Randomised Trial of Pre-operative Colour Duplex Marking in Primary Varicose Vein Surgery: Outcome is Not Improved

Transcription

1 Eur J Vasc Endovasc Surg 23, (2002) doi: /ejvs , available online at on Randomised Trial of Pre-operative Colour Duplex Marking in Primary Varicose Vein Surgery: Outcome is Not Improved J. J. Smith, L. Brown, R. M. Greenhalgh and A. H. Davies Department of Vascular Surgery, Imperial College of Science, Technology and Medicine, Charing Cross Hospital, London, U.K. Objective: the ability of colour duplex to accurately locate incompetent venous sites has been widely published; its value in pre-operative marking in simple primary varicose vein surgery is evaluated in this study. Design of study: prospective randomised controlled trial. Setting: regional vascular service, hospital patients. Subjects: consecutive patients (149) undergoing primary varicose vein surgery where the only difference was one group of patients received duplex marking prior to surgery. Six weeks and 12 months post-operatively all patients had a colour duplex scan to determine the accuracy of surgery and the presence of residual/recurrent varicose veins and completed the Aberdeen, SF-36, and EuroQol quality of life questionnaires. Interventions: varicose vein surgery. Main outcome measures: Duplex evidence of venous incompetence, quality of life measures using the SF-36 and Aberdeen Varicose Veins Questionnaire. Results: pre-operative marking of primary varicose veins by skilled duplex ultrasonography does not improve the accuracy or recurrence rate following surgery. Quality of life improved significantly following surgery in both groups, however there was no difference in this improvement between the groups. Conclusion: the trial has not demonstrated any additional benefit of pre-operative colour duplex marking over that of clinical and hand held Doppler marking in terms of satisfactory varicose vein surgery performance at 6 weeks or 12 months in patients with primary varicose veins of the long saphenous system. It role in the short saphenous system is less clear. Introduction recurrence of varicose veins has led to successful claims of negligence. 8 Over the years different techniques have been tried to reduce recurrence rates in varicose vein surgery. In the last few years PTFE patches have been used to cover the ligated saphenofemoral junction in re-re- current veins to try and prevent recurrence 9 occurring again. Even such physical barriers as this have a recurrence rate of 12% attributed to failure of the patch. A different group using reflected pectineus fascia to cover the junction 10 obtained similar results. Good surgery depends on a correct pre-operative assessment of patients. It has been repeatedly shown how poor clinical examination is for diagnosing reflux in varicose veins and even hand held Doppler has it limitations 16,17 with sensitivity of the investigation reported at 73% in the groin and 77% in the popliteal fossa. 16 To improve the results of surgery, we need to use the most accurate method of pre-operative assessment. The ability of colour duplex to accurately locate incompetent venous sites has been widely published, and as such duplex has become established as a method of pre-operative assessment of varicose veins. Varicose veins are one of the commonest of all surgical problems with a prevalence of 25% in women and 15% in men. 1,2 As a consequence operative rates are also high. Around operations were performed for varicose veins in the United Kingdom alone in Thirty years ago Lofgren stated Recurrence of varicose veins continues to be a problem of great magnitude to the patient and physician alike. 6 It was also stated that the recurrence rate of the time (20 26%) had not changed much over the preceding 10 years despite surgical advances. The situation is still the same today in terms of recurrence of varicose veins; a recent review of the subject has found that the recurrence rate of varicose veins is 21% over a 10-year period. 7 With such a high demand for a procedure that has a 1 in 5 recurrence rate, ways of improving the outcome of surgery need to be found. Furthermore, Please address all correspondence to: J. J. Smith, Department of Vascular Surgery, Charing Cross Hospital, Fulham Palace Road, London W6 8RF, U.K /02/ $35.00/ Published by Elsevier Science Ltd.

