Presence of varices after operative treatment: a review

Download this issue Back to summary
Part 1: This is the first of 2 chapters that will comprise the ‘PREsence of Varices After operatIve Treatment (PREVAIT)’. These 2 chapters will  be published in the journal consecutively.
Michel Perrin, MD
Vascular Surgery, 26 Chemin de Decines, F-69680 Chassieu, France

Abstract

Background: PREsence of Varices after operatIve Treatment (PREVAIT ) occurs in 13% to 65% of patients and remains a debilitating and  costly problem. The first part of this review provides an overview of the current understanding of the etiology and pathogenesis of PREVAIT . Methods: A PubMed search was conducted in English and French for the years 2000-2013 by using keywords (Duplex scanning,  endothermal ablation, neovaricoses, REVA S, sclerotherapy, varices recurrence, varicose vein, varicose vein surgery).
Results: Epidemiology and socioeconomic consequences were analyzed according to the initial operative treatment. Then a classification of  possible mechanisms and causes for PREVAIT are classified in terms of tactical and technical errors, evolution of the disease, considering  that the systematic use of ultrasound investigation has minimized the former.

Conclusion: The cause and underlying mechanisms for  recurrences of varicose veins are poorly understood. Large prospective studies should be performed to clear up the picture.

Background

The presence of varicose veins after operative treatment is a common, complex, and costly problem for both the patients and the physicians who cope with venous diseases. An international consensus meeting was held in Paris in 1998 and guidelines were proposed for the  definition and description of REcurrent Varices After Surgery (REVA S).1 In a related article from 2000, 94 references dealing with recurrence after operative treatment or including information on its presence or absence after operative treatment were listed. Since then, 140 additional publications in English and French have been identified.2-141

Classical surgery, which used to be the most frequent operative procedure for treating varicose veins in the last decade, has been progressively taken over by chemical and thermal ablation procedures, and to a slight extent, by mini-invasive surgeries including CHIVA (French acronym for ambulatory conservative hemodynamic management of varicose veins)142 and ASVAL (French acronym for tributary varices phlebectomy under local anesthesia).143,144 Therefore, the experts of the VEIN-TERM transatlantic interdisciplinary consensus meeting suggested replacing the classical surgery-related acronym REVA S with PREVAIT (PREsence of Varices After Interventional Treatment).145

During the same meeting, the following terms were defined:

1. Recurrent varices: Reappearance of varicose veins in an area previously treated successfully.
2. Residual varices: Varicose veins remaining after treatment.
3. PREVAIT : PREsence of Varices (residual or recurrent) After Interventional Treatment.

The concept of PREVAIT was developed for two reasons: (i) it is often difficult to correctly classify the results of initial procedures done by  others and consequently to differentiate recurrent varices from residual varices; and (ii) the term REVA S was limited to patients previously  treated by surgery as previously mentioned. The term PREVAIT encompasses both recurrent and residual varicose veins after any kind of operative treatment including open surgery and endovenous procedures, either thermal or chemical.

Table I. REVAS Classification sheet. Modified after reference 98: Perrin et al. Eur J Vasc Endovasc Surg. 2006;32:326-333.

Table I. REVAS Classification sheet.
Modified after reference 98: Perrin et al. Eur J Vasc Endovasc
Surg. 2006;32:326-333.

It was also argued that the term ‘interventional treatment’ was not equivalent to the term ‘operative treatment,’ ecause even noninvasive therapies such as venoactive drugs or compression therapy may modify the natural history of varicose veins and be considered as ‘interventional.’

In 2000, a REVA S classification form was elaborated for future studies (Table I). The REVA S classification was hen subject to intraobserver and interobserver reproducibility,98 and then used in an international survey.95,97 A form similar to this should be adapted to PREVAIT for possible future studies.

AIM

The purpose of this review is to analyze all available data on PREVAIT in order to help physicians identify the best operative treatment, if any, likely to prevent PREVAIT . Such analysis might help build a revised classification, as mentioned above.

Material and methods

A PubMed search was conducted to retrieve published articles in English and French for the years 2000-2013 using the keywords varices  recurrence, REVA S, endothermal ablation, sclerotherapy, varicose vein surgery, varicose vein, duplex scanning, neovaricose, and their  counterparts in French. Abstracts were not selected, only publications dealing with PREVAIT were chosen, some of them focused on  PREVAIT patients, others concerned patients presenting with varices and operatively treated whose follow-up specified the absence or presence of varices.

Results

Since the REVA S publication,1 140 articles on recurrent varices have been published.2-141 29 randomized trials were added to the references from the REVA S  articles list, taking the total papers regarding randomized trials to 34.6,7,13,16,17,42,52,61,62,66,69,80,83,90,92,103,107-111,117,118,120,122,124,136,137,140,146-152 Epidemiologic data and socioeconomic consequences will be analyzed according to the initial procedures, which will be followed by a discussion of the possible mechanisms for PREVAIT occurrence.

Magnitude of REVAS occurrence

With open surgery
The most documented outcomes are provided by classical surgery, but most studies are retrospective. In a 34-year follow-up study, varicose veins were present in  77% of the lower limbs examined and were mostly symptomatic: 58% were painful, 83% had a tired feeling, and 93% showed a reappearance of edema.50

Two prospective studies concerning classical surgery are available with a follow-up of 5 years.72,133 In both studies, patients were preoperatively investigated with duplex scanning (DS) and treated by high ligation, saphenous trunk stripping, and stab avulsion. In the Kostas et al series, 28 out of 100 patients had PREVAIT after 5 years: recurrent varices mainly resulting from neovascularization in eight limbs (8/28, 29%), new varicose veins as a consequence of disease progression in  seven limbs (7/28, 25%), residual veins due to tactical errors (eg, failure to strip the great saphenous vein) in three limbs (3/28, 11%), and complex patterns in ten  limbs (10/28, 36%).72 In the Van Rij et al series, 127 limbs (CEAP class C2–C6) were evaluated postoperatively by clinical examination, DS, and air  plethysmography (APG). At the clinical evaluation, recurrence of varicose veins was progressive from 3 months (13.7%) to 5 years (51.7%). In line with clinical changes, a progressive deterioration in venous function was measured by APG and a recurrence of reflux was assessed by DS.133

These 2 studies showed that recurrence of varicose veins after surgery is common, even in highly skilled centers, and even if the clinical condition of most affected  limbs after surgery improved compared with ‘before surgery.’ Progression of the disease and neovascularization are responsible for more than half of the  recurrences. Rigorous evaluation of patients and assiduous surgical techniques might reduce the recurrence resulting from technical and tactical failures.

Table II. Rasmussen 3-year clinical and DS outcome and reoperation percentages. Modified after reference 111: Rassmusen et al. J Vasc Surg: Venous and Lym Dis. 2013;1:349-356.

