Lymphovenous Canada: An Update on the Low Cost Management of Lymphoedema in the developing world

Low cost reduction of morbidity is a key aspect of management of the disease in the developing world and can be achieved through simple treatment practices such as limb elevation and movement, good skin care and breathing exercises (Table 1). This article will discuss some of the evidence base behind these practices. A new perspective of morbidity control in lymphoedema will be presented, in which the venous system is targeted in particular as the most significantly affected system using the basic interventions described above. It is suggested that correction of venous abnormalities in lymphoedema leads to a reduction in the extent of lymphatic failure.

The normal lymphatic system is a network of lymphatics in the upper dermis which preferentially drains into and along preferential lymphatic trunks leading directly to the lymph nodes. In lymphatic filariasis the main lymphatic channels become dilated and partially obstructed. The network of lymphatics becomes a "safety valve" and acts as a conduit through the skin until lymphatic trunks are found by the escaping lymph proximal to the limb and which are capable of acting as a low resistance or preferential pathway. This "safety valve" is a secondary system which fails if overloaded. The purpose of this paper is to emphasise that such overloading is usually due to a venous hypertension or inflammation of the skin.

Venous failure and its role in lymphatic failure

Increased dependency and stasis of the large lymphoedematous limb may lead to a state of venous hypertension, increased vascular permeability and angiogenesis. Chronic venous hypertension may therefore become an increasingly important feature of limbs suffering from lymphatic failure, and may further exacerbate oedema in a vicious circle as venous hypertension overloads the lymphatics. (Table 2)

Evidence of Angiogenesis and Leakage in Lymphoedema

Casley-Smith (1980) described increased blood vessel volume (0.009 ml/cm³ to 0.2 ml/cm³) in the deep fascia of experimental and congenital canine lymphoedema models. Snowden and Hanmerberg (1986) demonstrated this using angiography showing increase in vessel calibre and number in a filaria infected canine model.

Angiogenesis has also been widely observed in human lymphoedema. Histologically it is seen in the upper dermis where in normal skin a mobile fluid compartment of tissue fluid has been shown to exist (Eisenbeiss et al 2001, Hu et al 1998). The venules seen are thick walled, extended and tortuous (Fig 1). Roberts et al (1994) noticed an increase in the number of sub papillary capillaries in a series of 15 post mastectomy patients- this resulted in a constant capillary density despite a 14% increase in cutaneous area, strongly suggesting that neovascularization had occurred. This was confirmed by Mellors et al 2001.

It is well known that new vessels or the capillary bed involved in inflammation have an increase in permeability. The explanations include the leaky structure of new vessels and mediators of inflammation, which are associated with these pathological changes. Such mediators range from histamine to Vascular Endothelial Growth Factor (VEGF) (initially called Vascular Permeability Factor or VPF) which is, as an inducer of permeability, up to 5000 times as potent as histamine (Murohara, 1998). Plasma levels of VEGF are higher in patients with chronic venous disease (Shoab et al 1997).

VEGF production is higher in patients with venous hypertension (Coleridge-Smith, 2000) and in such patients an acute increase in venous pressure caused an increase in blood VEGF but not in the control group without venous disease. VEGF plays a fundamental role in the growth and differentiation of vascular and lymphatic endothelial cells. Lactate elicits its production in chronic wounds (Constant et al 2000). The VEGF family consist of VEGF-A, B, C and D (Karkkhainen and Petrova 2000). VEGF-C and D both act on VEGFR-3 (vascular endothelial growth factor receptor 3) which is a receptor for tyrosine kinase selectively expressed on lymphatic endothelium. Familial Milroy's lymphoedema has recently been attributed to mutations of the VEGFR-3 gene leading to disturbed transduction of the receptor. The epidermis has been found to be the likely source of VEGF (Pardoe et al. 1996) and it may be produced in response to any damage leading to hyperproliferation including transient ischaemia or mechanical stretch, both of which occur in lymphoedema. Increased vascular permeability can worsen oedema, but in addition it is becoming clear that chronic lymphoedema is a potent stimulator of angiogenesis through mechanisms that may be mediated by VEGF.

Angiogenesis could result from the failure of clearance of angiogenic factors such as normal metabolites, inflammatory agents and growth factors including VEGF. Changes in tissue compliance may also be a factor. Mechanical stretch encountered in oedema in itself may stimulate the epidermis to release factors such as VEGF. Tissue stretch by oedema may also separate cells and connective tissue, creating spaces into which vessels may grow more readily. (Ryan 1973 & 1995). The key point is that angiogenesis, however it is initiated, does exist in lymphoedema and may lead to increased venous hypertension and lymphatic overload in lymphatic disease through a number of mechanisms.

Venous hypertension in Lymphoedema

There is a long history of studies of venous hypertension and accelerated blood flow in lymphoedema. Mayall of Brazil and his co-workers have termed the hyperostomy syndrome following its initial description by Pratesi in 1952 (Campisi 2000).

Venous hypertension is a coexisting feature of lymphoedema. Kim et al. (1999) showed venous insufficiency in all 41 patients in their series with lymphoedema. All patients had increased venous pressure, venous volume, filling pressure and a decreased ejection fraction all of which were proportional to the size of the swollen leg.

Increased blood flow has been recorded in post mastectomy lymphoedema (Martin and Foldi, 1996, Svensson et al. 1991). This was thought to worsen the oedema present. Jacobsson (1967) using radio-active isotope clearance showed a 42% increase in blood flow in 25 such arms. An increase in arterial flow could have a knock-on effect by increasing venous filling and causing venous hypertension. Svensson (1994) noted that 70% of post mastectomy lymphoedema patients had abnormal arm venous outflow. Solti (1976) suggested that the increased blood flow through the arm together with arteriovenous shunting caused by angiogenesis in lymphoedematous arms may lead to increased venous pressure.

