Jürg Hafner, Andreas Kühne, Beatrice Schär, Thomas Bombeli, Markus Hauser, René Lüthi, Edgar Hänseler. Factor V Leiden Mutation in Postthrombotic and Non-postthrombotic Venous Ulcers. Arch Dermatol. 2001;137(5):599–603. doi:10-1001/pubs.Arch Dermatol.-ISSN-0003-987x-137-5-dst0025
To determine the prevalence of the factor V Leiden mutation in patients with postthrombotic and non-postthrombotic venous ulcers.
Department of Dermatology, University Hospital of Zurich, Zurich, Switzerland.
Seventy-three consecutive outpatients and inpatients with venous ulcers and 45 age- and sex-matched control subjects (matched to the 42 patients with postthrombotic syndrome).
Main Outcome Measures
Frequency of postthrombotic and non-postthrombotic findings in patients with venous ulcers. Prevalence of the factor V Leiden mutation in these different subgroups.
Postthrombotic syndrome was identified as the cause of 42 (58%; 95% confidence interval [CI], 45%-69%) of 73 venous ulcers, and the remainder were caused by primary valvular insufficiency. In postthrombotic ulcers, the prevalence of the factor V Leiden mutation was 38% (95% CI, 24%-54%) (16/42), which corresponds to an odds ratio of 13.2 (95% CI, 2.8-62.3; P<.001). In non-postthrombotic venous ulcers, the prevalence was 16% (95% CI, 5%-34%) (5/31), which corresponds to an odds ratio of 3.2 (95% CI, 1.0-10.0; P = .07).
The factor V Leiden mutation is highly prevalent in patients with postthrombotic venous ulcers. Even patients with non-postthrombotic venous ulcers show a moderately elevated prevalence of the factor V Leiden mutation. Some of the latter might be misclassified because of near-to-perfect revascularization after asymptomatic deep venous thrombosis. However, as long as the therapeutic consequences of the factor V Leiden mutation are not established, systematic screening cannot be recommended in patients with venous ulcers.
LEG ULCERS are a frequent disorder, with a point prevalence of 0.3% to 0.6% and a lifetime cumulative risk of 1.0% to 1.8%.1 The care of these chronic wounds generates considerable costs for health care systems (eg, £400 million per annum for Great Britain).1 Venous ulcers represent approximately 70% of all causes of leg ulceration and are considered the most advanced stage of chronic venous insufficiency. In contrast to the common belief that virtually all venous ulcers are due to postthrombotic disease, several more recent studies using duplex ultrasound and venography revealed that only 50% to 60% of venous ulcers are the result of postthrombotic vein damage and that the remainder develop in non-postthrombotic conditions, ie, in primary valvular insufficiency.2
Thrombophilia is defined as a propensity to develop venous thromboembolism. This can be acquired or determined genetically.3 The discovery of the factor V Leiden mutation4- 6 has added substantially to the elucidation of thrombophilia. Its prevalence has been reported at 21% in patients with a first episode of deep venous thrombosis (DVT)6 and at 50% in patients with a personal or family history of thrombophilia.3 To date, 4 clinical studies7- 10 have investigated the prevalence of the factor V Leiden mutation in patients with venous leg ulcers. A Danish case-control study7 found an elevated prevalence of activated protein C (APC) resistance in venous ulcers (12 of 46 or 26%), and a recent Dutch case-control study8 confirmed a prevalence of 21 carriers (23%) in 92 patients by molecular testing for the factor V Leiden mutation. Based on patient history, the Danish study7 found that the APC resistance is not more prevalent in postthrombotic syndrome than in non-postthrombotic primary venous insufficiency, whereas the Dutch study8 found a significant association between the factor V Leiden mutation and a history of venous thromboembolism. In an uncontrolled study of 29 patients with venous ulcers, APC resistance was found in 11 (38%), but molecular testing confirmed the factor V Leiden mutation in only 2 of these 11 cases (2 of 29 or 7%).9 A more recent French case-control study10 found APC resistance in 4 (11%) of 35 patients, but the factor V Leiden mutation was confirmed in only 1 of these 4 cases (1 of 35 or 3%). Fourteen (40%) of the 35 enrollees had a history of DVT.10
Assuming that 50% to 60% of venous ulcers are postthrombotic,2 most patients with thrombophilia are expected to be in this subgroup. This led us to perform a case-control study on the prevalence of the factor V Leiden mutation in patients with venous leg ulcers. All patients underwent extensive vascular examination to distinguish postthrombotic from non-postthrombotic venous disease. Screening for APC resistance and the factor V Leiden mutation4- 6 was performed in all patients and controls.
