Is arteriovenous shunting involved in the development of varicosities? A study of the intraluminal pressure and oxygen content in varicose veins

Introduction

Arteriovenous (AV) shunting has been proposed as the cause of varicose veins. ^1–4 It has also been associated with the skin disease and ulceration seen in the chronic venous insufficiency. ^5,6 Varicosities could result from localized increases in pressure and flow at the sites of AV communications. Such AV shunts are known to exist in the superficial circulation and may be activated under various physiological circumstances. ^2,7 This hypothesis has been encouraged by intraoperative observations of a brighter red colour of blood from varicose veins and apparent high pressure within the veins on puncture. These observations are supported by anatomical studies detailing the presence of fistulae and reports of Doppler waveform patterns in varicose veins consistent with underlying AV shunting. ³ A number of authors have tested the theory by measuring oxygen content of blood within varicosities and the femoral veins of patients with venous disease. The results have been inconsistent and subject to varying interpretation. ^4,8–12 To date, there have been no published studies of intravenous pressure profiles to investigate the AV fistula theory. Significant shunting would be expected to result in pulsatile venous hypertension in the proximity of the fistula. The aim of this study was to test the hypothesis that local venous hypertension caused by AV fistulae is involved in the development of varicose veins by measuring pressures and the oxygen content within varicosities.

Patients, materials and methods

Ethical approval for this study was obtained from the Oxfordshire Research Ethics Committee. Patients were recruited from the day case waiting list for venous surgery. These comprised patients with varicose veins with and without venous skin disease (clinical, aetiological, anatomical and pathological elements [CEAP] 2–6), all of who had incompetence in the long saphenous distribution. A control group was recruited from patients awaiting non-venous surgery with no clinical evidence of venous disease. All patients and control subjects had preoperative venous duplex scans. Patients with deep venous reflux or obstruction or with significant co-morbid conditions were excluded. All patients gave informed consent to participate in the study. Control group subjects had venous disease excluded through absence of a history of venous disease, normal lower limb examination and absence of reflux or obstruction on Duplex scanning. All patients gave informed consent to participate in the study. Scans and measurements were carried out on the morning of surgery in the same room maintained at a constant temperature of 22ºC.

Results

Patient demographics are given in Table 1. Intraluminal pressures and the blood oxygen content were measured in the varicose veins of the legs of 39 (19 women/20 men) patients and from the long saphenous veins of 10 normal controls. In 10 patients, additional measurements were made in above knee varicosities.

Mean pressure in varicose veins in the supine position was 12.3 mmHg (SD 3.6 mmHg). In this position, there were no differences between above and below knee varicosities and control subjects had similar pressures measured in the long saphenous vein. No pulsatile pressure tracings were obtained.

Varicosity pressures in the erect position averaged 66 mmHg (SD 9 mmHg). In all cases, the increase in pressure correlated with the distance of the varicosity from the heart. After exercise, pressure in varicosities reduced by a mean of only 22.2 mmHg, this pressure reduction was significantly less than those in control subjects. Recovery time (RT 90) was also significantly shorter than in the control group with the pressure in varicosities rapidly returning to baseline levels after exercise (Figures 1 and 2).

OPEN IN VIEWER

Figure 1

Pressure reduction with position and exercise

OPEN IN VIEWER

Figure 2

Time to return to baseline after exercise (RT 90). Mean and standard error of mean

Mean venous pO₂ in varicosities was 4.5 kPa (SD 1.0) in the supine position dropping to 3.9 kPa (SD 0.9) on standing; these values were not significantly different to samples from control subjects. Similar values and postural changes were seen in arm vein samples drawn in supine and standing positions (Figure 3).

OPEN IN VIEWER

Figure 3

Oxygen concentrations in erect and supine positions. Mean and standard error of mean

There were no differences between patients with venous ulcers (n = 6) compared with those without.