2 Varicose Vein Surgery 337 a diagnostic duplex scan in the outpatient department before consultation with the clinician. The surgical approach was then planned on the basis of the clinical, Doppler and duplex findings in the outpatient de- partment at first visit. After randomisation using the sealed envelope system the duplex scans were re- moved from the notes of all patients (approximately 90% of participants) in the clinical group by an in- dependent observer prior to being seen by a surgeon. An experienced vascular technologist performed all of the colour duplex scans with the patient standing with their weight taken mainly on the opposite foot. The vascular technologists had been previously aud- ited for intra-observer and inter-observer error as part of the Small Aneurysm 23 and Four Layer Bandaging studies. Reflux for this purpose was defined as reverse flow seen by colour change for greater than one second following calf pressure augmentation of flow. An Acu- son 2000 scanner with a 7.5 MHz linear array probe was used for all venous scans. An Imex continuous wave Doppler with an 8MHz probe was used for hand held Doppler analysis. The operative procedure was planned in the outpatient department based on all of the information available to the surgeon at the time. All patients underwent elective surgery under general anaesthetic performed by senior vascular higher surgical trainees, lecturers or consultants at the Regional Vascular Centre. Operative procedures on the long saphenous system involved flush sapheno-femoral transfixtion with division of surrounding branches and removal of the long saphenous vein to just below the knee if incompetent. Operative procedures on the short saphenous system involved flush sapheno-popliteal transfixion and removal of a section of the short sa- phenous vein. The removal of branch varicosities was performed via tiny stab incisions and the use of the Oesch phlebectomy hooks or a Kocherised Mosquito clip. These branch varicosities were marked by the operating surgeon with the patient standing im- mediately prior to surgery in both groups. Follow up was scheduled at 6 weeks and one year following surgery at which point all patients were assessed clinically and by colour duplex scan. All patients completed quality of life questionnaire booklets containing the Aberdeen Questionnaire, the SF-36 and the EuroQol, prior to surgery and again at 6 weeks and 12 months. These questionnaires were used to determine if there was any difference in quality of life between the patients randomised to duplex or non-duplex. Results are analysed by numbers of limbs and pro- cedures performed in order to compare any difference Few would doubt the ability of colour duplex to accurately diagnose and locate the sites of incompetent connections between the deep and superficial venous systems of the leg and there are some who consider mandatory 12,16 and necessary to plan the surgical approach. 22 In order to reduce recurrence rates further still some surgeons are also using colour duplex to mark these incompetent connecting sites immediately prior to surgery. However, is it necessary or beneficial prior to simple primary varicose vein surgery? Colour duplex is expensive and time consuming, and if the accuracy of surgery and recurrence rates are not significantly improved upon over the use of simpler technology such as Doppler then in today s financial climate it cannot be justified. The aim of this study was therefore to determine if the addition of colour duplex marking of the sites of deep to superficial incompetence improves the outcome of varicose vein surgery in terms of more accurate surgery, reduced recurrence rates and improved quality of life. The final variable was chosen, as it is the only objective way to measure the impact of an event on a patient s life. Methods and Patient Series Approval was sought and obtained from the ethical committee of the Regional Authority to conduct the study. All patients were given printed information sheets prior to entry into the study, which constituted part of their informed consent for participation. A prospective consecutive cohort of 152 patients with primary varicose veins without venous ulceration was randomised into one of two groups (Fig. 1) over a 14-month period. Randomisation was performed using a sealed envelope system provided from an independent unit; The Department of Health Sciences, University of York on the day before operation. Those