From April 1, 1997, to December 31, 1998, 172 consecutive patients with leg ulcers (112 women and 60 men; median age, 78.5 years; 95% confidence interval [CI], 49.1-91.0 years) admitted to the Department of Dermatology, University Hospital of Zurich, Zurich, Switzerland, were subjected to vascular screening with Doppler ultrasound. One hundred eight patients (63%) showed the typical clinical signs of chronic venous insufficiency (edema, dermatoliposclerosis, hemosiderotic pigmentation, and ulceration, commonly on the medial aspect of the gaiter area) and deep, superficial, or combined deep and superficial venous reflux on handheld Doppler ultrasound screening examination. Twenty-six patients with an ankle-brachial index of 0.9 or less were excluded from the study. This cutoff level has been shown to detect peripheral arterial disease with high sensitivity and specificity.11 The remaining 82 patients were diagnosed as having venous ulcers: 9 declined further examinations and 73 gave informed consent to participate in the study (47 women and 26 men; mean age, 75.1 years; 95% CI, 59.0-89.8 years) (Figure 1). Most included patients with leg ulcer were of Western European origin.
During further evaluation, all patients were examined using the CEAP classification (Clinical grade, Etiology, Anatomy, Pathophysiology) for reporting venous disease.2 Vascular examination comprised either antegrade (ascending) and retrograde (descending) venography or color-coded duplex ultrasound. The examiners performing venography or duplex ultrasound were masked to the results of the hemostasis tests.
Forty-one patients evaluated for possible vascular surgery (stripping of the greater or lesser saphenous vein, subfascial endoscopic perforator vein surgery) were examined by antegrade and retrograde venography of the affected leg as described elsewhere.12 Specific criteria to make the diagnosis of a postthrombotic syndrome on venography were 1 or more of the following: complete occlusion or partial recanalization, evidence of collateral circulation, or pronounced irregularities of vessel wall and vessel diameter along with distorted valves.12
Thirty-two patients were examined by means of duplex Doppler ultrasound using a 7-MHz linear array transducer (model 128XP10; Acuson Computed Sonography, Mountain View, Calif). Patients were examined in the supine and standing positions as described elsewhere.13 Specific criteria to make a diagnosis of a postthrombotic syndrome on duplex ultrasound were 1 or more of the following: complete occlusion, partial recanalization, and venous wall thickening.
Based on the patient's medical history and results of the vascular examination, the 73 enrollees were classified into 3 groups. Group 1 consisted of patients with a history of confirmed DVT (confirmed by ultrasound or venography) or specific postthrombotic findings according to the vascular workup. Group 2 consisted of patients who undoubtedly recalled a history of DVT that required oral anticoagulant therapy. However, this diagnosis was made clinically and was not confirmed by venography or ultrasound. The current vascular assessment did not show specific postthrombotic findings. Group 3 consisted of patients with no history of DVT or postthrombotic findings at the current vascular workup; they were referred to as having primary valvular insufficiency. Control subjects were recruited from a cohort of patients treated for basal cell carcinoma during the past 2 years at the Department of Dermatology, University Hospital of Zurich. They were otherwise healthy, and, specifically, they had no history of other cancers or venous thromboembolism. Clinical examination of the legs did not show any signs of chronic venous insufficiency. Controls were matched to the collective of groups 1 and 2, ie, patients in whom a postthrombotic syndrome was confirmed or considered likely.
Resistance to APC (functional test) and the factor V Leiden mutation (molecular test) were investigated in all 73 patients and 45 controls. Venous blood samples were drawn into vacuum tubes containing citrate (for APC resistance) and EDTA (for DNA extraction).