Evidence exists therefore, to associate the presence of venous hypertension with lymphoedema. That venous hypertension can cause changes in lymph flow has been shown in a number of studies.

Chen et al (1991) looked at the effect of increasing venous pressure on lymph flow transients (immediate changes in lymph flow) in the dog paw. They noted an increase in lymph flow on average 5.3 minutes after pressure elevation.

Impaired lymph transport secondary to venous thrombosis has been shown in at least two studies. Brautigam and Foldi (1998) looked at lymphatic drainage using radioactive tracers in the acute and chronic post thrombotic syndrome leg. They found that in the acute stage there was increased venous pressure and increased lymphatic drainage as shown by increased tracer uptake in the epifascial lymph nodes, but in the chronic phase there was no such increase in venous pressure but there was a decrease in the subfascial lymph node tracer uptake, suggesting lymphatic failure is present in long standing cases of post thrombotic syndrome. In another study, impaired lymph transport secondary to venous thrombosis was demonstrable shortly after deep vein thrombosis of the leg leading to venous hypertension (Askar 1969).

Indeed, the therapy by electromagnetic diathermia to warm the skin which is used to treat some cases of lymphoedema was found in a recent study to significantly improve venous flow in healthy patients but to have no effect on lymph flow. They concluded that "if thermal therapy is having a salutary benefit on lymphoedema it must be operating through some other physiologic mechanism perhaps involving greater venous return" (van der Veen et al. 2000).

Hence it can be concluded that chronic venous hypertension is a feature of lymphoedema, and can cause overload of lymphatics. This could be through a direct effect of increased capillary pressure and increased capillary filtration, due to mechanical effects of an enlarged venous system on adjacent lymphatics or due to reduced clearance of inflammatory mediators which damage the lymphatics- Coleridge-Smith (2000) has established that there is accumulation of such mediators in venous failure. It is the basis of the Matrix Trapping theory (Falanga and Eaglstein 1993) put forward as a cause of lipodermato sclerosis induced by venous hypertension.

Breathing and Lymphatic Failure

Breathing exercises have been used as a standard component of complex decongestive physiotherapy (CDP). The rationale for this is that the main lymphatic trunks lead to the thoracic duct, and unless they become emptied they act as a static reservoir into which the peripheral lymphatics drain, with little encouragement to flow and exit into the venous system (Ryan, 2001).

The clearance of central lymphatics is essential before peripheral manoeuvers are taken to increase lymphatic drainage. Boris et al (1998) report a significantly higher incidence (43% versus 3%) of genital oedema following the use of compression pumps for the lower limb to treat lymphoedema. This oedema persisted despite discontinued use of the pump and shows the dangers of trying to clear peripheral fluid before clearing central drainage routes. Fluid dispersed from the legs is displaced into more proximal tissues causing genital oedema as the main lymphatic trunks and thoracic duct are full and congested. Protein remains in the oedematous tissues resulting in recurring oedema.

Breathing acts by increasing thoracic duct flow helping to decongest the central lymphatics. This is explained by the existence of a respiratory pump for thoracic lymph flow. Schad et al (1978) found increased thoracic duct flow during hyperventilation in anaesthetized dogs, and decreased duct flow upon opening the thoracic cavity. Opening the thoracic cavity decreased intrathoracic pressure changes acting on the thoracic duct. They suggested that during inspiration there is a drop in intrathoracic pressure and increase in intra-abdominal pressure as the diaphragm pushes down, with consequent movement of lymph form the abdominal to the thoracic cavity. During expiration they suggested the increase in intrathoracic pressure causes lymph to be expelled from the intrathoracic duct into the veins of the upper thorax. Further evidence comes from Dery et al (2000) who noted that closed thorax compression on mice can result in lymph uptake from distant parts of the body.

Gabel and Drake (1992) found that increased venous pressure caused increased thoracic duct pressure in sheep, suggesting that thoracic duct flow is also dependent on the pressure of the vein into which it empties.

Breathing also affects intrathoracic venous pressure. Walker and Pickard (1962) noted that there was a negative pressure in the inferior vena cava immediately above the diaphragm with a positive pressure below it and noted both were augmented in deep inspiration. This would also apply to the veins into which the thoracic duct empties.

Hence increased thoracic duct flow during inspiration could be explained by an increased pressure gradient between the duct and the veins into which it empties- the venous pressure reduction during inspiration driving the flow of lymph out of the thoracic duct. It could also be explained by a direct effect of increased thoracic duct pressure during inspiration.

It is interesting to note that an effective form of breathing used for lymphoedema patients involves taking a deep, sharp breath in through the nose (rapid inspiration) holding of the breath, and then a long slow phase of breathing out through the mouth (slow expiration). The sharp inspiratory phase in theory would maximize thoracic duct lymph flow on the basis of the evidence provided above.

Of equal interest is that this system of breathing is used by many traditional systems of medicine and is widely used in Asia to promote health e.g. Chinese Dragon Breathing T'aichi and Yoga. (Hewitt 1983)

That breathing (hyperventilation) has an affect on the lymphatics of the skin has been demonstrated by Allegra et al 2001 in both primary lymphoedema and in patients with chronic venous insufficiency. Using a Servo Nulling system, a counter pressure pump and a pressure transducer hyperventilation was found to lower lymphatic pressure by approximately 50%.

Movement in Lymphatic Failure

Another key mainstay in treatment of venous ulcers is movement of the lower limb and especially the ankle. This can also be applied to lymphoedema.

Movement increases both blood flow and lymph flow through the limb. Small amplitude movements can produce these effects, and are better than vigorous exercise in lymphoedema as exercise can increase blood flow and tissue fluid formation in the limb.

