MODE OF ACTION OF TROMEXAN

MODE OF ACTION OF TROMEXAN

821 PREPARATIONS MODE OF ACTION OF TROMEXAN R. B. HUNTER M.B.E., M.B., F.R.C.P. Edin. PROFESSOR OF PHARMACOLOGY AND THERAPEUTICS G. R. TUDHOPE B.Sc...

504KB Sizes 2 Downloads 38 Views

821 PREPARATIONS

MODE OF ACTION OF TROMEXAN R. B. HUNTER M.B.E., M.B., F.R.C.P. Edin. PROFESSOR OF PHARMACOLOGY AND THERAPEUTICS

G. R. TUDHOPE B.Sc., M.B. St. And. ASSISTANT, DEPARTMENT

OF PHARMACOLOGY AND

THERAPEUTICS

UNIVERSITY OF ST. ANDREWS MEDICAL SCHOOL, DUNDEE

IT is now established that in the treatment of thromboembolic phenomenaTromexan ’(ethyl biscoumacetate, bis 3-3’-(4 oxycoumarinyl) ethyl acetate) is as effective as, and less dangerous than, dicoumarol, over which it

advantages. First, its action in reducing proactivity is much quicker (18-48 hours, coinwith 48-72 hours). Secondly, because of its more pared rapid destruction and excretion by the body, prothrombin activity and complications can be better controlled by varying the oral dosage. When the drug is discontinued prothrombin activity will increase sufficiently in 8-12 hours to control bleeding (Stirling and Hunter has two

thrombin

1951).).

-

The

original studies of the effect of dicoumarol in (Butt et al. 1941) showed a gradual fall of prothrombin activity in 48-72 hours. This suggests that dicoumarol might act by preventing the production of prothrombin by the liver, and that the delay of 48 hours is accounted for by the consumption of prothrombin already formed. In our early work with tromexan we noted the now well recognised rapid fall of prothrombin activity. In some of our experimental studies initial dosage of 1800 mg. brought about low levels of prothrombin activity in 9-12 hours. Two possibilities suggested themselves : either the life of prothrombin was much shorter than was originally thought, or tromexan acted mainly not on protlirombin but on some man

accelerator factor, interference with which caused the one-stage prothrombin test to give low estimates of

prothrombin activity.

By different methods Owren (1944 and 1949) and Quick (1943) arrived at the conclusion that normal plasma contained a prothrombin accelerator which

disappeared

from stored

plasma.

The term" tromexan-treated plasma " is used for taken from a patient to whom tromexan has been administered. The addition of tromexan to plasma in vitro is without effect on coagulation.

plasma

Factor V (Owren ca7zd Aas 1951).-This is prepared from ox-plasma. Nine volumes of ox blood are mixed with one volume 2-5% (w/v) solution of potassium oxalate monohydrate. After being spun, the plasma is first clarified by passing it through an 18% asbestos filter-pad, and then it is passed through a filter-pad containing 50% asbestos, which Owren states will adsorb prothrombin and factor vii but allow

passage of factor v. The filtrate is neutralised to 7-3 by N/2HCI, and then stored in small glass tubes at below -10°C.

the

pH

Factor VII (Owren and Aas 1951).-Stored human serum is used as a source of this factor. A small amount of thromboplastin (3 ml. to 100 ml. blood) is added to normal human blood immediately after withdrawal, and the mixture is stirred. Prothrombin disappears almost completely in 30 minutes, and any remaining factor v disappears after storage for a week.

Thromboplastin.-A commercial preparation of rabbit brain is used-’ Thrombokinase ’ (Geigy). One tablet (10 mg.) is crushed in a small test-tube and mixed with 2-5 ml. of normal saline. The suspension is warmed in a water-bath at 37°C for 10 minutes and is then ready for use. Barbitone Buffer Solution (Owren 1949).-This solution is prepared by adding 430 ml. JV/10 HCI and 5-67 g. NaCl to 570 rnl. 0-10 JYI sodium diethyl barbiturate and diluting the mixture with an equal volume of 0-9% NaCl solution. METHODS OF PROTHROMBIN ESTIMATION

tlethod.-A mixture of 0.2 ml. nrl. thromboplastin suspension is incubated in a water-bath at 37°C for 5 minutes. Then 0.2 ml. 0.2% calcium chloride solution is added and the time required for clotting is measured. The percentage prothrombin activity can be estimated for each plasma by the use of a correlation graph which is prepared from the average result obtained by estimating the clotting of at least ten dilutions of normal

(1) jtfo
Quick’s

0-2

plasma. (2) Owren’s

feod.—The plasma to be tested is obtained from oxalated blood by centrifuging, and is then diluted with the barbitone buffer solution to give 1 : 10 dilution of plasma. A small sample, 0-2 ml., is mixed with 0-2 ml. factor-v preparation and 0-2 ml.

