Endovenous Ablation (Radiofrequency and Laser) and Foam Sclerotherapy Versus Conventional Surgery For Great Saphenous Vein Varices (Review)

Endovenous ablation (radiofrequency and laser) and foam
sclerotherapy versus conventional surgery for great

saphenous vein varices (Review)
Nesbitt C, Eifell RKG, Coyne P, Badri H, Bhattacharya V, Stansby G
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2011, Issue 10
http://www.thecochranelibrary.com
Endovenous ablation (radiofrequency and laser) and foam sclerotherapy versus conventional surgery for great saphenous vein varices
(Review)
Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Analysis 1.1. Comparison 1 Laser ablation versus surgery, Outcome 1 Recanalisation - (number of limbs). . . . . 47
Analysis 1.2. Comparison 1 Laser ablation versus surgery, Outcome 2 Technical failure - (number of limbs). . . . 48
Analysis 2.1. Comparison 2 Radiofrequency ablation versus surgery, Outcome 1 Recurrence. . . . . . . . . . 48
Analysis 2.2. Comparison 2 Radiofrequency ablation versus surgery, Outcome 2 Recanalisation. . . . . . . . 49
Analysis 2.3. Comparison 2 Radiofrequency ablation versus surgery, Outcome 3 Neovascularisation. . . . . . . 50
Analysis 2.4. Comparison 2 Radiofrequency ablation versus surgery, Outcome 4 Technical failure. . . . . . . . 50
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 63
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Endovenous ablation (radiofrequency and laser) and foam sclerotherapy versus conventional surgery for great saphenous vein varices i
(Review)
[Intervention Review]
Endovenous ablation (radiofrequency and laser) and foam

sclerotherapy versus conventional surgery for great
saphenous vein varices
Craig Nesbitt1 , Ron KG Eifell2, Peter Coyne3 , Hassan Badri4 , Vish Bhattacharya4 , Gerard Stansby5
1
Department of Vascular Surgery, Queen Elizabeth Hospital, Gateshead, UK. 2 Department of Vascular and General Surgery, Western
Infirmary Glasgow, Glasgow, UK. 3 Department of General Surgery, Queen Elizabeth Hospital, Gateshead, UK. 4 Department of
General and Vascular Surgery, Queen Elizabeth Hospital, Gateshead, UK. 5 Department of Surgery, University of Newcastle upon Tyne,
Newcastle upon Tyne, UK
Contact address: Craig Nesbitt, Department of Vascular Surgery, Queen Elizabeth Hospital, Queen Elizabeth Avenue, Sheriff Hill,
Gateshead, Tyne and Wear, NE9 6SX, UK. craignesbitt@hotmail.co.uk.
Editorial group: Cochrane Peripheral Vascular Diseases Group.

Publication status and date: New, published in Issue 10, 2011.
Review content assessed as up-to-date: 24 September 2010.
Citation: Nesbitt C, Eifell RKG, Coyne P, Badri H, Bhattacharya V, Stansby G. Endovenous ablation (radiofrequency and laser) and
foam sclerotherapy versus conventional surgery for great saphenous vein varices. Cochrane Database of Systematic Reviews 2011, Issue
10. Art. No.: CD005624. DOI: 10.1002/14651858.CD005624.pub2.
ABSTRACT
Background
Minimally invasive techniques to treat great saphenous varicose veins include ultrasound-guided foam sclerotherapy (USGFS), ra-
diofrequency ablation (RFA) and endovenous laser therapy (EVLT). Compared with conventional surgery (high ligation and stripping
(HL/S)), proposed benefits include fewer complications, quicker return to work, improved quality of life (QoL) scores, reduced need
for general anaesthesia and equivalent recurrence rates.
Objectives
To review available randomised controlled clinical trials (RCT) data comparing USGFS, RFA, EVLT to HL/S for the treatment of
great saphenous varicose veins.
Search methods
The Cochrane Peripheral Vascular Diseases (PVD) Group searched their Specialised Register (July 2010) and CENTRAL (The Cochrane
Library 2010, Issue 3). In addition the authors performed a search of EMBASE (July 2010). Manufacturers of EVLT, RFA and sclerosant
equipment were contacted for trial data.
Selection criteria
All RCTs of EVLT, RFA, USGFS and HL/S were considered for inclusion. Primary outcomes were recurrent varicosities, recanalisation,
neovascularisation, technical procedure failure or need for re-intervention, patient quality of life (QoL) scores and associated compli-
cations. Secondary outcomes were type of anaesthetic, procedure duration, hospital stay and cost.
Data collection and analysis
CN, RE, VB, PC, HB and GS independently reviewed, assessed and selected trials which met the inclusion criteria. CN and RE
extracted data. The Cochrane Collaborations tool for assessing risk of bias was used. CN contacted trial authors to clarify details.
Endovenous ablation (radiofrequency and laser) and foam sclerotherapy versus conventional surgery for great saphenous vein varices 1
(Review)
Main results
Thirteen reports from five studies with a combined total of 450 patients were included. Rates of recanalisation were higher following
EVLT compared with HL/S, both early (within four months) (5/149 versus 0/100; odds ratio (OR) 3.83, 95% confidence interval
(CI) 0.45 to 32.64) and late recanalisation (after four months) (9/118 versus 1/80; OR 2.97 95% CI 0.52 to 16.98), although these
results were not statistically significant. Technical failure rates favoured EVLT over HL/S (1/149 versus 6/100; OR 0.12, 95% CI
0.02 to 0.75). Recurrence following RFA showed no difference when compared with surgery. Recanalisation within four months was
observed more frequently following RFA compared with HL/S although not statistically significant (4/105 versus 0/88; OR 7.86, 95%
CI 0.41 to 151.28); after four months no difference was observed. Neovascularisation was observed more frequently following HL/
S compared with RFA, but again this was not statistically significant (3/42 versus 8/51; OR 0.39, 95% CI 0.09 to 1.63). Technical
failure was observed less frequently following RFA compared with HL/S although this was not statistically significant (2/106 versus 7/
96; OR 0.48, 95% CI 0.01 to 34.25). No randomised clinical trials comparing HL/S versus USGFS met our study inclusion criteria.
QoL scores and operative complications were not amenable to meta-analysis.
Authors conclusions
Currently available clinical trial evidence suggests RFA and EVLT are at least as effective as surgery in the treatment of great saphenous
varicose veins. There are insufficient data to comment on USGFS. Further randomised trials are needed. We should aim to report and
analyse results in a congruent manner to facilitate future meta-analysis.
PLAIN LANGUAGE SUMMARY
Endovenous ablation (radiofrequency and laser) and foam sclerotherapy versus conventional surgery for varicose veins
Varicose veins are dilated, tortuous superficial veins. When they are in the legs they can be painful, particularly when standing and
walking, itchy and unsightly. Varicose veins are conventionally treated with surgery to remove the veins, by stripping them to the level
of the knee (so-called high ligation and stripping). New less invasive treatments seal the main leaking vein in the thigh using laser
(endovenous laser therapy), radiofrequency ablation (RFA) or foam sclerotherapy. These techniques may result in less pain after the
procedure, fewer complications and a quicker return to work and normal activities with improved patient quality of life, as well as
avoiding the need for a general anaesthetic.
Our review brought together all available randomised controlled trials that compared the new techniques to surgery in the treatment
of varicosities in the great saphenous vein. We found only five trials, with a combined total of 450 patients, which met our inclusion
criteria. Three trials compared laser therapy with surgery and two trials compared RFA with surgery. Laser therapy was associated with
less technical failure but a trend to higher rates of reopening of the treated vein (recanalisation) compared with surgery. No results were
available to compare the rates of recurrence. We found that RFA was associated with trends for fewer technical failures and less new vein
growth (neovascularisation) compared with surgery; the trend was for more recanalisation within four months with no demonstrated
difference in recurrence of varicose veins. No randomised controlled trial compared sclerotherapy with surgery. The results in the study
reports were presented as either the number of legs or number of patients, where some patients had varicose veins in both legs. The
outcomes were also measured at different times after the procedures for the different trials. This limited the findings of our review.