Table II. Rasmussen 3-year clinical and DS outcome and reoperation percentages.
Modified after reference 111: Rassmusen et al. J Vasc Surg: Venous and Lym Dis. 2013;1:349-356.

Table III. Pre and postoperative VCSS and AVVSS according to operative treatment. Modified after reference 111: Rassmusen et al. J Vasc Surg: Venous and Lym Dis. 2013;1:349-356.

Table III. Pre and postoperative VCSS and AVVSS according to operative treatment.
Modified after reference 111: Rassmusen et al. J Vasc Surg: Venous and Lym Dis. 2013;1:349-356.

In a four arm randomized controlled trial (RCT) by Rassmussen et al, endovenous laser ablation (EVLA ), radiofrequency ablation (RFA), ultrasound-guided foam  sclerotherapy (UGFS), and surgical stripping for great saphenous varicose veins (GSV ) were compared with a 3-year follow-up, the rate of PREVAIT was reported  in each arm (Table II).111 There was no significant difference between the 4 procedures (P=0.29) in terms of clinical recurrence, but the presence of persisting  reflux in the GSV was significantly higher in UGFS compared with the other 3 methods (P<0.0001) as well as the reoperation rate (P<0.0001).

Regardless of the procedure used, the severity of varicose disease as assessed with the Venous Clinical Severity Score (VCSS) was significantly reduced, and the  quality of life using the Aberdeen Varicose Veins Severity Score (AVV SS) was significantly improved after all operative treatments no matter which procedure was  used (P<0.0001; Table III)

With radiofrequency ablation
From a multicenter prospective study, recurrence rates after RFA with ClosurePlus® were reported. At the 5-year follow-up, PREVAIT was estimated at 27.4%.84 A  3-year follow-up RCT comparing ClosureFast®-RFA of the GSV with or without treatment of calf varicosities did not document the PREVAIT rate, but only the obliteration rate on DS investigation, venous clinical severity score (VCSS), and the presence of symptoms.102 In the four arm study by Rassmussen et al,111 there was no statistical difference regarding PREVAIT rates between RFA and the other operative procedures (P=0.29; Table II).

With endovenous laser ablation
At the 2-year follow-up, a RCT by Rass et al found no significant difference (P=0.15) when comparing EVLA with classical surgery (EVLA 16.2% vs 23.1%).107 An Italian group reported a PREVAIT rate of 6% at month 36.2 In a RCT comparing EVLA with GSV stripping with a 5-year follow-up, PREVAIT was reported in 36% and 37% of patients, respectively, with no statistical difference between groups (P=0.9). In this study, reoperative treatment was performed in 38.6% and 37.7%, respectively, mainly by UGFS.110 Again in the four arm study by Rassmussen et al,111 there was no statistical difference regarding PREVAIT rates between EVLA and the other operative procedures (P=0.29; Table II).

Ultrasound-guided foam sclerotherapy
Hamel-Desnos et al reported a 36% and 37% recanalization rate at the 2-year follow-up with UGFS, one injection with 1% and 3% polidocanol foam,  espectively.62 In a RCT of UGFS vs surgery for the incompetent GSV with a follow-up of 2 years, PREVAIT was identified in 9% vs 11.3%, respectively. P=0.407,   which is not significant. Conversely, reflux was significantly higher in UGFS (P=0.003).118

In the British long-term RCT by Kalodiki et al of UGFS combined with sapheno-femoral ligation vs standard surgery for GSV, clinical severity of venous disease assessed by VCSS and venous segmental disease score (VSDS) were equally reduced in both groups, and the quality of life equally improved as well (using AVV Q and 36-Item Short-Form).69 Unfortunately, PREVAIT was not reported in this study.

With procedures saving the saphenous trunk

CHIVA

PREVAIT was assessed when using the CHIVA method vs classical surgery in 2 RCT’s with a follow-up of 5 and 10 years.16,90 In both studies, the Hobbs classification was used to assess PREVAIT.148,149

If we add failure (presence of VV >0.5 cm) and slightly improved patients in terms of cosmetic appearance (presence of VV <0.5 cm), the outcomes were as follows: (i) At 5 years postsurgery, the PREVAIT rate in the group operated by stripping was 70.7% vs 55.6% in the CHIVA group (P>0.001).90 In the 10-year follow-up RCT by Carandina, the recurrence rate of varicose veins was significantly higher in the stripping group compared with the CHIVA group (CHIVA , 18%; stripping, 35%; P<0.04 Fisher’s exact test). The associated risk of recurrence at 10 years was doubled in the stripping group (odds ration [OR], 2.2; 95%  confidence interval [CI] 1-5; P=0.04).16 In both RCTs, the recurrence rate was lower with CHIVA ,90,16 yet there is a great discrepancy between the studies, PREVAIT was unexpectedly higher in the 5-year follow-up RCT,90 compared with the 10-year follow-up.16

ASVAL

No published data is available regarding the mid-term results.

Socioeconomic consequences
No socioeconomic data on PREVAIT has been published. When a redo surgery is performed, the cost is higher than the first surgery because of the number of peri- and postoperative complications. In one observational study, 40% of patients had complications after classical surgery for PREVAIT.64

Possible mechanisms leading to PREVAIT
They must be classified in 2 groups: tactical errors and technical problems.

Tactical errors
Tactical errors are common to all operative treatments. It includes wrong or incomplete diagnosis of the extent and/or location of varices, source of reflux,  onidentification of deep venous anomalies including pelvic reflux (Figures 1, Figure 2), primary vein compression or reflux, and posthrombotic syndrome. Fortunately, the systematic use of DS before any operative treatment has minimized this cause of error. In most of the articles published before systematic use of preoperative DS, tactical error was the most frequent mechanism leading to PREVAIT .

Figure1. PREVAIT clinical aspect. A. Pelvic vein leak. B. Selective pelvic venography from the same patient as A. (Courtesy of Drs Monedero and Zubicoa).

Figure1. PREVAIT clinical aspect.
A. Pelvic vein leak. B. Selective pelvic venography from the same
patient as A. (Courtesy of Drs Monedero and Zubicoa).

Figure 2. Selective pelvic venography. After a Valsalva maneuver. Reflux through the obturator vein feeding the nonsaphenous vein network. (Courtesy of Drs Monedero and Zubicoa).

Figure 2. Selective pelvic venography.
After a Valsalva maneuver. Reflux through the obturator vein
feeding the nonsaphenous vein network.
(Courtesy of Drs Monedero and Zubicoa).