Movement can improve lymph flow through 3 possible mechanisms:

1. Increased uptake by initial lymphatics due to increased tissue hydrostatic pressure variations and elastin "snap back" of the initial lymphatic wall
2. Increased pumping of lymphatics due to smooth muscle contraction in collecting lymphatics
3. Emptying the venous system and reducing overload of the failing lymphatics and allowing them to recover their function.

Mortimer et al. (1990) noted increased radioisotope clearance from the lymphatics of the lower limb on movement of the limb- even small amplitude ankle movements produced a significant increase in isotope clearance to the inguinal lymph nodes. This in itself would help in the clearance of oedema by facilitating lymphatic flow out of the affected limb. However, the effect of movement may actually be more important in counteracting venous hypertension and facilitating venous return from the swollen limb.

Daley et al (1965) noted that a passive pumping motion of the legs decreases saphenous venous pressure (i.e. empties the vein) within a few seconds. This suggests that small amplitude movements such as ankle movements may be effective in activating the muscular venous pumps of the lower limb leading to increased venous return. Gardner (1993) discusses evidence in dog and sheep models that simple dorsiflexion of the ankle joint leads to the ejection of blood from veins and the emptying of foot veins, and suggests that the venous system is most efficient in pumping venous blood back to the heart during normal walking. This mechanism is therefore reduced in lymphoedema where there is restriction of ankle movement.

There is also much evidence that movement can aid in healing of venous ulcers caused by venous hypertension. Reduced ankle mobility has been shown to be a risk factor for poor ulcer healing. Barwell (2001) showed that chronic venous ulcers were slower to heal in patients with reduced ankle mobility- 13% of ulcers healed in patients with less than 35^o ankle mobility, whereas 60% of ulcers healed in patients with greater than 35^o of mobility. Similarly, Doherty (2001) found that individuals with fixed ankle joints have a significantly longer healing time than those with mobile ankle joints and Brooks et al. (2001) found decreased recurrence of venous ulceration in legs of patients with full ankle movement.

The important distinction that walking velocity and joint mobility are independent factors in ulcer healing has been shown by Kigler et al. (2001). Their study measured peripheral venous pressure directly in two groups of healthy subjects- one group with joint restriction as a result of wearing a limb cast versus the control group. The study found that restriction of ankle mobility had the same effects as restriction of knee mobility- both impaired the lower limb muscle-vein pump, emphasizing the importance of ankle movement. Maximal reduction of venous pressure on standing was found to be reduced by 12-21.5% in limbs with restricted joint mobility.

This pressure reduction impairment signifies reduced venous emptying on standing and suggests that reduced ankle mobility may worsen venous hypertension in patients with chronic disease. Andrade 1999 reviewed the literature and compared ankle mobility in patients (120 limbs) with lymphoedema with that in 22 normal volunteers and found the range of movement between ankle flexion and extension was about halved (normal 61.1% lymphoedema patients 34%). Godoy et al 2001 in six patients showed a volume change of only 160ml on elevation but a volume change 830ml after passive extension and flexion of a foot.

Hence it can be concluded that movement is essential in order to drive blood through the veins out of the swollen limb and either to promote the normal function of healthy lymphatics or to relieve the overload on a failing lymphatic system. Continuous low amplitude ankle movements are extremely effective in doing this.

Limb Elevation and Lymphatic Failure

To the encouragement of movement for swollen limbs should be added exhortation to elevate. It is common sense that fluid flow through any tissue plane will be influenced by gravitational effects. However, the key point made here is that the main influence of gravity is on venous flow rather than lymphatic flow.

Swedborg et al (1993) found that elevation was effectively reduced the volume of control and lymphoedematous arms of patients suffering from postmastectomy lymphoedema. However with no difference in volume reduction between control and lymphoedematous arms following 5 hours of elevation, they concluded that elevation alone is not the most effective measure to reduce oedema. They state that elevation is thought to act mainly by decreasing the "hydrostatic pressure gradient from the blood vascular system to the tissues. This reduces the outflow of proteins and fluids from the vascular system". They also concluded that there was a limited effect of gravity on the lymphatics themselves.

However, Pippard and Roddie (1987) found that gravitational changes influence venous pressure but not the pressure within adjacent lymphatics of the hind limb of the sheep when tilted on a tilt table. This suggests that limb elevation serves mainly to increase venous return to the heart and reduce venous pooling of the lower limb, thus reducing venous pressure in the lower limb rather than increasing afferent lymphatic flow. This has clear implications in the reduction of venous pressure of the lymphoedematous limb.

The significant effect of elevation on the venous system has long been known in the field of venous ulcer care. Patients have long been told to elevate their limbs as much as is reasonably possible, with healing being facilitated by the degree of elevation of the limb (Tibbs, 1992). Indeed, Abu Own et al. (1994) found that elevation of the leg 30cm above heart level for 30minutes 2 or 4 times a day can reduce oedema and improve cutaneous microcirculation in patients with chronic venous disease.

Elevation of the lymphoedematous limb therefore seems to affect primarily the venous system, and reduces any venous hypertension present. This has been used in the treatment of venous ulcers, but the same mechanism of action applies in reducing oedema caused by lymphatic failure. While mainly investigated in the group of patients with limb swelling, the role of venous emptying in the management of genital oedema deserves more study.

Another factor much discussed in the Venous literature is the role of the protective arteriolar vasoconstriction response to lowering the limb. It is a reflex that several authors have claimed is lost in various pathologies such as Diabetes Mellitus, Sickle Cell Anaemia (Mohan et al 1997) and in venous disease of the lower legs (Allen et al 1988, Svedman et al 1998). Also worth visiting is the probability that the tissues are affected by an Ischaemia and reperfusion syndrome so that damage from oxygen free radicals is enhanced by exhaustion of protective antioxidants (Matthew et al 2001).

The Skin Barrier and Lymphoedema

Another important low cost intervention which should be considered is that of skin hygiene.

