- Email: [email protected]
The prevalence of thrombophilia in patients with chronic venous leg ulceration
The prevalence of thrombophilia in patients with chronic venous leg ulceration Rhoda K. MacKenzie, FRCSEd,a Christopher A. Ludlam, PhD, FRCPath,b C. Vaughan Ruckley, ChM, FRCSEd,a Paul L. Allan, FRCR,c Paul Burns, AFRCS,d and Andrew W. Bradbury, MD, FRCSEd,d Edinburgh and Birmingham, United Kingdom Background: Thrombophilia is increasingly recognized as a risk factor for deep venous thrombosis (DVT), which in turn is a major risk factor for chronic venous ulceration (CVU). However, the relationship between thrombophilia and CVU remains unknown. The aim of this study was to define the prevalence of thrombophilia in patients with CVU and to determine whether this is associated with a history or duplex scan evidence of DVT. Methods: Eighty-eight patients with CVU were prospectively studied. The patients underwent clinical assessment and duplex ultrasound scanning. Blood was drawn for antithrombin, proteins C and S, activated protein C resistance, factor V Leiden, prothrombin 20210A, lupus anticoagulant, and anticardiolipin antibodies. Results: The study included 35 men with a median age of 61 years (interquartile range, 45 to 72 years) and 53 women with a median age of 76 years (interquartile range, 69 to 82 years). Thirty-six percent of the patients had either a history or duplex scan evidence suggestive of previous DVT. The following abnormalities were detected: four, five, and six cases of antithrombin, protein C, and protein S deficiencies, respectively; 14 cases of activated protein C resistance; 11 cases of factor V Leiden mutation; three cases of prothrombin 20210A mutation; eight cases of lupus anticoagulant; and 12 cases of anticardiolipin antibodies. Thrombophilia was not significantly related to previous DVT, deep reflux, or disease severity. Conclusion: Patients with CVU have a 41% prevalence rate of thrombophilia. This rate is two to 30 times higher than the rate of the general population but is similar to that reported for patients with previous DVT. However, in patients with CVU, thrombophilia does not appear to be related to a history of DVT, a pattern of reflux, or severity of disease. Many patients with CVU may have unsuspected postthrombotic disease. (J Vasc Surg 2002;35:718-22.)
In the United Kingdom, the life-time risk of development of chronic venous ulceration (CVU) is 1%,1 and the annual cost of treatment is estimated at approximately ,£600 million ($1 billion), which constitutes 1% to 2% of total healthcare spending. 2 Similar figures pertain in other European countries and in North America.3 Thrombophilia is increasingly recognized as a risk factor for deep venous thrombosis (DVT), which in turn is a major risk factor for chronic venous insufficiency and CVU.4 Specifically, the available literature suggests that between 25% and 75% of patients with CVU have postthrombotic disease on the basis of a history of DVT or evidence of postphlebitic changes in the deep venous system on investigation.5 However, these estimates vary widely according to the population studied and the thoroughness with which history and investigative evidence of prior DVT are sought. The association between CVU and DVT
From the Vascular Surgery Unit,a Department of Haematology,b Department of Medical Imaging,c Royal Infirmary, and the University Department of Vascular Surgery, Heartlands Hospital.d Competition of interest: nil. Reprint requests: Prof A. W. Bradbury, Lincoln House (Research Institute), Birmingham Heartlands Hospital, Bordesley Green East, Birmingham, B9 5SS, United Kingdom (e-mail: [email protected]). Copyright © 2002 by The Society for Vascular Surgery and The American Association for Vascular Surgery. 0741-5214/2002/$35.00 + 0 24/1/121749 doi:10.1067/mva.2002.121749
718
is further confused because DVT may be subclinical, patients may confuse superficial thrombophlebitis with DVT when giving a history, and duplex ultrasound scanning is less sensitive than venography in the detection of minor postthrombotic changes. On theoretic grounds, it seems plausible to hypothesize that postthrombotic CVU in the presence of thrombophilia may be more severe because the thrombosis is likely to be recurrent, more extensive, and, possibly, more resistant to both treatment and endogenous fibrinolysis and recanalization. If a clear relationship between a proportion of CVU and certain thrombophilias could be shown, new preventative and treatment opportunities might open up. The aim of this study, therefore, was to determine the prevalence of thrombophilia in CVU and to examine, for the first time, whether thrombophilia is associated with a history and or duplex scan evidence of DVT and clinical severity of chronic venous disease. METHODS Clinical and duplex scan assessment. This study was approved as part of a larger study by the local ethical committee board. Informed, signed consent was obtained from all patients before investigation. Patients seen at a hospital-based, “one-stop,” leg ulcer clinic with ulcerated limbs underwent history and physical examination and measurement of ankle-brachial pressure index. Note was made of previous major trauma, major surgery (including orthopedic), obesity (with body mass index), and use of the
JOURNAL OF VASCULAR SURGERY Volume 35, Number 4
MacKenzie et al
719
Table I. Relationship between history of deep venous thrombosis, duplex scan evidence of previous deep venous thrombosis, and deep venous reflux Duplex scan evidence of previous DVT History of DVT
Yes
Yes No
No
19 (19) 3 (3)
10 (5) 56 (26)
Numbers in brackets are those patients with deep venous reflux in at least the popliteal segment. DVT, Deep venous thrombosis.
