Thrombotic Risk in Hereditary Antithrombin III, Protein C, or Protein S Deficiency
A Cooperative, Retrospective Study
Abstract A cooperative, retrospective study was performed on data from 8 coagulation laboratories and thrombosis units in Austria, Germany, and Switzerland to assess the risk for thrombosis in patients with hereditary antithrombin III (AT-III), protein C (PC), and protein S (PS) deficiencies; to compare the clinical manifestations of these 3 deficiency states; and to estimate the risk for development of thrombosis in high-risk situations. Two hundred thirty patients from 71 families with a documented hereditary deficiency of a natural coagulation inhibitor were included in the study. The patient group comprised 69 patients from 25 families with AT-III deficiency, 86 patients from 27 families with PC deficiency, and 75 patients from 19 families with PS deficiency. Diagnosis of the deficiency state was made at each participating center. Clinical data were documented in a questionnaire that was completed during each patient’s visit to the participating center. The questionnaires were sent to the coordinating center (Vienna) and analyzed centrally. The probability of developing thrombosis was 80% to 90% for all deficiency states by 50 to 60 years of age, and this figure did not change considerably after data from the propositi were excluded. AT-III–deficient females developed thrombosis earlier in life compared with PC- and PS-deficient females due to a high thrombotic risk during pregnancy (40% in patients with AT-III deficiency) and oral contraceptive use. The clinical features of thromboembolism were similar in the three deficiency states except for a higher frequency of superficial thrombophlebitis in patients with PC and PS deficiencies. Mesenteric vein thromboses occurred in 4% to 10% of patients and in 2 of 8 patients as a recurrent event. The recurrence rate was 63% (60% of recurrent events occurred spontaneously) and did not differ significantly among the three deficiency states. Before 14 years of age only 1 of 80 surgical procedures and 0 of 21 leg injuries were followed by thrombosis. After 14 years of age thromboembolic events occurred after abdominal surgery or leg injury in one third of patients. Between 40% and 50% of symptomatic patients reported that they felt handicapped by a postthrombotic syndrome. We conclude that diagnosis of a coagulation inhibitor deficiency state should be made before 14 years of age. During childhood thrombosis prophylaxis cannot be regularly recommended but should be instituted after 13 years of age during/after abdominal surgery, including appendectomy, and after leg injury in AT-III–, PC-, and PS-deficient patients. The high recurrence rate (60% spontaneous recurrence) and the relatively high frequency of mesenteric vein thrombosis as a recurrent event favor introduction of long-term oral anticoagulant treatment after the first thrombotic event in patients with a documented hereditary deficiency of AT-III, PC, or PS.
- antithrombin III deficiency
- protein C deficiency
- protein S deficiency
- natural inhibitor deficiency
- thrombotic risk
Reprint requests to I. Pabinger, MD, First Department of Medicine, Division of Hematology and Blood Coagulation, University of Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria.
Participants in the GTH Study Group on Natural Inhibitors are listed in the “Appendix.”
- Received August 11, 1995.
- Accepted February 29, 1996.
It is well established that patients who are deficient in one of the natural coagulation inhibitors (ie, AT-III, PC, or PS) have a markedly increased risk for venous thromboembolism.1 2 3 4 Retrospective studies were performed in large kindreds with PC5 and AT-III6 deficiencies or in multiple families with PS deficiency7 or were based on molecular biology8 and revealed an increased risk for thromboembolism in the inhibitor-deficient compared with the inhibitor-nondeficient individuals. These results were confirmed in a small, prospective cohort study.9 At present it has not been clearly established whether inhibitor-deficient patients should receive long-term oral anticoagulant treatment after their first thrombotic event, and opinions on this mode of treatment are conflicting.
There are only a few studies on the actual risk of thrombosis in inhibitor-deficient, related individuals.5 6 Furthermore, no direct comparison of the three deficiency states with regard to the lifetime probability of thrombosis has yet been performed in larger series of inhibitor-deficient patients. Nor has it been clearly established whether the thrombotic risk is different among patients with AT-III, PC, and PS deficiencies in high-risk situations and whether such an increased risk is already present during childhood and adolescence.
