- Email: [email protected]
Hemodynamic and Metabolic Changes in Hepatic Transplantation
Hemodynamic and Metabolie Changes in Hepatic Transplantation
STEVEN R. RETTKE, M.D., THEODORE A. JANOSSY, M.D., ROBERT C. CHANTIGIAN, M.D., Department of Anesthesiology; MARY F. BURRITT, Ph.D., Division of Laboratory Medicine; RUSSELL A. VAN DYKE, Ph.D., JAMES V. HARPER, M.D.,* Department of Anesthesiology; DUANE M. ILSTRUP, M.S., Section of Biostatistics; HOWARD F. TASWELL, M.D., Blood Bank and Transfusion Services; RUSSELL H. WIESNER, M.D., Division of Gastroenterology and Internal Medicine; RUUD A. F. KROM, M.D., Ph.D., Section of Transplantation Surgery In this study, we retrospectively analyzed the intraoperative hemodynamic, laboratory, and coagulation data on the first 83 patients who underwent an initial liver transplan tation procedure at our institution. The major hemodynamic changes at the time of reperfusion of the donor liver were significant decreases in arterial blood pressure, systemic vascular resistance, and pulmonary artery temperature and significant in creases in cardiac output and pulmonary capillary wedge pressure. The alterations in laboratory values reflected intraoperative therapeutic manipulations. Citrate toxicity is a concern, and the amount of calcium chloride administered reflected the volume of blood transfused. On reperfusion, the fibrinogen concentration decreased and both the prothrombin time and the activated partial thromboplastin time increased. This coagulopathy was also evident in the thromboelastographic values. Aggressive moni toring and prompt intervention are necessary to maintain hemodynamic and metabolic homeostasis in these patients.
Patients who undergo orthotopic liver transplan tation have numerous medical problems. Because of the generally debilitated condition of these pa tients and the difficulty of the surgical procedure, variations occur in hemodynamic and metabolic factors and present a management challenge to the anesthesiologist. In addition, liver transplan tation is associated with coagulopathies and mas sive blood transfusions, especially during reper fusion of the donor liver. In the current study, we retrospectively reviewed the intraoperative hemodynamic and metabolic
*Mayo Clinic Jacksonville, Jacksonville, Florida. Individual reprints of this article are not available. The entire Symposium on Liver Transplantation will be available for purchase as a bound booklet from the Proceedings Circulation Office in June. Mayo Clin Proc 64:232-240,1989
data obtained on the first 83 patients who un derwent an initial orthotopic liver transplanta tion at our institution. Our objectives were to analyze the trends in these data during the vari ous phases of liver transplantation and to deter mine whether hemodynamic and metaboUc homeo stasis could be maintained. MATERIAL A N D METHODS For analysis, liver transplantation can be divided into three distinct periods:1 phase I, which begins with the induction of anesthesia and continues until occlusion of blood flow to the patient's dis eased liver (the anhepatic time); phase II, or the anhepatic phase, which begins at the anhepatic time and continues until the donor liver is reperfused by the patient's circulating blood; and phase III, which begins at the time of reperfusion and continues until the end of the surgical proce232
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
dure. Phase III can be further subdivided into IIIA and IIIB, which encompass, respectively, the time from onset of reperfusion to 70 minutes after reperfusion and from that point until the end of the surgical procedure. Data Collection and Analysis.—Hemodynamic and laboratory data were obtained at six specific times (twice during each phase of the liver transplantation): (1) baseline—immediately after induction of general anesthesia; (2 and 3) 5 minutes before and 10 minutes after the anhepatic time; (4 and 5) 5 minutes before and 10 minutes after the reperfusion time; and (6) 70 minutes after reperfusion of the donor liver. The data on fluids and drugs were totals obtained at the end of the aforementioned phases I, II, and IIIA. All data were entered into an IBM XT com puter, and an Rbase System 5 relational data base program was used. Only raw data were entered, and when reports were generated, calcu lated values, systemic vascular resistance, pulmo nary vascular resistance, and cardiac index were automatically computed. All data are reported as means ± 1 standard deviation. Mean values were compared in the vari ous phases of liver transplantation with repeatedmeasures analysis of variance. 2 Individual means were compared against one another with use of the Student-Newman-Keuls multiple comparison procedure. 3 Differences in values were considered significant at the P<0.05 level. The protocol for review of these patients was approved by the Mayo Clinic Institutional Review Board. Hemodynamic Data.—The hemodynamic data collected were the systemic arterial blood pres sure, heart rate, pulmonary artery pressure, cen tral venous pressure, cardiac output, pulmonary capillary wedge pressure, and pulmonary artery blood temperature. Laboratory Data.—-The following laboratory data were collected: arterial blood gas tensions, potassium, hemoglobin (Instrumentation Labora tory, Inc., Lexington, Massachusetts), ionized cal cium (Radiometer, Copenhagen, Denmark), prothrombin time, activated partial thromboplastin time (General Diagnostics, Morris Plains, New Jer sey), fibrinogen (BBL, Division of Becton Dickinson and Company, Lincoln Park, New Jersey), platelet count (Sequoia-Turner Corporation, Mountain View, California), and thromboelastographic findings (Haemoscope Corporation, Morton Grove, Illinois).
233
Fluid and Drug Data.—We recorded the num ber of units of erythrocytes, autologous salvaged blood, fresh frozen plasma, cryoprecipitate, and platelets. These substances were provided to main tain vascular volume and to correct any concur rent coagulopathy. Drug data were obtained on sodium bicarbonate, which was used to correct acidosis, and calcium chloride, which was used to correct citrate toxicity. RESULTS Hemodynamic Data.—The arterial blood pres sure did not change significantly during phases I and II of liver transplantation, but on reper fusion of the donor liver, the arterial blood pres sure declined significantly (P<0.05) and then re turned toward normal values within 70 minutes after reperfusion (Fig. 1 A). The heart rate averaged 93 ± 18 beats/min on arrival of the patients in the operating room and increased significantly (P<0.05) during phase I of transplantation. By the end of phase II, the heart rate decreased to baseline values and then remained essentially unchanged during phase III (Fig. 1 A). The cardiac output declined significantly (P<0.05) during phase II of transplantation (Fig. IB). On reperfusion of the donor liver, it increased signif icantly (P<0.05) and then returned to baseline values within 1 hour. The pulmonary capillary wedge pressure de clined during phase I and reached a nadir dur ing phase II of liver transplantation (Fig. IB). On reperfusion, it immediately increased signif icantly (P<0.05) but subsequently declined dur ing the rest of phase III. The systemic vascular resistance did not change significantly during phases I and II of transplan tation, but on reperfusion of the donor liver, it decreased significantly (P<0.05) and then in creased slightly during phase III (Fig. 2 A). The pulmonary artery temperature declined during phase I of transplantation (Fig. 2 B). A significant decrease (P<0.05) occurred during phase II when the diseased liver was removed and the donor liver (at 4°C) was placed into the recipient's abdominal cavity. During phase III, the pulmonary artery temperature showed a further decline and then a slight increase. Laboratory Data.—On arrival in the operat ing room, the patients had a mean hemoglobin concentration of 9.5 g/dl, which increased signif-
234
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
15
I E
a.
