PREOPERATIVE DIGITAL PHOTOPLETHYSMOGRAPHY PREDICTS IMPROVEMENT IN VENOUS FUNCTION AFTER SUPERFICIAL VENOUS SURGERY FOR CHRONICALLY ULCERATED LIMBS
 Sullivan JG, Ghauri ASK, Whyman MR, Poskitt KR
Phlebology 1998; 13:142-47

ABSTRACT AND COMMENTARY BY:
Clifford M. Sales, MD
Westfield, New Jersey

This paper sets out to look at a most important issue in venous surgery - the use of preoperative noninvasive studies to predict outcome in patients with venous ulceration. Like many other papers on this subject, the authors fail to link the use of noninvasive testing to clinical outcomes. Rather, they have produced another study documenting the physiologic improvement (measured in this instance by digital plethysmography) with venous surgery.

The authors studied 30 ulcerated limbs with incompetence confined to the superficial system (reflux > 1 second). Digital plethysmography (d-PPG) was used to quantitate venous refill time and pump power. Preoperative measurements were made with and without a tourniquet in an effort to mimic ablation of saphenous reflux. Patients with documented reflux and ulceration underwent saphenectomy (greater and lesser as indicated). Improvement in venous refill time and pump power was demonstrated preoperatively with the use of an above-knee tourniquet and postoperatively at the time of testing. The venous refill time "correction" with a tourniquet correlated with a postoperative improvement but pump power correlation could not be demonstrated. Eighty percent (80%) of the ulcers healed with little difference in the venous refill time and pump power change noted between the healed and nonhealed groups.

COMMENTARY

This paper is another in a long line of attempts to assist the surgeon in the selection of patients for appropriate venous surgery. Again, unfortunately, a rather simple (if not outdated) test has failed to provide any help to the clinician treating patients with venous ulceration. The improvement in various physiologic parameters (venous refill time and pump power) is evident. However, the disheartening finding of little (if any) difference between the healed and nonhealed groups indicates that d-PPG will not adequately assist in patient selection for venous surgery.

This report again highlights the complex nature of venous disease. Unfortunately, to date, no single test has been identified which can be used exclusively to make clinical decisions. This paper adds to our growing knowledge of the treatment of venous insufficiency but does little to offer practical assistance. vdsal223


EDITOR'S NOTE

 Dr. Sales refers to the complex nature of venous disease and points out that no single test can predict clinical efficacy. In this particular study, an explanation for the lack of predictability could be the effect of pulsatile perforator outflow and leukocyte stimulation in causation of venous ulceration. Clearly, digital plethysmography with or without tourniquet testing would be unable to predict the efficacy of surgery on the local events which lead to the severe chronic venous insufficiency. It should be clear by now to all interested physicians that the general limb hemodynamics are not the single explanation for severe chronic venous insufficiency.




PREOPERATIVE EVALUATION IN PRIMARY VARICOSE VEINS
 Pfyffer M.
Schweiz Med Wochenschrift 1998; 128:1772-80

COMMENTARY BY:
Matthias K. Widmer, MD
Cardiovascular Surgery
University Hospital
Berne, Switzerland

This review is really a state-of-the-art article which summarizes the clinical findings and anatomic terminology of primary venous insufficiency. It points out that the objective of each diagnostic evaluation is to indicate proper therapy. It is especially true that in surgical treatment, it is important to know which segments of the veins should be removed and which ones should be preserved.

In the 1980s, photoplethysmography (PPG) was used for analysis of reflux and assessment of deep venous insufficiency. Today, with continuous-wave Doppler, superficial reflux can be examined easily and, together with the clinical examination, therapeutic decisions are possible. In cases of short saphenous or perforator vein insufficiency or recurrent varicosities, duplex examination is of great value. This is especially true if the surgeon is present during the examination. Phlebography, whether performed by ascending techniques or with Valsalva stimulation, and varicography are only necessary if duplex findings are unclear or a high-quality duplex technician is not available.

This group of Swiss physicians believes that CT and MRI are of no value for varicose vein patient evaluations but these techniques are useful for evaluation of congenital angiodysplasias.