Table II. Demographics of patients with and without abnormal thrombophilia screen results No P thrombophilia Thrombophilia value All patients No. of patients
52
36
50:50
25:75
.029†
Median age (IQR; years)
69 (56-76)
77 (65-87)
.005‡
Median age at first ulcer
54 (40-66)
66 (45-83)
.029‡
33%
42%
NS†
64
24
50:50
33:67
NS†
Median age (IQR; years)
70 (56-77)
75 (65-85)
NS‡
Median age at first ulcer (IQR; years)
56 (40-68)
62 (37-81)
NS‡
36%
38%
NS†
Male:female ratio (%)
Previous DVT Patients after exclusion* No. of patients Male:female ratio (%)
Previous DVT
*After exclusion of those patients with isolated antiphospholipid antibodies. †P value determined with 2 test. ‡P value determined with Mann-Whitney test. IQR, Interquartile range; NS, not significant; DVT, deep venous thrombosis.
contraceptive pill. Venous duplex ultrasonography (HDI 5000, Advanced Technology Laboratories, Bothell, Wash) of the ulcerated leg(s) was performed by a single observer (RKM). Nonulcerated limbs did not undergo duplex scanning because of time restrictions. The patients were examined in reverse Trendelenburg position at an angle of 45 degrees on a tilting couch. The following veins were examined: proximal and distal super- ficial femoral, above-knee and below-knee popliteal, posterior tibial, peroneal, saphenofemoral and saphenopopliteal junctions, above-knee and below-knee long saphenous, short saphenous, and gastrocnemius/soleal. Reflux was induced with a manual calf squeeze and was considered to be signifi-
Fig 1. Prevalence of thrombophilia in patients with chronic venous ulceration. AT, Antithrombin deficiency; PC, protein C deficiency; PS, protein S deficiency; APCR, activated protein C resistance; fVL, factor V Leiden; fIIL, (prothrombin 20210A) factor II Leiden; LA, lupus anticoagulant; ACLA, anticardiolipin antibody.
cant if it exceeded 0.5 seconds.6,7 Post-thrombotic features, such as deep vein obstruction, partial recanalization, vein wall irregularity, or valve cusp thickening, were noted. Deep venous reflux (DVR) was defined as significant reflux in the popliteal or superficial femoral vein. A history of DVT was taken from clinical records and from patient recollection of limb swelling, pain, or “white leg.” CVU was diagnosed on the basis of venous reflux on duplex scan results and exclusion of other causes. No patient had a history of DVT less than 1 year before the clinic appointment. Thrombophilia testing. A standard thrombophilia screen was performed that examined the congenital conditions of antithrombin, protein C, and protein S deficiency and factor V Leiden and prothrombin 20210A mutations. The acquired antiphospholipid antibodies (see subsequent) were also assayed. Venous blood was drawn into 3.8% trisodium citrate (1:9 volume) for thrombophilia investigations. The following tests were performed: antithrombin activity (Dade Behring [Deerfield, Ill], healthy range, 0.8 to 1.20 IU/mL), protein C (Chromatic Chromogenix [Mölndal, Sweden], healthy range, 0.67 to 1.38 IU/mL), protein S (healthy range, free 0.61 to 1.43 and total 0.64 to 1.54 IU/mL), activated protein C resistance (Coatest Chromogenix [Mölndal, Sweden], healthy range, >2.0), factor V Leiden and prothrombin 20210A with Multiplex allele-specific amplification polymerase chain reaction (MJ Research, Inc, Waltham, Mass), and the antiphospholipid antibodies (lupus anticoagulant; Diagen [Diagnostic Reagents Ltd, Thane, Oxon, UK], healthy range, >1.09:1) and anticardiolipin antibodies (Organon Technika [Rochville, Md], healthy range, immunoglobulin G < 13 G units/mL and immunoglobulin M < 10 M units/mL). All patients in
JOURNAL OF VASCULAR SURGERY April 2002
720 MacKenzie et al
Fig 2. Number of patients with multiple thrombophilic tendencies.
Table III. Severity of disease in patients with and without thrombophilia Median values (IQR) Total ulcer disease duration (months) Duration of this episode (months) Total area (cm2) No. of episodes Level of pain (0-3)
No thrombophilia (n = 52) 78 6 3.8 2 2
(24-234) (2-24) (0.1-9.5) (1-5) (1-3)
Thrombophilia (n = 36)
P value
48 (16-240) 10 (3-18) 2.9 (1.8-12) 2 (1-5) 2 (0-3)
NS* NS* NS* NS† NS†
*P
value determined with Mann Whitney test. value determined with 2 test. IQR, Interquartile range; NS, not significant. †P
whom an abnormality was detected (ie, less than healthy range for antithrombin, protein C, protein S, and activated protein C resistance or more than healthy levels for the antiphospholipid antibodies) were asked to return for repeat thrombophilia screen after a minimum of 3 months. Only patients with abnormalities on both occasions were considered to have a thrombophilia. Statistical methods. SPSS version 10.0.5 package (Statistical Package for Social Sciences Inc, Chicago, Ill) was used for all statistical analysis. For categoric data, 2 test or Kruskal-Wallis test were applied, and for continuous variables, the Mann-Whitney test was used. Exclusions. Six patients who were already undergoing oral anticoagulation therapy were excluded because of the known interference of warfarin with thrombophilia testing. One patient was excluded on the basis of having a thrombophilic abnormality on only one occasion. Two patients refused blood sampling. RESULTS Patients. This was a prospective study of 97 consecutive patients with CVU who attended a hospital-based leg ulcer clinic. Nine patients were excluded for the reasons shown previously, which left 88 patients in the study
