- Top of page
- PATIENTS AND METHODS
Behçet's disease (BD) is a multisystemic inflammatory disease of unknown origin characterized by recurrent oral aphtous ulcers, genital ulcers, uveitis, and skin lesions (1, 2). Vasculitis is the pathologic lesion common to most of the clinical manifestations of BD. Vascular lesions, although not included in the diagnostic criteria for BD by the International Study Group (3), are detected in 10–30% of patients (4). Superficial thrombophlebitis and involvement of the deep veins of the extremities are the most common vascular lesions.
The pathogenesis of thrombosis in BD is unknown, although the contribution of multiple factors is likely. Various abnormalities related to endothelial cell functions have been described in BD (4–9). Endothelial dysfunction resulting from vasculitis could play an important role in the pathogenesis of thrombotic manifestations observed in BD. However, thrombophilia may also play a key role. The factor V Leiden mutation is the most common cause of inherited thrombophilia. A single point mutation 1691G to A in the factor V gene produces a 506Arg to Gln replacement in the cleavage site of the factor V moiety, making it resistant to degradation by activated protein C. This mutation increases the risk of venous thrombosis (10–12). An odds ratio (OR) for venous thrombosis of 6.2 was estimated for factor V Leiden carriers in a large random sample of the Italian population (13). Discordant data have been reported on the association between deep vein thrombosis (DVT) and factor V Leiden mutation in BD (14–20).
A second mutation that is an established inherited cause of thrombophilia is the G20210A mutation in the 3′-untranslated region of the prothrombin gene. This mutation is reported to increase the risk of venous thrombosis by almost 3-fold (21). The 20210A allele is associated with elevated prothrombin levels, itself a risk factor of thrombosis (22). Interestingly, its geographic distribution parallels the prevalence of BD, the highest prevalence being found in southern Europe and the Middle East (23), where the prevalence of BD is higher. The association between this mutation and venous thrombosis in BD has been studied in Turkish and Spanish patients, with discordant data being reported (15, 20, 24).
The aim of this study was to determine whether or not these 2 mutations were associated with an increased risk of venous thrombosis in Italian patients with BD. We also evaluated the frequency and the types of vascular lesions in our series of Italian patients.
- Top of page
- PATIENTS AND METHODS
Table 1 shows the clinical and demographic characteristics of 118 Italian patients with BD. We observed 37 (31.4%) patients with vascular lesions. The 2 most common lesions were subcutaneous thrombophlebitis (12/118, 10.2%) and DVT of the legs (25/118, 21.2%). Two patients had isolated intracardiac thrombosis and 1 patient had associated Budd-Chiari syndrome and extensive inferior vena cava and leg vein thromboses. Therefore, a total of 27 patients had evidence of DVT. Arterial lesions were not present. There were no significant differences between patients with and without DVT (Table 1).
Table 1. Demographic and clinical features of 118 Italian patients with Behçet's disease*
| ||Total BD (n = 118)||BD without DVT (n = 91)||BD with DVT (n = 27)|
|Female/male||54/64 (45.8./54.2)||41/50 (45.1/54.9)||13/14 (48.1/51.9)|
|Age at disease onset, mean ± SD, years||30 ± 12||29 ± 11||31 ± 16|
|Disease duration, mean ± SD, years||12 ± 8||12 ± 8||12 ± 10|
|Oral ulcer||118 (100)||91 (100)||27 (100)|
|Genital ulcer||72 (61.0)||59 (64.8)||13 (48.1)|
|Cutaneous lesions||98 (83.1)||75 (82.4)||23 (85.2)|
| Erythema nodosum||49 (41.5)||37 (40.7)||12 (44.4)|
| Papulopustular lesions||67 (56.8)||51 (56.0)||16 (59.2)|
|Eye lesions||74 (64.4)||56 (61.5)||20 (74.1)|
| Anterior uveitis†||31 (26.3)||26 (28.6)||5 (18.5)|
| Posterior uveitis/retinal vasculitis||61 (51.7)||44 (48.4)||17 (63.0)|
|Arthritis||52 (44.1)||39 (42.9)||13 (48.1)|
|Subcutaneous thrombophlebitis||12 (10.2)||10 (11.0)||2 (7.4)|
|Central nervous system involvement||24 (20.3)||20 (22.0)||4 (14.8)|
|Epididymitis||6 (5.1)||5 (5.5)||1 (3.7)|
|Positive pathergy test result‡||26 (49.0)||17 (48.6)||9 (50.0)|
|HLA–B51§||54 (57.4)||41 (55.4)||13 (65.0)|
The allele and genotype frequencies of prothrombin gene G20210A and factor V gene G1691A polymorphisms in BD patients and in the control group are shown in Table 2. The distributions did not differ significantly between BD patients and healthy controls.
