| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1996;62:717-723
© 1996 The Society of Thoracic Surgeons
Departments of Surgery, Anesthesiology, and Pathology, The University of Texas Medical Branch, Galveston, Texas
Accepted for publication April 16, 1996.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. In a prospective, randomized study, 40 patients undergoing coronary artery bypass grafting with shed mediastinal blood collected in a cardiotomy reservoir were divided into two equal groups and studied during their first 4 hours in the intensive care unit. Shed mediastinal blood was directly infused in group I (n = 20), whereas in group II (n = 20), it was not. In group II, if a sufficient volume of red cells was present to allow processing (n = 5), washed red cells were infused. Variables studied before and after infusion were the amount of blood lost and infused, homologous blood transfused, complete blood count and differential, serum fibrinogen, fibrin split products, D-dimers, clotting factors, prothrombin time, activated partial thromboplastin time, thromboelastograms, plasma-free hemoglobin, complement factors C3 and C4, creatine kinase and its MB isoenzyme, and body temperature.
Results. After infusion of shed mediastinal blood, elevated levels of fibrin split products and D-dimers were found in significantly more patients in group I. The thromboelastogram index was normal in 76% of patients in group II but in only 12.5% in group I. Group I also had an increase in band neutrophils, a greater number of febrile patients, higher serum levels of creatine kinase, its MB isoenzyme, and plasma-free hemoglobin, and greater blood loss during hours 3, 4, and 5 in the intensive care unit. The volume of red cells in shed mediastinal blood (hematocrit, 9% to 10%) was small, resulting in clinically insignificant autotransfusion when infused directly, and insufficient for cell processing in most patients.
Conclusions. These data support those in previous studies that direct infusion of shed mediastinal blood does not save substantial amounts of autologous red cells and can cause a delayed coagulopathy and other adverse effects that may be harmful to patients postoperatively.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In an effort to decrease exposure to homologous blood transfusions, intraoperative salvage and postoperative collection and infusion of shed mediastinal blood (SMB) are now used by many surgeons. Schaff and associates [6] initially described a technique for collection and direct infusion of postoperative SMB and demonstrated that this blood was defibrinogenated but contained significantly higher concentrations of platelets and clotting factors than did homologous banked blood. This group also showed a significant reduction in the use of banked blood in patients who received SMB. Hartz and co-workers [7] found that SMB was defibrinated but, when autotransfused, resulted in significant savings of banked blood without evidence of fibrinolysis or disseminated intravascular coagulation. Cosgrove and colleagues [8] modified this approach by using the cardiotomy reservoir as the collection chamber. Several other studies [9–13] have shown that collected and infused SMB lowers the number of banked blood transfusions required. We routinely employed SMB infusion (as described by Cosgrove and co-workers [8]) for 8 years prior to this study.
More recent studies [14–16] have not shown a clear benefit from direct infusion of SMB and indeed have suggested that postoperative bleeding is increased because of SMB-induced coagulopathy. The elevated level of fibrin split products (FSP) found in patients after infusion of SMB is, in part, accounted for by the high levels in the shed blood that is infused but may also indicate active fibrinolysis; this may explain the observation that larger volumes of SMB infusion have been associated with increased delayed bleeding [17]. The purpose of the current study was to determine whether direct infusion of SMB caused a coagulopathy or other adverse effects.
To better characterize the delayed effects of SMB infusion on blood coagulation, we added thromboelastography (TEG) to other standard coagulation studies. The thromboelastogram is a dynamic study of clot development and stabilization and is reported to be a more sensitive indicator of fibrinolysis and platelet function [18]. This may help determine whether the elevated levels of fibrinolytic products observed after SMB infusion indicate presence of a coagulopathy [19].
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Our study group consisted of 32 men and 8 women (average age, 59 ± 10 years; range, 35 to 79 years). All were operated on by the same surgeon (V.R.C.). The 40 patients were divided by a random number generator into two groups of 20 patients each. The group assignment was blinded to all participants except the study coordinator until arrival in the intensive care unit (ICU).
Group I had our "routine" postoperative care of direct infusion of SMB collected in a cardiotomy reservoir (used during cardiopulmonary bypass) with an intrinsic underwater seal and suction regulatory system (CAP 35F; Gish Labs, Inc, Santa Ana, CA). Negative pressure of 20 cm H2O was applied to the reservoir, and the collected blood was infused into the patient through a 40-µm filter (SQ40S; Pall, Inc, Glen Cove, NY) by a standard infusion pump. The blood product infused thus consisted of filtered blood and pleural space and mediastinal fluid, with our protocol calling for an infusion rate that would use nearly the entire reservoir content over the subsequent hour.