Recurrent inflammatory episodes are a common complication of lymphoedema and exacerbate the condition further. It may play a part in the natural progression of lymphoedema through varying degrees of severity by causing overload and more lymphatic damage during each episode- Pani and Srividya (1995) found that the frequency of inflammatory episodes correlated positively with progression through the grades of lymphoedema.

Most episodes have been blamed on recurrent streptococcus infections. Patients develop a high fever, and increased redness, swelling and pain in the tissues. This has been termed dermatolymphangioadenitis (DLA). Doubts have been raised as to whether bacterial infection is the whole story. The episodes can regress within two days spontaneously and the "devastating flesh eating responses" of contemporary streptococcus infections are very rarely seen in these lymphoedematous limbs.

Control trials comparing the effects of long term antibiotics to simple washing (Shenoy et al. 1999) have shown that washing alone may decrease the number of inflammatory episodes per annum. Shenoy (1995) also found that neither DEC not antibiotics altered DLA attack frequency in their study of 65 patients with filarial lymphoedema. They found that "simple hygienic measures combined with good foot care and local antibiotics/fungal cream when required were effective in reducing the number of DLA attacks. Similarly the experience if Dreyer et al (1999) in Brazil suggests that simple hygiene can protect against infections.

It could be suggested that not all cases of DLA are caused by infection and that simple care of the skin may reduce the inflammatory mediators produced as a result of any kind of damage to the epidermis. It is a reservoir of large quantities of IL-1 and other cytokines like TNF made in response to irritation and secreted into the dermis. Olszewski et al (1992) showed the accumulation of several immune proteins and cytokines in filarial lymphoedema, suggesting ongoing inflammatory responses, despite the lack of overt dermatitis. These proteins, of which interleukin-1 is probably the most important, accumulate in the interstitium due to poor lymph clearance in the limb, and may cause DLA episodes in absence of infection.

Importantly, Foldi (1996) found recurrent DLA attacks were nearly eliminated in post mastectomy lymphoedema patients treated in Foldikilinic with only Complex Decongestive Physiotherapy (CDP) and not antibiotics. This suggests that reduction of oedema through improved lymphatic drainage can clear inflammatory mediators. Interestingly, systemic symptoms of DLA have been observed in some patients following CDP with no preceding signs of infection and may be caused by the release of accumulated inflammatory proteins from the lymphoedematous limb into tissue both locally and directly above the swollen portion of the limb or systemically throughout the body. Foldi (2000) also found a decrease in expression of pro-inflammatory genes after CDP, suggesting that CDP affects the inflammatory process in a fundamental way.

During the past decade the field of wound healing has explored the role of contamination versus infection and has cast some doubt on the need for antibiotics and antiseptics versus simple tap water washing of wounds. These disciplines draw attention to the barrier function of the skin and the importance of its breakdown as a factor determining clinical infection. Pre-antibiotic studies by Sulzberger and Baer (1950) and others showed that heavy contamination by a wide variety of organisms only produced inflammation responses if the barrier function of the skin was first impaired. Studies of skin resistance include a number of interventions that affect the barrier function of the skin

1. Removal of surface fats by acetone/alcohol or excessive washing with alkaline soaps
2. High turnover states of the epidermis as may be seen in psoriasis or even nutritional deficiency diseases or electrolyte disturbances
3. Sellotape stripping of surface layers of the skin

In addition to the above, skin barrier function correlates with a low pH, the acid mantle of the skin. Studies in IL-1 and TNF receptor deficient mice have recently shown a clear relationship between the induction of cytokines within the epidermis and loss of barrier function. The IL-1 receptor deficient mice had accelerated barrier function recovery following barrier disruption, suggesting that the increase in IL-1 following barrier disruption normally delays components of the repair response (Man et al 1999).

The relationship between the loss of barrier function and increased production of inflammatory cytokines, prostaglandins and other inflammatory mediators by the epidermis including VEGF in response to a wide range of stimuli deserves further study. The high mast cell content of lymphoedematus tissue first shown more than a century ago by Ehrlich merits re examination in view of contemporary studies of the H1 and H2 receptors of Keratinocytes and how H1 and H2 receptor antagonists accelerate skin barrier repair and prevent epidermal hyperplasia (Ashida et al 2001).

Thus there is currently a view of skin care procedures which recognizes the existence of subclinical vulnerability of the skin due to the above and the possibility of restoring the barrier function and other aspects of skin health by simple procedures such as washing and emollients. They are already the focus of study by industries manufacturing cosmetics, soaps and laundry products. Against the background of the eczematous changes that occur in venous disease and the gross warty changes that are seen in lymphoedema not withstanding the deep crevices and the effects of fungal infections there is clearly potential for barrier breakdown and enhanced production of inflammatory mediators.

Consequently, there are procedures widely used to reduce such eczematous/dermatitic changes include the washing away of debris and crusts and the replacement of skin surface natural emollients by emollient substitutes as well as the prescription of anti fungals.

A number of emollients have been suggested for recovery of skin barrier function- lanolin, mineral oil, vegetable oil and petrolatum. Good properties of emollients are exemplified by ultra-pure medical grade lanolin which has replaced conventional grade lanolin that was known to sensitise skin of some patients. It has similar physical and chemical properties to skin lipids, enabling it to imitate and augment skin lipid function. The main functions are:

1. Water absorption. This smoothes the skin surface, allowing minor fissures and cracks to close and thus improving barrier function. Lanolin can hold up to twice its own weight in water, forming a water in oil emulsion within the skin itself. Lanolin and other skin lipids therefore hold moisture in the skin.

2. Reduced Loss of water - Lanolin forms a semi-permeable membrane with water droplet sizes 0.5-300nm, protecting the skin and reducing trans-epidermal water loss by up to 30% (Spruit, 1971).