There is a consensus on the fact that saphenous ablation provides a better outcome when saphenous trunk incompetence is present and when classical surgery, thermal or chemical, is performed. Yet, the proponents of the CHIVA and ASVAL procedures contest this point by arguing that trunk conservation would provide good results. In the CHIVA procedure, the argument is that the preservation of the saphenous trunks together with sparing of their functions (cutaneous and subcutaneous drainage) is allowed thanks to appropriate shunt disconnections that breaks the higher-than-normal hydrostatic pressure and subsequently improves hemodynamics.16,90,142 In the ASVAL method, the ablation  of the reservoir incompetent tributaries leads to a reduction in the reflux in the saphenous trunk.143,144

Technical problems related to the first operative treatment (surgery, thermal, or chemical ablation)
Such problems can overlap in the same patient, and some are specific and related to the procedure used, while others are identified no matter what procedure was used.

Surgery
The most frequent technical error quoted in classical surgery was non flush ligation at the saphenofemoral junction (SFJ; Figure 3) or at the saphenopopliteal junction (SPJ; Figure 4). This point is now controversial as some series with conservation of the SFJ claim to achieve excellent results including patients with incompetent terminal valve.152 Several authors continue to state that non flush ligation of the saphenous termination is responsible for frequent recurrence,41,52 particularly over the long-term.55-57 In the CHIVA  technique, PREVAIT would be mainly related to wrong preoperative marking and inappropriate technique.90

Figure 3. PREVAIT clinical aspect. A. Massive groin recurrence related to non flush high ligation in a patient with an incompetent GSV terminal valve. B. Same patient with a B mode ultrasound. The terminal valve is identified at the saphenofemoral junction. (Courtesy of Dr Gillet). C. Same patient with a color duplex ultrasound. Massive reflux induced by a Valsalva maneuver. (Courtesy of Dr Gillet).

Figure 3. PREVAIT clinical aspect.
A. Massive groin recurrence related to non flush high ligation
in a patient with an incompetent GSV terminal valve. B. Same
patient with a B mode ultrasound. The terminal valve is identified
at the saphenofemoral junction. (Courtesy of Dr Gillet). C. Same
patient with a color duplex ultrasound. Massive reflux induced
by a Valsalva maneuver. (Courtesy of Dr Gillet).

Figure 4. PREVAIT clinical aspect. A. Popliteal fossa massive recurrence related to non-flush high ligation in a patient with an incompetent SSV terminal valve. B. Postoperative duplex scanning identified reflux in the SSVS, which feeds the varicose network after the compressiondecompression maneuver. (Courtesy of Dr Gillet).

Figure 4. PREVAIT clinical aspect.
A. Popliteal fossa massive recurrence related to non-flush
high ligation in a patient with an incompetent SSV terminal
valve. B. Postoperative duplex scanning identified reflux in the
SSVS, which feeds the varicose network after the compressiondecompression
maneuver. (Courtesy of Dr Gillet).

Thermal ablation
Inadequate technique consisting mainly of delivering insufficient energy, irradiance, or fluence in laser or radiofrequency procedures should be responsible for  short or long-term recanalization of the treated vein.

Chemical ablation
Inadequate technique as well as inappropriate sclerosing agent dose should be responsible for short or long-term recanalization of the treated vein.

Technical problems not related to initial treatment
The neovascularization phenomenon was discovered 25 years ago, but remains not fully elucidated.152 It occurs mainly at the SFJ (Figure 5) and less frequently  at the SPJ (Figures 6), and is considered, in many articles, as the main cause of PREVAIT after correct classical surgery.28,29,134,153,154 El Wajeh et al contests the term neovascularization and favors adaptive dilatation of preexisting venous channels (vascular remodeling), probably in response to abnormal  hemodynamic forces.43 According to Lemasle et al, this phenomenon is related to preexisting anatomical anomalies.79 Egan et al minimizes its frequency as well as its importance in groin recurrence. 41 However, neovascularization has been reported not only in procedures including SFJ or SPJ ligation, but also after thermal ablation,76 albeit at a lower frequency.71,124

Figure 5. PREVAIT clinical aspect. A. A varicose network at the thigh just below a previous groin incision related to neovascularization. B. Same patient with a duplex scan. Small refluxive veins identified above the CFV after a Valsalva maneuver. (Courtesy of Dr Gillet).

Figure 5. PREVAIT clinical aspect.
A. A varicose network at the thigh just below a previous groin
incision related to neovascularization. B. Same patient with a
duplex scan. Small refluxive veins identified above the CFV after
a Valsalva maneuver. (Courtesy of Dr Gillet).

Figure 6. PREVAIT clinical aspect. A. A varicose network at the popliteal related to neovascularization. B. Same patient with a duplex scan. Varicose network above a refluxive popliteal vein (Courtesy of Dr Gillet).

Figure 6. PREVAIT clinical aspect.
A. A varicose network at the popliteal related to neovascularization.
B. Same patient with a duplex scan. Varicose network
above a refluxive popliteal vein (Courtesy of Dr Gillet).

Evolution of the disease
It should never be forgotten that superficial venous disease is a chronic condition that tends to progress over time.104 In other words, previously unaffected  superficial veins or perforators may become incompetent. Varices may develop in the same territory initially treated including saphenous tributaries that were not incompetent at the time of the operative treatment or in another superficial vein territory.

Risks factors for chronic venous disease progression and, in particular, varices have been investigated in many prospective studies.155 However, underpinnings and constitution risk factors for disease progression are still poorly understood. It is generally accepted that there is a strong family predisposition, not only for presenting varicose veins, but also for developing recurrence related to disease evolution. The precise nature of the genetic basis for this family predisposition is far from clear. To shed more light on this issue, it will not be sufficient to study single genes, potentially implicated in varices. Instead, genome wide association  studies will be needed using very large sample sizes to further unravel the genetic basis of varices and chronic venous insufficiency.156

Corresponding author

Corresponding author

1. Perrin M, Guex JJ, Ruckley CV, et al.  Recurrent varices after surgery (REVA S), a  consensus document. Cardiovasc Surg.  2000;8:233-245.

2. Agus B, Mancini S, Magi G. The first  1000 cases of Italian Endovenous-Laser  Working Group (IEWG): Rational, and  long terms outcomes for the 1999-2003  period. Int Angiology. 2006;25:209-215.

3. Ali SM, Callam MJ. Results and  signifiance of colour duplex assessment  on the deep venous system in recurrent  Varicose veins. Eur J Vasc Endovasc Surg.  2007;34:97-101.

4. Allaf N, Welch M. Recurrent varicose  veins: inadequate surgery remains a  problem. Phlebology. 2005;20:138-114.

5. Allegra C, Antignani PL, Carlizza A.  Reccurent varicose veins following  surgical treatment: our experience with  five years follow-up. Eur J Vasc Endovasc  Surg. 2007;33:751-756.

6. Belcaro G, Nicolaides AN, Cesarone  NM, et al. Flush ligation of the saphenofemoral  junction vs simple distal ligation  A randomised, 10-year, follow-up. The  safe study. Angéiologie. 2002;54:19-23.

7. Belcaro G, Cesarone NM, Di Renzo  A, et al. Foam sclerotherapy, surgery,  sclerotherapy and combined treatment  for varicose veins A 10-year, prospective,  randomised, controlled trial (VEDICO trial  ). Angiology. 2003;54:307-315.

