Mechanism of action of sclerosing agents
and rationale for selection of a sclerosing solution

Copyright 2002 by Craig F. Feied, MD, FACEP

Goals of sclerotherapy

When we treat varicosities and telangiectasias, we want to remove or obliterate the abnormal vessels that carry retrograde flow, without damaging adjacent or connected vessels that carry normal antegrade flow. Obliterating a vessel is not easy: a small amount of damage will produce intravascular thrombus, but thrombosis alone usually does not result in obliteration of the vessel. Intact endothelium aggressively lyses thrombus, and a thrombosed vessel with intact endothelium will not be sclerosed.

Recanalization of thrombosed vessels

Vascular fibrosis and obliteration only occurs in response to irreversible endothelial cellular destruction and exposure of the underlying subendothelial cell layer.

If an injected sclerosant is too weak, there may be no endothelial injury at all. If the sclerosant is a little stronger, the varicose vessel is damaged, but recanalization occurs and an incompetent pathway for retrograde blood flow persists. If the injected sclerosant is too strong, the varicose vessel endothelium is destroyed, but the sclerosant also flows into adjacent normal vessels and causes damage there as well. The key goal is to deliver a minimum volume and concentration of sclerosant that will cause irreversible damage to the endothelium of the abnormal vessel to be sclerosed, while leaving adjacent normal vessels untouched. It is important to protect normal superficial vessels, and it is critically important to avoid injuring the endothelium of deep veins, because deep vein thrombosis places patients at risk of death from thromboembolism, as well as causing permanent disability from chronic deep venous valvular insufficiency. The rational treatment of varicosities and telangiectasias by chemical sclerosis depends upon our ability to produce vascular endothelial damage that is irreversible in the area under treatment, but that does not extend to adjacent normal vessels.

To limit endothelial injury to a controlled area, we exploit differences in flow dynamics between the abnormal veins being perfused with sclerosant and the adjacent normal vessels that should not be sclerosed. A thorough understanding of the mechanism of action of the principal sclerosing agents is essential, as is a firm grasp of the biophysical principles underlying the techniques of sclerotherapy.

Volume dilution and patient positioning

Sclerosant is diluted with blood as it diffuses away from the site of injection, thus if a strong sclerosant is injected there will be three zones of action. In zone 1, vascular endothelium is irreversibly injured: the vessel will be fully sclerosed and eventually will be completely replaced by a fibrous tissue. In zone 2, vascular endothelium is injured, and the vessel will be partially or completely thrombosed but will eventually recanalize. In zone 3 the sclerosant will be diluted below its injurious concentration, and there will be no endothelial injury.

Dilution by diffusion from the injection site

Because dilution of the sclerosant with blood occurs immediately upon injection, the original injected concentration is of no real importance. What is important is the diluted concentration of sclerosant at the surface of the endothelium. An injected concentration that is perfectly effective in a spider vein (where sclerosant displaces blood rather than mixing with it) may be ineffective in a reticular feeding vein or a truncal varix simply because dilution reduces the final concentration so low that there will be no endothelial injury whatsoever (no zone I or zone II). If the injected concentration is too high, dilution will leave the final concentration so high that endothelial damage will occur where it is not wanted (zone I and zone II are too large). If the injected concentration is just right, dilution will leave a final concentration that is sufficient to injure the local varicose endothelium, but not high enough to damage normal superficial or deep veins (most of the varicose vessel falls into zone I, a small amount falls into zone II, and all normal vessels fall into zone III).

When we select a particular volume and concentration of a chemical agent with which to sclerose a vessel, we are explicitly or implicitly adjusting the injected concentration and volume to take into account the dilution that will occur when the sclerosant is mixed with blood immediately after injection. We also must take into account the further dilution that will occur as the sclerosant flows or diffuses away from the site of injection. The importance of patient positioning in determining dilutional volume often is not properly appreciated by the novice in phlebology.

Because of the cylindrical geometry of blood vessels, the volume contained in a vessel depends on the square of the vesel radius: the volume of any cylinder is calculated as pi * r2 * L (where r is the radius and L is the length of the vessel). Vessels collapse to a smaller radius when the legs are elevated, thus the volume contained is reduced dramatically. For this reason, the position of the patient has a very powerful effect on the final diluted concentration of sclerosant at the surface of the vessel endothelium.

