deficiency in a large

Transcription

1 Hereditary protein S New jersey kindred deficiency in a large Donald K. Wilkerson, MD, Lillian Burrell, BS, Laura A. Cisar, PhD, Alan M. Graham, MD, and Hugh Kim, MD, New Brunswick, N.J. Purpose: Protein S is a vitamin K-dependent anticoagulant protein that serves as a cofactor for activated protein C. Deficiency of protein S has been associated with recurrent thrombotic events. To characterize better the risks of thrombosis in protein S deficiency, we studied 62 members in a large kindred. Method: All members were evaluated by a thorough clinical history. Plasma samples were assayed for total protein S antigen and protein S activity. Upper and lower extremity venous duplex examinations were performed in the majority of adult members. Result: Twenty-six (40%) of the 62 family members were classified as deficient on the basis of either low total protein S antigen levels or low protein S functional activity. Five members deficient in protein S had 16 venous thrombotic events. In all members the onset of thrombotic events occurred after 19 years of age, with a tendency for recurrence. Three lower extremity deep venous thromboses that had been occult previously were first diagnosed on surveillance duplex scanning. Only one member whose protein S level was not deficient had a single episode of superficial thrombophlebitis. Conclusion: Our findings in this large kindred confirm an autosomal-dominant inheritance pattern. Thrombotic events occurred after the age of 19 years in affected individuals and tended to be recurrent. The diagnosis of protein S deficiency is based on functional and immunologic plasma assays. In this study venous duplex scanning proved to be a useful diagnostic adjuvant. (J VASC SIrinG 1993;18:932-8.) Protein S is a vitamin K--dependent plasma protein that functions to inhibit blood clotting by serving as a cofactor for activated protein C (apc). Protein S circulates in two forms: as a free protein and in complex with C4b-binding protein (C4b-BP). Protein S bound to C4b-BP does not serve as a cofactor for apc. The existence of these two plasma pools of protein S, one functional and one nonfunctional, has traditionally complicated the diagnosis of protein S deficiency by readily available immunologic assays. 1 Recent reports have described hereditary deficiencies of protein S in patients with recurrent From the Divisions of Vascular Surgery and Hematology, UMDNJ-Robert Wood Johnson Medical School, New Brunswick. Presented at the Seventh Annual Meeting of the Eastern Vascular Society, Philadelphia, Pa., April 29-May 2, Reprint requests: Donald K. Wilkerson, MD, Department of Surgery, UMDNJ-Robert Wood Johnson Medical School, One Robert Wood Johnson PI., New Brunswick, NJ Copyright 1993 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter /93/$ /6/49984 thrombotic disease. Thrombosis in these patients is presumably the result of a deficient clot-regulatory mechanism. 2,3 To characterize better the risks of thrombosis in protein S deficiency and to compare the diagnostic capabilities of immunologic versus fimctional assays, we studied 62 members of a New Jersey kindred of four generations in whom the disorder was suspected. We report the clinical and diagnostic findings in this New Jersey pedigree. We also discuss the role of extremity venous duplex scanning as an adjunctive diagnostic tool. METHOD History. After the identification of the propositus in 1992, members of a New Jersey family were systematically evaluated for protein S deficiency. A careful personal history was obtained from each member. Oral history taking was supplemented by a standardized questionnaire. Whenever possible, medical records were obtained from physicians and hospitals. The diagnosis of deep vein thrombosis was confirmed at least once by ascending venography or extremity venous duplex scanning. The diagnosis of pulmonary embolism was confirmed at least once by 932

