in patients after autotransfusion (Table IV). Likewise, base-excess was significantly lower in SMB, but pH remained unchanged in the patient samples (Table IV). Discussion The present study is the first to evaluate the haemostatic capacity in SMB and after autotransfusion of SMB in cardiac surgery patients using whole blood coagulation analyses and evaluation of platelet function.
Table III. Platelet function measured by Multiplate® aggregometry before, after autotransfusion and from shed mediastinal blood. Area under the curve (AU*min) is indicated.
Agonists ADP 5 mM ADP 10 mM Collagen 1 mg/mL Collagen 2 mg/mL
Before autotransfusion Median (IQR) 671 741 1192 1411
(113; 1001) (243; 1082) (415; 1347) (764; 1772)
Shed mediastinal blood Median (IQR) 14 31 53 132
(0; 33) (0; 89) (16; 230) (53; 539)
IQR, interquartile range; *compared to before autotransfusion.
p-value* ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001
After autotransfusion Median (IQR) 668 843 1265 1338
(190; 981) (252; 1118) (594; 1626) (804; 1848)
p-value* 0.89 0.28 0.24 0.29
318
M. Pedersen et al.
Table IV. Standard coagulation tests and other biochemical analysis before, after autotransfusion and from shed mediastinal blood. Reference interval
Variables (⫻ 109/L)
Platelet count aPTT (s) Fibrinogen (μM) Thrombin time (s) Prothrombin time VIII:Clot (⫻ 103 IU/L) XIII:Clot (⫻ 103 IU/L) Haemoglobin (mM) Haematocrit pH Potassium (mM) Base excess (mM) Calcium (mM)
145–350 25–38 5–12 ⬍ 21 ⬎ 0.80 0.66–1.55 0.61–1.77 7.0–10.5 0.35–0.5 7.36–7.42 3.5–4.6 (⫺ 3)–3 1.18–1.32
Before autotransfusion Median (IQR) 147 34 6.6 19 0.71 1.17 0.69 6.2 0.31 7.35 4.6 ⫺ 2.0 1.17
(113; 174) (31; 39) (5.5; 8.1) (16; 28) (0.59; 0.75) (0.93; 1.42) (0.6; 0.8) (5.9; 7.5) (0.29; 0.37) (7.34; 7.38) (4.4; 4.8) (⫺ 3.6; ⫺ 1) (1.13; 1.2)
Shed mediastinal blood Median (IQR) 69 150 3.1 240 0.48 1.18 0.46 5.3 0.26 7.38 5.0 ⫺ 8.9 1.14
(52; 121) (102; 150) (2.1; 4.8) (137; 240) (0.21; 0.55) (0.94; 1.25) (0.35; 0.55) (4.3; 6.5) (0.25; 0.37) (7.31; 7.41) (4.7; 5.3) (⫺ 9.8; ⫺ 4.2) (1.08; 1.18)
p-value* ⬍ 0.001 0.001 0.001 0.001 0.001 0.43 0.001 0.01 0.01 0.71 0.004 ⬍ 0.001 0.02
After autotransfusion Median (IQR) 139 37 6.6 22 0.72 1.21 0.72 6.2 0.31 7.35 4.6 ⫺ 1.7 1.17
(100; 170) (32; 44) (5.3; 8.5) (18; 36) (0.61; 0.76) (1.03; 1.47) (0.59; 0.83) (5.3; 7.6) (0.29; 0.38) (7.33; 7.38) (4.5; 4.8) (⫺ 3.1; ⫺ 0.8) (1.14; 1.2)
p-value* 0.02 0.06 0.70 0.02 0.04 0.01 0.20 0.03 0.10 0.83 0.10 0.96 0.93
IQR, interquartile range; *compared to before autotransfusion.