The group II patients had the same collection system in place, but no SMB was directly infused. Shed blood was held in the reservoir for 3 hours, after which time it was transferred to an infusion bag and sent to the blood bank for washing (Cobe 2991 cell processor; Cobe Laboratories, Inc, Lakewood, CO). If sufficient volume was present to allow processing (>100 mL of red blood cells), the washed packed cells were immediately infused. Thus, the study compares the effects of directly infused SMB (group I) with the results in a group who did not receive any directly infused SMB (group II) and covers the 4-hour interval that started on the patient's arrival in the ICU from the operating room; blood loss data were collected for an additional 2 hours.
Data collection times were the same for both groups. Blood samples were obtained through a central venous line immediately on arrival in the ICU (baseline) and 4 hours after ICU admission after SMB was directly infused (final). Both baseline and final samples were collected through the same line in all patients (central line) after a 10-mL clearing sample was discarded. The cardiotomy blood sample was collected after a thorough mixing and in sufficient volume to allow accurate sampling of the reservoir contents. Patient temperature and pulmonary artery blood temperature (measured through the Swan-Ganz catheter) were recorded at the final time point. Hourly and total amounts of mediastinal blood collected and administered were recorded.
Procedures
Anesthesia was maintained with fentanyl, pancuronium bromide, midazolam hydrochloride, and isoflurane. Cardiopulmonary bypass was performed in all patients using identical equipment (Optima oxygenator model CM30; Cobe Laboratories, Inc), the same perfusion protocol (moderate hypothermia, nasopharyngeal temperature to 28°C), and an asanguineous fluid to prime. Myocardial preservation consisted of cold (8.0°C) hyperkalemic blood crystalloid solution (4:1) delivered antegrade, retrograde, or both. The activated clotting time (Hemochron 800; International Technidyne Corporation, Edison, NJ) was maintained at greater than 500 seconds with subsequent doses of heparin sodium, and, after bypass was terminated, heparin was neutralized with protamine sulfate (1:1 mg ratio with total heparin). The blood remaining in the pump after the completion of cardiopulmonary bypass was hemoconcentrated (Hemoconcentrator; Amicon Corp, Danvers, MA), transferred to an infusion bag, and immediately infused into the patient. All patients had their chest tubes connected to a cardiotomy reservoir for collection of blood shed during the immediate postoperative period. Homologous red blood cell transfusions were administered by protocol when the patient's hematocrit fell to less than 24%. No study patient received antifibrinolytic agents or aprotinin, and no patient had predonated autologous blood for use perioperatively.
Laboratory Procedures
The collected blood was assayed in accordance with normal clinical laboratory procedures for the following hematologic variables: complete blood count and differential including white blood cell count, red blood cell count, hemoglobin content, hematocrit, mean cell volume, mean cell hemoglobin, mean cell hemoglobin concentration, red cell distribution width, platelet count, and mean platelet volume. Coagulation was evaluated by assaying the following: serum fibrinogen; FSP; D-dimers; clotting factors II, V, and VIII (percentage of normal); antithrombin III (percentage of normal); and protein C (percentage of normal). The coagulation cascade was evaluated by the one-stage prothrombin time test and the semiautomated activated partial thromboplastin time.
Further analysis of the coagulation system was done with TEG (Thromboelastograph; Haemoscope, Morton Grove, IL). Thromboelastography was performed on baseline and final whole-blood samples using the thromboelastograph coagulation analyzer (Hemoscope Corporation, Skokie, IL) in accordance with the manufacturer's instruction manual. Four main variables were measured by TEG: reaction time (R time), K time, maximum amplitude (MA), and alpha angle (A). Reaction time measured initiation of the clotting process, which correlates with the whole blood clotting time, and signals the beginning of coagulation. This is the end point of most conventional clinical coagulation assays. K time measures the ability of the blood to form a clot to a fixed level of firmness. Maximum amplitude reflects the maximum shear modulus of the developed clot, measuring its strength mainly on the basis of the contribution of platelets and to a lesser extent the strength of the fibrin. The alpha angle is more comprehensive and measures the rate of clot growth. Together, K time and alpha angle describe polymerization of the structural elements involved in clotting. This describes the rate of clot growth and is related to platelet function and plasma components residing on the platelet surface. The thromboelastogram index (TI) identifies the coagulation state of a patient by the process of discriminant analysis giving weight to each of the measurements of the thromboelastogram in the following equation [20]: TI = -(0.1225)R time + (0.0092)K time + (0.1655)MA - (0.0241)A - 5.0220.
