Hemodialysis (HD) has improved the prospects of. Sonoclot Analysis in Cardiac Surgery in Dialysis- Dependent Patients

Transcription

1 CARDIOVASCULAR Sonoclot Analysis in Cardiac Surgery in Dialysis- Dependent Patients Toshihiko Shibata, MD, PhD, Yasuyuki Sasaki, MD, PhD, Koji Hattori, MD, PhD, Hidekazu Hirai, MD, Mitsuharu Hosono, MD, PhD, Hiromichi Fujii, MD, PhD, and Shigefumi Suehiro, MD, PhD Department of Cardiovascular Surgery, Osaka City University Medical School, Osaka, Japan Background. Dialysis-dependent patients have multiple disorders of hemostasis; however, there are no reports of viscoelastic changes during cardiac surgery in such patients. Methods. Hemostasis in dialysis-dependent patients during cardiac operations was evaluated. Thirty patients who underwent cardiopulmonary bypass (CPB) were studied: 6 with chronic renal failure undergoing dialysis (HD group), and 24 without hemodialysis. Blood samples were obtained at four points: before sternotomy, 30 and 90 minutes after the start of CPB, and after protamine administration. Results. Activated clotting time (ACT) measured with Sonoclot analyzer was significantly correlated with ACT measured traditionally (r 0.92; p < 0.001; y x). Values for ACT measured with Sonoclot analyzer as well as traditional ACT increased significantly during CPB. Values for ACT measured with Sonoclot analyzer in the HD group were significantly longer than those in the control group. Before CPB, both ACT measured with Sonoclot analyzer and traditional ACT in the HD group were significantly longer than those in the control group; however, there were no significant differences in ACT measured with Sonoclot analyzer between the groups after CPB. Clot rates and peak signal values were significantly decreased during CPB in both groups, and returned to preoperative values after protamine administration. There were no significant differences in clot rate and peak signal values between the two groups. There were no differences between the two groups in changes of time to peak. Platelet counts in the HD group were significantly higher (p < 0.05) than those in the control group. There were no differences in platelet counts after CPB between the two groups. Antithrombin III levels decreased below 50% during and after CPB. Antithrombin III in the HD group was significantly lower (p < 0.01) than those in the control group. A significant time-group interaction was observed in antithrombin III levels. Conclusions. Sonoclot signatures in HD patients showed no significant differences in viscoelastic changes compared with non-hd patients. (Ann Thorac Surg 2004;77:220 5) 2004 by The Society of Thoracic Surgeons Hemodialysis (HD) has improved the prospects of long-term survival for renal failure patients, and the number having cardiac operations has increased with lower perioperative mortality [1, 2]. Dialysis-dependent patients have multiple disorders of hemostasis [3, 4]. Cardiac surgeons have empirically recognized that hemostatic disorders are serious problems in patients undergoing dialysis [5]. The incidence of reexploration in such patients because of bleeding is higher [6, 7], leading to postoperative complications [8]. Aprotinin [9] and heparin-coated cardiopulmonary bypass (CPB) circuits [10] are reported to reduce postoperative bleeding and reexploration in dialysis patients. Although postoperative bleeding has been recognized as a common problem with cardiac surgery in dialysis-dependent patients, there are few studies of perioperative hemostasis and coagulofibrinolysis. During CPB large amounts of heparin are required for Accepted for publication Aug 1, Address reprint requests to Dr Shibata, Department of Cardiovascular Surgery, Osaka City University Medical School, 1-4-3, Asahi-machi, Abeno, Osaka , Japan; shibata@msic.med.osaka-cu.ac.jp. anticoagulation, and activated clotting time (ACT) has been used as an on-site monitor of anticoagulation. Activated clotting time measurement, although easy and reliable, does not reflect platelet function, but traditional laboratory coagulation tests are impractical because they are time-consuming. Accordingly, the use of on-site coagulation monitoring helps physicians more appropriately manage this during and after CPB. Sonoclot analysis, a