Platelet function can be impaired after exposure to tissue or artificial surfaces, i.e. collection from surgical cavities or plastic tubing [13]. In accordance with this, the platelet function in SMB was significantly reduced. The incentive to give SMB is to reduce the need of allogeneic blood transfusions. In our study, the median volume of SMB autotransfused was similar to that of one standard allogeneic red blood cell unit (200–350 mL), which has a haematocrit of 55–80% compared to 26% in the SMB in the present study [13]. This clearly favours allogeneic red blood cells in terms of oxygen-carrying capacity [13]. However, stored red blood cells undergo changes that adversely affect their ability to deliver oxygen to tissue [14]. Red blood cells in SMB, on the other hand, maintain longevity like circulating red blood cells and preserve 2,3 diphoshoglycerate [15,16]. Consequently, the higher quality of red blood cells supports the use of SMB. Haemoglobin levels remained unchanged statistically significantly after autotransfusion, even though SMB held a lower level compared to the baseline, but this was not clinically significant. In hypovolemic patients, the use of SMB can be used in limited amount without compromising haemostasis. To what extent SMB can be given without causing dilution of haemoglobin, coagulation factors and platelets needs to be evaluated in a larger clinical study. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. References [1] Jakobsen CJ, Ryhammer PK, Tang M, Andreasen JJ, Mortensen PE. Transfusion of blood during cardiac surgery is associated with higher long-term mortality in low-risk patients. Eur J Cardiothorac Surg 2012;42:114–20. [2] Marberg H, Jeppsson A, Brandrup-Wognsen G. Postoperative autotransfusion of mediastinal shed blood does not influence haemostasis after elective coronary artery bypass grafting. Eur J Cardiothorac Surg 2010;38:767–72.
[3] de Haan J, Boonstra PW, Monnink SH, Ebels T, van Oeveren W. Retransfusion of suctioned blood during cardiopulmonary bypass impairs hemostasis. Ann Thorac Surg 1995;59:901–7. [4] Westerberg M, Bengtsson A, Jeppsson A. Coronary surgery without cardiotomy suction and autotransfusion reduces the postoperative systemic inflammatory response. Ann Thorac Surg 2004;78:54–9. [5] Body SC, Birmingham J, Parks R, Ley C, Maddi R, Shernan SK, LC Siegel, EP Stover, MN D’Ambra, J Levin, DT Mangano, BD Spiess. Safety and efficacy of shed mediastinal blood transfusion after cardiac surgery: a multicenter observational study. Multicenter Study of Perioperative Ischemia Research Group. J Cardiothorac Vasc Anesth 1999;13:410–6. [6] Cross MH. Autotransfusion in cardiac surgery. Perfusion. 2001;16:391–400. [7] Kongsgaard UE, Hovig T, Brosstad F, Geiran O. Platelets in shed mediastinal blood used for postoperative autotransfusion. Acta Anaesthesiol Scand 1999;37:265–8. [8] Murphy GJ, Allen SM, Unsworth-White J, Lewis CT, Dalrymple-Hay MJ. Safety and efficacy of perioperative cell salvage and autotransfusion after coronary artery bypass grafting: a randomized trial. Ann Thorac Surg 2004;77:1553–9. [9] Vertrees RA, Conti VR, Lick SD, Zwischenberger JB, McDaniel LB, Shulman G. Adverse effects of postoperative infusion of shed mediastinal blood. Ann Thorac Surg 1996;62:717–23. [10] Martin J, Robitaille D, Perrault LP, Pellerin M, Page P, Searle N, Cartier R, Hébert Y, Pelletier LC, Thaler HT, Carrier M. Reinfusion of mediastinal blood after heart surgery. J Thorac Cardiovasc Surg 2000;120:499–504. [11] Sirvinskas E, Veikutiene A, Benetis R, Grybauskas P, Andrejaitiene J, Veikutis V, J Surcus. Influence of early re-infusion of autologous shed mediastinal blood on clinical outcome after cardiac surgery. Perfusion 2007;22:345–52. [12] Folkersen L, Tang M, Grunnet N, Jakobsen CJ. Transfusion of shed mediastinal blood reduces the use of allogenic blood transfusion without increasing complications. Perfusion 2011;26:145–50. [13] Klein HG, Spahn DR, Carson JL. Red blood cell transfusion in clinical practice. Lancet 2007;370:415–26. [14] Gerber DR. Transfusion of packed red blood cells in patients with ischemic heart disease. Crit Care Med 2008;36: 1068–74. [15] Schmidt H, Lund JO, Nielsen SL. Autotransfused shed mediastinal blood has normal erythrocyte survival. Ann Thorac Surg 1996;62:105–8. [16] Schmidt H, Folsgaard S, Mortensen PE, Jensen E. Impact of autotransfusion after coronary artery bypass grafting on oxygen transport. Acta Anaesthesiol Scand 1997;41:995–1001.
