Advertisement

Endovenous therapies of lower extremity varicosities: A meta-analysis

Open ArchivePublished:August 11, 2008DOI:https://doi.org/10.1016/j.jvs.2008.06.030

      Background

      Minimally invasive techniques such as endovenous laser therapy, radiofrequency ablation, and ultrasound-guided foam sclerotherapy are widely used in the treatment of lower extremity varicosities. These therapies have not yet been compared with surgical ligation and stripping in large randomized clinical trials.

      Methods

      A systematic review of Medline, Cochrane Library, and Cinahl was performed to identify studies on the effectiveness of the four therapies up to February 2007. All clinical studies (open, noncomparative, and randomized clinical trials) that used ultrasound examination as an outcome measure were included. Because observational and randomized clinical trial data were included, both the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) and Quality Of Reporting Of Meta-analyses (QUORUM) guidelines were consulted. A random effects meta-analysis was performed, and subgroup analysis and meta-regression were done to explore sources of between-study variation.

      Results

      Of the 119 retrieved studies, 64 (53.8%) were eligible and assessed 12,320 limbs. Average follow-up was 32.2 months. After 3 years, the estimated pooled success rates (with 95% confidence intervals [CI]) for stripping, foam sclerotherapy, radiofrequency ablation, and laser therapy were about 78% (70%-84%), 77% (69%-84%), 84% (75%-90%), and 94% (87%-98%), respectively. After adjusting for follow-up, foam therapy and radiofrequency ablation were as effective as surgical stripping (adjusted odds ratio [AOR], 0.12 [95% CI, –0.61 to 0.85] and 0.43 [95% CI, –0.19 to 1.04], respectively). Endovenous laser therapy was significantly more effective compared with stripping (AOR, 1.13; 95% CI, 0.40-1.87), foam therapy (AOR, 1.02; 95% CI, 0.28-1.75), and radiofrequency ablation (AOR, 0.71; 95% CI, 0.15-1.27).

      Conclusion

      In the absence of large, comparative randomized clinical trials, the minimally invasive techniques appear to be at least as effective as surgery in the treatment of lower extremity varicose veins.
      Lower-extremity venous insufficiency is a common medical condition and occurs in about 15% of men and 35% of women.
      • Callam M.J.
      Epidemiology of varicose veins.
      • Evans C.J.
      • Fowkes F.G.R.
      • Ruckley C.V.
      • Lee A.J.
      Prevalence of varicose veins and chronic venous insufficiency in men and women in the general population: Edinburgh Vein Study.
      • Margolis D.J.
      • Bilker W.
      • Santanna J.
      • Baumgarten M.
      Venous leg ulcer: incidence and prevalence in the elderly.
      The effect of venous insufficiency on patients' health-related quality of life (HRQOL) is substantial and comparable with other common chronic diseases such as arthritis, diabetes, and cardiovascular disease.
      • Andreozzi G.M.
      • Cordova R.M.
      • Scomparin A.
      • Martini R.
      • D'Eri A.
      • Andreozzi F.
      • Quality of Life Working Group on Vascular Medicine of S.I.A.P.A.V.
      Quality of life in chronic venous insufficiency An Italian pilot study of the Triveneto Region.
      In 1995 the overall cost associated with deep or superficial venous insufficiency, or both, was about 2.5% of the total health care budget in France and Belgium.
      • Van den Oever R.
      • Hepp B.
      • Debbaut B.
      • Simon I.
      Socio-economic impact of chronic venous insufficiency An underestimated public health problem.
      The treatment of varicose veins alleviates symptoms and, hopefully, reduces the complication rate of venous insufficiency. The traditional gold standard in the treatment of varicosity of great saphenous veins (GSVs) is a high ligation at the saphenofemoral junction (SFJ), followed by stripping; conventional treatment of small saphenous veins (SSVs) is ligation at the saphenopopliteal junction (SPJ), often without stripping.
      Surgery of varicose veins is usually performed under general or epidural anesthesia and may be associated with neurologic damage (about 7% in short and up to 40% in long stripping of GSVs),
      • Morrison C.
      • Dalsing M.C.
      Signs and symptoms of saphenous nerve injury after greater saphenous vein stripping: prevalence, severity, and relevance for modern practice.
      • Holmes J.B.
      • Skajaa K.
      • Holme K.
      Incidence of lesions of the saphenous nerve after partial or complete stripping of the long saphenous vein.
      scars, and postoperative pain. Despite the relatively high incidence, the neurologic damage has often little resultant morbidity. Although surgery is highly effective in the short term, the 5-year recurrence rates are approximately 30% for GSVs and 50% for SSVs, which may be due to neovascularization.
      • Hartmann K.
      • Klode J.
      • Pfister R.
      • Toussaint M.
      • Weingart I.
      • Waldermann F.
      • et al.
      Recurrent varicose veins: sonography-based re-examination of 210 patients 14 years after ligation and saphenous vein stripping.
      • Darke S.G.
      The morphology of recurrent varicose veins.
      Only <10% of these recurrences are clinically relevant.
      To improve effectiveness and patients' HRQOL and to reduce postoperative downtime, complications, and costs, new minimally invasive techniques such as ultrasound-guided foam sclerotherapy (UGFS),
      • Belcaro G.
      • Nicolaides A.N.
      • Ricci A.
      • Dugall M.
      • Errichi B.M.
      • Vasdekis S.
      • et al.
      Endovascular sclerotherapy, surgery, and surgery plus sclerotherapy in superficial venous incompetence: a randomized, 10-year follow-up trial–final results.
      radiofrequency ablation (RFA, VNUS Closure, VNUS Medical Technologies, San Jose, Calif),
      • Goldman M.P.
      Closure of the greater saphenous vein with endoluminal radiofrequency thermal heating of the vein wall in combination with ambulatory phlebectomy: preliminary 6-month follow-up.
      and endovenous laser ablation (EVLA)
      • Navarro L.
      • Min R.J.
      • Bone C.
      Endovenous laser: a new minimally invasive method of treatment for varicose veins–preliminary observations using an 810 nm diode laser.
      are now widely used in the treatment of lower extremity varicosities.
      Although case series and comparative studies suggest lower recurrence rates of these minimally invasive interventions compared with surgical stripping, no large, long-term, comparative randomized controlled trials (RCTs) have been performed yet, but some are ongoing.