We can conclude from the limited available evidence that RFA and EVLT are no worse than HL/S. Our review is not powered to make
any robust recommendations that affect clinical practice. We need more RCT data comparing these novel therapies to HL/S before we
really know their true potential.
Description of the condition

Varicose veins are dilated, tortuous superficial veins (varices) which
usually affect the lower extremities. There are two main superficial
BACKGROUND
(Review)
venous systems, the great (GSV) and small (SSV) saphenous veins skin from burns in addition to providing local anaesthesia. Imme-
and their tributaries. diate mobilisation is encouraged following the procedure. Wound
The surgical management of great saphenous vein (GSV) varices haematoma and infection are reported to be infrequent following
has traditionally consisted of saphenofemoral ligation and strip- this procedure, however superficial phlebitis, thermal skin injuries
ping of the GSV to the knee (HL/S) (Rutgers 1994; Sarin 1992). and nerve injuries have been reported. Recurrence of varices after
Visible truncal and non-saphenous varices in the calf as well as two years is between 10% and 15% (Merchant 2002; Pichot 2004)
varicose GSV branches in the thigh are treated with hook phle- and one author reported that this incidence remains unchanged
bectomies (stab avulsions), sclerotherapy or (more recently) tran- after three years (Nicolini 2005).
silluminated powered phlebectomy. These procedures have often 2. EVLT involves insertion of a bare tipped optical laser fibre into
been performed under general anaesthetic (Dwerryhouse 1999; the GSV under ultrasound guidance with partial compression ap-
Rautio 2002a) however increasingly no anaesthesia is required or plied to the vein as some intravascular blood is necessary for heat-
local block techniques are adopted (Hamel-Desnos 2008; Wright ing of the vein; intravascular coagulation therefore occurs with this
2006). The recurrence rate at five years following saphenofemoral method. The optical fibre is then withdrawn along the length of
junction ligation and GSV stripping (HL/S) is reported to be the vein from just below the saphenofemoral junction to the knee,
between 20% and 28% (Darke 1992; Fischer 2002) and in the producing obliteration of the vein lumen. This procedure can be
absence of GSV stripping (saphenofemoral ligation only) the re- performed under local or general anaesthetic with the addition of
currence rate is doubled (Sarin 1992). For primary GSV surgery tumescent anaesthesia. Again, patients are encouraged to mobilise
the overall complication rate is between 17% and 20% (Critchley immediately after the procedure. The recurrence rate is reported
1997; Wigger 1998) while that for recurrent varicose veins is 40% as 7% after two years (Min 2003) with most recurrences occurring
(Hayden 2001). These complications include wound haematoma, within the first three months. A low incidence of bruising and
wound infections, lymphatic leaks, common femoral vein and induration around the site of local (tumescent) anaesthetic infil-
artery injuries and neurological complications. Patients are en- tration has been reported, with some reports of either superficial
couraged to mobilise immediately after surgery and are usually or deep phlebitis or thermal skin and nerve injuries.
advised to return to work after 10 to 14 days. Bruising along the 3. Foam sclerotherapy relies on the basic principle of inducing
tract of the avulsed GSV is common and can last for up to six fibrosis of the vein and obliteration of the lumen by causing in-
weeks following surgery. flammation in the endothelial and subendothelial layers of the
vein wall. The GSV is cannulated under ultrasound guidance and
sclerosant foam injected. Several techniques have been proposed
Description of the intervention to produce sclerosant foam and the Tessari and Frullini techniques
appear to give the most favourable results (Frullini 2002; Tessari
Three newer techniques have been developed as alternatives to
2001). The foam replaces blood in the vein, which enhances the ef-
conventional surgery to treat GSV incompetence. They have been
ficacy of the sclerosing agent by reducing the volume of sclerosant
advocated to offer benefits over conventional surgical stripping of
required for treatment and increasing the effective surface area of
the above-knee portion of the GSV in eliminating saphenofemoral
the sclerosant in contact with the vein wall. Ultrasound moni-
reflux (Bergan 2002; Navarro 2001; Rautio 2002a) and in reduc-
toring during the procedure is necessary to prevent spread of the
ing complication rates, patient discomfort and length of time be-
foam into the femoral vein. A major proposed advantage of foam
fore return to work (Lurie 2003; Rautio 2002a). These three tech-
sclerotherapy over conventional sclerotherapy is that larger veins
niques are endovenous laser therapy (EVLT), radiofrequency ab-
appear to be successfully treated with the foam technique com-
lation (RFA) and foam injection sclerotherapy.
pared with liquid sclerotherapy (Baccaglini 1997; Tessari 2001).
Endovenous laser therapy and RFA require duplex ultrasound lo-
The success rate varies between 88% and 93% depending on the
calisation of the GSV for endovenous insertion of the device, iden-
technique employed to produce the foam (Frullini 2002) and the
tification of the device at the saphenofemoral junction and direct
types of complications and the complication rate are reported to
visualisation of the device during treatment. Foam sclerotherapy
be the same as with liquid sclerotherapy.
is also carried out under ultrasound visualisation.
1. RFA uses radiofrequency energy supplied via a catheter electrode
inserted into the GSV under ultrasound guidance. Compression
of the vein around the catheter is maintained as the catheter is
How the intervention might work
withdrawn from just below the saphenofemoral junction to the Jones et al have proposed that following conventional surgery groin
knee, resulting in heating of the vein wall which is kept at a con- haematoma and disrupted superficial pudendal venous drainage
stant temperature by a thermocouple. The procedure can be per- are stimulants for neovascularisation (development of new blood
formed under local or general anaesthetic with tumescent anaes- vessels) in the groin (Jones 1996), which is a source of venous reflux
thesia. Tumescent anaesthesia is the injection of local anaesthetic recurrence. The newer techniques preserve the pudendal and lower
around and along the vein length to form a cuff that protects the abdominal wall venous drainage, do not involve a surgical incision
(Review)
and prevent haematoma formation in the groin thus potentially Types of outcome measures
removing the stimulus for neovascularisation and perhaps reducing
longer-term recurrence (Pichot 2004). Additionally, as they are
less invasive they may be advantageous in relation to issues such Primary outcomes
as pain and recovery times compared with conventional surgery.
1. Recurrence or recanalisation, determined by ultrasound
imaging, at set time intervals post-procedure
2. Ultrasound evidence of neovascularisation
3. Need for re-intervention due to failure of a procedure
OBJECTIVES (technical failure)
4. Patient satisfaction determined by quality of life scores pre
To determine whether endovenous ablation (radiofrequency and and post-procedure
laser) and foam sclerotherapy have any advantages or disadvan- 5. Post-operative complications
tages in comparison with conventional surgical, saphenofemoral
junction ligation and stripping of great saphenous vein varices.
Secondary outcomes
1. The type of anaesthetic required for the procedures
METHODS 2. Length of the procedures
3. Hospital stay
4. Operative costs
Criteria for considering studies for this review
Search methods for identification of studies

Types of studies
Randomised controlled trials (RCTs) comparing RFA GSV ab-
lation, EVLT GSV ablation and foam sclerotherapy of the GSV Electronic searches
with conventional GSV high ligation and stripping were used. The Cochrane Peripheral Vascular Diseases Group searched their
Specialised Register (last searched July 2010) and the Cochrane
Central Register of Controlled Trials (CENTRAL) (The Cochrane
Types of participants Library 2010, Issue 3) at www.thecochranelibrary.com. See
Study participants were males and females of any age with varicose Appendix 1 for details of the search strategy used to search
veins affecting the GSV system, confirmed on duplex ultrasound CENTRAL. The Specialised Register is maintained by the Tri-
imaging, and who were suitable for any of the treatment options. als Search Co-ordinator and is constructed from weekly elec-
Patients who had undergone previous high ligation but had proven tronic searches of MEDLINE, EMBASE, CINAHL and AMED;
recurrent saphenofemoral junction (SFJ) incompetence and a pre- and through handsearching relevant journals. The full list of the
served and refluxing GSV to the groin were also included. Stud- databases, journals and conference proceedings which have been
ies that included combined treatments consisting of endovenous searched, as well as the search strategies used, are described in
treatment or foam sclerotherapy with surgical ligation of the SFJ the Specialised Register section of the Cochrane Peripheral Vas-
were not included. Studies looking at the treatment of varicose cular Diseases (PVD) Group module in The Cochrane Library (
veins affecting the small saphenous vein (SSV) system were not www.thecochranelibrary.com).
included. Studies that included patients who had undergone treat- In addition, the authors (RE, PC, HB and CN) searched EMBASE
ment of both great and small saphenous veins, and did not provide using the strategy described in Appendix 2.
any subanalysis of these groups, were excluded.
