HOME            FEEDBACK

CONTINUOUS  COMPRESSION  TECHNIQUE

OF  INJECTING  VARICOSE  VEINS

MB  MCh Dubl  FRCSI

Varicose veins have been treated by injection for more than a century. As early as 1835 patients were admitted to the surgical wards of the Massachusetts General Hospital for injections of their varicose veins (Linton 1963), but since then the popularity of this form of treatment has waxed and waned. The final results in the early cases were never fully assessed, but later surveys (Howard et al. 1931, Robertson 1947, Fenney 1951) showed that in many patients the end-result was infinitely worse than the pre-existing condition; the segment thrombosed by the injection almost always recanalised within a few months. The really serious objection to injection therapy, however, stems from the uncontrolled spread of thrombosis to unaffected valves in adjacent veins, thereby injuring them and reducing the efficiency of the peripheral pump (Agrifoglio and Edwards 1961).

The clinical reaction varies greatly. Sometimes the widespread thrombosis is almost terrifying, whereas on other occasions there is no evidence of any reaction at all. Less commonly, the injection results in a spastic, cord-like, painless short segment of vein. This led me to think that the vein that went into spasm and remained in spasm would be less likely to recanalise, and it was as an investigation of this notion that this work began.

Some years ago I realised (Fegan 1953) that the application and maintenance of pressure after injection could produce an inert cord-like vein. On the other hand, when pressure was omitted the leg became swollen and painful - the reaction resembling in every way a segment of thrombophlebitis. The consequences were the same if the pressure was released, even temporarily, before the occluded segment had finally become organised. To elucidate the histopathological changes, biopsy specimens obtained from a large series of our patients at various intervals after treatment were compared with specimens from patients who had traumatic thrombophlebitis.

In all, 135 specimens were examined. The extensive fluctuations of hydrostatic pressure in the unthrombosed segment of the vein act as a water hammer on the thrombosed segment, gorging a line of cleavage between the retracting clot and the vein wall. Unless compression is applied, slit-like channels are formed (fiig.10 at the same time as the thrombus is being actively invaded by thromboblasts and capillaries at its point of "firm2 attachment to the vein wall. The clot shrinks; the slit-like channels merge (fig.1); and the vein reopens. The capillaries arise from the vasa vasorum and can be seen to transgress the internal elastic lamina. Contrary to expectation, I found that the more pronounced the capillary fibroblastic invasion, the less was the likelihood of the vein reopening. In no specimen was any direct communication detectable between the slit-like channels and the arteriovenous network arising from the vasa vasorum. Obviously, therefore, the more uniform and complete the invasion of the thrombus by the fibroblastic capillary network, the less likely is the formation of channels and the consequent reopening of the vein. Thus, the maintenance of uninterrupted compression after injection is essential to counteract the hydrostatic pressure in the adjacent patent veins and to limit the formation of channels.

The successful treatment of varicose veins depends on restoration, if possible, of the peripheral pump (Dodd and Cockett 1957), which has considerable reserve power and which, in this respect resembles other major functional units in the body. Only when the cumulative effect of repeated injury exceeds this reserve is the pressure in the superficial veins not rhythmically reduced during movements of the leg. As a result, the musculo-elastic wall of the vein cannot recover its tone, and the vein becomes varicose. Hence, varicose veins are not an early manifestation of a minor development of the pump; and to cure them one does not have to remove every varicose vein in the leg.

The injection has hitherto been confined to the obvious varicosities and physiological circumstances have seldom been taken into account. Little, if any, attention has been paid to restoring the efficiency of the pump. This can be achieved by: (1) localisation of the significant leaks; and (2) permanent control of the leaks, with avoidance of damage to normally functioning valves and to veins capable of recovering their normal function.

Abnormally dilated veins can contract and become normal when the abnormal strains are removed. This is commonly seen in the puerperium, and in my clinic I have observed it following one well-placed injection. Because of this, I am opposed to the principle of removing large segments of veins by stripping.

13,352 patients have now been treated; of these 8150 have been treated by me, and a further 5102 by my registrars. There has been no selection of cases, and a large proportion of the patients had one or more of the complications associated with varicose veins (table 1). 2,760 patients were pregnant when first seen. Because adequate attention to varicose veins during pregnancy is essential for preventing severe varicosities during later life, pregnancy has not been regarded as a contraindication.

The most satisfactory results are obtained in those patients in whom the leaks - i.e., the incompetent perforating veins - have been accurately defined. To ensure that there is enough time for tracing the venous patterns of each patient, new patients are dealt with at separate appointments.

A complete history is taken, with particular note of any previous injury to the leg, superficial thrombophlebitis, or deep-vein thrombosis. A history of deep-vein thrombosis is not regarded as a contraindication to treatment. Previous treatment, such as ligation and stripping, is noted. If there is ulceration or any suggestion of nutritional deficiency, haemogloblin, blood-film, urine, and Wasserman examinations are carried out.