patients in group one (the duplex group ) underwent clinical assessment, hand held Doppler examination, plus the patient has pre-operative colour duplex mark- ing of the sites of deep to superficial incompetence on the day of surgery. For those patients in group two (the non-duplex group ) the only difference in assessment was that the patients did not have a colour duplex marking scan on the day of surgery. Patients with bilateral disease had both limbs randomised to the same arm of the trial. Patients were taken from the waiting list for varicose vein surgery and interviewed prior to entry with a full explanation of the study. Prior to commencing the trial it had been standard practice for patients to have

3 338 J. J. Smith et al. Fig. 1. Study design (according to the CONSORT template for randomised trials). made by pre-operative duplex marking. Post-operatively a recurrence was defined as reflux at the site of previous surgery. This was then further classified into inadequate surgery (persistence of reflux on the 6 week duplex scan), neovascularisation (presence of new areas of reflux seen on the 12 month duplex scan in an area where complete obliteration of reflux was seen on the 6 week scan) or progression of disease (new sites of reflux in areas previously proven competent on duplex). Incompetent long saphenous vein correction was declared as adequate if the vein was not visible in the upper two-thirds of the thigh on the 6 week duplex scan or if the remaining vein was competent. Statistical analysis was performed using Chi-squared tests with continuity correction or Fishers Exact Test where appropriate using SPSS (Statistical Package for the Social Sciences). Power calculations were performed prior to the start of the trial. Assuming one wants to improve recurrence rates from 20% to 5% and assuming a 5% significance level then one needs 78 limbs per group. Results Of the 152 patients entered into the study, 149 had an operative intervention. The discrepancy is accounted for as follows; one could not be traced in the hospital PAS (Patient Administration System) and two had their admission for surgery suspended (not otherwise specified). Figure 1 describes details of attendance for follow-up duplex scans and operative procedures performed in the two groups. Long saphenous system The sapheno-femoral junction One hundred and twenty-five patients (189 limbs) underwent high saphenous ligation and stripping of the long saphenous vein with or without multiple distal avulsions of branch varicosities from hereon known as long saphenous procedures. Bilateral procedures were performed in 51% of patients. In this

4 Varicose Vein Surgery 339 at six weeks and the number rose to 14 at 12 months. Of the 10 at 6 weeks, four were in the duplex arm and six were in the non-duplex arm of the study (p=0.26, Fisher s exact test). Of the 14 at 12 months six were in the duplex arm and eight in the clinical arm (p=0.17, Fisher s exact test). This gives a recurrence rate of 27% due to inadequate surgery, a recurrence rate of 10.8% due to neovascularisation and an overall rate of 37.8%. Perforator veins and branch varicosities The results described include both long and short Fig. 2. Sapheno-femoral junction incompetence. saphenous procedures and those patients undergoing avulsions alone. Incompetent perforator veins were group of long saphenous procedure patients 92 limbs present in 135 limbs prior to surgery. In six of 135 were randomised to pre-operative duplex marking limbs these incompetent perforators were still present and 97 to non-duplex marking. Two limbs were found at 6 weeks of which one was in the duplex group and to have sapheno-femoral reflux at the 6 week duplex five in the non-duplex group (p=0.11, Fisher s exact scan, but a further 16 had evidence of sapheno-femoral test). By 12 months a further 19 new incompetent reflux at the 12 month scan. Of the 2 residual problems perforators (in different position on the limb) were at six weeks, 1 was in the duplex group and 1 was in present that were not seen at the 6 week scan. Fifteen the non-duplex group (p=1.0). This is a total of 18 of these were in the non-duplex group and four in the limbs with reflux following surgery at 12 months (Fig. duplex group (p=0.012). This gives a recurrence 2) an overall recurrence rate of 9.5% for the sapheno- rate of 3% due to inadequate surgery. femoral junction. As for branch varicosities all except those in two limbs (both in the duplex