Searching other resources

Types of interventions We contacted the following manufacturers for details of published
1. Radiofrequency endovenous ablation and unpublished trials:
2. Laser endovenous ablation EVLT equipment (Diomed),
3. Foam sclerotherapy RFA device (VNUS Medical Technologies),
4. Saphenofemoral junction ligation and stripping of the great Sclerosant (sodium tetradecyl sulphate - 3% Fibro-Vein) (STD
saphenous vein Pharmaceuticals LTD).
(Review)
Data collection and analysis All included studies were randomised controlled trials.
The Cochrane Collaborations tool for assessing risk of bias was
used when assessing the risk of bias in the included studies (Higgins Results of the search
2011).
The search of CENTRAL yielded 500 reports which were screened
by reading the titles and abstracts (if available). Ninety-seven re-
Selection of studies ports were retrieved from the Specialised Register. None of the
manufacturing or pharmaceutical companies who were contacted
CN, RE, VB, PC, HB and GS independently reviewed the trials
provided details of any trials for analysis.
and selected those which met the above criteria.
Included studies
Data extraction and management
See Characteristics of included studies.
CN and RE extracted data and VB cross-checked the information
Thirteen reports relating to five studies were included in the review.
using pro formas designed by the Cochrane PVD Group.
Methods
Assessment of risk of bias in included studies
(i) Study design: Darwood 2008, Rautio 2002 and Subramonia
CN, RE and VB independently assessed the methodological qual-
2010 were randomised controlled single-centre clinical trials.
ity of included trials using the Cochrane Collaborations tool for
Rasmussen 2007 was a two-centre RCT and EVOLVeS Study was
risk of bias (Higgins 2011).
a multi-centre prospective RCT.
(ii) Method of randomisation: Darwood 2008, Rautio 2002 and
Measures of treatment effect Rasmussen 2007 used a sealed envelope technique. Subramonia
Statistical analysis was performed according to the Cochrane PVD 2010 and EVOLVeS Study used a computer-based randomisation
Groups statistical guidelines for review authors. Subgroup analyses method.
were performed as necessary. Odds ratio was used as the measure of (iii) Blinding: blinding operators and patients was not feasible
effect for each dichotomous outcome. Where there were sufficient for many of the outcomes. The post-operative assessors were not
data, a summary statistic for each outcome was calculated using blinded in any of the included studies.
both a fixed-effect model and a random-effects model. (iv) Exclusions post-randomisation: Rasmussen 2007 did not
Heterogeneity in the data was noted and explored using previously report any exclusions post-randomisation. Subramonia 2010,
identified characteristics of the studies, particularly assessments of Darwood 2008, EVOLVeS Study and Rautio 2002 all reported
quality. The I2 statistic was used to determine heterogeneity. A their exclusions, for a variety of reasons. Some discrepancy in
random-effects model was considered for I2 values greater than the number of exclusions was noted in the EVOLVeS trial, see
50%. Incomplete outcome data and Characteristics of included studies
We planned to undertake sensitivity analyses to examine the sta- for full details.
bility of the results in relation to a number of factors including (v) Losses to follow up: only Rautio 2002 reported no losses to
study quality, the source of the data (published and unpublished) follow up. The Characteristics of included studies details the losses
and patient type. Where continuous scales of measurement had to follow up in all other studies.
been used where we planned to assess the effects of treatment, these (vi) Intention-to-treat analysis: there was no intention-to-treat
data were analysed in continuous form (that is mean difference). analysis in both Darwood 2008 and EVOLVeS Study. In Rautio
If different scales were used in different studies, where possible the 2002 one patient was excluded after randomisation but not with-
results were standardised and then combined (that is standardised drawn from the study, indicating some intention-to-treat process;
mean difference). which group this patient retired from was not clear. One patient
randomised to surgery in Darwood 2008 switched over to the
laser arm and was followed up in the laser group. Only Rasmussen
2007 and Subramonia 2010 performed analysis on an intention-
to-treat basis.
RESULTS
(vii) Sources of funding: all included trials declared their sources
of funding. Subramonia 2010 and Darwood 2008 confirmed that
there had been no input from their sources of funding into the
Description of studies design, data collection, data analysis or interpretation of their tri-
See: Characteristics of included studies; Characteristics of excluded als. EVOLVeS Study declared that their sponsor (VNUS medi-
studies; Characteristics of ongoing studies. cal technologies) supported their data collection. Rautio 2002 re-
(Review)
ceived funding from the University of Oulu, Finland. Rasmussen Darwood 2008 was the only trial to exclude patients on warfarin,
2007 received a grant from the Public Health Insurance Research Rautio 2002 excluded patients with coagulopathies.
Foundation of Denmark and received endovenous laser catheters EVOLVeS Study and Rasmussen 2007 mentioned aberrant
from two pharmaceutical companies. They did not disclose any anatomy as an exclusion criteria.
information on the potential input of their sources of funding on Subramonia 2010 and EVOLVeS Study allowed subsequent ran-
the design, analysis or interpretation of their trial. domisation of a second leg; whereas Rautio 2002 excluded pa-
tients with bilateral varicose veins. The potential effects of treating
bilateral varicose vein patients is discussed in the Other potential
Participants sources of bias section.
Only Darwood 2008 and Rautio 2002 did not specify excluding
(i) Country: included studies had centres in UK, Austria, USA,
patients with ankle brachial pressure indexes of < 0.9. EVOLVeS
France, Denmark and Finland.
Study, Rasmussen 2007 and Subramonia 2010 excluded pa-
(ii) Setting or location: all included studies were hospital based
tients with a history of deep vein thrombosis. Rautio 2002 and
except Rasmussen 2007, where all patients were operated on in a
Subramonia 2010 specified exclusion of pregnant patients.
private clinic.
(iii) Number: in order to achieve congruity, sample sizes were
considered in terms of number of patients rather than number
of limbs. Sample sizes ranged from 28 patients (Rautio 2002) Interventions
to 121 patients (Rasmussen 2007). The total number of patients
randomised was 450 and 411 were analysed. Table 1 summarises
the study sample sizes.
(iv) Age: ages ranged from 22 to 79 years (Rasmussen 2007).
