April 21, 2008

Management of Chronic Venous Disorders of the Lower Limbs Guidelines According to Scientific Evidence

Disclaimer Due to the evolving field of medicine, new research may, in due course, modify the recommendations presented in this document. At the time of publication, every attempt has been made to ensure that the information provided is up to date and accurate. It is the responsibility of the treating physician to determine the best treatment for the patient. The authors, committee members, editors, and publishers cannot be held responsible for any legal issues that may arise from the citation of this statement.

Rules of evidence

Management of patients with chronic venous disorders has been traditionally undertaken subjectively among physicians, often resulting in less than optimal strategies. In this document, a systematic approach has been developed with recommendations based upon cumulative evidence from the literature. Levels of evidence and grades of recommendation range from Level I and Grade A to Level III and Grade C. Level I evidence and Grade A recommendations derive from scientifically sound randomized clinical trials in which the results are clear-cut. Level II evidence and Grade B recommendations derive from clinical studies in which the results among trials often point to inconsistencies. Level IH evidence and Grade C recommendations result from poorly designed trials or from small case series.1'2


Meta-analyses are included in the present document but there should be caution as to their possible abuse. Certain studies may be included in a meta-analysis carelessly without sufficiently understanding of substantive issues, ignoring relevant variables, using heterogenous findings or interpreting results with a bias.3 It has been demonstrated that the outcomes of 12 large randomized controlled trials were not predicted accurately 35% of the time by the meta-analyses published previously on the same topics.4




Chronic venous disease (CVD) of the lower limbs is often characterized by symptoms and signs as a result of structural or functional abnormalities of the veins. Symptoms include aching, heaviness, leg-tiredness, cramps, itching, burning sensations, swelling and the restless leg syndrome, as well as cosmetic dissatisfaction. Signs include telangiectasias, reticular and varicose veins, edema, and skin changes such as pigmentation, lipodermatosclerosis, dermatitis and ultimately ulceration.5,6

CVD is usually caused by primary abnormalities of the venous wall and valves and/or secondary abnormalities resulting from previous deep venous thrombosis (DVT) that can lead to reflux, obstruction or both. Rarely, congenital malformations lead to CVD.7

The clinical history and examination do not always indicate the nature and extent of underlying abnormalities. Consequently, several diagnostic techniques have been developed to define the anatomic extent and functional severity of obstruction and/or reflux, as well as calf muscle pump dysfunction. Difficulties in deciding which investigations to use and how to interpret the results has previously stimulated a consensus statement on investigations for CVD.8 The current document aims to provide an account of current concepts of CVD and guidelines for management.


Changes in superficial and deep veins

Varicose veins are a common manifestation of CVD and are believed to result from abnormal distensibility of connective tissue in the vein wall. Veins from patients with varicosities have different elastic properties than those from individuals without varicose veins.9,10

Primary varicose veins result from venous dilatation and valve damage without previous DVT. secondary varicose veins are the consequence of DVT or, less commonly, superficial thrombophlebitis. Recanalization may give rise to relative obstruction and reflux in deep, superficial and perforating veins.6 Approximately 30% of patients with deep venous reflux shown by imaging appear to have primary valvular incompetence rather than detectable post- thrombotic damage.11-13 Rarely, deep venous reflux is due to agenesis or aplasia. Varicose veins may also be caused by pelvic vein reflux in the absence of incompetence at the saphenofemoral junction, thigh or calf perforators. Retrograde reflux in ovarian, pelvic, vulval, pudental or gluteal veins may be also associated with clinical symptoms and signs of pelvic congestion.14-17

Following DVT, spontaneous lysis over days or weeks and recanalization over months or years can be observed in 50% to 80% of patients.18-20 Rapid thrombus resolution after DVT is associated with a higher incidence of valve competency.18,21 Such resolution depends on thrombus extent and location.22 Inadequate recanalization following DVT can lead to outflow obstruction. Less frequently, obstruction results from extramural venous compression (most commonly left common iliac vein compression by the right common iliac artery), intra-luminal changes,23-27 or rarely from congenital agenesis or hypoplasia.28

Most post-thrombotic symptoms result from venous hypertension due to valvular incompetence and/or outflow obstruction. Venous hypertension increases transmural pressure in post-capillary vessels leading to skin capillary damage, lipodermatosclerosis and, ultimately, ulceration.29

The reported prevalence of post-thrombotic syndrome following DVT has been variable (35% to 69% at 3 years and 49% to 100% at 5 to 10 years) and depends on the extent and location of thrombosis as well as treatment.30-40 Patients with both chronic obstruction and reflux have the highest incidence of skin changes or ulceration.40 The risk of ipsilateral post-thrombotic syndrome is higher in patients with recurrent thrombosis and is often associated with congenital or acquired thrombophilia.41-44 In recent studies, skin changes or ulceration have been less frequent (4% to 8% in 5 years) in patients with proximal thrombosis treated with adequate anticoagulation, early mobilisation, and long-term elastic compression.45

Incompetent perforating veins

Incompetent perforating veins (IPV) can be defined as those that penetrate the deep fascia and permit deep to superficial flow. The flow in IPV is often bidirectional. It is outward during muscular contraction and inward during relaxation. In the majority of patients with primary uncomplicated varicose veins the net flow is inward from superficial to deep. However, in the presence of severe damage to deep veins especially with persisting deep vein obstruction, the flow is predominantly outward.46,47

IPVs can result from superficial and/or deep venous reflux but are rarely found in isolation.4850 The prevalence of IPVs, their diameter, volume flow and velocity increase with clinical severity of CVD whether or not there is co-existing deep venous incompetence.47,51-56 Up to 10% of patients, often women, presenting with clinical CEAP 1 to 3 disease have non-saphenous superficial reflux in association with unusually placed IPVs.57

Molecular mechanisms affecting the venous wall

As mentioned above, varicose veins have different elastic properties to normal veins.9,10 The ratio between collagen I and collagen III is altered as are dermal fibroblasts from the same patients suggesting a systemic disorder with a genetic basis.58

Leukocyte activation, adhesion and migration through the endothelium as a result of altered shear stress S9-61 contribute to the inflammation and subsequent remodeling of the venous wall and valves.7, 62-64 Reduction in shear stress also stimulates production of tumor growth factor-beta1 (TGF-beta1) by activated endothelial cells and smooth muscle cells (SMCs) inducing SMC migration into the intima and subsequent proliferation. Fibroblasts proliferate and synthesize matrix metaloproteinases (MMPs) overcoming the effect of tissue inhibitors of metaloproteinases (TIMPs). The MMP/TIMP imbalance results in degradation of elastin and collagen.60,65 This may contribute to hypertrophie and atrophie venous segments and valve destruction as observed in varicose veins.60, 65,66 Remodelling of the venous wall and abnormal venous distension prevents valve leaflets from closing properly resulting in reflux.