Effect of position on varicose geometry

Standing

For a standing patient with a superficial varicosity of 2 cm in diameter, the final concentration at a distance from the injection site of 10 cm (4 inches) is 30 times lower than the initial concentration. Doubling the initial concentration serves only to double the final concentration, which will still be 15 times weaker than the concentration in the syringe. In other words, if 1 cc of a 3% solution is injected, the final concentration at the endothelial surface is 1% at a distance of 1 cm from the injection point, 0.5% at a distance of 2cm, 0.25% at a distance of 4 cm, and 0.2% at a distance of 5cm (2 inches) from the injection point. As we shall see, this means that it is very difficult to achieve sclerosis of a large vessel by injecting detergent sclerosants with the patient in a standing position: if the highest available concentration is injected, the dilution factor may still drop the final concentration below the threshold of effectiveness within 1.5 inches from the injection site.

Supine

What about the supine position? Varicose vessels that bulge when the patient is standing may collapse when the patient is supine, but duplex ultrasound readily demonstrates that the veins are not empty of blood. Both varicose and normal vessels contain a significant volume of blood with the legs extended in the supine position. A bulging varicosity that has a diameter of 2 cm in the standing position may have a diameter of 1 cm in the supine position and of 0.5 cm or less when the legs are elevated as high as possible. With such a patient in the supine position, injection of 1 cc of a 3% solution leads to a final concentration of approximately 1.7% at a distance of 1 cm and a concentration of about 0.6% at a distance of 5 cm (2 inches). This supine technique limits dilution enough to allow successful sclerosis of large vessels using detergent solutions, so long as sufficient concentrations and volumes of sclerosants are injected. The only problem is that if an injection of sclerosant at a high initial concentration is made directly into a perforating vessel, so that sclerosant flows directly into the deep system, dilution within the deep vessel will still permit zone I and zone II endothelial injury for a short distance within the deep vein. This can lead to deep vein valve damage and chronic venous insufficiency, to deep vein thrombosis, and to life-threatening pulmonary embolism.

Legs elevated

In contrast to the standing and supine positions, when a patient lies supine and the legs are raised vertically so that they are well above the central circulation, most superficial varices collapse to the point where they no longer contain any significant volume of blood. Repeating the calculation above for a patient in this position, injection of 1cc of a 3% solution leads to a final concentration of 2.5% at a distance of 1 cm from the injection, and a final concentration of 1.6% at a distance of 5 cm (2 inches). In fact, the final concentration will still be above 1% at a distance of 10cm from the injection site. Because the superficial varicosity is collapsed, there is very little dilution with distance so long as the sclerosant stays within the floppy-walled varicosity.

Graph: volume of dilution vs distance from injection site

What happens when sclerosant passes through into normal vessels? Although flow measurements reveal little or no spontaneous flow through varices and smaller superficial veins when the patient is in the leg-up position, a substantial intravenous volume and a substantial rate of flow still persists in the deep veins and in normal larger superficial veins, which have less collapsible walls. This difference between the volumes contained in floppy superficial varicose veins and the volumes contained in normal surface veins and deep veins may be exploited to cause damage that is almost perfectly localized to superficial varices. If an elevated, empty varicose vessel is perfused with a concentration of sclerosant so low that even without dilution it is just barely sufficient to cause endothelial injury, then any further dilution will reduce the concentration below the threshold of injury. Because larger superficial vessels and deep vessels continue to carry a volume of blood in the leg-up position, any sclerosant passing into these vessels will immediately be diluted to a safe and noninjurious concentration, sparing the endothelium of vessels that we wish to preserve. Injection of this 'threshold' concentration directly into a perforating vein (or even directly into a deep vein) will not cause any deep vein injury.

Types of sclerosants

Virtually any foreign substance can be utilized to cause venous endothelial damage. Historical methods for producing venous endothelial trauma have included 'a slender rod of iron', reportedly used by Hippocrates himself, absolute alcohol, introduced by Monteggio and by Leroy D'Etoilles in the 1840s, and ferric chloride, introduced by Charles-Gabriel Pravaz in 1851. In 1910 it was noted that the injection of antisyphilitic mercurial drugs caused obliteration of antecubital veins, and these agents were adapted for use in sclerotherapy. There were many other drugs and techniques used for venous sclerosis during these years, but all of them suffered from one or another problem that made them unacceptable for modern use. Early sclerosing agents caused many deaths from sepsis and from pulmonary embolism, as well as a high incidence of allergic reactions, local tissue necrosis, pain, and failed sclerosis.