      Current controlled trials. www.controlled-trials.com. Current Controlled Trials Ltd.

      The objective of this analysis is to systematically review and summarize the available studies on the surgical and new therapies and compare the effectiveness of these different options in order to assist physicians and patients in selecting the most appropriate intervention for lower extremity varicose veins in the current absence of well-designed RCTs.

      Methods

      Because of the heterogeneity of the included studies, both the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) and Quality Of Reporting Of Meta-analyses (QUORUM) guidelines were used.
      • Stroup D.F.
      • Berlin J.A.
      • Morton S.C.
      • Olkin I.
      • Williamson G.D.
      • Rennie D.
      • et al.
      Meta-analysis of observational studies in epidemiology: a proposal for reporting Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
      • Moher D.
      • Cook D.J.
      • Eastwood S.
      • Olkin I.
      • Rennie D.
      • Stroup D.F.
      Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement Quality of Reporting of Meta-analyses.

       Literature search

      We initiated an electronic search of Medline, Cochrane Library, and Cinahl up to February 2007. PubMed was searched by a clinical librarian using the following algorithm: (sclerocompression or sclerotherapy) or ([{thermal or radiofrequency} and {ablation or obliteration}] or VNUS) or (laser or laser surgery) or (endovascular or endovenous) or (stripping or stripped or strip or strips or stripper or Babcock) and (saphenous or saphena or varicose veins or varicosis) and (duplex or Doppler or ultrasonic or ultrasound). To broaden the search, the “related articles” function was also used. Specialty journals such as Dermatologic Surgery, Journal of Vascular Surgery, European Journal of Vascular and Endovascular Surgery, and Phlebology were also searched electronically and references of identified studies and reviews were hand-searched. We reviewed all abstracts, studies, and citations, irrespective of language. Clinical trial registries were also searched.

       Inclusion criteria

      Our meta-analysis included RCTs, clinical trials, and prospective and retrospective case series on the treatment of human lower extremity varicosities by surgical stripping (SFJ ligation and GSV stripping or SPJ ligation [and SSV stripping]), EVLA (all wavelengths and energy parameters were included), UGFS with foam (multiple treatments were allowed and no distinction was made between type or concentration of sclerosant), and RFA. We were unable to differentiate between GSVs and SSVs because most studies that included both did not differentiate the outcomes. Only studies that used US examination as the outcome measure were eligible because US is considered the gold standard in the assessment of venous insufficiency and it increases the homogeneity of the analysis. For comparative studies, the arms of interest were included separately. All follow-up periods were allowed. English, German, French, and Dutch studies were included.

       Exclusion criteria

      Studies that performed SFJ ligation without stripping were excluded because this approach is considered suboptimal.
      • Dwerryhouse S.
      • Davies B.
      • Harradine K.
      • Earnshaw J.J.
      Stripping the long saphenous vein reduces the rate of reoperation for recurrent varicose veins: five-year results of a randomized trial.
      Studies that explicitly examined combination therapies were excluded. Treatments of nontruncal varicose veins were not included. We excluded UGFS studies that used liquid sclerosant because it is considered less effective than foam.
      • Hamel-Desnos C.
      • Desnos P.
      • Wollmann J.C.
      • Ouvry P.
      • Mako S.
      • Allaert F.A.
      Evaluation of the efficacy of polidocanol in the form of foam compared with liquid form in sclerotherapy of the greater saphenous vein: initial results.
      To our knowledge, there are no comparative RCTs suggesting a type of sclerosant is superior in the treatment of saphenous trunks using UGFS. Moreover, a RCT showed no significant difference between polidocanol and sodium tetradecyl sulfate in the treatment of varicose and telangiectatic veins, suggesting that the effect of the specific sclerosant in our analysis is limited.
      • Rao J.
      • Wildemore J.K.
      • Goldman M.P.
      Double-blind prospective comparative trial between foamed and liquid polidocanol and sodium tetradecyl sulfate in the treatment of varicose and telangiectatic leg veins.
      If multiple articles reported the same study population, the publication with the longest follow-up was included.

       Data extraction

      The data of all eligible studies were analyzed by two authors (R. v. d. B. and T. N.) independently. The number of patients and treated limbs, the type of veins (GSV or SSV), the treatment procedure, the study type (retrospective or prospective), the duration of follow-up, the type of follow-up (mean follow-up, exact follow-up, or exact with loss of follow-up), the US outcome definitions, and success rate (if possible for GSVs and SSVs separately) were recorded. Because 89% of the included studies were case series, an extensive quality assessment of the studies was not performed, except that a distinction was made between retrospective and prospective data collection. Case series and the arms of interest of RCTs were entered separately in the analysis.

       Standardization of outcome measures

      All of the eligible studies used US as an outcome, but the definitions of treatment success by US examination varied considerably. Because the technical end point of each of the treatments is obliteration or complete removal (ie, anatomic success) of the insufficient vein, the definitions that closely reflected this objective were grouped by consensus of three authors (R. v. d. B., M. N., and T. N.). Therefore, US-based outcomes that used definitions such as absence of “detectable flow,” “recurrence of reflux,” “recanalization,” “vein reopening,” “recurrent or new varices,” “closed vein,” “occlusion,” “obliteration,” and “completely stripped vein” were considered to be equally successful. Studies that reported “clinical improvement,” “patient satisfaction,” “reflux at any site,” “varicose veins present anywhere,” and others were excluded.