To answer these questions a multicenter, retrospective study that included a large number of carefully selected patients with an inhibitor deficiency was initiated. The purpose of this study was to (1) compare the thrombotic risk in hereditary AT-III, PC, and PS deficiencies; (2) compare the clinical manifestations in these three groups of patients; and (3) estimate the thrombotic risk in high-risk situations.
Patients and Study Design
The patients were enrolled on a multicenter basis between 1989 and 1993. Eight centers with laboratory and clinical experience in diagnosing and managing patients with natural inhibitor deficiencies participated in the study. Seven of the 8 participating centers performed a preselection procedure for determination of AT-III, PC, or PS deficiency in symptomatic patients. The main selection criteria were juvenile thrombosis (first thromboembolic event before 45 years of age) and a positive family history. This screening procedure, however, accounted for only 10% to 30% of all patients with DVT and/or PE.
Diagnosis of the deficiency state was made in each participating center. The aim of the investigation was to include only those families for whom a complete or nearly complete data set was available. Therefore, patients who met the following criteria were included in the study: (1) established hereditary inhibitor deficiency, ie, the deficiency state was diagnosed in at least 2 family members and it was possible to enroll at least 2 family members in the study. Isolated cases or families for whom only a single patient could be enrolled were excluded; (2) association of thrombotic disease with the deficiency state. All index patients had a history of thromboembolism; (3) availability of sufficient data on clinical events, in particular thromboembolic events and special-risk situations.
More than two thirds of the patients were recruited at 2 centers (Vienna and Frankfurt). Data on the entire group of patients who tested positive for an AT-III, PC, or PS deficiency were available from 6 of the 8 participating centers. At these 6 centers 2514 unrelated patients with a history of venous thromboembolism were investigated. Of these 2514 patients, 61 propositi (2.4%) with a hereditary inhibitor deficiency were recruited and included in the study. At the participating centers, 15 other propositi with a hereditary inhibitor deficiency were identified, but because they did not consent to the study they were not recruited. At the coordinating center 8 unrelated patients were excluded for the following reasons: 4 patients were isolated cases, it was not possible to recruit other family members for 3 patients, and sufficient data on clinical events were not available for 1 patient. Thus, 2.4% of 2514 patients were investigated and included in our study because of a history of venous thromboembolism. In the literature a prevalence of 3% for hereditary AT-III, PC, or PS deficiency has been found,10 11 12 and therefore, we were able to identify and subsequently include in the present study the vast majority of patients with a hereditary deficiency of AT-III, PC, or PS.
A total of 230 patients from 71 families were included in the study. The average number of study subjects per family was 3.2 (range, 2 to 10). Additional data on the patients and laboratory values are listed in Table 1⇓. Four of 25 families, each represented by 2 members, with AT-III deficiency (type IIa or IIb) and 3 of 27 families, each likewise represented by 2 members, with a PC deficiency had a type II deficiency. All patients with PS deficiency had a type I deficiency (decreased total and free PS antigen) as defined by the Working Party of the Subcommittee for Protein C and Protein S. Patients known to be homozygous for the deficiency were not included. Although molecular analysis was not required, it was available for the majority of type II patients.
A questionnaire was completed for each patient during a visit to the participating center. The date of occurrence and site of the thrombotic event(s); predisposing clinical situations such as surgery, trauma, pregnancy, or immobility; and the onset and duration of oral contraceptive use were recorded. Furthermore, the date of hospital admission and duration of hospitalization and anticoagulant treatment were documented. Because a complete set of objective end points for the presence or absence of thromboembolic disease was not available, a thrombotic event was categorized as “established” when diagnosis of a thrombotic event was confirmed by phlebography, Doppler sonography, or perfusion lung scanning or when the event led to hospitalization and treatment with heparin or oral anticoagulants. The diagnosis of superficial thrombophlebitis was based on typical clinical symptoms. An event (eg, surgery or trauma) was defined as the “triggering” event if it was followed by a thrombotic episode within 6 weeks.