group) were removed at 6 The long saphenous vein weeks and none re-appeared by 12 months. How- In 11 of 189 limbs, the vascular technologists found ever, 17 limbs showed progression of disease (eight an incompetent segment of long saphenous vein in in duplex and nine in non-duplex) by the development the upper thigh at 6 weeks and the number increased of new branch varicosities by the time of the 12-month to 17 of 189 at 12 months of which 8 were in the duplex scan (p=1.0). duplex arm and 19 in the non-duplex arm of the study (p=1.0, Fisher s exact test). This gives an overall recurrence rate of 9% for the long saphenous vein. Quality of life Short saphenous system Improvement in quality of life was seen following surgery for both groups at 6 weeks and 12 months using both the Aberdeen Questionnaire and the SF-36 Thirty patients underwent ligation of the sapheno- (p>0.05) and has been reported elsewhere. 20 popliteal junction and removal of a segment of the short saphenous vein with/without multiple distal avulsions as necessary hereon known as short saphenous system procedures. Seven of these procedures Long saphenous system were bilateral and therefore a total of 37 short saphenous procedures were performed. Twenty-five of Pre-operatively there was no significant difference in the procedures were combined with a long saphenous quality of life measured with the Aberdeen Ques- system procedure. Of the 37 short saphenous system tionnaire (p=0.9), SF-36 (p >0.3 all domains) or the procedures 22 limbs were randomised to pre-operative EuroQol (p=0.451). duplex marking and 15 to non-duplex marking. At six weeks patients in the duplex group appeared Of the 37 short saphenous procedures, 10 had evidence to be doing better in terms of quality of life as measured of continuing reflux into a short saphenous vein with the Aberdeen Questionnaire (mean score 10.85

5 340 J. J. Smith et al. Fig. 3. Post-operative quality of life differences (Aberdeen Questionnaire) Long Saphenous System. Fig. 5. Short saphenous quality of life differences post-operative (SF-36). the SF-36 (p=0.09 in the domain of Physical Functioning ) and the EuroQol (p=0.03). Quality of life improved within the groups as a result of having the operation at 6 weeks and 12 months but there still continued to be a differences using the different questionnaires between the groups at both 6 weeks (p= 0.015, Aberdeen Questionnaire) and at 12 months (p= domain of Bodily Pain [SF-36] Fig. 5, p=0.5 [EuroQol]). Discussion Fig. 4. Short saphenous quality of life differences pre- and postoperative (Aberdeen Questionnaire). compared to 15.85, p=0.034 independent sample t- test), however this difference was not sustained by the time of the 12-month follow-up (Fig. 3). There was no difference between the two groups at six weeks (p>0.38 all domains) or 12 months (p>0.15 all domains) using the SF-36 as the quality of life measure. Furthermore, no significant difference was seen with question 2 of the SF-36, (health transition item, p>0.47) at six weeks or 12 months. The short saphenous system Prior to surgery there were significant differences in health between the duplex and non-duplex group using the Aberdeen Questionnaire (p=0.02 Fig. 4), Attention to detail in any branch of surgery will certainly affect results and this is very true in varicose vein surgery. Recurrence of varicose veins is periodically blamed on surgical trainees inadequately performing the operations. 24 Equally there are reports that trainees do as well as consultants. Appropriately supervised, basic surgical trainees can achieve very good results 25 therefore showing that with correct supervised train- ing all surgical trainees can achieve good results. Inappropriate assessment of varicose veins in the pre-operative phase will lead to inappropriate surgery and increased recurrence rates. It has been suggested that failure to get a duplex scan prior to surgery would result in the wrong operation being performed in up to a quarter of cases. 16,26 Colour duplex was introduced in the mid 1980s and as stated in the introduction many consider duplex to be mandatory in varicose vein assessment in order to perform the correct procedure and reduce recurrence rates. 12,16,27 29 It was also stated in the introduction that recurrence rates have

