Cutaneous necrosis revealing the coexistence of an antiphospholipid syndrome with acquired protein S deficiency, factor V Leiden and hyperhomocysteinemia

ARTICLE

Transitory or chronic hypercoagulable states are recognised but rare causes of cutaneous necrosis. The most frequent etiologies concern isolated abnormalities such as the Anti-Phospholipid Syndrome (APS), or congenital or acquired protein C or protein S deficiencies [1]. Moreover, recently thrombophilic factors have been identified such as the Activated Protein C Resistance (APC-R) [2] or moderate hyperhomocysteinemia [3]. So the observation, in a consistent number of cases, of the association, in the same patient, of various causative factors suggests that venous thrombosis can have multifactorial origins [4]. The cutaneous necrosis observation of this report is original because of the presence of 4 abnormalities associated with an increased risk of thrombosis. This indicates the need, in some clinical contexts, for investigating more systematically these associations.

Case report

A 72-year-old female had been suffering, for 5 months, from bilateral necrotic ulcers on the lower limbs. In addition to an arterial hypertension correctly controlled by a ramipril (Triatec^®) therapy, she was classified as having thromboembolic disease, as she developed 3 phlebitis in 1977, 1994 and 1996 complicated by pulmonary embolism, but she had had no abortion. She had not received antiplatelet agents or anticoagulant therapy. There was no familial history of thromboembolic disease. Ulcers developed suddenly, on the 3rd lowest part of the legs, and started with multiple necrotic, hyperalgic and insomniant plaques. Their extension produced large ulcerations. There was no systemic syndrome associated.

Upon admission, the patient was tired, but without any general alteration of her state, nor hyperthermia. Ulcers had yellowish borders and were covered with necrotic plaques (Fig. 1). More recently, she had developed cyanosis of the second toe, which was hyperalgic (Fig. 2). There was no livedo reticularis, no "nodule" on the lower limbs, nor any lupic eruption. Clinical examination was quite normal; peripheral pulses were weak, but they were perceived without any murmur in the large arteries. The Doppler examination showed superficial and deep venous insufficiencies, consecutive to phlebitis and a disseminated atheromatous charge without any hemodynamic consequence. The systolic pressure index was 0.7. A biopsy, performed at the ulcer borderline or at the cyanotic toe allowed us to eliminate the presence of cholesterol emboli, but identified multiple venous and arterial thromboses, without any sign of vaculitis or atherosclerosis (>Fig. 3).

Haemostasis tests are shown in Table I. A lupus anticoagulant was diagnosed ; the tests (APPT, TTD) were not corrected by the addition of platelets (1.39 versus < 1.2 for normal range). The presence of an anti-Protein S antibody, due to decreased protein level, was suspected and researched; it was negative when the ELISA technique was used. Antithrombin III and Protein C activities were normal. Despite hyperhomocysteinemia, there was no deficiency in folates or vitamin B12. No polymorphism G20210A on the prothrombin gene was found. The anti-PF4-heparin antibodies were measured despite a normal platelet count but remained negative.

Antinuclear antibodies were positive for the 1:320 dilution, with an anti-RNP at a level of 1.1 (normal range < 1), and an anti Sm positive at 2.2 (normal range < 1). There was no other positive test for autoimmunity (Waler Rose, anti ECT or anti native DNA).

There was no renal insufficiency, no cryoprotein, nor any of the complications currently observed in APS or lupus (hematologic disorder, neurological, cardiac or renal complications).

The initial therapy included a heparin treatment, use of a vaso-dilatator drug and application of fat wound dressing. As the ulcers extended and as the pain was persistent, the patient was treated with iloprost (Ilomedine^®) at the posology of 50 mug/day, with a maximal flow of 40 ml/hr. At the end of the first week, there was a favourable evolution with a decrease of pain, and a beginning of ulcer healing, which was facilitated by some grafts. After 1 month of therapy, there was a normalisation of protein S level (85%). Heparin was replaced by a long-term dicoumarol therapy (INR of about 3). After 2 years of follow-up, there was no recurrence of ulcers, but the moderate hyperhomocysteinemia and APS remained, with the presence of a lupus anticoagulant activity and a low level of antiphospholipid antibodies.

Discussion

This patient presented extensive multiple necrotic bilateral ulcers which evoked a necrotic angiodermitis, inasmuch as she developed an arterioatheromatous hypertension.

However, there was the very atypical presentation of the toe ischemia. Histology did not show any sign of necrotic angiodermitis [1], nor any cholesterol emboli nor a superficial necrotic vasculitis. By contrast, the presence of multiple venous and arterial thrombosis suggest that a thrombotic process was the cause of this necrosis. The etiological investigation showed the associated presence of acquired and congenital disorders which could explain, through their unusual association, the severity of the clinical manifestations. The trigger of the acute episode could be explained by the occurrence of the lupus anticoagulant. This suggests the development of an APS, which is confirmed by the persistence over a one-year period of anticardiolipin antibodies associated with a lupus anticoagulant activity, and anti beta2GPI dependent autoantibodies. No drug induced causative factor was found; furthermore, there was no myasthenia, nor any other factor for suspecting a systemic vasculitis, haemopathis or cancer. Antinuclear antibodies were positive for the 1:320 dilution, and, above all, an anti-Sm was present; nevertheless there were not enough criteria for establishing a diagnosis of systemic lupus erythematosus (SLE).

