The Treatment of Postperfusion Bleeding Using wlminocaproic Acid, Cryoprecipitate, Fresh-Frozen Plasma, and Protamine Sulfate Cary J. Lambert, M.D., Alain J. Marengo-Rowe, M.D., James E. Leveson, Ph.D., Robert H. Green, M.D., J. Peter Thiele, M.D., Gerald F. Geisler, M.D., Maurice Adam, M.D., and Ben F. Mitchel, M.D. ABSTRACT The evaluation of excessive hemorrhage was carried out in 774 patients after cardiopulmonary bypass. Excessive hemorrhage was defined in any adult patient as chest tube drainage of more than 600 ml within the first eight hours after operation. Using the prothrombin time, partial thromboplastin time, fibrinogen level, and h i - F titer tests, it was possible to differentiate medical from surgical bleeding. Hyperfibrinolytic bleeding was the most frequently identifiable coagulation disorder and occurred in 159 patients (2O0/o). All these patients were successfully treated with Amicar ( E aminocaproic acid) alone, or with Amicar supplemented with cryoprecipitate or fresh-frozen plasma. Three patients (0.4%)were noted to have residual heparin and required additional protamine sulfate. Five patients (0.6%) had normal coagulation studies and required immediate reexploration. The overall blood consumption per patient was 2.1 units of packed cells. Whole blood and platelets were not used.
ized oozing, which is both frustrating and difficult to control. In another report we [ll] define what could be regarded as normal blood loss after a bypass operation and correlate specific blood coagulation tests with excessive bleeding. Abnormal or excessive bleeding is defined as chest tube drainage exceeding 600 ml in an adult patient within the first eight hours after open-heart operation. Using a combination of the prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen level, and tri-F titer (TFT), it was possible to categorize any hemorrhagic tendency into five distinctly treatable entities. These are as follows: (1) hyperfibrinolysis, compensated; (2) hyperfibrinolysis, uncompensated, with hypofibrinogenemia and other coagulation factor deficiencies; (3) unneutralized heparin or heparin rebound; (4) surgical bleeding, and (5) combinations of these. The purpose of this communication is to describe how each patient who bleeds excessively after bypass can be appropriately diagnosed and treated.
The diagnosis and management of patients with excessive hemorrhage after cardiopulmonary bypass can be problematical. In general, more advanced hemorrhagic problems develop Material and Methods in these patients than in patients with classic or A group of 774 consecutive patients undergoing inherited bleeding disorders, and the diagnosis cardiopulmonary bypass operation were suris more difficult to establish. Postperfusion veyed. The operation was done by one or more hemorrhage presents a complex situation in- of a group of five established, cardiovascular volving surgical as well as nonsurgical factors, surgeons. No patient was included in the study and sometimes both. The need for reexplora- until an experience of 250 patients had been tion has been high and often the cardiac sur- accomplished. This was done in order to exgeon has been confronted by untidy, general- clude conditions that might be attributable either to technical or to learning differences [7]. From the Departments of Thoracic Surgery and Special During cardiopulmonary bypass, a bloodless Hematology, Baylor University Medical Center and Wadley Central Blood Bank, Dallas, TX. prime, a bubble oxygenator, and a roller pump Presented at the Fifteenth Annual Meeting of The Society of at moderate to hypothermic levels were used, Thoracic Surgeons, Jan 15-17, 1979, Phoenix, AZ. complemented by cooling with topical saline Address reprint requests to Dr. Marengo-Rowe, Special Hematology Department, Baylor University Medical solution with or without hypothermic root cardioplegia. The pump prime consisted of two Center, 3500 Gaston Ave, Dallas, TX 75246.
440
0003-4975/79/110440-05$01.25 @ 1978 by Cary J. Lambert
441 Lambert et al: Treatment of Postperfusion Bleeding
liters of lactated Ringer’s solution to which 6,000 units of heparin were added E2-41. Systemic heparinization consisted of 300 units per kilogram of body weight initially, followed by a further 50 units per kilogram given every 30 minutes on bypass. Throughout the operation, heparin levels were monitored at 30-minute intervals and measured by an actual plasma heparin assay procedure [51. At the termination of bypass, heparin was neutralized with a dose of protamine sulfate equal to one or one and a half times that of the heparin level measured at the last assay. In the vast majority of patients, the amount of blood suctioned during operation was so small, less than 200 ml, that it was discarded. The routine utilization of a cell saver system could not be justified on a need or cost basis. However, any significant preperfusion or postperfusion hemorrhage resulting in the pooling of blood was recovered by cardiotomy suckers and returned to the oxygenating circuit. This blood was routinely filtered through one macropore and one micropore filter. On the completion of bypass, all the perfusate was returned to the patient if his hemodynamic state permitted. Any residual volume, usually 500 to 1,000 ml, that was not returned to the patient in the operating room was collected and usually given to the patient in the intensive care unit. Preoperatively, each patient’s hemostatic function was evaluated clinically and by laboratory means. The laboratory tests included an examination of the peripheral smear; clot retraction; bleeding time (Ivy); the prothrombin time (PT) (Dade); the partial thromboplastin time (PTT) (Hyland); platelet count with platelet aggregation (Biodata) to adenosine diphosphate, collagen, ristocetin, thrombin, epinephrine; fibrinogen (Dade); tri-F titer (TFT) [8]; and fibrin(ogen) split products (FSP). With the exception of the bleeding time, the same laboratory tests were repeated immediately after the completion of operation and then as indicated by abnormal bleeding or abnormal test values. All blood samples for laboratory testing were collected either by a clean venipuncture stick or from indwelling lines after the first 20 ml was discarded. For coagulation studies other than
clot retraction, FSP, and platelet counts, blood was drawn into plastic tubes containing the correct amount of 3.2% sodium citrate anticoagulant. All testing was performed within 30 minutes of blood collection. Blood loss, as estimated by the volume of chest tube drainage, was measured at hourly intervals after operation until no more occurred. Then the chest tubes were removed. At all times only blood components were used for treatment [lo, 161. These components included packed red cells, cryoprecipitate and fresh-frozen plasma. Whole blood, fresh blood, platelet-rich plasma, and platelet concentrates were not used.