       Statistical analysis

      After deriving the natural logarithm of the odds of success for all studies, we calculated pooled estimates of success rate and the 95% confidence interval (CI) for all four treatments using SAS PROC MIXED software (SAS Institute Inc, Cary, NC). A random-effect model
      • Houwelingen H.C.van
      • Zwinderman K.H.
      • Stijnen T.
      Bivariate approach to meta-analysis.
      • DerSimonian R.
      • Laird N.
      Meta-analysis in clinical trials.
      • van Houwelingen H.C.
      • Arends L.R.
      • Stijnen T.
      Advanced methods in meta-analysis: multivariate approach and meta-regression.
      was used because a likelihood ratio test showed that the random-effect model fitted the data significantly better than did a fixed-effect model (χ24 = 32.7, P < .001).
      We compared a random-effect model with one general random intercept to a multivariate random-effect model in which each treatment has its own random intercept. Because the latter did not improve the model significantly (χ23 = 3.8, P = .28), we used the random-effect model with one general random intercept only for all treatments. The treatments were used as covariates in the model, and the differences between the estimated log odds of the treatments automatically resulted in the log odds ratios (OR) to compare the treatments with each other.
      Because follow-up time varied considerably within and between the four treatment groups and the decline of success percentages over time may differ per treatment, a meta-regression with follow-up time per treatment as a covariate was performed to present success rates for different time intervals (ie, 3 months, 1, 3, and 5 years). Furthermore, we performed subgroup analysis based on the type of study (prospective vs retrospective) and study size (more or less than 60 limbs). The between-study variances of the models with and without these covariates were compared to assess whether heterogeneity in the covariates can explain part of the between-study variances.

      Results

       Literature search

      Of all screened abstracts and titles, 119 reports were reviewed in detail, and 64 studies (with a total of 72 arms) fulfilled the eligibility criteria. Of these, 13 (18%) reported on stripping, 10 (14%) on UGFS, 30 (42%) on EVLA, and 19 (26%) on RFA (Table I). We excluded 55 studies for several reasons (Fig 1).
      Table ICharacteristics of studies included in meta-analysis
      No.First authorYear
      Year of publication.
      CountryStudy type
      Type 1 is prospective case series, type 2 is retrospective case series, and type 3 is a randomized clinical trial.
      No. of included limbsTherapyFollow- up
      Follow-up in months.
      Success rateDefinition of failure
      TotalGSVSSVTotalGSVSSV
      1Allegra2007Italy11326862132Surgery600.750.870.7Recurrent/new varices
      2Bountouroglu2006UK330300Surgery30.780.78NAClosed/occlusion/obliteration
      3De Maseneer2002Belgium11721720Surgery120.750.75NARecurrent/new varices
      4De Medeiros2005Brazil320200Surgery11.001.00NARe-opening
      5Dwerryhouse1999UK152520Surgery600.50.5NARecurrent/new varices
      6Fischer2001Swiss21251250Surgery4080.40.4NARecurrent/new varices
      7Frings2004Germany15005000Surgery30.950.95NADetectable flow/reflux
      8Hartmann2006Germany224522025Surgery1680.690.660.88Recurrent/new varices
      9Hinchliffe2006UK316160Surgery1.50.880.88NAIncomplete strip
      10Lurie2005USA336360Surgery240.790.79NARecurrent/new varices
      11Perala2005Finland313130Surgery360.770.77NARecurrent/new varices
      12Sarin1994UK243430Surgery210.510.51NADetectable flow/reflux
      13Smith2002UK122618937Surgery120.860.910.62Detectable flow/reflux
      14Barrett 1
      Year of publication.
      2004NZ2997920UGFS23.70.69
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      15Barrett 1
      Type 1 is prospective case series, type 2 is retrospective case series, and type 3 is a randomized clinical trial.
      2004NZ217143UGFS24.50.77
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      16Barrett 22004NZ21009823UGFS22.50.77
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      17Belcaro 1
      The surgery arm of this study was not included because only ligation without stripping was performed.
      2000Italy339390UGFS1200.810.81NADetectable flow/reflux
      18Belcaro 2
      The surgery arm of this study was not included because only ligation without stripping was performed.
      2003Italy32112110UGFS1200.490.49NARecurrent/new varices
      20Darke2006UK114311528UGFS1.50.880.860.96Closed/occlusion/obliteration
      21Hamel-Desnos2003France145450UGFS0.750.840.84
      Not documented separately for GSV and SSV.
      Detectable flow/reflux
      19Smith2006UK11411886263UGFS110.860.880.82Closed/occlusion/obliteration
      22Tessari2001Italy12497UGFS1111Closed/occlusion/obliteration
      23Yamaki2004Japan137370UGFS120.680.68NAClosed/occlusion/obliteration
      24Agus2006Italy11068105216EVLA360.97
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      25De Medeiros2005Brazil120200EVLA20.950.95NAClosed/occlusion/obliteration
      26Disselhoff2005Netherlands193930EVLA30.840.84NAClosed/occlusion/obliteration
      27Gerard2002France120200EVLA10.90.9NAClosed/occlusion/obliteration
      28Gibson2007USA1210156210EVLA40.96NA0.96Recanalization
      29Goldman2004USA124240EVLA811NADetectable flow/reflux
      30Huang2005China119190EVLA0.511NAClosed/occlusion/obliteration
      31Kabnick2006USA160600EVLA120.930.93NADetectable flow/reflux
      32Kavuturu2006USA166660EVLA120.970.97NAClosed/occlusion/obliteration
      33Kim 12006USA134340EVLA12.211NARecanalization
      34Kim 22006USA160600EVLA6.80.970.97NAClosed/occlusion/obliteration
      35Marston2006USA231310EVLA00.840.84NAClosed/occlusion/obliteration
      36Morrison2005USA150500EVLA120.660.66NAClosed/occlusion/obliteration
      37Min 12001USA190900EVLA90.960.96NAClosed/occlusion/obliteration
      38Min 22003USA14994990EVLA240.930.93NAClosed/occlusion/obliteration
      39Navarro2001USA140400EVLA4.211NADetectable flow/reflux
      40Oh2003Korea115150EVLA311NAClosed/occlusion/obliteration
      41Perkowski2004USA119115437EVLA0.50.97
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Recanalization
      42Petronelli2006Italy152520EVLA120.930.93NARecanalization
      43Proebstle 12002Germany131310EVLA10.970.97NAClosed/occlusion/obliteration
      44Proebstle 22003Germany141410EVLA60.950.95NARecanalization
      45Proebstle 32004Germany21061060EVLA30.90.9NAClosed/occlusion/obliteration
      46Proebstle 42005Germany12822820EVLA30.980.98NAClosed/occlusion/obliteration
      47Proebstle 52006Germany12632630EVLA120.960.96NAClosed/occlusion/obliteration
      48Puggioni2005USA277770EVLA0.250.940.94NARecanalization
      49Ravi2006USA11091990101EVLA0.50.960.970.91Recanalization
      50Sadick2004USA130300EVLA240.970.97NAClosed/occlusion/obliteration
      51Sharif2006UK11451450EVLA120.760.76NAClosed/occlusion/obliteration
      52Theivacumar2007UK168068EVLA60.880.88Closed/occlusion/obliteration
      53Timperman2005USA1100830EVLA90.960.96NADetectable flow/reflux
      54Dunn2006USA185850RFA60.90.9NAClosed/occlusion/obliteration
      55Fassiadis2003UK159590RFA311NAClosed/occlusion/obliteration
      56Goldman2000USA112120RFA611NAClosed/occlusion/obliteration
      57Goldman2002USA141410RFA130.680.68NAClosed/occlusion/obliteration
      58Hinchliffe2006UK316160RFA1.50.810.81NAClosed/occlusion/obliteration
      59Hingorani2004USA173730RFA0.30.960.96NAClosed/occlusion/obliteration
      60Lurie2005USA146460RFA240.860.86NARecurrent/new varices
      61Marston2006USA258580RFA00.880.88NAClosed/occlusion/obliteration
      62Merchant 12002USA1318RFA240.85
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      63Merchant 22005USA11222115452RFA600.87
      Not documented separately for GSV and SSV.
      Not documented separately for GSV and SSV.
      Closed/occlusion/obliteration
      64Morrison2005USA150500RFA120.80.8NAClosed/occlusion/obliteration
      65Ogawa2005Japan125250RFA111NAClosed/occlusion/obliteration
      66Perala2005Finland315150RFA360.670.67NARecurrent/new varices
      67Pichot2004France263630RFA250.910.91NAClosed/occlusion/obliteration
      68Puggioni2005USA253530RFA0.230.910.91NARecanalisation
      69Sybrandy2002Netherlands126260RFA120.890.89NAClosed/occlusion/obliteration
      70Wagner2003USA228280RFA311NAClosed/occlusion/obliteration
      71Weiss2002USA11401400RFA00.90.9NAClosed/occlusion/obliteration
      72Welch2006USA21841840RFA00.80.8NAClosed/occlusion/obliteration
      EVLA, endovenous laser ablation; GSV, great saphenous vein; NA, nonapplicable; NZ, New Zealand; RCT, randomized clinical trial; RFA, radiofrequency ablation; SSV, short saphenous vein; UGFS, ultrasound-guided foam sclerotherapy; UK, United Kingdom; USA, United States of America.
      a Year of publication.
      b Type 1 is prospective case series, type 2 is retrospective case series, and type 3 is a randomized clinical trial.
      c Follow-up in months.
      d Not documented separately for GSV and SSV.
      e The surgery arm of this study was not included because only ligation without stripping was performed.
      Figure thumbnail gr1
      Fig 1Schematic flow chart of literature search.