The legs are inspected with the patient erect, and visible dilated veins are marked with a blue skin pencil. The leg is nest palpated circumferentially, and other palpable but less obvious high pressure veins are pencilled in. With the patient still erect, a dilated vein is repeatedly percussed while the opposite hand palpates the surface of the leg. This combination of percussion and palpation often brings to light a hidden high pressure superficial vein, which could be missed by inspection or palpation alone. If such a vein is present, it is marked with the skin pencil. The mapping procedure, without short cuts, is essential, since the pattern when fully traced helps greatly to elucidate the diagnosis.

The patient now lies down, and the leg is elevated to a right angle and remains thus for one to two minutes. Preferably with the eyes closed to increase tactile sensitivity, the surgeon next palpates the leg with the tips of the fingers of both hands. The fascial orifices of the incompetent perforating vein are more readily identified by palpating the raised leg. It is important not to look at the leg, because can be influenced by the marking of the superficial veins. As the palpating fingertips repeatedly skirt up and down the leg, weaknesses or orifices in the fascia covering the soft flabby muscle bags reveal themselves as definitely as the inguinal orifices of the abdominal wall. The surgeon rings them with skin pencils of different colour. Fingers are then introduced into each orifice, and the patient is told to stand. The fingers are now withdrawn one by one from the orifices, and the filling of the venous complex by the leaking vein from the deep-vein pump system is easily observed. This point is marked with a skin pencil of different colour.

It is to this segment of vein that attention is directed, and if permanent and effective obstruction can be produced at this point the remaining veins will often recover much of their tone; functionless valves restore their function; and the vicious cycle will be reversed. Indeed sapheno-varices have been found to disappear after injection of an incompetent hunterian communicating vein (Fegan 1961). How this comes about is illustrated in fig. 2.

In many patients, one finger placed in the correct fascial orifice will control the filling of all the veins, and one seldom has to apply three or four points of compression to control filling of the venous complex, however extensive this may be. This observation has led me to question the importance usually attached to the operation of "flush ligation" of the long saphenous vein at its termination. In 13,000 patients treated by us, none has had to have any treatment at the saphenous opening. In all, the highest point of treatment has been the upper hunterian communicating vein.

During the past twelve years the following principles have governed treatment in this clinic:

1. Treatment should be simple.
2. The mortality should be nil.
3. The morbidity should be as low as possible.
4. The cost of patient's time and stay in hospital should be as small as possible.
5. The treatment should be painless.
6. The beneficial effect should be permanent or of long duration.
7. The treatment should have a sound scientific basis.
8. It should be applicable to a large series of unselected patients, with proper documentation and follow-up examination for at least five years.
9. The treatment should work as well in the hands of other people as in the hands of the originator.

The method described here fulfils these requirements. When the point of incompetence of the vein has been found, the patient sits upright on a couch, and a needle is introduced into the affected segment of vein. Aspiration of a few drops of blood confirms that the needle is in the vein. While syringe is held steady against the leg, the patient lies down and elevates the leg to empty the veins. A finger is now placed on each side of the needle along the course of the vein, and the desired segment is isolated by digital pressure. The distance between the fingers should be approximately 5 cm. Sodium tetradecyl (0.5 ml. Of a 3% solution) is then injected, and the needle is withdrawn. Finger pressure is maintained for thirty to sixty seconds after the injection, and the leg is then bandaged from the sight of the injection to the toes.

The fingers are removed only as the bandage is applied over the points of pressure. This procedure must be carried out with care, because, to avoid injury to adjacent competent valves, it is important that the reaction should be confined to a short segment of vein. A pad of sorbo rubber is placed over the area of injection, and a further bandage is applied. The leg and bandage are next covered with an elastic stocking, which the patients are instructed to wear continuously, day and night, until their next attendance.

The correct application of bandages is sufficiently important to merit a detailed description of the technique used. Compression must be adequate - that is, sufficient to compress the vein without interfering with the arterial supply to the leg. If there is any suggestion of previous arterial inadequacy great care must be exercised.

The method of support provides three types of compression:

1. Specific local compression by the rubber pad.
2. Isometric compression by the bandages.
3. Isotonic compression by the elastic stocking.

The importance of these three types of compression is only appreciated when it is realised that the leg may either swell or shrink during the interval between visits. The isometric compression is important while the leg is swelling, and the isotonic action of the stocking is vital for maintaining the compression after swelling disappears.

The bandages are applied in such a way as to take into account the contours of the various parts of the leg. The fan-shaped thigh is covered by a folded bandage, the knee by a figure-of-eight (crossing at the back), the spindle-shaped leg by trellis bandage, and the ankle by a figure-of0eight crossing anteriorly. A sorbo rubber pad in the popliteal space prevents abrasion of the skin by the bandage. The upper edge of the bandage on the thigh is prevented from rolling downwards by incorporating a small pad of sorbo rubber projecting slightly above the edge.

