Heparin Rebound: A Comparative Study of Protamine Chloride and Protamine Sulfate in Patients Undergoing Coronary Artery Bypass Surgery Anne H. Kuitunen, MD, Markku T. Salmenper87, MD, Jussi Heinonen, MD, Vesa P. Rasi, MD, and Gunnar Myllyll, MD Heparin rebound has been suggested to occur when protamine sulfate, but not protamine chloride, is used to neutralize heparin. This study was undertaken to compare these two protamine salts in 32 patients undergoing coronary artery bypass surgery. Initial heparin and subsequent protamine doses were determined by constructing a heparin-activated coagulation time response curve. Heparin was neutralized either with protamine sulfate or protamine chloride. The total protamine/heparin dose ratio was 0.71 f 0.05 for protamine sulfate and 0.77 * 0.07 (mg/lOO U) for protamine chloride. The initial neutralization effect, the subsequent
H
EPARIN REBOUND is a potential bleeding disorder associated with cardiopulmonary bypass (CPB). It is defined as the resurgence of the anticoagulant activity in a patient whose blood heparin has been adequately neutralized with protamine.’ There is no uniform theory of the cause of heparin rebound, but the following mechanisms have been suggested: (1) heparin may be released by red blood cell breakdown’; (2) a proportion of heparin’s effect may temporarily be neutralized by some endogenous antagonists, and heparin activity is again demonstrated when these antagonists have been cleared from plasma3; (3) heparin may redistribute into the circulation from extravascular spaces or from endothelium4.5; or (4) a naturally occurring protaminase in plasma may cause breakdown of protamine, thereby leaving heparin free in plasma.‘~6~7 Various protocols have been used to neutralize heparin with protamine after CPB. The approach of administering a fixed dose of protamine for each 100 U of heparin given has been partially replaced by methods determining protamine dose on the basis of measured heparin activity, heparin concentration, or by using a protamine titration in vitro.8-‘3 These methods have resulted in decreased protamine/ heparin dose ratios.8-1’Although the initial neutralization might be adequate with these decreased protamine doses, it is possible that a price has to be paid in the form of an increased incidence of heparin rebound.r4 Protamine is available as sulfate and chloride salts. Protamine chloride has been shown to have a more rapid onset of neutralization in patients than protamine sulfate.” In vitro, protamine chloride is more stable in plasma than protamine sulfate.‘6.‘7 Based on these studies, heparin rebound has been suggested to occur when protamine sulfate but not protamine chloride is used to neutralize heparin.16 The present study was undertaken to compare the clinical effects of the two protamine salts with special reference to their effect on the occurrence of heparin rebound. MATERIALS
AND METHODS
After institutional approval, neutralization of heparin was studied in 32 patients undergoing coronary artery bypass surgery. None of the patients received antiplatelet drugs, oral anticoagulants, or heparin prior to surgery, and none of them had preoperative
behavior of the plasma heparin level, and the various coagulation parameters did not differ significantly between the groups. Two hours after neutralization, a small and temporary increase of plasma heparin level was observed in both groups. The postoperative blood losses were comparable in both groups. Thus, protamine chloride was not a clinically superior antidote to heparin than protamine sulfate. The observed heparin rebound levels were low and clinically insignificant in terms of blood loss, but they were associated with slight changes in coagulation monitoring. Copyright o 1991 by W.B. Saunders Company