group. There were 35 men with a median age of 61 years (interquartile range, 45 to 72 years) and 53 women with a median age of 76 years (interquartile range, 69 to 82 years; P = .001). Twelve men (34%) and 20 women (38%) gave a history of ipsilateral lower limb DVT. Duplex scan findings. There was only a weak relationship between a history of DVT and evidence of previous DVT on duplex scan results, although a history of DVT was strongly associated with DVR (P < .001; Table I). Thus, of the 29 patients who gave a history of DVT, only 19 (66%) had evidence of previous DVT on duplex scan results, although 24 of 29 (83%) had DVR. A further three patients had evidence of previous DVT on scan results but were unaware they had had a DVT and had no record of this in their hospital notes. Thrombophilia. Overall, 36 of 88 patients (41%) had at least one abnormality on thrombophilia test results (Fig 1; Table II). The significance of the presence of antiphospholipid antibodies alone is less certain8 than the other factors in terms of causing venous thromboembolism, and the condition is fairly common in the population, particularly among older women. If isolated antiphospholipid antibodies are excluded from the overall analysis results, there was no age or gender difference between those patients with and without
JOURNAL OF VASCULAR SURGERY Volume 35, Number 4
MacKenzie et al
Table IV. Prevalence of thrombophilia in patients with and without a history of ipsilateral deep venous thrombosis History of DVT Antithrombin deficiency Protein C deficiency Protein S deficiency Activated protein C resistance Factor V Leiden Prothrombin G20210A Lupus anticoagulant Anticardiolipin antibody
3 2 3 5 1 4 2 6
No history of DVT
(10%) (7%) (10%) (17%) (3%) (14%) (7%) (20%)
1 3 3 9 2 7 6 6
(2%) (5%) (5%) (15%) (3%) (12%) (10%) (10%)
Table VI. Patterns of reflux (>0.5 seconds) in patients with chronic venous ulceration with and without thrombophilia
P value* NS NS NS NS NS NS NS NS
721
No thrombophilia (n = 52) DVR only SVR only DVR and SVR Any DVR Any SVR
4 20 26 30 46
(8%) (38%) (50%) (58%) (88%)
Thrombophilia (n = 36) 5 11 19 24 30
(14%) (31%) (53%) (67%) (83%)
P value* NS NS NS NS NS
*P
value determined with 2 test. DVR, Deep venous reflux; SVR, superficial venous reflux; NS, not significant.
*P
value determined with 2 test. DVT, Deep venous thrombosis; NS, not significant.
Table V. Prevalence of thrombophilia in patients with and without duplex scan evidence of previous ipsilateral deep venous thrombosis Median values (IQR)
Table VII. Prevalence of specific thrombophilias in patients with chronic venous ulceration, in patients with first episode of venous thromboembolism, and in the general population Patients with CVU (n = 88)
Duplex scan No duplex scan P evidence of DVT evidence of DVT value*
Antithrombin deficiency Protein C deficiency Protein S deficiency Activated protein C resistance Factor V Leiden Prothrombin G20210A Lupus anticoagulant Anticardiolipin antibody
3 2 2 5
(14%) (9%) (9%) (23%)
1 3 4 9
(2%) (5%) (6%) (14%)
.02 NS NS NS
2 3 1 6
(9%) (14%) (5%) (27%)
1 8 7 6
(2%) (12%) (11%) (9%)
NS NS NS .03
*P
value determined with 2 test. IQR, Interquartile range; DVT, deep venous thrombosis; NS, not significant.
thrombophilia (Table II). Thrombophilia was not related to the clinical severity of the ulcer disease (ie, duration of disease, total area, number of episodes, or level of pain; Table III), previous history of DVT (Table IV), duplex scan evidence of DVT (apart from antithrombin and anticardiolipin antibodies; Table V), or the pattern of deep and superficial reflux (Table VI). No pattern to the position of ulceration or lipodermatosclerosis was identified between the groups. One patient gave a family history of DVT, but affected family members had not undergone testing for thrombophilia. There was no association between the number of thrombophilic abnormalities and the severity of disease or the pattern or severity of reflux (Fig 2). Interestingly, the patient with five abnormalities had no history of DVT and no evidence of deep venous disease. A history of trauma, major surgery (including orthopedic), cancer, obesity (body mass index >25), and use of the contraceptive pill did not increase the prevalence of DVT in patients with thrombophilia. DISCUSSION The principal finding of this study is that more than 40% of patients with CVU seen at a hospital-based leg
LA ACA Antithrombin PC PS APCR FVL Prothrombin 20210A
8 12 4 5 6 14 11 3
(9%) (14%) (5%) (6%) (7%) (16%) (13%) (4%)
Patients with first General episode of VTE population 10% 20% 1% 3% 5% 20% 18% 6%
2% 10% 0.2% 0.3% 0.2% 6% 7% 2%
First venous thromboembolism and general population prevalence figures from world literature.8,17-26 CVU, Chronic venous ulceration; VTE, venous thromboembolism; LA, lupus anticoagulant; ACA, anticardiolipin antibodies; PC, protein C; PS, protein S; APCR, activated protein C resistance; FVL, factor V Leiden.
ulcer clinic have at least one identifiable thrombophilia. This rate is two to 30 times higher than the rate in the general population but similar to that reported for patients with DVT (Table VII). Given that DVT is the single most common identifiable risk factor for the development of CVU and that patients with postthrombotic disease have a significantly worse prognosis in response to medical9 and surgical therapy,10,11 it is perhaps surprising that little information exists regarding the relationship between thrombophilia and CVU. The studies that do exist are small and often methodologically flawed and give a conflicting picture with regard to the relationship between thrombophilia and a clinical history of DVT. Munkvad and Jorgensen,12 Ribeaudeau et al,13 Alcaraz et al,14 and Hafner and Felten15 found that, in patients with CVU, there was no difference in DVT rates between those with and without thrombophilia. By contrast, Maessen-Visch et al16 found that in patients with CVU, 91% of patients with positive results for factor V Leiden gave a history of DVT as compared with 48% of those without the mutation.