3. Deep skin penetration and long lasting effects - Lanolin is unusual in that its effects can last up to at least 8 hours after initial administration, and that it is absorbed as deep as the stratum granulosum. Most other cheaper emollients typically last for 2-3 hours.

A strong case can be made for the study of traditional soaps and emollients to seek whether they have similar properties to the above. Some are known to have useful properties such as antisepsis. At least 300 traditional plant based soaps have been identified and they have the advantage of being locally available at low cost as well as a sustainable source of supply (Burford & Ryan in preparation).

The second important part of skin care is bathing. Emollients can be used as a soap or after a bath when the skin is still wet and water will be trapped. There is no consensus on the frequency of bathing or on the mode of bathing (shower versus bath tub) in the treatment of dry, damaged skin - but a common view is that superhydration of the skin is important and compliments emollient use (Hai, 1992). Dermatologists and the nursing profession emphasise the damage done by overwashing such as pre-surgical "scrubbing". The minimum length of bathing has been suggested to be 10 minutes by Brown et al. (1983) who found that this significantly improved dry feet over a period of 2 weeks, and Hardy (1990) who found a 10 minute soak daily significantly improved the skin condition of 15 institutionalized elders.

There are no clear pointers as to the advantages of using hot or cold water. The optimum temperature for most biological systems in the human is 37^oC. The skin of peripheries is adapted for a cooler metabolism. Mortimer and Ryan (in press) have shown that warmth increases the clearance of technesium colloid from the skin by the lymphatic system. Shimotoyodone et al 2001 used topical warming 40-45°C for 2 minutes only to supplement massage of the gingiva and clearance of C¹⁴ methylated albumin. Cool water had no such effect.

Ohkuma (2001) has also found that the effects of vibration and lymphatic flow are enchanced by warming. It has been shown that wound healing is delayed by the use of cold water (Ryan et al, 2000). There have been fashions in the use of heating or cooling for inflammation. It is relevant that the removal of fibrosis by collagenases requires an optimum temperature in the region of 37^oC. It can be argued that the ideal management of the fibrotic lymphoedematous tissue is to remove the mechanical tension on the fibroblast, by the systems described above for the management of lymphoedema and venous disease, while encouraging the collagenases by preventing cooling. The use of cold water should be discouraged and water that has been cooled after boiling that is fit for drinking is preferable at just above body temperature.

Finally, in the exploration of the inflammatory responses of the epidermis in lymphoedema one can now recognize that bacteria have several roles to play in the generation of inflammatory responses of which super antigens and shock proteins are only part of the story. On the other hand the production in mice by damaged epidermis of antimicrobial peptides such as LL-37-enhanced by interferon gamma (IFN gamma) have been shown to influence the survival and multiplication of streptococcus A in wounds, (Pentonjamasp & Gallo 2001). The influence of washing and emollients on these peptides deserve further study in the human.

Ringworm and thrush infection

Dermatologists recognise a condition of pretibal cellulitis as a marker of fungal infection between the toes. They do not recognise it as a marker of lymphoedema unlike the profession of Lymphologists. A case can be made that lymph from the toes normally travels via deep lymphatics and it is only when the latter are malfunctioning that skin lymphatics act as collaterals. Such collaterals require movement of the skin for complete clearance of their lymph contents. It is over the tibia that the skin is moved very little. A fact that has been offered as an explanation of the location of erythema nodosum at that site.

The role of fungi in destroying the epidermal barrier and encouraging the abnormal production of cytokines deserves further study. In the days when anti fungal agents were less well developed there were studies of skin vulnerability and the role of blood supply in pre-disposing to fungal infections. Baer and Rosenthal 1966 attempted to infect 150 young adults by bathing for 30 minutes in foot baths containing Trichophyton rubrum or T. mentagrophytes but none became infected.

By producing inflammatory changes in the skin using croton oil and cantharidin 10 out of 33 persons developed transient asymptomatic infection. Persons who developed blisters were more likely to develop active disease. Thus 20 of 71 persons had clinical infection at one to four weeks after developing blisters. Five persons took three months to develop active disease.

The part played by trauma and the degree of preceding damage to the skin, has been frequently shown to be an important factor in the development of fungus infection Kogoj et al. (quoted by Aravijskij, 1962) showed that trauma to the skin of guinea pigs encouraged active disease following the intracardiac injection of fungi. Aravijskij (1962) describes a surgeon in whom control of fungus of the hands by usually effective therapy only occurred after scrubbing up regimens had been made less traumatic.

Cremer (1953) showed that chilblains encouraged Trichophyton rubrum infection. Andriasian (1967) found that 76 per cent. of 117 patients suffering from thrombophlebitis of the legs were suffering from epidermophytosis of the soles. Out of 143 patients with varicose veins 73.4 per cent. were suffering from epidermophytosis of the soles and 90 per cent. of patients with obliterative endarteritis were suffering from epidermophytids. Healing of the epidermophytosis led to considerable improvement of the main disease. He quotes Knight, Thompson and Hallmann as having suggested that small vessel changes are secondary to the fungus infection in obliterative endarteritis.

Certainly when one observes the microcirculation of the skin or nail bed in persons infected with Candida or ringworm fungi one notes considerable hypertrophy of the blood vessels supplying the epidermis. Frequently there are microhaemorrhages and complete irreversible thrombosis of the papillary vasculature. Since there is a relation between the structure of keratin and its invasion by dermatophytes and yeasts (Achten, 1967) and since keratin formation normally depends on healthy epidermal capillary relationships there is a tendency for fungal infection to encourage the kind of disturbances in both keratin and blood vessels which provide an ideal environment for their growth.

It is possible that some emollients or soaps that act on the skin barrier function might play a role in epidermal or blood vessel responses to the presence of a fungus and thereby promoting host resistance or at least a less complete breakdown of the barrier.