6 Varicose Vein Surgery 341 vein to the knee level. There are still some centres that practise long saphenous vein preservation 35 despite the fact that a residual long saphenous vein has been shown to increase recurrence rates. 1,5 The recurrence rate in the long saphenous group in this series (9%) may increase over the next 10 years. For this reason these patients are being followed up over a longer time period. As far as the short saphenous system is concerned, the recurrence rate at six weeks (inadequate surgery) was 27%, which confirms that the anatomy is difficult in the area of the popliteal fossa. The deep fascia must be opened and the short saphenous vein has to be followed to the popliteal vein and divided flush and also relevant medial and lateral branches divided and ligated. This has clearly not been done adequately in some of these patients. There were relatively more cases of inadequate surgery in the non-duplex group compared to the duplex group, as the numbers in the former were smaller. This difference however was not significant (p=0.17). Clearly a much larger study needs to be performed looking specifically at popliteal fossa varicose vein surgery with far more numbers than was in this series in order to get a definitive answer as has been obtained with the long saphenous system above. The number of new incompetent perforators that appeared in new places (not previously seen or op- erated on) between the 6-week and 12-month duplex is likely to represent progression of the disease and should not be taken to mean recurrence as none re- appeared at a site of previous surgery. No explanation can be offered as to why the number of new in- competent perforators increased more in the nonduplex group (p=0.012) and one has to assume this is a result of chance as this represents new disease, and there were no differences between the groups for long or short saphenous system incompetence. The quality of life studies have shown that the patients in the long saphenous surgery group were comparable in their randomisation to duplex or non- duplex by having no differences in the pr-operative health status measured with the Aberdeen Ques- tionnaire or the SF-36. At six weeks the duplex group appeared to be doing better in terms of quality of life (Aberdeen Questionnaire) but there was no significant difference at 12 months. Duplex marking therefore confers no benefit in terms of quality of life im- provement in the long saphenous system. However, in patients with short saphenous system problems there were significant differences in their health status. This was completely unintentional as the randomisation was done by a sealed envelope system. Patients in the non-duplex group had significantly worse quality of life pre-operatively and at changed little over the last several decades. Despite the use of duplex scanning this is still the case today. Three recent papers have reported on recurrence rates after varicose vein surgery all of which have been performed in the duplex era and still show that recurrence rates remain at around 20%. Two of the papers were published in 1998 and reported recurrence rates of 18% 30 and 38.7%, 31 and one paper in 1999 reporting a rate of 13.6%. 32 The above represents only three recent papers, with a mean recurrence rate of 23%. One could conclude that the introduction of duplex has done little to improve recurrence rates. This of course is not true, recurrence of varicose veins is a multifactorial problem and the arguments about whom or what is to blame will continue to rage as they have done so in the past. Is recurrence due to trainees, 24,25 neovascularisation and/or neoreflux, 33,34 or progression of disease? The answer is probably all of these in various combinations. With the introduction of structured training programs and closer supervision of basic and higher surgical trainees there should be a reduction in the number of inadequate procedures performed. Until the biology of neovascularisation is fully understood this will not be able to preventable; the same can be said for progression of disease.at the moment the area to target is inadequate surgery. Some units are pre-operatively marking the position of incompetent junctions as de- scribed in this paper with colour duplex. This paper does not attempt to refute the value of duplex diagnosis and planning but seeks to find out if duplex marking can improve the results of surgery. Other studies have also reported that following surgery for primary var- icose veins new sites of reflux can be found as early on as 6 weeks following surgery, some of which resolves at one year. 33 In the study