722 MacKenzie et al
However, these patients did not all have documented DVT and did not undergo investigation with venography or duplex ultrasound scan. Between 25% and 75% of patients with CVU are thought to have sustained a previous DVT on the basis of history or investigation.5 However, as present data confirm, a history of DVT is unreliable with both false positive and negative results. Furthermore, duplex scan is less capable than venography, rarely performed these days in the United Kingdom, in distinguishing postthrombotic deep venous disease from deep primary valvular incompetence. This is especially so in the veins of the calf. Interestingly, in this study, there appeared to be no relationship between the presence of thrombophilia and the severity of disease, pattern of reflux, history of DVT, or duplex scan evidence of postthrombotic disease. Neither specific coagulation defects nor the presence of a hypercoagulable state led to the worsening of ulcer duration, number of episodes of recurrence, total area of ulceration, or pain from the ulcer. There was also no pattern to the location of ulceration or lipodermatosclerosis between the groups. Although, because of small numbers, one must caution against overinterpretation of the data, these observations do raise the intriguing possibility that many patients with CVU in association with apparently isolated superficial, or superficial and limited deep disease, may have in fact sustained distal macrovascular, or even microvascular, thrombosis as a result of unrecognized thrombophilia. Such disease may be below the threshold of detection for duplex ultrasonography, even venography. The existence of such unrecognized thrombosis in association with thrombophilia might explain why some patients with gross superficial reflux have pristine skin and others with lesser degrees of superficial reflux and (near) healthy deep veins have skin changes and, eventually, develop ulceration. In conclusion, thrombophilia is increasingly recognized as a major risk factor for DVT, which in turn is the major risk factor for intractable CVU. More than 40% of patients with CVU have at least one thrombophilia. The association between thrombophilia and the development of chronic venous insufficiency and ulceration needs further elucidation through larger cross-sectional and longitudinal studies. REFERENCES 1. Callam M, Ruckley C, Harper D, Dale J. Chronic ulceration of the leg: extent of the problem and provision of care. BMJ 1985;290: 1855-6. 2. Laing W. Chronic venous diseases of the leg. London: Office of Health Economics; 1992. 3. Fowkes FGR. Epidemiology of chronic venous insufficiency. Phebology 1996;11:2-5. 4. Browse NL, Burnand KG. The cause of venous ulceration. Lancet 1982;2:243-5.
JOURNAL OF VASCULAR SURGERY April 2002
5. Milne A, Ruckley C. Venous insufficiency following deep vein thrombosis. Vasc Med Rev 1994;5:241-8. 6. Bradbury AW, Evans CJ, Allan PL, Lee AJ, Ruckley CV, Fowkes FGR. The relationship between lower limb symptoms and superficial and deep venous reflux on duplex ultrasonography: the Edinburgh Vein Study. J Vasc Surg 2000;32:921-31. 7. Stuart W, Adam D, Allan P, Ruckley C, Bradbury A. Saphenous surgery does not correct perforator incompetence in the presence of deep venous reflux. J Vasc Surg 1998;28:834-8. 8. Ginsberg JS, Wells PS, Brill-Edwards P, et al. Antiphospholipid antibodies and venous thromboembolism. Blood 1995;86:3685-91. 9. Brittenden J, Bradbury A, Allan P, Prescott R, Harper D, Ruckley C. Popliteal vein reflux reduces the healing of chronic venous ulcer. Br J Surg 1998;85:60-2. 10. Burnand K, Thomas ML, O’Donnell T, Browse NL. Relationship between post-phlebitic changes in the deep veins and results of surgical treatment of venous ulcers. Lancet 1976;1:936-8. 11. Gloviczki P. Subfascial endoscopic perforator vein surgery: indications and results. Vasc Med 1999;4:173-80. 12. Munkvad S, Jorgensen M. Resistance to activated protein C: a common anticoagulant deficiency in patients with venous leg ulceration. Br J Dermatol 1996;134:296-8. 13. Ribeaudeau F, Senet P, Cayuela JM, Fund X, Paul C, et al. Br J Dermatol 1999;141:259-63. 14. Alcaraz I, Levfevre I, Wiart T, Lafon C, Forzy G, Modiano P. Leg ulcers and antiphospholipid antibodies. Prospective study of 48 cases. Annales de Dermatologie et de Venereologie 1999;126:313-6. 15. Hafner J, Felten A. Resistance to activated protein C in patients with venous leg ulcers. Dermatology 1997;195:413-4. 16. Maessen-Visch MB, Hamulyak K, Tazelaar DJ, Crombag NH, Neumann HA. The prevalence of Factor V Leiden mutation in patients with leg ulcers and venous insufficiency. Arch Dermatol 1999;135:41-4. 17. Van der Meer FJM, Koster T, Vandenbrouke JP, Briët E, Rosendaal FR. The Leiden Thrombophilia Study (LETS). Thromb Haemost 1997; 78:631-5. 18. Koster T, Rosendaal FR, de Ronde H, et al. Venous thrombosis due to poor anticoagulant response to activated Protein C: Leiden thrombophilia study. Lancet 1993;342:1503-6. 19. Rodeghiero F, Tosetto A. The epidemiology of inherited thrombophilia: the VITA Project. Vicenza Thrombophilia and Atherosclerosis Project. Thromb Haemost 1997;78:636-40. 20. Lowe G, Rumley A, Woodward M, Reid E, Rumley J. Activated protein C resistance and the FV:R506Q mutation in a random population sample. Thromb Haemost 1999;81:918-24. 21. Tait RC, Walker ID, Reitsma PH, et al. Prevalence of Protein C deficiency in the healthy population. Thromb Haemost 1995;73:87-93. 22. Dahlbäck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognised mechanism characterised by poor anticoagulant response to activated protein C: prediction of a co-factor to activated protein C. Proc Natl Acad Sci U S A 1993;90:1004-8. 23. Tosetto A, Missiaglia E, Frezzato M, Rodeghiero F. The VITA project: prothrombin G20210A mutation and venous thromboembolism in the general population. Thromb Haemost 1999;82:1395-8. 24. Pineo GF, Hull RD. Thrombophilia: disorders predisposing to venous thromboembolism. Baillieres Clin Haematol 1998;11:525-40. 25. Lane D, Mannucci P, Bauer K, et al. Inherited thrombophilia: part 1. Thromb Haemost 1996;76:651-62. 26. Lane D, Mannucci P, Bauer K, et al. Inherited thrombophilia: part 2. Thromb Haemost 1996;76:824-34.