Changes in microcirculation as a result of venous hypertension

Techniques such as laser Doppler,67,68 measurements of transcutaneous PO^sub 2^,69 interstitial pressure capillaroscopy,70 microlymphography71 and skin biopsy 72,73 have provided the means to study the extent of changes in skin microcirculation of limbs with CVD.

In patients with venous hypertension, capillaries become markedly dilated, elongated, and tortuous, especially at skin sites with hyperpigmentation and lipodermatosclerosis. These changes are associated with a high overall microvascular blood flow66,74 in the dermis and a decreased flow in nutritional capillaries.75,76 A striking feature in the skin of patients with venous hypertension is a "halo" formation around dilated capillaries observed on capillaroscopy. This is associated with microedema, pericapillary fibrin 77 and other proteins that possibly prevent normal nutrition of skin cells predisposing to ulceration. Microlymphangiopathy78,79 and outward migration of leucocytes exacerbate microedema and inflammation.80-84 As a late phenomenon, capillary thromboses successively lead to reduction in nutritional skin capillaries and transcutaneous PO2.70-85 Pathophysiology of stasis dermatitis and dermal fibrosis

Mechanisms modulating leukocyte activation, fibroblast function and dermal extracellular matrix alterations have been the focus of investigation in the 1990s. As stated above, CVD is caused by persistent venous hypertension leading to chronic inflammation. It is hypothesized that the primary injury is extravasation of macromolecules (i.e. fibrinogen and alpha^sub 2^-macroglobulin) and red blood cells into the dermal interstitium.73,86-88 Red blood cell degradation products and interstitial protein extravasation are potent chemoattractants that represent the initial underlying chronic inflammatory signal responsible for leukocyte recruitment. These cytochemical events are responsible for increased expression of intercellular adhesion molecule-1 (ICAM-I) on endothelial cells of microcirculatory exchange vessels observed in CVI dermal biopsies.89,90ICAM-1 is the activation dependent adhesion molecule utilized by macrophages, lymphocytes and mast cells for diapedesis.

Cytokine regulation and tissue fibrosis

As indicated above, CVD is characterized by leukocyte recruitment, tissue remodeling and dermal fibrosis. These physiologic processes are prototypical of disease states regulated by TGF-beta1. TGF-beta1 is present in pathologic quantities in the dermis of patients with CVD and increases with disease severity.91 TGF-beta1 is secreted by interstitial leukocytes and becomes bound to dermal fibroblasts and extracellular matrix proteins. Platelet- derived growth factor receptor alpha and beta (PDGFR-alpha and PDGFR- beta) and vascular endothelial growth factor (VEGF) have also been identified in the dermis of CVD patients.92 It has been postulated that these molecules regulate leukocyte recruitment, capillary proliferation and interstitial edema in CVD by upregulation of adhesion molecules leading to leukocyte recruitment, diapidesis and release of chemical mediators.91

Dermal fibroblast function

Aberrant phenotypic behavior has been observed in fibroblasts isolated from venous ulcer edges when compared to fibroblasts obtained from ipsilateral thigh biopsies of normal skin in the same patients.93 Collagen production by fibroblasts is increased by 60% in a dose-dependent manner in control skin whereas venous ulcer fibroblasts are unresponsive. Unresponsiveness in ulcer fibroblasts is associated with a fourfold decrease in TGF-beta1 type II receptors.93 This is associated with decrease in phosphorylation of TGF-beta1 receptor substrates SMAD 2 and 3 as well as p42/44 mitogen activated protein kinases,94 and decrease in collagen and fibronectin production from venous ulcer fibroblasts when compared to normal controls.95

Venous ulcer fibroblast growth rates become markedly suppressed when stimulated with bFGF, EGF and IL-I96 and this growth inhibition can be reversed with bFGF.97 The proliferative response of CVI fibroblasts to TGF-beta1 decreases with increased disease severity,98 and phenotypically, venous ulcer fibroblasts appear to become morphologically similar to fibroblasts undergoing cellular senescence.

Role of matrix Metalloproteinases (MMPs) and their inhibitors in CVD

The signaling event responsible for development of a venous ulcer and the mechanisms responsible for slow healing are poorly understood. Wound healing is an orderly process that involves inflammation, re-epithelialization, matrix deposition and tissue remodeling. Matrix deposition and tissue remodeling are processes controlled by matrix metalloproteinases (MMPs) and tissue inhibitors of matrix metalloproteinases (TIMPs). In general, MMPs and TIMPs are induced temporarily in response to exogenous signals such as various proteases, cytokines or growth factors, cell-matrix interactions and altered cell-cell contacts. Gelatinases MMP-2 and MMP-9 as well as TIMP-I appear to be increased in exudates from venous ulcers compared to acute wounds."-101 However, analyses of biopsy specimens have demonstrated variable results. Herouy et al, reported that MMP1, 2 and TIMP-I are increased in patients with lipodermatosclerosis compared to normal skin.102 In a subsequent investigation, biopsies from venous ulcer patients were found to have increased levels of the active form of MMP-2 compared to normal skin.103 In addition, increased immunoreactivity to extracellular inducer of MMP (EMMPRIN), membrane Type 1 and 2 metaloproteinases (MTl-MMP and MT2-MMP) were detected in the dermis and perivascular regions of venous ulcers.104 Saito et al. were unable to identify differences in overall MMP-I, 2, 9 and TIMP-I protein levels or activity in CVD patients with clinical CEAP class 2 through 6 disease compared to normal controls.105 However, within a clinical class, MMP-2 levels were elevated compared to MMP-1,9 and TIMP-I in patients with CEAP class 4 and 5 disease. These data indicate that active tissue remodeling is occurring in patients with CVD. Which matrix metalloproteinases are involved and how they are activated and regulated is currently unclear. It appears that MMP-2 may be activated by urokinase plasminogen activator (uPA). Herouy et al. observed increased uPA and uPAR mRNA and protein levels in venous ulcers compared to normal skin.106 Elevated levels of active TGFbeta- 1 in the dermis of CVI patients suggest a regulatory role for TGF1- I in MMP and TIMP synthesis and activity but this, needs to be verified by further studies.

Magnitude of the problem

Early epidemiological studies have shown that CVD has a considerable socio-economic impact in western countries due to its high prevalence, cost of investigations and treatment, and loss of working days.107,108 Varicose veins are present in 25-33% of female and 10-20% of male adults.109-119 In the Framingham study, the incidence of varicose veins per year was 2.6% in women and 1.9% in men,120 The prevalence of edema and skin changes such as hyperpigmentation and eczema due to CVD varies from 3.0% i09 to 11%111 of the population.