(i) Treatment(s)
Rasmussen 2007 and Subramonia 2010 had comparatively higher We did not identify any RCTs comparing conventional foam scle-
mean ages. Darwood 2008, despite an adequate randomisation rotherapy with conventional surgery. Two studies compared EVLT
process, had slightly older surgical patients compared to their laser of GSV with conventional surgery (Darwood 2008; Rasmussen
treatment cohort. 2007). Three studies compared RFA of GSV with conventional
(v) Sex: all included studies had more woman than men, this was surgery (EVOLVeS Study; Rautio 2002; Subramonia 2010).
most apparent in Rautio 2002. This reflects the natural epidemi- Surgical technique: conventional surgery consisted of SFJ expo-
ology of the disease. See Table 2 for full details of the trial partic- sure and division of all tributaries followed by GSV stripping to
ipants age and sex. the knee or upper calf. Authors used one of two techniques when
(vi) Inclusion criteria: all studies included ultrasound scan (USS) stripping: invagination using either a flexible or rigid PIN stripper,
confirmed exclusive SFJ to GSV incompetence. Only Darwood or intraluminal stripper with an olive. Rautio 2002 stripped the
2008 and Rautio 2002 specified symptomatic varicose veins, how- GSV in the thigh from knee to groin, which is considered a variant
ever Subramonia 2010 included only varicose veins where patient of conventional surgery. We have considered both of these tech-
and surgeon agree intervention is required. Rasmussen 2007, niques as conventional stripping techniques and have not drawn
EVOLVeS Study and Darwood 2008 included patients with bi- any distinction between them in our analysis.
lateral varicose veins. Radio frequency ablation (RFA) technique: all studies used the
Only Rautio 2002 did not specify any limits in the age range. VNUS closure system with electrode sizes ranging from 5 to 8 Fr
EVOLVeS Study and Rasmussen 2007 specified the CEAP clas- depending on the size of the vein. This was decided by the relevant
sification of venous disease of the legs, class C2 to C4, but none operators. Subcutaneous tumescence was employed in all studies,
of the trials mentioned inclusion of patients with healed or active but Rautio 2002 used saline only without an anaesthetic agent.
venous ulceration. All of the VNUS procedures in this study were performed under
Subramonia 2010, Rautio 2002 and EVOLVeS Study mentioned general anaesthetic. Rautio 2002 inserted the catheter at the level
patient factors such as suitability for general anaesthetic and day of the ankle but only treated the GSV above the medial condyle
case intervention, availability for the follow-up protocol of the of the tibia, in comparison with Subramonia 2010 and EVOLVeS
studies, ambulatory status and ability to give informed consent. Study who canalised the GSV at the level of the knee, and knee
Full details on the inclusion criteria of the studies can be found at or upper calf respectively.
Characteristics of included studies. Laser (EVLT) technique: Darwood 2008 included a cohort of
(vii) Exclusion criteria: all trials excluded patients with veins patients in whom they used a continuous laser and withdrawal
unsuitable for endovenous therapy, consensus in veins that were technique. Rasmussen 2007 used only a pulsed technique. Both
> 12 mm in diameter or too tortuous, or both, to permit passage studies treated the vein from the groin to the level of the knee.
of the endovenous catheter. All trials excluded patients with SSV Table 3 summarises these techniques. For the purposes of this
reflux. meta-analysis, no subtechnique analysis was performed.
(Review)
Foam technique: as above, we did not identify any RCTs compar- score (VDS). Patency of GSV and presence of neovascularisation
ing conventional foam sclerotherapy with conventional surgery were also assessed.
in our literature searches. Although it was not initially declared, the EVOLVeS Study later
Concomitant phlebectomy: all trials except Darwood 2008 in- presented rates of neovascularisation in the groin and recurrence
cluded concomitant phlebectomies in both treatment arms. Their at two years.
EVLT patients had only their GSV treated initially and sclerother- A summary of the outcome measures of the trials is made in Table
apy was offered at six weeks post-operatively for any residual vari- 5.
cosities. See Table 4.
Excluded studies
(ii) Control or placebo Eighty-seven reports relating to 62 studies were excluded from the
In the interpretation of all studies, conventional surgery was con- review.
sidered the control. Studies which did not include a therapeutic arm of conventional
(iii) Duration: all trials closely followed up their participants surgery were excluded. The RECOVERY Trial is a robust RCT
in the first three months. Only EVOLVeS Study reviewed their comparing RFA and EVLT but no randomised comparison was
patients as soon as 72 hours after surgery. Darwood 2008 pre- made to conventional surgery and hence it was excluded.
sented one-year follow-up results. EVOLVeS Study, Rasmussen Some studies included treatment of patients with GSV and pa-
2007 published two-year follow up results and Rautio 2002 have tients with SSV incompetence. Ogawa 2008 was one such ex-
three-year follow-up results. Darwood 2008 did report two-year ample, where eight of the 92 patients had treatment of SSV re-
follow-up results (Theivacumar 2009) but they were imbedded flux without stratification in their results for SSV and GSV; the
within the results of non-randomised patients, without subgroup study was subsequently excluded. Every effort was made to con-
analysis. The two-year follow-up results were therefore excluded tact authors of such papers to ascertain if any subgroup analysis
from this Cochrane review. The most recent trial (Subramonia was made. Any study which included SSV incompetence without
2010) had no results beyond five weeks. stratification in the results was excluded.
Many papers used liquid sclerotherapy, for example Neglen 1993.
Liquid sclerotherapy is a somewhat historical treatment of GSV
Outcomes
incompetence and hence these papers were excluded.
(i) Primary outcomes: it is not clear from Rasmussen 2007, Lin 2007 was a Chinese study which described a random, com-
EVOLVeS Study or Rautio 2002 what their specific primary or parative study of EVLT and GSV stripping. The study was written
secondary measures were. Rasmussen 2007 set out to assess sa- and published in Chinese. Despite translation it was not possible
fety, efficacy, post-operative morbidity, sick leave, QoL and costs. to extract any meaningful data from the paper and it was therefore
EVOLVeS Study designed their study to compare procedure-re- excluded.
lated complications, patient recuperation and QoL outcomes. Abela 2008 and Bountouroglou 2006 are both RCTs comparing
Rautio 2002 aimed to evaluate outcome in terms of pain, sick foam sclerotherapy to conventional surgery. However they were
leave, health-related QoL and cost. excluded because both studies performed SFJ ligation prior to
Darwood 2008 declared treatment success, with abolition of SFJ or foam sclerotherapy. This is not considered conventional foam scle-
GSV reflux and assessment of the change in the Aberdeen varicose rotherapy treatment.
vein symptom score (AVVSS) as their primary outcomes. Kalteis 2008 and De Medeiros 2006 were two RCTs comparing
Subramonia 2010 was the only trial to include a solitary primary EVLT to conventional surgery. They were excluded because the
outcome, time taken to return to normal household activities, as EVLT patients also underwent SFJ ligation; this is not considered
their primary outcome. conventional laser treatment.
(ii) Secondary outcomes: Darwood 2008 assessed post-procedure The Characteristics of excluded studies table is discussed in detail
pain, time to return to normal activity and work, cosmesis and in Characteristics of excluded studies.
overall satisfaction at three months.
Subramonia 2010 listed intraoperative complications, duration of
the procedure, post-operative morbidity, time to return to driving,
patient satisfaction and QoL.
Risk of bias in included studies
Rautio 2002 assessed further outcomes at three years including Full details on risk of bias can be found in the tables within the
recurrence, satisfaction, venous clinical severity score (VCSS), ve- Characteristics of included studies section as well as in Figure 1
nous segmental disability score (VSDS) and the venous disability and Figure 2.
(Review)
Figure 1. Methodological quality graph: review authors judgements about each methodological quality
item presented as percentages across all included studies.
(Review)
Figure 2. Methodological quality summary: review authors judgements about each methodological quality
item for each included study.
(Review)
Allocation
2008 and Subramonia 2010 clearly distinguished between primary
Darwood 2008, Rasmussen 2007 and Rautio 2002 used an enve- and secondary outcome measures. All other trials included aims
lope technique of randomisation. Few authors provided adequate only.
details of this process in their published trials but, after contact- Rasmussen 2007 reported additional outcome measures in a sub-
ing them, all were able to provide appropriate evidence. Darwood sequent publication with two-year results (Rasmussen 2010) re-
2008 claimed that they stratified their patients by consultant, to porting recurrence rates, which were not a pre-specified outcome
allow for any minor variations in technique. No clear details on measure. However, this did not introduce any bias or inaccuracy
how this stratification was achieved were published. into the trial.
EVOLVeS Study and Subramonia 2010 used computer-based pro- Although it was not initially declared, the EVOLVeS Study later
grams. Again, insufficient data on these methods were published. presented rates of neovascularisation in the groin and recurrence
The authors of the EVOLVeS Study confirmed that they stratified at two years. This was not considered to have introduced any bias
by centre, using block randomisation with block sizes of five cases. into the study.