COMMENTARY

This review article underscores the importance of duplex examination in preoperative planning of varicose vein surgery and especially in recognizing perforator vein insufficiency as well as the variability of the termination of the short saphenous vein. Even though no data are presented by these authors concerning sensitivity and specificity of duplex examination, the literature which they review points out that duplex is the examination of choice today. Close relationships and permanent communication between the angiologist, the surgeon, and the radiologist can help to improve operative results. This is true in the field of highly selective surgery as described by P. Wigger in a separate article in the same issue of this journal.1

The duplex technique allows identification of varicose veins preoperatively but it is even more valuable in documenting the surprisingly high incidence of recurrent varicose veins after greater saphenous vein stripping and perforator dissection.2 In order to have optimal results in reflux scanning, it is not only important to use the duplex but also a standardized cuff inflation/deflation technique instead of Valsalva stimulation or manual compression. Standardized testing is an important goal to strive for in evaluating patients with primary varicose veins. vdwid200

REFERENCES

1. Wigger P. Surgical treatment of primary varicose veins. Schweiz Med Wochenschr 1998; 128:1781-88.

2. Jeanneret Ch, Fisher R, Galeazzi RL et al. Recurrence of insufficient perforators after SEPS procedure. VASA 1998(Suppl) 52:78.

3. Jeanneret Ch, Aschwanden M, Labs KH, Jager K. Duplex ultrasound for the assessment of venous reflux. Current Prob Dermatol 1999 (in press).




CLINICOPATHOLOGICAL EVIDENCE THAT NEOVASCULARIZATION IS A CAUSE OF RECURRENT VARICOSE VEINS
 Nyamekye I, Shephard NA, Davies B, Heather BP, Earnshaw JJ
Eur J Vasc Endovasc Surg 1998; 15:412-15

COMMENTARY BY:
Peter Conrad, FRCS, FRCS(Ed), FRACS, FACS(US)
Sydney, Australia

The authors seek to prove that neovascularization at the saphenofemoral junction after previous saphenofemoral high ligation is due to the formation of new thin-walled, serpentine veins which have certain features when examined histologically. This includes that since they are immature veins, they are tortuous, thin walled, and lack mural nerves on S100 staining.

A total of 28 legs (20 patients) with previous saphenofemoral ligation were studied. Diagnosis preoperatively was by hand-held Doppler (19) and color duplex scanning (9). At operation, a block of tissue was taken and studied histologically. In 19 cases, neovascularization conforming to the above criteria was identified. The new veins connected the common femoral vein to thigh varicosities. The other nine groin recurrences appeared to be due to a missed residual vein. Analyzing the original operations, 19 groins had a saphenofemoral disconnection alone and nine had additional stripping of the long saphenous vein in the thigh. It was found that in 8 of the 9 cases where recurrence was due to residual veins, some histological evidence of neovascularization remained.

The authors conclude that neovascularization is a major cause of recurrent saphenofemoral incompetence. They postulate that angiogenesis, a characteristic of healing wounds, is the beginning of the process and may cause saphenofemoral recurrence by connecting the femoral vein near the saphenofemoral junction to nearby divided tributaries or an unstripped long saphenous vein. If true, they conclude it would be logical to avulse tributaries carefully near the saphenofemoral junction and to strip out the long saphenous vein to prevent neovascularization. They also state that barrier methods and use of growth factor inhibitors should be further investigated.

COMMENTARY

This study supports the known fact that neovascularization, new vein formation by immature, thin-walled, tortuous veins lacking intramural small nerves, is a definite cause of saphenofemoral recurrence. The authors hint, though do not conclude, that neovascularization is the most common cause of recurrence when the long saphenous vein has not been stripped.

My own experience in reexploring the saphenofemoral junction for recurrence is that there is often neovascularization connecting the saphenofemoral stump and a non-stripped long saphenous vein. This is supported by the authors in that 19 of their 28 cases had saphenofemoral disconnection alone without stripping. I believe the most important conclusion of the paper is that angiogenesis is a feature of all healing wounds and that saphenofemoral recurrence likely depends on the presence of larger veins within reach of developing vessels. Thus it is logical that surgical techniques include careful avulsion of all peripheral tributaries within the vicinity of the ligated saphenofemoral junction and stripping out of the long saphenous vein. Of particular importance is that the surgeon look for the anterolateral vein of the thigh which can act as a long saphenous vein collateral which, when not stripped, may act as a target of neovascularization.