alterations in blood coagulation (platelet count, activated partial thromboplastin time [APT-T], prothrombin time [PT], plasma fibrinogen level) or liver function. The patients were randomly assigned to two groups according to the protamine salt to be used for heparin neutralization: either protamine sulfate or protamine chloride was given in a double-blind manner using the dose ratio of 1.3 mg protamine/lOO U heparin as estimated from the heparinactivated coagulation time (ACT) curve (see below). Two patients of the 32 were excluded because of failure of compliance with the protocol by the intensive care unit (KU) staff not participating in the study. Clinical details of the 30 patients studied are shown in Table 1. The patients were premeditated with morphine, 0.2 mg/kg, and scopolamine, 0.006 mglkg, intramuscularly. Anesthesia was induced with fentanyl? 50 ug!kg, diazepam, 0.1 mgkg, and pancuronium, 0.1 mg/kg, and maintained with additional doses of fentanyl and pancuronium, as well as low concentrations of enflurane. Patients were ventilated with a mixture of oxygen and air (F,O, 0.5). Dopamine and nitroglycerin were given postoperatively when clinically indicated. After induction of anesthesia, 1 U of blood was withdrawn and replaced with 1,000 mL of Ringer’s acetate solution. The perfusion circuit included a Shiley 100 A bubble oxygenator (Irvine, CA) that was primed with 2,000 mL of Ringer’s acetate and 5,000 U of heparin. The patients were subjected to moderate systemic hypothermia (28°C to 3O”C), topical myocardial cooling, and cold crystalloid potassium cardioplegia solution. During the extracorporeal circulation, the hematocrit was maintained between 20% and 25%, with packed red cells being added to the oxygenator if necessary. Anticoagulation was achieved with porcine mucosal sodium heparin (N.V. Organon, Oss, The Netherlands). After determination of the baseline celite ACT, the heparin-ACT dose response curve was constructed with the aid of an initial heparin dose (200 U/kg) as recommended by Bull et al.” The ACT was maintained greater than 480 seconds during CPB. After discontinuation of
From the Department ofAnesthesiology Helsinki University Central Hospital, and Finnish Red Cross Blood Transfusion Service, Helsinki, Finland. Supported in part by the Paolo Foundation and Helsinki University Central Hospital, Helsinki, Finland. Address reprint requests to Anne H. Kuitunen, MD, Department of Anesthesiology, Helsinki University Central Hospital, Haartmaninkatu 4, SF-00290 Helsinki, Finland. Copyright 0 1991 by W.B. Saunders Company 1053-0770/91J0503-0.00l0 221
222
KUITUNEN ET AL
Table 1. Clinical Details of the Patients Protamine Chloride
Protamlne Sulfate
No. of patients
14
16
Sex (female/male)
519
818
Age (vr) Weight (kg)
57.2 + 2.3
55.9 ?X2.0
78.0 + 3.0
77.4 t 3.7
Height(m)
1.68 + 0.02
1.67 + 0.02
Perfusion time (min)
126?
7
Surgical time (min)
296 ? 13
276 f 13
127 + 9
No. of grafts
3.6 2 0.4
3.9 +- 0.4
NOTE. Values are given as mean t SEM.
CPB, heparin was neutralized according to the study group either with protamine chloride or protamine sulfate. Immediately after weaning the patient from CPB, the heparin level in the circulating blood was determined according to the heparin-ACT dose response curve.8 Thereafter, protamine, 1.3 mgilO0 U of calculated heparin, was given as an infusion into a peripheral vein over 15 minutes. Simultaneously with the administration of this protamine infusion, the autologous blood unit was given and the available residual oxygenator perfusate was returned to the patient. The amount of this perfusate was not different between the two groups: 645 ? 10.5 mL (mean t SEM) in the protamine sulfate group and 690 + 75 mL in the protamine chloride group. An additional dose of either protamine sulfate or protamine chloride (according to the group assigned) was given to the patient after the administration of the perfusate. This additional dose of protamine was 25 mg/500 mL of perfusate. All the blood samples were collected into plastic syringes through an unheparinized radial artery catheter. Samples were taken at the following times: (1) preoperatively after induction of anesthesia; (2) after the end of CPB before the infusion of protamine; (3) 20 minutes; (4) 80 minutes; (5) 140 minutes; (6) 