Subramonia 2010 decided to use two stratification variables (age Similarly, Rautio 2002 also published additional outcome mea-
and sex) which they judged most likely to influence outcome in sures: firstly in their early paper (Rautio 2002), including com-
the two groups. This could introduce a potential bias. No further plications and time of procedure, and again in their three-year
details on their randomisation process was attained despite con- results (Perala 2005), including recurrence, satisfaction, severity
tacting the authors. score measures and presence of neovascularisation. However, it was
felt that these additions did not introduce any bias into the trial.
Blinding
It was not possible for any of the study authors to blind either their
Other potential sources of bias
patients or the operators. However many subjective measures were
made at subsequent follow-up visits, for example recurrence as Darwood 2008 declared that their sample sizes were insufficient to
noted by the observer, and no study blinded their post-operative permit statistical testing for equivalence. Results were not analysed
assessors. on an intention-to-treat basis since one patient was randomised to
surgery but underwent laser treatment and was analysed with the
laser patients.
Incomplete outcome data Rasmussen 2007 performed all procedures in an office-based set-
All included studies reported their incomplete outcome data. ting, in a private clinic under tumescent anaesthesia. All other
One patient randomised to receive surgery in Darwood 2008 re- authors used general anaesthesia in a hospital-based setting. It is
ceived laser therapy. No reason for the switch over was given but possible the private clinic setting and local anaesthesia could con-
it was declared that the patient was followed up in the laser group, found results on absence from work and normal activity, QoL,
demonstrating a lack of intention-to-treat analysis. pain scores etc.
An imbalance in the treatment groups owing to exclusions post- Darwood 2008, EVOLVeS Study, Subramonia 2010 and
randomisation was seen in the EVOLVeS Study: 86 limbs ran- Rasmussen 2007 included patients who underwent treatment of
domised (46 RFA, 40 surgery) and 45 RFA, 36 surgery after exclu- bilateral varicose veins. Darwood 2008 and Rasmussen 2007 ran-
sions. There was also a discrepancy of one patient when the results domised patients once to receive the same treatment during the
of their two-year follow up were scrutinised; and differences given same operation. Results were presented as number of limbs or legs
for the reasons for excluded patients (Lurie 2005). and no stratification was made of those patients who underwent
In Rautio 2002 one patient was excluded after randomisation but bilateral treatment. Some outcome measures can be affected by bi-
not withdrawn from the study. The treatment group this patient lateral treatment, for example pain, QoL measures, time to return
retired from was not made clear. to work, therefore introducing a potential bias in the interpretation
In all the other studies the exclusions were not considered to have of these measures. This is especially true in Darwood 2008 where
had an impact on the intervention groups, or any added risk of 15 patients had bilateral laser ablation compared with just four
bias. surgical patients who underwent bilateral stripping. Subramonia
2010 and the EVOLVeS Study waited for full recovery (> 6 weeks)
before randomising patients second limb thus minimising this
Selective reporting potential bias. Rautio 2002 excluded patients with bilateral veins.
In all of the included studies the aims were made clear. All included The effect of presenting results as limbs rather than patients is
studies reported their pre-specified outcomes. Only Darwood discussed in Overall completeness and applicability of evidence.
(Review)
Both Subramonia 2010 and Rasmussen 2007 included patients Recanalisation
who had undergone previous ligation of the SFJ. They were in- We defined obliteration as complete anatomical obliteration on
cluded on the grounds that they had recanalised their GSV and duplex ultrasound. Partial obliteration without reflux was consid-
had a patent, refluxing SFJ and GSV. It was acknowledged by the ered obliteration failure. Standard criteria for reflux was used: at
authors that surgery for such recurrent patients would not be con- least 0.5s of reflux.
sidered conventional. This formed the basis for our exclusion of Recanalisation in the surgical group was defined as incompetent
Hinchliffe 2006. Unlike Hinchliffe 2006, the other two trials in- GSV refilling from any saphenofemoral neovascularisation follow-
cluded only small numbers of recurrent varicose veins: 16 patients ing stripping, which could occur if primary surgery was a failure
(8 surgery, 8 EVLT) in Rasmussen 2007, and 5 in Subramonia (incomplete stripping) or a duplicate GSV was not identified or
2010 (3 RFA, 2 surgery). CN contacted both authors to see if any stripped during the primary operation. Recanalisation can occur
subgroup analysis had been made of these recurrent patients and with or without recurrent varicosities.
in both instances they had not. Owing to both the small num- Recanalisation was reported at varying time intervals across the
bers and the equal distribution of the recurrent patients in the studies. We have taken four months as a cut off between early and
treatment cohorts it was decided to include both of these papers. late recanalisation.
Wound complications, pain scores, time to return to work or nor-
mal activities appeared to be no higher in these trials compared to
the other included papers. (i) Laser versus surgery
Unfortunately, despite enquiries to relevant authors, we were only
Effects of interventions able to report recanalisation in numbers of legs and not numbers
of patients. Consequently Analysis 1.1 is potentially misleading
and confidence intervals may be unreliable due to some patients
Primary outcomes receiving bilateral treatments. CN contacted Darwood et al and
Rasmussen et al for results per patient. Darwood et al were able
to provide three-month abolition of GSV reflux (by patient): 37/
Recurrence 38 patients for EVLA1, 25/27 patients for EVLA2 (62/65 EVLT
Recurrence of varicosities is a clinical diagnosis. We defined recur- patients overall) and 26/30 for surgery. Although Rasmussen et
rent varicosities as new non-truncal varicose veins. The EVOLVeS al published recanalisation rates per patient, their figures were
Study defined recurrence as new varicose veins below the knee. actually described as per limb. The authors were not able to clarify
Rautio 2002 did not provide their description of recurrent vari- if their recanalisation rates were per patient.
cose veins. Darwood 2008 did not specifically report non-truncal
recurrence in their study. Rasmussen 2007 considered recurrent Results of limbs comparing laser ablation and conventional
varicose veins as veins which had not been observed before or not surgery showed no early recanalisation (within four months)
previously marked by the patient on the Aberdeen varicose vein within the surgery legs compared with five in those legs undergo-
severity score (AVVSS) form. ing laser ablation (OR 3.83, 95% CI 0.45 to 32.64). Beyond four
Only three papers have published results on recurrence (EVOLVeS months only one recanalisation was seen in surgical legs compared
Study; Rasmussen 2007; Rautio 2002). Only one paper reported with nine in laser limbs (OR 2.97, 95% CI 0.52 to 16.98).
recurrence in the laser versus surgery group and so no meta-analy-
sis was possible. Analysis 2.1 shows the meta-analysis of the avail-
able results for patients who underwent RFA versus surgery. Only (ii) Radiofrequency ablation versus surgery
recurrence noted by the clinicians was used. Little difference was No early (within four months) recanalisation was reported in pa-
seen overall, with 10 recurrences noted with the RFA patients and tients who underwent surgery compared with four RFA patients
eight with conventional surgery (odds ratio (OR) 1.04, 95% con- (OR 7.86, 95% CI 0.41 to 151.28). There were reports of three
fidence interval (CI) 0.37 to 2.91). RFA and three surgical patients with recanalisation beyond four
Table 6 shows results of neovascularisation and recurrence as noted months (OR 0.79, 95% CI 0.15 to 4.23). See Analysis 2.2.
by clinicians and patients.