The APS is well known for its arterial and/or venous thrombosis and presence of systemic manifestations. The cutaneous complications concern 40% of patients and can be the very first signs of the disease. They are multiple (Table II) [5]. Cutaneous ulcerations are observed in 30% of cases with painful ulcers at "serpiginous" borders, which are predominant on pretibial and ankle zones. They can be associated with distal cutaneous necrosis, frequently severe. On histological studies, the presence of thrombosis in small vessels is very characteristic [6, 7]. Nevertheless, they are not specific, as they can be observed in other pathologies (which were ruled out in our patient, such as a cryoglobulin, vascular emboli, warfarin or heparin induced necrosis, protein C deficiency). The mechanism which induces cutaneous necrosis is discussable; it is very likely that antiphospholipid antibodies are implicated in the triggering cause. These antibodies induce an increased perturbation of Protein C pathway [8], already altered by the presence of APC-R, resulting from the factor V Leiden.

The presence of APC-R, without factor V Leiden, has been observed in 38% of patients with APS [9]. In addition, acquired Protein S deficiencies associated or not with an anti-Protein S antibody, have been described in this context [10, 11]. Activated Protein C controls and regulates blood activating by cleaving and inactivating factors Va and VIIIa. Activated protein C acts at the phospholipid surface in the presence of its cofactor, protein S. It is then understandable that the presence of antiphospholipid antibodies with acquired protein S deficiency can favour the development of thrombosis (Fig. 4).

Constitutional, but rare, coagulation abnormalities can also be present, which enhance the thrombotic tendency in the APS. This is the case in our patient, who has APC-R with the factor V Leiden mutation, and hyperhomocysteinemia with the C to T homozygous mutation at position 677 on the MTHFR gene. These defects are probably the origin of the previous thrombotic episodes. APCR-R was first described by Dahlbäck [2] in 1993. It is frequently consecutive to a mutation of the factor V gene. The factor V has, at position 506, a glutamine residue instead of the arginin, and is usually called factor V Leiden. When this mutation is present, mutated factor Va is inhibited at a 10-fold lower rate than normal factor Va. The heterozygous form of this mutation concerns 3 to 8% of the population [12]. The transmission is autosomal and dominant. The relative thrombotic risk is estimated to be multiplied by 5 to 10 in heterozygous and by 50 to 100 in homozygous. APC-R is reported frequently associated with venous thrombosis [12]. Its presence has been reported in venous ulcers consecutive to phlebitis [9]. A more exceptional case of superficial cutaneous necrosis, favoured by the association with a Protein S deficiency, has been reported in presence of APC-R [13]. The association (APS and factor V Leiden) is rare [14, 15] and is probably circumstantial [16, 17], but it obviously presents an increased risk factor induced by the occurrence of antiphospholipid antibodies [18].

Hyperhomocysteinemia has also been reported as a venous and arterial thrombogenic risk factor, with a risk increase estimated at 2.5 when the concentration is above 2 standard deviations of the normal value [3].

At least in vitro, hyperhomocysteinemia is involved at various stages [19]. In particular, it increases factor V activation and on the endothelial cell surface, it influences the protein C pathway, by decreasing thrombomodulin expression, the factor which is necessary for protein C activation. This abnormality should be linked with the presence of the homozygous C677T thermolabile variant of MTHFR. Its efficacy could then be reduced by 50 to 60%, which induces a moderate hyperhomocysteinemia as our patient. This mutation is frequent and concerns 4 to 15% of the population. If the risk associated with hyperhomocysteinemia and/or with the MTHFR homozygous variant is still discussed [20, 21], by contrast, the association with factor V Leiden is thought to increase the venous thrombotic risk by a factor of 20 [22]. This defect could have been involved as a fourth risk predisposing to thrombosis in our patient (Fig. 4).

CONCLUSION

In conclusion, in our patient the presence of various congenital or acquired disorders has probably contributed to the more or less complete blockage of the protein C pathway (Fig. 4) and explains the severe venous and arterial pathology observed. These multiple associations, which seem to be rare at present, should be investigated more systematically in presence of cutaneous necrosis, when there are individual or familial thromboembolic disease antecedents. It is obvious that the association of the different risk factors, acquired or congenital, increases the risk of thrombosis. The multifactorial origin of thrombosis is more and more frequently evoked. Rosendaal [4] recently reported that on 7 selected congenital or acquired thrombogenic risk factors, the coexistence of 2 factors increases the risk of thrombosis by a factor of 7.7, and this risk is multiplied by 20 when at least 3 factors are associated.