The existence of this

prothrombin accelerator-pro-accelerin (Owren prothrombin accelerator (Fantl and Nance

1949), 1946),

factor v (Owren 1947)-is

now accepted, and in this paper it is termed factor v. Warner et al. (1939a and b) have suggested that there was another prothrombin accelerator which is stable-co-thromboplastin (Mann and Hurn 1951), S.P.C.A. (Alexander et al. 1949), proconvertin (Owren 1947), factor VII (Koller et al. 1951), and in this paper it is referred to as factor VII. Owren (1950) and Koller et al. (1951) found that during dicoumarol therapy the concentration of factor v is practically unaltered but that prothrombin and also, more notably, factor VII are decreased. Owren (1949) has also suggested that the one-stage Quick (1935) method of estimating prothrombin-time and the two-stage method of Warner et al. (1936) can be improved by the addition of factor v to the clotting system. Owren claims that this method gives an exact quantitative estimation of prothrombin regardless of the amount of factor v present. We here set out information concerning : (1) the practical merits of Owren’s and Quick’s methods of

determining plasma ; on

the

prothrombin, using tromexan-treated (2) the effect of factor v and factor VII clotting-times of tromexan-treated plasma. and

Results of prothrombin tests by Quick’s method and by Owren’s method in 80 samples of tromexanised plasma.

822

suspension. The mixture is placed in a water-bath at 37°C for 5 minutes, 0-2 nil. 0.33% calcium chloride solution is added, and the time required for clotting is measured. A correlation graph is prepared from dilutions of normal plasma. The clotting-time of 1 : 10 dilution of normal plasma in barbitone buffer is regarded as equivalent to 100% prothrombin activity. Higher dilutions of the normal plasma are made with a barbitone buffer containing 30 mg. potassium oxalate in 100 ml.

thromboplastin

TABLE

II-EFFECT

(SEC.)

OF

VII

FACTOR

ON

THE

CLOTTING-TIME

OF TROMHXAN-TREATED PLASMA

solution. The results

accompanying figure shows the distribution of by Quick’s and Owren’s methods of determining prothrombin activity in 80 cases where tromexan was given in varying amounts. Statistical analysis of the figures does not reveal any significant difference between the two sets of observations. In the therapeutic range of activity-i.e., 10-25% of prothrombin activitythere were 37 results by one method and 39 by the other. The only pronounced difference between two groups of results was at 15-20% and 20-25% of prothrombin activity. In our opinion this is immaterial. The experiment also provides evidence that factor v is not affected by the administration of tromexan. In view of our experimental results, we believe that it is unnecessary to add a sample of factor v to every routine prothrombin test, and that Owren’s one-stage prothrombin test has no practical advantage over that of Quick’s.

Owren’s method may have certain definite advantages from the research point of view. A typical series of results is shown in table I. The clotting-time of normal plasma by the modified Quick method using thromboplastin (Geigy) is 15-6-18.2 seconds. The addition of factor v does not shorten the clotting-time significantly : when factor vii is added there is a distinct shortening of the clotting-time which

(SEC.)

TABLE I-CLOTTING-TIME AND WITH THE ADDITION

OF NORMAL PLASMAS WITHOUT

OF

FACTOR

V OR FACTOR VII

T.P. = Tromexan-treated plasma. F. v. F. vii = Factor VII. T. = Thromboplastin.

=

Factor v. Ca Calcium. =

Several of the

clotting-times when factor vii is added region of 14 seconds, which is shorter than the clotting-time, measured by the modified Quick method, of normal plasma using thromboplastin (Geigy) are

in the

(table II). It is evident therefore that the addition of serum factor vii will greatly shorten the clottingtimes of tromexan-treated plasma. The only adequate explanation for our findings is that the prothrombin of tromexan-treated plasma is still present but its conversion to thrombin is delayed because of a deficiency of factor vii. Owen and Bollman (1948) and Koller et al. (1951) describe similar findings for dicoumarol-treated

containing

plasma. DISCUSSION .