2 Volume 18, Number 6 Wilkerson et al. 933 'I '1 '1,! Fig. 1. New Jersey pedigree. Propositus is represented by arrow, pe~usion lung scanning. All diagnoses of superficial thrombophlebitis were established by either a physician on the basis of physical examination or extremity venous duplex scanning. Plasma assays. Total protein S antigen levels were determined quantitatively by an enzyme-linked in~nunosorbent assay (Asserachrom Protein S; American Bioproducts Co., Diagnostica Stago, Asnieres-Sur-Seine, France), as described previously. 4,s Total protein S antigen levels were considered deficient if they fell below the lower limit of the normal range found in 24 healthy control subjects (63% to 96%). Protein S activity was measured by a functional clot-based assay (Staclot Protein S., Diagnostica Stago) as described previously. Briefly, the assay is based on the protein S cofactor activity for apc inhibition of activated factor V. 6 Protein S functional activity levels were considered deficient if they fell below the lower limit of the normal range found in 20 healthy control subjects (75% to 157%). Extremity venous duplex scanning. Upper and lower extremity venous duplex examinations were performed in 20 of 29 adult family members. All ultrasound examinations were performed in the vascular laboratory with a color Doppler imaging system (Acuson 128P Scanner; Acuson Inc., Mountain View, Calif.). A 7.5 MHz probe was used to study superficial veins and a 5.0 MHz probe was used to study more deeply situated venous segments. The assessments included the systematic longitudinal and transverse tracking of relevant venous segments. Real-time images of all veins and simultaneous Doppler-flow information were recorded on videotape and hard-copy color prints. Scans were interpreted at the time of examination by the vascular surgery attending staff. The criteria for a positive scan result included the visualization of intraluminal thrombus, the absence of spontaneous flow, the absence of phasic flow on respiration, and incompressibility of the vein on probe pressure Data analysis. Statistical analysis was performed with SAS software (SAS Institute, Cary, N,C.) implemented on a personal computer. Fisher's exact test was used to correlate the occurrence of thromboembolic events with assay-determined protein S deficiency. Pearson's correlation coefficient was used to determine the relationship between protein S functional activity and measured total protein S levels. RESULTS The New Jersey pedigree is shown in Fig. 1 and the results of the protein S immunologic and functional assays in family members in Table I. Of the 62 members studied, 43 were female and 19 were male. Twenty-nine adults (aged > 18 years) were studied along with 33 children. The mean age was 21 years. Seventeen (27%) of 62 family members have protein S deficiency on the basis of low total protein S antigen levels. Twenty-three (37%) of 62 family members have protein S deficiency on the basis of low protein S activity on functional assay. Fourteen family members were deficient in both total protein S antigen and protein S functional activity. An individual was considered protein S deficient if he or she was deficient in either protein S antigen or functional activity. Five of the members deficient in protein S had 16 venous thrombotic events (Fig. 2). Arterial thromboses have not occurred. Venous thrombotic events included five episodes of lower extremity deep vein thrombosis, eight episodes of superficial thrombophlebitis, and two episodes of pulmonary

3 934 Wilkerson et al. December 1993 Table I. Results of the protein S immunologic and functional assays in family members Protein S assays (%) Generation Age (yr) Thromboembolism Total antigen s Functional I-1 68 I-2 83 II-1 63 II-2 60 II-3 58 II-4 54 II-5 48 III-1 41 III-2 39 III-3 37 III-4 35 III-5 34 III-6 31 III-7 29 III-8 27 III III III III III III III III III III III III III III III III III IV-1 6 IV-2 6 IV-3 12 IV-4 10 IV-5 7 IV-6 12 IV-7 10 IV-8 10 IV-9 7 1V-10 2 IV IV IV-15 6 IV IV-18 7 IV-19 4 IV IV-21 8 IV-22 8 IV-23 5 IV-24 3 IV-25 7 IV-26 4 IV-27 9 IV-28 4 IV-29 6 IV-30 4 IV-31 1/2 IV-32 5 IV-33 1 IV-36 2 Not done O ~ 81 5O 47 49~ O O O 73 7O O O Not done ~ ~ ~Normal value 63% to 96% (mean ± SD 79.6% ± 8.1%). tnormal value 75% to 157% (mean ± SD 116% ± 20.3%). ~Taking an oral anticoagulant. Value below the lower limit of the normal range.