200 minutes; (7) 260 minutes; and (8) 380 minutes after the end of the protamine infusion. Whole blood coagulation time was determined by a semiactivated technique with special glass tubes containing either diatomaceous earth (ACT) or saline solution (saline coagulation time [SCT]) and a small magnet. The tubes were rotated at 37°C in a Hemochron device (International Technidyne Corporation, Edison, NJ). A thromboelastograph (TEG) tracing (Hellige Company, Freiburg, Germany) was obtained from titrated whole blood. Coagulation was initiated by adding calcium chloride to the cuvettes. TEG tracings were analyzed by measuring reaction time (R, min), coagulation time (R + K, min), clot formation rate (alpha angle, “) and maximum amplitude (MA, mm).” ACT, SCT, and TEG tracings were performed from two identical blood samples at the same time. Protamine was added to one of the samples (protamine concentration in the sample was 5 kg/mL) and the samevolume of saline solution to the other. The purpose of this protamine administration was to demonstrate the presence of free heparin in the blood sample. The APTT and thrombin time (TT) were determined from titrated frozen (-70°C) plasma samples using standard techniques. TT was determined by adding thrombin in a concentration (10 National Institute of Health UnitsimL) that produced a clotting time of 5 seconds for normal human plasma. Even with this high thrombin concentration, the distribution of TTs was skewed toward long clotting times necessitating logarithmic transformation to reach normal distribution for statistical comparisons. Plasma heparin level was determined by a photometric assay using the
chromogenic substrate S 2222, which shows excess of activated Factor X (Heparin Coatest, Kabi, Sweden).” In order to compare various methods of protamine dose estimation, the hypothetical protamine doses needed by the patients were calculated in two alternative ways: (1) by determining protamine response in vitro,” and (2) by using directly measured plasma heparin levels (chromogenic heparin assay).” Protamine response in vitro was evaluated using two ACTvalues: a status (heparinized) ACT and an ACT obtained after in vitro addition of liquid protamine (protamine concentration in the sample, 9.1 &mL). These ACT’s were determined simultaneously using the standard Hemochron procedure. Protamine concentration needed to bring the ACT to baseline (unheparinized) values was extrapolated from the data. To calculate the dose of protamine, the patients were assumed to be normovolemic and their blood volume was calculated using a formula previously reported by Allen et al.“’ The same blood volume was used in calculation of the protamine dose on the basis of plasma heparin level. In these protamine dose calculations, a protamineiheparin dose ratio of I .3 mgi100 IU was used. Blood loss through the chest tubes was measured from the end of the operation until the last blood sample. Postoperatively, hematocrit was maintained at approximately 30% by giving packed red cells and Ringer’s acetate. Whole blood was used to replace blood loss. The amount of transfused blood products was recorded during the study period. After the initial neutralization of heparin. extra protamine (2.5 mg) was given to bleeding patients (chest tube drainage > 200 mL/h) if the TEG values were outside the normal of the ACT and SCT, and if the in range, there was prolongation vitro test protamine (5 &mL) decreased the hypocoagulability. One- and two-way analysis of variance (ANOVA) for repeated measures were first applied for the statistical comparisons within and between the groups, respectively. In case of significance after one-way ANOVA (P < 0.05), a paired t test with Dunnett‘h allowance for multiple comparisons was used to seek contrasts to the control values. Between the groups the comparisons for the individual data points were made with unpaired t test using Bonferroni’s allowance as necessary. Simple least squares linear regression analysis was used to correlate the variables.