Copyright of Scandinavian Journal of Clinical & Laboratory Investigation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Autotransfusion of a restricted volume of shed mediastinal blood does not affect the haemostatic capacity in patients following cardiac surgery
MELISSA PEDERSEN1, MICHAEL KREMKE1, ANNE-METTE HVAS2 & HANNE B. RAVN1,3 Department of 1Anaesthesia and Intensive Care, 2Clinical Biochemistry, Aarhus University Hospital, Aarhus N and 3Department of Cardiothoracic Anaesthesia, The Heart Centre, Rigshospitalet, Denmark Abstract The aim was to investigate the haemostatic capacity after autotransfusion of shed mediastinal blood in patients following cardiac surgery. Fifteen cardiac surgery patients with a chest tube drainage ⱖ 300 mL blood within the first 6 hours postoperatively were included. The haemostatic capacity was evaluated using whole blood thromboelastometry (ROTEM®), impedance aggregometry (Multiplate®) and conventional coagulation tests. Measurements were carried out in (1) mediastinal blood, and in blood samples obtained, (2) before autotransfusion, and (3) after autotransfusion of mediastinal blood. In shed mediastinal blood, ROTEM® analyses showed reduced clot firmness in the EXTEM (p ⬍ 0.001), INTEM (p ⬍ 0.001), and FIBTEM assay (p ⫽ 0.002). Platelet function and conventional coagulation parameters were significantly impaired (p ⬍ 0.001). However, ROTEM®, platelet function and conventional coagulation tests remained unchanged after autotransfusion. Conclusion: Shed mediastinal blood has a substantially reduced haemostatic capacity, but autotransfusion of an average of 350 mL did not affect the overall haemostatic capacity. Key Words: Blood coagulation, blood coagulation tests, operative blood salvage, platelet function tests, thoracic surgery, thromboelastography
Introduction Transfusion of blood products after cardiac surgery is associated with increased morbidity and mortality [1]. The risks associated with blood transfusion and the scarcity of resources has led to various strategies to reduce the need for blood transfusions. In this respect, autotransfusion of shed mediastinal blood (SMB) is a possible supplement. Previous studies have demonstrated conflicting results regarding the efficacy of SMB in reducing the need for allogeneic blood transfusion [2–6]. However, there has been concern that the use of SMB may increase postoperative bleeding due to transfusion of fibrin degradation products, tissue-type plasminogen activator, and blood containing a low number of platelets [3,7]. In terms of haemostatic capacity, previous studies have mainly focused on measuring concentrations of various coagulation components [5,8].
The aim of the present study was to evaluate the effect of autotransfusion of SMB on the haemostatic capacity by thromboelastometry, platelet function tests, and conventional coagulation assays. Our hypothesis was that autotransfusion of SMB had a negative effect on the haemostatic capacity in cardiac surgery patients.
Methods Study population Patients who underwent elective cardiac surgery with cardiopulmonary bypass (CPB) and a postoperative stay on the cardiac recovery unit were eligible. Fifteen patients were included on different, randomly chosen days during the period November 2010– December 2011. Patients could be included if chest tube drainage volumes were ⱖ 300 mL within the
Correspondence: Hanne B. Ravn, Prof. DMSc, PhD, Department of Cardiothoracic Anaesthesia, The Heart Centre, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. Tel: ⫹ 45 3545 1710. E-mail: [email protected] (Received 1 June 2014 ; accepted 23 November 2014 ) ISSN 0036-5513 print/ISSN 1502-7686 online © 2015 Informa Healthcare DOI: 10.3109/00365513.2014.992943
Autotransfusion and haemostasis
315
first 6 postoperative hours. Exclusion criteria were: contraindications to autotransfusion (systemic infection, endocarditis, haemolysis or malignant disease), vitamin K-antagonist or antiplatelet treatment later than five days prior to surgery, or hypothermia at the time autotransfusion was initiated (core temperature ⬍ 36°C). Patients in need for surgical re-exploration and patients with excessive postoperative chest drain loss, necessitating immediate transfusion of blood products, were also excluded. Clinical data were drawn from the Western Denmark Heart Registry. A registry comprising detailed clinical and demographic data on cardiac surgery patients during the entire perioperative period. Chest drain loss and the amount of autotransfusion were gathered