4 Volume 18, Number 6 Wilkerson et al. 935 S*Fp 1-1 I MVT I, II- 2 STP I STP STP STPSTP DVT I I I J L.! ODVT III - 8 ODVT i DVT STP PE PE DVT III - 9 L L L_ ODVT! Ill- 11 DVT I SIP STP Ill- 13 [ I PEDIGREE NUMBER I I I I I I I I I I I I 70 I AGE (YEARS) Fig. 2. Temporal relationship between occurrence ofthrombofic events (thinverticalline), age, and oral anticoagulation treatment (black bar) in protein S-deficient persons from pedigree in Fig. 1. Right-hand limit of lines indicates member's age at time of investigation. DVT, Lower extremity deep vein thrombosis; ODVT, occult lower extremity deep vein thrombosis by venous duplex scan; MVT, mesenteric vein thrombosis; STP, superficial thrombophlebitis; PE, pulmonary embolism. embolism. Of the lower extremity deep vein thromboses, three had been occult previously and were diagnosed as a result of chronic changes seen on screening extremity duplex scans. In the five members with symptoms, the first onset of thrombotic events occurred after age 19 years, with a tendency for recurrence in three of five members (Fig. 2). The family's patriarch, who was not tested, had a fatal episode of mesenteric venous thrombosis at the age of 68 years. The increased incidence of thromboembolic events in the members deficient in protein S was significant (p < 0.05). Although there was a significant correlation between the total protein S levels and functional protein S activity (r = 0.76;p < ), a fair amount of variability was observed between the two assay measurements (Table I). Faro@ members deficient in protein S and younger than age 19 years have not had thrombotic complications. Only one member whose protein S levels were not deficient (III-10) had a thrombotic event, a single episode of superficial thrombophlebitis associated with an intravenous infusion. Duplex examinations were negative for silent deep vein thromboses in the nine adult family members studied whose protein S levels were not deficient. DISCUSSION Protein S is a vitamin K-dependent glycoprotein that serves as a cofactor for the expression of the anticoagulant and profibrinolytic activities of apc Protein S is composed of a single polypeptide chain with a molecular weight of about 84,000 daltons. Protein S is synthesized mostly by hepatocytes, but endothelial cells, megakaryocytes, and osteoblasts are other sites of synthesis, is Protein S is synthesized as an inactive precursor. The functional form is obtained after the carboxylation of glutamic residues by a vitamin K-dependent carboxylase during platelet activation. Preliminary data have indicated that the biologic half-life of protein S is about 60 hours. 16 Protein S potentiates the anticoagulant and profibrinolyric properties of apc. Activated protein C inhibits activated coagulant factors V and VIII. Protein S circulates in the plasma in at least two forms: as free protein S (40%) and in reversible complex with C4b-BP (60%), an inhibitor of the classic complement pathway. On binding to C4b-BP, protein S loses its ability to function as a cofactor for apc. Persons who have genetic or acquired deficiencies in protein S are prone to severe, recurrent episodes of venous thrombosis. There have been anecdotal reports of arterial thrombotic events. 17:s

5 936 Wilkerson et al. December 1993 The immunologic, as well as functional, diagnosis of protein S deficiency is complicated by the existence of the two pools of protein S in normal plasma. The use of commonly available immunologic techniques alone has proved deceptive in several studies. 19 Although immunoassays provide good estimates of the antigenic level of circulatory protein S, they do not distinguish between free and functionally inactive protein S. We suspect that the correlation between our total antigen levels and functional activity measurements represents an underestimation of total protein concentration by our enzyme-hnked immunosorbent assay method. Recent reports have suggested such an underestimation when this assay is employed with incubation times and plasma dilutions currently recommended by the manufacturer. 2 It is postulated that higher plasma sample dilutions and a further prolongation of incubation time of plasma with assay antibody allow a more complete dissociation of the protein S-C4b-BP complex and a more accurate estimation of protein S concentrations. Similar difficulties in defining ideal immunologic assay conditions have led some authorities to suggest that protein S deficiency be diagnosed only when there is evidence of diminished levels of protein S functional activity. These authorities would use the additional information provided by immunoassays to subclass@ individuals deficient in protein S further. 