Submitted Jul 4, 2001; accepted Oct 10, 2001.
In the United Kingdom, the life-time risk of development of chronic venous ulceration (CVU) is 1%,1 and the annual cost of treatment is estimated at approximately ,£600 million ($1 billion), which constitutes 1% to 2% of total healthcare spending. 2 Similar figures pertain in other European countries and in North America.3 Thrombophilia is increasingly recognized as a risk factor for deep venous thrombosis (DVT), which in turn is a major risk factor for chronic venous insufficiency and CVU.4 Specifically, the available literature suggests that between 25% and 75% of patients with CVU have postthrombotic disease on the basis of a history of DVT or evidence of postphlebitic changes in the deep venous system on investigation.5 However, these estimates vary widely according to the population studied and the thoroughness with which history and investigative evidence of prior DVT are sought. The association between CVU and DVT
From the Vascular Surgery Unit,a Department of Haematology,b Department of Medical Imaging,c Royal Infirmary, and the University Department of Vascular Surgery, Heartlands Hospital.d Competition of interest: nil. Reprint requests: Prof A. W. Bradbury, Lincoln House (Research Institute), Birmingham Heartlands Hospital, Bordesley Green East, Birmingham, B9 5SS, United Kingdom (e-mail: [email protected]). Copyright © 2002 by The Society for Vascular Surgery and The American Association for Vascular Surgery. 0741-5214/2002/$35.00 + 0 24/1/121749 doi:10.1067/mva.2002.121749
718
is further confused because DVT may be subclinical, patients may confuse superficial thrombophlebitis with DVT when giving a history, and duplex ultrasound scanning is less sensitive than venography in the detection of minor postthrombotic changes. On theoretic grounds, it seems plausible to hypothesize that postthrombotic CVU in the presence of thrombophilia may be more severe because the thrombosis is likely to be recurrent, more extensive, and, possibly, more resistant to both treatment and endogenous fibrinolysis and recanalization. If a clear relationship between a proportion of CVU and certain thrombophilias could be shown, new preventative and treatment opportunities might open up. The aim of this study, therefore, was to determine the prevalence of thrombophilia in CVU and to examine, for the first time, whether thrombophilia is associated with a history and or duplex scan evidence of DVT and clinical severity of chronic venous disease. METHODS Clinical and duplex scan assessment. This study was approved as part of a larger study by the local ethical committee board. Informed, signed consent was obtained from all patients before investigation. Patients seen at a hospital-based, “one-stop,” leg ulcer clinic with ulcerated limbs underwent history and physical examination and measurement of ankle-brachial pressure index. Note was made of previous major trauma, major surgery (including orthopedic), obesity (with body mass index), and use of the
JOURNAL OF VASCULAR SURGERY Volume 35, Number 4
MacKenzie et al
719
Table I. Relationship between history of deep venous thrombosis, duplex scan evidence of previous deep venous thrombosis, and deep venous reflux Duplex scan evidence of previous DVT History of DVT
Yes
Yes No
No
19 (19) 3 (3)
10 (5) 56 (26)
Numbers in brackets are those patients with deep venous reflux in at least the popliteal segment. DVT, Deep venous thrombosis.
Table II. Demographics of patients with and without abnormal thrombophilia screen results No P thrombophilia Thrombophilia value All patients No. of patients
52
36
50:50
25:75
.029†
Median age (IQR; years)
69 (56-76)
77 (65-87)
.005‡
Median age at first ulcer
54 (40-66)
66 (45-83)
.029‡
33%
42%
NS†
64
24
50:50
33:67
NS†
Median age (IQR; years)
70 (56-77)
75 (65-85)
NS‡
Median age at first ulcer (IQR; years)
56 (40-68)
62 (37-81)
NS‡
36%
38%
NS†
Male:female ratio (%)
Previous DVT Patients after exclusion* No. of patients Male:female ratio (%)
Previous DVT
*After exclusion of those patients with isolated antiphospholipid antibodies. †P value determined with 2 test. ‡P value determined with Mann-Whitney test. IQR, Interquartile range; NS, not significant; DVT, deep venous thrombosis.