       Study characteristics for included trials

      We included 64 studies (72 study arms) with a total of 12,320 treated limbs, of which 2804 (23%) were stripped, 2126 (17%) were treated by UGFS, 4876 (40%) by EVLA, and 2514 (20%) by RFA. The reports were published between January 1994 and February 2007, and 92% in the last 5 years (Table I). Of the 72 study arms, 58 (81%) were prospective. Although follow-up duration ranged from 1 day to 34 years, 51 of the 72 studies had a follow-up of between 3 months and 10 years. The number of included limbs was 12 to 1411. Nine studies reported the separate success rates of SSV and GSV therapy, and seven were RCTs that included two intervention arms. Nine of the 10 UGFS studies used aethoxysclerol (polidocanol), one study only used sodium tetradecyl sulfate, and three studies used both sclerosants.

       Success rates for each therapy

      The crude success rates of each of the four therapies independent of follow-up time according to the random-intercept model suggest that the success rate of EVLA (93.3%; 95% CI, 91.0-95.0) and RFA (87.5%; 95% CI, 82.5-91.3) are higher than for stripping and UGFS (Fig 2). For stripping, UGFS, and RFA, the effectiveness of the therapies decreased over time from ≥80% success rates at 3 months to <80% after 5 years. The success percentages of EVLA remained at ≥92.9% (Table II, Fig 3). The estimated success rates declined significantly for stripping (P = .004), but no significant negative trend was detected for UGFS (P = .08), RFA (P = .25), or EVLA (P = .61) over time.
      Figure thumbnail gr2
      Fig 2Forest plots with log(odds) of each study ordered per treatment.
      Table IIThe pooled proportion of patients with anatomical successful outcome after different time intervals
      Type of intervention3 months1 year3 year5 year
      Success rate (%)95% CISuccess rate (%)95% CISuccess rate (%)95% CISuccess rate (%)95% CI
      Surgery80.472.3-86.579.771.8-85.877.870.0-84.075.767.9-82.1
      UGFS82.172.5-88.980.971.8-87.677.468.7-84.373.562.8-82.1
      RFA88.883.6-92.587.783.1-91.284.275.2-90.479.959.5-91.5
      EVLA92.990.2-94.893.391.1-95.094.587.2-97.795.479.7-99.1
      CI, Confidence intervals; EVLA, endovenous laser ablation; RFA, radiofrequency ablation; UGFS, ultrasound guided foam sclerotherapy.
      Figure thumbnail gr3
      Fig 3Anatomic success rate for surgical stripping, ultrasound-guided foam sclerotherapy (UGFS), endovenous laser ablation (EVLA), and radiofrequency ablation (RFA) in time. The estimated success rates declined significant for stripping (P = .004), but no significant negative trend was detected for UGFS (P = .08), RFA (P = .25), and EVLA (P = .61) over time.