Results Of the 774 patients who underwent aortocoronary bypass, 167 (22%) bled more than 600 ml within the first eight hours after operation. Five of them had normal coagulation studies and required immediate reexploration. In each, a definitive bleeding point requiring surgical intervention was found. All the remaining 162 patients had coagulation abnormalities severe enough to give rise to excessive bleeding and were successfully treated medically. In these patients, excessive bleeding postoperatively was best correlated with a PTT greater than 45 seconds, a PT greater than 19 seconds, a fibrinogen level less than 225 mg per deciliter, and a TFT equal to or less than 1:32. These laboratory findings, which occurred singly or in combination, formed the basis for rational therapy. In this clinical study, 255 patients (33%) required no blood or blood components; 352 (46%) received only packed red cells; and 164 (21%) received Amicar with or without cryoprecipitate or fresh-frozen plasma. Three patients (0.4%) required additional protamine sulfate. Five patients (0.6%) required reexploration for surgical bleeding. The overall average of blood components used per patient for the entire series was as follows: packed red cells, 2.1 units; cryoprecipitate, 2.5 units; and freshfrozen plasma, less than 1unit. Diagnosis and Treatment In the patient with excessive bleeding after perfusion, the coagulopathies induced by oper-
442
The Annals of Thoracic Surgery Vol 28 No 5 November 1979
Classification and Treatment of Patients with Excessive Postoperative Bleeding Cause of Bleeding Compensated hyperfibrinolysis Uncompensated hyperfibrinolysis Heparin Surgical
PT > 19 sec
PTT > 45 sec
Fibrinogen
< 225 mg/dl
TFT 1:32
Treatment
No
No
No
Yes
Amicar
YeslNo
Yes
Yes
Yes
Amicar, cryo-
Heparin effect
FFP Protamine
No
Reexploration
precipitate, YeslNo
No
Yes No
No No
PT = prothrombin time; PTI = partial thromboplastin time; TFT = tri-F titer; FFP = fresh-frozen plasma.
ation were explainable on the basis of a hyperfibrinolytic syndrome or inadequate heparin neutralization after coming off bypass. In 159 patients, there was excessive hemorrhage because of hyperfibrinolytic activity. This coagulopathy was either fully compensated or uncompensated, giving rise to various coagulation factor deficiencies. All patients with hyperfibrinolytic bleeding were treated with Amicar alone or with Amicar supplemented by cryoprecipitate and fresh-frozen plasma. In this series, because of rigorous heparin monitoring, only 3 patients were noted to have residual circulating heparin after operation and required treatment with protamine sulfate. The classification and treatment of patients with excessive postoperative bleeding is outlined in the Table. The characteristics of the various coagulopathies are as follows. Compensated hyperfibrinolysis (fibrinolysis without coagulation factor deficiencies) was identified by a falling saline TFT, a normal PTT, a normal PT, and the absence of any heparin effect in the TFT. The fibrinogen level was greater than 225 mg per deciliter. In every instance, treatment consisted of an intravenously administered bolus of Amicar (10 gm) given over 5 minutes, supplemented by another 10 gm at the rate of 2 gm per hour. Uncompensated hyperfibrinolysis (fibrinolysis with coagulation factor deficiencies) was identified by a falling saline TFT, a PTT greater than 45 seconds, a PT less than 19 seconds, a fibrinogen less than 225 mg per deciliter, and the absence of any heparin effect in the TFT. Treatment consisted of the Amicar regi-
men as for compensated hyperfibrinolysis and the rapid infusion of 20 units of cryoprecipitate. However, if the PT was prolonged beyond 19 seconds, this regimen was supplemented with 4 units of fresh-frozen plasma. Unneutralized heparin was identified by the TFT by the lack of clot formation in the lower dilutions of the patient’s plasma in saline and Amicar. The PTT and PT can also be prolonged, depending on the heparin concentration, but fibrinogen remains greater than 225 mg per deciliter. In these instances the patient was treated with an appropriate dose of protamine sulfate as determined by a heparin assay. In patients with a hemorrhagic diathesis associated with hyperfibrinolysis, compensated or uncompensated, the dose of Amicar is judged by the degree of fibrinolytic activity as determined by serial TFT, PTT, and PT. With an adequate urinary output (greater than 35 ml per hour), dosages up to 60 gm within the first hour were tolerated without adverse sequelae. Some bleeding patients with persistently prolonged PTTs required up to 40 units of cryoprecipitate within the first two hours. By contrast, patients with prolonged PTs have not required freshfrozen plasma in excess of the initial 4 units. Whenever coagulopathies occur, it is essential to implement treatment early and to pursue management aggressively until coagulation studies are normalized and hemorrhage stops.
