THROMBOSIS RESEARCH 60; 63-70,199O 0049-3848/90 $3.00 + .OO Printed in the USA. Copyright (c) 1990 Pergamon Press pk. All rights reserved.
COMPARISON OF DIFFERENT PROTHROMBIN COMPLEX CONCENTRATES IN VITRO AND IN VIVO STUDIES
M. Kchler, M. Heiden, G. Harbauer*,
C. Miyashlta! S. Morsdorf,
B. Braun, P. ErneR*, E.Wenzel, S. Rose*, G. Pindur. Department of Clinical Haemostaseology *Department
and Transfusion Medicine
of Experimental Surgery
University of Saarland D- 6650 Homburg/Saar F.R.G.
(Received
19.7.1990;
accepted
in revised form 58.1990
by Editor H. Vinazzer)
ABSTRACT
The potency and tests for thrombogenicity
were studied prospectively
of each brand, A - G) prothrombin complex concentrates
in 7 different (two lots
(PCC). Human albumine (H) and a
factor IX concentrate (I), served as controls. The potency of coagulation factors and inhibitors varied considerably.
Two brands (E, F) contained no protein S, additionally
one brand con-
tained no protein C. Two preparations exhibited high amidolytic activities, especially towards the thrombin-sensitive
chromogenic
substrate S-2238, in vitro. These activities could be
quenched in part by the addition of hirudin or antithrombin
Ill. The heparin and antithrombin
Ill content of the PCCs was significantly different, and, after addition of antithrombin increase of thrombin-antithrombin
Ill an
Ill complexes in 2 preparations (B, D) was observed in vitro.
Additionally, three brands (8, D, F) caused more severe cardio-pulmonary
reactions in rabbits,
associated with an increase of fibrin split products for brands B and D. We conclude that the use of these preparations hypercoagulable
Key words: prothrombin
in patients, in whom an acquired
protein C or S defect, or a
state, can be suspected, cannot be recommended.
complex concentrate,
protein C, protein S, thrombogenicity,
animal model
64
PROTHROMBIN COMPLEX CONCENTRATES
Vol. 60, No. 1
INTRODUCTION Prothrombin complex concentrates (PCC) are frequently used for the treatment of hereditary coagulation factor deficiencies (factors II, VII, X, and IX) and even patients with protein C deficiency have received PCCs [I]. The high risk of transmitting viral infections has apparently been overcome by different virus inactivation procedures [for a review see 2, 31. The observation that virucidal treatment may also alter the protein structure of clotting factors is a cause for concern since a different biological behaviour, i.e. decreased half-life or increased thrombogenicity, may be the consequence [4]. The available clinical data on virusinactivated PCC’s, however, suggest a normal half-life in man [5, 6, 71. However, a problem hitherto unsolved is the thrombogenic potential of PCC’s. To date, more than 60 cases of disseminated intravascular coagulation (DIC), thrombosis or myocardial infarction in haemophiliacs attributable to the infusion of PCC have been reported to the ICTH subcommittee on FVIII/FIX [Lusher, personal communicarion 79891.The risk of developing thrombosis is even higher when treating patients with acquired defects of haemostasis such as those resulting from blood loss, oral anticoagulant therapy or severe liver disease. It is almost impossible, to determine the thrombogenicity of PCCs in these patients, thus numerous attempts have been made to measure thethrombogenicity invitro and in animal experiments. In order to control and compare the quality of virus-inactivated PCC we studied the in vitro potency and in vivo tolerance of these preparations in rabbits. METHODS 1. Design: Two lots of seven brands of PCCs were prospectively investigated in vitro (Al -Gl , A2-G2) and one lot of each brand was studied in an animal model (A2-G2). The brands are coded with letters A-G, the first series with 1, the second with 2, i.e. Al and A2. 2. Preparations: The PCC-preparations (A-G) were from the following manufacturers (or distributors): Alpha, Langen; Behringwerke, Marburg; Biotest, Dreieich; Immune, Heidelberg; Medac, Hamburg; Serapharm, Mijnster and Troponwerke, Win. Brands 6 and D were from one manufacturer. Additionally, albumin solution (H) and a factor IX concentrate (I), served as controls. The lyophilized prepations were dissolved in distilled water to obtain a factor IX activity of 25 U/ml (as declared by the manufacturers). This solution was injected i.v. in six rabbis. For in vitro studies further dilutions by Owren’s veronal buffer, pH 7.35, were carried out in order to meet the measuring range. 