contraceptive pill. Venous duplex ultrasonography (HDI 5000, Advanced Technology Laboratories, Bothell, Wash) of the ulcerated leg(s) was performed by a single observer (RKM). Nonulcerated limbs did not undergo duplex scanning because of time restrictions. The patients were examined in reverse Trendelenburg position at an angle of 45 degrees on a tilting couch. The following veins were examined: proximal and distal super- ficial femoral, above-knee and below-knee popliteal, posterior tibial, peroneal, saphenofemoral and saphenopopliteal junctions, above-knee and below-knee long saphenous, short saphenous, and gastrocnemius/soleal. Reflux was induced with a manual calf squeeze and was considered to be signifi-
Fig 1. Prevalence of thrombophilia in patients with chronic venous ulceration. AT, Antithrombin deficiency; PC, protein C deficiency; PS, protein S deficiency; APCR, activated protein C resistance; fVL, factor V Leiden; fIIL, (prothrombin 20210A) factor II Leiden; LA, lupus anticoagulant; ACLA, anticardiolipin antibody.
cant if it exceeded 0.5 seconds.6,7 Post-thrombotic features, such as deep vein obstruction, partial recanalization, vein wall irregularity, or valve cusp thickening, were noted. Deep venous reflux (DVR) was defined as significant reflux in the popliteal or superficial femoral vein. A history of DVT was taken from clinical records and from patient recollection of limb swelling, pain, or “white leg.” CVU was diagnosed on the basis of venous reflux on duplex scan results and exclusion of other causes. No patient had a history of DVT less than 1 year before the clinic appointment. Thrombophilia testing. A standard thrombophilia screen was performed that examined the congenital conditions of antithrombin, protein C, and protein S deficiency and factor V Leiden and prothrombin 20210A mutations. The acquired antiphospholipid antibodies (see subsequent) were also assayed. Venous blood was drawn into 3.8% trisodium citrate (1:9 volume) for thrombophilia investigations. The following tests were performed: antithrombin activity (Dade Behring [Deerfield, Ill], healthy range, 0.8 to 1.20 IU/mL), protein C (Chromatic Chromogenix [Mölndal, Sweden], healthy range, 0.67 to 1.38 IU/mL), protein S (healthy range, free 0.61 to 1.43 and total 0.64 to 1.54 IU/mL), activated protein C resistance (Coatest Chromogenix [Mölndal, Sweden], healthy range, >2.0), factor V Leiden and prothrombin 20210A with Multiplex allele-specific amplification polymerase chain reaction (MJ Research, Inc, Waltham, Mass), and the antiphospholipid antibodies (lupus anticoagulant; Diagen [Diagnostic Reagents Ltd, Thane, Oxon, UK], healthy range, >1.09:1) and anticardiolipin antibodies (Organon Technika [Rochville, Md], healthy range, immunoglobulin G < 13 G units/mL and immunoglobulin M < 10 M units/mL). All patients in
JOURNAL OF VASCULAR SURGERY April 2002
720 MacKenzie et al
Fig 2. Number of patients with multiple thrombophilic tendencies.
Table III. Severity of disease in patients with and without thrombophilia Median values (IQR) Total ulcer disease duration (months) Duration of this episode (months) Total area (cm2) No. of episodes Level of pain (0-3)
No thrombophilia (n = 52) 78 6 3.8 2 2
(24-234) (2-24) (0.1-9.5) (1-5) (1-3)
Thrombophilia (n = 36)
P value
48 (16-240) 10 (3-18) 2.9 (1.8-12) 2 (1-5) 2 (0-3)
NS* NS* NS* NS† NS†
*P
value determined with Mann Whitney test. value determined with 2 test. IQR, Interquartile range; NS, not significant. †P
whom an abnormality was detected (ie, less than healthy range for antithrombin, protein C, protein S, and activated protein C resistance or more than healthy levels for the antiphospholipid antibodies) were asked to return for repeat thrombophilia screen after a minimum of 3 months. Only patients with abnormalities on both occasions were considered to have a thrombophilia. Statistical methods. SPSS version 10.0.5 package (Statistical Package for Social Sciences Inc, Chicago, Ill) was used for all statistical analysis. For categoric data, 2 test or Kruskal-Wallis test were applied, and for continuous variables, the Mann-Whitney test was used. Exclusions. Six patients who were already undergoing oral anticoagulation therapy were excluded because of the known interference of warfarin with thrombophilia testing. One patient was excluded on the basis of having a thrombophilic abnormality on only one occasion. Two patients refused blood sampling. RESULTS Patients. This was a prospective study of 97 consecutive patients with CVU who attended a hospital-based leg ulcer clinic. Nine patients were excluded for the reasons shown previously, which left 88 patients in the study
group. There were 35 men with a median age of 61 years (interquartile range, 45 to 72 years) and 53 women with a median age of 76 years (interquartile range, 69 to 82 years; P = .001). Twelve men (34%) and 20 women (38%) gave a history of ipsilateral lower limb DVT. Duplex scan findings. There was only a weak relationship between a history of DVT and evidence of previous DVT on duplex scan results, although a history of DVT was strongly associated with DVR (P < .001; Table I). Thus, of the 29 patients who gave a history of DVT, only 19 (66%) had evidence of previous DVT on duplex scan results, although 24 of 29 (83%) had DVR. A further three patients had evidence of previous DVT on scan results but were unaware they had had a DVT and had no record of this in their hospital notes. Thrombophilia. Overall, 36 of 88 patients (41%) had at least one abnormality on thrombophilia test results (Fig 1; Table II). The significance of the presence of antiphospholipid antibodies alone is less certain8 than the other factors in terms of causing venous thromboembolism, and the condition is fairly common in the population, particularly among older women. If isolated antiphospholipid antibodies are excluded from the overall analysis results, there was no age or gender difference between those patients with and without
JOURNAL OF VASCULAR SURGERY Volume 35, Number 4
MacKenzie et al
Table IV. Prevalence of thrombophilia in patients with and without a history of ipsilateral deep venous thrombosis History of DVT Antithrombin deficiency Protein C deficiency Protein S deficiency Activated protein C resistance Factor V Leiden Prothrombin G20210A Lupus anticoagulant Anticardiolipin antibody
3 2 3 5 1 4 2 6
No history of DVT
(10%) (7%) (10%) (17%) (3%) (14%) (7%) (20%)
1 3 3 9 2 7 6 6
(2%) (5%) (5%) (15%) (3%) (12%) (10%) (10%)
Table VI. Patterns of reflux (>0.5 seconds) in patients with chronic venous ulceration with and without thrombophilia
P value* NS NS NS NS NS NS NS NS
721
No thrombophilia (n = 52) DVR only SVR only DVR and SVR Any DVR Any SVR
4 20 26 30 46
(8%) (38%) (50%) (58%) (88%)
Thrombophilia (n = 36) 5 11 19 24 30
(14%) (31%) (53%) (67%) (83%)
P value* NS NS NS NS NS
*P
value determined with 2 test. DVR, Deep venous reflux; SVR, superficial venous reflux; NS, not significant.