Venous ulcers occur in about 0.3% of the adult population in western countries.112, 1200 -128 The prevalence of active and healed ulcers combined is about 1%.129,130 Healing of venous ulcers may be delayed in patients of low social class and those who are single.131 Data from the Brazilian security System show that CVD is the 14th most-frequently quoted disease for temporary work absenteeism and the 32nd most frequent cause of permanent disability and public financial assistance.132

Some older studies were based on clinical assessment or questionnaires only. Different definitions of venous disease, were used and populations selected contained different age groups and other non-representative factors so that it was difficult to compare epidemiological data. Introduction of the CEAP classification in the mid 1990s and improved diagnosic techniques have allowed studies to become more comparable.

Thus, in recent studies from France,133 Germany 134 and Poland135 the CEAP classification (see below) has been used to differentiate between the different classes of CVD even although selection criteria remain different. The prevalence in the French, German and Polish studies are shown in Table I.

Socioeconomic aspects

The considerable socioeconomic impact of CVD is due to the large numbers concerned, cost of investigations and management and morbidity, and suffering it produces which are reflected in a deterioration in quality of life and loss of working days. The problem is compounded by the fact that CVD is progressive and has a propensity to recur.

Measures to reduce the magnitude of the problem include awareness of the problem, early diagnosis and care, careful consideration of the necessity and choice of investigations, discipline in the choice of management based on clinical effectiveness and cost. These requirements imply specific training in all aspects of this condition.


Direct costs are associated with medical, nursing and ancillary manpower together with costs for investigations and treatment whether in hospital or as an out-patient. Indirect costs relate to loss of working days. The cost in human terms must also be considered and this can be quantified by assessment of quality of life. Manpower costs alone are important: 22% of district nurses' time is spent treating ulcers of the legs.136 Estimations of the overall annual costs of CVD vary from 600 to 900 million euro* (US$720 million-1 billion) in Western European countries 137-139 representing 1-2% of the total health care budget, to 2.5 billion euro (US$3 billion) in the USA.140 Often, the costs for treatment include reimbursements by the State and are affected by government policies.141

Detailed figures for France in 1991 142 showed a total expenditure for CVD of 2.24 billion euro (US$2.7 billion) of which 41% was for drugs, 34% for hospital care and 13% for medical fees. There were 200,000 hospitalizations for CVD during that year of which 50% were for varicose veins which was the 8th most common cause for hospitalization. These costs represented 2.6% of the total health budget for that year. A prospective study from France has broken down the cost for treating venous ulceration and of the total cost, 48% was for care, 33% for medication, 16% for hospitalization and 3% for loss of work.143

Similarly high costs have been found in Germany 144 which have increased by 103% between 1980 and 1990 to reach about 1 billion euro (US$1.2 billion) with in-patient direct costs of 250 million euro (US$300 million), out patient costs of 234 million euro (US$280 million) and drug costs of 207 million euro (US$248 million).

In Belgium, medical care costs for CVD in 1995 amounted to 250 million euro (US$300 million) which is 2% to 2.25% of total health care budget.145 In Sweden, the average weekly cost for treating venous leg ulcers in 2002 was 101 euro (US$121) with an estimated annual cost of 73 million euro (US$88 million) 146 and these costs were slightly less than in previous years which was attributed to a more structured management program.

In the USA, a cost estimate of long-term complications for deep vein thrombosis (DVT) after total hip replacement gave figures varying from 700 euro to 3180 euro (US$839 to 3817) per patient in the first year and 284 euro to 1400 euro (US$341 to 1677) in subsequent years depending on the severity of the postthrombotic syndrome.147 The cost of a pulmonary embolus (PE) was 5500 euro (US$6604).

Many of the above costs are based on estimations and assumptions and strict comparisons are difficult as there is no agreed definition of "costs". Furthermore, the figures need to be related to the country's population or to Gross National Product. However, they do illustrate the considerable cost of venous diseases.

Phlebotropic drugs that are prescribed as an alternative to elastic stockings essentially for relief of leg heaviness, pain and edema 14S in women who are either standing or sitting for long periods at work result in considerable expenditure. This cost amounts to 63.2 million euro (US$76 million) in Spain, 25 million euro (US$30 million) in Belgium and 457 million euro (US$548 million) in France,145, 149 representing 3.8% of the sales of refundable medicines. Two very similar surveys in Germany 150 and France 151,152 showed that nearly 50% of the population aged over 15 years reported leg vein problems of whom 90.3% purchased a phlebotropic drug: 71% were women of whom 30% were "obese, relatively underpriviledged in terms of age, occupational status, hours of work, working conditions, leisure, income and health".

Indirect costs of venous disease in terms of working days lost is quoted as "the most important cost factor" in 1990 in Germany, amounting to 270 million euro (US$324 million).144 In the USA, venous ulcers cause loss of 2 million work-days per year.140 In France, 6.4 million days of work were lost in 1991.142 Another study in France found that about 7% of the working population is off work because of venous disease (CEAP: C1-C6) with an overall "estimation" of 4 million working days lost in a year at an estimated cost of 320 million euro (US$384 million) to the economy.148 153 These costs are higher than the amount spent for the treatment of arterial disease.

Quality of life

Good Quality of Life (QOL) has been defined by the World Health Organization (WHO) as "a state of complete physical, mental and social well-being".154 QOL reflects the patient's perception of "well-being" at any time. Thus, it is an important element in the general assessment of any patient. Illness has repercussions on QOL. In this way, a measure of QOL is also a measure of the "cost" of any disease in terms of human suffering. It also considerably helps to assess a patient's perception of the result of any treatment.

Various quantitative instruments in the form of questionnaires, both generic and specific for venous disease, have been developed and some have been validated.108, 154-157 They show conclusively that QOL is adversely affected by venous disease.108 148, 154-160 Similarly, reduction in severity of disease, for example after treatment, is reflected in the QOL.154, 158, 160-162 There is a significant association between QOL and severity of venous disease and also with the CEAP classification.154, 158, 161-165 a recent study also shows an association in women between venous disease and working conditions which is reflected in the QOL.148 In conclusion, CVD is very costly both economically and in terms of human suffering. However, prevention of the condition and cost-effective management should lead to a reduction in costs.

Cost-effectiveness of prevention and treatment

The need to contain the increasing cost of CVD is evident. The methods used, whether aimed at prevention or treatment must essentially be shown to be effective but must also take into consideration the cost in relation to the proven effectiveness.