21 The incidence of venous thrombotic events was 38% in adults deficient in protein S. Previous clinical studies have reported higher incidences (55% to 60%) of venous thrombotic events in families deficient in protein S. The age of first occurrence is typically after 15 years. Interestingly, four of the five individuals suffering from venous thrombosis were first-order relatives. These individuals have also been assessed for several other hypercoagulable parameters including prothrombin time, partial thromboplastin time, protein C, antithrombin III, F 1 + 2, and D-dimer levels (F1+2 and D-dimer are in vivo coagulation activation markers). This workup has uncovered no additional abnormalities. The increased susceptibility of this cluster of family members is unclear. The diagnosis of three previously occult deep vein thromboses in this symptom-flee, high-risk population was made on surveillance duplex examination. 22 Phlebographic confirmation of this discovery led us to treat one of these patients first with heparin therapy followed by indefinite oral anticoagulation. 2s Although recent reports have suggested less than optimal sensitivities when duplex scanning is used for surveillance of deep vein thrombosis in patients at high risk, based on our finding we currently recommend serial extremity venous duplex scans in any adult patient diagnosed as having protein S deficiency. Our finding in this large kindred confirms an autosomal-dominant inheritance pattern. In affected members, thrombotic events occurred after age 19 years and tended to be recurrent. The diagnosis of protein S deficiency is established by documenting either low protein S antigen levels or low levels of protein S functional activity. Surveillance duplex imaging proved a useful adjuvant in this population of susceptible individuals. We gratefially acknowledge the assistance of Mr. Gary Ribar for his illustrations. REFERENCES 1. Comp PC, Clouse L. Plasma proteins C and S: the function and assay of two natural anticoagulants. Lab Management 1985; 23: Engesser L, Broekmans A, Briet E, Brommer E, Bertina R. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1982;106: Comp P, Esmon C. Recurrent venous thromboembolism in patients with a partial deficiency of protein S. N Engl J Med 1984;311: Boyer-Neumann C, Wolf M, Amiral J, Guyuder AM, Meyer D, Larrieu MJ. Familial type I protein S deficiency associated with severe venous thrombosis: a study of five cases [Letter]. Thromb Haemost 1988;60: Wolf M, Boyer-Neumann C, Leroy-Matheron C, et al. Functional assay of protein S in seventy patients with congenital and acquired disorders. Blood Coagul Fibrinolysis 1991;2: Wolf M, Boyer-Neumann C, Martinoli JL, et al. A new fianctional assay for human protein S activity using activated factor V as substrate. Thromb Haemost 1989;62: Cranley JJ, Higgins RF, Berry RE, Ford CR, Comerota AJ, Griffin LH. Near parity in the final diagnosis of deep venous thrombosis by duplex scan and phlebography. Phlebology 1989;4: Kerr T, Lutter K, Cranley R, Cranley J. Diagnosis of upper extremity venous thrombosis by duplex scanning. In: Bergan JJ, Yao JST, eds. Venous disorders. Philadelphia: WB Saunders, 1991:chap Lensing A, Prandoni P, Brandjes D, et al. Detection of deep vein thrombosis by real-time B-mode ultrasonography. N Engl J Med 1989;320: Killewich L, Bedford G, Beach K, Strandness D. Diagnosis of deep venous thrombosis: a prospective study comparing duplex scanning to contrast venography. Circulation 1989; 79: Falk R, Smith D. Thrombosis of upper extremity thoracic inlet veins: diagnosis with duplex Doppler sonography. AJR Am J l~oentgenol 1987;149: Van Hinsbergh VWM, Bertina RM, Van Wijngaarden A, Van Tilburg NH, Emesis JJ, Haverkate R. Activated protein C decreases plasminogen activator inhibitor activity in endothelial cell-conditioned medium. Blood 1985;65:

6 Volume 18, Number 6 Wilkerson et al Walker FJ. Regulation of activated protein C by protein S: the role of phospholipid in factor Va inactivation, j" Biol Chem 1981;156: DeFonn NJ, Haverkate F, Bertina R, Koopman T, Van Wijngaarden A. The cofactor role of protein S in the acceleration of whole blood clot lysis. Blood 1986;67: MaiUard A, Berruyer M, Serre CM, Dechavanne S, Delmas PD. Protein-S, a vitamin K-dependent protein, is a bone matrix component synthesized and secreted by osteoblasts. Endocrinology 1992;130: Broekmans AW, Bertina RM, Reinalda-Poot J, et al. Hereditaly protein S deficiency and venous thromboembolism: a study in three dutch families. Thromb Haemost 1985;53: Eason JD, Mills JL, Beckett WC. Hypercoagulable states in arterial thromboembolism. Surg Gynecol Obstet 1992;174: Coller B, Owen J, Jesty J, et al. Deficiency of plasma protein S, protein C, or antithrombin III and arterial thrombosis. Arteriosclerosis 1987;7: Comp P, Doray D, Patton D, Esmon C. An abnormal plasma distribution of protein S occurs in fimctional protein S deficiency. Blood 1986;67: Tripodi A, Bertina RM, Conard J, Pabinger I, Sala N, Marmucci PM. Multicenter evaluation of three commercial methods for measuring protein S antigen. Thromb Haemost 1992;68: Dahlback B, Bertina R, Griffin J, Salem HIt. Presented to the Subcommittee Meeting on Protein C and S, at the International Society on Thrombosis and Haemostasis Thirty-eighth Annual Meeting of the Scientific and Standardization Committee, July 6-9, 1992; Munich, Germany.) 22. Mattos MA, Londrey GL, Leutz DW, et al. Color-flow duplex scanning for the surveillance and diagnosis of acute deep venous thrombosis. J VAsc SURG 1992;15: Michaels J, Stibbe J, Bertina R, Broekmans A. Effectiveness of long-term oral anticoagulation treatment in preventing venous thrombosis in hereditary protein S deficiency. Br Med J 1987;195: Submitted May 12, 1993; accepted July 9, DISCUSSION Dr. Kenneth Ouriel (Rochester, N.Y.). Protein S deficiency is one of a variety of inherited hypercoagnlability syndromes, each of which is defined by mutations in the genes coating for specific components of the coagulation pathways. To date five major primary hypercoagulable states have been described. Inherited disorders are responsible for at least four of these: antithrombin HI deficiency, protein S and C deficiencies, disorders ofplasminogen, and hyperfibrinogenemia. To date lupus anticoagulant has not been described as an inherited disorder, but this may change. We as surgeons would like to blame many of our graft thromboses on the presence of a hypercoagulable state. However, the described hypercoagulable syndromes are most frequently associated with venom disease. In a series of patients with acute deep venous thromboses, it was found that 8% of these patients had an inherited hypercoagulable state. That is illustrated here; we can see that the protein S and C deficiencies are the most common hypercoagulable syndromes found in patients with deep vein thrombosis, but also antithrombin III and plasminogen problems were noted. Protein S, like protein C, is formed in the hepatocytes through y-carboxylation. Its synthesis is therefore vitamin K dependent and inhibited by warfarin. Protein S acts as a cofactor to activate protein C, promoting its anticoagulant inhibitory effects on factor V and factor VIII. Familial deficiency of protein S was first described by two independent investigators about 10 years ago. The dinical manifestations of protein S deficiency include venous thrombosis in up to 55% of heterozygotes. The thrombotic events characteristically occur before age 30 years. Coumadin-induced skin necrosis characterized by small-vessel thrombosis may develop when warfarin is begun without concurrent heparinization in the alreadylowered protein S concentrations, which fail faster than the prothrombotic factors 2, 7, 9, and lo. Your study is somewhat unusual in its low frequency of thrombotic complications in heterozygotes; nevertheless different families may have different mutations, and different functional levels of the proteins may occur in different families. In addition, in vitro functional assays may not represent the true in vivo activity of the mutated protein. Is anticoagulant therapy indicated for asymptomatic heterozygotes? Second, what is the best means of treating these patients given the risks of Coumadin-induced skin necrosis? Finally, why are these patients at risk for the development of venous thrombotic events but arterial thromboses are rare? Dr. Donald K. Wilkerson. In terms of the question, Is anticoagulation necessary in symptom-free individuals? It would seem on review of the literature that it is not, unless that individual was at significant risk for thrombosis. Again, risks would include a postsurgical period, a period during which the patient would have forced immobility, pregnancy, because protein S levels go down dramatically with pregnancy, and smoking. I think in those instances the anticoagulation for symptom-free people is probably indicated. In terms of the second question, the best means of anticoagulation for protein S-deficient patients, for years it was thought that this patient population was somewhat