Crêpe bandages without rubber fibre are applied at a tension just enough to render firm the soft fat and muscles. The bandages are allowed to follow the contour of the leg so that both edges are applied with equal force. The bandaging is not considered to be complete until the leg is completely covered from the most proximal site of injection to the roots of the toes, and until it is possible to run the hand up and down the leg and feel a firm leg with smooth contour without the bandages becoming disturbed in the process.

A full-length two-way stretch elastic stocking with rubber fibre is pulled over the bandages and supported by a suspender belt. In men a webbing belt with attached suspenders is used.

The importance of walking immediately and the avoidance of standing is impressed on the patient. If the bandage loosens, the patients are told to lie down immediately and to have it re-applied with the leg raised.

After a week, the patient is re-examined, and any remaining leaks are dealt with in an identical manner. The use of multiple injections at the first visit greatly reduces the number of attendances needed and yields superior results. The injections given at the same time have been found to give better results than three injections given on separate days. It is easy to inject at the site of the ankle perforating vein and to apply compression. It is then possible to move up to the tibial perforator and to inject at this site and again apply compression. Finally, one can move on to the knee and pick up the geniculate perforator or higher up to one of the lower hunterian perforators; and having injected these, to then apply bandages and rubber compression. It should be made clear that we do not use a tourniquet, and that with very little experience it is as easy to inject the empty, or near-empty, vein as the distended vein. The pressure is maintained for six weeks after the final injection.

To obtain the best results, treatment in a large clinic devoted exclusively to this disorder is necessary.

On re-examination of the injection site after a week, a small, hard, non-tender segment of vein will be found. Microscopic examination has shown that this is the ideal result, and for purposes of description it has been called "endosclerosis" to distinguish it from local thrombophlebitis. Maintenance of compression for at least six weeks permits complete organisation of the endosclerotic segment. In a review of 700 unselected patients (table II) treated here over a period of six years 81.6% were considered to have satisfactory results (table II). In the years before our clinic at the Rotunda hospital was devoted to the injection treatment there was a special ulcer clinic. All patients attending were pregnant and had ulcers during pregnancy, we have now not a single ulcer case. By contrast, here are some of the results reported after surgical treatment. Mathieson (1953) found that 82% of 293 patients showed improvement on follow-up after three to eight years; 38% were completely satisfactory. Arenander (1960) reported that 80% of 107 patients seen two to five years after operation showed complete recovery. Agrifoglio and Edwards (1961) had excellent results in 71% of 416 patients who were seen from one to fourteen years after operation.

If the compression is interrupted after injection, owing either to inadequate application or to interference by the patient, a tender nodule becomes palpable at the site of injecton. This is referred to as "localised thrombosis". Occasionally the reaction is more widespread; but with proper maintenance of pressure the subsequent result will be satisfactory, although the time required is lengthened by several weeks. The period of compression is proportional to the amount of thrombus which has to become organised.

The accidental extravascular injection of sclerosant is not likely to be serious, provided that the amount injected is small (0.25 ml.), and that it is kept well away from periosteum, nerve, or skin. If the sclerosant is inadvertently injected subcutaneously, a small area of skin immediately over the site of injection may slough. The likelihood of this happening is greatly reduced if, as described, a sorbo rubber pad is immediately applied over the injection site.

The possible extension of sclerosis into deep veins is generally regarded as a major drawback to the treatment of varices by injection. With our technique the sclerosant is confined for a long time to the selected segment of vein. This has been verified by injecting radio-opaque material with the sclerosant and subsequent radiography of the patient’s leg. There has been no case of clinically recognisable deep-vein damage with this technique, nor has pulmonary embolism been noted.

After the treatment of venous insufficiency in patients with intermittent claudication, the claudication pain disappeared completely in some cases, and the arterial blood-supply of skin and muscle improved in all.

Gross obesity renders maintenance of compression extremely difficult. It is now established practice in this clinic to insist on weight reduction to within 10% of the normal based on age and height for each patient. Also, patients are more likely to make serious attempts at weight reduction if treatment, apart from elastic stockings, is refused until they reach the required standard.

A technique of sclerotherapy based on physiological considerations has been used successfully in over 13,000 patients. Histologically, this method has been shown to effect permanent occlusion of the portion of the selected vein.

Correct diagnostic procedure and continuous compression after injection are essential to success.

A preliminary survey of 760 patients treated over a period of six years has shown a recurrence-rate of less than 15%. Pregnancy and previous deep-vein thrombosis are not regarded as contraindications to this method of treatment.

I am grateful for the help of Dr. E. FitzGerald, who is in receipt of a full-time research fellowship from the Research Council of Ireland; and of Mr. J. C. Milliken, my senior registrar.

HOME          FEEDBACK