Barrier methods have been advocated in the past as a means of preventing neovascularization. These include the covering of the saphenofemoral stump with a transposition of pectineus fascia as advocated by Sheppard in 1978.1 Other authors have suggested covering the saphenofemoral stump with Dacron or other synthetic mesh. These barrier operations are not practiced widely and there is no hard data to support their effectiveness. ivdfcon3

REFERENCE

1. Sheppard M. A procedure for the prevention of recurrent saphenofemoral incompetence. Aust NZ J Surg 1978; 48:322-26.


EDITOR'S NOTE

 The present paper was selected for abstracting because of the controversial aspects of the authors' conclusions. Others have taken a different view of the causes of recurrent varicose veins but have agreed that careful stripping out of the tributaries to the saphenofemoral junction is crucial to decreasing the incidence of recurrent varicosities. Further, it is becoming almost universally agreed that stripping out of the saphenous vein in the thigh is a second crucial step in reducing recurrences. Conrad, in his commentary, has made the correct observation that the lateral anterior tributary to the saphenous vein can act as a conduit for recurrent varicosities in the thigh and elsewhere. Certainly if it can be stripped out during the primary operation, this should be done.




ANGIOSCOPIC EXTERNAL VALVULOPLASTY IN THE TREATMENT OF VARICOSE VEINS
 Satokawa H, Hoshino S, Igari T, et al.
Phlebology 1997; 12:136-41

ABSTRACT AND COMMENTARY BY:
Harold Welch, MD
Department of Vascular Surgery
Lahey Hitchcock Medical Center
Burlington, Massachusetts

With the hope of preserving the greater saphenous vein (GSV) for future use as a bypass conduit, the authors performed angioscopic diagnosis of valvular incompetence of the proximal GSV in patients undergoing surgery for varicose veins. They also performed external valvuloplasty of the GSV in selected patients. From December 1989 to June 1997, 306 limbs in 187 patients underwent intraoperative angioscopy as part of their surgical procedure for primary varicose veins. Preoperative duplex scanning had identified reflux in the thigh segment of the GSV. Exclusion criteria for valvuloplasty included tortuosity of the thigh segment of the GSV, a diameter of 8 mm or more by standing duplex examination, deep venous reflux by duplex, and CEAP clinical class greater than 3. A total of 232 limbs had valvuloplasty attempted.

Valvuloplasty was performed on the subterminal valve using three different techniques: 1) total plication of the dilated annulus by running propylene suture, 2) plication by autogenous femorofascial sleeve or Dacron-reinforced silicone, or 3) plication of the commissure with shortening of the cusps from outside the vein wall. In addition, partial stripping or segmental ligation was performed for varicosities below the knee with suprafascial ligation of incompetent perforating veins. Valvuloplasty was successful in 62 limbs (20% of limbs undergoing angioscopy and 27% of attempted valvuloplasties).

Limb blood flow was studied using strain-gauge plethysmography to measure venous capacitance, photoplethysmography to measure refill time, and duplex ultrasonography to measure GSV reflux by the standing cuff-deflation technique. Duplex values of reflux time, venous diameter, and peak reflux velocity were measured. Time-averaged flow rates were calculated. Morphology of the valves were classified according to Dr. Satokawa's previously described system.1 Successful valvuloplasty was performed in 62 limbs (20%) by a variety of techniques. Partial stripping from below the knee and suprafascial ligation of perforators was simultaneously performed in selected patients.

Results showed that the venous capacitance did not change significantly from preoperative to postoperative in the valvuloplasty group or the stripping group but venous filling time was significantly prolonged in both groups postoperatively. In the valvuloplasty group, mean GSV diameter decreased from 7.3 mm to 5.4 mm and reflux was abolished in 77% of limbs. Both reflux time and time-averaged flow also decreased significantly. Followup showed that the time-averaged flow increased in the first three years following valvuloplasty and then reached a plateau. There were two (3.2%) GSV thromboses postoperatively (19 and 22 months) and a varicose vein recurrence rate of 6.5%.

COMMENTARY

Vein valvuloplasty is an enjoyable but technically demanding procedure which is very rewarding when successful. These authors have once again demonstrated that they can perform the procedure very well. Although valvuloplasty can be successful in the long saphenous vein, the question of whether it should be done remains unanswered.