Comment Patients undergoing a cardiopulmonary bypass operation are classified as having an acceptable
443 Lambert et al: Treatment of Postperfusion Bleeding
blood loss if chest tube drainage measures less than 600 ml within the first eight postoperative hours. In our series of 774 consecutive patients who underwent aortocoronary bypass, 607 (78%) fell within this range of normality and 167 (22%) demonstrated excessive blood loss. Of the patients who bled excessively, 5 (0.6%) had normal coagulation studies and required immediate reexploration. The remaining 162 (21%) patients with excessive hemorrhage had abnormal coagulation studies due to hyperfibrinolytic activity or unneutralized heparin, and all were treated medically. No correlation was found to exist between excessive bleeding and time on bypass or volume of suctioned blood. The most useful laboratory tests for hemostatic dysfunction were found to be a PTT greater than 45 seconds, a PT greater than 19 seconds, a fibrinogen less than 225 mg per deciliter, and a TFT equal to or less than 1:32. Platelet counts and function studies, as well as FSP estimations, bear no relevance to the clinical state and were not found to be of value. Hyperfibrinolytic bleeding was the most frequently identifiable coagulation disorder and was diagnosed in 159 (20%) patients. All were successfully treated with Amicar alone or with Amicar supplemented with cryoprecipitate and fresh-frozen plasma. Three patients (less than 1%)were noted to have residual heparin activity and were treated with protamine sulfate. Amicar or eaminocaproic acid is a 6 carbon monaminocarboxylic acid with a molecular weight of 143 daltons. The function of Amicar is inhibition of plasma plasminogen activator substances. Consequently, plasma plasminogen cannot be activated and fibrinolysis is inhibited [l, 61. Antifibrinolytic agents, such as Amicar, cannot prevent bleeding unless clotting can occur in the first place. The pharmacological effect of Amicar helps to arrest lysis of the protective blood clot, thus preventing hemorrhage. Amicar cannot be used successfully prophylactically because its action is very short-lived and, being a very small molecule, is rapidly excreted in the urine. In our study, Amicar in doses up to 90 gm within a two-hour period have been employed, without demonstrable ill effects, for ongoing hyperfibrinolytic activity associated with excessive hemorrhage. We did not encounter an increased incidence of
blood coagulability or pulmonary embolism. The latter complication occurred in this study in 2 patients who had received 20 gm of Amicar each and in 4 patients who had not received Amicar. The absence of thromboembolic complications is in accord with the findings of other investigators [14, 151. Amicar also has been used successfully in the control of bleeding due to hyperfibrinolysis following transurethral resection [91, cardiac bypass operation in children with congenital heart disease 1121, and dental extractions in hemophiliacs [171, and in the prevention of early recurrence of spontaneous subarachnoid hemorrhage [13]. The poor reputation of Amicar in certain areas is due to its use in circumstances in which it should not have been employed. Only blood components, red cells, cryoprecipitate, and fresh-frozen plasma were used as treatment. Fresh blood, whole blood, and platelets were not needed. In this series 255 patients (33%) required no blood or blood components; 352 (46%) received only packed cells in order to correct persistent normovolemic anemia; and 164 (21%)received cryoprecipitate with or without fresh-frozen plasma to treat coagulopathies. The average of blood component used per patient was as follows: packed cells, 2.1 units; cryoprecipitate, 2.5 units; and fresh-frozen plasma, less than 1 unit. In this series of patients, no instance occurred in which excessive bleeding was caused by both surgical and hematological reasons. As the study has continued over the years, however, a few patients with this combination have been noted. In each instance, the blood coagulation profile was abnormal at the termination of operation, and the patient continued to bleed despite correction of the coagulation factors with appropriate therapy. In all bleeding situations, the value of rapid hematological assessment of the patient cannot be overemphasized. References 1. Ablondi FF, Hagan JJ, Phillips M, et al: Inhibition of plasmin, trypsin and the streptokinase-activated fibrinolytic system by epsilonaminocaproic acid. Arch Biochem Biophys 82: 153, 1959 2. Beattie HW, Evans G , Garnet ES, et al: Albumin and water fluxes during cardiopulmonary bypass. J Thorac Cardiovasc Surg 67:926, 1974
444
The Annals of Thoracic Surgery Vol 28 No 5 November 1979