Phases of OLT
Phases of OLT
Fig. 1. Hemodynamic data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Heart rate and systolic and diastolic blood pressure (BP). B, Cardiac output (CO) and pulmonary capillary wedge pressure (PCWP).
icantly (P<0.05) during phase I of transplanta tion (Fig. 3 A). A decline in the hemoglobin early in phase II was followed by a steady increase and then a significant decrease (P<0.05) after reper fusion of the donor liver. The arterial blood pH decreased during phases I and II of transplantation. A significant decline (P<0.05) occurred initially after reperfusion of the donor liver; later during phase III, the pH in creased significantly (P<0.05). The serum potas sium remained stable during phases I and II but showed a sharp increase on reperfusion of the donor liver followed by a decline to normal values during phase III (Fig. 3 B). Serum ionized calcium levels were stable in phases I and II of transplantation but increased significantly (P<0.05) in phase III (Fig. 4 A). The
platelet count increased during phase I, possibly because of platelet transfusion, and declined slightly in phase II. The decrease on reperfusion was significant (P<0.05) and was followed by an increase during phase III due to platelet trans fusion (Fig. AB). Coagulation Data.—The fibrinogen concen tration decreased during the transplantation pro cedure, whereas the prothrombin time and the activated partial thromboplastin time increased. The most dramatic changes occurred at the time of reperfusion of the donor liver (Fig. 5 A). On thromboelastography, the reaction time and the coagulation time increased significantly (P<0.05) and the clot formation rate, the maxi mal amplitude, and the amplitude 30 minutes after maximal amplitude decreased significantly
'U '-«-WA — - 1 — -
Phases of OLT
Phases of OLT
Fig. 2. Hemodynamic data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Systemic vascular resistance. B, Pulmonary artery temperature.
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
235
Average # units ofRBCs administered Phase I 4.1 Phase II 6.1 Phase IIIA 4.6
E ω
I
Phases of OLT
Phases of OLT
Fig. 3. Laboratory data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Hemoglobin concentration. B, pH and serum potassium. RBC = red blood cells.
(P<0.05) by 10 minutes after reperfusion of the donor liver (Fig. 5 B). An hour later, the thromboelastographic values began to normalize. We analyzed the thromboelastographic find ings in patients with the three most common disease states (chronic active hepatitis, primary biliary cirrhosis, and primary sclerosing cholangitis) individually at 10 minutes after reper fusion of the donor liver (Table 1). Thromboelastography revealed a significant deterioration (P<0.05) in the clot formation rate, maximal am plitude, and amplitude 30 minutes after maximal amplitude on reperfusion in patients with chronic active hepatitis in comparison with those who had primary biliary cirrhosis but no significant difference in comparison with those who had primary sclerosing cholangitis.
Phases of OLT
DISCUSSION Hemodynamic Data.—A hyperdynamic circu lation state exists in patients with cirrhosis. 4,5 Consequently, cardiac output is increased and systemic vascular resistance is decreased. The high cardiac output at baseline has been de scribed previously 6 and is typical of patients in hepatic failure. 7 During phase II of liver trans plantation, cardiac output decreases. This out come may be the result of a decrease in venous return to the heart, even if venovenous bypass is used, as it was in all our adult patients (mean flow, 2,500 ml/min). The decrease in cardiac out put may also be due to the removal of the liver, a highly metabolic organ that normally receives about 25% of the cardiac output. 8 Early in phase III, the significant increase in cardiac output is
Phases of OLT
Fig. 4. Laboratory data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Serum ionized calcium. B, Platelet count.
236
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
Fig. 5. Coagulation data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *i"<0.05 compared with previous value. A, Activated partial thromboplastin time (APTT), prothrombin time (PT), and fibrinogen. Cryo = cryoprecipitate; FFP = fresh frozen plasma. B, Thromboelastographic values. Angle = clot formation rate; MA = maximal amplitude; MA + 30 = amplitude 30 minutes after maximal amplitude; R - reaction time; R + K= coagulation time.
mainly due to a decreased afterload and a mild increase in preload and contractility. After reperfusion, the heart rate has been rela tively constant in our experience, but some in vestigators have reported an associated bradycardia (called the postreperfusion syndrome). 9 The changes in heart rate in our study patients, although statistically significant, were not clini cally significant, and the mean change from each data point was less than 8 beats/min. The cause of postreperfusion syndrome remains unknown, but the disorder may be due to the release of acid metabolites, potassium, cold perfusate, or vasoactive substances from the donor liver.10,11 On the basis of a study by Khoury and associates, 12 it seems unrelated to neurotensin or vasoactive in testinal peptides released from the gut. In a study by Martin and colleagues, 13 administration of calcium chloride before reperfusion prevented a decline in the cardiac index but did not reverse
hypotension, bradycardia, and dysrhythmias. The hypotension associated with reperfusion has also been attributed to hypovolemia. 14 This con dition was diagnosed with use of transesophageal echocardiography during orthotopic liver transplantation; investigators have determined that the pulmonary capillary wedge pressure can not be used to predict the preload, as determined by the left ventricular end-diastolic diameter. The increase in pulmonary capillary wedge pressure evident on reperfusion may also be due to an increase in transfusion or transient heart failure (or both). 15 Not all liver transplant patients have an unstable condition after reperfusion; thus, we treat such patients symptomatically as altera tions manifest. Body temperature decreases throughout liver transplantation even though attempts are made to minimize such decreases through the use of warming blankets, warming fluids, and heated
Table 1.—Thromboelastographic Values* 10 Minutes After Reperfusion of the Donor Liver in Patients Undergoing Liver Transplantation at the Mayo Clinic MA + 30 (mm) MA (mm) R(mm) R + K (mm) Angle (degrees) Diagnosist >50 50-70 12-16 20-24 46-65 Normal range 41.7 21.5Î 29.0J 29.2 CAH 23.5Î 40.4 33.4 43.6 22.7 PBC 40.6 22.1 46.8 27.2 37.1 PSC 31.8 *R = reaction time; R + K = coagulation time; angle = clot formation rate; MA = maximal amplitude; MA + 30 = amplitude 30 minutes after maximal amplitude. fCAH = chronic active hepatitis; PBC = primary biliary cirrhosis; PSC = primary sclerosing cholangitis. tSignificantly different from PBC value but not from PSC value (P<0.05).