from the patient data management system (Picis Care Suite, Picis, Wakefield, MA, USA). The primary end-point was change in clotting time (CT) in the EXTEM-assay after mediastinal blood had been autotransfused back to the patient. On the basis of data from our laboratory obtained from patients undergoing heart surgery (non-published data), the mean CT was estimated to be 65 sec with a standard deviation of 10 sec. We chose the minimum relevant difference (MIREDIF) to be 9 sec. With a significance level at 5% (2α) and a test power at 90% (1β) we had to include 13 patients. We chose to include 15 patients to be sure to obtain a complete data set. We had to restrict the actual amount of SMB to less than 500 mL, as a greater amount of autotransfused fluid within 30 min might have compromised the patient’s haemodynamics. The study was approved by the Danish Data Protection Agency (Journal Number: 2007-58-0010). We received a confirmation from The Regional Committee on Biomedical Research Ethics, that the present study did not require informed patient consent due to its purely descriptive character, without therapeutic implications and patient discomfort. Patients were treated according to the institution’s standard care and results obtained during the study period did not influence the patient treatment
the right jugular vein. Haemodynamic monitoring was extended to include a pulmonary artery catheter and/or insertion of a transesophageal echo-probe at the discretion of the anaesthetist in charge. Antifibrinolytic therapy was given to all patients, who received 4 g of intravenous tranexamic acid divided into separate doses during surgery. Cardiac surgery was carried out with CPB in normo- or mild hypothermia.The heart-lung machine (Maquet, Hirrlingen, Germany) consisted of either a Quadrox oxygenator with a VHK 2001 venous reservoir (Maquet, Hirrlingen, Germany) or a PrimO2X oxygenator with a Synthesis R-venous reservoir (Sorin Group, Milano, Italy). The choice was made at the discretion of the perfusionist. The priming fluid was 1650 mL Ringer’s Lactate with 5000 IU heparin. The myocardium was protected by intermittent antegrade crystalloid cardioplegia. Prior to CPB, the patients were heparinised with unfractioned heparin (300 IU/kg) to ensure an activated clotting time (ACT) ⬎ 400 sec. At the end of CPB, the effect of heparin was reversed with one mg protamine sulphate per 100 IU heparin. If the aPTT was higher than aPTTpreoperative ⫹ 5 sec, additional protamine sulphate was administered. Before chest closure, between one and three chest tubes were placed in the mediastinum and the pleural cavities. Postoperatively, the cardiotomy reservoir was used for the collection of shed mediastinal blood.
Anaesthesia and surgery
Blood sampling and laboratory investigations
All patients followed the standard procedure with respect to anaesthesia, cardiac surgery and postoperative care. Preoperatively, a catheter was inserted in the right radial artery for blood sampling and blood pressure monitoring. The activated partial thromboplastin time (aPTT) (APTT Kaolin, Stago Diagnostica, Asnieres, France) was determined before the patient received any medication. Anaesthesia was induced and maintained by sufentanil and propofol. Insertion of an endotracheal tube was performed after muscle relaxation with intravenous rocuronium. All patients had a central venous catheter inserted in
Blood samples were obtained three times: (i) from the patient immediately before autotransfusion, (ii) from the SMB, and (iii) from the patient within 5 min after completion of autotransfusion. Blood samples before and after autotransfusion were drawn from the radial artery catheter. Blood samples from SMB were obtained via the infusion tubes during autotransfusion. The first 8 mL of blood in each sample were discarded. For thromboelastometry (ROTEM®, Pentapharm GmbH, Munich, Germany) analyses, tubes containing 3.2% sodium citrate (Terumo, Lueven,
The cardiac recovery section When chest drain loss reached a minimum of 300 mL within 6 h postoperatively, the venous reservoir was connected to infusion tubes with a 40 μm filter (Sorin Group, Milano, Italy or Pall Corp., NY, USA) primed with 100 mL saline. Reinfusion of SMB was initiated by an infusion pump (Baxter Healthcare, Deerfield, USA) with an infusion rate at 1000 mL/hour in the central venous catheter. None of the patients received intravenous infusions potentially affecting coagulation (i.e. colloids, blood products or haemostatic drugs) during autotransfusion.