       Comparison of therapies

      Compared with stripping, UGFS was as effective and EVLA and RFA were significantly more effective in the treatment of lower extremity varicose veins (Table III). After adjusting for duration of follow-up, however, we observed no significant differences between stripping and RFA. Of the three minimally invasive techniques, EVLA was superior to UGFS (P = .013) and RFA (P = .016) after adjusting for follow-up time, but there was no significant difference between UGFS and RFA (P = .27).
      Table IIIComparisons of four different treatment options for lower extremity varicose veins
      ComparisonsUnadjusted for follow-upAdjusted for follow-up
      Crude OR95% CIPAdjusted OR95% CIP
      UGFS vs strip0.15–0.49 to 0.80.640.12–0.61 to 0.85.73
      EVLA vs strip1.541.02 to 2.07<.00011.130.40 to 1.87.006
      RFA vs strip0.870.29 to 1.45.0030.43–0.19 to 1.04.16
      EVLA vs UGFS1.390.81 to 1.97<.00011.020.28 to 1.75.013
      RFA vs UGFS0.710.08 to 1.34.030.31–0.29 to 0.91.27
      EVLA vs RFA0.680.17 to 1.18.0090.710.15 to 1.27.016
      CI, Confidence intervals; EVLA, endovenous laser ablation; OR, odds ratio; RFA, radiofrequency ablation; UGFS, ultrasound guided foam sclerotherapy.

       Subgroup analysis

      Restricting the analysis to the 58 prospective studies confirmed that EVLA was significantly more effective than stripping (P < .0001), UGFS (P < .0001), and RFA (P = .01). However, no significant differences in effectiveness were observed between RFA vs stripping (P = .14) and RFA vs UGFS (P = .13).
      The results of the analyses of the 35 largest studies that treated >60 limbs were comparable with the complete meta-analysis: EVLA remained significantly more successful than stripping (P < .0001), UGFS (P < .0001), and RFA (P = .04); and RFA was superior to stripping (P = .048) and UGFS (P = .04). Excluding the SSV and restricting the analysis to 62 studies that presented success rates for GSVs (separately) confirmed the finding that EVLA was significantly more effective than the other therapies (P < .0001).

      Discussion

      The results of this meta-analysis suggest that endovenous treatments of lower extremity varicosities are better in achieving anatomic success (ie, obliteration or disappearance of veins) than surgery and UGFS. Of the endovenous therapies, EVLA is significantly more effective than RFA to obliterate the insufficient veins. These findings, however, should be confirmed in large, long-term, comparative RCTs.
      The estimated success rates of the studied therapies and the comparison between therapies are in agreement with most of the available studies. A small paired analysis
      • De Medeiros C.A.
      • Luccas G.C.
      Comparison of endovenous treatment with an 810 nm laser versus conventional stripping of the great saphenous vein in patients with primary varicose veins.
      and a nonrandomized pilot study that compared EVLA with stripping of the GSV
      • Mekako A.I.
      • Hatfield J.
      • Bryce J.
      • Lee D.
      • McCollum P.T.
      • Chetter I.
      A nonrandomised controlled trial of endovenous laser therapy and surgery in the treatment of varicose veins.
      showed that the clinical efficacy parameters were comparable in the short term. A recent RCT showed that EVLA was as effective as stripping after 6 months and was associated with less postoperative pain and bruising.
      • Rasmussen L.H.
      • Bjoern L.
      • Lawaetz M.
      • Blemings A.
      • Lawaetz B.
      • Eklof B.
      Randomized trial comparing endovenous laser ablation of the great saphenous vein with high ligation and stripping in patients with varicose veins: short-term results.
      In the long term, however, it is likely that the recurrence rate of surgery is higher than that of EVLA because of neovascularization, as is confirmed by the findings of the current analysis. One retrospective study suggested that RFA and EVLA were equally effective
      • Puggioni A.
      • Kalra M.
      • Carmo M.
      • Mozes G.
      • Gloviczki P.
      Endovenous laser therapy and radiofrequency ablation of the great saphenous vein: analysis of early efficacy and complications.
      and another that EVLA was superior.
      • Almeida J.I.
      • Raines J.K..
      Radiofrequency ablation and laser ablation in the treatment of varicose veins.
      Three small, short-term RCTs showed that RFA and surgery were about equally effective, but RFA-treated patients reported less postoperative pain and physical limitations, faster recovery, fewer adverse events, and superior HRQOL compared with patients who underwent surgical stripping.
      • Rautio T.
      • Ohinmaa A.
      • Perala J.
      • Ohtonen P.
      • Heikkinen T.
      • Wiik H.
      • et al.
      Endovenous obliteration versus conventional stripping operation in the treatment of primary varicose veins: a randomized controlled trial with comparison of the costs.
      • Lurie F.
      • Creton D.
      • Eklof B.
      • Kabnick L.S.
      • Kistner R.L.
      • Pichot O.
      • et al.
      Prospective randomized study of endovenous radiofrequency obliteration (closure procedure) versus ligation and stripping in a selected patient population (EVOLVeS Study).
      • Perala J.
      • Rautio T.
      • Biancari F.
      • Ohtonen P.
      • Wiik H.
      • Heikkinen T.
      • 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.
      An earlier RCT showed that liquid UGS was less effective than surgical stripping,
      • Rutgers P.H.
      • Kitslaar P.J.
      Randomized trial of stripping versus high ligation combined with sclerotherapy in the treatment of the incompetent greater saphenous vein.
      but that study used liquid sclerosant, which is washed out relatively quickly and induces less vasospasm and sclerous formation than foam sclerosant.
      • Hamel-Desnos C.
      • Desnos P.
      • Wollmann J.C.
      • Ouvry P.
      • Mako S.
      • Allaert F.A.
      Evaluation of the efficacy of polidocanol in the form of foam compared with liquid form in sclerotherapy of the greater saphenous vein: initial results.
      Clinical trial registries indicate that several important RCTs of RFA vs stripping and UGFS vs surgery are currently ongoing.

      Current controlled trials. www.controlled-trials.com. Current Controlled Trials Ltd.