8. Bhatti TS, Whitman B, Harradine  K, et al. Causes of re-recurrence  after polytetrafluoroethylene patch  saphenoplasty for recurrent varicose  veins. Br J Surg. 2000;87:1356-1360.
9. Beresford T, Smith JJ, Brown L,  Greenhalgh RM, Davies AHA.  Comparison of health-related quality of  life of patients with primary and recurrent  varicose veins. Phlebology. 2003;18:35-  37.

10. Blomgren L, Johansson G, Dahlberg-  Akerman A, et al. Recurrent varicose  veins: Incidence, risk factors and groin  anatomy. Eur J Vasc Endovasc Surg.  2004;27:269-274.

11. Blomgren L, Johansson G, Dahlberg-  Akerman, et al. Changes in superficial  and perforating vein reflux after varicose  vein surgery. J Vasc Surg. 2005;42:315-  320.

12. Brake M. Pathogenesis and etiology  of recurrent varicose veins. J Vasc Surg.  2013;57:860-868.

13. Bountourouglou DG, Azzam M,  Pathmarajh M, et al. Ultrasound guided  foam sclerotherapy combined with  sapheno-femoral ligation compared to  surgical treatment of varicose veins: early  results of a randomised contolled trial.  Eur J Vasc Endovasc Surg. 2006;31:93-  100.

14. Bridget M, Donnelly M, Tierney S.  Recurrent varicose veins after surgery. Br J  Surg. 2006;38:49.

15. Cardia G, Catalano G, Rosafio I,  Granatiero M, De Fazio M. Recurrent  varicose veins of the legs. Analysis of a  social problem. G Chir. 2012;33:450-  454.

16. Carandina S, Mari C, De Palma M, et al.  Varicose vein stripping vs haemodynamic  correction (CHIVA ): a long term  randomised trial. Eur J Vasc Endovasc  Surg. 2008;35:230-237.

17. Carradice D, Mekako AI, Mazari  FA, Samuel N, Hatfield J, Chetter IC.  Randomized clinical trial of endovenous  laser ablation compared with  conventional surgery for great saphenous  varicose veins. Br J Surg. 2011:98:501-  510.

18. Castenmiller PH, de Leur K, de Jong  TE, van der Laan L. Clinical results after  coil embolization of the ovarian vein  in patients with primary and recurrent  lower-limb varices with respect to vulval  varices. Phlebology. 2013;28:234-238.

19. Chong CS, Toh BC, Chiang V, Cheng SC,  Lee CW. Pattern of recurrence of lower  limb varicose veins post EVLT : a single  center experience. Ultrasound Med Biol.  2011;37:S159.

20. Smith PC. Chronic venous disease treated  by ultrasound guided foam sclerotherapy.  Eur J Vasc Endovasc Surg. 2006;32:577-  583.

21. Creton D. Surgery of great saphenous  vein recurrences: the presence of diffuse  varicose veins without a draining residual  saphenous trunk is a factor of poor  prognosis for long-term results.  J Phlebolymphology. 2002;2:83-89.

22. Creton D. Surgery for recurrent  saphenofemoral incompetence using  expanded polytetrafluoroethylene patch  interposition in front of the femoral vein:  long-term outcome in 119 extremities.  Phlebology. 2002;16:93-97.

23. Creton D. 125 réinterventions pour  récidives variqueuses poplitées  après exérèse de la petite saphène.  Hypothèses anatomiques et  physiologiques du mécanisme de la  récidive. J Mal Vasc. 1999;24:30-36.

24. Creton D, Uhl JF. Foam Sclerotherapy  Combined with Surgical Treatment for  Recurrent Varicose Veins: Short Term  Results. Eur J Vasc Endovasc Surg.  2007;33:619-624.

25. Creton D. A non draining saphenous  system is a factor of poor prognosis  for long-term results in surgery of great  saphenous vein recurrences. Dermatol  Surg. 2004;30:744-749.

26. Darke SG, Baker SJ. Ultrasound-guided  foam sclerotherapy for the treatment of  varicose veins. Br J Surg. 2006;93:969-  974.

27. Darvall KA, Bate GR, Adam DJ, Silverman  SH, Bradbury AW. Duplex ultrasound  outcomes following ultrasound-guided  foam sclerotherapy of symptomatic  recurrent great saphenous varicose veins.  Eur J Vasc Endovasc Surg. 2011;42:107-  114.

28. De Maeseneer MG. The role of  postoperative neovascularisation in  recurrence of varicose veins: from  historical background to today’s  evidence. Acta Chirurgica Belgica.  2004;104:281-287.

29. De Maeseneer MG, Ongena KP, Van  den Brande F, Van Schil PE, De Hert  SG, Eyskens EJ. Duplex ultrasound  assessment of neovascularisation  after saphenofemoral or saphenopoliteal  junction ligation. Phlebology.  1997;12:64-68.

30. De Maeseneer MG, Tielliu IF, Van Schil  PE, De Hert SG, Eyskens EJ. Clinical  relevance of neovascularization on  duplex ultrasound in long-term follow-up  after varicose vein operation. Phlebology.  1999;14:118-122.

31. De Maeseneer MG, Giuliani DR, Van  Schil PE, De Hert SG. Can interposition of  a silicone implant after sapheno-femoral  ligation prevent recurrent varicose veins.  Eur J Vasc Endovasc Surg. 2002;24:445-  449.

32. De Maeseneer MG, Vandenbroeck  CP, Van Schil PE. Silicone patch  saphenoplasty to prevent repeat  recurrence after surgery to treat recurrent  saphenofemoral incompetence: Longterm  follow-up study. J Vasc Surg.  2004;40:98-105.

33. De Maeseneer MG. Recurrent varicose  veins after surgery [thesis]. Antwerpen:  Universiteit Antwerpen; 2005.

34. De Maeseneer MG, Vandenbroeck CP,  Hendriks JM, Lauwers PR, Van Schil PE.  Accuracy of duplex evaluation one year  after varicose vein surgery to predict  recurrence at the sapheno-femoral  junction after five years. Eur J Vasc  Endovasc Surg. 2005;29:308-312.

35. De Maeseneer M. Recurrence of varicose  veins is usually defined as re-emergence  of varicosities after previous surgery.  Angeiologie. 2006;58:28-31.

36. De Maeseneer MG, Philipsen TE,  Vandenbroeck CP, et al. Closure  of the cribriform fascia: an efficient  anatomical barrier against  postoperative neovascularisation at the  saphenofemoral junction? (A prospective  study). Eur J Vasc Endovasc Surg.  2007;34:361-366.

37. De Maeseneer M. Surgery for recurrent  varicose veins: Toward a less-invasive  approach? Perspect Vas Surg Endovasc  Ther. 2011;23:244-249.