3. Brazier J, Cooper N, Maloney JF, et al: The adequacy of myocardial oxygen delivery and acute normovolemic anemia. Surgery 75:508, 1974 4. Cohen LH, Fosbert AM, Anderson WP, et al: The effects of phlebotomy, hemodilution and autologous transfusion on systemic oxygenation and whole blood utilization in open heart surgery. Chest 68:283, 1975 5. Dacie JV, Lewis SM: Practical Haematology. Fifth edition. Edinburgh and London, Churchilll Livingstone, 1975, p 413 6. Griffin JD, Ellman L: Epsilon-aminocaproic acid (EACA). Semin Thromb Hemostas 5:27, 1978 7. Lambert CJ: Cardiopulmonary bypass hemorrhage: a surgeon’s point of view. Semin Thromb Hemostas 3:90, 1976 8. Lambert CJ, Marengo-Rowe AJ, Leveson J, et al: The tri-F titer: a rapid test for estimation of plasma fibrinogen and detection of fibrinolysis, fibrin(ogen) split products, and heparin. Ann Thorac Surg 18:357, 1974 9. Madsen P, Strauch A: The effect of aminocaproic acid on bleeding following transurethral prostatectomy. J Urol 96:255, 1966 10. Marengo-Rowe AJ: The control of hemorrhage, in Human Hemostasis 1975. Prepared for the 28th Annual Meeting of the American Association of Blood Banks, Chicago, IL, Nov 1975. Edited by TS Zuck. Washington, DC, AABB Publications, 1975, p 51 11. Marengo-Rowe AJ, Lambert CJ, Mitchel BF, et al: The evaluation of hemorrhage in cardiac patients who have undergone extracorporeal circulation. Transfusion 19:62, 1979 12. McClure PD, Izsak J: The use of epsilonaminocaproic acid to reduce bleeding during cardiac bypass in children with congenital heart disease. Anesthesiology 40:604, 1974 13. Nibbelink DW, Torner MS, Henderson WF: Intracranial aneurysms and subarachnoid hemorrhage: a cooperative study: antifibrinolytic therapy in recent onset subarachnoid hemorrhage. Stroke 6:622, 1975 14. Nilsson IM, Anderson L, Bjorkman SE: Epsilonaminocaproic acid (EACA) as a therapeutic agent based on 5 years’ clinical experience. Acta Med Scand [Suppll 448:5, 1966 15. Sengupta RP, Sing CS, Villarejo-Ortega FJ: Use of epsilon-aminocaproic acid (EACA) in the preoperative management of ruptured intracranial aneurysm. J Neurosurg 44:479, 1976 16. Shorr JB, Marx GF: New trends in intra-operative blood replacement. Anesth Analg (Cleve)49546, 1970 17. Walsh PN, Rizza CR, Evans BR, et al: The therapeutic role of epsilon-aminocaproic acid (EACA) for dental extractions in hemophiliacs. Ann NY Acad Sci 20:267, 1975
Discussion DR. ROBERT s. LITWAK (New York, NY): I enjoyed the paper by Dr. Lambert and his colleagues. Their work is an interesting and helpful attempt to systematically analyze and treat the cause of postperfusion bleeding. But the problem, of course, is that one can never really be sure of a single cause of inappropriate bleeding, which-thanks to improved perfusion techniques-has become rather unusual. It is noteworthy that in this series, the decision to reexplore patients with excessive bleeding was based on the prior determination that the coagulation studies were all normal. Certainly, there can be no disagreement that prompt operative reentry is indicated in this subgroup. But is the converse true? That is, is it appropriate to defer early reexploration of a bleeding patient just because he has an abnormal coagulation pattern? We believe that there is far greater danger in delaying reexploration of any patient who is bleeding. It is entirely possible for surgical hemostasis to be imperfect in a patient who also happens to have a deranged coagulation spectrum. Indeed, it seems to me that we must assume the probability of inadequate surgical hemostasis in any patient who is bleeding actively after open intracardiac operation. What is the danger of early reexploration? In our experience the risk has been virtually nil. This was not the case, however, in early years when we delayed reentry because of an abnormal coagulation pattern in the patient who was bleeding after perfusion and who was being treated with blood replacement, various coagulation factors, and plasminogen activator inhibitors (such as eaminocaproic acid). Even if a patient with a coagulation abnormality is managed nonoperatively with apparent success and finally stops bleeding, the potential for tamponaderelated reduced cardiac performance is heightened because of retained intrapericardial clots. In summary, the nice study of Dr. Lambert and his colleagues is a useful contribution to our understanding of the nature of postperfusion bleeding. But experience dictates a clear message: regardless of the coagulation spectrum, if a patient is bleeding actively, assume a surgical hemostatic problem and reexplore early. DR. LAMBERT: I appreciate Dr. Litwak’s remarks. I have had considerable experience with reexploring patients, and I do not believe that the inability to make a diagnosis and provide effective treatment is a justification to continue operating. Besides, if there is a hemostatic deficiency, operative intervention will provide no real benefit. Instead, the problem of bleeding will remain. I do not mean to imply that a patient should be allowed to die. He must be made hemodynamically stable, and we practice this program. No one in our series died because of bleeding.