The use of antibiotics

The organisms that invade the skin and cause cellulitis exist on the skin or in the throat in 10-20% of persons and in most of these they are not causing any pathology. They have the potential to do great harm, especially if they penetrate deeply and are not destroyed by the host. Most recurrent inflammatory episodes (DLA) produce local inflammation of the tissues but rarely causes their destruction, nor shock or death of the patient. The streptococcus that is most commonly blamed remains sensitive to penicillins, tetracyclines and several other antibiotics.

Host defences are diminished by diabetes mellitus, AIDS, alcoholism and locally by lymphoedema. In the wound healing field there is an interest in Biofilm - a Matrix of extracellular polysaccharide that surrounds and protects bacteria from antiseptics. Surfactants are necessary to reduce such sanctuary sites. The possibility of sanctuary sites within lymphoedematous tissue should be considered. For most cases of DLA one can afford to await resolution as a result of the measures described above, but antibiotics should be prescribed if improvement is not quickly apparent. Aggressive pathology such as blisters and extreme pain with high fever out of proportion to the local appearance may require intravenous antibiotics and surgical debridement.

Prophylactic long term antibiotics are often prescribed in the developed world where they are fully affordable. Fortunately the risks of antibiotic resistance seem exceedingly low in the case of the organisms responsible for bacterial inflammatory events of the DLA kind.

Bandaging and Hosiery With Manual Lymphatic Drainage

Bandaging and hosiery with manual lymphatic drainage are components of the gold standard management of lymphoedema. They are largely unavailable in the developing world due to the lack of trained personnel and the cost of materials. It is however a concern that the good effects of manual lymphatic drainage are mostly unavailable outside specialist centres. The pioneering efforts of Godoy in Brazil (Godoy and Godoy 1999) providing a simplified self help system in a well illustrated booklet should be more widely disseminated. In the long term all countries need this expertise and for the difficult case, there needs to be investment into the training of specialists. This paper however is for the time being promoting low cost, self help.

Conclusion

On the basis of bringing together the opinions currently expressed in the literature concerned with lymphatic venous and skin disease we conclude that low cost self help management of lymphoedema can be advocated. It is an amalgam of systems already in use and although the evidence base is not yet perfect there have been many studies which give support to the management schemes described above.

The interaction of factors determining loss of barrier function (to infective organisms, allergens and irritants) and the chronic effects of venous hypertension with lymphatic failure are amenable to therapy. Even the grossest of skin hypertrophy can be reversed over time so that the handicap of lymphatic filariasis can be eliminated as well as its causative organism. It requires the cooperation of the patient and a concordance program that includes the patients community, fits in with their lifestyle, and assuages conflicting elements. This program is designed to do just that.

If this programme works it is likely to be of benefit in the management of many other important diseases including the skin of patients with AIDS, or for the management of breaks in the surface continuity of epithelium that are highly prevalent in the tropics or in metabolic diseases such as diabetes mellitus or in vulnerable groups such as the bedridden, elderly.

It requires recognition that the grotesque changes that occur in the most severe cases of lymphoedema are not merely a consequence of lymphatic failure alone. There is especially in all cases a strong component of both venous and epidermal failure which responds to therapies specifically aimed at the veins and epidermis.

References

• Abu-Own A, Scurr JH, Coleridge-Smith PD. Effect of leg elevation on the skin microcirculation in chronic venous insufficiency. J Vasc Surg 1994 Nov;20(5):705-10
  
• Achten G (1967) Les Onychomycosos13^th Int Congr Derm Munich ed W Jadassohne and C G Schirren. Springer Berlin. 828-829.
  