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
237
and humidified inspired gases. The patient's ex ing phase II. The increase in ionized calcium tremities and head are wrapped to provide extra during phases II and III is due to the adminis insulation. During phase II of the transplanta tration of calcium chloride. The sharp increase tion procedure, heat is also lost through the veno into the normal range evident on Figure 4 A reflects venous bypass pump and the cold donor liver in our practice of administering calcium chloride the abdominal cavity. On reperfusion, the body immediately before reperfusion to help prevent temperature continues to decrease as the venous cardiovascular depression after reperfusion 17 ' 18 blood passes through the cold liver. A heat ex and offset the anticipated increase in serum potas changer is not used in the venovenous bypass sium. 19 The total amount of calcium chloride ad system because the patient's blood is not hepa- ministered is a reflection of the volume of blood rinized, and any turbulence-producing device transfused (Fig. 6). Citrate toxicity does occur in would be a nidus for clot formation and would patients undergoing liver transplantation, and thereby impose the potential for emboli to the rapid determination of the serum ionized calcium heart and lungs. level is imperative. Laboratory Data.—The laboratory changes The serum potassium level remains stable during in our patients were primarily due to manipula phases I and II of transplantation but increases tions performed intraoperatively and therefore sharply on reperfusion of the donor liver, a result reflect the goals we were attempting to achieve. thought to be due to the output of metabolic by The increase in hemoglobin concentration during products in the donor liver. 13 The serum potas phase I was due to transfusion of erythrocytes, sium is within the normal range at 10 minutes and the decline at the beginning of phase II was after reperfusion. The use of washed blood to partially due to a dilutional effect of the priming prevent hyperkalemia on reperfusion has been volume from the venovenous bypass pump enter advocated, 19 but in our experience, this measure ing the patient's circulation. The continued in has been unnecessary because of the administra crease during phase II was due to transfusion of tion of calcium chloride and sodium bicarbonate erythrocytes. Our goal is to increase the hemo and hyperventilation of the patient. The signif globin concentration to about 12 g/dl before re- icant decrease in potassium later in phase III perfusion because of the expected loss of blood may be due to potassium reentering the hepatovolume after circulation is restored to the new cytes of the donor liver.10,20 In our series of pa liver. We prefer to replace this lost volume with tients, potassium chloride was administered in clotting factors (that is, fresh frozen plasma, 58% of the cases to maintain normal serum levels platelets, or cryoprecipitate). of potassium. Some of the changes in potassium A mild metabolic acidosis occurs during phases level may reflect the pH because acidosis causes I and II of transplantation, and the pH signif icantly decreases on reperfusion, probably be cause of an acid load entering the patient's cir culation from the ischemie liver,16 despite attempts M A to compensate for this anticipated event. Our goal A A at the end of the procedure is to have the patient A A 15 in a mild metabolic acidosis because a metabolic A A alkalosis develops in these patients after trans O A' A plantation—perhaps related to metabolism of ci * A* Ö 10 trate from the transfused blood by the new liver. A At A A (0 \ " A* o The serum ionized calcium does not change 5 significantly during phases I and II of liver trans plantation because calcium chloride is adminis tered to maintain normal blood levels. The ionized 0 20 40 60 80 95 calcium levels must be monitored closely because Total RBCs transfused, units of the possibility of citrate toxicity, 17 which can eventuate because of the inability of the diseased liver to metabolize the citrate used in the blood Fig. 6. Amount of calcium chloride administered plotted against preservative and because no liver is present dur total number of units of red blood cells (RBCs) transfused. A
A
AÉÊLA
IJA
ìp"' '
238
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
release of potassium from cells into the extra cellular fluid. Coagulation Data.—During orthotopic liver transplantation, profound changes in coagula tion occur. The most noticeable change occurs abruptly at the time of reperfusion and is re flected by a decrease in platelet count and fibrinogen concentration in conjunction with an increase in prothrombin time and activated par tial thromboplastin time. The increase in platelet count during phase I is probably due to trans fusion of platelets during this phase. The leveling during phase II reflects our practice of not trans fusing platelets while a patient is on venovenous bypass, in order to decrease the possibility of clot formation in the bypass tubing and the produc tion of emboli.21,22 The increase in platelet count after reperfusion is due to platelet transfusion. The fibrinogen level decreases significantly when venovenous bypass is initiated. This phenomenon has been reported previously, 23 and the level de creases again on reperfusion of the donor liver, a possible indication of a fibrinolytic process. 21,22 In our patients, the mean prothrombin time was above normal (10.9 to 12.8 seconds) at baseline, an expected finding in patients with liver fail ure. 21 The mean activated partial thromboplastin time was slightly elevated. Of concern is the substantial increase on reperfusion, which is pos sibly attributable to activation of fibrinolysis and developing coagulopathy. 21,22 ' 24 Observation of diffuse oozing on reperfusion of the donor liver is common.
cause of function of the donor liver, within an hour the thromboelastographic values begin to nor malize. Also, patients with widespread parenchymal liver damage (chronic active hepatitis) in general have more severe clotting abnor malities than do those with chronic cholestasis (primary biliary cirrhosis or primary sclerosing cholangitis). 27,28 The patient who has a coagulopathy, in addition to derangements in hemodynamic and metabolic states, presents a major problem. Kang and as sociates 22,24 have conducted extensive studies of the coagulopathy associated with orthotopic liver transplantation. We used the thromboelastograph to determine the clotting status of our patients initially and made corrections, if necessary, dur ing phase I of the transplantation. In two studies, circulating heparin has been shown to be a major cause of reperfusion coagulopathy. 29,30 The pres ervation fluid used for the donor livers in these two reports contained heparin (100 mg/liter). In 13 patients who received donor livers that had been preserved in fluid that contained no heparin, a thrombin time and a Reptilase test were performed. The thrombin time was signif icantly prolonged in only 2 of the 13 patients, an indication that heparin or a heparin-like sub stance plays a minor role in coagulopathy. 22 We do not use heparin in preservation fluid for donor livers. Some investigators advocate the use of ε-aminocaproic acid (EACA) to treat fibrinolysis that occurs after reperfusion.24 EACA can be used after the following strict criteria are fulfilled:
The thromboelastograph (Fig. 7) is an instru ment that measures the sheer elasticity of a blood clot from the time the first fibrin strands are formed to completion of clot formation, including fibrinolysis. 22 Its use and comparison with other coagulation tests have been reported,21,22,25,26 and it is helpful in guiding blood product transfusion therapy. 21 Preliminary information (on reaction time, coagulation time, and clot formation rate) can be obtained within 15 minutes, a briefer period than that required for standard laboratory tests of coagulation. Thromboelastography shows that the reaction time and the coagulation time increase significantly along with a significant decrease in clot formation rate, maximal ampli tude, and amplitude 30 minutes after maximal amplitude by 10 minutes after reperfusion. This pattern is indicative of a coagulopathy. With transfusion of blood products and possibly be
Fig. 7. Diagram of variables measured by thromboelastography. a = clot formation rate (normal, more than 50°); MA = maxi mal amplitude (normal, 50 to 70 mm); MA + 30 = amplitude 30 minutes after maximal amplitude (normal, 46 to 65 mm); R = reaction time (normal, 12 to 16 mm); R + K = coagulation time (normal, 20 to 24 mm). (Normal values are from reference 22.)