38. De Maeseneer MG, Cavezzi A. Etiology  and pathophysiology of varicose vein  recurrence at the saphenofemoral or  saphenopopliteal junction: an update.  Veins and Lymphatics. 2012;1:4.

39. De Maeseneer M, Pichot O, Cavezzi  A, et al; Union Internationale de  Phlebologie. Duplex ultrasound  investigation of the veins of the lower  limbs after treatment for varicose veins:  UIP Consensus Document. Eur J Vasc  Endovasc Surg. 2011;42:89-102.

40. Edwards AG, Donaldson D, Bennets C,  Mitchell DC. The outcome of recurrent  varicose veins surgery: the patient’s  perspective. Phlebology. 2005;20:57-59.

41. Egan B, Donnelly M, Bresnihan M,  Tierney S, Feeley M. Neovascularization:  An innocent bystander in recurrent  varicose veins. J Vasc Surg.  2006;44:1279-1284.

42. Elkaffas KH, Elkashef O, Elbaz W.  Great saphenous vein radiofrequency  ablation versus standard stripping in the  management of primary varicose veinsa  randomized clinical trial. Angiology.  2010;62:49-54.

43. El Wajeh Y, Giannoukas AD, Gulliford  CJ, Suvarna SK, Chan P. Saphenofemoral  venous channels associated with  recurrent varicose veins are not  neovascular. Eur J Vasc Endovasc Surg.  2004;28:590-594.

44. Englund R. Duplex scanning for  recurrent varicose veins. Aust N Z J Surg.  1996;66(9):618-620.

45. Farrah J, Shami SK. Patterns of  incompetence in patients with recurrent  varicose veins: a duplex ultrasound study.  Phlebology. 2001;16:34-37.
46. Fassiadis N, Kianifard B, Holdstock JM,  Whiteley MS. A novel approach to the  treatment of recurrent varicose veins. Int  Angiol. 2002;21(3):275-276.

47. Ferrara F, Bernbach HR. La sclérothérapie  des varices récidivées. Phlébologie.  2005;58:147-150.

48. Fischer R, Chandler JG, De Maeseneer  MG, et al. The unresolved problem of  recurrent saphenofemoral reflux. J Am  Coll Surg. 2002;195:80-94.

49. Fischer R, Linde N, Duff C, Jeanneret  C, Chandler JG, Seeber P. Late  recurrent saphenofemoral junction  reflux after ligation and stripping of the  greater saphenous vein. J Vasc Surg.  2001;34:236-240.

50. Fischer R, Linde N, Duff C. Cure and  reappearance of symptoms of varicose  veins after stripping operation–A 34 year  follow-up. J Phlebology. 2001;1:49-60.

51. Fischer R, Chandler JG, Stenger  D, Puhan MA, De Maeseneer  MG, Schimmelpfennig L. Patients  characteristics and physician-determined  variables affecting saphenofemoral reflex  recurrence after ligation and stripping of  the great saphenous vein. J Vasc Surg.  2006;43:81-87.

52. Frings N, Nelle A, Tran P, Fischer R,  Krug W. Reduction of neoreflux after  correctly performed ligation of the  saphenofemoral junction. A randomized  trial. Eur J Vasc Endovasc Surg.  2004;28:246-252.

53. Gad MA, Saber A, Hoklam EN.  Assessment of causes and patterns of  recurrent varicose veins after surgery.  North Am J Med Sci. 2012;4:45-48.

54. Gauw SA, Pronk P, Mooij MC, Gaastra  MTW, Lawson JA, van Vlijmen-van Keulen  CJ. Detection of varicose vein recurrence  by duplex ultrasound: intra- and  interobserver reproducibility. Phlebology.  2013;28:109-111.

55. Geier B, Stücker M, Hummel T, et al.  Residual stumps associated with inguinal  varicose vein recurrence: a multicenter  study. Eur J Vasc Endovasc Surg.  2008;36:207-210.

56. Geier B, Olbrich S, Barbera L, Stücker M,  Mumme A. Validity of the macroscopic  identification of neovascularization  at the saphenofemoral junction by  the operating surgeon. J Vasc Surg.  2005;41:64-68.

57. Geier B, Mumme A, Hummel T, Marpe B,  Stücker M, Asciutto G. Validity of duplexultrasound  in identifying the cause  of groin recurrence after. J Vasc Surg.  2009;49:968-972.

58. Gillet JL, Perrin M. Exploration echodoppler  des récidives variqueuses postchirurgicales.  Angéiologie. 2004;56:26-  31.

59. Gillet JL. Traitement des récidives  chirurgicales de la jonction saphènofémorale  et saphéno-poplitée par echosclérose.  Phlébologie. 2003;56:241-245.

60. Gohel MS, Davies AH. Choosing  between varicose vein treatments: looking  beyond occlusion rates. Phlebology.  2008;23:51-52.

61. Haas E, Burkhardt T, Maile N. Recurrence  rate by neovascularisation following a  modification of long saphenous vein  operation in the groin: a prospectived  randomized duplex-ultrasound controlled  study. Phlebologie. 2005;34:101-104.

62. Hamel-Desnos C, Ouvry P, Benigni  JP, et al. Comparison of 1% and 3%  polidocanol foam in ultrasound guided  sclerotherapy of the great saphenous  vein: A randomized, double-blind trial  with 2 year-follow-up:The 3/1 study. Eur J  Vasc Endovasc Surg. 2007;34:723-729.

63. Hartman K, Klode J, Pfister R, et al.  Recurrent varicose veins: Sonographybased  re-examination of 210 patients  14 years after ligation and saphenous  stripping. VASA. 2006;35:21-26.

64. Hayden A, Holdsworth J. Complications  following re-exploration of the groin for  recurrent varicose veins. Ann R Coll Surg  Engl. 2001;83:272-273.

65. Heim D, Negri M, Schlegel U, De  Maeseneer M. Resecting the great  saphenous stump with endothelial  inversion decreases neither  neovascularization nor thigh varicosity  recurrence. J Vasc Surg. 2008;47:1028-  1032.

66. Hinchliffe RJ, Ubhi J, Beech A, Ellison  J, Braithwaite BD. A prospective  randomised controlled trial of VNUS  Closure versus surgery for the treatment  of recurrent long saphenous varicose  veins. Eur J Vasc Endovasc Surg.  2006;31:212-218.

67. Jiang P, van Rij AM, Christie R, Hill G,  Solomon C, Thomson I. Recurrent varicose  veins: patterns of reflux and clinical  severity. Cardiovasc Surg. 1999;7:322-  329.

68. Kakkos SK, Bountouroglou DG, Azzam  M, Kalodiki E, Daskalopoulos M,  Geroulakos G. Effectiveness and safety  of ultrasound-guided foam sclerotherapy  for recurrent varicose veins: immediate  results.  J Endovasc Ther. 2006;13:357-364.