RESULTS
The total dose of heparin given was 514 2 39 U/kg in the protamine chloride group and 516 ‘- 29 U/kg in the protamine sulfate group. The amount of protamine actually given to the patients was calculated from the heparin dose-ACT response curve applying the protamine/heparin ratio of 1.3 mg/lOO U. For comparison, the required protamine dose from the protamine titration curve or by using the directly measured plasma heparin levels was also calculated applying the protamine/heparin ratio of 1.3 mg/lOO U. The fourth method used for calculation of the hypothetical protamine dose was the application of the widely used fixed protamine/given heparin dose ratio (1 mg protamine/lOO U of given heparin). The doses of protamine and the total protamine/heparin ratios estimated by these four methods are shown in Table 2. With initial neutralization, the mean heparin concentration decreased from 3.25 -f 0.29 UimL to 0.018 ? 0.006 U/mL in the protamine chloride group and from 3.26 -t 0.19 U/mL to 0.010 ? 0.03 U/mL in the protamine sulfate group. This initial response and the subsequent plasma free heparin levels did not significantly differ between the groups (Fig 1). However, plasma heparin concentrations
HEPARIN REBOUND IN CARDIAC SURGERY
223
circulating free heparin concentration was shown in both these patients (0.056 U/mL and 0.119 U/mL). There were also other patients with equally high plasma heparin concentrations, but they did not have increased bleeding. After protamine administration, coagulation parameters (ACT, SCT, APTI’, TT, TEG) did not differ significantly between the groups during the study period (Fig 1). The appearance of minima1 amounts of heparin after initial neutralization did not affect the mean value of ACT. However, significantly higher peak heparin concentrations were observed in those patients having their ACTS elevated more than 10 seconds above the baseline (0.059 ? 0.008 U/mL [n = 181 v 0.036 * 0.006 U/mL [n = 121; P < 0.05). Concomitant slight changes toward decreased coagulability were observed in SCT, APTT, logTT, and, from TEG tracing parameters, c-w-angle.The changes from the preoperative values in most of these parameters correlated significantly with the peak free heparin levels (Table 3). The best correlation was with 1ogTT and the weakest with ACT. Six hours after protamine administration, the coagulation times had shortened as compared with the preoperative values, but this change was significant only in ACT and APTT in both groups and in TT in the protamine chloride group. Three patients, all receiving protamine chloride, underwent reexporation due to excessive bleeding. A single bleeding site was identified and hemostasis was achieved in all these cases. These patients were included in the study and samples for coagulation analyses were collected up to
Table 2. Comparison of the Four Methods for Estimation of the Protamine Dose Needed for Neutralization Method
of
Estimation
Fixed protamine/heparin*
of Heparin
Protamine
Protamine
Chloride (mgl
Sulfate (mg)
396.7 + 31.411
396.6 -t 26.711
(1 .oo f 0.00)
(1.00 -t 0.00)
Heparin-ACT curvet
306.1 + 37.0
269.9 -t 15.7
(0.77 -t 0.06)
(0.71 f 0.04)
Plasma heparin level*
209.7 IT 15.1ll
210.5 + 16.411
(0.54 + 0.03)
(0.55 f 0.04)
Protamine titration§
152.9 f 20.911
151.8 zk 20.411
(0.40 + 0.05)
(0.38 + 0.05)
NOTE. Values are given as mean 2 SEM. The protamine/heparin
total
dose ratios (mg/lOO U) are presented in parentheses. ‘1 mg of protamine for 100 U given of heparin. tProtamine/heparin *Directly measured
ratio of 1.3 mgll00 U. plasma heparin and application of protamine/
heparin ratio of 1.3 mg/lOO U. 01.3 mg of protamine for 100 U of heparin estimated indirectly by protamine titration. llf < 0.05 v protamine dose obtained from heparin-ACT curve. llP < 0.001 v protamine dose obtained from heparin-ACT curve.
increased significantly in both groups after initial neutralization. Two hours after protamine administration, the mean heparin concentration was 0.045 f 0.01 U/mL in the protamine chloride group and 0.034 f 0.07 U/mL in the protamine sulfate group. An additional protamine dose of 25 mg was given to one patient with excessive bleeding in each group. An elevated
240
Fig 1. Plasma heparin, celiteACT, SCT, APlT, TT, and TEG a-angle after heparin neutralization with protamine sulfate (n = 16) or protamine chloride (n = 14). ?? **p < 0.001, ?? *p < 0.01, *P < 0.05 compared with the values 20 minutes after protamine (the 0 time). Values are given as mean 2 SEM.
1
---t
T
160
’
0
3 *i&l;.