      In addition to anatomic success rates, patient-reported outcomes such as HRQOL, treatment satisfaction, symptom relief, and side effects are pivotal in a comparison between invasive and noninvasive therapies for venous insufficiency. Compared with surgery, EVLA-treated patients appreciated EVLA more than surgery because they reported fewer side effects and their HRQOL improved better and faster.
      • De Medeiros C.A.
      • Luccas G.C.
      Comparison of endovenous treatment with an 810 nm laser versus conventional stripping of the great saphenous vein in patients with primary varicose veins.
      • Mekako A.I.
      • Hatfield J.
      • Bryce J.
      • Lee D.
      • McCollum P.T.
      • Chetter I.
      A nonrandomised controlled trial of endovenous laser therapy and surgery in the treatment of varicose veins.
      Patient-reported outcomes are especially important when two therapies are equally effective. For example, this current meta-analysis suggests that the anatomic success rates of UGFS and surgery are comparable, but patients' opinions may differ between these therapies.
      Also, cost-effectiveness assessments are lacking and should be included in clinical trials. One study suggested that the RFA procedure was cost-saving from a societal perspective compared with surgery because the patient's physical function was restored faster and endovenous therapies can be performed in an outpatient setting, resulting in lower nonmedical costs.
      • Rautio T.
      • Ohinmaa A.
      • Perala J.
      • Ohtonen P.
      • Heikkinen T.
      • Wiik H.
      • et al.
      Endovenous obliteration versus conventional stripping operation in the treatment of primary varicose veins: a randomized controlled trial with comparison of the costs.
      Minor and relatively common postoperative complications of ligation and stripping are wound infection, hematoma, lymphorrhagia, and hypertrophic scarring. Other complications of surgery are nerve injury (7%) and deep vein thrombosis (<2%).
      • Cox S.J.
      • Wellwood J.M.
      • Martin A.
      Saphenous nerve injury caused by stripping of the long saphenous vein.
      • Docherty J.G.
      • Morrice J.J.
      • Bell G.
      Saphenous neuritis following varicose vein surgery.
      • Morrison C.
      • Dalsing M.C.
      Signs and symptoms of saphenous nerve injury after greater saphenous vein stripping: prevalence, severity, and relevance for modern practice.
      • Sam R.C.
      • Silverman S.H.
      • Bradbury A.W.
      Nerve injuries and varicose vein surgery.
      • Wood J.J.
      • Chant H.
      • Laugharne M.
      • Chant T.
      • Mitchell D.C.
      A prospective study of cutaneous nerve injury following long saphenous vein surgery.
      • Holme J.B.
      • Skajaa K.
      • Holme K.
      Incidence of lesions of the saphenous nerve after partial or complete stripping of the long saphenous vein.
      Because the sclerosant enters the deep venous system, UGFS may be associated with several specific complications such as migraine, temporal brain ischaemia, and scotomas, especially among patients with a foramen ovale.
      • Smith P.C.
      Chronic venous disease treated by ultrasound guided foam sclerotherapy.
      As in surgery, most patients will experience ecchymosis and pain (often described as “a pulling chord”) for 1 to 2 weeks after endovenous therapies. Dysesthesia, phlebitis, and skin burns have been reported in a small proportion and deep vein thrombosis in <1% of patients after EVLA and RFA.
      • Agus G.B.
      • Mancini S.
      • Magi G.
      IEWG. The first 1000 cases of Italian Endovenous-laser Working Group (IEWG) Rationale, and long-term outcomes for the 1999-2003 period.
      • Proebstle T.M.
      • Moehler T.
      • Gul D.
      • Herdemann S.
      Endovenous treatment of the great saphenous vein using a 1,320 nm Nd:YAG laser causes fewer side effects than using a 940 nm diode laser.
      • Kabnick L.S.
      Outcome of different endovenous laser wavelengths for great saphenous vein ablation.
      • Merchant R.F.
      • Pichot O.
      Closure Study Group Long-term outcomes of endovenous radiofrequency obliteration of saphenous reflux as a treatment for superficial venous insufficiency.
      To our knowledge, this is the first meta-analysis and meta-regression analysis comparing different treatment options for lower extremity varicose veins, and the results suggest that there are significant differences between interventions. The detected differences are in accordance with the few available comparative studies suggesting a good face validity of our findings. More than 60 studies met our inclusion criteria. To increase homogeneity of the comparison, we restricted the analysis to studies that used US as primary end point. Because of the variation in follow-up duration, we adjusted the comparison between the therapies for this difference. Several sensitivity analyses were performed to assess the effects of study design, duration of follow-up, and sample size on our findings and they confirmed our initial results.
      Meta-analysis is associated with several limitations. A major limitation of this analysis is that it included a heterogeneous mix of case series and RCTs. This rather unusual but methodologically and clinically sound approach was chosen because of the lack of comparative RCTs in phlebology, as was illustrated by the systematic review. To increase the quality of analysis, both the MOOSE and QUORUM guidelines were followed as much as possible.
      • Stroup D.F.
      • Berlin J.A.
      • Morton S.C.
      • Olkin I.
      • Williamson G.D.
      • Rennie D.
      • et al.
      Meta-analysis of observational studies in epidemiology: a proposal for reporting Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
      • Moher D.
      • Cook D.J.
      • Eastwood S.
      • Olkin I.
      • Rennie D.
      • Stroup D.F.
      Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement Quality of Reporting of Meta-analyses.
      The objective of this study was to inform physicians about four therapies commonly used in the treatment of lower extremity varicose veins and compare their efficacy based on the available data.
      An aggregation or ecologic bias, which occurs because group rates may not resemble individual rates, is unavoidable. Because we were unable to precisely describe the heterogeneous study populations, different inclusion criteria may have affected our findings (eg, case series of endovenous therapies may have included more primary, nontortuous, interfascial GSVs than UGFS and stripping, and RFA is limited to veins of ≤12 mm due to the catheter size). Although we restricted the analysis to studies that used US to increase comparability, the standardization of the different definitions of success, which was based on consensus, may have affected our results.
      To minimize the effect of publication bias, an extensive English and non-English literature search was performed, including registries of clinical trials. Small studies were not excluded to reduce publication bias because their impact was weighted and the proportion of total weight of these studies was limited. A subanalysis limited to studies with >60 patients showed findings similar to those presented, confirming that the effect of the smaller studies was not substantial.
      The EVLA studies with limited follow-up are likely to reflect the centers' initial experience (ie, learning curve), and the relatively large proportion of these studies may explain the lower success rates after 3 months compared with later intervals. Several studies from the 1970s and 1980s were excluded because US examination was not an outcome measurement. To further increase homogeneity of the analysis, it was restricted to studies that used ligation and stripping because this is the gold standard of surgical care and restricted to foam in sclerotherapy because it is superior to liquid sclerosant.
      • Hamel-Desnos C.
      • Desnos P.
      • Wollmann J.C.
      • Ouvry P.
      • Mako S.
      • Allaert F.A.
      Evaluation of the efficacy of polidocanol in the form of foam compared with liquid form in sclerotherapy of the greater saphenous vein: initial results.
      Also, we did not differentiate between concentration of sclerosant, which varied from 1% to 3%. However, a recent RCT demonstrated that the concentration of sclerosant (1% vs 3%) was not a significant predictor of outcome in UGFS.
      • Ceulen R.P.
      • Bullens-Goessens Y.L.
      • Pi-VAN DE Venne S.J.
      • Nelemans P.J.
      • Veraart J.C.
      • Sommer A.
      Outcomes and side effects of duplex-guided sclerotherapy in the treatment of great saphenous veins with 1% versus 3% polidocanol foam: results of a randomized controlled trial with 1-year follow-up.
      Because 89% of the studies were case series, a thorough quality assessment was not performed, but subgroup analysis suggests that the results of retrospective and prospective studies were not substantially different.