viscoelastic measurement using a Sonoclot analyzer (Sienco Inc, Wheat Ridge, CO), offers a global assessment of whole blood clotting within 30 minutes [11, 12] and has shown its usefulness for evaluating intraoperative hemostatic changes [13]. Viscoelastic changes in blood have been measured by thromboelastograph; however, a Sonoclot analyzer can measure them with less intrusion. Viscoelastic changes of blood using a Sonoclot analyzer have not been reported during cardiac surgery involving dialysis patients. We assumed that dialysis patients have some differences in their viscoelastic and other coagulofibrinolytic variables during and after CPB. Thus, in this study we compared the viscoelastic changes in dialysis-dependent 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /s (03)

2 Ann Thorac Surg SHIBATA ET AL 2004;77:220 5 SONOCLOT ANALYSIS IN DIALYSIS PATIENTS Table 1. Subject Characteristics Variables HD Group (n 6) Control Group (n 24) p Value Operation CABG 4 15 Valvular surgery Male/female 5/1 20/ Age (y) Body weight (kg) BUN (mg/dl) Creatinine (mg/dl) Bleeding-time (min) a Data are expressed as the mean standard deviation. BUN blood urea nitrogen; CABG coronary artery bypass grafting; HD hemodialysis. patients against patients without HD during cardiac surgery. 221 extracorporeal ultrafiltration method was used during CPB in dialysis patients. Protamine sulfate was given to neutralize the heparin after discontinuation of CPB. Additional protamine was given when ACT was longer than the preoperative control value. Packed red blood cells were given when hemoglobin fell below 7 g/dl during CPB. Sonoclot Analysis Sonoclot measures the viscoelastic drag of an oscillating Sono-probe, producing a graph called the Sonoclot signature (Fig 1). For Sonoclot analysis, blood samples (0.4 ml) were placed in the cuvette, which contains a high concentration of activator. The vibrating probe is immersed in the sample cuvette. The onset time (SonACT) is the time until the beginning of fibrin formation; this liquid phase of blood corresponds to the ACT. The rate of fibrin formation (clot rate) is indicated by the primary slope, and is expressed as a percentage of the peak CARDIOVASCULAR Patients and Methods Thirty patients who underwent elective cardiac surgery gave informed consent to be enrolled in this study, which had been approved by our hospital s institutional review board. Six dialysis-dependent patients with chronic renal failure underwent a cardiac operation with CPB (HD group). In the same period, 24 patients not presenting renal failure were chosen for the control group. The renal patients had been dialyzed for between 1 and 20 years (median, 12 years). Preoperative exclusion criteria in the control group were a previous cardiac operation, renal dysfunction (serum creatinine 1.5 mg/dl), hepatic dysfunction, acute myocardial infarction, aspirin administration, and heparin infusion. No patients in either group needed an intraaortic balloon pump. The clinical characteristics of the patients are summarized in Table 1. There were no differences between the groups in age, sex, or preoperative bleeding times. Hemodialysis was performed the day before surgery on all HD patients. Surgery was performed on both groups through a midline sternotomy, by the same team of surgeons. All materials in the CPB circuit were coated with covalently bonded heparin (Carmeda Bioactive Surface; Medtronic Cardiopulmonary, Anaheim, CA). The membrane oxygenator (Maxima model CB 1380; Medtronic Cardiopulmonary) was also coated with Carmeda Bioactive Surface. Patients received an initial heparin dose of 300 IU/kg just before CPB. Before CPB was initiated, the ACT (Hemochron; International Technidyne, Inc, Edison, NJ) exceeded 400 seconds. The ACT was determined every 30 minutes during CPB, and additional heparin was given to maintain anticoagulation levels. In both groups, CPB was maintained with a perfusion rate of 2.8 L min 1 m 2 under moderate hypothermia (rectal temperature approximately 32 C). Cardiac arrest was induced by aortic cross-clamping followed by the infusion of cold blood cardioplegia. The Fig 1. Sonoclot signature before sternotomy (A) and during cardiopulmonary bypass (B). (SonACT activated clotting time measured by Sonoclot analyzer.)