4
*
mlhta chlorldc
KUITUNEN ET AL
224
Table 3. Correlation of the Peak Postprotamine Concentrations
Free Heparin
with the Corresponding Changes in Coagulation
Parameters From the Preoperative
Baseline Values
Peak Heparin Versus ASCT
AACT
AAPTT
ATEGa
AlogTT
r
0.356
0.569
0.535
0.603
0.801
P
0.053
0.001
0.002
< 0.001
H
EPARIN REBOUND is a potential bleeding disorder associated with cardiopulmonary bypass (CPB). It is defined as the resurgence of the anticoagulant activity in a patient whose blood heparin has been adequately neutralized with protamine.’ There is no uniform theory of the cause of heparin rebound, but the following mechanisms have been suggested: (1) heparin may be released by red blood cell breakdown’; (2) a proportion of heparin’s effect may temporarily be neutralized by some endogenous antagonists, and heparin activity is again demonstrated when these antagonists have been cleared from plasma3; (3) heparin may redistribute into the circulation from extravascular spaces or from endothelium4.5; or (4) a naturally occurring protaminase in plasma may cause breakdown of protamine, thereby leaving heparin free in plasma.‘~6~7 Various protocols have been used to neutralize heparin with protamine after CPB. The approach of administering a fixed dose of protamine for each 100 U of heparin given has been partially replaced by methods determining protamine dose on the basis of measured heparin activity, heparin concentration, or by using a protamine titration in vitro.8-‘3 These methods have resulted in decreased protamine/ heparin dose ratios.8-1’Although the initial neutralization might be adequate with these decreased protamine doses, it is possible that a price has to be paid in the form of an increased incidence of heparin rebound.r4 Protamine is available as sulfate and chloride salts. Protamine chloride has been shown to have a more rapid onset of neutralization in patients than protamine sulfate.” In vitro, protamine chloride is more stable in plasma than protamine sulfate.‘6.‘7 Based on these studies, heparin rebound has been suggested to occur when protamine sulfate but not protamine chloride is used to neutralize heparin.16 The present study was undertaken to compare the clinical effects of the two protamine salts with special reference to their effect on the occurrence of heparin rebound. MATERIALS
AND METHODS
After institutional approval, neutralization of heparin was studied in 32 patients undergoing coronary artery bypass surgery. None of the patients received antiplatelet drugs, oral anticoagulants, or heparin prior to surgery, and none of them had preoperative
behavior of the plasma heparin level, and the various coagulation parameters did not differ significantly between the groups. Two hours after neutralization, a small and temporary increase of plasma heparin level was observed in both groups. The postoperative blood losses were comparable in both groups. Thus, protamine chloride was not a clinically superior antidote to heparin than protamine sulfate. The observed heparin rebound levels were low and clinically insignificant in terms of blood loss, but they were associated with slight changes in coagulation monitoring. Copyright o 1991 by W.B. Saunders Company
alterations in blood coagulation (platelet count, activated partial thromboplastin time [APT-T], prothrombin time [PT], plasma fibrinogen level) or liver function. The patients were randomly assigned to two groups according to the protamine salt to be used for heparin neutralization: either protamine sulfate or protamine chloride was given in a double-blind manner using the dose ratio of 1.3 mg protamine/lOO U heparin as estimated from the heparinactivated coagulation time (ACT) curve (see below). Two patients of the 32 were excluded because of failure of compliance with the protocol by the intensive care unit (KU) staff not participating in the study. Clinical details of the 30 patients studied are shown in Table 1. The patients were premeditated with morphine, 0.2 mg/kg, and scopolamine, 0.006 mglkg, intramuscularly. Anesthesia was induced with fentanyl? 