      Conclusion

      The results of this meta-analysis support the increasing use of minimally invasive interventions in the treatment of lower extremity varicosities. In the absence of comparative RCTs, it appears that EVLA is more effective than surgery, UGFS, and RFA. However, large, long-term comparative RCTs that include patient-reported outcomes, cost-effectiveness analyses, and safety assessment are needed to achieve the highest level of evidence for these novel therapies.

      Author contributions

      • Conception and design: TN, LA, MN
      • Analysis and interpretation: TN, LA
      • Data collection: RB, TN, MK
      • Writing the article: RB, TN, LA
      • Critical revision of the article: MK, MN, LA
      • Final approval of the article: RB, MK, MN, LA, TN
      • Statistical analysis: LA
      • Obtained funding: Not applicable
      • Overall responsibility: TN

      References

        • Callam M.J.
        Epidemiology of varicose veins.
        Br J Surg. 1994; 81: 167-173
        • Evans C.J.
        • Fowkes F.G.R.
        • Ruckley C.V.
        • Lee A.J.
        Prevalence of varicose veins and chronic venous insufficiency in men and women in the general population: Edinburgh Vein Study.
        J Epidemiol Community Health. 1999; 53: 149-153
        • Margolis D.J.
        • Bilker W.
        • Santanna J.
        • Baumgarten M.
        Venous leg ulcer: incidence and prevalence in the elderly.
        J Am Acad Dermatol. 2002; 46: 381-386
        • Andreozzi G.M.
        • Cordova R.M.
        • Scomparin A.
        • Martini R.
        • D'Eri A.
        • Andreozzi F.
        • Quality of Life Working Group on Vascular Medicine of S.I.A.P.A.V.
        Quality of life in chronic venous insufficiency.
        Int Angiol. 2005; 24: 272-277
        • Van den Oever R.
        • Hepp B.
        • Debbaut B.
        • Simon I.
        Socio-economic impact of chronic venous insufficiency.
        Int Angiol. 1998; 17: 161-167
        • Morrison C.
        • Dalsing M.C.
        Signs and symptoms of saphenous nerve injury after greater saphenous vein stripping: prevalence, severity, and relevance for modern practice.
        J Vasc Surg. 2003; 38: 886-890
        • Holmes J.B.
        • Skajaa K.
        • Holme K.
        Incidence of lesions of the saphenous nerve after partial or complete stripping of the long saphenous vein.
        Acta Chir Scand. 1990; 156: 145-148
        • Hartmann K.
        • Klode J.
        • Pfister R.
        • Toussaint M.
        • Weingart I.
        • Waldermann F.
        • et al.
        Recurrent varicose veins: sonography-based re-examination of 210 patients 14 years after ligation and saphenous vein stripping.
        Vasa. 2006; 35: 21-26
        • Darke S.G.
        The morphology of recurrent varicose veins.
        Eur J Vasc Surg. 1992; 6: 512-517
        • Belcaro G.
        • Nicolaides A.N.
        • Ricci A.
        • Dugall M.
        • Errichi B.M.
        • Vasdekis S.
        • et al.
        Endovascular sclerotherapy, surgery, and surgery plus sclerotherapy in superficial venous incompetence: a randomized, 10-year follow-up trial–final results.
        Angiology. 2000; 51: 529-534
        • Goldman M.P.
        Closure of the greater saphenous vein with endoluminal radiofrequency thermal heating of the vein wall in combination with ambulatory phlebectomy: preliminary 6-month follow-up.
        Dermatol Surg. 2000; 26: 452-456
        • Navarro L.
        • Min R.J.
        • Bone C.
        Endovenous laser: a new minimally invasive method of treatment for varicose veins–preliminary observations using an 810 nm diode laser.
        Dermatol Surg. 2001; 27: 117-122
      1. Current controlled trials. www.controlled-trials.com. Current Controlled Trials Ltd.