The two main and costly manifestations of CVD are varicose veins with or without skin changes and venous ulceration. At the present time, there is no way to effectively prevent the onset of varicose veins. However, there are known risk factors, some of which are proven (e.g. obesity), and many are not (heredity, gender, pregnancies, age). Much work has been done to prevent CVD developing in patients with early varicose veins or following venous thrombosis and all measures that contribute to preventing a venous ulcer will have a strong impact on the human and socioeconomic costs.

There is a growing awareness of the need to demonstrate cost- effectiveness in many aspects of the management of CVD and this is shown by the volume of publications on this subject. Costeffectiveness in CVD takes into consideration the progressive nature of the symptoms and their tendency to recur and this implies continuous follow-up. In the case of venous ulcers, assessment of the recurrence rate is as important as the healing rate. However, at present there is a paucity of evidence-based studies of the most cost-effective way to manage primary varicose veins.

Selection of the most appropriate investigation has been established.8 Initial outlay for duplex ultrasound has a cost but this is justified by its cost-effectiveness.166,167

Hospital admissions are costly; for example, treatment of a venous ulcer costs 24 times more in hospital than at home.168 Realization of this fact has led to more management outside hospital whenever possible and has opened new fields such as day surgery for varicose veins and home treatment of DVT in suitable cases. Prevention and management of venous thrombosis outside hospital has been shown to be not only as clinically effective as in hospital but also more costeffective.169 It has also been shown that treatment of venous ulcers in dedicated centers with a set protocol of treatment is very cost-effective and gives faster healing times than treatment in nondedicated centers without a set protocol. 140, 168, 170, 171 The most cost-effective method to manage venous ulcers is by simple dressings and multilayered bandaging to provide good pressure.140, 172-185 a recent study 185 concluded that for long-term management of venous ulcers, education of the patient and good compression with effective compliance would save 5270 euro (US$6326) in medical costs per patient per whole life together with a further saving of 14228 euro (US$17080) due to fewer working days lost. A further study 173 demonstrated that high compression hosiery was more cost-effective than moderate compression for preventing ulcer recurrence and was particularly cost-saving if combined with patient education.186

There is now evidence for cost-effectiveness of phlebotropic drugs when used as adjuvant therapy to increase the rate of healing of venous ulcers.187-188

Many women suffering from CVD have found that their symptoms were made worse by their working conditions resulting in many days off work. It has been suggested that simple changes in working conditions such as providing high stools, adequate rest periods and medical counseling could be very cost-effective.148,151,152

The CEAP classification of chronic venous disorders (CVD)

The CEAP classification was published in the mid 1990s in 25 journals and books in 8 languages (Table II). Several revisions by the ad hoc committee of the American Venous Forum in conjunction with the International ad hoc committee have resulted in the classification summarized below that has been adopted worldwide to facilitate meaningful communication about and description of all forms of CVD. The term CVD includes all morphological and functional abnormalities of the venous system in the lower limb. Some of these like telangectasia are highly prevalent in the adult population and in many cases the use of the term 'disease' is, therefore, inappropriate. The term chronic venous insufficiency (CVI) is entrenched in the literature and has been used to imply a functional abnormality (reflux) of the venous system and is usually reserved for patients with more advanced disease including those with edema (C3), skin changes (C4) or venous ulcers (C5/6). In the revised CEAP classification 189 the previous overall structure of CEAP has been maintained but more precise definitions have been added. The following recommended definitions apply to the clinical C classes in CEAP.

Telangiectasia: a confluence of dilated intradermal venules of less than 1 mm in caliber. Synonyms include spider veins, hyphen webs, and thread veins.

Reticular veins: dilated bluish subdermal veins usually from 1 mm in diameter to less than 3 mm in diameter. They are usually tortuous. This excludes normal visible veins in people with transparent skin. Synonyms include blue veins, subdermal varices, and venulectasies.

Varicose veins: subcutaneous dilated veins equal to or more than 3 mm in diameter in the upright position. These may involve saphenous veins, saphenous tributaries, or non-saphenous veins. Varicose veins are usually tortuous, but refluxing tubular saphenous veins may be classified as varicose veins. Synonyms include varix, varices, and varicosities.

Corona phlebectatica: this term describes a fanshaped pattern of numerous small intradermal veins on the medial or lateral aspects of the ankle and foot. This is commonly thought to be an early sign of advanced venous disease. Synonyms include malleolar flare and ankle flare.

Edema: this is defined as a perceptible increase in volume of fluid in the skin and subcutaneous tissue characterized by indentation with pressure. Venous edema usually occurs in the ankle region, but it may extend to the leg and foot.

Pigmentation: brownish darkening of the skin initiated by extravasated blood, which usually occurs in the ankle region but may extend to the leg and foot. Eczema: erythematous dermatitis, which may progress to a blistering, weeping, or scaling eruption of the skin of the leg. It is often located near varicose veins but may be located anywhere in the leg. Eczema is usually caused by CVD or by sensitization to local therapy.

Lipodermatosclerosis (LDS): localized chronic inflammation and fibrosis of the skin and subcutaneous tissues sometimes associated with scarring or contracture of the Achilles tendon. LDS is sometimes preceded by diffuse inflammatory edema of the skin which may be painful and which is often referred to as hypodermitis. This condition needs to be distinguished from lymphangitis, erysipelas or cellulitis by their characteristic local signs and systemic features. LDS is a sign of severe chronic venous disease.

Atrophie blanche or white atrophy, localized, often circular whitish and atrophie skin areas surrounded by dilated capillary spots and sometimes with hyperpigmentation. This is a sign of severe chronic venous disease. Scars of healed ulceration are excluded from this definition.

Venous ulcer; full thickness defect of the skin most frequently at the ankle that fails to heal spontaneously sustained by CVD.

Revised CEAP 189


C0: no visible or palpable signs of venous disease.

C1 : telangiectasies or reticular veins.

C2: varicose veins.

C3: edema.

C4a: pigmentation and/or eczema.

C4b: lipodermatosclerosis and/or atrophie blanche.

C5: healed venous ulcer.

C6: active venous ulcer.

S: symptoms including ache, pain, tightness, skin irritation, heaviness, muscle cramps, as well as other complaints attributable to venous dysfunction.

A: asymptomatic.


Ec: congenital.

Ep: primary.

Es: secondary (post-thrombotic).

En: no venous etiology identified.


As: superficial veins.

Ap: perforator veins.

Ad: deep veins.

An: no venous location identified.


Basic CEAP:

__ Pr: reflux.

__ Po: obstruction.

__ Pr,o: reflux and obstruction.

__ Pn: no venous pathophysiology identifiable.

Advanced CEAP:

__ Same as Basic with the addition that any of 18 named venous segments can be utilized as locators for venous pathology.