50 ug!kg, diazepam, 0.1 mgkg, and pancuronium, 0.1 mg/kg, and maintained with additional doses of fentanyl and pancuronium, as well as low concentrations of enflurane. Patients were ventilated with a mixture of oxygen and air (F,O, 0.5). Dopamine and nitroglycerin were given postoperatively when clinically indicated. After induction of anesthesia, 1 U of blood was withdrawn and replaced with 1,000 mL of Ringer’s acetate solution. The perfusion circuit included a Shiley 100 A bubble oxygenator (Irvine, CA) that was primed with 2,000 mL of Ringer’s acetate and 5,000 U of heparin. The patients were subjected to moderate systemic hypothermia (28°C to 3O”C), topical myocardial cooling, and cold crystalloid potassium cardioplegia solution. During the extracorporeal circulation, the hematocrit was maintained between 20% and 25%, with packed red cells being added to the oxygenator if necessary. Anticoagulation was achieved with porcine mucosal sodium heparin (N.V. Organon, Oss, The Netherlands). After determination of the baseline celite ACT, the heparin-ACT dose response curve was constructed with the aid of an initial heparin dose (200 U/kg) as recommended by Bull et al.” The ACT was maintained greater than 480 seconds during CPB. After discontinuation of
From the Department ofAnesthesiology Helsinki University Central Hospital, and Finnish Red Cross Blood Transfusion Service, Helsinki, Finland. Supported in part by the Paolo Foundation and Helsinki University Central Hospital, Helsinki, Finland. Address reprint requests to Anne H. Kuitunen, MD, Department of Anesthesiology, Helsinki University Central Hospital, Haartmaninkatu 4, SF-00290 Helsinki, Finland. Copyright 0 1991 by W.B. Saunders Company 1053-0770/91J0503-0.00l0 221
222
KUITUNEN ET AL
Table 1. Clinical Details of the Patients Protamine Chloride
Protamlne Sulfate
No. of patients
14
16
Sex (female/male)
519
818
Age (vr) Weight (kg)
57.2 + 2.3
55.9 ?X2.0
78.0 + 3.0
77.4 t 3.7
Height(m)
1.68 + 0.02
1.67 + 0.02
Perfusion time (min)
126?
7
Surgical time (min)
296 ? 13
276 f 13
127 + 9
No. of grafts
3.6 2 0.4
3.9 +- 0.4
NOTE. Values are given as mean t SEM.
CPB, heparin was neutralized according to the study group either with protamine chloride or protamine sulfate. Immediately after weaning the patient from CPB, the heparin level in the circulating blood was determined according to the heparin-ACT dose response curve.8 Thereafter, protamine, 1.3 mgilO0 U of calculated heparin, was given as an infusion into a peripheral vein over 15 minutes. Simultaneously with the administration of this protamine infusion, the autologous blood unit was given and the available residual oxygenator perfusate was returned to the patient. The amount of this perfusate was not different between the two groups: 645 ? 10.5 mL (mean t SEM) in the protamine sulfate group and 690 + 75 mL in the protamine chloride group. An additional dose of either protamine sulfate or protamine chloride (according to the group assigned) was given to the patient after the administration of the perfusate. This additional dose of protamine was 25 mg/500 mL of perfusate. All the blood samples were collected into plastic syringes through an unheparinized radial artery catheter. Samples were taken at the following times: (1) preoperatively after induction of anesthesia; (2) after the end of CPB before the infusion of protamine; (3) 20 minutes; (4) 80 minutes; (5) 140 minutes; (6) 200 minutes; (7) 260 minutes; and (8) 380 minutes after the end of the protamine infusion. Whole blood coagulation time was determined by a semiactivated technique with special glass