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
fibrinolysis is detected by thromboelastography, EACA halts fibrinolysis in vitro as noted on the thromboelastograph, and generalized oozing oc curs in a previously dry surgical field. In order to meet these criteria, up to 2 hours can elapse. We have treated the fibrinolytic states with fresh frozen plasma, platelets, and cryoprecipitate while monitoring progress with the thromboelasto graph. The use of EACA shows promise, but a rapid means of determining fibrinolysis needs to be developed. In our series of 83 patients, fibrinolysis on reperfusion occurred 25% of the time. Such episodes of fibrinolysis were corrected with conventional transfusion therapy because the use of EACA was still somewhat controversial.24 CONCLUSION The management of anesthesia for patients under going orthotopic liver transplantation is a chal lenge because of the need to correct the many hemodynamic and metabolic variables during a technically demanding procedure. In our expe rience, the use of venovenous bypass has helped to maintain hemodynamics, and thromboelastog raphy has proved to be a rapid and reliable pro cedure for monitoring coagulation. Further studies are needed to enhance our understanding of the hemodynamic and hématologie abnormalities that occur in patients during liver transplantation. ACKNOWLEDGMENT We thank Michelle R. Engel for preparation of the submitted manuscript.
REFERENCES 1. Krom RAF: Liver transplantation at the Mayo Clinic. Mayo Clin Proc 61:278-282,1986 2. Morrison DF: Multivariate Statistical Methods. Second edition. New York, McGraw-Hill Book Company, 1976, pp 141-160 3. Winer BJ: Statistical Principles in Experimental Design. Second edition. New York, McGraw-Hill Book Company, 1971, pp 191-195; 215-218 4. Shaw BW Jr, Martin DJ, Marquez JM, Rang YG, Bugbee AC Jr, Iwatsuki S, Griffith BP, Hardesty RL, Bahnson HT, Starzl TE: Venous bypass in clinical liver transplan tation. Ann Surg 200:524-534,1984 5. Johnson G Jr, Dart CH Jr, Peters RM, Macfie JA: Hemodynamic changes with cirrhosis of the liver. Ann Surg 163:692-703,1966 6. Farman JV: Anaesthesia in the presence of liver disease and for hepatic transplantation. Br J Anaesth 44:946952,1972
239
7. Martini GA, Baltzer G, Arndt H: Some aspects of cir culatory disturbances in cirrhosis of the liver. Prog Liver Dis 14:231-250,1972 8. Svensson KL, Sonander HG, Henriksson B-A, Stenqvist O: Whole-body oxygen consumption during liver trans plantation. Transplant Proc 19 (Suppl 3):56-58,1987 9. Aggarwal S, Kang Y, Freeman JA, Fortunato FL, Pinsky MR: Postreperfusion syndrome: cardiovascular collapse following hepatic reperfusion during liver transplanta tion. Transplant Proc 19 (Suppl 3):54-55,1987 10. Carmichael FJ, Lindop MJ, Farman JV: Anesthesia for hepatic transplantation: cardiovascular and metabolic alterations and their management. Anesth Analg 64:108116,1985 11. Farman JV: Anaesthesia for transplant surgery. Int Anesthesiol Clin 16:92-119,1978 12. Khoury GF, Kaufman RD, Musich JA, Mogard M: Neurotensin and vasoactive intestinal peptide levels during orthotopic liver transplantation (abstract). Anesth Analg 65 (Suppl):S79,1986 13. Martin DJ, Marquez JM, Kang YG, Shaw BW Jr, Pinsky MR: Liver transplantation: hemodynamic and electro lyte changes seen immediately following revascularization (abstract). Anesth Analg 63:246,1984 14. Lichtor JL, Ellis JE, Uitvlugt A, Chung M, Lin CY, Thistlethwaite JR, Emond J, Broelsch CE, Roizen MF: Transesophageal echocardiography during liver trans plantation (abstract). Anesth Analg 66 (Suppl):S104,1987 15. Ellis JE, Lichtor JL, Chung MR, Polk SL, Emond J, Broelsch C, Thistlethwaite JR, Feinstein SB, Roizen MF: Mechanism of myocardial dysfunction during liver transplantation: the role of isolated right ventricular failure (abstract). Anesthesiology 67 (Suppl):A82, 1987 16. Fortunato FL Jr, Kang Y, Aggarwal S, Freeman JA, Pinsky MR: Acid-base status during and after orthotopic liver transplantation. Transplant Proc 19:59-60,1987 17. Marquez J, Martin D, Virji MA, Kang YG, Warty VS, Shaw B Jr, Sassano J J , Waterman P, Winter PM, Pinsky MR: Cardiovascular depression secondary to ionic hypocalcemia during hepatic transplantation in humans. Anesthesiology 65:457-461,1986 18. Kost GJ, Jammal MA, Ward RE, Safwat AM: Monitoring of ionized calcium during human hepatic transplanta tion. Am J Clin Pathol 86:61-70, 1986 19. Ellis R, Beeston JT, Witherington SS, Allen EL, Keenan RL: Liver transplantation: effect of washing bank blood on intraoperative control of hyperkalemia. Transplant Proc 19 (Suppl 3):73-74, 1987 20. Poison RJ, Park GR, Lindop MJ, Farman JV, Calne RY, Williams R: The prevention of renal impairment in pa tients undergoing orthotopic liver grafting by infusion of low dose dopamine. Anaesthesia 42:15-19,1987 21. Owen CA Jr, Rettke SR, Bowie EJW, Cole TL, Jensen CC, Wiesner RH, Krom RAF: Hemostatic evaluation of patients undergoing liver transplantation. Mayo Clin Proc 62:761-772,1987 22. Kang YG, Martin DJ, Marquez J, Lewis JH, Bontempo FA, Shaw BW Jr, Starzl TE, Winter PM: Intraoperative changes in blood coagulation and thrombelastographic monitoring in liver transplantation. Anesth Analg 64:888896 1985 23. Denmark SW, Shaw BW Jr, Starzl TE, Griffith BP: Veno venous bypass without systemic anticoagulation in canine and human liver transplantation. Surg Forum 34:380-382,1983 24. Kang Y, Lewis JH, Navalgund A, Russell MW, Bontempo FA, Niren LS, Starzl TE: Epsilon-aminocaproic acid for treatment of fibrinolysis during liver transplantation. Anesthesiology 66:766-773,1987
240
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
25. Zuckerman L, Cohen E, Vagher JP, Woodward E, Ca prini JA: Comparison of thrombelastography with com mon coagulation tests. Thromb Haemost 46:752-756, 1981 26. Howland WS, Schweizer O, Gould P: A comparison of intraoperative measurements of coagulation. Anesth Analg 53:657-662,1974 27. Bontempo FA, Lewis JH, Van Thiel DH, Spero JA, Ragni MV, Butler P, Israel L, Starzl TE: The relation of pre operative coagulation findings to diagnosis, blood usage, and survival in adult liver transplantation. Transplan tation 39:532-536,1985
Mayo Clin Proc, February 1989, Vol 64
28. Roberts HR, Cederbaum AI: The liver and blood coag ulation: physiology and pathology. Gastroenterology 63:297-320,1972 29. Howland WS, Ryan GM, Bettigole RE, Fortner JG: Co agulation abnormalities associated with liver transplan tation. Surgery 68:591-596,1970 30. Bellani KG, Estrin JA, Ascher NL, Najarian JS, Bush man J, Buckley J J : Reperfusion coagulopathy during human liver transplantation. Transplant Proc 19 (Suppl 3):71-72,1987
STEVEN R. RETTKE, M.D., THEODORE A. JANOSSY, M.D., ROBERT C. CHANTIGIAN, M.D., Department of Anesthesiology; MARY F. BURRITT, Ph.D., Division of Laboratory Medicine; RUSSELL A. VAN DYKE, Ph.D., JAMES V. HARPER, M.D.,* Department of Anesthesiology; DUANE M. ILSTRUP, M.S., Section of Biostatistics; HOWARD F. TASWELL, M.D., Blood Bank and Transfusion Services; RUSSELL H. WIESNER, M.D., Division of Gastroenterology and Internal Medicine; RUUD A. F. KROM, M.D., Ph.D., Section of Transplantation Surgery In this study, we retrospectively analyzed the intraoperative hemodynamic, laboratory, and coagulation data on the first 83 patients who underwent an initial liver transplan tation procedure at our institution. The major hemodynamic changes at the time of reperfusion of the donor liver were significant decreases in arterial blood pressure, systemic vascular resistance, and pulmonary artery temperature and significant in creases in cardiac output and pulmonary capillary wedge pressure. The alterations in laboratory values reflected intraoperative therapeutic manipulations. Citrate toxicity is a concern, and the amount of calcium chloride administered reflected the volume of blood transfused. On reperfusion, the fibrinogen concentration decreased and both the prothrombin time and the activated partial thromboplastin time increased. This coagulopathy was also evident in the thromboelastographic values. Aggressive moni toring and prompt intervention are necessary to maintain hemodynamic and metabolic homeostasis in these patients.