Until 1943 it

tends to approach the optimum clotting-time of 12-15 seconds (Biggs et al. 1953). Further work on this aspect of blood coagulation will be described in a later communication.

generally believed that the low clotin dicoumarol poisoning and in of blood ting power vitamin-K deficiency was due to a fall in prothrombin It was believed that even in the presence content. of an abundant supply of thromboplastin and calcium ions thrombin was formed in insufficient amounts to ensure proper conversion of fibrinogen to fibrin. The work of Quick, Owren, and others suggests that what is estimated as prothrombin " may in fact be a complex of two or more substances. We have prepared by Owren’s method plasma and serum samples which contain factor v and factor vn, the latter being free from prothrombin. When the factor-v preparation is added to tromexan-treated plasma the clotting-time is not shortened ; this finding agrees with those from experiments with dicoumaroltreated plasma and suggests that factor v is little, if at all, affected by tromexan. We have also reported evidence that factor vii which is present in plasma and serum (Koller et al. 1951, Biggs and Macfarlane 1953a and b) accelerates the clotting of tromexan-treated plasma and that this defect results from the administration of tromexan. This evidence as to the mode of action of tromexan agrees with that of Owren (1950) and Owen et al. (1951), who studied dicoumarol-treated plasma. Owen et al. state : was

"



PROTHROMBIN-TIME OF TROMEXAN -TREATED PLASMA WITH ADDED FACTOR V AND FACTOR VII

Specimens of plasma were obtained from patients receiving tromexan. Clotting-times were estimated by the method of Quick, modified as above and also by the addition of one other reagent. Thus 0.2 ml. factor-vn preparation or 0-2 ml. factor-v preparation, or in the control 0-2 ml. of normal saline, was added to 0-2 ml. plasma., 0-2 ml. thromboplastin suspension, and 0-2 ml. calcium chloride solution and the time taken for the mixture to clot determined in a water-bath at

37°C. The results of this experiment are shown in table II. The addition of plasma containing factor v did not make any significant difference to the clotting-times. Addition of serum samples containing factor vn shortened ,the clotting-times in every case, and statistical analyses of the two groups show that the result is highly significant.

,

"

The rate with which prothrombin converts to thrombin would seem to be a function of these accessory factors

823

exdmivety, for

when

normal plasma

was

mixed with

purified

HÆMORRHAGIC DIATHESIS DUE TO ABSENCE OF CHRISTMAS FACTOR*

prothrombin in various proportions, the conversion rate was independent of the prothrombin concentrations, but paralleled

of the accessory factors." in this field would at the of workers The majority the view that the various factors time accept present mentioned are prothrombin accelerators. Biggs and and b) and Biggs et al. (1953), howMacfarlane ever, have recently reported very good evidence suggesting that five factors are concerned in thromboplastin generation-the platelets, factor v, factor VII, antiIismophilic globulin, and the Christmas factor (Biggs et al. 1952). They suggest that brain thromboplastin, whether it be derived from animal or human source, is not really thromboplastin at all but is a mixture of Mtihsemophilic globulin, platelet factor, and the Christmas factor. Thromboplastin preparations, therefore, such as we have used in this study do not contain complete thromboplastin but are deficient in both factor v and factor vii. It is only on the addition of these factors that thromboplastin formation proceeds, and this reaction takes a short but definite time-10-12 seconds. The use of brain thromboplastin reveals factor-v and factor-vn deficiency as well as prothrombin deficiency. Thus thromboplastin generation does not

directly the concentrations

S.

place quickly in tromexan-treated plasma because deficiency of factor vii. There is also evidence {Koller et al. 1951, Owren and Aas 1951, Biggs and Macfarlane 1953a and b) that prothrombin is reduced in amount by the administration of dicoumarol or tromexan ;; but this reduction is neither so rapid nor so great as that of factor vii, particularly with tromexan.

M. M. P. PAULSSEN Pharm.D. Amsterdam

Groningen

OF BIOCHEMICAL LABORATORY

HEAD

PEDIATRICS PÆDIATRIC

CLINIC,

UNIVERSITY

OF AMSTERDAM

LATELY some cases have been reported- of a haemordiathesis which clinically resembled haemophilia but in which there was no deficiency of the antihaemophilic factor or any coagulation factor previously recognised. This condition has been attributed to the absence of another coagulation factor recently discovered in the plasma. The disease has been described by Aggeler et al. (1952), Schulman and Smith (1952), Biggs et al. (1952), and Poole (1953), and the familial haemorrhagic disorder described by Koller et al. (1950) is probably the same. Biggs et al. (1952) named the condition Christmas disease after their first patient and described seven cases. We report here a case of Christmas disease which illustrates the difference between it and another haemorrhagic diathesis (hypoproconvertinaemia) which has also been recently described for the first time.

rhagic



of

THE

CASE AND THE DIAGNOSIS

A boy, aged 10 years, the only child of healthy parents, showed the first signs of a haemorrhagic diathesis after an injury at the age of 15 months. Before this trauma, however, his parents had noticed that he was easily bruised.