*P
value determined with 2 test. DVT, Deep venous thrombosis; NS, not significant.
Table V. Prevalence of thrombophilia in patients with and without duplex scan evidence of previous ipsilateral deep venous thrombosis Median values (IQR)
Table VII. Prevalence of specific thrombophilias in patients with chronic venous ulceration, in patients with first episode of venous thromboembolism, and in the general population Patients with CVU (n = 88)
Duplex scan No duplex scan P evidence of DVT evidence of DVT value*
Antithrombin deficiency Protein C deficiency Protein S deficiency Activated protein C resistance Factor V Leiden Prothrombin G20210A Lupus anticoagulant Anticardiolipin antibody
3 2 2 5
(14%) (9%) (9%) (23%)
1 3 4 9
(2%) (5%) (6%) (14%)
.02 NS NS NS
2 3 1 6
(9%) (14%) (5%) (27%)
1 8 7 6
(2%) (12%) (11%) (9%)
NS NS NS .03
*P
value determined with 2 test. IQR, Interquartile range; DVT, deep venous thrombosis; NS, not significant.
thrombophilia (Table II). Thrombophilia was not related to the clinical severity of the ulcer disease (ie, duration of disease, total area, number of episodes, or level of pain; Table III), previous history of DVT (Table IV), duplex scan evidence of DVT (apart from antithrombin and anticardiolipin antibodies; Table V), or the pattern of deep and superficial reflux (Table VI). No pattern to the position of ulceration or lipodermatosclerosis was identified between the groups. One patient gave a family history of DVT, but affected family members had not undergone testing for thrombophilia. There was no association between the number of thrombophilic abnormalities and the severity of disease or the pattern or severity of reflux (Fig 2). Interestingly, the patient with five abnormalities had no history of DVT and no evidence of deep venous disease. A history of trauma, major surgery (including orthopedic), cancer, obesity (body mass index >25), and use of the contraceptive pill did not increase the prevalence of DVT in patients with thrombophilia. DISCUSSION The principal finding of this study is that more than 40% of patients with CVU seen at a hospital-based leg
LA ACA Antithrombin PC PS APCR FVL Prothrombin 20210A
8 12 4 5 6 14 11 3
(9%) (14%) (5%) (6%) (7%) (16%) (13%) (4%)
Patients with first General episode of VTE population 10% 20% 1% 3% 5% 20% 18% 6%
2% 10% 0.2% 0.3% 0.2% 6% 7% 2%
First venous thromboembolism and general population prevalence figures from world literature.8,17-26 CVU, Chronic venous ulceration; VTE, venous thromboembolism; LA, lupus anticoagulant; ACA, anticardiolipin antibodies; PC, protein C; PS, protein S; APCR, activated protein C resistance; FVL, factor V Leiden.
ulcer clinic have at least one identifiable thrombophilia. This rate is two to 30 times higher than the rate in the general population but similar to that reported for patients with DVT (Table VII). Given that DVT is the single most common identifiable risk factor for the development of CVU and that patients with postthrombotic disease have a significantly worse prognosis in response to medical9 and surgical therapy,10,11 it is perhaps surprising that little information exists regarding the relationship between thrombophilia and CVU. The studies that do exist are small and often methodologically flawed and give a conflicting picture with regard to the relationship between thrombophilia and a clinical history of DVT. Munkvad and Jorgensen,12 Ribeaudeau et al,13 Alcaraz et al,14 and Hafner and Felten15 found that, in patients with CVU, there was no difference in DVT rates between those with and without thrombophilia. By contrast, Maessen-Visch et al16 found that in patients with CVU, 91% of patients with positive results for factor V Leiden gave a history of DVT as compared with 48% of those without the mutation.