Superficial veins:

1. Telangiectasies/reticular veins.

2. Great saphenous vein (GSV) above knee.

3. GSV below knee.

4. Small saphenous vein.

5. Non-saphenous veins.

Deep veins:

6. Inferior vena cava.

7. Common iliac vein.

8. Internal iliac vein.

9. External iliac vein.

10. Pelvic: gonadal, broad ligament veins, other.

11. Common femoral vein.

12. Deep femoral vein.

13. Femoral vein.

14. Popliteal vein.

15. Crural: anterior tibial, posterior tibial, peroneal veins (all paired).

16. Muscular: gastrocnemial, soleal veins, other Perforating veins:

17. Thigh

18. Calf

Date of classification

CEAP is not a static classification, and the patient can be reclassified at any point in time. Therefore, the classification should be followed by the date.

Level of investigation

A Roman numeral (e.g. LII) describes the level (L) of intensity of investigation (see below) and will be discussed in the next section.


A patient presents with painful swelling of the leg and varicose veins, lipodermatosclerosis and active ulceration. Duplex scanning on May 17, 2004 showed axial reflux of GSV above and below the knee, incompetent calf perforators and axial reflux in the femoral and popliteal veins. No signs of post-thrombotic obstruction.

Classification according to basic CEAP: C6, S, Ep, As,p,d, Pr (2004-05-17, LII)

Classification according to advanced CEAP: C2,3,4b,6,S, Ep, As,p,d, Pr2,3,18,13,14 (2004-05-17, LII).

Basic and advanced CEAP

Basic CEAP includes all four components. Use of the C- classification alone inadequately describes CVD. The majority of patients have a duplex scan that provides data on E, A, and P. The highest descriptor is used for clinical class. Advanced CEAP is for the researcher and for reporting standards. This is a more detailed and precise classification where the extent of disease can be allocated to one or more 18 named venous segments.


General remarks

There is no single test that can provide all information needed to make clinical decisions and plan a management strategy. Understanding the pathophysiology is the key to selecting the appropriate investigations.

When a patient presents with symptoms and signs suggestive of CVD, a physician should ask a number of clinically relevant questions. The first question is to ask whether CVD is present. If it is then investigations should follow that determine the presence or absence of reflux, obstruction, calf muscle pump dysfunction and the severity of each.8

Detection of reflux and obstruction

The clinical presentation is assessed with the history and physical examination which may include an initial evaluation with a 'pocket' Doppler or duplex scan. Such an evaluation helps to identify the presence and sites of reflux and potential occlusion of proximal veins. A proportion of patients may require additional investigation (see below).

Duplex scanning

Duplex ultrasound is superior to phlebography and is considered to be the method of choice to detect reflux in any venous segment.56, 190-197 Imaging is usually performed with colour flow scanners using high frequency probes for superficial veins and lower frequency probes when deep penetration is required. The entire superficial and deep venous systems as well as the communicating and perforating veins are examined. Elements of the examination that are often germane to further management include:

1. standing position for the femoral and great saphenous veins or sitting position for popliteal and calf veins;

2. measurement of the duration of reflux;

3. size of perforators;

4. diameter of saphenous veins;

5. size and competence of major saphenous tributaries.


Quantification of venous obstruction is difficult. Traditional methods that measure arm-foot pressure differential,198 outflow fraction 199,200 and outflow resistance by plethysmography8 express functional obstruction but do not quantify local anatomic obstruction. Intravascular ultrasound (IVUS) and direct pressure measurements demonstrate relative degrees of obstruction at the involved venous segment more reliably, but they are not useful for infra-inguinal obstruction.

Investigation of patients in different CEAP clinical classes

A precise diagnosis is the basis for correct classification of the venous problem. A way to organize the diagnostic evaluation of the patient with CVD is to utilize one or more of three levels of testing, depending on the severity of the disease:

Level I: The office visit with history and clinical examination, which may include use of a 'pocket' Doppler or a color flow duplex.

Level II: The non-invasive vascular laboratory with mandatory duplex colour flow scanning, with or without plethysmography.

Level III: The addition of invasive investigations or complex imaging studies including ascending and descending phlebography, varicography, venous pressure measurements, CT scan, venous helical scan, MRI or IVUS.

A simple guide to the level of investigation in relation to CEAP clinical classes is given below. This may be modified according to clinical circumstances and local practice.


Level I investigations are usually sufficient. However, symptoms such as ache, pain, heaviness, legtiredness and muscle cramps in the absence of visible or palpable varicose veins are an indication for duplex scanning to exclude reflux which often precedes the clinical manifestation of varices.


Level II (duplex scanning) should be used in the majority of patients and is mandatory in those being considered for intervention. Level III may be needed in certain cases.


Level II investigations are utilized to determine whether or not reflux or obstruction in the deep veins is responsible for the edema. If obstruction is demonstrated or suspected as a result of duplex scanning, level III studies to investigate the deep venous system should be considered. Lymphoscintigraphy may be indicated to confirm the diagnosis of lymphedema in certain patients.


Level II investigations will be required in virtually all patients. Selected cases, such as those being considered for deep venous intervention, will proceed to level III. Level I investigations may be sufficient in some patients with irreversible muscle pump dysfunction due to neurological disease, severe and non- correctable reduction of ankle movement or where there is a contraindication to surgical intervention. Some investigations may have to be deferred, particularly in patients with painful ulcers.



Compression therapy

Therapy that applies pressure to the lower extremities is a fundamental component for managing CVD.


Long stretch bandages extend by more than 100% of their original length, short-stretch bandages extend to less than 100% and stiff bandages such as zinc plaster bandages (Unna's boot) and Velcro devices do not extend at all.201

Medical compression hosiery and classes

Medical compression stockings are made of elasticated textile. According to their length, they are classified as knee-length, thigh- length and tights (panty style). They may be custom-made or off the shelf and are available in standard sizes.

Different compression classes are available according to the pressure exerted. The pressure profile for each compression class varies among different countries and is measured by various non- standardized methods. The European Prestandard on medical compression hosiery proposed by the Commite Europeen de Normalisation (CEN) provides five compression classes as shown in Table III.202 Measurement of interface pressure and stiffness in vivo 203

There is a need to standardize measurements of interface pressures and fabric stiffness in vivo to allow comparison between different compression systems, both for clinical practice and research. Fabric stiffness is determined by the increase of interface pressure per centimetre increase of the leg circumference due to muscular contraction during walking or standing.202 For equal resting pressures, the peak pressure and bandwidth of pressure change at the ankle is much higher with short stretch material. Addition of several layers of compression bandages and superimposition of stockings increase both the interface pressure and stiffness of the cumulative compression.