69. Kalodiki E, Lattimer CR, Azzam M,  Shawish E, Bountouroglou D, Geroulakos  G. Long term results of a randomized  controlled trial on ultrasound guided  foam sclerotherapy combined with  sapheno-femoral ligation vs standard  surgery for varicose veins. J Vasc Surg.  2012;55:451-457.

70. Kambal A, De’ath AD, Albon H, Watson  A, Shandall A, Greenstein D. Endovenous  laser ablation for persistent and recurrent  venous ulcers after varicose vein surgery.  Phlebology. 2008;23:193-195.

71. Kianifard B, Holdstock JM, Whiteley MS.  Radiofrequency ablation (VNUS closure)  does not cause neo-vascularisation at  the groin at one year: results of a case  controlled study. Surgeon. 2006;4:71-74.

72. Kostas T, Loannou CV, Toulouopakis E, et  al. Recurrent varicose veins after surgery:  A new appraisal of a common and  complex problem in vascular surgery. Eur  J Vasc Endovasc Surg. 2004;27:275-282.

73. Kostas TT , Ioannou CV, Veligrantakis  M, Pagonidis C, Katsamouris AN. The  appropriate length of great saphenous  vein stripping should be based on the  extent of reflux and not on the intent to  avoid saphenous nerve injury. J Vasc  Surg. 2007;46:1234-1241.

74. Kofoed SC, Qvamme GM, Schroeder  TV , Jakobsen BH. Causes of need for  reoperation following surgery for varicose  veins in Denmark. Ugeskr Laeger.  1999;8:779-783.

75. Labropoulos N, Touloupakis E,  Giannoukas AD, Leon M, Katsamouris A,  Nicolaides AN. Recurrent varicose veins:  investigation of the pattern and extent  of reflux with color flow duplex scanning.  Surgery. 1996;119:406-409.

76. Labropoulos N, Bhatti A, Leon L,  Borge M, Rodriguez H, Kalman  P. Neovascularization after great  saphenous ablation. Eur J Vasc Endovasc  Surg. 2006;31:219-222.

77. Lane RJ, Cuzzilla ML, Coroneos JC,  Phillips MN, Platt JT. Recurrence rates  following external valvular stenting of the  saphenofemoral junction: a contralateral  stripping of the great saphenous vein.  Eur J Vasc Endovasc Surg. 2007;34:595-  603.

78. Leal Monedero J, Zubicoa Ezpeleta  S, Castro Castro J, Calderon Ortiz M,  Sellers Fernandez G. Embolization  treatment of recurrent varices of pelvic  origin. Phlebology. 2006;21:3-11.

79. Lemasle Ph, Lefebvre-Villardebo M, Uhl JF,  Vin F, Baud JM. Récidive variqueuse postopératoire:  et si la neovascularisation  inguinale n’était que le développement  d’un réseau pré-existant. Phlébologie.  2009:62:42-48.

80. Lurie F, Creton D, Eklof B, et al.  Prospective randomized study of  endovenous radiofrequency obliteration  (Closure) versus ligation and vein  stripping (EVOLV eS) two-year follow-up.  Eur J Vasc Endovasc Surg. 2005;29:67-  73.

81. Lv W, Wu XJ, Collins M, Han ZL, Jin  X. Analysis of a series of patients with  varicose vein recurrence. J Int Med Res.  2012;40:1156-1165.

82. Mc Donagh B, Sorenson S, Gray C,  et al. Clinical spectrum of recurrent  postoperative varicose veins and efficacy  of sclerotherapy management using  the compass technique. Phlebology.  2003;18:173-185.

83. Menyhei G, Gyevnár Z, Arató E, Kelemen  O, Kollár L. Conventional stripping versus  cryostripping: a prospective randomised  trial to compare improvement in quality  of life and complications. Eur J Vasc  Endovasc Surg. 2008;35:218-223.

84. Merchant RF, Pichot O; for the Closure  study group. Long-term outcomes of  endovenous radiofrequency obliteration  of saphenous reflux as a treatment for  superficial venous insufficiency. J Vasc  Surg. 2005;42:502-509.

85. Mikati A. Indications et résultats de  la ligature coelioscopique des veines  perforantes incontinentes dans les  récidives variqueuses compliquées.  Phlébologie. 2010;63:59-67.

86. Milone M, Salvatore G, Maietta P, Sosa  Fernandez LM, Milone F. Recurrent  varicose veins of the lower limbs after  surgery. Role of surgical technique  (stripping vs. CHIVA ) and surgeon’s  experience. G Chir. 2011;32:460-463.  87. Mouton WG, Bergner M, Zehnder T,

von Wattenwyl R, Naef M, Wagner HE.  Recurrence after surgery for varices in the  groin is not dependent on body mass  index. Swiss Med Wkly. 2008;138(11-  12):186-188.

88. Nwaejike N, Srodon PD, Kyriakides  C. Endovenous laser ablation for the  treatment of recurrent varicose vein  disease – A single centre experience. Int J  Surg. 2010;8:299-301.

89. O’Hare JL, Parkin D, Vandenbroeck  CP, Earnshaw JJ. Mid-term results of  ultrasound guided foam sclerotherapy  for complicated and uncomplicated  varicose veins. Eur J Vasc Endovasc Surg.  2008;36:109-113.

90. Parés JO, Juan J, Tellez R, et al. Varicose  vein surgery. Stripping versus the CHIVA  method: a randomized controlled trial.  Ann Surg. 2010;251:624-631.

91. Pavei P, Vecchiato M, Spreafico G, et  al. Natural history of recurrent varices  undergoing reintervention: a retrospective  study. Dermatol Surg. 2008;34:1676-  1682.

92. Perala J, Rautio T, Biancari F, et al.  Radiofrequency endovenous obliteration  versus stripping of the long saphenous  vein in the management of primary  varicose veins: 3-year outcome of a  randomized study. Ann Vasc Surg.  2005;19: 669-72.

93. Perrin M. Recurrent varicose veins after  surgery. Phlebolymphology. 31:14-20.

94. Perrin M. Recurrent varices after surgery.  Hawaii medical journal. 2000;59:214-  216.

95. Perrin M, Labropoulos N, Leon LR.  Presentation of the patient with recurrent  varices after surgery (REVA S). J Vasc Surg.  2006;43:27-34.

96. Perrin M, Gillet JM. Management of  recurrent varices at the politeal fossa  after surgical treatment. Phlebology.  2008:23:64-68.

97. Perrin M. Le profil du patient REVA S:  résultats d’une enquête internationale.  Angéiologie. 2006;58:44-45.

98. Perrin M, Allaert, FA. Intra-and interobserver  reproducibility o the recurrent  varicose veins after surgery (REVA S)  classification. Eur J Vasc Endovasc Surg.  2006;32:326-333.