The cardiac enzyme creatine kinase and its MB isoenzyme and lactate dehydrogenase electrophoresis and index were measured by immunoassay (Baxter-Stratus, Miami, FL). Plasma free hemoglobin was determined by a colorimetric method (Sigma Chemical Co, St. Louis, MO). Complement factors C3 and C4 were measured by nephelometry (Beckman Instruments Corp, Brea, CA).
Statistical Analysis
Statistical analysis was accomplished by the software program Sigmastat Version 1.02 (Jandel Scientific Corp, San Rafael, CA). Cardiotomy reservoir samples were compared by Student's t test to assess differences between groups. In the final samples, between-group comparisons were made by t test to assess the effect of the infusion of unwashed SMB on the patient's coagulation profile. Intragroup comparisons of data before and after treatment required the paired t test for comparison. Nonparametric data were analyzed using the Mann-Whitney U test. Differences were considered significant at the 95% confidence level (p < 0.05), taking into account multiple comparisons (Bonferroni). Data are displayed as the mean ± the standard error of the mean.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
. Table 2 shows the amounts of SMB collected (total SMB output) and infused (infused volume) during the 4-hour study period and the average homologous transfusion in the two study groups. There was no difference in the need of red blood cell transfusions between groups, and no patient received any other blood product. Group I had more SMB collected (729 ± 70 mL) than group II (607 ± 61 mL), but the difference was not significant. Three patients (15%) in group I and no patient in group II had more than 1,000 mL of SMB. Although there was no significant difference between groups in the total volume of SMB during the study, Figure 1 indicates that there was a tendency for more mediastinal drainage continuing in group I after 3 hours. Blood lost during each successive hour, postoperative hours 3, 4, and 5, was significantly greater in group I. This was particularly prominent in hour 3 where an output of more than 100 mL/h occurred in 11 patients in group I versus only 2 patients in group II (p = 0.0497 by Fisher's exact test).
|
|
|
Coagulation Variables
Comparisons between groups for the baseline (ICU arrival) coagulation variables showed no significant differences (Table 3). No significant differences between groups were noted for coagulation values determined on the SMB; all values in both groups showed elevated D-dimers, elevated FSP, and very low levels of fibrinogen and clotting factors. All of these SMB reservoir samples were significantly different from the baseline values, findings indicating a significant alteration in the coagulation state of the SMB as a result of the process of bleeding and collection. Few patients in either group had an elevation of FSP (group I, 2 patients; group II, 1) or D-dimers (group I, 4 patients; group II, 3 patients) when they arrived in the ICU, but after SMB infusion, significantly more patients had elevations of these variables in group I (FSP, 15 patients; D-dimers, 16) than in group II (FSP, 0; D-dimers, 2).
|
Thromboelastography
Thromboelastography was performed at the two sampling times of baseline (ICU arrival) and final. Sixteen patients in group I and 17 patients in group II underwent TEG; the remaining 7 patients did not because at the time of their studies, the device was under repair; however, the study continued so as not to alter the patient randomization. Reaction times were significantly prolonged at the final point only in group I (39.4 ± 7.1 versus 28.5 ± 4.7; p < 0.05) (Table 4). K times for both groups were prolonged, with the group I final sample significantly greater than all other values. The baseline alpha angles of both groups were normal, but the final sample was significantly lower in group I. At the final sample, the thromboelastogram index was normal (0 ± 2.0) in 76% of patients in group II (-0.8 ± 0.6) but in only 12.5% in group I (-2.75 ± 0.6; p < 0.05). Collectively, the thromboelastographic results in group I are consistent with abnormal coagulation. Clotting was prolonged and a softer clot was formed after direct infusion of unprocessed SMB.
|
) and size, white blood cell count and polymorphonuclear differential of segmented and band neutrophil distribution, and quantity of monocytes and lymphocytes. At the final comparison, the polymorphonuclear distribution showed the band neutrophil percentage to be significantly higher in group I compared with either the group II (37.0% ± 1.7% versus 21% ± 1.8%; p < 0.05) or the group I baseline value (37.0% ± 1.7% versus 22.6% ± 1.4%; p < 0.05). This reflects a significant shift toward more immature neutrophil forms (bands) and away from the mature (segmented) forms in group I but not in group II as a result of the infused SMB. There were no other differences in hematologic variables found between the two experimental groups at the final sampling time.
