The mode of action of phenylindanedione.

Since the discovery (Soulier and Gueguen, 1947) that phenylindanedione (P.I.D.) was an anticoagulant which effectively altered the one-stage prothrombin time it has been used extensively in the treatment of cases of thrombosis (Blaustein, 1950; Blaustein, Shnayerson, and Wallach, 1953; Dennis, Coon, Hodgson, and Duff, 1952; Toohey, 1953). Clinically it appears to produce the same effect as dicoumarol, but its action is more rapid and less prolonged (Blaustein, 1950; Jaques, Gordon, and Lepp, 1950; Bjerkelund, 1950). Sufficient experience has not yet been accumulated to compare its efficacy with that of tromexan (ethyl-biscoumacetate), though the available data suggest that both are valuable in the treatment of thrombosis, leading to a reduction in thromboembolic complications. There is no clear evidence of the precise mode of action of phenylindanedione. If a drug causes a prolongation of the clotting time in the one-stage prothrombin test this does not necessarily mean that prothrombin itself is involved. Confusion has arisen because some workers have made this assumption on the basis of the one-stage test, and as a result it is usually stated that P.I.D. depresses prothrombin production by the liver. It is now accepted that at least three factors influence the clotting time in this test, xiz., factor V, factor VII, and prothrombin itself. Although the test, which was originally designed (Quick, Stanley-Brown, and Bancroft, 1935) to measure prothrombin as such, remains a satisfactory one for routine control of anticoagulant therapy, it is not a satisfactory research or diagnostic method because it does not distinguish between deficiencies of the three factors. Owren (1951) and Owren and Aas (1951) stated that both prothrombin and proconvertin (factor VII) were decreased by phenylindanedione. Bjerkelund (1950) found that phenylindanedione depressed prothrombin in man, but the method he used (Owren, 1949) for estimating prothrombin makes no distinction between prothrombin and factor VII, being a measure of both. The same criticism applies to the experiments of Molho (1952) in rabbits: he reported that the main effect was on prothrombin. Richards (quoted by Jaques, 1950) and Seegers (quoted by Jaques, 1950) both found a marked fall in prothrombin in dogs given phenylindanedione, with a slight fall in factor V acglobulin). Once again, however, it would appear that " prothrombin " as described by these workers includes both prothrombin and factor VII, so that either or both could be affected by the administration of phenylindanedione. In order to determine which of the factors is principally affected by phenylindanedione the following methods were used.