316
M. Pedersen et al.
Belgium) were used. Four standard assays (the EXTEM®, INTEM®, FIBTEM® and the HEPTEM®) were performed. The INTEM® assay assesses whole blood coagulation after activation via the contact phase by addition of phospholipid, ellagic acid and calcium. INTEM® is sensitive to factor deficiencies of the intrinsic system and heparinization. In the EXTEM® assay initiation of coagulation is achieved by the means of thromboplastin (tissue factor) and calcium to evaluate the external pathways of whole blood coagulation. The FIBTEM® assay consist of activation with thromboplastin. To block platelet function cytochalasin D is added. Thus, the achieved whole blood coagulation profile reflects fibrinogen levels and the ability of fibrinogen to polymerize. The HEPTEM® assay is comparable to the INTEM®, but as heparinase is added, the eventual heparin effect will not be reflected in the whole blood coagulation profile, and comparison of the INTEM® and HEPTEM® reveals whether the patient is heparinized. From the ROTEM® we obtained dynamic parameters of clot initiation (clotting time: CT, s) and clot propagation, such as the time until maximum velocity of clot propagation (t, MaxVel, s). Whole blood clot strength was assessed by evaluation of the maximum clot firmness (MCF, mm ⫻ 100). By the addition of heparinase, the HEPTEM® assay revealed residual heparin if INTEM® clotting time was extended by more than 25% compared to the HEPTEM® measurement. Blood samples for analysis of platelet function were obtained using tubes containing hirudin (MP0600, Dynabyte Medical). Platelet function was assessed by employing multiple electrode platelet aggregometry by the Multiplate® analyzer (Roche Diagnostics, Mannheim, Germany). Analyses were performed using the following agonists: adenosine diphosphate (ADP) 5 μM and 10 μM (Fluka, SigmaAldrich, Denmark), and collagen 1 μg/mL and 2 μg/mL (Horm, Medinor, Nycomed, Austria). Platelet function was expressed as area under the aggregation curve (AUC, AU*min). All measurements were performed within 30–120 min after blood sampling. Blood samples for conventional coagulation analyses were obtained using tubes containing 3.2% sodium citrate (Terumo, Lueven, Belgium). APTT (APTT Kaolin, Stago Diagnostica, Asnieres, France), fibrinogen (functional), thrombin time and prothrombin time (neoplastin, Stago Diagnostica) were determined on the STA-R analyser (Stago Diagnostica, Asnieres, France). In case the thrombin time exceeded 21 sec, the STA-R analyser added protamine sulphate to reverse any potential effect of heparin. The ratio between the thrombin time and this analysis was calculated to confirm the presence of heparin. Factor XIII:Clot and factor VIII:Clot (Hart Biologicals, UK) were also determined (ACL TOP, Instrumentation Laboratory, Barcelona, Spain). For platelet count analysis (Sysmex, 2100-XE, Kobe, Japan), blood was
collected in 3.0 mL tubes containing EDTA (Terumo, Leuven, Belgium). Haemoglobin levels, haematocrit, pH, potassium, base excess, calcium (pH ⫽ 7.4) and creatinine were measured with bedside analysers (ABL7000; Radiometer, Copenhagen, Denmark). Statistics Normality was tested by performing histograms and Q-Q-plot of the parameters. As data were not normally distributed, non-parametric statistical analyses were carried out. Descriptive data are presented as median with and interquartile range. Wilcoxon matched-pairs signed-ranks test were performed to compare values at different time points. P-values less than 0.05 were considered significant. All statistical analyses were performed using GraphPad Prism (GraphPad Software, Inc., CA, USA).
Results Characteristics of the 15 patients and information related to the cardiac surgery are shown in Table I. Seven patients underwent coronary artery bypass grafting (CABG), five patients an aortic valve replacement and the remaining patients a combined valve and CABG procedure (n ⫽ 3). The median amount of SMB autotransfused was 350 mL, which was collected during approximately 4 h. Table II shows the ROTEM® results. Compared to the patient samples, the haemostatic capacity was reduced in the SMB as indicated by an increased clotting time and increased time to maximum velocity in the INTEM assay. Furthermore, clot strength was decreased, indicated by significantly reduced MCF in the EXTEM, INTEM and the FIBTEM. Noteworthy, the FIBTEM assay demonstrated a 50% reduction in MCF. The patients’ haemostatic capacity after autotransfusion of SMB were unchanged in all ROTEM® assays compared to baseline measurements (Table II). Heparin was found in excess in the SMB, as judged by HEPTEM, aPTT, and thrombin time (Tables II and IV). When comparing patient samples after Table I. Characteristics of patients (n ⫽ 15) and surgery-related information. Variables
Median (IQR)
Men; n ⫽ 12 (80%) Weight (kg) Age (years) Surgery duration (min) ECC duration (min) Time from surgery until autotransfusion (min) Autotransfusion amount (mL) Shed blood 4 h after autotransfusion (mL)
80 (71; 93) 65 (55; 73) 191 (168; 240) 99 (77; 129) 262 (220; 364) 350 (300; 400) 145 (90; 225)
IQR, interquartile range; ECC, extra corporal circulation.