Table 2. Allele and genotype frequencies of prothrombin gene G20210A and factor V gene G1691A polymorphisms in Italian Behçet's disease patients and controls*
|Variable||Healthy controls (n = 132)||Behçet's disease (n = 118)||P||Odds ratio (95% CI)|
|Factor V G1691A|| || || || |
| Alleles|| || || || |
| A||5/264 (1.9)||9/236 (3.8)||NS||1.4 (0.9–2.0)|
| G||259/264 (98.1)||227/236 (96.2)|| ||0.7 (0.3–1.4)|
| Genotypes|| || || || |
| AA||0/132||0/118|| || |
| GA||5/132 (3.8)||9/118 (7.6)||NS|| |
| GG||127/132 (96.2)||109/118 (92.4)|| || |
|Prothrombin G20210A†|| || || || |
| Alleles|| || || || |
| A||5/264 (1.9)||8/234 (3.4)||NS||1.3 (0.8–2.0)|
| G||259/264 (98.1)||226/234 (96.6)|| ||0.7 (0.4–1.4)|
| Genotypes|| || || || |
| AA||0/132||0/117|| || |
| GA||5/132 (3.8)||8/117 (6.8)||NS|| |
| GG||127/132 (96.2)||109/117 (93.2)|| || |
The comparisons of the frequencies of carriage rates of prothrombin gene G20210A and factor V gene G1691A polymorphisms in controls and in BD patients with and without DVT are shown in Table 3. No significant differences were found. No patient was carrying both factor V Leiden and prothrombin gene G20210A mutations.
Table 3. Comparisons of carriage rates of prothrombin gene G20210A and factor V gene G1691A polymorphisms in controls and Italian Behçet's disease patients with and without deep vein thrombosis*
|Carriage rate||BD with DVT (n = 27) A||BD without DVT (n = 91) B||Control (N = 132) C||A versus C P||B versus C P||A versus B P|
|Factor V G1691A|| || || || || || |
| A||1 (3.7)||8 (8.8)||5 (3.8)||NS||NS||NS|
| G||26 (96.3)||83 (91.2)||127 (96.2)|| || || |
|Prothrombin G20210A†|| || || || || || |
| A||2 (7.4)||6 (6.7)||5 (3.8)||NS||NS||NS|
| G||25 (92.6)||84 (93.3)||127 (96.2)|| || || |
The HLA–B51 allele frequency was significantly higher in BD patients compared with healthy controls (57.4% versus 19.2%; P = 0.0001, OR 5.7, 95% CI 3.1–10.3). We also investigated possible associations of the 2 polymorphisms studied with BD with or without DVT stratifying on HLA–B51. Although this analysis was limited by the low number of patients studied, no significant associations were observed in HLA–B51-positive or HLA–B51-negative patients (data not shown).
The associations between prothrombin gene G20210A and factor V gene G1691A polymorphisms and BD clinical manifestations defined in Table 1 were evaluated in the 118 BD Italian patients comparing patients with and without manifestations. The frequency of 20210A allele was significantly higher in BD patients with ocular disease than in those without (10.8% versus 0%; P = 0.03, OR 11.4, 95% CI 5.2–24.9), particularly in the patients with posterior uveitis/retinal vasculitis (13.1% versus 0%; P = 0.004, OR 18.2, 95% CI 8.3–39.8).