7 938 Wilkerson et al. December 1993 different from patients suffering from protein C deficiency. It is well known that a patient suffering from a protein C deficiency, if not first given heparin, experiences a further reduction in the protein C levels once starting warfarin, and might have paradoxic thromboses of small arterioles and suffer Coumadin-induced skin necrosis. Many clinicians for years were quite comfortable giving protein S-deficient patients warfarin in an outpatient setting. Unfortunately, recently in the literature several case studies have been published that show that this patient population can also suffer from Coumadin-induced skin necrosis. Because of that, at our institution, when one of these patients needs an anticoagulant, we admit him, give him heparin, and, after 3 clays or so, begin conversion to warfarin. We maintained the prothrombin time in a range of 2 to 3 in terms of the international ratio. Regarding the last question, why do these patients suffer from venothrombotic events and why do we never see them coming in because their arterial anatomy has occluded? I initially thought that the answer to such a question might be related to flow, with relatively lower flows found in venous systems. I have since been informed that flow probably does not have such an important role. In fact I am sure Dr. Ouriel has recently demonstrated that, in an ex vivo system, when similar flows are used injured veins tend to thrombose in response to changes in coagulation, whereas injured arteries tend to respond negatively when there is evidence of platelet abnormality. I wonder if there is a differential response in endothelinm, in the two systems, but I have no proof to suggest that there might be. Dr. Maids J. Tsapogas (Northport, N.Y.). Protein S and C deficiencies are gradually being better recognized as additional causes of thromboembolic events. They may be associated with disturbances of other major antithrombins. Acquired forms of these reductions were also observed in seriously ill patients, as for example those with carcinoma of the lung or interstitial nephropathy. A reported interesting finding is diminished S and C proteins in pregnancy, together with relatively low levels of other physiologic anticoagulants such as antithrombin III and heparin cofactor II. What is your experience on the manifestations of the acquired protein S and C deficiency with various pathologic conditions? Do you believe that reduction of these proteins during pregnancy, in addition to the other known factors, contributes to thromboembolic complications? Dr. Wilkerson. It is true that protein S levels, as well as other anticoagulation factors, are diminished in the pregnancy state. That is in comparison with normal control subjects, so I think that the decreased level of this factor in the pregnancy state is a normal response. I would not treat it. I am not saying that if a patient was protein S deficient and told me she was pregnant I would not treat her. I think in that instance I would probably offer her prophylactic heparin therapy throughout the pregnancy. There is literature to suggest that that patient population does exist and they do have a high risk of thrombotic events during pregnancy if they are not so treated. Dr. John J. Ricotta (Buffalo, N.Y.). You did say that there was a correlation between your protein S-deficient levels and symptomatic thromboses. Can you give us some idea of what level of deficiency was required to have symptomatic venous thrombosis? Second, why do you use an INR of 2 to 3 when the current recommendations for anticoagulation in other disease states talk about an INR of 1.3 to 1.5? Is this just a decision you have made or are there data to substantiate your use of a higher level of anticoagulation? Dr. Wilkerson. The higher international ratio has been suggested in the literature; it is not just something we have arbitrarily decided on. In terms of your first question, I agree that our total protein S antigen levels did show a correlation with thromboembolic phenomenon. Total protein S antigen levels were considered deficient if they fell below the lower limit of the normal range found in 24 healthy control subjects.