"

This first

SUMMARY

Evidence is described which supports the view that tromexan acts on the mechanism of blood coagulation similarly to dicoumarol, and that its action is principally factor vii. A comparison of Quick’s and Owren’s methods of estimating " prothrombin " indicates that Quick’s onestage method is suitable for routine use. Reference is made to recent views on thromboplastin on

generation. We are grateful to Dr. Ann C. Donald for assistance in carrying out a number of estimations and to Mr. E. A. Asher, of Pharmaceutical Laboratories Geigy Ltd., for supplies of thromboplastin. Part of the expenses of the research were defrayed by a grant from Pharmaceutical Laboratories

Geigy Ltd.,

CREVELD

PROFESSOR OF

(1953a

take

VAN

M.D.

haemorrhage due to trauma was followed by many especially in midsummer and midwinter. The haemorrhages, especially from the nose, often began and stopped suddenly. Not rarely a haemorrhage occurred some time after a preceding haemorrhage had apparently stopped. The patient had also repeated haemorrhages into the joints (ankles and knees), but there was no ankylosis of these joints or restriction of their function, such as develops in haemophilia. Dr. G. J. van Lookeren Campagne, who sent us this patient, had many years ago found a prolonged coagulation-time and had diagnosed haemophilia. No haemorrhagic diathesis could be traced in the family. severe

ones,

When admitted to the Children’s Clinic the boy had of the venous (table I) a much prolonged

coagulation-time

TABLE I-RESULTS IN DIFFERENT DETERMINATIONS IN BLOOD

Manchester. REFERENCES

Alexander, B., de Vries, A., Goldstein, R. (1949) Blood, 4, 247. Biggs, R., Douglas, A. S., Macfarlane, R. G. (1953) J. Physiol. 119, 89. Dacie, J. V., Pitney, W. R., Merskey, C., O’Brien, J. R (1952) Brit. med. J. ii, 1378. Macfarlane, R. G. (1953a) Human Blood Coagulation and its Disorders. Oxford. (1953b) Personal communication. Butt, H. R., Allen, E. V., Bollman, J. L. (1941) Proc. Mayo Clin. —







— —

16, 388.

Fantl, P., Nance, M. (1946) Nature, Lond. 158, 708. Koller, F., Loeliger, A., Duckert, F. (1951) Acta hœmat. 6, 8. Mann, F. D., Hurn, M. (1951) Amer. J. Physiol. 164, 105. Owen, C. A. jun., Bollman, J. L. (1948) Proc. Soc. exp. Biol., N.Y. 67, 231.

— Magath, T. B., Bollman, 166, 1. Owren, P. A. (1944) Aarbok for Oslo ; p. 21.

J.

L.

(1951) Amer. J. Physiol.

det Norske

Videnskapsakademi.

(1947) Acta med. scand. suppl. 194. (1949) Scand. J. clin. Lab. Invest. 1, no. 1. (1950) Int. Congr. Hœmat. pp. 379, 475. Aas, K. (1951) Ibid, 3, 201. Quick, A. J., Stanley-Brown, M., Bancroft, F. W. (1935) Amer. J. med. Sci. 190, 501. — (1943) Amer. J. Physiol. 140, 212. Stirling, M., Hunter, R. B. (1951) Lancet, ii, 611. Warner, E. D., Brinkhous, K. M., Smith, H. P. (1936) Amer. J. Physiol. 114, 667. ( 939a) Proc. Soc. exp. Biol., N.Y. 40, 197. ( 939b) Amer. J. Physiol. 125, 296. —





-

— — — — — —

blood ; normal values for bleeding-time, blood-clot retraction, fibrinogen content, and thrombocyte-count ; normal capillary resistance; and insufficient consumption of

prothrombin on coagulation. boy was not (we concluded) suffering from haemophilia, because : The

(1) His plasma in various dilutions (table 11) exerted a normal coagulation-promoting activity on the blood of a known haemophiliac with a greatly prolonged coagulationtime. work was supported by a grant from the Nederlandse Organisatie voor toegepast Natuurwetenschappelijk Onderzoek (Dutch Organisation for Applied Scientific Research).

* This