Darwood et al presented two-year results on both neovascularisa-
tion and recurrence (Theivacumar 2009) but they were presented Neovascularisation
in conjunction with all other patients who were treated (non-ran- Three trials (Darwood 2008; EVOLVeS Study; Rautio 2002) re-
domly) at the same time of the original RCT. There was no strat- ported rates of neovascularisation, limiting meta-analysis. Analysis
ification in these results of the randomised and non-randomised 2.3 shows rates within the radiofrequency versus surgery cohort,
patients. These results could therefore not be utilised in this re- with eight surgical patients and three RFA patients having evi-
view. dence of neovascularisation (OR 0.39, 95% CI 0.09 to 1.63). As
(Review)
above, Table 6 summarises the rates of neovascularisation along intervals used for assessments within this range. The only con-
with recurrence. EVOLVeS Study defined neovascularisation as gruity in the assessment intervals was between Rasmussen 2007
multiple small vessels in the groin reconnecting a more proximal and Darwood 2008 who both assessed severity scores pre-opera-
vein or its tributaries and the distal patent vein below the site of tively and at three months post-operatively. For these reasons the
interruption, surgery or RFA, determined after two years on du- QoL measures could not be meta-analysed. A table demonstrating
plex ultrasound examination. Rautio 2002 regarded neovasculari- the variety of measures used is seen in Table 8. It was not possi-
sation as small superficial branches at the area of the SFJ and re- ble to show the VCSS scores for Rasmussen 2007 and Darwood
ported rates at three years. Interestingly Rautio 2002 commented 2008 in a table because Rasmussen 2007 presented the mean and
that neither patient identified with neovascularisation had clinical range for their VCSS scores pre-treatment and at three months,
recurrent varicose veins. Darwood 2008 mentioned neovascular- while Darwood 2008 presented median values and an inter-quar-
isation as a cause of SFJ incompetence and GSV reflux in one tile range (IQR) for pre-treatment values and improvement val-
patient, but only followed up 12 of the 32 patients randomised to ues at three months.
surgery. They did not mention neovascularisation as a finding in
any of their patients who underwent laser treatment.
Re-intervention or technical failure Post-operative complications
All studies declared and reported post-operative complications.

(i) Laser versus surgery Terminology was the main factor that prevented accurate meta-
analysis. For example, post-operative wound infection, wound
Unfortunately, despite enquiries to relevant authors, we were only
discharge and wound breakdown were reported separately across
able to report re-intervention or failure rates in numbers of legs
the studies. Similarly, post-operative paraesthesia along the dis-
and not numbers of patients. Six technical failures were reported
tribution of the great saphenous nerve, numbness and saphe-
in limbs undergoing surgery compared with one limb undergoing
nous nerve injury were reported separately. In an attempt to dis-
laser ablation. Consequently Analysis 1.2 is potentially misleading
play such a wide and varying collection of complications we have
and confidence intervals may be unreliable due to some patients
constructed a table which combines many of these similar events
receiving bilateral treatments (OR 0.12, 95% CI 0.02 to 0.75).
(Table 9). Across the studies, adverse events were recorded at vary-
ing times post-operatively. We have divided them into early (within
(ii) Radiofrequency ablation versus surgery three months) and late (beyond three months), again for simplic-
Seven technical failures were reported in patients undergoing ity. We have also characterised adverse events as major and minor
surgery compared with two in patients undergoing RFA (OR 0.48, depending on whether or not these events required intervention:
95% CI 0.01 to 34.25). See Analysis 2.4. minor requiring no intervention and major requiring interven-
Re-intervention was only reported in two of the studies ( tion.
Rasmussen 2007; Rautio 2002), which prevented meaningful When events were presented more than once within the first three
meta-analysis. The available results are presented in Table 7. months, as in the EVOLVeS Study, which presented adverse events
at 72 hours, one and three weeks post-operatively, the highest value
of any given event was recorded in our summary table. Darwood
Quality of Life measures 2008 reported excessive bruising only; Rasmussen 2007 reported
All included studies evaluated changes in venous severity scores bruising but did not comment on the extent. All bruising was
and quality of life (QoL) measurements using a variety of different reported in our summary table without stratification.
questionnaires. Subramonia 2010, EVOLVeS Study, Rasmussen Overall there were 135 minor adverse events in the patients un-
2007 and Darwood 2008 used disease-specific severity scores and dergoing surgery, 135/192 (70%) early (< 3 months) and 13/149
QoL questionnaires whereas Rautio 2002 used the generic health- (9%) late (> 3 months). There were 25 minor adverse events in
related QoL questionnaire RAND-36. The Aberdeen varicose vein the patients undergoing laser, 25/149 (17%) early (< 3 months)
severity score (AVVSS) was used by Rasmussen 2007, Subramonia and none late (> 3 months); and 76 minor adverse events in the
2010 and Darwood 2008 to assess patient satisfaction. The ve- patients undergoing RFA, 76 (72%) early (< 3 months) and one
nous clinical severity score (VCSS) was used by Rautio 2002 and late (2%) (> 3 months). There were six major complications <
Rasmussen 2007. Rautio 2002 also reported the venous segmen- 3 months following surgery (6/341, 18%). There were no major
tal disability score (VSDS) and the venous disability score (VDS). complications associated with either laser or RFA. Of note, one
Subramonia 2010 used the total clinical severity score (TCSS) patient undergoing surgery in Darwood 2008 aspirated post-op-
and VDS. The assessment intervals of the studies ranged from eratively and required seven days intensive treatment unit (ITU)
one week to three years post-operatively, with several different support.
(Review)
Secondary outcomes The time to return to work or normal activities following interven-
tion was reported in all trials. Unfortunately, the different studies
reported these results as either parametric or non-parametric data
Type of anaesthetic required (mean, median, range, IQR) precluding accurate meta-analysis.
Table 10 summarises the type of anaesthetic used in each trial. Rautio 2002 presented the mean number of sick leave days and
the subjective patients own assessment of the length of the re-
quired sick leave. Subramonia 2010, as well as providing data on
Length of the procedure or operative time time to return to work and normal activities, also presented results
Only the trials comparing RFA versus surgery presented data on for time to return to driving. Table 14 summarises the results that
procedure times. Subramonia 2010 gave values for operative and were available.
theatre time, presenting median values and an IQR. Rautio 2002
presented operating time, operating room time and recovery time
Post-procedure pain
using the mean and standard deviation. Lurie et al (EVOLVeS
Study) admitted in his results that the separate recording times There was a wide variation in how pain was analysed and reported,
of venous access time, treatment time and adjunctive procedure which precluded any meaningful meta-analysis. A break down of
time were not collected uniformly, resulting in confusion with these various pain measures is outlined below.
their overall figures. They presented total treatment times only.
No units of time were provided. The lack of congruity with the
(i) Laser versus surgery
presented results prevented any meaningful meta-analysis. Table
11 shows the results that were available. Rasmussen 2007 reported pain in three ways, with results of the
0 to 10 visual analogue scale (VAS), mean analgesic intake, and
through the SF-36 quality of life measure of bodily pain. Mean
Duration of hospital stay analgesic intake was 12.9 tablets in laser patients versus 12 tablets
Three studies (Darwood 2008; EVOLVeS Study; Rasmussen in surgery patients within the six months post-procedure. VAS
2007) presented data on length of hospital stay thus preventing scores were recorded for 10 days post-procedure and plotted on
meta-analysis. Table 12 details the available data. The majority of a graph. There were no statistically significant differences in the
patients were operated on as day cases. VAS scores between the two groups, although pain was initially
slightly higher after surgery. Bodily pain results were reported for
six months post-procedure. Again there was no statistical differ-
Operative costs ence between surgery and laser patients in their reports of bod-
Cost analysis was provided by only two of the studies (Rasmussen ily pain and both groups showed statistically significant improve-
2007; Rautio 2002). Table 13 summarises the cost information ments at three months from their pre-treatment scores.
that was presented. Darwood 2008 used a 100 mm VAS score recorded daily for
Rasmussen 2007 presented their procedural costs in euros. Using the first post-procedure week; they also reported the day patients
the exchange rate from the beginning of the year of the papers pub- stopped using analgesics. VAS results for the first week were re-
lication, we have presented these costs in UK pounds. Rasmussen ported as median with IQR. Although we considered both EVLT
2007 divided their costs into direct procedural costs and the cal- techniques (continuous and pulsed) together in all other out-
culated indirect cost from lost work days, however they did not comes, it was not possible to combine figures for post-operative
explain how this was achieved. Similarly Rautio 2002 calculated pain and hence results are given separately: pulsed EVLT 11 (IQR
indirect cost from lost work days but based this on the national 4 to 31), continuous EVLT 18 (IQR 10 to 45), surgery 14 (IQR
average wage in Finland at the time of the study. They presented 3 to 53). Results for the day patients stopped using analgesia were
their figures in US dollars. In our table of costs we have converted reported with a median value and IQR: pulsed EVLT 5 (IQR 2
this into UK pounds using the exchange rate from the beginning to 7), continuous EVLT 6 (IQR 3 to 7), surgery 4 (IQR 1 to 7).
of the year of the papers publication.