Determination of AT-III, PC, and PS
Because this was a multicenter, retrospective study AT-III, PC, and PS measurements were performed by different methods. For determination of AT-III activity chromogenic assays from 3 different manufacturers (Boehringer Mannheim, n=24 patients; Kabi/Chromogenix, n=36; and Pharmacia LKB, n=9) were used. AT-III antigen was determined by either Laurell electroimmunoassay (antibody from Behring, n=30) or radial immunodiffusion according to Mancini (partigen plates from Behring, n=39). PC activity was determined with commercially available assays using Protac activation and measurement of activated PC with chromogenic substrates from 2 different manufacturers (Behring, n=55; Kabi/Chromogenix, n=31). PC antigen was determined by Laurell immunoelectrophoresis (antibody from Behring, n=52) or ELISA (Boehringer Mannheim, n=37). Free PS antigen was determined by Laurell immunoelectrophoresis (antibody from Baxter-Dade, n=4; American Diagnostica, n=66) or ELISA (Boehringer Mannheim, n=5) after polyethylene glycol precipitation of PS bound to C4b binding protein according to Comp.13 Total PS was determined by Laurell immunoelectrophoresis (antibodies from Boehringer Mannheim, n=2; American Diagnostica, n=11) or ELISA (Boehringer Mannheim, n=58). One laboratory determined bound instead of total PS by measuring PS in the polyethylene glycol precipitate by Laurell immunoelectrophoresis (antibody from Baxter-Dade, n=4).14
For statistical analysis sas software was used.15 Standard formulas were used to calculate frequencies, means, standard deviations, medians, and percentiles. Depending on the scale and type of data distribution, Student’s t test, the Wilcoxon rank-sum test, or χ2 analysis was used to compare the different groups. Survival analysis methods were used to determine the time from birth to the time of the first thrombotic episode, and the survival estimator was computed by the Kaplan-Meier method. To compare survival curves, the Wilcoxon test, which places more weight on early events, and the log-rank test, which places more weight on later events, were used.
Probability of Developing Thrombosis
More than half of the patients (142/230, or 62%) with AT-III, PC, or PS deficiency had experienced at least one thromboembolic event. The probability for the first thromboembolic episode is shown in the Figure⇓. Only a few inhibitor-deficient patients had their first thrombotic event before the age of 12. At 14 years of age the risk increases steeply, and by 50 to 60 years of age the probability is 80% to 90% for all deficiency states. For AT-III–deficient patients a 50% probability is reached at an age of 21 years; for PC- and PS-deficient patients the 50% probability occurs at an age of 26 years. Both Wilcoxon and log-rank tests show a small but significant difference for all patients (panel a). This difference is due to earlier development of thrombosis in AT-III–deficient females compared with PC- and PS-deficient females, whereas in males the risk is equal in the three deficiency states.
After exclusion of the propositi (panel b), the probability of developing thrombosis was 75% to 90% by an age of 50 to 60 years. The probability of developing thrombosis was 50% at 27 years in AT-III– and PS-deficient patients and at 36 years in PC-deficient patients. There was a significant difference in probabilities among the female AT-III–, PC-, and PS-deficient patients. AT-III–deficient females developed thrombosis earlier in life than did PC- and PS-deficient females. For AT-III–, PC-, and PS-deficient males the probability of developing thrombosis was almost identical among the three groups.
Clinical Features of Thromboembolism
The most common sites of thrombosis were the veins of the pelvic area and deep veins of the leg (54% to 89%); in approximately 50% of patients PE also occurred (Table 2⇓). Recurrent SVT was more common in PC- and PS-deficient than in AT-III–deficient patients. Ten patients (7%) had MVT, and it is interesting to note that 8 of 10 had various thromboembolic manifestations (mostly DVT) prior to MVT. The median age of MVT occurrence was 40 years (range, 22 to 52). Other thrombotic manifestations are listed in Table 2⇓. Differences in thromboembolic manifestations among the three deficiency states were not observed.