ized oozing, which is both frustrating and difficult to control. In another report we [ll] define what could be regarded as normal blood loss after a bypass operation and correlate specific blood coagulation tests with excessive bleeding. Abnormal or excessive bleeding is defined as chest tube drainage exceeding 600 ml in an adult patient within the first eight hours after open-heart operation. Using a combination of the prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen level, and tri-F titer (TFT), it was possible to categorize any hemorrhagic tendency into five distinctly treatable entities. These are as follows: (1) hyperfibrinolysis, compensated; (2) hyperfibrinolysis, uncompensated, with hypofibrinogenemia and other coagulation factor deficiencies; (3) unneutralized heparin or heparin rebound; (4) surgical bleeding, and (5) combinations of these. The purpose of this communication is to describe how each patient who bleeds excessively after bypass can be appropriately diagnosed and treated.
The diagnosis and management of patients with excessive hemorrhage after cardiopulmonary bypass can be problematical. In general, more advanced hemorrhagic problems develop Material and Methods in these patients than in patients with classic or A group of 774 consecutive patients undergoing inherited bleeding disorders, and the diagnosis cardiopulmonary bypass operation were suris more difficult to establish. Postperfusion veyed. The operation was done by one or more hemorrhage presents a complex situation in- of a group of five established, cardiovascular volving surgical as well as nonsurgical factors, surgeons. No patient was included in the study and sometimes both. The need for reexplora- until an experience of 250 patients had been tion has been high and often the cardiac sur- accomplished. This was done in order to exgeon has been confronted by untidy, general- clude conditions that might be attributable either to technical or to learning differences [7]. From the Departments of Thoracic Surgery and Special During cardiopulmonary bypass, a bloodless Hematology, Baylor University Medical Center and Wadley Central Blood Bank, Dallas, TX. prime, a bubble oxygenator, and a roller pump Presented at the Fifteenth Annual Meeting of The Society of at moderate to hypothermic levels were used, Thoracic Surgeons, Jan 15-17, 1979, Phoenix, AZ. complemented by cooling with topical saline Address reprint requests to Dr. Marengo-Rowe, Special Hematology Department, Baylor University Medical solution with or without hypothermic root cardioplegia. The pump prime consisted of two Center, 3500 Gaston Ave, Dallas, TX 75246.
440
0003-4975/79/110440-05$01.25 @ 1978 by Cary J. Lambert
441 Lambert et al: Treatment of Postperfusion Bleeding
liters of lactated Ringer’s solution to which 6,000 units of heparin were added E2-41. Systemic heparinization consisted of 300 units per kilogram of body weight initially, followed by a further 50 units per kilogram given every 30 minutes on bypass. Throughout the operation, heparin levels were monitored at 30-minute intervals and measured by an actual plasma heparin assay procedure [51. At the termination of bypass, heparin was neutralized with a dose of protamine sulfate equal to one or one and a half times that of the heparin level measured at the last assay. In the vast majority of patients, the amount of blood suctioned during operation was so small, less than 200 ml, that it was discarded. The routine utilization of a cell saver system could not be justified on a need or cost basis. However, any significant preperfusion or postperfusion hemorrhage resulting in the pooling of blood was recovered by cardiotomy suckers and returned to the oxygenating circuit. This blood was routinely filtered through one macropore and one micropore filter. On the completion of bypass, all the perfusate was returned to the patient if his hemodynamic state permitted. Any residual volume, usually 500 to 1,000 ml, that was not returned to the patient in the operating room was collected and usually given to the patient in the intensive care unit. Preoperatively, each patient’s hemostatic function was evaluated clinically and by laboratory means. The laboratory tests included an examination of the peripheral smear; clot retraction; bleeding time (Ivy); the prothrombin time (PT) (Dade); the partial thromboplastin time (PTT) (Hyland); platelet count with platelet aggregation (Biodata) to adenosine diphosphate, collagen, ristocetin, thrombin, epinephrine; fibrinogen (Dade); tri-F titer (TFT) [8]; and fibrin(ogen) split products (FSP). With the exception of the bleeding time, the same laboratory tests were repeated immediately after the completion of operation and then as indicated by abnormal bleeding or abnormal test values. All blood samples for laboratory testing were collected either by a clean venipuncture stick or from indwelling lines after the first 20 ml was discarded. For coagulation studies other than
clot retraction, FSP, and platelet counts, blood was drawn into plastic tubes containing the correct amount of 3.2% sodium citrate anticoagulant. All testing was performed within 30 minutes of blood collection. Blood loss, as estimated by the volume of chest tube drainage, was measured at hourly intervals after operation until no more occurred. Then the chest tubes were removed. At all times only blood components were used for treatment [lo, 161. These components included packed red cells, cryoprecipitate and fresh-frozen plasma. Whole blood, fresh blood, platelet-rich plasma, and platelet concentrates were not used.