99. Perrin M, Gillet JL. Récidive de varices  à l’aine et à la fosse poplitée après  traitement chirurgical. J Mal Vasc.  2006;31:236-246.

100. Pittaluga P, Chastanet S, Locret T, Rousset  O. Retrospective evaluation of the need  of a surgery at the groin for the surgical  treatment of varicose vein. J Vasc Surg.  2010;51:1442-1450.

101. Pourhassan S, Zarras K, Mackrodt  HG, Stock W. Recurrent varicose veins.  Surgical procedure-results. Zentralbl Chir.  2001;126(7):522-525.

102. Proebstle TM, Aim J, Göckeritz O, et al;  European Closure Fast Clinical Study  Group. Three-year European follow-up  of endovenous radiofrequency-powered  segmental thermal ablation of the  great saphenous vein with or without  treatment of calf varicosities. J Vasc Surg.  2011;54:146-152.

103. Pronk P, Gauw SA, Mooij MC, et al.  Randomised Controlled Trial Comparing  Sapheno-Femoral Ligation and Stripping  of the Great Saphenous Vein with  Endovenous Laser Ablation (980 nm)  Using Local Tumescent Anaesthesia: One  Year Results. Eur J Vasc Endovasc Surg.  2010;40:649-656.

104. Rabe E, Pannier F, Ko A, Berboth  G, Hoffmann B, Hertel S. Incidence  of varicose veins, chronic venous  insufficiency, and progression of the  disease in the Bonn Vein Study II. J Vasc  Surg. 2010;51:791.

105. Rashid HI, Ajeel A, Tyrrell MR. Persistent  popliteal fossa reflux following  saphenopopliteal disconnection. Br J  Surg. 2002;89:748-751.

106. Reich-Schupke S, Mumme A, Altmeyer  P, Stuecker M. Expression with stump  recurrence and neovascularization  after varicose vein surgery a pilot study.  Dermatol Surg. 2011;37:480-485.

107. Rass K, Frings N, Glowack P, et al.  Comparable effectiveness of endovenous  laser ablation and high ligation with  stripping of the great saphenous vein.  Arch Dermatol. 2012;148:49-58.

108. Rasmussen LH, Bjoern L, Lawaetz  M, Lawaetz B, Blemings A, Eklöf B.  Randomized trial comparing endovenous  laser ablation with stripping of the great  saphenous vein: clinical outcome and  recurrence after 2 years. Eur J Vasc  Endovasc Surg. 2010;39:630-635.

109. Rasmussen LH, Lawaetz M, Bjoern  L, Vennits B, Blemings A, Eklof B.  Randomized clinical trial comparing  endovenous laser ablation,  radiofrequency ablation, foam  sclerotherapy and surgical stripping for  great saphenous varicose veins. Br J  Surg. 2011;98:1079-1087.

110. Rasmussen L, Lawaetz M, Bjoern L,  Blemings A, Eklof B. Randomized  trial comparing endovenous laser  ablation and stripping of the great  saphenous vein with clinical and duplex  outcome after 5-years. J Vasc Surg.  2013;58(2):421-426.

111. Rasmussen LA , Lawaetz M, Bjoern  L, et al. Randomized clinical trial  comparing endovenous laser ablation,  radiofrequency ablation, foam  sclerotherapy and surgical stripping for  great saphenous varicose veins with  3-year follow-up. J Vasc Surg: Venous  and Lym Dis. 2013;1:349-356.

112. Raussi M, Pakkanen J, Varlia E, Kupi H,  Saarinen J. Transilluminated powered  phlebectomy in the treatment of primary  and recurrent varicose disease: sixmonth  follow-up of 135 legs. Phlebology.  2006;21:110-114.

113. Rewerk S, Noppeney T, Winkler M, et al.  Pathogenese der Primär- und Rezidivvarikosis  an der Magna-Krosse (Die  Rolle von VEGF und VEGF-Rezeptor).  Phlebologie. 2007;36:137-142.

114. Roka F, Binder M, Bohler-Sommeregger  K. Mid-term recurrence rate of  incompetent perforating veins after  combined superficial vein surgery and  subfascial endoscopic perforating vein  surgery.  J Vasc Surg. 2006;44(2):359-363.

115. Rutgers PH, Kitslaar PJ. Randomized  trial of stripping versus high ligation  combined with sclerotherapy in  the treatment of the incompetent  greater saphenous vein. Am J Surg.  1994;168:311-315.

116. Rutherford EE, Kianifard B, Cook SJ,  Holdstock JM, Whiteley MS. Incompetent  perforating veins are associated with  recurrent varicose veins. Eur J Vasc  Endovasc Surg. 2001;21:458-460.

117. Saarinen J, Suominen V, Heikinen  M, et al. The profile of leg symptoms,  clinical disability and reflux in legs with  previously operated varices. Scand J  Surg. 2005;94:51-55.

118. Shadid N, Ceulen R, Nelemans P,  et al. Randomized clinical trial of  ultrasoundguided foam sclerotherapy  versus surgery for the incompetent  great saphenous vein. Br J Surg.  2012;99:1062-1070.

119. Stonebridge P, Chalmers N, Beggs I.  Recurrent varicose veins: a varicographic  analysis leading to a new practical  classification. Br J Surg. 1995;82:60-62.

120. Stötter L, Schaaf I, Bockelbrink A.  Comparative outcomes of radiofrequency  endoluminal ablation, invagination  stripping and cryostripping in the  treatment of great saphenous vein.  Phlebology. 2006;21:60-64.

121. Stücker M, Netz K, Breuckmann F,  Altmeyer P, Mumme A. Histomorphologic  classification of recurrent saphenofemoral  reflux. J Vasc Surg. 2004;39:816-822.

122. Klem TM, Schnater JM, Schütte PR,  Hop W, van der Ham AC, Wittens CH.  A randomized trial of cryo stripping  versus conventional stripping of the  great saphenous vein. J Vasc Surg.  2009;49:403-439.

123. Theivacumar NS, Dellagrammaticas D,  Darwood RJ, Mavor AI, Gough MJ. Fate  of the great saphenous vein following  endovenous laser ablation: does recanalisation  mean recurrence. Eur J Vasc  Endovasc Surg. 2008;36:211-215.

124. Theivacumar NS, Darwwod R, Gough  MJ. Neovascularisation and recurrence  2 years after varicose vein treatment for  sapheno-femoral and great saphenous  vein reflux: a comparison of surgery and  endovenous laser ablation. Eur J Vasc  Endovasc Surg. 2009;38:203-207.

125. Theivacumar NS, Gough MJ.  Endovenous laser ablation (EVLA ) to  treat recurrent varicose veins. Eur J Vasc  Endovasc Surg. 2011;41:691-696.

126. Tsang FJ, Davis M, Davies AH. Incomplete  saphenopopliteal ligation after short  saphenous vein surgery: a summation  analysis. Phlebology. 2005;20:106-109.

