To analyze the clinical manifestations of protein S deficiency, we evaluated 136 members of 12 families with the disorder. Seventy-one persons were found to be heterozygous for protein S deficiency, which is inherited as an autosomal dominant trait. Venous thrombotic events occurred in 39 patients (55%) and were recurrent in 77%. Most symptomatic patients had various combinations of deep venous thrombosis (74%), superficial thrombophlebitis (72%), and pulmonary embolism (38%), either in succession or simultaneously. On five occasions thrombosis was found at unusual sites, like the axillary, mesenteric, and cerebral veins. The age at the first thrombotic event ranged from 15 to 68 years (mean, 28 years), and at age 35 the probability to be still free of thrombosis was only 32%. Fifty-six percent of the thrombotic events were not preceded by a precipitating condition. In these respects protein S deficiency is similar to protein C deficiency.

The molecular defects in two congenital abnormal fibrinogens, Umuiden and Nimegen, were determined by sequence analysis of genomic DNA amplified by the polymerase chain reaction. Both fibrinogens were heterozygous, UImuiden having a BfiArg14-)Cys substitution and Nolmegen having a BfIArg44--Cys substitution. Clotting induced by thrombin or Reptilase was impaired in both fibrinogens, indicating defective fibrin polymerization. Immunoblot analysis of both purified fibrinogens demonstrated that some of the abnormal molecules were linked by disulfide bonds to albumin. In addition, abnormal high molecular weight fibrinogen complexes with Mrs between 600,000 and 700,000 were present.Fibrinogen-albumin and high molecular weight complexes were also detected in the patients' plasmas. Quantative analysis demonstrated that of the total plasma fibrinogen in the Umuliden patient, 20% was linked to albumin and 10% was present as high molecular weight complexes. In plasma Nijmegen, 13% was linked to albumin and 15% was present as high molecular weight complexes. These results demonstrate that the additional abnormal cysteine in fibrinogens Umuiden and Nimegen resulted in the formation of disulfide-linked complexes with other proteins, predominantly albumin. We also found that a significant fraction ofthe abnormal fibrinogen molecules contained free sulfhydryl groups. These findings complicate interpretation of functional studies of these altered fibrinogens.The fibrinogen molecule is a dimer of three polypeptides, Aa, Bf3, and y, linked together by 29 disulfide bonds (1). During coagulation fibrinogen is converted to an insoluble fibrin matrix by thrombin-catalyzed removal of fibrinopeptides from the Aa and the BP chains. Congenital dysfibrinogenemia is a disorder in which a fibrinogen structural abnormality results in altered functional characteristics. A number of abnormal fibrinogens have been described (2), some associated with bleeding disorders and some associated with thrombosis. The structural defects for -75 cases are known, with 21 cases of arginine substituted by cysteine-14 cases of

SummaryProtein S, a vitamin K-dependent coagulation factor, is involved in the regulation of the anticoagulant activity of activated protein C.Using an immunoradiometric assay for total protein S in plasma we identified 14 patients (7 male and 7 female) in three unrelated Dutch families as fulfilling the criteria for an isolated protein S deficiency. In 9 patients who were not receiving oral anticoagulant treatment the mean total protein S antigen concentration was 0.50 ± 0.08 U/ml (± S.D.) and the calculated free protein S concentration was 0.15 ± 0.01 U/ml (± S.D.). In the five patients who were on oral anticoagulant treatment the mean total protein S antigen was 0.23 ± 0.05 U/ml (± S.D.).Seven of the 14 patients had a history of venous thromboembolism occurring at a mean age of 25 years and often without an apparent cause. Protein S deficiency is inherited as an autosomal dominant trait.

SummaryHeparin cofactor II (HC II) levels were measured by electroimmunoassay in healthy volunteers, and patients with liver disease, DIC, proteinuria or a history of venous thrombosis. Analysis of the data in 107 healthy volunteers revealed that plasma HC II increases with age (at least between 20 and 50 years). HC II was found to be decreased in most patients with liver disease (mean value: 43%) and only in some patients with DIC. Elevated levels were found in patients with proteinuria (mean value 145%). In 277 patients with a history of unexplained venous thrombosis three patients were identified with a HC II below the lower limit of the normal range (60%). Family studies demonstrated hereditary HC II deficiency in two cases. Among the 9 heterozygotes for HC II deficiency only one patient had a well documented history of unexplained thrombosis. Therefore the question was raised whether heterozygotes for HC II deficiency can also be found among healthy volunteers. When defining a group of individuals suspected of HC II deficiency as those who have a 90% probability that their plasma HC II is below the 95% tolerance limits of the normal distribution in the relevant age group, 2 suspected HC II deficiencies were identified among the healthy volunteers. In one case the hereditary nature of the defect could be established.It is concluded that hereditary HC II deficiency is as prevalent among healthy volunteers as in patients with thrombotic disease. Further it is unlikely that heterozygosity for HC II deficiency in itself is a risk factor for the development of venous thrombosis.

SummaryTwo hundred and three patients with venous thrombophilia were investigated in order to find out whether an elevated plasma concentration of plasminogen activator inhibitor (PAI) could be a cause of their tendency to thrombosis. The patients were studied in an asymptomatic period about 3 months after their last thromboembolic episode. PAI activity was found to be elevated in 19 patients (9%), and a corresponding elevation of PAI-1 antigen was observed.In 16 out of the 19 patients with elevated PAI activity, followup could be performed after an additional asymptomatic period of about 1 year: in eight patients the elevation of PAI was transient and in eight it was persistent. Out of the eight patients with a persistent elevation of PAI, seven had a positive family history of thrombosis. Investigation of these families excluded a hereditary elevation of PAI activity in two families. In only two other families was elevated PAI activity also found among family members. The occurrence of elevated PAI activity, however, did not coincide with the occurrence of thrombosis in these individuals: except for the probands, all investigated family members who had a history of thrombosis had a normal PAI activity.We therefore conclude that, at least in our material, familial thrombophilia can not be attributed to an inherited, persistent elevation of the blood level of PAI.