Practical use of bandages

There are no definitive data on the superiority of different bandaging techniques (spiral, figure of eight, circular etc.). However, an important feature of a good compression bandage is that it develops a sufficiently high pressure peak during walking to enable intermittent compression of the veins while allowing a tolerable resting pressure. Bandages should maintain their nominal pressures during application for several days and nights. They should be washable and reusable.

Multilayer bandages better meet the above requirements than single layer bandages.

Pads or rolls of different materials can be used to increase the local pressure over a treated venous segment following sclerotherapy or over a venous ulcer situated behind the medial malleolus.

Practical use of compression stockings

Stockings should only be prescribed if patients are able to apply them on a regular basis. Different devices have been developed to facilitate application of stockings. They are best put on in the morning.204 New stockings should be prescribed after 3-6 months if used daily.

Intermittent pneumatic compression devices (IPC)

IPC devices consist of single or preferably multiple inelastic cuffs that are intermittently and/or sequentially inflated. Limited data based on randomized controlled studies are currently available demonstrating encouraging clinical outcome when IPC is used as part of the care for venous ulcers.205

Quality of life and compliance

Several studies have shown improvement in quality of life with compression treatment.160,162, 206,207 Compliance is crucial to prevent ulcer recurrence.45,208-211 Regular daily use of compression stockings for at least two years after DVT can reduce the incidence and severerity of the postthrombotic syndrome.45,210

Mode of action

The beneficial effects of compression treatment and methods used to measure these effects are summarized in Table IV.

Clinical applications

A summary of evidence-based indications for compression therapy is listed in Table V.

Grade A recommendations for the use of compression therapy are available for management of venous ulceration and prevention of the postthrombotic syndrome. Application of continuous compression may be contraindicated in patients with advanced peripheral arterial disease or severe sensory impairment. Grade B and C recommendations apply to other frequent indications for compression treatment such as venous edema and lymphedema (Table VI).




Venoactive drugs (VADs) are a heterogenic group of drugs from vegetal or synthetic origin (Table VII).291, 292

Numerous randomized controlled double blind studies have demonstrated the anti-edematous effect and effective attenuation of symptoms of CVD such as heavy legs, pain and restless legs by VADs so that they have become an established component of the therapeutic armamentarium for all stages of disease. VADs may accentuate the effects of compression on symptoms and some of them accelerate healing of leg ulcers.

Mode of action

VADs have two pathophysiological mechanisms of action. They alter macrocirculatory changes in the venous wall and venous valves that cause hemodynamic disturbances to produce venous hypertension6 and they alter microcirculatory effects of venous hypertension that lead to venous microangiopathy.6 The mode of action varies depending on the drug product.

Action at the macrocirculatory level

Mechanisms of action on the venous wall and valves are summarised in Table VIII.293-325 Until recently, the most popular theory was that weakness of the vein wall produced venous dilatation causing secondary valvular incompetence. For this reason, research on VADs was focused for a long time on their effect on venous tone. Most VADs have been shown to increase venous tone by a mechanism related to the noradrenaline pathway. Micronized purified flavonoid fraction (MPFF) 293,294,303-305 prolongs noradrenergic activity, hydroxyethylrutosides295,314 act by blocking inactivation of noradrenaline, and ruscus extracts 296-302 act by agonism on venous ccl-adrenergic receptors. A high affinity for the venous wall was found for MPFF326 and hydroxyethylrutosides.309-311 The precise mechanism by which other drugs increase venous tone is not known.

More recently, as indicated in Part I, it has been realized that chronic venous disease is related to primary failure of venous valves that are affected by inflammation.308,327 Currently available drugs have been shown to attenuate various elements of the inflammatory cascade, particularly the leucocyte-endothelial interactions 306, 317-319, 322,325 that are important in many aspects of the disease.63,328, 329 Results of a recent trial performed on an animal model of acute venous hypertension revealed that MPFF showed an anti-inflammatory effect under this acute situation that may result in protection of venous valves in chronic conditions.307

Action on the microcirculation.-VAD effects on capillary resistance, lymphatic drainage, protection against inflammation, and blood flow are summarized in Table DC.330-395

Capillary resistance.-Numerous studies have shown that VADs are able to increase capillary resistance and reduce capillary filtration. This is seen for MPFF,33o-347rutosides,351-355 escin,379 ruscus extracts,349, 350,356 proanthocyanidines,348,358, 359 and calcium dobesilate.382-385 The capillary protective effect of MPFF may be related to inhibition of leukocyte adhesion to capillaries.334,335,343, 345-347 This is enhanced by micronisation.364

Lymphatic drainage.-The efficacy of coumarin on Iymphedema has been described by Casley Smith.395 Coumarin combined with rutin reduce high protein edema by stimulating proteolysis.377 MPFF improves lymphatic flow and increases the number of lymphatic vessels 361363 and calcium dobesilate enhances lymphatic drainage.386-389

Protection against inflammation.-In animal models of skin inflammation, VADs appear to attenuate the inflammatory response by various mechanisms. Numerous reports have confirmed free-radical- scavenging, anti-elastase and antihyaluronidase properties of most VADs (rutosides,357,360 escin,380 ruscus extracts,381 proanthocyanidines,358, 359 calcium dobesilate,390-392 and MPFF 366- 368).

Hemorrheological disorders. -Hemorrheological changes are constant in CVD appearing as a basic trait with increased blood viscosity due to plasma volume contraction and increased fibrinogen as a consequence of inflammation.396 The presence of huge red cell aggregates in the vicinity of venules reduces blood flow to cause poor oxygen delivery from red cells. Erythrocyte aggregability and blood viscosity increase with greater severity of disease.396 Some VADs limit red cell aggregation (Gingko biloba),396 decrease blood viscosity (MPFF,374,375 calcium dobesilate,394), and increase red cell velocity (MPFF).375

Therapeutic efficacy of oral VADs on venous-related symptoms

The main indications for VADs are symptoms related to varicose veins or attributed to CVD (heavy legs, "heaviness", "discomfort", pruritus, pain along varicose vein paths) or less specific but frequently associated symptoms (paresthesiae, night time cramps or restless leg syndrome) and edemata, 390.397

Two reviews of VADs published recently by Martinez et al..398 (Cochrane review) and by Ramelet et al.291 studied the efficacy of the drugs in detail. The paper by Ramelet et al. represented proceedings of an International Medical Consensus Meeting on "Veno- active drugs in the management of chronic venous disease" held in the framework of the 13th Conference of the European Society for Clinical Hemorrheology (ESCH) in Siena, Italy.291

Data from randomized, double-blind, placebocontrolled trials (RCTs) for the efficacy of VADs at any stage of disease were extracted by independent reviewers who also assessed the quality of trials according to quality criteria specified in the Cochrane Handbook 398 or evidence-based medicine predefined criteria or their own experience.291 Outcomes included edema, venous ulcers, trophic disorders, and symptoms (pain, cramps, restless legs, itching, heaviness, swelling and paraesthesiae)398 or symptoms only at any stage of the disease.291