Data on the site of the first thrombotic event are listed in Table 3⇓. DVT+PE and PE were frequent (almost 50%) in patients with AT-III deficiency, although the difference was not statistically significant. In 2 patients with PC deficiency, MVT was the first thrombotic event.
Sixty-three percent of symptomatic patients had recurrent events. The recurrence rate did not differ among the different inhibitor deficiencies (Table 2⇑). The recurrence rate for patients in whom the first thrombotic event had occurred spontaneously was lower (54%) than in those who had developed the first thrombotic event during or after a precipitating condition (72%); this difference was significant at a low level (P<.05). In patients with a first spontaneous event, recurrence was likewise spontaneous in 72%. In those patients in whom the first thrombotic event was not spontaneous, the second event was spontaneous in only 45%. Only 7 of 230 (3%; 1 patient with PC deficiency, 2 with PS deficiency, and 4 with AT-III deficiency) patients had a history of coronary heart disease, stroke, or peripheral artery disease.
Thrombotic Risk After Surgery
Data on 203 surgical procedures (including tonsillectomy, appendectomy, herniotomy, other abdominal operations, caesarean sections, surgery of varicose veins, and orthopedic and gynecologic operations) were obtained. Eighty operations were performed at 14 years of age or younger and without thromboprophylaxis. In only 1 case did a postoperative thrombosis occur (nephrectomy at the age of 4 years because of a neoplasm [Wilms’ tumor] followed by caval vein thrombosis).
More than half of the operations (123) were performed in patients older than 14, but 38 of 123 procedures were excluded from study because either appropriate postsurgical thromboprophylaxis had been done or the patients did not know whether they had received any thromboprophylaxis or not. In the remaining 85 procedures 25 thromboembolic events (20 DVTs and/or PEs, 1 caval vein thrombosis, and 4 SVTs) occurred. Thromboembolic events occurred after appendectomy (n=5), orthopedic operations (n=5), caesarean section (n=3), other gynecologic operations (n=4), varicose vein stripping (n=1), abdominal surgery other than appendectomy or caesarean section (n=3), herniotomy (n=3), and tonsillectomy (n=1). Surgical procedures with a high risk of postoperative thrombosis were appendectomy (18 cases/5 thromboses), other abdominal operations (7/3), herniotomy (6/3), orthopedic surgery (7/3), and caesarean section (5/3). Only 1 of 53 tonsillectomies was followed by SVT.
Because it appeared that appendectomy was a major triggering factor for thrombosis, detailed data on the thrombotic risk after this procedure are shown (Table 4⇓). Twenty-five appendectomies in patients <14 years old were not followed by thromboembolism, but in 5 of 18 adults appendectomy triggered the thrombosis (DVT in every case at ages 16, 17, 21, 22, and 31 years).
Thrombotic Risk After Injury
Information on 122 injuries was available, and in 21 of these events the patients either received or did not know if they had received thromboprophylaxis; these events were therefore excluded from further evaluation. The injuries were categorized as those of the leg with cast fixation (55 events), the arm with cast fixation (26 events), and other (20 events; mostly soft-tissue trauma to the leg). After 17 of 101 injuries (17%) thromboembolism occurred. The age of patients when injury was followed by thrombosis was significantly higher (mean, 28.5 years) than that of those in whom injury was not followed by thrombosis (mean, 18.5 years; P=.0017). The youngest age at which posttraumatic thrombosis occurred in AT-III and PS deficiencies was 14 years and in PC deficiency 19 years. None of the 26 patients with cast fixation of the arm developed thrombosis. There was a relatively sharp distinction regarding age and thrombotic risk during/after cast fixation of the leg (Table 4⇑). Whereas none of the 21 injuries in patients younger than 14 years was complicated by venous thrombosis, venous thromboembolism occurred after one third of injuries in patients older than 14 years (and in 5 events before the age of 20 years). In 5 patients thromboembolism occurred after soft-tissue trauma to the leg without cast fixation.