Resistance to APC was determined as the ratio between the activated partial thromboplastin time with and without the addition of APC (Coatest APC resistance; Chromogenix, Möndal, Sweden). The factor V Leiden mutation (G1691A) was determined by allele-specific polymerase chain reaction (allele-specific amplification).14,15 Polymerase chain reaction fragments were analyzed on 7% polyacrylamide gels and visualized by ethidium bromide staining. All experiments were repeated at least twice.
Continuous data are expressed as medians with 95% CIs, and percentages are expressed with 95% CIs. StatView 5.0 software (SAS Institute Inc, Cary, NC) was used for statistical calculations. Discrete (case-control) data were compared using the Fisher exact test. Odds ratios (ORs) with 95% CIs were calculated on the basis of binomial distribution. P<.05 was considered significant.
The study was approved by the local ethical committee, and written informed consent was obtained from all participants.
Forty-two patients with venous ulcers (58%; 95% CI, 45%-69%) were diagnosed as having postthrombotic syndrome (either confirmed or likely, ie, group 1 plus group 2) and 31 patients (42%; 95% CI, 31%-55%) had primary valvular incompetence (group 3). The results of the vascular workup are summarized in Table 1.
Group 1 comprised 22 patients (9 men and 13 women) with a median age of 73.6 years (95% CI, 54.8-87.8 years). Five of these patients (23%; 95% CI, 8%-45%) were carriers of the factor V Leiden mutation (all heterozygous) (Table 2).
Group 2 comprised 20 patients (8 men and 12 women) with a median age of 76.7 years (95% CI, 61.0-92.0 years). Eleven of these patients (55%; 95% CI, 32%-77%) were carriers of the factor V Leiden mutation (10 heterozygous and 1 homozygous) (Table 2).
Group 3 comprised 31 patients (9 men and 22 women) with a median age of 75.4 years (95% CI, 48.1-89.0 years). Five of these patients (16%; 95% CI, 5%-34%) were carriers of the factor V Leiden mutation (all heterozygous) (Table 2).
Forty-five controls (18 men and 27 women) were matched to the 42 patients in groups 1 and 2. The median age was 75.1 years (95% CI, 60.0-86.8 years). Two controls (4%; 95% CI, 1%-15%) were carriers of the factor V Leiden mutation (both heterozygous) (Table 2).
Taking groups 1 and 2 together (n = 42), the factor V Leiden mutation was found in 16 of 42 patients with postthrombotic syndrome (38%; 95% CI, 24%-54%) (OR, 13.2; 95% CI, 2.8-62.3; P<.001). Fifteen patients were heterozygous and 1 was homozygous for this mutation. The prevalence of the factor V Leiden mutation in patients with primary valvular insufficiency was 16% (95% CI, 5%-34%) (OR vs controls, 4.1; 95% CI, 0.7-22.9; P = .11; OR vs groups 1 and 2, 3.2; 95% CI, 1.0-10.0; P = .07). The functional test for APC resistance and the molecular test for the factor V Leiden mutation were congruent in all participants (patients and controls).
In groups 1 and 2, no association between the factor V Leiden mutation and recurrent (≥2) DVTs could be found. Seven of 16 carriers had a history of 2 or more DVTs compared with 9 of 26 noncarriers (P = .74).
The present study confirms an increased prevalence of the factor V Leiden mutation in patients with venous ulcers (21 of 73 or 29%) and especially in patients with postthrombotic venous ulcers (16 of 42 or 38%).
The distinction of postthrombotic syndrome from primary valvular incompetence represents a principal problem in the assessment of patients with advanced chronic venous insufficiency. The first problem is that 50% of DVTs remain clinically asymptomatic.16 Therefore, the history of previous DVT in a patient with venous ulcer might be unreliable. Second, venous ulcers are much more frequent in older age.1 This means that many patients have experienced DVT at a time when venography was not a routine examination establishing the diagnosis of DVT and B-mode ultrasound was not available. Third, specific postthrombotic changes (see the "Participants and Methods" section) are detectable in only 35% of DVT events. Hence, complete recanalization takes place in 65% over time. Complete morphological and functional restitution (without any reflux) can be observed in 12% to 25% of patients with DVT.17,18 Deep venous reflux in the absence of morphological signs of postthrombotic damage cannot specifically be used to distinguish postthrombotic syndrome from primary valvular insufficiency of the deep venous system.