Results Of the 774 patients who underwent aortocoronary bypass, 167 (22%) bled more than 600 ml within the first eight hours after operation. Five of them had normal coagulation studies and required immediate reexploration. In each, a definitive bleeding point requiring surgical intervention was found. All the remaining 162 patients had coagulation abnormalities severe enough to give rise to excessive bleeding and were successfully treated medically. In these patients, excessive bleeding postoperatively was best correlated with a PTT greater than 45 seconds, a PT greater than 19 seconds, a fibrinogen level less than 225 mg per deciliter, and a TFT equal to or less than 1:32. These laboratory findings, which occurred singly or in combination, formed the basis for rational therapy. In this clinical study, 255 patients (33%) required no blood or blood components; 352 (46%) received only packed red cells; and 164 (21%) received Amicar with or without cryoprecipitate or fresh-frozen plasma. Three patients (0.4%) required additional protamine sulfate. Five patients (0.6%) required reexploration for surgical bleeding. The overall average of blood components used per patient for the entire series was as follows: packed red cells, 2.1 units; cryoprecipitate, 2.5 units; and freshfrozen plasma, less than 1unit. Diagnosis and Treatment In the patient with excessive bleeding after perfusion, the coagulopathies induced by oper-
442
The Annals of Thoracic Surgery Vol 28 No 5 November 1979
Classification and Treatment of Patients with Excessive Postoperative Bleeding Cause of Bleeding Compensated hyperfibrinolysis Uncompensated hyperfibrinolysis Heparin Surgical
PT > 19 sec
PTT > 45 sec
Fibrinogen
< 225 mg/dl
TFT 1:32
Treatment
No
No
No
Yes
Amicar
YeslNo
Yes
Yes
Yes
Amicar, cryo-
Heparin effect
FFP Protamine
No
Reexploration
precipitate, YeslNo
No
Yes No
No No
PT = prothrombin time; PTI = partial thromboplastin time; TFT = tri-F titer; FFP = fresh-frozen plasma.
ation were explainable on the basis of a hyperfibrinolytic syndrome or inadequate heparin neutralization after coming off bypass. In 159 patients, there was excessive hemorrhage because of hyperfibrinolytic activity. This coagulopathy was either fully compensated or uncompensated, giving rise to various coagulation factor deficiencies. All patients with hyperfibrinolytic bleeding were treated with Amicar alone or with Amicar supplemented by cryoprecipitate and fresh-frozen plasma. In this series, because of rigorous heparin monitoring, only 3 patients were noted to have residual circulating heparin after operation and required treatment with protamine sulfate. The classification and treatment of patients with excessive postoperative bleeding is outlined in the Table. The characteristics of the various coagulopathies are as follows. Compensated hyperfibrinolysis (fibrinolysis without coagulation factor deficiencies) was identified by a falling saline TFT, a normal PTT, a normal PT, and the absence of any heparin effect in the TFT. The fibrinogen level was greater than 225 mg per deciliter. In every instance, treatment consisted of an intravenously administered bolus of Amicar (10 gm) given over 5 minutes, supplemented by another 10 gm at the rate of 2 gm per hour. Uncompensated hyperfibrinolysis (fibrinolysis with coagulation factor deficiencies) was identified by a falling saline TFT, a PTT greater than 45 seconds, a PT less than 19 seconds, a fibrinogen less than 225 mg per deciliter, and the absence of any heparin effect in the TFT. Treatment consisted of the Amicar regi-
men as for compensated hyperfibrinolysis and the rapid infusion of 20 units of cryoprecipitate. However, if the PT was prolonged beyond 19 seconds, this regimen was supplemented with 4 units of fresh-frozen plasma. Unneutralized heparin was identified by the TFT by the lack of clot formation in the lower dilutions of the patient’s plasma in saline and Amicar. The PTT and PT can also be prolonged, depending on the heparin concentration, but fibrinogen remains greater than 225 mg per deciliter. In these instances the patient was treated with an appropriate dose of protamine sulfate as determined by a heparin assay. In patients with a hemorrhagic diathesis associated with hyperfibrinolysis, compensated or uncompensated, the dose of Amicar is judged by the degree of fibrinolytic activity as determined by serial TFT, PTT, and PT. With an adequate urinary output (greater than 35 ml per hour), dosages up to 60 gm within the first hour were tolerated without adverse sequelae. Some bleeding patients with persistently prolonged PTTs required up to 40 units of cryoprecipitate within the first two hours. By contrast, patients with prolonged PTs have not required freshfrozen plasma in excess of the initial 4 units. Whenever coagulopathies occur, it is essential to implement treatment early and to pursue management aggressively until coagulation studies are normalized and hemorrhage stops.