Discussion

A haemodynamic basis for the development of varicose veins is suspected but as yet unproven. Many surgeons broadly accept the theory of descending incompetence put forward by Trendelenburg at the end of the 19th century. ¹³ According to this theory, leaking valves in the saphenofemoral region give rise to higher pressures on the lower valves in a progressive manner, ultimately leading to high pressure in the distal veins causing them to dilate and become varicose. Several authors have questioned this theory and its validity is contradicted by a number of observations. Of particular note, varicosities often appear in veins draining into competent channels and the pattern of distribution of varicosities rarely follows the distribution of reflux. ¹⁴

The theory that varicose veins arise from AV fistulae is credited to Pigeaux ¹³ who noted blood from varicosities was a brighter red than expected. This was supported by Blalock ⁴ who demonstrated high oxygen concentrations in the veins of patients with venous disease. A number of authors have repeated these results using modern methods. Baron and Cassaro demonstrated higher oxygen concentrations in varicose veins compared with controls in the recumbent position, ⁹ Similarly, Scott et al ^10,11 found higher oxygen concentrations in varicose veins in the supine position but not when standing. Although the radiolabelled albumin studies of Partsch ¹⁵ in the 1970's cast serious doubt on the AV fistulae theory, further evidence in its support has come from Haimovici ³ using Doppler waveform analysis. He reported evidence of AV fistulae in 80% of the varicose vein patients he studied.

Proponents of the AV fistula theory explain the development of varicosities as the reaction of a weakened vein to the hydrodynamic insult of high pressure oxygenated arterial blood on the venous wall or simply as a dilatation in response to increased pressure.

AV fistulae are shown experimentally to increase the pressure in the distal vein and although the magnitude of pressure increase diminishes with chronicity of the fistula a pulsatile waveform is maintained. ^16–18

This study for the first time tests the AV fistula theory by recording pressure profiles in a series of patients with varicose veins and across a spectrum of mild to complicated varicose veins. We found pressures similar to the normal pressure values recorded in early physiological studies ¹⁹ in both patients and control subjects and we were unable to demonstrate any abnormal elevations in pressures within varicosities in either supine or standing positions. Moreover, in contrast to Haimovici's findings using Doppler, we did not detect any pulsatile pressure fluctuations.

An impaired ability to reduce pressure in the varicosities in response to exercise (ambulatory venous pressure) was, however, seen in all varicose vein limbs consistent with venous incompetence. ¹³ Ambulatory venous pressure as recorded in a dorsal foot vein has a well established correlation with venous skin disease and ulceration. ²⁰ Our finding that this failure to reduce venous pressure with exercise is consistently reflected in the varicosities is consistent with the association of varicose veins and valvular incompetence, and may support the role of descending incompetence in the pathogenesis of varicosities. However, the question of whether valvular incompetence precedes varicosities or rather is a result of the same inflammatory process ²¹ which gives rise to them remains unresolved.

We found venous oxygen concentrations of similar values to those reported previously. In the current series, however, there was no significant difference between patients and control subjects in either supine or standing positions. Our study again demonstrates the drop in venous O₂ on standing reported by Scott et al. We have shown, however, that the increase in O₂ on lying down is also seen in arm veins and therefore most likely related to postural changes in cardiac output.

These data cannot exclude the presence of microscopic fistulae but a normal oxygen content coupled with a failure to demonstrate any resting pressure differences between varicose veins and normal veins does make it unlikely that localized venous hypertension from AV fistulae is responsible for varicosis even if these AV fistulae exist.

Bright red venous blood observed during surgery can be explained at least in some cases by the fact that blood is supersaturated with oxygen with a high inspired oxygen content delivered by the anaesthetist during surgery. Profuse bleeding from varicosities seen at surgery is not related to high intraluminal pressures.

In conclusion, pressures within varicose veins are determined by hydrostatic forces and contain venous blood with the normal oxygen content. Varicose veins are not associated with elevated intraluminal pressures in the resting state but are associated with impaired reduction of intra-varicosity pressure on exercise. This study does not support a role for AV fistula in limbs with varicose veins.