Thrombotic Risk During and After Pregnancy
Twenty-five females with AT-III, 23 with PC, and 23 with PS deficiency had been pregnant at least once. The average number of pregnancies was significantly different among the three deficiency states: 2.2 for AT-III–deficient females; 2.8 for PC-deficient females; and 3.6 for PS-deficient females. In Table 5⇓ data on thrombotic events during and after pregnancy are shown. Twenty-nine pregnancies were excluded because either heparin was administered prophylactically or the patients were not able to give exact information on the clinical history. Development of DVT and/or PE during pregnancy was frequent in AT-III–deficient females (≈40% of pregnancies). In PS-deficient females only 1 pregnancy was complicated by DVT. Postdelivery thrombosis was relatively frequent in PS-deficient females. In AT-III–deficient females thromboembolic events occurred mostly during the first (6 events) and second (9 events) trimesters of pregnancy and in PC-deficient females during the second (3 events) and third (1 event) trimesters. Spontaneously aborted pregnancies were reported by 4 females with AT-III, 4 with PC, and 8 with PS deficiency.
Influence on Quality of Life in Inhibitor-Deficient Patients
The patients were asked to report their symptoms of the postthrombotic syndrome (Table 2⇑). Approximately 50% of symptomatic patients complained of leg swelling and 30% of pain, and ≈20% had had a varicose ulcer at least once. Between 40% and 50% of symptomatic patients felt handicapped by the postthrombotic syndrome, and a similar proportion of patients stated that their quality of life had been negatively influenced by their inhibitor deficiency. Approximately 50% of symptomatic patients also stated that they would have accepted oral anticoagulant treatment before the first thrombotic event if they had been informed of their high risk for thrombosis.
In the present study we obtained detailed data from a large number of patients on the clinical manifestations of hereditary AT-III, PC, or PS deficiency. Great care was taken to include only those patients and relatives with a documented hereditary deficiency (symptomatic and asymptomatic). Only patients in whom the deficiency state was proved by family studies and for whom more than 1 additional family member was eligible for study were included. On average, >3 members per family were recruited.
The probability of developing thrombosis was high (80% to 90% by the fifth to sixth decade of life), a finding that agrees with other published data.1 5 6 Even after data for the propositi were excluded, the risk was still 70% to 90% by age 50 to 60 years. Interestingly, these results in patients with AT-III, PC, or PS deficiency are different from those of patients with resistance to APC, since Svensson and Dahlbäck16 reported only a 30% risk of thrombosis at age 60 for APC-resistant individuals after exclusion of the data for propositi. Hereditary AT-III, PC, and PS deficiencies seem to be stronger risk factors for thrombosis than is APC resistance. However, in general, it should be pointed out that the propositi in our study represented a highly selected group of patients with venous thromboembolism because many of them were screened for a history of venous thromboembolism at a young age. Inhibitor-deficient patients who were identified by applying different selection criteria might have a different risk. Therefore, it is not entirely possible to compare different studies of patients with different risk factors.
Although a significant difference among the three deficiency states with regard to age at the first thrombotic event was not detected in males, females with AT-III deficiency developed thrombosis significantly earlier in life compared with females with PC or PS deficiency. This difference is due to the extremely high thrombotic risk associated with pregnancy and oral contraceptive use in AT-III–deficient females.17 18
Significant differences among the three deficiency states were not demonstrated with regard to the site of thromboembolic events except for the higher prevalence of recurrent superficial thrombophlebitis in patients with PC or PS deficiency. MVT, a potentially life-threatening event, was observed in 4% to 10% of symptomatic patients. In most patients (8/10) MVT occurred as a second or third event except in 2 patients with PC deficiency, in whom MVT was the first thrombotic event. We were unable to confirm the results of another group that had found a higher prevalence of arterial thrombotic manifestations in PC/PS-deficient patients compared with AT-III–deficient patients.19
The event recurrence rate was similar in the three deficiency states. However, it is important to note that the recurrence rate was lower in those patients who had developed their first thrombotic event spontaneously. We surmise that this is not due to a lower thrombotic risk but to more intense diagnostic, therapeutic, and prophylactic regimens in patients with spontaneous thromboses. Spontaneous recurrence of thrombosis (72%) was more frequent in patients in whom their first thromboembolism was also spontaneous. However, spontaneous recurrences can be expected in almost 50% of patients in whom their first thrombosis occurred after a triggering event.