Many VADs consisting of natural products (flavonoids: rutosides, french maritime pine bark extract, grape seed extract, micronized diosmine and hidrosmine, disodium flavodate; saponosides: centella asiatica) and synthetic products (calcium dobesilate, naftazone, aminaftone and chromocarbe) 398 were explored. Escin was excluded from the Cochrane review of Martinez et al. 398 but was evaluated in the Cochrane review of Pittier and Ernst399 and was covered by the consensus paper.291

Studies were classified as level I (low risk of bias), level II (moderate risk of bias) or level III (high risk of bias).398 Alternatively, they were associated with grade of recommendations: grade A (RCTs with large sample sizes, meta-analyses combining homogeneous results), grade B (RCTs with small sample size, single RCT) or grade C (other controlled trials, non-randomized controlled trials).291 One hundred and ten RCTs were identified in the Cochrane review,398 but eventually only 44 of them were included in the efficacy analysis. Eighty three trials of VADs were analysed in the consensus paper 291 with 31 of these retained 258, 399, 400, 403,406.408-411.415,421-434 for assessing the grade of recommendations for each medication (25 RCTs and 6 meta-analyses). The efficacy of VADs on both symptoms and signs related to CVD estimated by relative risk applying a random effects statistical model is displayed in columns 2 and 3 of Table X398 with the grade of recommendations per individual medication shown in columns 4 and 5.291

One of the limitations in the Cochrane reviews 398 is that while all studied the full spectrum of conditions seen in CVD, only 23% of the studies reported the diagnostic classification used. Of the studies that did report it, Widmer's classification was used most frequently,403,409,410,415,421 followed by the CEAP classification.400,405 Only symptoms were considered in the consensus paper291 allowing a better uniformity of outcomes.

Therapeutic efficacy of oral VADs on edema of venous origin

Although edema is a non-specific sign, it is one of the most frequent and typical complaints of CVD. All other causes should be excluded to confirm the venous origin of edema. Chronic venous disease-related edema is described as a sporadic unilateral or bilateral edema limited to the legs which may also involve proximal parts of the lower extremities. It is enhanced by prolonged orthostatic posture and improved by leg elevation.435

Several well-conducted, controlled trials versus placebo 404,406.412.415,436-438 or stockings 409-439have shown efficacy of oral VADs such as micronized purified flavonoid fraction,412 rutosides,436,437,439 horse chestnut seed extract,438 calcium dobesilate,406 proanthocyanidines 404 and coumarin rutin.415 In these trials, evaluation of the antiedema efficacy was based on objective measures such as leg circumference assessment, strain- gauge plethysmography and water displacement. Other large-scale trials performed internationally,154 on air-travel edema,440,441 on healthy volunteers 432 or in patients with varicose veins or postthrombotic syndrome 442 have shown the value of VADs in reducing leg edema. Results of meta-analyses have confirmed the anti-edema efficacy of such medications.398,443

Pharmacological treatment of leg ulcers

Healing of venous leg ulcers (stage C6) has been shown to be accelerated in double-blind studies using "micronised purified flavonoid fraction"(MPFF).187,411-444 This was confirmed in 2005 by a meta-analysis of 5 trials using MPFF as an adjunct to standard treatment in 723 patients of stage C6 of the CEAP classification.445

Among VADs, the use of horse chestnut seed extract or of hydroxyrutosides failed to demonstrate superiority over compression in advanced chronic venous insufficiency 258,446 or in preventing venous ulcer recurrence.447

A small number of other drugs have been used with varying success. Stanozolol, a fibrinolytic anabolic steroid was expected to break down pericapillary fibrin cuffs but did not increase the rate of ulcer healing.448 Abnormalities of coagulation observed in patients with venous disease have been improved by aspirin 449 but there is a lack of data supporting its use for preventing thromboembolic events in patients with CVD.6 A thromboxane receptor antagonist (Ifetroban) failed to show benefit over compression therapy in ulcer healing.450 Several trials have suggested that pentoxifylline may improve venous ulcer healing rates although the magnitude of the effect appears to be small and its role in patient management is unclear.6,451,452

Safety of oral VADs

Safety of VADs is in general good, except for hepatotoxicity from coumarin and benzarone. Adverse events most commonly associated with VADs are gastrointestinal (e.g. abdominal pain, gastric discomfort, nausea, dyspepsia, vomiting and diarrhoea) or autonomie (e.g. insomnia, drowsiness, vertigo, headache and tiredness). They occur in approximately 5% of patients treated (Table XQ.453-455 Some VADs have been used without any problems during the second and third trimester of pregnancy but there are no long-term series documenting this. Thus, caution is recommended when administering VADs to patients who are breast feeding because of absence of data concerning diffusion of these medications into breast milk.

Indications for oral VADs

In France where VADs are widely prescribed, recommended prescribing practices for "Venotropics in venous insufficiency of the legs" 456 state that it is not appropriate to prescribe VADs in the absence of disease-related symptoms (heavy legs, pain, restless legs on going to bed) or in varicose veins if they are not associated with symptoms. In addition, VADs should not be prescribed for more than 3 months except in the event of recurrence of symptoms after treatment discontinuation. It is not appropriate to combine several VADs in the same prescription.

Although trials of VADs on the improvement of symptoms are numerous, the anti-edema effect of VADs has been objectively demonstrated in double-blind trials. VADs may be indicated as a firstline treatment for CVD-related symptoms and edema in patients at any stage of disease. In more advanced stages, VADs may be used in conjunction with sclerotherapy, surgery and/or compression therapy.291,453

A meta-analysis of micronised purified flavonoid fraction further confirmed its valuable contribution for healing leg ulcers as an adjunct to standard treatment.445

Combination of oral VADs with other methods such as compression

VADs may accentuate the effect of compression. A double-blind trial demonstrated that the combination of compression and VADs was more effective than compression alone 409,439 and may be prescribed instead of compression when compression is contra-indicated as in the presence of arterial insufficiency or neuropathies or where compression is poorly tolerated (individual reactions, summer heat). There is only one randomized study comparing VADs versus stockings to prevent edema.258

Topical treatment

VADs and heparinoids are blended in topical preparations. The formulation, especially in gels, has a relieving effect on some symptoms. Natural heparin and heparinoids have anti-inflammatory properties, an analgesic effect by inactivating histamine, and anti- thrombotic effects. The transcutaneous effectiveness of VADs and heparinoids depends on their concentration. Several brands are associated with other active substances such as polidocanol or a local anesthetic agent. A double blind study has been performed to prevent edema in long flights with a rutoside gel, which proved to be more effective than its excipient.