Patients who undergo orthotopic liver transplan tation have numerous medical problems. Because of the generally debilitated condition of these pa tients and the difficulty of the surgical procedure, variations occur in hemodynamic and metabolic factors and present a management challenge to the anesthesiologist. In addition, liver transplan tation is associated with coagulopathies and mas sive blood transfusions, especially during reper fusion of the donor liver. In the current study, we retrospectively reviewed the intraoperative hemodynamic and metabolic
*Mayo Clinic Jacksonville, Jacksonville, Florida. Individual reprints of this article are not available. The entire Symposium on Liver Transplantation will be available for purchase as a bound booklet from the Proceedings Circulation Office in June. Mayo Clin Proc 64:232-240,1989
data obtained on the first 83 patients who un derwent an initial orthotopic liver transplanta tion at our institution. Our objectives were to analyze the trends in these data during the vari ous phases of liver transplantation and to deter mine whether hemodynamic and metaboUc homeo stasis could be maintained. MATERIAL A N D METHODS For analysis, liver transplantation can be divided into three distinct periods:1 phase I, which begins with the induction of anesthesia and continues until occlusion of blood flow to the patient's dis eased liver (the anhepatic time); phase II, or the anhepatic phase, which begins at the anhepatic time and continues until the donor liver is reperfused by the patient's circulating blood; and phase III, which begins at the time of reperfusion and continues until the end of the surgical proce232
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
dure. Phase III can be further subdivided into IIIA and IIIB, which encompass, respectively, the time from onset of reperfusion to 70 minutes after reperfusion and from that point until the end of the surgical procedure. Data Collection and Analysis.—Hemodynamic and laboratory data were obtained at six specific times (twice during each phase of the liver transplantation): (1) baseline—immediately after induction of general anesthesia; (2 and 3) 5 minutes before and 10 minutes after the anhepatic time; (4 and 5) 5 minutes before and 10 minutes after the reperfusion time; and (6) 70 minutes after reperfusion of the donor liver. The data on fluids and drugs were totals obtained at the end of the aforementioned phases I, II, and IIIA. All data were entered into an IBM XT com puter, and an Rbase System 5 relational data base program was used. Only raw data were entered, and when reports were generated, calcu lated values, systemic vascular resistance, pulmo nary vascular resistance, and cardiac index were automatically computed. All data are reported as means ± 1 standard deviation. Mean values were compared in the vari ous phases of liver transplantation with repeatedmeasures analysis of variance. 2 Individual means were compared against one another with use of the Student-Newman-Keuls multiple comparison procedure. 3 Differences in values were considered significant at the P<0.05 level. The protocol for review of these patients was approved by the Mayo Clinic Institutional Review Board. Hemodynamic Data.—The hemodynamic data collected were the systemic arterial blood pres sure, heart rate, pulmonary artery pressure, cen tral venous pressure, cardiac output, pulmonary capillary wedge pressure, and pulmonary artery blood temperature. Laboratory Data.—-The following laboratory data were collected: arterial blood gas tensions, potassium, hemoglobin (Instrumentation Labora tory, Inc., Lexington, Massachusetts), ionized cal cium (Radiometer, Copenhagen, Denmark), prothrombin time, activated partial thromboplastin time (General Diagnostics, Morris Plains, New Jer sey), fibrinogen (BBL, Division of Becton Dickinson and Company, Lincoln Park, New Jersey), platelet count (Sequoia-Turner Corporation, Mountain View, California), and thromboelastographic findings (Haemoscope Corporation, Morton Grove, Illinois).
233
Fluid and Drug Data.—We recorded the num ber of units of erythrocytes, autologous salvaged blood, fresh frozen plasma, cryoprecipitate, and platelets. These substances were provided to main tain vascular volume and to correct any concur rent coagulopathy. Drug data were obtained on sodium bicarbonate, which was used to correct acidosis, and calcium chloride, which was used to correct citrate toxicity. RESULTS Hemodynamic Data.—The arterial blood pres sure did not change significantly during phases I and II of liver transplantation, but on reper fusion of the donor liver, the arterial blood pres sure declined significantly (P<0.05) and then re turned toward normal values within 70 minutes after reperfusion (Fig. 1 A). The heart rate averaged 93 ± 18 beats/min on arrival of the patients in the operating room and increased significantly (P<0.05) during phase I of transplantation. By the end of phase II, the heart rate decreased to baseline values and then remained essentially unchanged during phase III (Fig. 1 A). The cardiac output declined significantly (P<0.05) during phase II of transplantation (Fig. IB). On reperfusion of the donor liver, it increased signif icantly (P<0.05) and then returned to baseline values within 1 hour. The pulmonary capillary wedge pressure de clined during phase I and reached a nadir dur ing phase II of liver transplantation (Fig. IB). On reperfusion, it immediately increased signif icantly (P<0.05) but subsequently declined dur ing the rest of phase III. The systemic vascular resistance did not change significantly during phases I and II of transplan tation, but on reperfusion of the donor liver, it decreased significantly (P<0.05) and then in creased slightly during phase III (Fig. 2 A). The pulmonary artery temperature declined during phase I of transplantation (Fig. 2 B). A significant decrease (P<0.05) occurred during phase II when the diseased liver was removed and the donor liver (at 4°C) was placed into the recipient's abdominal cavity. During phase III, the pulmonary artery temperature showed a further decline and then a slight increase. Laboratory Data.—On arrival in the operat ing room, the patients had a mean hemoglobin concentration of 9.5 g/dl, which increased signif-
234
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
15
I E
a.