I do not believe valvuloplasty has usefulness in the United States at this time because of the increased costs it adds to varicose vein surgery. Not only does the procedure add length to the procedure time, but with only a 20% valvuloplasty success rate, the cost of angioscopy is not justified in this cost-cutting era. The larger question is how many patients have benefitted from this procedure? While we may find out in ten years with a followup report on varicose degeneration of the GSV in the valvuloplasty group, I suspect it will be 20 years before we know how many patients had a preserved GSV used as an arterial conduit. vdwel194

REFERENCE

1. Satokawa H, Iwaya F, Igari T et al. Morphological studies of the venous valves by angioscopy. Mayakkangaku (J Jap Coll Angio) 1992; 32:395-401 (in Japanese).




POWER-BASED COLOR-CODED DUPLEX SONOGRAPHY FOR EVALUATION OF CALF VEINS
 Baumgartner I, Braunschweig M, Triller J, Mahler F
Int Angio 1998; 17:43-48

COMMENTARY BY:
Nicos Labropoulos, M.D.
Loyola University Medical Center
Maywood, Illinois, USA

This prospective study compares color Doppler (CD) and power Doppler (PD) ultrasound in the detection of deep venous thrombosis (DVT) in patients presenting with acute symptoms. A total of 47 consecutive patients underwent phlebography and duplex scanning within 12 hours of each other. The calf veins were examined both by CD and PD. On phlebography, three views were obtained in the calf veins and two views in the popliteal vein. The phlebograms were read by two radiologists independently.

Acute DVT was found in 23 patients (49%), acute recurrent DVT was found in 5 (10.5%), chronic postthrombotic changes were found in 5 (10.5%) and the remaining 14 (30%) were normal. Location of the thrombi was proximal in only 2 (7.1%), proximal and calf in 20 (71.4%), and calf only in 6 (21.4%). Phlebograms were suboptimal in 14 patients (30%) due to underfilling of muscular calf veins.

The overall sensitivity, specificity, and accuracy of color Doppler were 93%, 89%, and 91% (95% CI, 81-98%). The accuracy of detecting DVT in paired calf veins was 96% (95% CI, 85-99%). In the muscular calf veins there were three false-positive and three false-negative cases for a sensitivity of 57%, specificity of 88%, and accuracy of 81% (95%, 63-93%). The addition of the power Doppler significantly increased visualization of the paired calf veins but did not improve accuracy of color Doppler.

COMMENTARY

Despite the fact that phlebography is still the accepted gold standard for DVT detection, several reports have shown its limitations. The authors recognize and adequately discuss the limitations of phlebography. They also point out the technical limitations of ultrasound and the potential bias of the single observer. The description of their technique and the discussion of their results and potential problems in the study reveal that these authors are experienced in the diagnosis of DVT. Color ultrasound has an excellent diagnostic accuracy in detecting proximal and distal DVT in symptomatic patients. However, its accuracy has been questioned in high-risk symptomatic patients. Several studies including a recent meta-analysis have shown mediocre to poor sensitivity in these patients. For this reason, all Level I studies use phlebography to detect DVT.

We have found that B-mode ultrasound together with CD has an excellent accuracy even in asymptomatic patients. That is also confirmed in other centers with large experience in this field. With venography being obsolete in such centers, it is difficult to make more comparative studies.

The major weakness of this current paper is the small sample size. This is reflected on the wide 95% CI despite the great accuracy of both CD and PD. In addition, 70% of patients had acute DVT, postthrombotic changes, or a combination of both. In most prospective studies using phlebography or ultrasound, the prevalence of DVT is less than 30%. Since this study used consecutive patients, their yield rate for DVT was quite high. If the patients with postthrombotic changes alone were excluded, the prevalence of acute DVT would still be high at 60%.

Although PD increased the visualization of calf veins, it did not improve the accuracy of CD. PD is too sensitive and is subject to motion artifacts while CD has low-flow settings capable of detecting slow flow. While CD may not detect flow in the muscular veins, this is not a problem because simple compression on B-mode can be used to detect thrombi. In a recent study we found that 40% of patients with calf DVT have isolated soleus or gastrocnemius vein involvement. The propagation rate of thrombi in the soleus and gastrocnemius veins is unknown and cannot be assumed to be similar to the peroneal and posterior tibial veins. However, in one study of 75 patients, propagation into proximal veins occurred more often from isolated soleus vein thrombosis than peroneal and posterior tibial vein thrombosis. Therefore, muscular veins should be included in the examination.