• Allegra C, Bartolo M, Carioti B, Carlizza A. Microlymphology and clinical pattern. Europ J Lymphology 2001 9 58 (Abstract)
• Allen A J. Wright D,D,LL. McCollum CN. Tooke JE. Impaired postural vasoconstriction; a contributory cause of Oedema in patients with chronic venous insufficiency. Phlebology 1988 3 163-8.
• Andrade M F C. European J Lymphology 1999 7.
• Andriasian GK (1967) Abs 13^th Int Congr Derm. Munich Springer, Berlin p72.
• Aravijskij A N (1962) The Onychomycosos and the Profession in Symposium Dermatologorum Vol 3. Universitas Caroline, Prague. 67-70.
• Ashida Y, Mitsuhiro D, Telsuj, H. Histamine H1 & H2 receptor antagonists accelerates skin barrier repair and prevent epidermal hyperplasia induced by barrier disruption in a dry environment. J. Invest Derm 2001 116 261-265
• Askar OM. Communicating lymphatics and lympho-venous communication in relation to deep venous occlusion of the leg. Lymphology (1969) 2:56-63
• Baer RL, Rosenthal SA Biology of Fungus Infections of the Feet. J. Amer med Ass (1996) 197, 1017-1020.
• Barwell JR, Taylor M, Deacon J, Davies C, Whyman MR and Poskitt KR. Ankle motility is a risk factor for healing of chronic venous leg ulcers Phlebology 16 38-40 2001.
• Boris M, Weindorf S, Lasinski BB. The risk of genital edema after external pump compression for lower limb lymphedema. : Lymphology 1998 Mar;31(1):15-20
• Brautigam P, Foldi E, Schaiper I, Krause T, Vancheioll W and Moser E. Analysis of lymphatic drainage in various forms of leg edema using two compartment lymphoscintigraphy. Lymphology 1998 Jun;31(2):43-55
• Brooks J, Lloyd A, Ersser S, Ryan TJ. A quasi-experimental study to evaluate the effectiveness of a community based, structured, patient concordance programme to prevent the occurrence of venous leg ulcers. In press
• Brown MM, Boosinger J, Black J, Gaspar T, Sather L. Nursing innovation for dry skin care of the feet in the elderly: a demonstration project. J Gerontol Nurs 1982 Jul;8(7):393-5
• Campisi C 2000. Mayall's Syndrome Lymphology 2000 33 Suppl 81-89
• Casley-Smith JR, Clodius L, Piller NB. Tissue changes in chronic experimental lymphedema in dogs. Lymphology (1980) 13:130-141.
• Chen HI, Granger HJ and Taylor AE. Lymph flow transients following elevation of venous pressure in the dog hind paw. Lymphology (1991) 24:155-160
• Coleridge-Smith PD. Leg Ulcers: Biochemical Factors. Phlebology (2000) 15:156-161
• Constant JS, Feng JJ, Zabel DD, Yuan H, Suh DY, Scheuenstuhl H, Hunt TK, Zamirul H. Lactate elicits vascular endothelial growth factor from macrophages: A possible alternative to Hypoxia Wound Rep Reg 8 253-360 2000.
• Cremer G Special Granulomatous Forms of Mycosis on Lower Legs Caused by Trichophyton Rubrum Castellani. Dermatologica (Basel), (1953) 107, 28.
• Daley WJ, Krum Holz RA and Ross JC. The venous pump in the legs as a determinant of pulmonary filling capacity. J Clin Invest (1965) 44:271
• Dery MA, Yonuschot G, Winterson BJ. The effects of manually applied intermittent pulsation pressure to rat ventral thorax on lymph transport. Lymphology 2000 Jun;33(2):58-61
• Doherty DC, Moffatt CJ, Franks PJ. Mobility as a risk factor for leg ulcer healing. Abstract: 2001 Conference of European Wound Management Association
• Dreyer G, Medeiros Z, Netto MJ, Leal NC, de Castro LG, Piessens WF. Acute attacks in the extremities of persons living in an area endemic for bancroftian filariasis: differentiation of two syndromes. Trans R Soc Trop Med Hyg 1999 Jul-Aug;93(4):413-7
• Eisenbeiss C, Welzel J, Eichler W & Klotz K. Influence of body water distribution and skin thickness measurements using ultrasound. Br.J.Derm 2001 144 947-951.
• Falanga V, Eaglestein WH. The "Trap" hypothesis of venous ulceration Lancet 1993 341 1006-7.
• Foldi E, Sauerwald A, Hennig B. Effect of complex decongestive physiotherapy on gene expression for the inflammatory response in peripheral lymphedema. Lymphology 2000 Mar;33(1):19-23
• Foldi E. Prevention of dermatolymphangioadenitis by combined physiotherapy of the swollen arm after treatment for breast cancer. Lymphology 1996 Jun;29(2):48-9
• Gabel JC, Drake RE. Increased venous pressure causes increased thoracic duct pressure in awake sheep. J Appl Physiol 1992 Aug;73(2):654-6
• Gardner AMN. The return of blood to the heart. John Libbey (1993), London. pp69.
• Godoy J M P de, M de F G Godoy. A new approach to manual lymphatic drainage. Brazilian Camera Book, Lin Comunicado Sao Jose de Rio Preto - Sao Paulo Brazil 1999. ISBN 85-87. 384.01-3
• Godoy J M P & Gody M F G. Development and evaluation of an apparatus for lymph drainage. Europ J Lymphology 2001 9 121
• Hai M A. "Dry Skin Care" in: Nursing interventions: essential nursing treatments, WB Saunders (Philadelphia, PA) 1992 ed 2, pp 34-47
• Hardy MA. A pilot study of the diagnosis and treatment of impaired skin integrity: dry skin in older persons. Nurs Diagn 1990 Apr-Jun;1(2):57-63
• Hewitt J. The yoga breathing posture and meditation. London Random House 1983 89-91.
• Hu D, Phan TT, Cherry GW, Ryan TJ. Dermal oedema assessed by high frequency ultrasound in venous leg ulcers. Br J Derm 138 815-820.
• Jacobsson S. Blood circulation in lymphoedema of the arm. Br J Plast Surg (1967) Oct;20(4):355-8
• Karkkainen MJ, Petrova TV. Vascular endothelial growth factor receptors in the regulation of angiogenesis and lymphangiogenesis Oncogene 2000 19 5598-5605.
• Kim PI, Huh S, Huang JH, Kim YI, Lee BB. Venous dynamics in leg lymphoedema. Lymphology (1999) 32:11-14
• Kigler C, Strunk M, Rudofsky G. Venous pressure dynamics of the healthy human leg. Journal of Vascular Research (2001) 38:20-29
• Man M-Q, Wood L, Elias PM, Feingold R. Cutaneous barrier repair and pathophysiology following barrier diruption in IL-1 and TNF type 1 receptor deficient mice. Exp Dermatol 1999:8:261-266