        • Stroup D.F.
        • Berlin J.A.
        • Morton S.C.
        • Olkin I.
        • Williamson G.D.
        • Rennie D.
        • et al.
        Meta-analysis of observational studies in epidemiology: a proposal for reporting.
        JAMA. 2000; 283: 2008-2012
        • Moher D.
        • Cook D.J.
        • Eastwood S.
        • Olkin I.
        • Rennie D.
        • Stroup D.F.
        Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement.
        Lancet. 1999; 354: 1896-1900
        • Dwerryhouse S.
        • Davies B.
        • Harradine K.
        • Earnshaw J.J.
        Stripping the long saphenous vein reduces the rate of reoperation for recurrent varicose veins: five-year results of a randomized trial.
        J Vasc Surg. 1999; 29: 589-592
        • Hamel-Desnos C.
        • Desnos P.
        • Wollmann J.C.
        • Ouvry P.
        • Mako S.
        • Allaert F.A.
        Evaluation of the efficacy of polidocanol in the form of foam compared with liquid form in sclerotherapy of the greater saphenous vein: initial results.
        Dermatol Surg. 2003; 29: 1170-1175
        • Rao J.
        • Wildemore J.K.
        • Goldman M.P.
        Double-blind prospective comparative trial between foamed and liquid polidocanol and sodium tetradecyl sulfate in the treatment of varicose and telangiectatic leg veins.
        Dermatol Surg. 2005; 31: 631-635
        • Houwelingen H.C.van
        • Zwinderman K.H.
        • Stijnen T.
        Bivariate approach to meta-analysis.
        Stat Med. 1993; 12: 2273-2284
        • DerSimonian R.
        • Laird N.
        Meta-analysis in clinical trials.
        Control Clin Trials. 1986; 7: 177-188
        • van Houwelingen H.C.
        • Arends L.R.
        • Stijnen T.
        Advanced methods in meta-analysis: multivariate approach and meta-regression.
        Stat Med. 2002; 21: 589-624
        • De Medeiros C.A.
        • Luccas G.C.
        Comparison of endovenous treatment with an 810 nm laser versus conventional stripping of the great saphenous vein in patients with primary varicose veins.
        Dermatol Surg. 2005; 31: 1685-1694
        • Mekako A.I.
        • Hatfield J.
        • Bryce J.
        • Lee D.
        • McCollum P.T.
        • Chetter I.
        A nonrandomised controlled trial of endovenous laser therapy and surgery in the treatment of varicose veins.
        Ann Vasc Surg. 2006; 20: 451-457
        • Rasmussen L.H.
        • Bjoern L.
        • Lawaetz M.
        • Blemings A.
        • Lawaetz B.
        • Eklof B.
        Randomized trial comparing endovenous laser ablation of the great saphenous vein with high ligation and stripping in patients with varicose veins: short-term results.
        J Vasc Surg. 2007; 46: 308-315
        • Puggioni A.
        • Kalra M.
        • Carmo M.
        • Mozes G.
        • Gloviczki P.
        Endovenous laser therapy and radiofrequency ablation of the great saphenous vein: analysis of early efficacy and complications.
        J Vasc Surg. 2005; 42: 488-493
        • Almeida J.I.
        • Raines J.K..
        Radiofrequency ablation and laser ablation in the treatment of varicose veins.
        Ann Vasc Surg. 2006; (20547–52)
        • Rautio T.
        • Ohinmaa A.
        • Perala J.
        • Ohtonen P.
        • Heikkinen T.
        • Wiik H.
        • et al.
        Endovenous obliteration versus conventional stripping operation in the treatment of primary varicose veins: a randomized controlled trial with comparison of the costs.
        J Vasc Surg. 2002; 35: 958-965
        • Lurie F.
        • Creton D.
        • Eklof B.
        • Kabnick L.S.
        • Kistner R.L.
        • Pichot O.
        • et al.
        Prospective randomized study of endovenous radiofrequency obliteration (closure procedure) versus ligation and stripping in a selected patient population (EVOLVeS Study).
        J Vasc Surg. 2003; 38: 207-214
        • Perala J.
        • Rautio T.
        • Biancari F.
        • Ohtonen P.
        • Wiik H.
        • Heikkinen T.
        • 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: 669-672
        • Rutgers P.H.
        • Kitslaar P.J.
        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
        • Cox S.J.
        • Wellwood J.M.
        • Martin A.
        Saphenous nerve injury caused by stripping of the long saphenous vein.
        BMJ. 1974; 1: 415-417
        • Docherty J.G.
        • Morrice J.J.
        • Bell G.
        Saphenous neuritis following varicose vein surgery.
        Br J Surg. 1994; 81: 698
        • Morrison C.
        • Dalsing M.C.
        Signs and symptoms of saphenous nerve injury after greater saphenous vein stripping: prevalence, severity, and relevance for modern practice.
        J Vasc Surg. 2003; 38: 886-890
        • Sam R.C.
        • Silverman S.H.
        • Bradbury A.W.
        Nerve injuries and varicose vein surgery.
        Eur J Vasc Endovasc Surg. 2004; 27: 113-120
        • Wood J.J.
        • Chant H.
        • Laugharne M.
        • Chant T.
        • Mitchell D.C.
        A prospective study of cutaneous nerve injury following long saphenous vein surgery.
        Eur J Vasc Endovasc Surg. 2005; 30: 654-658
        • Holme J.B.
        • Skajaa K.
        • Holme K.
        Incidence of lesions of the saphenous nerve after partial or complete stripping of the long saphenous vein.
        Acta Chir Scand. 1990; 156: 145-148
        • Smith P.C.
        Chronic venous disease treated by ultrasound guided foam sclerotherapy.
        Eur J Vasc Endovasc Surg. 2006; 32: 577-583
        • Agus G.B.
        • Mancini S.
        • Magi G.
        IEWG. The first 1000 cases of Italian Endovenous-laser Working Group (IEWG).
        Int Angiol. 2006; 25: 209-215
        • Proebstle T.M.
        • Moehler T.
        • Gul D.
        • Herdemann S.
        Endovenous treatment of the great saphenous vein using a 1,320 nm Nd:YAG laser causes fewer side effects than using a 940 nm diode laser.
        Dermatol Surg. 2005; 31: 1678-1683
        • Kabnick L.S.
        Outcome of different endovenous laser wavelengths for great saphenous vein ablation.
        J Vasc Surg. 2006; 43: 88-93
        • Merchant R.F.
        • Pichot O.
        Closure Study Group.
        J Vasc Surg. 2005; 42: 502-509
        • Ceulen R.P.
        • Bullens-Goessens Y.L.
        • Pi-VAN DE Venne S.J.
        • Nelemans P.J.
        • Veraart J.C.
        • Sommer A.
        Outcomes and side effects of duplex-guided sclerotherapy in the treatment of great saphenous veins with 1% versus 3% polidocanol foam: results of a randomized controlled trial with 1-year follow-up.
        Dermatol Surg. 2007; 33: 276-281