Phases of OLT
Phases of OLT
Fig. 1. Hemodynamic data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Heart rate and systolic and diastolic blood pressure (BP). B, Cardiac output (CO) and pulmonary capillary wedge pressure (PCWP).
icantly (P<0.05) during phase I of transplanta tion (Fig. 3 A). A decline in the hemoglobin early in phase II was followed by a steady increase and then a significant decrease (P<0.05) after reper fusion of the donor liver. The arterial blood pH decreased during phases I and II of transplantation. A significant decline (P<0.05) occurred initially after reperfusion of the donor liver; later during phase III, the pH in creased significantly (P<0.05). The serum potas sium remained stable during phases I and II but showed a sharp increase on reperfusion of the donor liver followed by a decline to normal values during phase III (Fig. 3 B). Serum ionized calcium levels were stable in phases I and II of transplantation but increased significantly (P<0.05) in phase III (Fig. 4 A). The
platelet count increased during phase I, possibly because of platelet transfusion, and declined slightly in phase II. The decrease on reperfusion was significant (P<0.05) and was followed by an increase during phase III due to platelet trans fusion (Fig. AB). Coagulation Data.—The fibrinogen concen tration decreased during the transplantation pro cedure, whereas the prothrombin time and the activated partial thromboplastin time increased. The most dramatic changes occurred at the time of reperfusion of the donor liver (Fig. 5 A). On thromboelastography, the reaction time and the coagulation time increased significantly (P<0.05) and the clot formation rate, the maxi mal amplitude, and the amplitude 30 minutes after maximal amplitude decreased significantly
'U '-«-WA — - 1 — -
Phases of OLT
Phases of OLT
Fig. 2. Hemodynamic data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Systemic vascular resistance. B, Pulmonary artery temperature.
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
235
Average # units ofRBCs administered Phase I 4.1 Phase II 6.1 Phase IIIA 4.6
E ω
I
Phases of OLT
Phases of OLT
Fig. 3. Laboratory data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Hemoglobin concentration. B, pH and serum potassium. RBC = red blood cells.
(P<0.05) by 10 minutes after reperfusion of the donor liver (Fig. 5 B). An hour later, the thromboelastographic values began to normalize. We analyzed the thromboelastographic find ings in patients with the three most common disease states (chronic active hepatitis, primary biliary cirrhosis, and primary sclerosing cholangitis) individually at 10 minutes after reper fusion of the donor liver (Table 1). Thromboelastography revealed a significant deterioration (P<0.05) in the clot formation rate, maximal am plitude, and amplitude 30 minutes after maximal amplitude on reperfusion in patients with chronic active hepatitis in comparison with those who had primary biliary cirrhosis but no significant difference in comparison with those who had primary sclerosing cholangitis.
Phases of OLT
DISCUSSION Hemodynamic Data.—A hyperdynamic circu lation state exists in patients with cirrhosis. 4,5 Consequently, cardiac output is increased and systemic vascular resistance is decreased. The high cardiac output at baseline has been de scribed previously 6 and is typical of patients in hepatic failure. 7 During phase II of liver trans plantation, cardiac output decreases. This out come may be the result of a decrease in venous return to the heart, even if venovenous bypass is used, as it was in all our adult patients (mean flow, 2,500 ml/min). The decrease in cardiac out put may also be due to the removal of the liver, a highly metabolic organ that normally receives about 25% of the cardiac output. 8 Early in phase III, the significant increase in cardiac output is
Phases of OLT
Fig. 4. Laboratory data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *P<0.05 compared with previous value. A, Serum ionized calcium. B, Platelet count.
236
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
Fig. 5. Coagulation data (mean ± 1 SD) at beginning and end of each of three phases of orthotopic liver transplantation (OLT) for 83 first-time transplantations. *i"<0.05 compared with previous value. A, Activated partial thromboplastin time (APTT), prothrombin time (PT), and fibrinogen. Cryo = cryoprecipitate; FFP = fresh frozen plasma. B, Thromboelastographic values. Angle = clot formation rate; MA = maximal amplitude; MA + 30 = amplitude 30 minutes after maximal amplitude; R - reaction time; R + K= coagulation time.
mainly due to a decreased afterload and a mild increase in preload and contractility. After reperfusion, the heart rate has been rela tively constant in our experience, but some in vestigators have reported an associated bradycardia (called the postreperfusion syndrome). 9 The changes in heart rate in our study patients, although statistically significant, were not clini cally significant, and the mean change from each data point was less than 8 beats/min. The cause of postreperfusion syndrome remains unknown, but the disorder may be due to the release of acid metabolites, potassium, cold perfusate, or vasoactive substances from the donor liver.10,11 On the basis of a study by Khoury and associates, 12 it seems unrelated to neurotensin or vasoactive in testinal peptides released from the gut. In a study by Martin and colleagues, 13 administration of calcium chloride before reperfusion prevented a decline in the cardiac index but did not reverse
hypotension, bradycardia, and dysrhythmias. The hypotension associated with reperfusion has also been attributed to hypovolemia. 14 This con dition was diagnosed with use of transesophageal echocardiography during orthotopic liver transplantation; investigators have determined that the pulmonary capillary wedge pressure can not be used to predict the preload, as determined by the left ventricular end-diastolic diameter. The increase in pulmonary capillary wedge pressure evident on reperfusion may also be due to an increase in transfusion or transient heart failure (or both). 15 Not all liver transplant patients have an unstable condition after reperfusion; thus, we treat such patients symptomatically as altera tions manifest. Body temperature decreases throughout liver transplantation even though attempts are made to minimize such decreases through the use of warming blankets, warming fluids, and heated
Table 1.—Thromboelastographic Values* 10 Minutes After Reperfusion of the Donor Liver in Patients Undergoing Liver Transplantation at the Mayo Clinic MA + 30 (mm) MA (mm) R(mm) R + K (mm) Angle (degrees) Diagnosist >50 50-70 12-16 20-24 46-65 Normal range 41.7 21.5Î 29.0J 29.2 CAH 23.5Î 40.4 33.4 43.6 22.7 PBC 40.6 22.1 46.8 27.2 37.1 PSC 31.8 *R = reaction time; R + K = coagulation time; angle = clot formation rate; MA = maximal amplitude; MA + 30 = amplitude 30 minutes after maximal amplitude. fCAH = chronic active hepatitis; PBC = primary biliary cirrhosis; PSC = primary sclerosing cholangitis. tSignificantly different from PBC value but not from PSC value (P<0.05).