The perfect sclerosant

The best imaginable sclerosant would have no systemic toxicity. It would be effective only above some threshold concentration, so that its effects could be precisely localized through dilution. It would require a long period of contact to be effective, so that it would be relatively more effective in areas of stasis and relatively safer in the deep veins where there is high flow. It would be non-allergenic. It would be strong enough to sclerose even the largest vessels, yet it would produce no local tissue injury if extravasated. It would not cause staining or scarring. It would not cause telangiectatic matting. It would be perfectly soluble in normal saline. It would be painless upon injection. It would be inexpensive. It would be approved by the United States Food and Drug Administration.

No currently available sclerosant possesses all of the attributes of the perfect sclerosing agent. All currently available sclerosants fall short in one way or another, yet the variety of available agents is such that virtually every situation in which sclerotherapy is indicated can be safely and effectively handled by one or another of the available sclerosants, used alone or in combination.

Detergents

In the 1930's the class of drugs known as detergents, or as fatty acids and fatty alcohols, came into use with the introduction of sodium morrhuate and sodium tetradecyl sulphate. Detergent sclerosants work by a mechanism known as protein theft denaturation, in which an aggregation of detergent molecules forms a lipid bilayer in the form of a sheet, a cylinder, or a micelle, which then disrupts the cell surface membrane and may steal away essential proteins from the cell membrane surface.

Self-assembly of a cylindrical macroaggregate of polidocanol

The loss of these essential cell surface proteins causes a delayed cell death: when endothelial cell membranes are exposed to detergent micelles, irreversible cellular morphological changes are seen within minutes by scanning electron microscopy, but the fatal cellular changes that are visible by normal light microscopy do not become apparent for many hours. Unlike many other agents, the detergent sclerosants do not cause hemolysis, nor do they provoke direct intravascular coagulation.

Determinants of activity of detergent solutions

1) Concentration

At low concentrations, most detergent molecules are individually dissolved in solution, and there are very few micellar aggregates. When the concentration reaches some threshold (known as the critical micellar concentration, or CMC) nearly all further detergent molecules added to the solution will enter into micelles. Micelles can cause protein theft denaturation, but individual detergent molecules have no toxicity to the vascular endothelium, thus for each detergent sclerosant, there is some threshold concentration below which the agent causes no injury. This physical property means that detergent sclerosants offer significant benefits over most of the agents previously used, because they are potent agents that nonetheless have a clear-cut threshold below which they have absolutely no injurious effect on venous endothelium.

2) Temperature

The solubility of detergents is inversely temperature dependent. Because of the highly polar nature of water, and the entropic dependence of the hydrophobic effect, detergent molecules are much more soluble in cold solutions than in hot ones. This effect is easily seen in everyday life: dishwashing detergent produces a large amount of persistent foam in warm water, while cold water rinses away the soapy foam easily. The solubility of sclerosing agents such as polidocanol is likewise much higher in cold solutions, and because single dissolved molecules are ineffective, the strength of the sclerosing effect is higher at warmer temperatures.

3) Mixing

Detergent micellar formation can reach a maximum level based upon the temperature and upon the concentration of the detergent in solution. Micellar formation is a steric process, however, and the geometry of macroassemblages often prevents maximal micellar formation. The surface area of lipid bilayer structures such as sheets, cylinders, and micelles is maximized when the solution is shaken to produce a foam. Because it is the surface of these structures that causes protein theft denaturation, a solution that has been shaken will be a more effective sclerosant than one that has not. Unfortunately, foamy bubbles that are injected into spider veins or varicose veins can pass through a patent foramen ovale to lodge in the ocular and cerebral circulation, where they have produced temporary ischemic attacks with temporary blindness and other central nervous system effects.