described above this effect is also seen. One can assumed that reflux at six weeks in the saphenofemoral junction is due to inadequate surgery, which gives a recurrence rate of 1% due to inadequate surgery. One can also assume that reflux in veins at the sapheno femoral junction seen at 12 months and not at 6 weeks is due to neovascularisation if on the 6-week scan there are no veins present in this area, and hence the recurrence rate due to neovascularisation is 8.5% for the sapheno-femoral junction. As for the long saphenous vein, it is to be presumed that they were certainly left behind in 11 of these limbs and of the remaining six; they were probably residual and became more easily detected at 12 months. This gives a recurrence rate of 9% all due to inadequate surgery. As far as the saphenous vein in the distal thigh is concerned, one can surmise that greater effort must be made to remove the saphenous

7 342 J. J. Smith et al. 6 weeks using the Aberdeen Questionnaire. At 12 References months quality of life was still worse in this group, however this was not significant. Similar differences 1Callam MJ. Epidemiology of varicose veins. Br J Surg 1994; 81: were seen with the SF-36. Less of a difference was 2Allan PL, Bradbury AW, Evans CJ et al. Patterns of reflux and seen at six weeks and 12 months; however, the non- severity of varicose veins in the general population Edinburgh duplex group still had worse health. It is difficult to Vein Study. Eur J Vasc Endovasc Surg 2000; 20: Robbins MA, Frankel SJ, Nanchahal K, Coast J, Withans be certain why these patients had a worse quality of MH. Varicose vein treatments. Health care needs assessment: life, but as has been said before the numbers in this the epidemiologically based needs assessment reviews. (13). group are very small and should be interpreted with University of Bristol: Dept of Social Medicine, Lees T, Singh S, Beard J, Spencer P, Rigby C. Prospective audit caution. Furthermore, 70% of the patients in this group of surgery for varicose veins [see comments]. Br J Surg 1997; 84: also had a simultaneous operation on their long sa phenous system (all single side except for one patient), 5Baker SR, Stacey MC, Jopp-McKay AG, Hoskin SE, Thompson PJ. Epidemiology of chronic venous ulcers. Br J Surg 1991; 78: which may also account for some of the differences Removing these patients from the analysis would make 6Lofgren EP, Lofgren KA. Recurrence of varicose veins after the numbers far too small for analysis and there was the stripping operation. Arch Surg 1971; 102: Smith JJ, Davies AH, Greenhalgh RM. Does colour duplex no difference anyway in health status between those scanning improve the durability of varicose vein surgery? In: patients in the long saphenous group or those in Greenhalgh RM, ed. The durability of vascular and endovascular the short saphenous group measured with any in- surgery. London: WB Saunders, 1999: Tennant WG, Ruckley CV. Medicolegal action following treatstrument at any of the three time points. ment for varicose veins. Br J Surg 1996; 83: Even though some patients had recurrent/residual 9DeBacker G. Epidemiology of chronic venous insufficiency. varicose veins only three further operations were re- Angiology 1997; 48: Sadick NS. Predisposing factors of varicose and telangiectatic quired for persistent varicose veins in the study group. leg veins. J Dermatol Surg Oncol 1992; 18: All of the operations were limited to further avulsions 11 Labropoulos N, Leon M, Nicolaides AN et al. Superficial venous insufficiency: correlation of anatomic extent of reflux only and all were due to cosmetic reasons. with clinical symptoms and signs. J Vasc Surg 1994; 20: Varicose vein surgery should not only be a targeted 12 Basmajian JV. The distribution of valves in the femoral, external procedure based on accurate pre-operative assessment iliac and common iliac veins and their relationship to varicose veins. Surg Gynec Obstet 1952; 95: 537. but also targeted to symptoms. The Edinburgh Vein 13 Bradbury AW, Stonebridge PA, Callam MJ et al. Recurrent Study has shown that lower leg venous symptoms varicose veins: assessment of the sapheno femoral junction. Br poorly correlate with the presence or severity of trunk J Surg 1994; 81: DePalma RG, Hart MT, Zanin L, Massarin EH. Physical varices and that reflux may be present without sympexamination, Doppler ultrasound and colour flow duplex scantoms Patients with symptomatic varicose veins ning: Guides to therapy