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
237
and humidified inspired gases. The patient's ex ing phase II. The increase in ionized calcium tremities and head are wrapped to provide extra during phases II and III is due to the adminis insulation. During phase II of the transplanta tration of calcium chloride. The sharp increase tion procedure, heat is also lost through the veno into the normal range evident on Figure 4 A reflects venous bypass pump and the cold donor liver in our practice of administering calcium chloride the abdominal cavity. On reperfusion, the body immediately before reperfusion to help prevent temperature continues to decrease as the venous cardiovascular depression after reperfusion 17 ' 18 blood passes through the cold liver. A heat ex and offset the anticipated increase in serum potas changer is not used in the venovenous bypass sium. 19 The total amount of calcium chloride ad system because the patient's blood is not hepa- ministered is a reflection of the volume of blood rinized, and any turbulence-producing device transfused (Fig. 6). Citrate toxicity does occur in would be a nidus for clot formation and would patients undergoing liver transplantation, and thereby impose the potential for emboli to the rapid determination of the serum ionized calcium heart and lungs. level is imperative. Laboratory Data.—The laboratory changes The serum potassium level remains stable during in our patients were primarily due to manipula phases I and II of transplantation but increases tions performed intraoperatively and therefore sharply on reperfusion of the donor liver, a result reflect the goals we were attempting to achieve. thought to be due to the output of metabolic by The increase in hemoglobin concentration during products in the donor liver. 13 The serum potas phase I was due to transfusion of erythrocytes, sium is within the normal range at 10 minutes and the decline at the beginning of phase II was after reperfusion. The use of washed blood to partially due to a dilutional effect of the priming prevent hyperkalemia on reperfusion has been volume from the venovenous bypass pump enter advocated, 19 but in our experience, this measure ing the patient's circulation. The continued in has been unnecessary because of the administra crease during phase II was due to transfusion of tion of calcium chloride and sodium bicarbonate erythrocytes. Our goal is to increase the hemo and hyperventilation of the patient. The signif globin concentration to about 12 g/dl before re- icant decrease in potassium later in phase III perfusion because of the expected loss of blood may be due to potassium reentering the hepatovolume after circulation is restored to the new cytes of the donor liver.10,20 In our series of pa liver. We prefer to replace this lost volume with tients, potassium chloride was administered in clotting factors (that is, fresh frozen plasma, 58% of the cases to maintain normal serum levels platelets, or cryoprecipitate). of potassium. Some of the changes in potassium A mild metabolic acidosis occurs during phases level may reflect the pH because acidosis causes I and II of transplantation, and the pH signif icantly decreases on reperfusion, probably be cause of an acid load entering the patient's cir culation from the ischemie liver,16 despite attempts M A to compensate for this anticipated event. Our goal A A at the end of the procedure is to have the patient A A 15 in a mild metabolic acidosis because a metabolic A A alkalosis develops in these patients after trans O A' A plantation—perhaps related to metabolism of ci * A* Ö 10 trate from the transfused blood by the new liver. A At A A (0 \ " A* o The serum ionized calcium does not change 5 significantly during phases I and II of liver trans plantation because calcium chloride is adminis tered to maintain normal blood levels. The ionized 0 20 40 60 80 95 calcium levels must be monitored closely because Total RBCs transfused, units of the possibility of citrate toxicity, 17 which can eventuate because of the inability of the diseased liver to metabolize the citrate used in the blood Fig. 6. Amount of calcium chloride administered plotted against preservative and because no liver is present dur total number of units of red blood cells (RBCs) transfused. A
A
AÉÊLA
IJA
ìp"' '
238
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
Mayo Clin Proc, February 1989, Vol 64
release of potassium from cells into the extra cellular fluid. Coagulation Data.—During orthotopic liver transplantation, profound changes in coagula tion occur. The most noticeable change occurs abruptly at the time of reperfusion and is re flected by a decrease in platelet count and fibrinogen concentration in conjunction with an increase in prothrombin time and activated par tial thromboplastin time. The increase in platelet count during phase I is probably due to trans fusion of platelets during this phase. The leveling during phase II reflects our practice of not trans fusing platelets while a patient is on venovenous bypass, in order to decrease the possibility of clot formation in the bypass tubing and the produc tion of emboli.21,22 The increase in platelet count after reperfusion is due to platelet transfusion. The fibrinogen level decreases significantly when venovenous bypass is initiated. This phenomenon has been reported previously, 23 and the level de creases again on reperfusion of the donor liver, a possible indication of a fibrinolytic process. 21,22 In our patients, the mean prothrombin time was above normal (10.9 to 12.8 seconds) at baseline, an expected finding in patients with liver fail ure. 21 The mean activated partial thromboplastin time was slightly elevated. Of concern is the substantial increase on reperfusion, which is pos sibly attributable to activation of fibrinolysis and developing coagulopathy. 21,22 ' 24 Observation of diffuse oozing on reperfusion of the donor liver is common.
cause of function of the donor liver, within an hour the thromboelastographic values begin to nor malize. Also, patients with widespread parenchymal liver damage (chronic active hepatitis) in general have more severe clotting abnor malities than do those with chronic cholestasis (primary biliary cirrhosis or primary sclerosing cholangitis). 27,28 The patient who has a coagulopathy, in addition to derangements in hemodynamic and metabolic states, presents a major problem. Kang and as sociates 22,24 have conducted extensive studies of the coagulopathy associated with orthotopic liver transplantation. We used the thromboelastograph to determine the clotting status of our patients initially and made corrections, if necessary, dur ing phase I of the transplantation. In two studies, circulating heparin has been shown to be a major cause of reperfusion coagulopathy. 29,30 The pres ervation fluid used for the donor livers in these two reports contained heparin (100 mg/liter). In 13 patients who received donor livers that had been preserved in fluid that contained no heparin, a thrombin time and a Reptilase test were performed. The thrombin time was signif icantly prolonged in only 2 of the 13 patients, an indication that heparin or a heparin-like sub stance plays a minor role in coagulopathy. 22 We do not use heparin in preservation fluid for donor livers. Some investigators advocate the use of ε-aminocaproic acid (EACA) to treat fibrinolysis that occurs after reperfusion.24 EACA can be used after the following strict criteria are fulfilled:
The thromboelastograph (Fig. 7) is an instru ment that measures the sheer elasticity of a blood clot from the time the first fibrin strands are formed to completion of clot formation, including fibrinolysis. 22 Its use and comparison with other coagulation tests have been reported,21,22,25,26 and it is helpful in guiding blood product transfusion therapy. 21 Preliminary information (on reaction time, coagulation time, and clot formation rate) can be obtained within 15 minutes, a briefer period than that required for standard laboratory tests of coagulation. Thromboelastography shows that the reaction time and the coagulation time increase significantly along with a significant decrease in clot formation rate, maximal ampli tude, and amplitude 30 minutes after maximal amplitude by 10 minutes after reperfusion. This pattern is indicative of a coagulopathy. With transfusion of blood products and possibly be
Fig. 7. Diagram of variables measured by thromboelastography. a = clot formation rate (normal, more than 50°); MA = maxi mal amplitude (normal, 50 to 70 mm); MA + 30 = amplitude 30 minutes after maximal amplitude (normal, 46 to 65 mm); R = reaction time (normal, 12 to 16 mm); R + K = coagulation time (normal, 20 to 24 mm). (Normal values are from reference 22.)