The discovery of these multiple associations should lead us to sometimes consider a long-term anticoagulant therapy with caution. In certain deficiencies i.e. protein C deficiency, an initial treatment with anti-vitamin K can be accompagnied by thrombosis so it is necessary to start, in this case, an anticoagulant treatment with heparin as the initial treatment.

Article accepted on 5/2/02

REFERENCES

1. Saada V, Chosidow O, Frances C. Nécroses cutanées. Encycl Med Chir (Elsevier, Paris). Dermatologie 12.685 A 10. 1996; 6 p.

2. Dahlbäck B. Resistance to actived protein C caused by the factor V R506Q mutation is a common risk factor for venous thrombosis. Thromb Haemost 1997; 78: 483-8.

3. Den Heijer M, Rosendaal FR, Blom HJ, Gerrits WBJ, Bos GMJ. Hyperhomocysteinemia and venous thrombosis: a meta analysis. Thromb Haemost 1998; 80: 874-7.

4. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999; 353: 1167-73.

5. Nahass G.T. Antiphospholipid antibodies and the antiphospholipid syndrome. J Am Acad Dermatol 1997; 36: 149-68.

6. Alegre VA, Gastineau DA, Winkelmann. Skin lesions associated with circulating lupus anticoagulant. Br J Dermatol 1989; 120: 419-29.

7. Gibson GE, Su WPD, Pittelwow MR. Antiphospholipid syndrome and the skin. J Am Acad Dermatol 1997; 36: 970-82.

8. Atsumi T, Khamashta MA, Amengual O, et al. Binding anticardiolipin antibodies to proteine C via beta 2 glycoprotein 1: a possible mechanism in inhibitory effect of anticardiolipin antibodies on the protein C system. Clin Exp Immunol 1998; 112: 325-33.

9. Grossman D, Heald PW, Wang C, Rinder HM. Activated protein C resistance and anticardiolipin antibodies in patients with venous leg ulcers. J Am Acad Dermatol 1997; 37: 409-13.

10. Amster M.S, Conway J, Zeid M, Pincus S. Cutaneous necrosis resulting from protein S deficiency and increased antiphospholipid antibody in a patient with systemic lupus erythematosus. J Am Acad Dermatol 1993; 29: 853-7.

11. Morange PE, Alessi MC, Barthet MF, et al. Acquired protein S deficiency, likely due to anti-PS autoantibodies following a thrombotic event in a patient with a systemic lupus erythematosus. Thromb Haemost 1997; 78: 1416-7.

12. Hillarp A, Dalhbäck B, Zöller B. Activated protein C resistance: from phenotype to genotype and clinical practice. Blood 1995; 9: 201-12.

13. Loche F, Marc V, Carabone C, Bazex J. Manifestations cutanées d'une résistance à la protéine C activée. Ann Med Interne 1998; 149: 102-4.

14. Alarcon Segovia D, Ruiz-Argüelles GJ, Garcés-Eisele J, Ruiz-Argüelles A. Inherited activated protein C resistance in a patient with familial primary antiphospholipid syndrome. J Rheumatol 1996; 23: 2162-5.

15. Ruiz-Argüelles GJ, Garcés-Eisele J, Alarcon Segovia D, Ruiz-Argüelles A. Activated protein C resistance phenotype and genotype in patients with primary antiphospholipid syndrome. Blood Coagulat Fibrinol 1996; 73: 344-8.

16. Deschiens MA, Conard J, Horellou MH, et al. Coagulation studies, factor V Leiden, and anticardiolipin antibodies in 40 cases of cerebral venous thrombosis. Stroke 1996; 27: 1724-30.

17. Dizon-Townson D, Hutchinson C, Silver R, Branch DW, Ward K. The factor V Leiden mutation which predisposed to thrombosis is not common in patients with antiphospholipid syndrome. Thromb Haemost 1995; 74: 1029-31.

18. Brenner B, Vulfsons L, Lanir N, Nahir M. Coexistence of familial antiphospholipid syndrome and factor V Leiden: impact on thrombosis diathesis. Br J Haematol 1996; 94: 166-7.

19. Guba SC, Fonseca V, Fink LM. Hyperhomocysteinemia and thrombosis. Semin Thromb Hemostasis 1999; 25: 291-309.

20. Brattstrom L, Wilcken EL, Öhrvik J, Brudin L. Common methylene tetrahydrofolate reductase gene mutation leads to hyperhomocysteinemia but not to vascular disease. Circulation 1998; 98: 2520-6.

21. Lane DA, Grant PJ. Role of hemostatic gene polymorphism in venous and arterial thrombotic disease. Blood 2000; 95: 1517-32.

22. Ridker PM, Hennekens CH, Selhub J, et al. Interrelation of hyperhomocysteinemia, factor V leiden and risk of future venous thromboembolism. Circulation 1997; 95: 1777-82.