tubes containing either diatomaceous earth (ACT) or saline solution (saline coagulation time [SCT]) and a small magnet. The tubes were rotated at 37°C in a Hemochron device (International Technidyne Corporation, Edison, NJ). A thromboelastograph (TEG) tracing (Hellige Company, Freiburg, Germany) was obtained from titrated whole blood. Coagulation was initiated by adding calcium chloride to the cuvettes. TEG tracings were analyzed by measuring reaction time (R, min), coagulation time (R + K, min), clot formation rate (alpha angle, “) and maximum amplitude (MA, mm).” ACT, SCT, and TEG tracings were performed from two identical blood samples at the same time. Protamine was added to one of the samples (protamine concentration in the sample was 5 kg/mL) and the samevolume of saline solution to the other. The purpose of this protamine administration was to demonstrate the presence of free heparin in the blood sample. The APTT and thrombin time (TT) were determined from titrated frozen (-70°C) plasma samples using standard techniques. TT was determined by adding thrombin in a concentration (10 National Institute of Health UnitsimL) that produced a clotting time of 5 seconds for normal human plasma. Even with this high thrombin concentration, the distribution of TTs was skewed toward long clotting times necessitating logarithmic transformation to reach normal distribution for statistical comparisons. Plasma heparin level was determined by a photometric assay using the
chromogenic substrate S 2222, which shows excess of activated Factor X (Heparin Coatest, Kabi, Sweden).” In order to compare various methods of protamine dose estimation, the hypothetical protamine doses needed by the patients were calculated in two alternative ways: (1) by determining protamine response in vitro,” and (2) by using directly measured plasma heparin levels (chromogenic heparin assay).” Protamine response in vitro was evaluated using two ACTvalues: a status (heparinized) ACT and an ACT obtained after in vitro addition of liquid protamine (protamine concentration in the sample, 9.1 &mL). These ACT’s were determined simultaneously using the standard Hemochron procedure. Protamine concentration needed to bring the ACT to baseline (unheparinized) values was extrapolated from the data. To calculate the dose of protamine, the patients were assumed to be normovolemic and their blood volume was calculated using a formula previously reported by Allen et al.“’ The same blood volume was used in calculation of the protamine dose on the basis of plasma heparin level. In these protamine dose calculations, a protamineiheparin dose ratio of I .3 mgi100 IU was used. Blood loss through the chest tubes was measured from the end of the operation until the last blood sample. Postoperatively, hematocrit was maintained at approximately 30% by giving packed red cells and Ringer’s acetate. Whole blood was used to replace blood loss. The amount of transfused blood products was recorded during the study period. After the initial neutralization of heparin. extra protamine (2.5 mg) was given to bleeding patients (chest tube drainage > 200 mL/h) if the TEG values were outside the normal of the ACT and SCT, and if the in range, there was prolongation vitro test protamine (5 &mL) decreased the hypocoagulability. One- and two-way analysis of variance (ANOVA) for repeated measures were first applied for the statistical comparisons within and between the groups, respectively. In case of significance after one-way ANOVA (P < 0.05), a paired t test with Dunnett‘h allowance for multiple comparisons was used to seek contrasts to the control values. Between the groups the comparisons for the individual data points were made with unpaired t test using Bonferroni’s allowance as necessary. Simple least squares linear regression analysis was used to correlate the variables.