Given the inability of phlebography to demonstrate flow in the muscular veins, it is difficult to compare PD and CD with the accepted gold standard. This fact can explain the low sensitivity of ultrasound in the muscular veins. In the absence of a true gold standard, this may remain a diagnostic problem. However, for those who use ultrasound and have the experience and skill, calf DVT would only be a rare diagnostic challenge. ivdflab




PELVIC CONGESTION SYNDROME: EARLY CLINICAL RESULTS AFTER TRANSCATHETER OVARIAN VEIN EMBOLIZATION
 Cordts PR, Eclavea A, Buckley PJ, DeMaioribus CA, Cockerill ML, Yeager TD
J Vasc Surg 1998; 28:862-68

COMMENTARY BY:
Marianne Vandendriessche, MD
John T. Hobbs, MD, FRCS
Gent, Belgium

Chronic pelvic pain is a common disorder accounting for up to 10% of outpatient gynecologic visits and one-third of diagnostic laparoscopies. Many causes have been identified. During the last 30 years, pelvic venous congestion due to ovarian vein incompetence has become recognized in some of these patients. The pain is of variable intensity, worse premenstrually, and aggravated by standing, fatigue, and coitus. Often there is perineal heaviness and urgency of micturition. Many methods of treatment have been advocated including medical (i.e.,m dihydroergotamine, ovarian suppression, rheologic agents) and surgical (uterine ventrosuspension, hysterectomy, and ovarian vein ligation and excision).

This paper describes the early results of ovarian vein embolization, considered a less invasive approach. The early radiographic and clinical results of attempted ovarian vein embolization in 11 women with symptoms suggestive of pelvic congestion syndrome are presented. All were multiparous (mean age 33.1 years). Eight were referred because of leg or vulvar varices and three were referred because of tubo-ovarian varices. After a clinical diagnosis of pelvic congestion syndrome, the women were investigated. Initial examination of the ovarian veins by duplex scanning was not successful. Ovarian vein venography was done in five women and ovarian and iliac vein venography was done in six. The catheter was introduced via the right common femoral vein or the internal jugular vein. The left ovarian vein was evaluated during normal breathing and during Valsalva maneuver. Inferior vena cava and left renal vein pressures were measured. An enlarged or incompetent left ovarian vein was treated with transcatheter coil embolization with Gianturco stainless steel coils or platinum microcoils. An average of 16 coils was used for each embolized ovarian vein (6 to 24 coils). No liquid sclerosants were used.

The criteria for embolization included an ovarian vein diameter of more than 10 mm, moderate to severe congestion of the ovarian plexus, uterine venous engorgement, and filling of the contralateral pelvic veins. Parallel ovarian vein trunks were also embolized. A selective right ovarian venography was then performed and dealt with in a similar fashion. A standard questionnaire was administered before and after embolization. The women expressed symptomatic relief as a percentage of each presenting symptom.

A total of 11 women were referred to the vascular clinic for evaluation of lower extremity varicosities (6), vulvar varices (2), tubo-ovarian varices diagnosed at laparoscopy (1) or by transvaginal ultrasound scan (1), and laparotomy for chronic pelvic pain (1). Seven had pelvic symptoms after the second or third pregnancy and four after the first pregnancy. Physical examination revealed vulvar varices in nine and varices over the buttock in three.

All patients were examined by selective venography and nine underwent embolization (mean followup 13.4 months). Complications included contrast extravasation (3) and a left lower lobe coil pulmonary embolus (1). The authors concluded that ovarian vein venography with embolization for pelvic congestion syndrome can be performed with a high degree of technical success but recognized that identification of the right ovarian vein is difficult. Of the nine patients treated by embolization, eight reported excellent initial relief of symptoms but symptoms returned in two. One reported a 40% improvement but repeat venography failed to identify the cause for persistent pain. Anatomic variations were common as is frequent in the venous system. One woman showed evidence of previous left ovarian thrombophlebitis.

COMMENTARY

Incompetence of the ovarian veins is now widely recognized as a cause of recurrent varicose veins of the legs arising from atypical veins in the upper medial thigh, usually associated with the pelvic congestion syndrome. In a small number of these patients, the legs are not involved.