3 CARDIOVASCULAR 222 SHIBATA ET AL Ann Thorac Surg SONOCLOT ANALYSIS IN DIALYSIS PATIENTS 2004;77:220 5 test. Two-way analysis of variance with repeated measures was used to evaluate the difference between groups and group time interaction. If analysis of variance revealed significant effects, Student s t test was used for each sample. Univariate regression analysis was used. Blood transfusion was expressed as median and range, and compared with the Mann-Whitney U test. All data were evaluated with computer software (StatView 5.0; Abacus Concepts, Berkeley, CA). A p value of less than 0.05 was considered significant. Fig 2. Correlation between activated clotting time measured traditionally (ACT) and by Sonoclot analyzer (SonACT). The dotted line represents the correlation in the hemodialysis (HD) group (n 24; r 0.96; p 0.001; y x) and the solid line represents the correlation in the control group (n 96; r 0.92; p 0.001; y x). amplitude per unit time (normal range, 15 to 45 signals/ min). An inflection point on the upstroke represents the start of the contraction of fibrin strands by the action of platelets. The secondary slope reflects further fibrinogenesis, fibrin polymerization, and platelet-fibrin interaction. Two variables can be obtained from the peak. The peak signal reflects the completion of fibrin formation. The time to peak amplitude is an index of the rate of conversion of fibrinogen to fibrin. A time to peak of less than 30 minutes is considered normal. A downward slope is produced after the peak because platelets induce further contraction of the completed clot as serum is squeezed out of the clot matrix. Results The SonACT was significantly correlated with the ACT in 120 measured points (r 0.92; p 0.001; y x). Significant correlation between SonACT and ACT was observed in each group (Fig 2). The SonACT as well as the ACT increased significantly during CPB (Table 2). In the analysis with two-way analysis of variance in repeated measures, values for SonACT in the HD group were significantly longer than those in the control group. There were no significant differences in ACT between the two groups. Before CPB, both SonACT and ACT in the HD group were significantly longer than those in the control group; however, there were no significant differences in SonACT in either group after CPB. Clot rates were significantly decreased during CPB in both groups (HD group, from 25 8 to signals/min; control group, from 28 9 to signals/min), and returned to preoperative values after protamine administration (Fig 3). Peak signal values were significantly decreased during CPB in both groups (HD group, from to signals; control group, from to signals), and returned to Blood Sampling Blood samples were obtained from a radial arterial catheter or the arterial side of the CPB circuit at four time points: before sternotomy, 30 and 90 minutes after the start of CPB, and 5 minutes after protamine administration. Hematocrit and platelet counts were measured with an automatic cell counter. Plasma was separated by centrifugation at 3,000 g for 10 minutes and stored at 20 C to measure fibrinogen, antithrombin III (AT-III), thrombin-antithrombin III complex (TAT), and plasmin- 2 plasmin inhibitor complex. No correction for hemodilution was performed. Data Analysis The data are expressed as the mean standard deviation. Comparisons of variables of patient characteristics between groups were performed using Student s t test. The discrete variables were compared with Fisher s exact Fig 3. Changes in clot rate. Data represent the mean standard deviation. (CPB30 and CPB90 30 and 90 minutes after the start of bypass; HD hemodialysis group; POST after protamine administration; PRE before sternotomy.)

4 Ann Thorac Surg SHIBATA ET AL 2004;77:220 5 SONOCLOT ANALYSIS IN DIALYSIS PATIENTS Fig 4. Changes in peak signal value. Data represent the mean standard deviation. (CPB30 and CPB90 30 and 90 minutes after the start of bypass; HD hemodialysis group; POST after protamine administration; PRE before sternotomy.) preoperative values after protamine administration (Fig 4). There were no significant differences in clot rate and peak signal values between the two groups. The time to peak was not estimated during CPB because of the prolonged SonACT with heparinization. There were no differences between the two groups in changes of time to peak (Fig 5). Blood transfusion during CPB was significantly greater in the HD group (mean, 9 units; range, 8 to 10 units) as compared with the control group (mean, 0 units; range, 0 to 6 units). There were no significant differences in hematocrit between the two groups, but significant time group interaction was detected (Table 2). Hematocrit in the HD group was significantly lower before operation, but higher after CPB because of more transfusions. Platelet counts in the HD group were significantly higher (p 0.05) than those in the control group. There were no differences in platelet counts after CPB between the two groups. Antithrombin III levels decreased below 50% during and after CPB. Antithrombin III in the HD group was significantly lower (p 0.01) than in the control group. A significant time group interaction was observed in AT- III. There were no significant differences between the two groups in fibrinogen levels. The TAT gradually increased during CPB, and the increase was prominent after protamine administration in both groups. There were no significant differences between the two groups in the changes in TAT. The changes in the plasmin- 2 plasmin inhibitor complex showed no significant difference between the two groups. 223 evaluated in cardiac surgery. Significant differences between the two groups, evaluated with two-way analysis of variance with repeated measures, were recognized only in SonACT, platelet counts, and AT-III in this study. We expected that some difference in Sonoclot signature between the groups would be recognized, especially after CPB. Unexpectedly, no significant differences between the two groups were recognized in any indices of Sonoclot signature after protamine administration. There are several reports of hemostatic management in cardiac surgery using a Sonoclot analyzer; however, detailed changes of Sonoclot signature during and after CPB were not mentioned [13 16]. Inasmuch as the typical changes in Sonoclot signature during cardiac surgery have not been recognized, we used 24 nondialysis patients to determine accurate control values of the Sonoclot indices. This is why the number of patients was different between the two groups. The SonACT reflects the beginning of fibrin clot formation, and is thought to be a similar measurement as ACT measured with Hemochron. Actually, in the present study, SonACT and ACT had excellent correlation. This correlation was recognized in each group as well. When comparing the two groups, values for SonACT in dialysis patients were longer than those in the control group, but not ACT. Because the same dose of heparin was used in each group, no differences between the two groups should be detected in ACT and SonACT. Moreover, our data demonstrated significantly lower levels of AT-III in dialysis patients. Because AT-III plays an important role in anticoagulation with heparin, decreased preoperative CARDIOVASCULAR Comment In this study, changes in the Sonoclot signature and in coagulofibrinolytic markers of dialysis patients were Fig 5. Time to peak. Data represent the mean standard deviation. (HD hemodialysis group; POST after protamine administration; PRE before sternotomy.)