Comment Patients undergoing a cardiopulmonary bypass operation are classified as having an acceptable
443 Lambert et al: Treatment of Postperfusion Bleeding
blood loss if chest tube drainage measures less than 600 ml within the first eight postoperative hours. In our series of 774 consecutive patients who underwent aortocoronary bypass, 607 (78%) fell within this range of normality and 167 (22%) demonstrated excessive blood loss. Of the patients who bled excessively, 5 (0.6%) had normal coagulation studies and required immediate reexploration. The remaining 162 (21%) patients with excessive hemorrhage had abnormal coagulation studies due to hyperfibrinolytic activity or unneutralized heparin, and all were treated medically. No correlation was found to exist between excessive bleeding and time on bypass or volume of suctioned blood. The most useful laboratory tests for hemostatic dysfunction were found to be a PTT greater than 45 seconds, a PT greater than 19 seconds, a fibrinogen less than 225 mg per deciliter, and a TFT equal to or less than 1:32. Platelet counts and function studies, as well as FSP estimations, bear no relevance to the clinical state and were not found to be of value. Hyperfibrinolytic bleeding was the most frequently identifiable coagulation disorder and was diagnosed in 159 (20%) patients. All were successfully treated with Amicar alone or with Amicar supplemented with cryoprecipitate and fresh-frozen plasma. Three patients (less than 1%)were noted to have residual heparin activity and were treated with protamine sulfate. Amicar or eaminocaproic acid is a 6 carbon monaminocarboxylic acid with a molecular weight of 143 daltons. The function of Amicar is inhibition of plasma plasminogen activator substances. Consequently, plasma plasminogen cannot be activated and fibrinolysis is inhibited [l, 61. Antifibrinolytic agents, such as Amicar, cannot prevent bleeding unless clotting can occur in the first place. The pharmacological effect of Amicar helps to arrest lysis of the protective blood clot, thus preventing hemorrhage. Amicar cannot be used successfully prophylactically because its action is very short-lived and, being a very small molecule, is rapidly excreted in the urine. In our study, Amicar in doses up to 90 gm within a two-hour period have been employed, without demonstrable ill effects, for ongoing hyperfibrinolytic activity associated with excessive hemorrhage. We did not encounter an increased incidence of
blood coagulability or pulmonary embolism. The latter complication occurred in this study in 2 patients who had received 20 gm of Amicar each and in 4 patients who had not received Amicar. The absence of thromboembolic complications is in accord with the findings of other investigators [14, 151. Amicar also has been used successfully in the control of bleeding due to hyperfibrinolysis following transurethral resection [91, cardiac bypass operation in children with congenital heart disease 1121, and dental extractions in hemophiliacs [171, and in the prevention of early recurrence of spontaneous subarachnoid hemorrhage [13]. The poor reputation of Amicar in certain areas is due to its use in circumstances in which it should not have been employed. Only blood components, red cells, cryoprecipitate, and fresh-frozen plasma were used as treatment. Fresh blood, whole blood, and platelets were not needed. In this series 255 patients (33%) required no blood or blood components; 352 (46%) received only packed cells in order to correct persistent normovolemic anemia; and 164 (21%)received cryoprecipitate with or without fresh-frozen plasma to treat coagulopathies. The average of blood component used per patient was as follows: packed cells, 2.1 units; cryoprecipitate, 2.5 units; and fresh-frozen plasma, less than 1 unit. In this series of patients, no instance occurred in which excessive bleeding was caused by both surgical and hematological reasons. As the study has continued over the years, however, a few patients with this combination have been noted. In each instance, the blood coagulation profile was abnormal at the termination of operation, and the patient continued to bleed despite correction of the coagulation factors with appropriate therapy. In all bleeding situations, the value of rapid hematological assessment of the patient cannot be overemphasized. References 1. Ablondi FF, Hagan JJ, Phillips M, et al: Inhibition of plasmin, trypsin and the streptokinase-activated fibrinolytic system by epsilonaminocaproic acid. Arch Biochem Biophys 82: 153, 1959 2. Beattie HW, Evans G , Garnet ES, et al: Albumin and water fluxes during cardiopulmonary bypass. J Thorac Cardiovasc Surg 67:926, 1974
444
The Annals of Thoracic Surgery Vol 28 No 5 November 1979