Currently available detergent agents

A) Sodium morrhuate

This detergent sclerosant is made up of a mixture of saturated and unsaturated fatty acids extracted from cod liver oil. It was introduced in 1920's, and is still available today. Because it was in general use before there was any requirement to demonstrate safety or efficacy it has been exempted from the need for approval by the Food and Drug Administration (FDA) for sale in the United States, but there are several problems with the product that make it a less than ideal agent for sclerotherapy. It is a biological extract rather than a synthetic preparation, and the composition varies somewhat from lot to lot. Its components have been incompletely characterized, and a significant fraction of its fatty acids and alcohols are of chain lengths that probably do not contribute to its effectiveness as a sclerosant. It is unstable in solution, causes extensive cutaneous necrosis if extravasated, and has been responsible for many cases of anaphylaxis.

B) Ethanolamine oleate

Ethanolamine oleate, a synthetic preparation of oleic acid and ethanolamine, has weak detergent properties because its attenuated hydrophobic chain lengths make it excessively soluble and decrease its ability to denature cell surface proteins. High concentrations of the drug are necessary for effective sclerosis, and its effectiveness in esophageal varices depends upon mural necrosis. Allergic reactions are uncommon, but there have been reports of pneumonitis, pleural effusions, and other pulmonary symptoms following the injection of ethanolamine oleate into esophageal varices. Like sodium morrhuate, this agent was exempted from the need for approval by the Food and Drug Administration (FDA) for sale in the United States. The principal disadvantages of the drug are a high viscosity that makes injection difficult, a tendency to cause red cell hemolysis and hemoglobinuria, the occasional production of renal failure at high doses, the possibility of pulmonary complications, and a relative lack of strength compared with other available sclerosants.

C) Sotradecol

Sodium tetradecyl sulfate (Sodium 1-isobutyl-4-ethyloctyl sulfate) is a synthetic long chain fatty acid that has seen extensive industrial use as a synthetic surfactant (soap). It is sold for medical use as a solution of up to 3% concentration with 2% benzoyl alcohol used as a stabilant. It is effective as a venous sclerosing agent in concentrations from 0.1% to 3%. Like sodium morrhuate and ethanolamine oleate, it was 'grandfathered' by the Food and Drug Administration (FDA) for sale in the United States, but unlike sodium morrhuate, sotradecol has proven to be a reliable, safe and effective sclerosant. The principal clinical problems with the drug are a tendency to cause hyperpigmentation in up to 30% of patients, a significant incidence of epidermal necrosis upon extravasation, and occasional cases of anaphylaxis.

D) Polidocanol

Polidocanol (hydroxy-polyethoxy-dodecane) is a synthetic long-chain fatty alcohol sold under many trade names (Sclerovein, Aetoxysclerol, Aethoxysklerol, Etoxisclerol, Sotrauerix, Laureth 9).

Structure of polidocanol monomer

All commercially available formulations contain some small quantity of ethanol. The drug was originally developed and marketed in the 1950s under the name Sch 600 as a non-amide, non-ester local anaesthetic that was useful for injected local anaesthesia as well as for epidural anaesthesia and for topical mucosal anaesthesia. It was first used as a sclerosing agent in Germany in the 1960's, and was quickly adopted for that use in most countries. The drug is not yet approved by the FDA for sale in the United States as a sclerosing agent, but is nonetheless widely used because it offers certain advantages over many other available drugs. As a local anaesthetic, Polidocanol is painless upon injection. It does not produce necrosis if injected intradermally, and has been reported to have a very low incidence of allergic reactions. The drug has been intensely studied and extremely well characterized, and has a high therapeutic index. The LD50 in rabbits is 200 mg/kg (approximately 5 times greater than that of novocaine), and the LD50 in mice is even greater, at 1200 mg/kg. For human use the German manufacturer of polidocanol recommends a maximum daily dose of 2 mg per kg, although at least one author has reported the routine use of much higher doses.{1353} For all its advantages, polidocanol is not without problems as a sclerosant. Occasional anaphylactic reactions have been reported. In some patients it may produce hyperpigmentation, although to a lesser extent than many other agents. Telangiectatic matting after sclerotherapy with polidocanol is as common as with any other agent.

E) Scleremo

Scleremo, a compound of 72% chromated glycerin, is a polyalcohol that often is considered a chemical irritant sclerosant. It is classified here with the detergents because it is similar to the detergents in the way it causes cell surface protein denaturation. It is very popular in Europe, but it has not been approved by the FDA for use in the United States, where it is virtually unknown. Compared to other sclerosants it is a very weak sclerosant (it is approximately 1/4 the strength of Polidocanol at the same concentration and volume) and is principally useful in the sclerosis of small vessels. Its principal advantage is that it rarely causes hyperpigmentation or telangiectatic matting, and that it very rarely causes extravasation necrosis. The main problems with scleremo are that it is hard to work with because it is extremely viscous, that it can be quite painful on injection, that the chromate moiety is highly allergenic, and that it has occasionally been reported to cause ureteral colic and hematuria.