3. Assays: Coagulationfactorswere measured using one-stage assays and deficient plasmafrom Immune, Vienna, and a Schnitger Gross coagulometer as described previously [8]. The solution containing 25 U/ml FIX was further diluted using Owrens veronal buffer (usually 5 dilutions ranging from I:40 to 1:lOO). The calibration was performed using ‘Reference Plasma lOO%‘, lmmuno Vienna, which is calibrated against the WHO standards for factors II, VII, IX, and X (1st international standard). Protein C was measured by an enzyme immunoassay and a coagulometric assay [9] using reagents from Boehringer, Mannheim. Protein S was determined by electro-immunodiff usion and reagents from Immune, Heidelberg. The potency is given in percent of declared (factor IX) potency, i.e. when in 500 U PCC, as declared by the manufacturer, 400 U protein C antigen were found, the potency is stated to be 80% protein C. Thrombin generation time (TGt50) and non-activated partial thromboplastin time (NAPTT) were performed as decribed by Sas et a/. [9] and Prowse et a/. [lo], respectively. The chromogenic substrate assays were performed as follows: 0.1 ml of PCC (25 U/ml) and 0.8 ml assay buffer were incubated for 5 min at 37” C. Subsequently, 0.1 ml solution (3 mmol/l) of different chromogenic substrates was added. The chromogenic substrates utilized were S-2238, S-2222, S-2444, S-2368 and S-2251 from KabiVitrum, Munich, and the assay buffers were those recommended by KabiViirum. Antithrombin Ill and heparin activity were determined using chromogenic substrate test kits from Boehringer, Mannheim. Thrombin-antithrombin lllcomplexes (TAT) were measured using an enzyme immunoassay from Behringwerke, Marburg [I 11. Plasma D-dimer levels were determined by an enzyme immunoassay from Boehringer, Mannheim [13]. Prior to infusion of PCC, the mean D-dimer levels were 5,2 ng/ml in 54 rabbits (standard deviation 6,8 ng/ml). 3. lnhlbltlon studies: The activity against the chromogenic substrate S-2238 was determined to be equivalent less than 100 mu/ml thrombin. The amount of antithrombin Ill (AT Ill) or hirudin required to neutralize this amount of thrombin was added to PCC dilutions. Recombinant hirudin was kindly provided
Vol. 60, No. 1
PROTHROMBIN
COMPLEX CONCENTRATES
65
by Professor Markwardt, Erfurt. The preparation was dissolved in distilled water and tri-fluoric-acid and subsequently diluted in saline to obtain a final concentration of 1000 antithrombin units per ml. The reaction mixture consisted of 0.75 ml assay buffer, 0.05 ml hirudin solution, and 0.1 ml PCC (25 U/ml). After incubation at 37°C for 5 min, O.lml of chromogenic substrate solution (S-2236, 3mmol/l) was added and the color reaction recorded. AT Ill concentrate (KabiViirum, Munich) was diluted to a concentration of 1 U/ml and heparin (Roche, Grenzach) to 50 U/ml. Equal volumes of AT Ill/heparin solution (or buffer) and PCC were mixed and incubated for 30 min at 37” C. The activity against the chromogenic substrate and the concentration of TAT were determined as described above. 4. Animal studles: Each brand (A 2 - I 2) was administered to 6 rabbits. A dose of 50 U/kg body weight was injected into the marginal ear vein within 10 seconds [I 31. The following parameters were recorded for 10 min: blood pressure (BP), pulse rate (PR), ventilation frequency (VF) and electrocardiography. Additionally, a blood specimen (0.4 ml 0.109 mol/l sodium citrate and 1.6 ml blood) was drawn prior to and 10 min after the injection of PCC. In previous experiments, the limits of these parameters were established after injection of one lot of PCC into IO rabbits. The ‘normal range’ was defined as the mean values +/- 2 standard deviation (SD). Each measurement at 2, 5, and 10 min after injection of PCC outside these limits was counted as + 1. The final score is calculated by adding all counts in one rabbi (0= no change at all), and the score of one lot is the mean score obtained from 6 rabbits. When thrombin containing PCC’s or Wessler-serum is injected, a score >6 can be observed; the cardiopulmonary changes can be described by an initial rapid decrease of BP, associated by an increase of PR and VF, followed by an increase of BP. RESULTS 1. In vitro studies: The potencies of coagulation factors and inhibitors determined in our study are shown in table 1. Both lots of brands B and D had a low factor IX potency. Brands E and F had a low FVII potency, while brand G had a high content of FVII, when compared with the other concentrates. TABLE 1: Potency of coagulation factors, protein C activity, and protein S antigen concentration in % of declared potency of prothrombin complex concentrates (A-G). Additionally, 2 lots of a factor IX concentrate are shown. The results are expressed as percentage of declared (factor IX) potency. Brand
Factor
Protein
IX
II
VII
X
C
S
Al A2
106 157
123 109
25 59
266 336
115 360
106 46
Bl 82
31 22
227
62 99
232 174
116 200
147 21
Cl c2
134 96
139 161
50 27
142 155
144 360
140 126
Dl D2
46 37
226 267
72 70
216 196
66 192
110 16
El E2
62 69
102 125
16 11
131 105
102 126
10 16
Fl F2 Gl G2
107 176 119 150
116 206 126 221
4 4 137 116
124 62 151 119
5 2 102 256
5 4 213 167
I1 I2
124 166