Biochemical Variables
As shown in Table 5, baseline biochemical variables showed no significant differences between groups. The samples of SMB from both groups showed marked elevation of enzyme and plasma-free hemoglobin levels compared with normal values. Levels of creatine kinase, its MB isoenzyme, lactate dehydrogenase, and plasma-free hemoglobin were markedly elevated after SMB infusion in group I compared with group II and were significantly higher than on ICU arrival only in group I. No patient had a level of the MB isoenzyme of creatine kinase higher than 55 or clinical findings consistent with perioperative myocardial infarction. These data indicate that unwashed SMB contains elevated levels of serum enzymes and plasma free hemoglobin and that on infusion, this results in elevated circulating blood levels.
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The initial studies that showed substantial savings in the use of homologous blood by SMB infusion comprised patients with substantially higher blood loss than was present in our study. Whereas these studies [6, 7] had SMB volumes of 700 to 2,000 mL with a hematocrit of 20% to 25%, the patients in our study had only about 600 mL of blood that could be infused with an average hematocrit of 11%. Also, most of our patients had the left pleural space drained when an internal mammary artery dissection was performed, thereby no doubt increasing the proportion of drainage that came from this source rather than blood lost from the heart and mediastinum. Therefore, we could see no benefit regarding transfusion requirement or preservation of red cell mass in these patients. Although average hourly blood loss was not excessively high in either group, we did see a definite tendency for increased blood loss after hour 3 only in the group that had direct infusion of SMB (group I). Those advocating the use of SMB infusion point out that to see significant clinical benefit in blood use, one would have to look at patients who lose more than approximately 1 L of blood early after operation [11], which occurred in only 3 patients in this study.
Earlier studies did not find evidence of a coagulopathy caused by the infusion of unaltered SMB. It was shown that SMB contained high levels of FSP and D-dimers, but the elevated levels in serum were thought to be due to the presence of these variables in the infused SMB and not to activation of intrinsic fibrinolysis in the patient [18]. Further, standard clotting studies and platelet function studies did not show significant derangements in earlier trials [23]. Like others, we also found high levels of D-dimers and FSP after the infusion of unwashed SMB and did not see differences in the usual laboratory studies done to identify coagulopathy: activated clotting time, prothrombin time, partial thromboplastin time, concentration of fibrinogen, and platelet counts.
However, our studies using TEG, which may be more sensitive to functionally important derangements in the coagulation mechanism, showed clear evidence of a delayed onset of clot formation and decreased clot strength in the group who had unwashed SMB infused (group I). Because we followed the clot characteristics for only 30 to 45 minutes and understand that to reliably detect fibrinolysis, the thromboelastogram must be followed for at least 60 minutes, we did not see direct evidence of fibrinolysis by TEG. In group I, the thromboelastogram index was less than the lower normal limit, a finding suggestive of evidence of lytic activity; the K time was much more prolonged in group I, and this indicates impaired association of clotting factors; and the maximum amplitude was low, which is suggestive of a platelet-consumptive activity. During all this, the prothrombin time, the partial thromboplastin time, and the platelet count were normal.
Although multifactorial issues may account for these changes, the most likely explanation is excessive fibrinolytic activity, which can result from the increased levels of D-dimers and FSP found in group I as a result of the direct infusion of SMB [24]. Other factors that could have affected these results, such as the dilutional effect after bypass and the persistent heparin effect, would have been present equally in both groups and are unlikely to account for the significant differences observed.
Other potentially deleterious effects observed after infusion of SMB were found. As previously observed by others [25, 26], SMB contains high levels of the enzymes used to determine cardiac injury, and infusion of this blood markedly elevated these levels in patients early after operation, thus making these values less useful for determining perioperative infarction. There are also significant levels of free hemoglobin in SMB, thus exposing the patient to further increases in plasma free hemoglobin with its known adverse effects [27]. Finally, we saw an increase in the level of immature neutrophils (bands) in patients after SMB infusion and a tendency for patients receiving SMB to experience fever early after operation, thus suggesting an inflammatory response to the infusion of SMB.