RESULTS
The total dose of heparin given was 514 2 39 U/kg in the protamine chloride group and 516 ‘- 29 U/kg in the protamine sulfate group. The amount of protamine actually given to the patients was calculated from the heparin dose-ACT response curve applying the protamine/heparin ratio of 1.3 mg/lOO U. For comparison, the required protamine dose from the protamine titration curve or by using the directly measured plasma heparin levels was also calculated applying the protamine/heparin ratio of 1.3 mg/lOO U. The fourth method used for calculation of the hypothetical protamine dose was the application of the widely used fixed protamine/given heparin dose ratio (1 mg protamine/lOO U of given heparin). The doses of protamine and the total protamine/heparin ratios estimated by these four methods are shown in Table 2. With initial neutralization, the mean heparin concentration decreased from 3.25 -f 0.29 UimL to 0.018 ? 0.006 U/mL in the protamine chloride group and from 3.26 -t 0.19 U/mL to 0.010 ? 0.03 U/mL in the protamine sulfate group. This initial response and the subsequent plasma free heparin levels did not significantly differ between the groups (Fig 1). However, plasma heparin concentrations
HEPARIN REBOUND IN CARDIAC SURGERY
223
circulating free heparin concentration was shown in both these patients (0.056 U/mL and 0.119 U/mL). There were also other patients with equally high plasma heparin concentrations, but they did not have increased bleeding. After protamine administration, coagulation parameters (ACT, SCT, APTI’, TT, TEG) did not differ significantly between the groups during the study period (Fig 1). The appearance of minima1 amounts of heparin after initial neutralization did not affect the mean value of ACT. However, significantly higher peak heparin concentrations were observed in those patients having their ACTS elevated more than 10 seconds above the baseline (0.059 ? 0.008 U/mL [n = 181 v 0.036 * 0.006 U/mL [n = 121; P < 0.05). Concomitant slight changes toward decreased coagulability were observed in SCT, APTT, logTT, and, from TEG tracing parameters, c-w-angle.The changes from the preoperative values in most of these parameters correlated significantly with the peak free heparin levels (Table 3). The best correlation was with 1ogTT and the weakest with ACT. Six hours after protamine administration, the coagulation times had shortened as compared with the preoperative values, but this change was significant only in ACT and APTT in both groups and in TT in the protamine chloride group. Three patients, all receiving protamine chloride, underwent reexporation due to excessive bleeding. A single bleeding site was identified and hemostasis was achieved in all these cases. These patients were included in the study and samples for coagulation analyses were collected up to
Table 2. Comparison of the Four Methods for Estimation of the Protamine Dose Needed for Neutralization Method
of
Estimation
Fixed protamine/heparin*
of Heparin
Protamine
Protamine
Chloride (mgl
Sulfate (mg)
396.7 + 31.411
396.6 -t 26.711
(1 .oo f 0.00)
(1.00 -t 0.00)
Heparin-ACT curvet
306.1 + 37.0
269.9 -t 15.7
(0.77 -t 0.06)
(0.71 f 0.04)
Plasma heparin level*
209.7 IT 15.1ll
210.5 + 16.411
(0.54 + 0.03)
(0.55 f 0.04)
Protamine titration§
152.9 f 20.911
151.8 zk 20.411
(0.40 + 0.05)
(0.38 + 0.05)
NOTE. Values are given as mean 2 SEM. The protamine/heparin
total
dose ratios (mg/lOO U) are presented in parentheses. ‘1 mg of protamine for 100 U given of heparin. tProtamine/heparin *Directly measured
ratio of 1.3 mgll00 U. plasma heparin and application of protamine/
heparin ratio of 1.3 mg/lOO U. 01.3 mg of protamine for 100 U of heparin estimated indirectly by protamine titration. llf < 0.05 v protamine dose obtained from heparin-ACT curve. llP < 0.001 v protamine dose obtained from heparin-ACT curve.
increased significantly in both groups after initial neutralization. Two hours after protamine administration, the mean heparin concentration was 0.045 f 0.01 U/mL in the protamine chloride group and 0.034 f 0.07 U/mL in the protamine sulfate group. An additional protamine dose of 25 mg was given to one patient with excessive bleeding in each group. An elevated
240
Fig 1. Plasma heparin, celiteACT, SCT, APlT, TT, and TEG a-angle after heparin neutralization with protamine sulfate (n = 16) or protamine chloride (n = 14). ?? **p < 0.001, ?? *p < 0.01, *P < 0.05 compared with the values 20 minutes after protamine (the 0 time). Values are given as mean 2 SEM.
1
---t
T
160
’
0
3 *i&l;.
4
*
mlhta chlorldc
KUITUNEN ET AL
224
Table 3. Correlation of the Peak Postprotamine Concentrations
Free Heparin
with the Corresponding Changes in Coagulation
Parameters From the Preoperative
Baseline Values
Peak Heparin Versus ASCT
AACT
AAPTT
ATEGa
AlogTT
r
0.356
0.569
0.535
0.603
0.801
P
0.053
0.001
0.002
< 0.001