Therefore, we decided to address the previously mentioned problems by classifying patients with venous ulcers into 3 groups (see the "Participants and Methods" section). In our consecutive series of 73 patients with venous ulcers, 22 (30%) were classified as having confirmed postthrombotic syndrome, 20 (27%) as having postthrombotic syndrome based on a suggestive history of DVT alone, and 31 (42%) as having primary valvular incompetence. This distribution is in accordance with the literature.2,19
It would have been preferable to use a single method for the vascular assessment of all patients, ie, noninvasive duplex ultrasound. At the time when this study was conducted, vascular surgeons at our institution preferred to use ascending and descending venography in patients evaluated for vein surgery. These patients did not have duplex ultrasound at the same time. We believe that use of either venography (ascending and descending) or duplex ultrasound during the present study was justified because both methods are highly sensitive and specific in the detection of postthrombotic changes.17,20,21
Group 1 had a lower prevalence of the factor V Leiden mutation (5/22, 23%; 95% CI, 8%-45%) than group 2 (11/20, 55%; 95% CI, 32%-77%). This difference is likely to be caused by the small patient numbers in the subgroups, which results in less precise prevalence data, as reflected by large CIs. The prevalence of the factor V Leiden mutation in postthrombotic venous ulcers can be estimated by combining the rates of groups 1 and 2 (16/42, 38%; 95% CI, 24%-54%). Thus, a subsequent study with a larger number of patients is still necessary to determine the prevalence of the factor V Leiden mutation in patients with leg ulcers and a confirmed postthrombotic syndrome.
The 38% prevalence of factor V Leiden mutation in patients with postthrombotic venous ulcers is remarkable compared with the 21% in unselected patients experiencing a first episode of DVT.6 In the literature, recurrent DVT is associated with severe chronic venous insufficiency.22 In fact, the aforementioned Dutch study8 confirmed a strong association between the factor V Leiden mutation and a history of recurrent DVT. This was not confirmed in our study. Finally, patient selection bias at tertiary referral centers might contribute to higher prevalences of the factor V Leiden mutation in studies that are conducted at university hospitals.
The prevalence of primary valvular insufficiency might be overestimated by our classification of patients with venous ulcers. Patients experiencing an asymptomatic DVT with subsequent complete recanalization will not show specific postthrombotic changes at vascular workup and, hence, will be classified as having primary valvular insufficiency. As a matter of fact, the frequency of the factor V Leiden mutation was also elevated in our patients classified as having primary valvular insufficiency (16%; P = .07), indicating that a few patients classified as postthrombotic might have been misclassified. The theoretical prevalence of the factor V Leiden mutation in patients with primary valvular insufficiency would have been expected to be approximately 5%.
The therapeutic consequences of the previously mentioned findings are unknown. A recent study23 has shown that extended anticoagulant therapy was effective in preventing recurrences after a first episode of idiopathic venous thromboembolism. Other investigators,24 however, have shown by means of a decision analysis model that the risk of major hemorrhage with prolonged oral anticoagulant therapy (1-5 years) would probably exceed the benefit of prevented pulmonary embolism in factor V Leiden carriers who experienced a first episode of DVT. In conclusion, a large-scale epidemiologic study should be undertaken to establish the prevalence of the factor V Leiden mutation in patients with postthrombotic and non-postthrombotic venous ulcers more exactly; a randomized controlled trial of prolonged vs standard25 anticoagulant therapy in factor V Leiden–positive patients with venous ulcers who have a new episode of DVT would be required to establish the risk-benefit ratio of this intervention. As long as therapeutic consequences are not clarified, there is no need for systematic screening for the factor V Leiden mutation in patients with venous leg ulcers.
Accepted for publication June 18, 2000.
This project was supported by funds from the Division of Haematology and Haemostasis (functional test for APC resistance) and from the Institute of Clinical Chemistry (molecular test for the factor V Leiden mutation), University Hospital of Zurich.
We thank B. Seifert, PhD, for statistical advice and E. P. Scheidegger, MD, for proofreading the manuscript.
Corresponding author and reprints: Jürg Hafner, MD, Department of Dermatology, University Hospital of Zurich, CH-8091 Zurich, Switzerland (e-mail: firstname.lastname@example.org).