for primary varicose veins. Phlebology should also have quality of life assessment done prior 1993; 8: McIrvine A, Corbett CR, Aston NO et al. The demonstration to surgery to examine the impact of the disease on of sapheno femoral junction incompetence; Doppler ultrasound their lives, as symptoms per se are a poor predictor of compared with clinical tests. Br J Surg 1984; 74: severity Burkitt DP. Varicose veins: facts and fantasy. Arch Surg 1976; 111: Colour duplex scanning for varicose vein surgery 17 McMullin GM, Coleridge, Smith PD. An evaluation of Dopdoes have significant cost implications, 39 it is therefore pler ultrasound and photoplethysmography in the investigation important to minimise these costs whilst still main- of venous insufficiency. AustNZJSurg 1992; 62: Katsamouris AN, Kardoulas DG, Gourtsoyiannis N. The taining the very high standards we expect from varnature of lower extremity venous insufficiency in patients with icose vein surgery. primary varicose veins. Eur J Vasc Endovasc Surg 1994; 8: Quigley FG, Raptis S, Cashman M, Faris IB. Duplex ultrasound mapping of sites of deep to superficial incompetence in primary varicose veins. AustNZJSurg 1992; 62: Smith JJ, Garratt AM, Guest M, Greenhalgh RM, Davies AH. Evaluating and improving health-related quality of life in patients with varicose veins. J Vasc Surg 1999; 30: Conclusion 21 Thibault PK, Lewis WA. Recurrent varicose veins. Part 1: Evaluation utilizing duplex venous imaging. J Dermatol Surg Oncol 1992; 18: Pre-operative duplex marking of the sites of deep to 22 Hanrahan LM, Kechejian GJ, Cordts PR et al. Patterns of superficial incompetence confers no additional benefit venous insufficiency in patients with varicose veins. Arch Surg 1991; 126: in terms of outcome in primary varicose vein surgery 23 Powell JT, GRUSATP. Mortality results for randomised confor long saphenous system varicose veins. The position trolled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. Lancet 1998; 352: is less clear for the short saphenous system and will be answered by a larger study specifically looking at 24 King ESJ. The genesis of varicose veins. Aust NZ J Surg 1950; this area. 20: 126.

8 Varicose Vein Surgery Eger SA, Casper SL. Etiology of varicose veins from an anatomic 33 Beaglehole R. Epidemiology of varicose veins. World J Surg aspect based on dissections of 38 adult cadavers. JAMA 1943; 1986; 10: : Nyamekye I, Shephard NA, Davies B, Heather BP, Earnshaw 26 van der Heijden FH, Bruyninckx CM. Preoperative colourof JJ. Clinicopathological evidence that neovascularisation is a cause coded duplex scanning in varicose veins of the lower extremity. recurrent varicose veins. Eur J Vasc Endovasc Surg 1998; 15: Eur J Surg 1993; 159: Duchosal F, Allerman H, Widmer LK, Breil H, Leu HJ. 35 Schultz-Ehrenburg U, Weindorf N, Matthes U, Hirche H. Varikosis Alter Korpergewicht. Z Kreislaufforsch 1968; 57: 380. [An epidemiologic study of the pathogenesis of varices. The 28 US Department of Health Education and Welfare.Health Bochum study I III]. Phlebologie 1992; 45: Statistics from the US National Health Survey. Chronic 36 Bradbury AW, Evans CJ, Allan P, Lee A, Ruckley CV. What are the symptoms of varicose veins? Edinburgh vein study cross conditions causing limitation of activities. United States July sectional population survey. BMJ 1999; 318: June Washington DC: Widmer LK. Prevalence and socio medical importance. Ob- 29 Trendelenberg F. Über die Unterbindung der Vena saphena servations in 4529 apparently health persons. In: Anonymous magna bei Unterschenkel Varicen. Beitr Klin Chir 1891; 7: 195. peripheral venous disorders. Bern: Hans Huber, Wali MA, Sheehan SJ, Colgan MP, Moore DJ, Shanik GD. 38 Evans CJ, Fowkes FGR, Ruckley CV, Lee A. Prevalence of Recurrent varicose veins. EAfrMedJ1998; 75: varicose veins and chronic venous insufficiency in men andn 31 Wills V, Moylan D, Chambers J. The use of routine duplex women in the general population: Edinburgh vein study. J scanning in the assessment of varicose veins. Aust NZ J Surg Epidemiol Community Health 1999; 53: ; 68: Jutley RS, Cadle I, Cross KS. Preoperative assessment of prim- 32 Dwerryhouse S, Davies B, Harradine K, Earnshaw JJ. Strip- ary varicose veins: a duplex study of venous incompetence. Eur ping the long saphenous vein reduces the rate of reoperation J Vasc Endovasc Surg 2001; 21: for recurrent varicose veins: five-year results of a randomized trial. J Vasc Surg 1999; 29: Accepted 21 January 2002