5 CARDIOVASCULAR 224 SHIBATA ET AL Ann Thorac Surg SONOCLOT ANALYSIS IN DIALYSIS PATIENTS 2004;77:220 5 Table 2. Hematologic and Coagulofibrinolytic Data a Variable Group Pre CPB30 CPB90 Post Hematocrit (%) Control HD c b PLT ( 10 4 /mm 3 ) Control HD c c ACT (s) Control HD SonACT (s) Control HD c c AT III (%) Control HD c 33 5 a b FIB (mg/dl) Control HD TAT (ng/ml) Control HD PIC (ng/ml) Control HD a Data are expressed as the mean SD. b p c p 0.01 compared with control. ACT activated clotting time (Hemochron); AT-III antithrombin III; CPB30 and CPB90 30 and 90 minutes after the start of bypass; FIB fibrinogen; PIC plasmin- 2 plasmin inhibitor complex; PLT platelet; Post after protamine administration; Pre before sternotomy; SonACT activated clotting time measured with Sonoclot analyzer; TAT thrombin antithrombin III complex. AT-III levels cause reduced response to heparin [17]. Therefore the ACT and SonACT in dialysis patients would be shorter owing to lower AT-III levels. The mechanism of prolonged SonACT in dialysis patients cannot be explained with only AT-III levels in our study. Ebrest and Berkowitz [3] reviewed hemostasis in renal disease, and mentioned reduced coagulation factors levels. Some other factors might contribute to prolonged SonACT in dialysis patients. Our data demonstrated no significant differences between the two groups in other variables in the Sonoclot signature. The time to peak represents the duration from the start of measurement to the time of the highest Sonoclot signal. It can be divided into two parts: horizontal (SonACT) and a slope (Fig 1A). Because SonACT is extremely prolonged in the presence of large doses of heparin, the time to peak is greatly influenced by extremely prolonged SonACT (Fig 1B). Accordingly, we consider that the time to peak is a less valuable variable during CPB. Therefore, we compared this variable before and after protamine administration. Pivalizza and colleagues [18] measured the viscoelastic changes in uremic patients using a thromboelastograph and Sonoclot and demonstrated a high clot rate and low time to peak in uremic patients. Holloway and associates [19] showed hypercoagulability and high fibrinogen levels in dialysis patients, whereas our data showed no significant differences in fibrinogen levels. We compared the indices of the Sonoclot signature and coagulofibrinolytic variables (TAT and plasmin- 2 plasmin inhibitor complex). During CPB, TAT levels are gradually increased even when adequate heparin is used [20]. In our data, TAT continued to increase after initiation of CPB, and the patterns of the changes in TAT levels were similar in both groups. Significant increases of TAT were noted after protamine administration in both groups, but no differences in TAT levels were observed between the two groups at any point. There were no significant changes in plasmin- 2 plasmin inhibitor complex levels between the two groups. These changes of coagulofibrinolytic variables were not likely to affect the Sonoclot signature. Another interest was whether dialysis patients actually have bleeding diathesis or not. Unfortunately we performed empiric platelet transfusion following blood sampling after protamine administration in some dialysis patients, but not in control patients. Therefore blood loss during operation could not be compared in this study. At any rate, the platelet transfusion in the HD group did not affect the data in this study. Platelet transfusion is often performed empirically in the management of bleeding, with little scientific basis. Although there are many available methods to determine platelet function [21], many of these are not suitable for care monitoring because they are time-consuming. The time to peak in Sonoclot signature is thought to correlate with general platelet function [14]. The present data showed that any significant changes between the two groups in the time to peak suggested that the platelet function was not aggravated after CPB in dialysis patients. Moreover, there were no differences in platelet counts between the two groups after CPB. These results suggest that empiric platelet transfusion should not be necessary in dialysis patients in respect to the quantity and function of the platelets. This study was limited because of the small number of patients and more transfusions during CPB in the HD group. Blood transfusion was inevitable in practice because dialysis patients generally have anemia. The Sonoclot signature in dialysis patients showed no significant