Autotransfusion and haemostasis
317
Table II. ROTEM® parameters before and after autotransfusion, and from shed mediastinal blood.
ROTEM®
variables
Shed mediastinal blood
Before autotransfusion Median (IQR)
After autotransfusion
p-value*
Median (IQR)
Median (IQR)
p-value*
EXTEM® Clotting time (s) Max clot firmness (mm) Time to max velocity (s) INTEM® Clotting time (s) Max clot firmness (mm) Time to max velocity (s) FIBTEM® Max clot firmness (mm) HEPTEM® INTEM® CT/HEPTEM® CT
64 (58; 90) 59 (57; 61) 143 (65; 154)
75 (69; 128) 47 (36; 49) 156 (122; 185)
0.07 ⬍ 0.001 0.13
69 (64; 74) 58 (57; 60) 141 (73; 162)
0.88 0.15 0.88
183 (171; 200) 59 (57; 63) 212 (193; 244)
314 (273; 820) 46 (28; 50) 354 (266; 520)
⬍ 0.001 ⬍ 0.001 0.02
194 (180; 203) 59 (57; 63) 225 (210; 236)
0.44 0.34 0.81
12 (8; 15)
6 (3; 6)
1.08 (1.02; 1.13)
0.002 ⬍ 0.001
1.9 (1.5; 4.4)
12 (9; 14)
0.38
1.1 (1.0; 1.1)
0.29
IQR, interquartile range; *compared to before autotransfusion.
SMB demonstrated a significantly decreased haemostatic capacity judged from both ROTEM® analysis and platelet function testing. Furthermore, the FIBTEM assay suggested impaired polymerization and a decreased fibrinogen level. Despite this and slight, but not clinical relevant changes in a few of the standard coagulation tests, autotransfusion of SMB did not alter the whole blood coagulation profile in patients. This may be due to the limited volume of SMB used for autotransfusion in the present study. Vertrees et al. observed impaired coagulation after autotransfusion of 582 mL SMB, indicated by prolonged clot formation time, prolonged clotting time, and decreased maximum clot firmness [9]. Likewise, Martin and co-workers infused an average volume of 722 mL SMB, which was associated with increased postoperative bleeding indicated by an increased drain loss in patients receiving autotransfusion compared to the control group [10]. In contrast, three other studies demonstrated that patients receiving SMB (367–508 mL) were less likely to be transfused with allogeneic red blood cells and had reduced postoperative drain losses than control groups [8,11,12]. Hence, moderate volumes of SMB appear not to harm the haemostatic response in the patients and may prevent the need for allogeneic red blood cell transfusions. However, there is no doubt that administration of large amounts of autotransfusion could potentially lead to a vicious circle with dilution coagulopathy and subsequently more bleeding. At what point the disadvantages outweigh the benefit cannot be concluded from the present study.
autotransfusion of SMB with baseline measurements, we found that this heparin excess only caused a minor increase in thrombin time (p ⬍ 0.05), whereas aPTT and HEPTEM remained unchanged (Table IV). As shown in Table III, platelet function was significantly reduced in SMB compared to platelet function. Similar to ROTEM® analyses, the platelet function remained unchanged in patients after autotransfusion (Table III). The platelet count, fibrinogen, prothrombin time and factor XIII:Clot were significantly reduced in SMB, and aPTT and thrombin time were prolonged (p ⬍ 0.001; Table IV). The platelet count was significantly decreased, thrombin and prothrombin time as well as factor VIII:Clot were significantly increased after autotransfusion of SMB. SMB contained significantly lower levels of haemoglobin and a higher potassium concentration (p ⬍ 0.01). Despite these changes, both the haemoglobin and potassium concentration were unchanged in patients after autotransfusion (Table IV). Likewise, base-excess was significantly lower in SMB, but pH remained unchanged in the patient samples (Table IV). Discussion The present study is the first to evaluate the haemostatic capacity in SMB and after autotransfusion of SMB in cardiac surgery patients using whole blood coagulation analyses and evaluation of platelet function.