722 MacKenzie et al
However, these patients did not all have documented DVT and did not undergo investigation with venography or duplex ultrasound scan. Between 25% and 75% of patients with CVU are thought to have sustained a previous DVT on the basis of history or investigation.5 However, as present data confirm, a history of DVT is unreliable with both false positive and negative results. Furthermore, duplex scan is less capable than venography, rarely performed these days in the United Kingdom, in distinguishing postthrombotic deep venous disease from deep primary valvular incompetence. This is especially so in the veins of the calf. Interestingly, in this study, there appeared to be no relationship between the presence of thrombophilia and the severity of disease, pattern of reflux, history of DVT, or duplex scan evidence of postthrombotic disease. Neither specific coagulation defects nor the presence of a hypercoagulable state led to the worsening of ulcer duration, number of episodes of recurrence, total area of ulceration, or pain from the ulcer. There was also no pattern to the location of ulceration or lipodermatosclerosis between the groups. Although, because of small numbers, one must caution against overinterpretation of the data, these observations do raise the intriguing possibility that many patients with CVU in association with apparently isolated superficial, or superficial and limited deep disease, may have in fact sustained distal macrovascular, or even microvascular, thrombosis as a result of unrecognized thrombophilia. Such disease may be below the threshold of detection for duplex ultrasonography, even venography. The existence of such unrecognized thrombosis in association with thrombophilia might explain why some patients with gross superficial reflux have pristine skin and others with lesser degrees of superficial reflux and (near) healthy deep veins have skin changes and, eventually, develop ulceration. In conclusion, thrombophilia is increasingly recognized as a major risk factor for DVT, which in turn is the major risk factor for intractable CVU. More than 40% of patients with CVU have at least one thrombophilia. The association between thrombophilia and the development of chronic venous insufficiency and ulceration needs further elucidation through larger cross-sectional and longitudinal studies. REFERENCES 1. Callam M, Ruckley C, Harper D, Dale J. Chronic ulceration of the leg: extent of the problem and provision of care. BMJ 1985;290: 1855-6. 2. Laing W. Chronic venous diseases of the leg. London: Office of Health Economics; 1992. 3. Fowkes FGR. Epidemiology of chronic venous insufficiency. Phebology 1996;11:2-5. 4. Browse NL, Burnand KG. The cause of venous ulceration. Lancet 1982;2:243-5.
JOURNAL OF VASCULAR SURGERY April 2002
5. Milne A, Ruckley C. Venous insufficiency following deep vein thrombosis. Vasc Med Rev 1994;5:241-8. 6. Bradbury AW, Evans CJ, Allan PL, Lee AJ, Ruckley CV, Fowkes FGR. The relationship between lower limb symptoms and superficial and deep venous reflux on duplex ultrasonography: the Edinburgh Vein Study. J Vasc Surg 2000;32:921-31. 7. Stuart W, Adam D, Allan P, Ruckley C, Bradbury A. Saphenous surgery does not correct perforator incompetence in the presence of deep venous reflux. J Vasc Surg 1998;28:834-8. 8. Ginsberg JS, Wells PS, Brill-Edwards P, et al. Antiphospholipid antibodies and venous thromboembolism. Blood 1995;86:3685-91. 9. Brittenden J, Bradbury A, Allan P, Prescott R, Harper D, Ruckley C. Popliteal vein reflux reduces the healing of chronic venous ulcer. Br J Surg 1998;85:60-2. 10. Burnand K, Thomas ML, O’Donnell T, Browse NL. Relationship between post-phlebitic changes in the deep veins and results of surgical treatment of venous ulcers. Lancet 1976;1:936-8. 11. Gloviczki P. Subfascial endoscopic perforator vein surgery: indications and results. Vasc Med 1999;4:173-80. 12. Munkvad S, Jorgensen M. Resistance to activated protein C: a common anticoagulant deficiency in patients with venous leg ulceration. Br J Dermatol 1996;134:296-8. 13. Ribeaudeau F, Senet P, Cayuela JM, Fund X, Paul C, et al. Br J Dermatol 1999;141:259-63. 14. Alcaraz I, Levfevre I, Wiart T, Lafon C, Forzy G, Modiano P. Leg ulcers and antiphospholipid antibodies. Prospective study of 48 cases. Annales de Dermatologie et de Venereologie 1999;126:313-6. 15. Hafner J, Felten A. Resistance to activated protein C in patients with venous leg ulcers. Dermatology 1997;195:413-4. 16. Maessen-Visch MB, Hamulyak K, Tazelaar DJ, Crombag NH, Neumann HA. The prevalence of Factor V Leiden mutation in patients with leg ulcers and venous insufficiency. Arch Dermatol 1999;135:41-4. 17. Van der Meer FJM, Koster T, Vandenbrouke JP, Briët E, Rosendaal FR. The Leiden Thrombophilia Study (LETS). Thromb Haemost 1997; 78:631-5. 18. Koster T, Rosendaal FR, de Ronde H, et al. Venous thrombosis due to poor anticoagulant response to activated Protein C: Leiden thrombophilia study. Lancet 1993;342:1503-6. 19. Rodeghiero F, Tosetto A. The epidemiology of inherited thrombophilia: the VITA Project. Vicenza Thrombophilia and Atherosclerosis Project. Thromb Haemost 1997;78:636-40. 20. Lowe G, Rumley A, Woodward M, Reid E, Rumley J. Activated protein C resistance and the FV:R506Q mutation in a random population sample. Thromb Haemost 1999;81:918-24. 21. Tait RC, Walker ID, Reitsma PH, et al. Prevalence of Protein C deficiency in the healthy population. Thromb Haemost 1995;73:87-93. 22. Dahlbäck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognised mechanism characterised by poor anticoagulant response to activated protein C: prediction of a co-factor to activated protein C. Proc Natl Acad Sci U S A 1993;90:1004-8. 23. Tosetto A, Missiaglia E, Frezzato M, Rodeghiero F. The VITA project: prothrombin G20210A mutation and venous thromboembolism in the general population. Thromb Haemost 1999;82:1395-8. 24. Pineo GF, Hull RD. Thrombophilia: disorders predisposing to venous thromboembolism. Baillieres Clin Haematol 1998;11:525-40. 25. Lane D, Mannucci P, Bauer K, et al. Inherited thrombophilia: part 1. Thromb Haemost 1996;76:651-62. 26. Lane D, Mannucci P, Bauer K, et al. Inherited thrombophilia: part 2. Thromb Haemost 1996;76:824-34.
Submitted Jul 4, 2001; accepted Oct 10, 2001.