6 Ann Thorac Surg SHIBATA ET AL 2004;77:220 5 SONOCLOT ANALYSIS IN DIALYSIS PATIENTS differences in viscoelastic changes compared with nondialysis patients. From our study, we could not detect any significant evidence that dialysis patients were in bleeding diathesis. However, the fragility of vessels and tissue is another important factor in surgical bleeding, and further study is needed to reveal the hemostatic ability of dialysis patients in cardiac surgery. References 1. Nakamura Y, Sajata R, Ura M, Miyamoto TA. Coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 1999;68: Frenken M, Krian A. Cardiovascular operations in patients with dialysis-dependent renal failure. Ann Thorac Surg 1999;68: Eberst ME, Berkowitz LR. Hemostasis in renal disease: pathophysiology and management. Am J Med 1994;96: Gordge MP, Faint RW, Rylance PB, et al. Platelet function and the bleeding time in progressive renal failure. Thromb Haemost 1988;60: Franza DL, Krata JM, Crumbley J, et al. Early and long-term results of coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 2000;70: Moulton MJ, Creswell LL, Mackey ME, et al. Reexploration for bleeding is a risk factor for outcomes after cardiac operations. J Thorac Cardiovasc Surg 1996;111: Horst M, Mehlhorn U, Hoerstrup SP, et al. Cardiac surgery in patients with end-stage renal disease: 10-year experience. Ann Thorac Surg 2000;69: Durmaz I, Buket S, Atay Y, et al. Cardiac surgery with cardiopulmonary bypass in patients with chronic renal failure. J Thorac Cardiovasc Surg 1999;118: Lemmer JH, Metzdorff MT, Krause AH, et al. Aprotinin use 225 in patients with dialysis-dependent renal failure undergoing cardiac operation. J Thorac Cardiovasc Surg 1996;112: Suehiro S, Shibata T, Sasaki Y, et al. Cardiac surgery in patients with dialysis-dependent renal failure. Ann Thorac Cardiovasc Surg 1999;5: Hett DA, Walker D, Pilkington SN, et al. Sonoclot analysis. Br J Anaesth 1995;75: Lyew MA, Spaulding WC. Template for rapid analysis of the Sonoclot signature. J Clin Monit 1997;13: Tuman KJ, Spiess BD, McCarthy RJ, et al. Comparison of viscoelastic measures of coagulation after cardiopulmonary bypass. Anesth Analg 1989;69: Miyashita T, Kuro M. Evaluation of platelet function by Sonoclot analysis compared with other hemostatic variables in cardiac surgery. Anesth Analg 1998;87: Stern MP, DeVos-Doyle K, Viguera MG, Lajos TZ. Evaluation of post-cardiopulmonary bypass Sonoclot signatures in patients taking nonsteroidal anti-inflammatory drugs. J Cardiothorac Vasc Anesth 1989;3: Kamada Y, Yamakage M, Niiya T, Tsujiguchi N, Chen S, Namiki A. Celite-activated viscometer Sonoclot can measure the suppressive effect of tranexamic acid on hyperfibrinolysis in cardiac surgery. J Anesth 2001;15: Dietrich W, Brauin S, Spannagl M, Richter JA. Low preoperative antithrombin activity causes reduced response to heparin in adult but not in infant cardiac-surgical patients. Anesth Analg 2001;92: Pivalizza EG, Abramson DC, Harvey A. Perioperative hypercoagulability in uremic patients: a viscoelastic study. J Clin Anesth 1997;9: Holloway DS, Vagher JP, Caprini JA, et al. Thromboelastography of blood from subjects with chronic renal failure. Thromb Res 1987;45: Kumano H, Suehiro S, Hattori K, et al. Coagulofibrinolysis during heparin-coated cardiopulmonary bypass with reduced heparinization. Ann Thorac Surg 1999;68: Mackenzie ME, Gurbel PA, Levine DJ, Serebruany VL. Clinical utility of available methods for determining platelet function. Cardiology 1999;92: CARDIOVASCULAR