Hypertonic and ionic solutions

Strong solutions of hypertonic saline and other salt solutions are part of a class of solutions that are often referred to as osmotic sclerosants. These solutions have long been regarded as causing endothelial death by osmotic cellular dehydration. Although it is true that osmotic dehydration at the point of injection is sufficient to rupture red blood cells and to dehydrate some nearby endothelial cells, the evidence suggests that these sclerosants are effective even after dilution has reduced the osmotic gradient far too low to account for the effects seen. Thermodynamic and physical chemical calculations suggest that these and other strong ionic solutions probably work by causing conformational denaturation of cell membrane proteins in situ. Like the detergents, they can be diluted to the point where they have no further cellular toxicity.

A) Hypertonic saline

Hypertonic solutions of saline became popular agents for sclerotherapy after they were adopted for that use by Linser in 1926. The most common preparations are a 20% or 23.4% solution. The principal advantage of the agent is the fact that it is a naturally occurring bodily substance with no molecular toxicity. It has not been approved by the FDA for use in sclerotherapy, but it has been used successfully for that purpose by several generations of physicians. There are several reasons why it is not universally accepted as a desirable sclerosing agent. Because of dilutional effects, it is difficult to achieve adequate sclerosis of large vessels without exceeding a tolerable salt load. It can cause significant pain on injection, and significant cramping after a treatment session. If extravasated, it almost invariably causes significant necrosis. Because it causes immediate red blood cell hemolysis and rapidly disrupts vascular endothelial continuity, it is prone to cause marked hemosiderin staining that is not very cosmetically acceptable. All of these problems can be overcome to some extent by meticulous technique and with experience, but patient satisfaction remains lower than with some other available agents. In an effort to reduce the complications, hypertonic saline has been mixed with procaine and heparin in a compound known as Heparsol. This approach has not proven effective, and is rarely used today.

B) Sclerodex

Sclerodex is a mixture of 25% dextrose and 10% sodium chloride, with a small quantity of phenethyl alcohol. A primarily hypertonic agent, its effects are similar to those of pure hypertonic saline, but the reduced salt load offers certain benefits. It is not approved by the FDA for sale in the United States. Like pure hypertonic saline, it is somewhat painful on injection, and epidermal necrosis continues to be the rule whenever extravasation occurs.

C) Polyiodinated iodine

Polyiodinated iodine (Variglobin, Sclerodine) is a mixture of elemental iodine with sodium iodide, along with a small amount of benzyl alcohol. It is rapidly ionized and rapidly protein-bound when injected, and most likely works by localized ionic disruption of cell surface proteins in situ. In vivo conversion of ionized iodine to iodide renders the solution ineffective as a sclerosant, thus localizing the sclerosing effects to the immediate area of injection. The agent is not approved by the FDA for sale in the United States, but is widely used in Europe. The problems with this agent are its high tendency to cause extravasation necrosis, its limited effectiveness at a distance from the injection site, and the risks of anaphylaxis and of renal toxicity that are associated with ionic iodinated solutions.

Cellular toxins

Other chemical sclerosants exist that probably act by a direct or indirect chemical toxicity to endothelial cells: by poisoning some aspect of cellular activity that is necessary for endothelial cell survival. Such agents are less useful to the extent that they also poison other bodily cells. They also lack another of the key attributes of a good sclerosant: they remain toxic to some degree even after extreme dilution, so that there is no real threshold below which injury will not occur.

Summary

The guiding principle of modern sclerotherapy is to cause irreversible endothelial injury in the desired location, while avoiding any damage to normal vessels that may be interconnected with the abnormal vessel we are treating. Our aim is to deliver the minimum volume and minimum concentration of the most appropriate sclerosant, and to inject it under conditions that will achieve the minimum effective exposure. Sclerosant concentration, volume, temperature, mixing, and patient positioning are more important in this endeavor than the choice of the actual sclerosing agent. With attention to these details, an accomplished phlebologist can achieve good results with virtually any currently available sclerosing agent.