Although many workers have attempted to investigate the ovarian veins with duplex, only Richardson's group in Wagga Wagga, Australia have consistently been able to use this method successfully.1 A few years ago, we operated on two patients in Vancouver where the radiologists had successfully demonstrated dilatation and incompetence of the left ovarian veins; however, in both cases they were unable to cannulate the right ovarian vein. While passing through Vancouver, Richardson had demonstrated a large right ovarian vein with reflux in both patients and at operation we confirmed this.

We also operated on two patients whose symptoms persisted and gross reflux was still present after embolization. In one, the large coil was flattened against the side of the vein and covered with intima. In the other, 24 coils had been inserted (15 on the left and 9 on the right). Despite this, symptoms persisted. After introduction of the coils, the patient experienced pain on the anterior upper right thigh. When the ovarian veins were resected, we noted that the ends of the coils were protruding through the vein, particularly in the area of the root of the right femoral nerve. Removal of veins and coils relieved the patient's original symptoms and eliminated the femoral nerve pain.

The anatomy of the ovarian veins is highly variable and often is not a single enlarged vein. Some six to nine veins arise from each ovary as a pampiniform plexus and join together to form a single vessel on each side. However, the pattern shows wide variation. Sometimes a single vessel forms immediately above the ovary and sometimes it does not become single until its proximal end. We believe this wide anatomical variation explains the high incidence of failure after embolization since one can place coils only in those veins which have been visualized. Other veins may have been missed and it is not always possible to introduce a catheter into the right ovarian vein.

At open operation however, the precise anatomy is clearly seen on both sides. Great care is taken to ensure that the complete large vein or network and all its tributaries from the ovarian plexus is resected. This requires a small gridiron approach to the retroperitoneal space on each side. Since the abdominal cavity is not opened, recovery is rapid and patients are discharged on the second or third day. Varicose veins of the legs are dealt with at the same time. In a series of more than 100 ovarian veins resected in the past 20 years, the only complication was a wound hematoma in one patient.

Another approach which has been suggested is ligation of the vein by laparoscopy.2 This technique is difficult when the venous plexus is complex and is worrisome because the vein is very thin-walled and can be grossly dilated and fragile. Open operation allows better control.

In conclusion, we believe that surgery is the best approach because of the anatomical complexity. ivdfvan2

REFERENCES

1. Boomsma JHB, Potocky V, Kievit CEL, Verhulsdonck JCJ, Gooskens VHJ, Weemhof RA. Phlebology and embolization in women with pelvic vein insufficiency. Medicamundi 1998; 42(2):22-29.

2. Grabham JA, Barrie WW. Laparoscopic approach to pelvic congestion syndrome. Br J Surg 1997; 84:1264.




MINI ABSTRACTS
John J. Bergan, M.D.
Items of Interest Which Have Crossed the Editor's Desk
(Provided for reference purposes and general interest)


Efficacy of a Low-Molecular-Weight Heparin Administered Intravenously or Subcutaneously in Comparison with Intravenous Unfractionated Heparin in the Treatment of Deep Venous Thrombosis
Kirchmaier CM, Wolf H, Sch”fer H, Ehlers B, Breddin HK.
Int Angiol 1998; 17:135-45

This interesting study measures thrombus regression associated with two different methods of treatment of deep venous thrombosis. As it turns out, subcutaneous treatment with low-molecular-weight heparin is at least as effective as unfractionated heparin.


Cutaneous Inflammation Limited to the Region of the Ulcer in Chronic Venous Insufficiency
Hahn J, J¸nger M, Friedrich B, et al.
VASA 1997; 26:277-81

Data in this paper implies that up-regulation of endothelial adhesion molecules (ICAM-1) and dermal infiltration by T lymphocytes and macrophages in chronic venous insufficiency is limited to the region of the ulcer or at least to skin areas with severe microangiopathy.


Arteriovenous Fistula After Endoscopic Dissection of Perforating Veins of the Calf in a Patient with Chronic Venous Distention Using the Neodym:YAG Laser (French)
C–lsch C, Rauber K, Langer C.
VASA 1998; 43

The title says it all.


Socio-economic Impact of Chronic Venous Insufficiency: An Underestimated Public Health Problem
Van Den Oever R, Hepp B, Debbaut B, Simon I.
Int Angiol 1998; 17:161-67

Readers of the Venous Digest appreciate the economic burden of chronic venous insufficiency. This is an up-to-date reference on the subject which may be useful.