Of some significance is the cost of additional procedures for resid-
(ii) Radiofrequency ablation versus surgery
ual or recurrent varicosities and none of the studies provided any
estimation of these costs. In the EVOLVeS Study patients recorded the maximum effect of
pain on their daily activities (1 = minimum, 5 = maximum effect).
These results were then translated into a scale of 0 to 100 where 0
Additional outcome measures = least affect on daily activities. The authors reported statistically
significant differences in the pain scores recorded at 72 hours and 1
week post-procedure in favour of RFA. Results are shown in Table
Time to return to work or normal activities 15. The authors adjusted their results for the type of anaesthesia
(Review)
used and the number of adjunctive phlebectomies performed. As In their 2007 publication, Rasmussen 2007 showed the short-
explained by the authors, this was achieved by entering results as term efficacy and safety of endovenous laser to be similar to con-
covariates in a genlinear model. ventional surgery. They found no difference in the frequency of
Rautio 2002 reported VAS scores in their patients when standing, adverse events between the two treatments apart from bruising,
walking and at rest. They also provided all of their patients with which was more frequent following conventional surgery (Table 9).
ibuprofen prescriptions and recorded their use of analgesics. Visual Post-operative pain scores were initially slightly higher following
analogue scores were recorded daily for two weeks. Mean scores at surgery but this was not statistically significant and no change was
rest, standing and walking were lower in the RFA group, especially seen beyond 12 days. The authors did not show a difference in the
between the fifth to 14th post-operative days. These differences time to return to work or normal activity following intervention,
were not statistically significant (P = 0.36). The RFA patients used casting uncertainty on the clinical implications of reduced post-
less ibuprofen analgesia: average daily tablets 0.4 (SD 0.49) com- operative pain. Quality of life and VCSS were increased similarly
pared with 1.3 (SD 1.09) in surgical patients. This was statistically in both treatment groups at three months. Cost analysis, which
significant (P = 0.001). No data was presented on whether patients included calculation of loss of productivity, showed a USD 339
used any other analgesia. saving per treatment associated with conventional surgery (Table
Subramonia 2010 reported pain data from a 10 cm unmarked 13). In their 2010 publication Rasmussen 2007 show no differ-
VAS, a patient-reported grading of severity (none, very mild, mild, ences in recurrence and reflux between the treatment groups and
moderate, severe and very severe) and duration of requirements a similar sustained improvement in quality of life. This study was
for analgesia. Median values for VAS scores within the first post- powered to show a 15% difference in success rate and hence carries
procedure week were: 1.7 (IQR 0.5 to 4.3) in RFA patients, and 4 a 20% type 2 error risk.
(IQR 2.35 to 6.05) in surgical patients. This was statistically sig- Darwood 2008 showed few differences when comparing EVLT
nificant (P = 0.001). Severity grading, although highly subjective, to conventional surgery. Indeed they commented that EVLT and
showed less pain associated with RFA, see Table 16. Duration of surgery were equally effective in abolishing SFJ and GSV reflux.
analgesia use was significantly less for RFA patients (P = 0.001): There were statistically significant differences in pain scores and
median number of days was 2 (IQR 0 to 7) and 10 (IQR 5 to 14) analgesia use. Disease-specific QoL measures were similar apart
for RFA and surgery respectively. Ten patients required no anal- from a reduced physical functioning score at week one in surgical
gesia after RFA compared with only one patient who underwent patients; the authors did not present figures or indicate if this was
conventional surgery. a statistically significant difference. EVLT was associated with a
Sensitivity analyses to examine the stability of the results were not quicker return to work and normal activity in the absence of a
undertaken in this review due to the limited number of studies difference in post-operative pain scores. This improved recovery
in the meta-analyses. We hope to conduct such analyses in future following minimally invasive procedure may well be due to a pa-
updates of this review. tients mind-set. The incidence of phlebitis was higher following
EVLT; all other adverse events showed no difference.
DISCUSSION (ii) Radiofrequency ablation versus surgery

Of the outcomes that were amenable to meta-analysis, recanalisa-
tion within the four months following surgery was observed less
frequently following surgery compared with RFA but this differ-
Summary of main results
ence did not reach statistical significance; after four months no
difference was observed (Analysis 2.2). Recurrence showed no dif-
ference (Analysis 2.1). Neovascularisation was observed less fre-
(i) Laser versus surgery quently following RFA (Analysis 2.3) but again this observation
Of the outcomes that were amenable to meta-analysis, rates of was not statistically significant. Technical failure was observed less
recanalisation were lower following surgery, both at early (< 4 frequently following RFA although this trend was not statistically
months) and late (> 4 months) time intervals (Analysis 1.1), how- significant (Analysis 2.4). The overall lack of statistical significance
ever the values must be taken with caution as the confidence inter- means that these observed differences between RFA and surgery
vals included the null value 1. Technical failure rates favour laser must be interpreted with caution. Fixed-effect model tests of het-
(Analysis 1.2) and although tighter confidence intervals (0.02 to erogeneity for recurrence (I2 = 40%), overall recanalisation (I2 =
0.75) were observed, the failure to present rates by number of 47%) and neovascularisation (I2 = 0%) suggest moderate levels of
patients limits the validity of this analysis. Fixed-effect model tests or little heterogeneity. The fixed-effect model test of heterogeneity
of heterogeneity for overall recanalisation (I2 = 0%) and technical for technical failure has an I2 of 76% suggesting a high level of
failure (I2 = 0%) suggest no heterogeneity. heterogeneity. A random-effects model analysis was undertaken.
(Review)
The EVOLVeS Study showed an earlier return to work and normal with bilateral veins on the same day and ensured that both limbs
activities, less pain, quicker recovery, fewer adverse events and im- received the same treatment.
proved early cosmesis in the patients undergoing RFA compared Rasmussen 2007 only published results according to the number
to conventional surgery. Three of five QoL measures were also im- of limbs rather than using the number of patients as the unit of
proved in RFA patients in the period up to four months. The only analysis. This affects the standard error of the treatment effect
measures with statistical significance were the early advantages of and subsequently meant the results of these trials were precluded
RFA with regards to discomfort and cosmesis. The two-year fol- from meta-analysis. CN contacted the author to see if they could
low-up efficacy results (clinical and haemodynamic studies) with provide information by patient but this was not available.
RFA were comparable to surgical patients. Numbers were not suf- Darwood 2008 also published results according to the number of
ficient to prove any of these parameters, with statistical signifi- limbs, however CN contacted the author who was able to pro-
cance, but in RFA-treated limbs the recurrence rates were lower vide data on three month abolition of GSV reflux by patient (37/
and neovascularisation rates were lower (Table 6); quality of life 38 patients for EVLA1, 25/27 patients for EVLA2 and 26/30 for
measures remained improved in both treatment groups, especially surgery). The author also confirmed that their AVVSS was by pa-
for those who underwent RFA. tient as were the results for return to normal activity, pain, satisfac-
Rautio 2002 reported less post-operative pain, shorter sick leave tion and cosmesis. All other results, however, have been excluded
and faster recovery of physical function in patients undergoing from the meta-analysis.