Table III. Platelet function measured by Multiplate® aggregometry before, after autotransfusion and from shed mediastinal blood. Area under the curve (AU*min) is indicated.
Agonists ADP 5 mM ADP 10 mM Collagen 1 mg/mL Collagen 2 mg/mL
Before autotransfusion Median (IQR) 671 741 1192 1411
(113; 1001) (243; 1082) (415; 1347) (764; 1772)
Shed mediastinal blood Median (IQR) 14 31 53 132
(0; 33) (0; 89) (16; 230) (53; 539)
IQR, interquartile range; *compared to before autotransfusion.
p-value* ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001
After autotransfusion Median (IQR) 668 843 1265 1338
(190; 981) (252; 1118) (594; 1626) (804; 1848)
p-value* 0.89 0.28 0.24 0.29
318
M. Pedersen et al.
Table IV. Standard coagulation tests and other biochemical analysis before, after autotransfusion and from shed mediastinal blood. Reference interval
Variables (⫻ 109/L)
Platelet count aPTT (s) Fibrinogen (μM) Thrombin time (s) Prothrombin time VIII:Clot (⫻ 103 IU/L) XIII:Clot (⫻ 103 IU/L) Haemoglobin (mM) Haematocrit pH Potassium (mM) Base excess (mM) Calcium (mM)
145–350 25–38 5–12 ⬍ 21 ⬎ 0.80 0.66–1.55 0.61–1.77 7.0–10.5 0.35–0.5 7.36–7.42 3.5–4.6 (⫺ 3)–3 1.18–1.32
Before autotransfusion Median (IQR) 147 34 6.6 19 0.71 1.17 0.69 6.2 0.31 7.35 4.6 ⫺ 2.0 1.17
(113; 174) (31; 39) (5.5; 8.1) (16; 28) (0.59; 0.75) (0.93; 1.42) (0.6; 0.8) (5.9; 7.5) (0.29; 0.37) (7.34; 7.38) (4.4; 4.8) (⫺ 3.6; ⫺ 1) (1.13; 1.2)
Shed mediastinal blood Median (IQR) 69 150 3.1 240 0.48 1.18 0.46 5.3 0.26 7.38 5.0 ⫺ 8.9 1.14
(52; 121) (102; 150) (2.1; 4.8) (137; 240) (0.21; 0.55) (0.94; 1.25) (0.35; 0.55) (4.3; 6.5) (0.25; 0.37) (7.31; 7.41) (4.7; 5.3) (⫺ 9.8; ⫺ 4.2) (1.08; 1.18)
p-value* ⬍ 0.001 0.001 0.001 0.001 0.001 0.43 0.001 0.01 0.01 0.71 0.004 ⬍ 0.001 0.02
After autotransfusion Median (IQR) 139 37 6.6 22 0.72 1.21 0.72 6.2 0.31 7.35 4.6 ⫺ 1.7 1.17
(100; 170) (32; 44) (5.3; 8.5) (18; 36) (0.61; 0.76) (1.03; 1.47) (0.59; 0.83) (5.3; 7.6) (0.29; 0.38) (7.33; 7.38) (4.5; 4.8) (⫺ 3.1; ⫺ 0.8) (1.14; 1.2)
p-value* 0.02 0.06 0.70 0.02 0.04 0.01 0.20 0.03 0.10 0.83 0.10 0.96 0.93
IQR, interquartile range; *compared to before autotransfusion.