Other drugs


Method of action,-Pentoxifylline is a vasoactive drug that reduces leucocyte adhesion and has rheological action on erythrocytes and a mild fibrinolytic action.460

Effectiveness.-In a systematic review, Jull et al. identified 8 clinical trials (547 adults) published from 1983 to 1999 comparing pentoxifylline to placebo, either associated with compression (n=445) or not (n= 102).461 They conclude that "our results suggest that pentoxifylline gives additional benefit to compression for venous leg ulcers, and possibly is effective for patients not receiving compression". However, positive global findings are strongly influenced by old studies with obsolete methodology. Diagnostic methods confirming a venous etiology of the ulcers are not reported in 2 of the 8 trials; while the diagnosis is based on clinical signs only in 4 and by Doppler ultrasound in only 2 of the 8 trials.

Results of recent studies are not conclusive. One trial with pentoxifylline and placebo did not reach statistical significance.452 However, the placebo double blind studies of Falanga 462 and Belcaro 463 indicated that pentoxifylline was effective for healing leg ulcers. In an open randomized trial with debatable methodology (inpatients were not distinguished from outpatients), Nikolovska 464 obtained good results from treating ulcers with pentoxifylline in the absence of compression. In one study,462 a higher dose of pentoxifylline (800 mg three times a day) was more effective than the lower dose (1200 mg daily).

Combination with other methods such as compression.- Pentoxyfilline therapy increased the rate of ulcer healing when combined with compression in some studies451,452,462,463,465,466 or given on its own.464,467 However, the use of such an adjuvant drug without adequate compression therapy should be considered only when compression is not tolerated or contra-indicated.

General recommendations for use.-Although pentoxyfilline is relatively well tolerated, its value for treating leg ulcers remains debatable until new data become available.

Prostaglandins E

Introduction.-Few studies have been devoted to the efficacy of prostaglandins (PG) for venous leg ulcers. Systemic or local PG are rather indicated for arterial ischemie ulcers. The method of action of PG is not well defined in published trials. Probable actions may include small vessel dilatation and augmented blood flow in the capillaries, increased fibrinolytic activity, effects on reducing platelet and leucocyte aggregation and adherence to endothelium, and reduction of white cell activation.

Intra-venous PGE.-In a double-blind, placebocontrolled study by Rudofsky,468 42 patients were randomly given either one i.v. infusion over 3 hours of 3 ampoules of Prostavasin (60 micrograms PGEl) or 3 ampoules of placebo daily diluted in 250 ml saline over a 6 week period. In the PGEl group (n = 20) there was a significant improvement in the ulcer status compared to placebo (n = 22) (P

Topical prostacyclin analogues - Iloprost.-In a multicenter, randomized, double-blind, placebocontrolled study in patients with venous leg ulcers, the efficacy and tolerability of topical applications of a prostacylin hydrogel (iloprost) was investigated 471 with 34 patients allocated to placebo treatment and 65 patients to iloprost treatment given in two concentrations. Both iloprost concentrations were well tolerated. In a second paper,471 the same team compared placebo to two iloprost concentrations in a larger number of patients with 49 patients allocated to treatment 1 (placebo solution), 49 patients to treatment 2 (0.0005% iloprost solution) and 50 patients to treatment 3 (0.002% iloprost solution). The solutions were applied to the ulcer edge and surrounding skin twice weekly for eight weeks. No significant difference was found in favor of the iloprost treatment in either study.

Absorption of topical iloprost may be variable. In a study by Meyer,472 iloprost could not be detected in the plasma in 40% of patients, whereas iloprost was absorbed through the ulcer base in variable degrees in the others. There was no direct relation between the ulcer size and amount of iloprost absorbed.

Intravenous or perilesion injection of PGE1.-In a study by Tondi,473 80 patients suffering from ischemic ulcers were enrolled. Treatment for 25 patients was with injection of low doses of alprostadil around the ulcers and intravenous saline infusion, and in a further 25 by intravenous alprostadil infusion and local injections of saline, while the control group of 30 patients received saline injections around the ulcers and intravenous saline infusions. All patients treated with PGEl showed statistically significant improvement in ulcer diameter, pain, and transcutaneous oxygen pressure compared to controls. Both intravenous and local subcutaneous alprostadil may be useful for treating ischemie leg ulcers, but subcutaneous administration is less expensive and easier to perform. A similar study in patients with venous ulcers has not been performed.

Indications.-Chronic leg ulcers (C6) may be an indication for either intravenous or topical PG but there are only few data on this topic and no recent studies for venous ulcers.

General recommendations for use of PG.-As the efficacy has not yet been fully demonstrated, no recommendation can be made.

Topical therapy for venous ulcers

A wide range of topical agents and dressings has been advocated to promote desloughing, granulation and re-epithelialization of venous ulcers, including hydrogels, alginates, hydrocolloids, enzymatic agents, growth factors, foams and films.474-498 Tissue- engineered skin equivalents based on cultured keratinocytes and fibroblasts have been shown to accelerate healing.4"-501 However, there is no level I evidence that the other agents provide additional benefit over simple wound dressing and compression therapy.

The use of topical antibiotics in patients with venous ulcerations is discouraged because of emergence of resistant organisms and increased risk of contact dermatitis.474,502 However, systemic antibiotics are indicated in the presence of sshemolytic streptococcus and evidence of soft tissue infection. Topical antiseptics exhibit cellular toxicity that exceeds their bactericidal activities and they have been found to impair wound epithelialization.503


Liquid sclerotherapy

Outcome after treatment of varicose veins is commonly described by the rate of recurrence. It is generally accepted that sclerotherapy is effective for treating Cl and some C2 CVD. However, sclerotherapy is reported to fail for all other clinical levels with increased frequency the longer patients are followed reaching 90% at 10 years.504507 In this respect, randomized trials have shown surgery to be superior to sclerotherapy for treating main stem GSV and SSV disease soe-sio unless the incompetent saphenofemoral junction is Hgated first.504,505 Ultrasound-guided techniques may improve early results 511.512 DUt long-term benefit has not been established.513 In practice, sclerotherapy is frequently combined with other interventions.

Foam sclerotherapy

When delivered as foam, detergent sclerosant is more active within the vein because it is not diluted by blood and persists in the treated vessels. Foam can be readily visualized by ultrasound and can be used to treat C2-C6 CVD.5H-5i6ReSults out to more than 5 years demonstrate clinical effectiveness rates exceeding 80%.517- 519Foam has been shown to be superior to liquid sclerotherapy in the GSV in terms of clinical and hemodynamic outcome.520,521 Treatment of C4-6 has been particularly rewarding.518,522 There are two RCTs publishe