- Top of page
- PATIENTS AND METHODS
We observed a 31.4% prevalence of vascular lesions in our consecutive series of Italian patients with BD. This frequency is similar to that reported in other studies. In particular, Koç et al, in a study designed to evaluate the prevalence of vascular involvement in BD, showed a prevalence of 27.7% in a series of Turkish patients (4). The 2 most common lesions observed in our study were subcutaneous thrombophlebitis and DVT of the legs. Two patients had isolated intracardiac thrombosis and 1 patient had associated hepatic vein thrombosis and extensive inferior vena cava and leg vein thromboses. Similar to the experience cited by Bayraktar et al (26), this patient had a catastrophic course with death occurring a few months after the development of vascular lesions. Arterial lesions were not present in our series of Italian patients.
We also compared the demographic and clinical characteristics of the patients with and without DVT. No significant differences were found. Koç et al reported a statistically significantly higher frequency of pathergy positivity, erythema nodosum, and eye involvement in patients with vascular lesions (4). Eye lesions, in particular posterior uveitis/retinal vasculitis, were also more frequent in our Italian patients with DVT, although the difference was not significant.
The factor V Leiden mutation is the most common cause of inherited thrombophilia (12). Seven studies have evaluated factor V Leiden as a potential risk factor for DVT in BD patients (14–20) (Table 4). Three studies have examined Turkish patients (14–16), 1 study examined patients from Saudi Arabia (17), 1 studied Palestinian and Jordan patients (18), and 2 studies examined European patients (19, 20). The results were controversial. Two of the studies reported a significant association between factor V Leiden mutation and venous thrombosis in BD (14, 17), but in the 4 other studies this association was not confirmed (15, 18–20). The seventh study was limited by the low number of patients with DVT (only 5) (16). Two of the studies were well-planned, case-control studies that evaluated 2 series of Turkish patients with BD referred to Ankara and Istanbul Universities (14, 15). Their results were conflicting. Gül et al (14) showed that heterozigosity for factor V gene mutation was associated with an almost 6-fold increase in the risk of venous thrombosis, whereas Toydemfir et al (15) showed that this mutation was not a risk factor for venous thrombosis in BD.
Table 4. Prevalence of the prothrombin gene G20210A and factor V gene G1691A mutations in Behçet's disease patients with and without vascular lesions: results from the literature*
|Reference||Geographic origin||Type of vascular lesion||Prevalence in BD patients with VL A||Prevalence in BD patients without VL B||Prevalence in controls C||P (OR, 95% CI) A versus B|
|Factor V gene G1691A mutation|| || || || || || |
| Gül et al (14)||Turkey||DVT||12/32 (37.5)||3/32 (9.4)||11/107 (10.3)||0.008 (5.8, 1.4–23.2)|
| Toydemir et al (15)||Turkey||DVT||10/30 (33.3)||6/30 (20.0)||4/100 (4.0)||NS|
| Öner et al (16)||Turkey||DVT||3/5 (60)||7/39 (17.9)||6/81 (7.1)||NR|
| Mammo et al (17)||Saudi Arabia||DVT||3/8 (37.5)||0/15||NR||0.03 (NR)|
| Verity et al (18)||Palestine, Jordan||DVT||8/24 (33.3)||21/82 (25.6)||23/120 (19.2)||NS|
| || ||RVO||11/25 (44.0)||5/31 (16.1)†|| ||0.02 (1.06, NR)|
| Lesprit et al (19)||France‡||DVT, AT||0/15||-||-||-|
| Espinosa et al (20)||Spain‡||DVT, stroke||0/14||0/24||2/100 (2.0)||NS|
| Present study||Italy||DVT||1/27 (3.7)||8/91 (8.8)||5/132 (3.8)||NS|
|Prothrombin gene G20210A mutation|| || || || || || |
| Gül et al (24)||Turkey||DVT||10/32 (31.3)||1/32 (3.1)||-||0.003 (14.1,1.7–118.2)|
| Toydemir et al (15)||Turkey||DVT||1/30 (3.3)||0/30||3/100 (3.0)||NS|
| Espinosa et al (20)||Spain‡||DVT, stroke||1/14 (7.1)||0/24||1/100 (1.0)||NS|
| Present study||Italy||DVT||2/27 (7.4)||6/90 (6.7)||5/132 (3.8)||NS|
In our study, we did not find any association between factor V Leiden mutation and venous thrombosis in Italian patients with BD. Similarly, the factor V gene mutation was not found in 2 series of BD patients with thrombosis, 1 from France (15 patients), the other from Spain (14 patients) (19, 20). Therefore, the factor V Leiden mutation seems not to be a risk factor for venous thrombosis in European patients with BD.