Platelet function can be impaired after exposure to tissue or artificial surfaces, i.e. collection from surgical cavities or plastic tubing [13]. In accordance with this, the platelet function in SMB was significantly reduced. The incentive to give SMB is to reduce the need of allogeneic blood transfusions. In our study, the median volume of SMB autotransfused was similar to that of one standard allogeneic red blood cell unit (200–350 mL), which has a haematocrit of 55–80% compared to 26% in the SMB in the present study [13]. This clearly favours allogeneic red blood cells in terms of oxygen-carrying capacity [13]. However, stored red blood cells undergo changes that adversely affect their ability to deliver oxygen to tissue [14]. Red blood cells in SMB, on the other hand, maintain longevity like circulating red blood cells and preserve 2,3 diphoshoglycerate [15,16]. Consequently, the higher quality of red blood cells supports the use of SMB. Haemoglobin levels remained unchanged statistically significantly after autotransfusion, even though SMB held a lower level compared to the baseline, but this was not clinically significant. In hypovolemic patients, the use of SMB can be used in limited amount without compromising haemostasis. To what extent SMB can be given without causing dilution of haemoglobin, coagulation factors and platelets needs to be evaluated in a larger clinical study. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. References [1] Jakobsen CJ, Ryhammer PK, Tang M, Andreasen JJ, Mortensen PE. Transfusion of blood during cardiac surgery is associated with higher long-term mortality in low-risk patients. Eur J Cardiothorac Surg 2012;42:114–20. [2] Marberg H, Jeppsson A, Brandrup-Wognsen G. Postoperative autotransfusion of mediastinal shed blood does not influence haemostasis after elective coronary artery bypass grafting. Eur J Cardiothorac Surg 2010;38:767–72.
[3] de Haan J, Boonstra PW, Monnink SH, Ebels T, van Oeveren W. Retransfusion of suctioned blood during cardiopulmonary bypass impairs hemostasis. Ann Thorac Surg 1995;59:901–7. [4] Westerberg M, Bengtsson A, Jeppsson A. Coronary surgery without cardiotomy suction and autotransfusion reduces the postoperative systemic inflammatory response. Ann Thorac Surg 2004;78:54–9. [5] Body SC, Birmingham J, Parks R, Ley C, Maddi R, Shernan SK, LC Siegel, EP Stover, MN D’Ambra, J Levin, DT Mangano, BD Spiess. Safety and efficacy of shed mediastinal blood transfusion after cardiac surgery: a multicenter observational study. Multicenter Study of Perioperative Ischemia Research Group. J Cardiothorac Vasc Anesth 1999;13:410–6. [6] Cross MH. Autotransfusion in cardiac surgery. Perfusion. 2001;16:391–400. [7] Kongsgaard UE, Hovig T, Brosstad F, Geiran O. Platelets in shed mediastinal blood used for postoperative autotransfusion. Acta Anaesthesiol Scand 1999;37:265–8. [8] Murphy GJ, Allen SM, Unsworth-White J, Lewis CT, Dalrymple-Hay MJ. Safety and efficacy of perioperative cell salvage and autotransfusion after coronary artery bypass grafting: a randomized trial. Ann Thorac Surg 2004;77:1553–9. [9] Vertrees RA, Conti VR, Lick SD, Zwischenberger JB, McDaniel LB, Shulman G. Adverse effects of postoperative infusion of shed mediastinal blood. Ann Thorac Surg 1996;62:717–23. [10] Martin J, Robitaille D, Perrault LP, Pellerin M, Page P, Searle N, Cartier R, Hébert Y, Pelletier LC, Thaler HT, Carrier M. Reinfusion of mediastinal blood after heart surgery. J Thorac Cardiovasc Surg 2000;120:499–504. [11] Sirvinskas E, Veikutiene A, Benetis R, Grybauskas P, Andrejaitiene J, Veikutis V, J Surcus. Influence of early re-infusion of autologous shed mediastinal blood on clinical outcome after cardiac surgery. Perfusion 2007;22:345–52. [12] Folkersen L, Tang M, Grunnet N, Jakobsen CJ. Transfusion of shed mediastinal blood reduces the use of allogenic blood transfusion without increasing complications. Perfusion 2011;26:145–50. [13] Klein HG, Spahn DR, Carson JL. Red blood cell transfusion in clinical practice. Lancet 2007;370:415–26. [14] Gerber DR. Transfusion of packed red blood cells in patients with ischemic heart disease. Crit Care Med 2008;36: 1068–74. [15] Schmidt H, Lund JO, Nielsen SL. Autotransfused shed mediastinal blood has normal erythrocyte survival. Ann Thorac Surg 1996;62:105–8. [16] Schmidt H, Folsgaard S, Mortensen PE, Jensen E. Impact of autotransfusion after coronary artery bypass grafting on oxygen transport. Acta Anaesthesiol Scand 1997;41:995–1001.
Copyright of Scandinavian Journal of Clinical & Laboratory Investigation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