3. Brazier J, Cooper N, Maloney JF, et al: The adequacy of myocardial oxygen delivery and acute normovolemic anemia. Surgery 75:508, 1974 4. Cohen LH, Fosbert AM, Anderson WP, et al: The effects of phlebotomy, hemodilution and autologous transfusion on systemic oxygenation and whole blood utilization in open heart surgery. Chest 68:283, 1975 5. Dacie JV, Lewis SM: Practical Haematology. Fifth edition. Edinburgh and London, Churchilll Livingstone, 1975, p 413 6. Griffin JD, Ellman L: Epsilon-aminocaproic acid (EACA). Semin Thromb Hemostas 5:27, 1978 7. Lambert CJ: Cardiopulmonary bypass hemorrhage: a surgeon’s point of view. Semin Thromb Hemostas 3:90, 1976 8. Lambert CJ, Marengo-Rowe AJ, Leveson J, et al: The tri-F titer: a rapid test for estimation of plasma fibrinogen and detection of fibrinolysis, fibrin(ogen) split products, and heparin. Ann Thorac Surg 18:357, 1974 9. Madsen P, Strauch A: The effect of aminocaproic acid on bleeding following transurethral prostatectomy. J Urol 96:255, 1966 10. Marengo-Rowe AJ: The control of hemorrhage, in Human Hemostasis 1975. Prepared for the 28th Annual Meeting of the American Association of Blood Banks, Chicago, IL, Nov 1975. Edited by TS Zuck. Washington, DC, AABB Publications, 1975, p 51 11. Marengo-Rowe AJ, Lambert CJ, Mitchel BF, et al: The evaluation of hemorrhage in cardiac patients who have undergone extracorporeal circulation. Transfusion 19:62, 1979 12. McClure PD, Izsak J: The use of epsilonaminocaproic acid to reduce bleeding during cardiac bypass in children with congenital heart disease. Anesthesiology 40:604, 1974 13. Nibbelink DW, Torner MS, Henderson WF: Intracranial aneurysms and subarachnoid hemorrhage: a cooperative study: antifibrinolytic therapy in recent onset subarachnoid hemorrhage. Stroke 6:622, 1975 14. Nilsson IM, Anderson L, Bjorkman SE: Epsilonaminocaproic acid (EACA) as a therapeutic agent based on 5 years’ clinical experience. Acta Med Scand [Suppll 448:5, 1966 15. Sengupta RP, Sing CS, Villarejo-Ortega FJ: Use of epsilon-aminocaproic acid (EACA) in the preoperative management of ruptured intracranial aneurysm. J Neurosurg 44:479, 1976 16. Shorr JB, Marx GF: New trends in intra-operative blood replacement. Anesth Analg (Cleve)49546, 1970 17. Walsh PN, Rizza CR, Evans BR, et al: The therapeutic role of epsilon-aminocaproic acid (EACA) for dental extractions in hemophiliacs. Ann NY Acad Sci 20:267, 1975
Discussion DR. ROBERT s. LITWAK (New York, NY): I enjoyed the paper by Dr. Lambert and his colleagues. Their work is an interesting and helpful attempt to systematically analyze and treat the cause of postperfusion bleeding. But the problem, of course, is that one can never really be sure of a single cause of inappropriate bleeding, which-thanks to improved perfusion techniques-has become rather unusual. It is noteworthy that in this series, the decision to reexplore patients with excessive bleeding was based on the prior determination that the coagulation studies were all normal. Certainly, there can be no disagreement that prompt operative reentry is indicated in this subgroup. But is the converse true? That is, is it appropriate to defer early reexploration of a bleeding patient just because he has an abnormal coagulation pattern? We believe that there is far greater danger in delaying reexploration of any patient who is bleeding. It is entirely possible for surgical hemostasis to be imperfect in a patient who also happens to have a deranged coagulation spectrum. Indeed, it seems to me that we must assume the probability of inadequate surgical hemostasis in any patient who is bleeding actively after open intracardiac operation. What is the danger of early reexploration? In our experience the risk has been virtually nil. This was not the case, however, in early years when we delayed reentry because of an abnormal coagulation pattern in the patient who was bleeding after perfusion and who was being treated with blood replacement, various coagulation factors, and plasminogen activator inhibitors (such as eaminocaproic acid). Even if a patient with a coagulation abnormality is managed nonoperatively with apparent success and finally stops bleeding, the potential for tamponaderelated reduced cardiac performance is heightened because of retained intrapericardial clots. In summary, the nice study of Dr. Lambert and his colleagues is a useful contribution to our understanding of the nature of postperfusion bleeding. But experience dictates a clear message: regardless of the coagulation spectrum, if a patient is bleeding actively, assume a surgical hemostatic problem and reexplore early. DR. LAMBERT: I appreciate Dr. Litwak’s remarks. I have had considerable experience with reexploring patients, and I do not believe that the inability to make a diagnosis and provide effective treatment is a justification to continue operating. Besides, if there is a hemostatic deficiency, operative intervention will provide no real benefit. Instead, the problem of bleeding will remain. I do not mean to imply that a patient should be allowed to die. He must be made hemodynamically stable, and we practice this program. No one in our series died because of bleeding.