Toydemfir et al (15) found a high frequency of factor V Leiden mutation in BD patients without DVT and suggested an association between this mutation and BD. In our study, the frequencies of factor V Leiden mutation in the total BD patients and in the subgroup without DVT were similar to that of controls.
Another study established inherited cause of thrombophilia is the G20210A mutation in the 3′-untranslated region of the prothrombin gene (21). We described 2 cases of BD with thrombotic lesions that were heterozygous for the prothrombin G20210A mutation (27). We thus decided to plan the present study to confirm a possible association. We did not find any significant differences in the frequency of prothrombin gene mutation between Italian BD patients with and without DVT. A recent study from Spain, confirming our results, did not find any association between this mutation and thrombosis (prevalently DVT) in patients with BD (Table 4) (20). Discordant data are reported on Turkish BD patients. A heterozygous prothrombin gene G20210A mutation was found to be significantly associated with DVT by Gül et al (24); this was not the case in the Toydemfir et al study (15). In these 2 Turkish studies, there were large differences in the prevalence of this mutation in BD patients with and without DVT (31.3% versus 3.3%).
There are different possible explanations for the conflicting results regarding the association of these 2 thrombophilic mutations with BD; allelic heterogeneity existing between ethnic groups, modification of the association by other genetic or environmental factors that vary between the populations studied, selection bias in sampling of cases or controls, and publication bias. However, one of the most important causes of failure to replicate findings in genetic association studies is the inadequacy of sample sizes (28). So far, the maximum number of patients with DVT studied has been only 32. To test the association between thrombophilic gene polymorphisms and DVT in BD, a larger group of cases with DVT would be needed. Therefore, multicenter collaborations to recruit an adequate number of cases are required.
In our study, we also evaluated whether prothrombin gene G20210A and factor V gene G1691A polymorphisms were associated with specific clinical findings comparing patients with and without certain manifestations. The frequency of the 20210A allele was significantly higher in BD patients with ocular disease than in those without. These data suggest that a prothrombin gene G20210A mutation is a risk factor for the development of ocular disease. However, because the strength of the association was higher in BD patients with more severe ocular disease (posterior uveitis/retinal vasculitis), this polymorphism may also influence the severity of ocular involvement. Additional studies are needed to replicate the association between gene G20210A mutation and ocular disease. Recurrent vascular occlusion due to an obliterative retinal vasculitis and vascular thrombosis is a major cause of visual loss in BD. A prothrombotic condition may be implicated in the pathogenesis of ocular inflammatory disease in BD.
Verity et al (18) found a significant association between factor V Leiden mutation and retinal vascular occlusion in BD (Table 4). However, the strength of the association was weak (OR 1.06) and the significance was lost upon comparison of ocular patients with and without vasoocclusive disease. We failed to find any association between this polymorphism and ocular lesions.
In conclusion, our study shows a frequency of vascular lesions in Italian BD patients similar to that reported in studies from other countries. However, we did not find any association between venous thrombosis and the factor V Leiden mutation or the G20210A mutation in the 3′-untranslated region of the prothrombin gene. Thrombotic lesions in Italian BD patients seem more probably related to the endothelial activation induced by the vasculitic process. Furthermore, our data support the hypothesis that prothrombin gene G20210A mutation may influence the development and severity of ocular involvement in BD.