Because the study design was retrospective, objective tests were not available for all patients at the time of thromboembolism. Therefore, a thromboembolic event was defined as established when the diagnosis of each event was confirmed by phlebography, Doppler sonography, or perfusion lung scanning or when the event led to hospitalization and/or treatment with heparin or oral anticoagulants. The fact that not every thromboembolic event could be objectively confirmed is a disadvantage of this type of study; however, we believe that possible inaccuracy due to “missed” events is equal in the three deficiency states. On the other hand, the retrospective design was a major advantage, in that it was possible to obtain data that are difficult (if not impossible) to obtain from a prospective study. Evaluation of triggering conditions, such as surgery, injury, or pregnancy, was also possible, because the deficiency state had been unknown in the majority of patients and prophylactic measures were not performed.
One of the most important aims of the study was to evaluate the importance of triggering events (such as surgery, injury, and pregnancy) for the development of thrombosis in inhibitor-deficient individuals. From our data thrombotic risk appears to be very low in childhood. We conclude that in children <14 years old prophylaxis for thrombosis cannot be regularly recommended. After the 14th year, however, the risk increases considerably. In one third of adolescent and adult patients thromboembolism occurred after abdominal surgery (including appendectomy) or leg injury (with or without cast fixation). This is in agreement with other published data.20 Consequently our data underline the importance of thrombosis prophylaxis in inhibitor-deficient patients >13 years of age during/after abdominal surgery or leg injury with or without cast fixation. In none of the patients was tonsillectomy or cast fixation of the upper limb followed by DVT or PE. Therefore, thromboprophylaxis is not necessary during and after these procedures.
Our data on the high risk of thromboembolism in AT-III–deficient females during pregnancy are in agreement with those of Conard et al18 and De Stefano et al.20 In our patients the risk for severe events (DVT and/or PE) was almost 40% in AT-III–deficient females but very low in PS-deficient females. In PS-deficient females the risk for thromboembolism was considerably higher in the postpartum period. The fact that PS-deficient females had almost twice as many pregnancies as AT-III–deficient females is most probably due to the high complication rate in AT-III–deficient females during pregnancy. An increased risk of spontaneous abortion was not found in our patient population.
In the present study at least 1 member (propositus) of each family was symptomatic. The conclusions of this study are therefore most likely true for this very specific patient population only, ie, individuals from families with symptomatic hereditary deficiencies of AT-III, PC, or PS. It may be speculated that in asymptomatic individuals in whom the diagnosis is made by chance, the risk for thrombosis is lower.
Recently, APC resistance has been recognized as an additional risk factor for thrombosis in patients with PC deficiency, and in fact a prevalence of 19% among symptomatic PC-deficient patients has been found.21 Possibly, knowledge of this additional risk factor may allow better quantification of thrombotic risk in families with both PC deficiency and APC resistance. However, there are still 80% of symptomatic PC-deficient patients for whom additional risk factors cannot be defined.
The data of our study have major implications for patient management. Precipitating conditions like surgery, injury, or pregnancy are relevant for developing thrombosis at a young age. This should motivate physicians and patients to conduct family studies. If the deficiency state is known, then education and counseling of patients and their physicians on thrombotic risk is possible. Diagnostic evaluation of children should be performed before 14 years of age, but it seems unnecessary to investigate very young children (<2 years of age) unless a special clinical situation like malignancy is present. Thrombosis prophylaxis should be instituted after 13 years of age during high-risk situations, early in the course of pregnancy in females with AT-III deficiency, and in all inhibitor-deficient females after delivery. The optimal prophylactic regimen with regard to type and duration has yet to be established in prospective studies. Although life expectancy has not been found to be lower in AT-III–deficient individuals,22 the high event recurrence rate (>60%), the fact that >50% of recurrent events occur spontaneously, the relatively high risk for MVT as a recurrent event, and the disabling effect of the postthrombotic syndrome in ≈50% of symptomatic patients would argue for long-term oral anticoagulant treatment after the first thrombotic event in this highly selected patient group. However, this question has yet to be addressed in appropriately designed trials.