Mayo Clin Proc, February 1989, Vol 64
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
fibrinolysis is detected by thromboelastography, EACA halts fibrinolysis in vitro as noted on the thromboelastograph, and generalized oozing oc curs in a previously dry surgical field. In order to meet these criteria, up to 2 hours can elapse. We have treated the fibrinolytic states with fresh frozen plasma, platelets, and cryoprecipitate while monitoring progress with the thromboelasto graph. The use of EACA shows promise, but a rapid means of determining fibrinolysis needs to be developed. In our series of 83 patients, fibrinolysis on reperfusion occurred 25% of the time. Such episodes of fibrinolysis were corrected with conventional transfusion therapy because the use of EACA was still somewhat controversial.24 CONCLUSION The management of anesthesia for patients under going orthotopic liver transplantation is a chal lenge because of the need to correct the many hemodynamic and metabolic variables during a technically demanding procedure. In our expe rience, the use of venovenous bypass has helped to maintain hemodynamics, and thromboelastog raphy has proved to be a rapid and reliable pro cedure for monitoring coagulation. Further studies are needed to enhance our understanding of the hemodynamic and hématologie abnormalities that occur in patients during liver transplantation. ACKNOWLEDGMENT We thank Michelle R. Engel for preparation of the submitted manuscript.
REFERENCES 1. Krom RAF: Liver transplantation at the Mayo Clinic. Mayo Clin Proc 61:278-282,1986 2. Morrison DF: Multivariate Statistical Methods. Second edition. New York, McGraw-Hill Book Company, 1976, pp 141-160 3. Winer BJ: Statistical Principles in Experimental Design. Second edition. New York, McGraw-Hill Book Company, 1971, pp 191-195; 215-218 4. Shaw BW Jr, Martin DJ, Marquez JM, Rang YG, Bugbee AC Jr, Iwatsuki S, Griffith BP, Hardesty RL, Bahnson HT, Starzl TE: Venous bypass in clinical liver transplan tation. Ann Surg 200:524-534,1984 5. Johnson G Jr, Dart CH Jr, Peters RM, Macfie JA: Hemodynamic changes with cirrhosis of the liver. Ann Surg 163:692-703,1966 6. Farman JV: Anaesthesia in the presence of liver disease and for hepatic transplantation. Br J Anaesth 44:946952,1972
239
7. Martini GA, Baltzer G, Arndt H: Some aspects of cir culatory disturbances in cirrhosis of the liver. Prog Liver Dis 14:231-250,1972 8. Svensson KL, Sonander HG, Henriksson B-A, Stenqvist O: Whole-body oxygen consumption during liver trans plantation. Transplant Proc 19 (Suppl 3):56-58,1987 9. Aggarwal S, Kang Y, Freeman JA, Fortunato FL, Pinsky MR: Postreperfusion syndrome: cardiovascular collapse following hepatic reperfusion during liver transplanta tion. Transplant Proc 19 (Suppl 3):54-55,1987 10. Carmichael FJ, Lindop MJ, Farman JV: Anesthesia for hepatic transplantation: cardiovascular and metabolic alterations and their management. Anesth Analg 64:108116,1985 11. Farman JV: Anaesthesia for transplant surgery. Int Anesthesiol Clin 16:92-119,1978 12. Khoury GF, Kaufman RD, Musich JA, Mogard M: Neurotensin and vasoactive intestinal peptide levels during orthotopic liver transplantation (abstract). Anesth Analg 65 (Suppl):S79,1986 13. Martin DJ, Marquez JM, Kang YG, Shaw BW Jr, Pinsky MR: Liver transplantation: hemodynamic and electro lyte changes seen immediately following revascularization (abstract). Anesth Analg 63:246,1984 14. Lichtor JL, Ellis JE, Uitvlugt A, Chung M, Lin CY, Thistlethwaite JR, Emond J, Broelsch CE, Roizen MF: Transesophageal echocardiography during liver trans plantation (abstract). Anesth Analg 66 (Suppl):S104,1987 15. Ellis JE, Lichtor JL, Chung MR, Polk SL, Emond J, Broelsch C, Thistlethwaite JR, Feinstein SB, Roizen MF: Mechanism of myocardial dysfunction during liver transplantation: the role of isolated right ventricular failure (abstract). Anesthesiology 67 (Suppl):A82, 1987 16. Fortunato FL Jr, Kang Y, Aggarwal S, Freeman JA, Pinsky MR: Acid-base status during and after orthotopic liver transplantation. Transplant Proc 19:59-60,1987 17. Marquez J, Martin D, Virji MA, Kang YG, Warty VS, Shaw B Jr, Sassano J J , Waterman P, Winter PM, Pinsky MR: Cardiovascular depression secondary to ionic hypocalcemia during hepatic transplantation in humans. Anesthesiology 65:457-461,1986 18. Kost GJ, Jammal MA, Ward RE, Safwat AM: Monitoring of ionized calcium during human hepatic transplanta tion. Am J Clin Pathol 86:61-70, 1986 19. Ellis R, Beeston JT, Witherington SS, Allen EL, Keenan RL: Liver transplantation: effect of washing bank blood on intraoperative control of hyperkalemia. Transplant Proc 19 (Suppl 3):73-74, 1987 20. Poison RJ, Park GR, Lindop MJ, Farman JV, Calne RY, Williams R: The prevention of renal impairment in pa tients undergoing orthotopic liver grafting by infusion of low dose dopamine. Anaesthesia 42:15-19,1987 21. Owen CA Jr, Rettke SR, Bowie EJW, Cole TL, Jensen CC, Wiesner RH, Krom RAF: Hemostatic evaluation of patients undergoing liver transplantation. Mayo Clin Proc 62:761-772,1987 22. Kang YG, Martin DJ, Marquez J, Lewis JH, Bontempo FA, Shaw BW Jr, Starzl TE, Winter PM: Intraoperative changes in blood coagulation and thrombelastographic monitoring in liver transplantation. Anesth Analg 64:888896 1985 23. Denmark SW, Shaw BW Jr, Starzl TE, Griffith BP: Veno venous bypass without systemic anticoagulation in canine and human liver transplantation. Surg Forum 34:380-382,1983 24. Kang Y, Lewis JH, Navalgund A, Russell MW, Bontempo FA, Niren LS, Starzl TE: Epsilon-aminocaproic acid for treatment of fibrinolysis during liver transplantation. Anesthesiology 66:766-773,1987
240
HEMODYNAMIC CHANGES IN LIVER TRANSPLANTATION
25. Zuckerman L, Cohen E, Vagher JP, Woodward E, Ca prini JA: Comparison of thrombelastography with com mon coagulation tests. Thromb Haemost 46:752-756, 1981 26. Howland WS, Schweizer O, Gould P: A comparison of intraoperative measurements of coagulation. Anesth Analg 53:657-662,1974 27. Bontempo FA, Lewis JH, Van Thiel DH, Spero JA, Ragni MV, Butler P, Israel L, Starzl TE: The relation of pre operative coagulation findings to diagnosis, blood usage, and survival in adult liver transplantation. Transplan tation 39:532-536,1985
Mayo Clin Proc, February 1989, Vol 64
28. Roberts HR, Cederbaum AI: The liver and blood coag ulation: physiology and pathology. Gastroenterology 63:297-320,1972 29. Howland WS, Ryan GM, Bettigole RE, Fortner JG: Co agulation abnormalities associated with liver transplan tation. Surgery 68:591-596,1970 30. Bellani KG, Estrin JA, Ascher NL, Najarian JS, Bush man J, Buckley J J : Reperfusion coagulopathy during human liver transplantation. Transplant Proc 19 (Suppl 3):71-72,1987