127. Turton EPL, Scott DJA, Richards SP, et  al. Duplex derived evidence of reflux  after varicose vein surgery: neo reflux or  neovascularisation. Eur J Vasc Endovasc  Surg. 1999;17:230-233.

128. Theivacumar NS, Gough MJ.  Endovenous Laser Ablation (EVLA ) to  Treat Recurrent Varicose Veins. Eur J Vasc  Endovasc Surg. 2011;41:691-696.

129. van Groenendael L, van der Vliet  A, Flinkenflögel L, et al. Treatment of  recurrent varicose veins of the great  saphenous vein by conventional surgery  and endovenous laser. J Vasc Surg.  2009;50:1106-1113.

130. van Groenendael L, van der Vliet  JA, Flinkenflögel L, Roovers EA,  van Sterkenburg SM, Reijnen MM.  Conventional surgery and endovenous  laser ablation of recurrent varicose  veins of the small saphenous vein: A  retrospective clinical comparison and  assessment of patient satisfaction.  Phlebology. 2010;25(3):151-157.

131. van Neer P, Kessels A, de Haan E, et  al. Residual varicose veins below the  knee are not related to incompetent  perforating veins. J Vasc Surg.  2006;44:1051-1054.

132. van Neer P, de Haan MW, de Veraart  JCJM, Neumann HAM. Recurrent  varicose veins below the knee after  varicose vein surgery. Phlebologie.  2007;36:132-136.

133. van Rij AM, Jiang P, Solomon C, Christie  RA, Hill GB. Recurrence after varicose  vein surgery: a prospective long-term  clinical study with duplex ultrasound  scanning and air plethysmography. J  Vasc Surg. 2003;38:935-943.

134. van Rij AM, Jones GT, Hill GB, Jiang  P. Neovascularization and recurrent  varicose veins : More histologic and  ultrasound evidence. J Vasc Surg.  2004;40: 296-302.

135. Vin F, Chleir F. Aspect échographique des  récidives variqueuses postopératoires du  territoire de la veine petite saphène. Ann  Chir. 2001;126:320-324.

136. Winterborn RJ, Foy C, Earnshaw JJ.  Causes of varicose vein recurrence: Late  results of a randomized controlled trial of  stripping the long saphenous vein. J Vasc  Surg. 2004;40:634-639.

137. Winterborn RJ, Earnshaw JJ. Randomised  trial of polytetrafluoroethylene patch  insertion for recurrent great saphenous  varicose veins. Eur J Vasc Endovasc Surg.  2007;34:367-373.

138. Wong JKF, Duncan JL, Nichols DM.  Whole-leg duplex mapping for varicose  veins: Observation on patterns of reflux in  recurrent and primary legs, with clinical  correlation. Eur J Vasc Endovasc Surg.  2003;25:267-275.

139. Wright D, Rose KG, Young E, McCollum  CN. Recurrence following varicose vein  surgery. Phlebology. 2002;16:101-105.

140. Wright D, Gobin JP, Bradbury AW, et  al; The Varisolve® European Phase  III Investigators Group. Varisolve®  polidocanol microfoam compared  with surgery or sclerotherapy in the  management of varicose veins in the  presence of trunk vein incompetence:  European randomized controlled trial.  Phlebology. 2006;21:180-190.

141. Zan S, Varetto G, Maselli M, Scovazzi  P, Moniaci D, Lazzaro D. Recurrent  varices after internal saphenectomy:  Physiopathological hypothesis and  clinical approach. Minerva Cardioangiol.  2003;51(1):79-86.

142. Franceschi C. ‘Theorie et Pratique de la  Cure Conservatrice et Hémodynamique  de l’Insuffisance Veineuse en  Ambulatoire’ Precy-sous-Thil: Editions de  l’Armancon, 1988.

143. Pittaluga P, Chastanet, Rea B, et al. Midterm  results of the surgical treatment of  varices by phlebectomy with conservation  of a refluxing saphenous vein. J Vasc  Surg. 2009;50:107-118.

144. Pittaluga P, Chastanet S, Rea B, Barbe  R. The effect of isolated phlebectomy  on reflux and diameter of the great  saphenous vein: A prospective study. Eur  J Vasc Endovasc Surg. 2010;40:122-128.

145. Eklof B, Perrin M, Delis KT, Rutherford RB,  Gloviczki P; American Venous Forum;  European Venous Forum; International  Union of Phlebology; American College  of Phlebology; International Union of  Angiology. Updated terminology of  chronic venous disorders: the VEIN-TERM  transatlantic interdisciplinary consensus  document. J Vasc Surg. 2009;48:498-  501.

146. Campanello M, Hammarsten J,  Forsberg C, Bernland P, Henrikson O,  Jensen J. Standard stripping versus  long saphenous vein saving surgery for  primary varicose veins: a prospective,  randomized study with the patients  as their own controls. Phebology.  1996;11:45-49.

147. Einarsson E, Eklof B, Neglén P.  Sclerotherapy or surgery as treatment for  varicose veins. A prospective randomized  study. Phlebology. 1993;8:22-26.

148. Hobbs JT. Surgery and sclerotherapy in  the treatment varicose veins: a 6-year  random trial. Arch Surg. 1974;109:793-  796.

149. Hobbs JT. Surgery or sclerotherapy  for varicose veins: 10year results of a  random trial. Lancet. 1978;27:1149.

150. Dwerryhouse S, Davies B, Harradine  K, Earnshaw JJ. Stripping the long  saphenous vein reduces the rate of  reoperation for recurrent varicose veins. J  Vasc Surg. 1999;29:589-592.

151. Pittaluga P, Chastanet S, Guex JJ.  Great saphenous vein stripping with  preservation of sapheno-femoral  confluence: hemodynamic and clinical  results. J Vasc Surg. 2008;47:1300-1305.

152. Glass GM. Neovascularization  in recurrence of varices of the  great saphenous vein in the groin:  phlebography. Angiology. 1988;39:577-  582.

153. Jones L, Braithwaite BD, Selwyn D, Cooke  S, Earnshaw JJ. Neovascularisation is  the principal cause of varicose vein  recurrence: results of a randomised trial  of stripping the long saphenous vein. Eur  J Vasc Endovasc Surg. 1996;12:442-445.

154. Nyamekye I, Shephard NA,  Davies B, Heather BP, Earnshaw JJ.  Clinicopathological evidence that  neovascularisation is a cause of recurrent  varicose veins. Eur J Vasc Endovasc Surg.  1998;15:412-415.

155. Kostas T, Ioannou CV, Drygiiannakis I, et  al. Chronic venous disease progression  and modification of predisposing factors.  J Vasc Surg. 2010;51:900-907.

156. Krysa J, Jones GT, van Rij AM. Evidence  for a genetic role in varicose veins and  chronic venous insufficiency. Phlebology.  2012;27:329-335.