Although the number of patients in this study did not permit us to see significant differences in the clinically important end points of homologous blood usage, reentry for bleeding or other forms of perioperative morbidity, the results support the idea that SMB infusion does cause a coagulopathy in some patients and has other clearly undesirable consequences. One may see a coagulopathy of sufficient magnitude to cause the need of reentry or significant morbidity in only 5% or less patients exposed to SMB infusion, a frequency that would not have been detected by this study. However, our data, together with previous studies that support the tendency of SMB infusion to cause bleeding (especially when larger volumes are infused [18]), strongly suggest that this is not a good method to protect patients from exposure to homologous blood transfusion. After analysis of our data, we no longer directly infuse SMB. In the occasional patient with an initial volume of drainage of more than approximately 1 L in the first 3 hours after operation, SMB is washed and spun in the blood bank so that when sufficient red cell content is present, the washed cells can be infused.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
We acknowledge posthumously the contribution of one coauthor, Dr Laura B. McDaniel, in the preparation of this article.
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Ranucci, G. Bozzetti, A. Ditta, M. Cotza, G. Carboni, and A. Ballotta Surgical Reexploration After Cardiac Operations: Why a Worse Outcome? Ann. Thorac. Surg., November 1, 2008; 86(5): 1557 - 1562. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Boodhwani, H. J. Nathan, T. G. Mesana, F. D. Rubens, and Cardiotomy Investigators Effects of Shed Mediastinal Blood on Cardiovascular and Pulmonary Function: A Randomized, Double-Blind Study Ann. Thorac. Surg., October 1, 2008; 86(4): 1167 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sirvinskas, A. Veikutiene, R. Benetis, P. Grybauskas, J. Andrejaitiene, V. Veikutis, and J. Surkus Influence of early re-infusion of autologous shed mediastinal blood on clinical outcome after cardiac surgery Perfusion, September 1, 2007; 22(5): 345 - 352. [Abstract] [PDF]
|
|
|
|
|
|
The Society of Thoracic Surgeons Blood Conservatio, V. A. Ferraris, S. P. Ferraris, S. P. Saha, E. A. Hessel II, C. K. Haan, B. D. Royston, C. R. Bridges, R. S.D. Higgins, G. Despotis, et al. Perioperative Blood Transfusion and Blood Conservation in Cardiac Surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists Clinical Practice Guideline Ann. Thorac. Surg., May 1, 2007; 83(5_Supplement): S27 - S86. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. J. Murphy, S. M. Allen, J. Unsworth-White, C. T. Lewis, and M. J. R. Dalrymple-Hay Safety and efficacy of perioperative cell salvage and autotransfusion after coronary artery bypass grafting: a randomized trial Ann. Thorac. Surg., May 1, 2004; 77(5): 1553 - 1559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Dial, D. Nguyen, and D. Menzies Autotransfusion of Shed Mediastinal Blood: A Risk Factor for Mediastinitis After Cardiac Surgery? Results of a Cluster Investigation Chest, November 1, 2003; 124(5): 1847 - 1851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Johnell, G. Elgue, R. Larsson, A. Larsson, S. Thelin, and A. Siegbahn Coagulation, fibrinolysis, and cell activation in patients and shed mediastinal blood during coronary artery bypass grafting with a new heparin-coated surface J. Thorac. Cardiovasc. Surg., August 1, 2002; 124(2): 321 - 332. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kottke-Marchant and S. Sapatnekar Hemostatic Abnormalities in Cardiopulmonary Bypass: Pathophysiologic and Transfusion Considerations Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2001; 5(3): 187 - 206. [Abstract] [PDF]
|
|
|
|
|
|
J. Martin, D. Robitaille, L. P. Perrault, M. Pellerin, P. Page, N. Searle, R. Cartier, Y. Hebert, L. C. Pelletier, H. T. Thaler, et al. Reinfusion of mediastinal blood after heart surgery J. Thorac. Cardiovasc. Surg., September 1, 2000; 120(3): 499 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Eagle, R. A. Guyton, R. Davidoff, G. A. Ewy, J. Fonger, T. J. Gardner, J. P. Gott, H. C. Herrmann, R. A. Marlow, W. C. Nugent, et al. ACC/AHA guidelines for coronary artery bypass graft surgery: A report of the American College of Cardiology/ American Heart Association task force on Practice Guidelines (Committee to revise the 1991 Guidelines for Coronary Artery Bypass Graft Surgery) J. Am. Coll. Cardiol., October 1, 1999; 34(4): 1262 - 1347. [Full Text] [PDF]
|
|
|
|
|
|
G. Shulman, C. McQuitty, R. A. Vertrees, and V. R. Conti Acute Normovolemic Red Cell Exchange for Cardiopulmonary Bypass in Sickle Cell Disease Ann. Thorac. Surg., May 1, 1998; 65(5): 1444 - 1446. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