• Martin KP, Foldi E. Are haemodynamic factors important in arm lymphedema after treatment of breast cancer? Lymphology (1996) 29:155-157
• Matthew S, Lau LL, Martin A, Lewis A, Hannon R J, Lee B, Young S, Soong CV. Venous ulceration is associated with less plasma concentration of endogenous antioxidants Phlebology abstrat 16 49 2001.
• Mellor RH, Stanton AWB, Menadue L, Levick JR & Mortimer PS. Cutaneous angiogenesis in post mastectomy oedema shown using flourescence angiography. J.Vascular research 2001 38 191 (abstract)
• Mortimer PS, Simmonds R, Rezvani M, Robbins M, Hopewell JW, Ryan TJ. The measurement of skin lymph flow by isotope clearance--reliability, reproducibility, injection dynamics, and the effect of massage. J Invest Dermatol 1990 Dec;95(6):677-82
• Murohara T, VEGF/ VPF enhances vascular permeability via NO and prostacyclin. Circulation (1998) 97:99-107.
• Ohkuma M. 2001. The skin blood flow change after a new lymphoedema therapy by magnetic fields, vibration and heat. Eur J. Lymphology 9, 83
• Olszewski WL, Jamal S, Lukomska B, Manokaran G, Grzelak I. Immune proteins in peripheral tissue fluid-lymph in patients with filarial lymphedema of the lower limbs. Lymphology 1992 Dec;25(4):166-71
• Pani SP, Srividya A. Clinical manifestations of bancroftian filariasis with special reference to lymphoedema grading. Indian J Med Res 1995 Sep;102:114-8
• Pardoe HD, Coleridge-Smith PD, Scurr JH. Is the Epidermis the source of the Angiogenic Stimulus in Chronic Vascular Disease? Phlebology (Abstract) 11 171 1996.
• Pestonjamasp VK and Gallo RL. Antimicrobial peptides are effectors of wound repair. Wound Repair and Regeneration 2001 9 158 (abstract)
• Pippard C, Roddie IC. Comparison of fluid transport systems in lymphatics and veins. Lymphology 1987 Dec;20(4):224-9
• Roberts CC, Stanton AW, Pullen J, Bull RH, Levick JR, Mortimer PS. Skin microvascular architecture and perfusion studied in human postmastectomy oedema by intravital video-capillaroscopy. Int J Microcirc Clin Exp (1994) Nov-Dec;14(6):327-34
• Ryan TJ. Exhange and the mechanical properties of the skin: Oncotic and hydrostatic forces control by blood supply and lymphatic drainage. Wound Rep Reg 1995;3:258:64.
• Ryan TJ, De Berker D. The interstitium, the connective tissue environment of the lymphatic and angiogenesis in human skin. Clinics in dermatology (1995) 13:451-8
• Ryan TJ. Warming and Wound Healing (Warm-Up ®) active wound therapy International Congress and Symposium series 257. Royal Society of Medicine. Editors Ryan TJ, Cherry GW, Harding KG 2000 P1-91.
• Ryan TJ. Lymphatic Filariasis: Management of Lymphoedema. Africa Health (2001) 23(3):9-11
• Schad H, Flowaczny H, Brechtelsbauer H, Birkenfeld G. The significance of respiration for thoracic duct flow in relation to other driving forces of lymph flow. Pflugers Arch 1978 Dec 28;378(2):121-5
• Shenoy RK, Sandhya K, Suma TK, Kumaraswami V. A preliminary study of filariasis related acute adenolymphangitis with special reference to precipitating factors and treatment modalities. Southeast Asian J Trop Med Public Health 1995 Jun;26(2):301-5
• Shenoy RK, Suma TK, Rajan K, Kumaraswami V. Prevention of acute adenolymphangiitis in Brugian Filariasis: comparison of the efficacy of Ivermectin and diethylcarbamazine, each combined with local treatment of the affected limb. Annals Trop Med & Parasitology 1998 92 587-594.
• Shenoy RK, Kumaraswami V, Suma TK, Rajan K, Radhakuttyamma G. A double-blind, placebo-controlled study of the efficacy of oral penicillin, diethylcarbamazine or local treatment of the affected limb in preventing acute adenolymphangitis in lymphoedema caused by brugian filariasis. Ann Trop Med Parasitol 1999 Jun;93(4):367-77
• Shimotoyodone A, Okajima M, Souno H, Kobayashi H, Hase T, Fujuimura A, Nosaka Y. Facilitation of the gingival lymphangial function by heat and massage and improvement of the Gingivitis condition
• Shoab SS, Porter J, Scurr JH, Coleridge-Smith PD. Increased plasma vascular endothelial growth factor in patients with chronic venous disease. A prospective controlled study (abstract) Phlebology 12 159 1997.
• Snowden K, Hanmerberg B. Vascular patterns in the Filaria Infected canine limb. Lymphology (1986) 19:77-78
• Solti F. Haemodynamic changes in extremital lymphedema. In: Ergebruisse der Angiologie Schahauer, FR (Ed.), Suttgart-New York, 1976, pp12
• Spruit D. The interference of some substances with the water vapour loss of human skin. Dermatologica 1971;142(2):89-92
• Sulzberger MB and Baer RL. "Treatment of Pyodermas common pus forming infectinos of the skin" 1950 Yearbook of dermatology and syphilology ed. Sulzberger and Baer, pp9-52. Yearbook publishers, Chicago (1950).
• Svedman C, Cherry GW, Ryan TJ 1998. The veno-arteriolar reflex in venus leg ulcers. Patients studied by laser doppler imaging. Acta Derm Venereol (Stockholm) 1998 78 258-261.
• Svensson WE, Mortimer PS and Tohno E. Colour Doppler demonstrates venous flow abnormalities in breast cancer patients with chronic arm swelling. European Journal of Cancer (1994) 30:657-660
• Svensson WE, Mortimer PS, Tohno E. The use of colour Doppler to define venous abnormalities in the swollen arm following therapy for breast carcinoma. Clin Radiol (1991) 44:249-252
• Swedborg I, Norrefalk JR, Piller NB, Asard C. Lymphoedema post-mastectomy: is elevation alone an effective treatment? Scand J Rehabil Med 1993 Jun;25(2):79-82
• Tibbs DJ, Varicose Veins and Related Dissorders. (1992) Butterworth Heinemann, London. pp492
• van der Veen P, Kempenaers F, Vermijlen S, Van Waeyenberghe C, Kerckhofs E, Bossuyt A, Van den Brande P, Lievens P. Electromagnetic diathermia: a lymphoscintigraphic and light reflection rheographic study of leg lymphatic and venous dynamics in healthy subjects. Lymphology (2000) Mar;33(1):12-8
• Walker WF, and Pickard C. Respiration and the Inferior Vena Cava. Bull int Soc Chir (1962) 3:257