Selected Abbreviations and Acronyms
|APC||=||activated protein C|
|ELISA||=||enzyme-linked immunosorbent assay|
|MVT||=||mesenteric vein thrombosis|
|SVT||=||superficial vein thrombosis|
The following investigators were members of the GTH (Gesellschaft für Thrombose- und Hämostaseforschung) Study Group on Natural Inhibitors: I. Scharrer and V. Hach-Wunderle, Department of Angiology, Department of Medicine, University Hospital, Frankfurt, Germany; K. Lechner, S. Eichinger, P.A. Kyrle, and M. Heistinger, Department of Hematology and Blood Coagulation, First Department of Medicine, University Hospital, Vienna, Austria; H. Vinazzer, Blood Coagulation Laboratory, Linz, Austria; B. Lämmle and F. Demarmels-Biasiutti, Central Hematology Laboratory, Inselspital, University Hospital, Bern, Switzerland; V. Tilsner and G. Marx, Department of Surgery, University Hospital, Hamburg, Germany; E. Seifried and A. Gabelmann, Medical University Clinic III, Ulm, Germany; G. Aspöck, Central Laboratory, Krankenhaus Wels, Austria; and M. Fischer and W.M. Halbmayer, Central Laboratory, Krankenhaus der Stadt Wien-Lainz, Vienna, Austria.
Griffin JH, Evatt B, Zimmermann TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest. 1981;68:1370-1373.
Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S in familial thrombotic disease. Blood. 1984;64:1297-1300.
Comp PC, Nixon RR, Cooper MR, Esmon CT. Familial protein S deficiency is associated with recurrent thrombosis. J Clin Invest. 1984;74:2082-2088.
Bovill EG, Bauer KA, Dickermann JD, Callas P, West B. The clinical spectrum of heterozygous protein C deficiency in a large New England kindred. Blood. 1989;73:712-717.
Demers C, Ginsberg JS, Hirsh J, Henderson P, Blajchman MA. Thrombosis in antithrombin III-deficient persons: report of a large kindred and literature review. Ann Intern Med. 1992;116:754-761.
Engesser L, Broekmans AW, Briet E, Brommer EJP, Bertina RM. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med. 1987;107:677-682.
Pabinger I, Brücker S, Kyrle PA, Schneider B, Korninger HC, Niessner H, Lechner K. Hereditary deficiency of antithrombin III, protein C and protein S: prevalence in patients with a history of venous thrombosis and criteria for a rational patient screening. Blood Coagulation Fibrinol. 1992;3:547-553.
Comp PC, Doray D, Patton D, Esmon CT. An abnormal plasma distribution of protein S occurs in functional protein S deficiency. Blood. 1986;67:504-508.
Thommen D, Buhrfeind E, Felix R, Sulzer I, Furlan M, Lämmle B. Hämostaseparameter bei 55 Patienten mit venösen und/oder arteriellen Thrombembolien. Schweiz Med Wschr. 1989;119:493-499.
SAS Language. Version 6, 1st edition. Cary NC: SAS Institute Inc; 1990.
De Stefano V, Leone G, Mastrangelo S, Tripodi A, Rodeghiero F, Castaman G, Barbui T, Finazzi G, Bizzi B, Mannucci PM. Clinical manifestations and management of inherited thrombophilia: retrospective analysis and follow up after diagnosis of 138 patients with congenital deficiency of antithrombin III, protein C, protein S. Thromb Haemost. 1994;72:352-358.
Koeleman BPC, Reitsma PH, Allart CF, Bertina RM. Activated protein C resistance as an additional risk factor for thrombosis in protein C deficient families. Blood. 1994;84:1031-1035.