RFA than for those undergoing conventional surgery (Table 14). In the EVOLVeS Study one patient (in the RFA group) received
Early results confirmed equal efficacy in eliminating the GSV. bilateral treatment. CN contacted the author and confirmed that
Complication rates were also similar in the two groups. Radiofre- a period of three months elapsed between the treatment of each
quency ablation was shown overall to be a more lengthy proce- limb. Due to the three-month recovery period, each limb has
dure but it was felt that this was due to the learning curve of been regarded as a separate patient episode and the results from
the operators (Table 11). In their three-year follow-up results an EVOLVeS have hence been included in the meta-analysis. Meta-
increase in GSV recanalisation was demonstrated (Analysis 2.2), analysing limbs and patients does introduce bias, however the
with a decrease in GSV occlusion rate from 93.4% to 85.2%, effect in this instance has been minimised. CN contacted the au-
suggesting a progressive failure of the procedure. Although clini- thors of Subramonia 2010 who confirmed they had 12 patients
cal recurrence figures were not significantly increased and VCSS with bilateral varicose veins in their trial. They were treated sepa-
scores were comparable, sonographic findings led the authors to rately, after a minimum of six weeks to avoid recall bias. The re-
conclude that RFA shows poorer results compared to conventional sults from Subramonia 2010 were included in the meta-analysis.
surgery at three years. Rasmussen 2007 performed both the surgery and laser as office-
Subramonia 2010 showed less pain, reduced analgesic require- based procedures. This eliminated the potential bias of general
ments and quicker return to normal activities, work and driving in anaesthesia, but it is not common practice in the UK to perform
patients who underwent RFA. Although this study was not pow- conventional surgery for GSV varicose veins under sedation and
ered to show a difference in QoL outcomes they were shown to be tumescent anaesthesia. Rasmussen 2007 also included patients
consistently improved following RFA, up to five weeks. who had undergone previous high ligation of their SFJ with ul-
trasound evidence of residual SFJ reflux. However, the number
of such limbs was small and was unlikely to affect the outcome
variables.
Overall completeness and applicability of No results are available on recurrence rates, neovascularisation or
evidence re-intervention rates for Darwood 2008. Although they have pub-
lished longer-term outcomes (Theivacumar 2009), in this later
Despite an apparent congruity in the outcome measures of the
publication they included non-randomised patients, without sub-
studies (Table 5), there was a serious lack of compatible data with
group analysis, and subsequently no results were extrapolated for
which any meaningful meta-analysis could be performed. This has
analysis on the randomised cohort. Darwood 2008 admitted that
seriously limited the overall effectiveness of this Cochrane review.
this trial was undertaken early in the units EVLT experience. This
However, all of the included studies were considered applicable to
was the only trial in the Cochrane review to randomise patients
this review, reporting most if not all of the reviews primary and
to two different EVLT techniques (continuous versus pulsed) and
secondary outcomes. The studies all include relevant participants
no significant differences were shown; they have been considered
and interventions.
together in this Cochrane review.
The inclusion and treatment of patients with bilateral great
An additional point that should not be forgotten is the strict in-
saphenous varicose veins has also impeded accurate meta-analy-
clusion criteria used in all trials regarding patients suitability for
sis. As previously documented Rasmussen 2007, Darwood 2008,
endovenous treatment. When considering treatment options for
Subramonia 2010 and the EVOLVeS Study permitted bilateral
patients, if they are deemed unsuitable for endovenous manage-
treatment. Rasmussen 2007 and Darwood 2008 treated patients
(Review)
ment (> 1.2 cm, tortuous veins etc) conventional surgery remains Clearly this review has been limited by a lack of RCTs. No trials
their only viable option, for which there is robust, long-term evi- comparing conventional surgery and foam sclerotherapy (varicose
dence for its safety and efficacy. veins from the great saphenous vein only) were available and subse-
quently no analysis of this novel therapy is possible. Limited meta-
analysis was performed on the five included trials meaning that
the results are of limited value. A poor coherence when reporting
Quality of the evidence outcomes has also limited meaningful meta-analysis.
Overall the quality of available evidence was limited. Of the 95 Where uncertainly was encountered on aspects of the included
RCTs that were assessed, only five were included after screening. studies, authors of the said papers were contacted. All authors who
No RCTs were included comparing conventional surgery with were contacted provided swift and full replies which are docu-
foam sclerotherapy. Only two compared surgery and EVLT, and mented in full within the summary of findings tables for the trials.
three compared RFA and surgery. Where trials had included treatment of both great and small saphe-
Of the results that were available for analysis, huge variations in nous veins, the authors were contacted to ascertain if any subgroup
the reporting of these results limited meaningful meta-analysis for analysis had been performed which would enable us to include the
the majority of proposed outcome measures. trial. Not all authors responded. Of those that did, on no occasion
As above, the biggest limitations to this Cochrane review were had any author performed such an analysis and subsequently these
both the lack of RCTs and a failure to present similar outcomes trials were excluded from the review.
in a congruous manner, thus precluding meta-analysis. A robust Two-year recurrence and neovascularisation figures were published
conclusion can not therefore be reached from the available evi- by Darwood et al (Theivacumar 2009). Unfortunately this pub-
dence. lication included both the randomised patients and patients who
Individually the trials were on the whole well designed and con- had refused randomisation. No subanalysis was made and subse-
ducted. Rasmussen 2007 is a robust, well planned and conducted quently no two-year follow-up data were available for the Dar-
RCT. They recruited sufficient numbers to show statistical signif- wood et al publication.
icance in their outcomes. The selection criteria of all the included trials introduces a bias
Darwood 2008 is also a well designed RCT. However there are into any interpretation of these procedures for the population as a
few details given on allocation concealment in their randomisa- whole; any potential advantage of these novel therapies is obviated
tion process, they did not analyse patients on an intention-to-treat in patients with veins deemed unsuitable for endovascular tech-
basis and they failed to recruit enough patients to perform equiv- niques. In these patients surgery remains the only viable treatment
alence analysis. Their long-term results are not amenable to anal- option.
ysis. Observers were not blinded and only performing mini-phle- The inclusion in some trials of patients with bilateral varicose veins
bectomies on the surgical patients adds a potential confounding and the presentation of results by numbers of limbs introduce a
bias when analysing results of post-operative pain, time to return large potential bias. Clearly bilateral treatment has implications
to work etc. for length of hospital stay, pain scores, operative time and adverse
Despite being a large multi-centre RCT, the EVOLVeS Study failed events. Without stratification of results on a per patient basis, this
to recruit sufficient numbers of patients to demonstrate statistically remains a potential source of confounding bias.
significant differences in some of their study outcomes. Losses to Individually, biases existed in all included trials.
follow up limited the significance of their recurrence rates and Although Rasmussen 2007 reported reduced pain post-operatively
rates of neovascularisation, although admittedly this wasnt the following laser therapy, they did not control for the number of
aim of their study from the outset. Despite a robust randomisation mini-phlebectomies. This adds a potential confounding bias. The
process they ended up with uneven numbers in their two groups, authors relied on the randomisation process to assume an equal
which could potentially affect the results. spread of the number and site of mini-phlebectomies. The au-
Rautio 2002 presented results from a relatively small study. Their thors felt the social security system in Denmark, which ensures
results are therefore limited in their significance. Despite this they full sick leave pay, could have discouraged patients from returning
were able to follow up all of their patients, reporting no losses to to work earlier in the laser group; and the addition of tumescent
follow up at three years. anaesthesia could have concurrently reduced the time off sick in
Subramonia 2010 have published a robust RCT demonstrating the surgical patients. Clearly these variables need further investi-
early results that were of statistical significance. They standardised gation. A further bias was the lack of blinding. The authors admit
their anaesthetic and inter-operator variability thus reducing bias. that although it is not possible to blind patients or surgeons, it is
possible to blind the observer at follow up and this wasnt done.
However they argue that because measures such as pain, quality
of life and recuperation time were patient reported, and recorded
Potential biases in the review process independent of the operating surgeon, this reduces potential bias.
(Review)
Rasmussen 2007 did not record patients occupations, a potential The systematic review instituted by the Society for Vascular
confounding bias when analysing time to return to work figures. Surgery (SVS) in collaboration with the American Venous Forum
In their 2010 publication Rasmussen 2007 criticised their own (AVF) reviewed available evidence in 2009, comparing endove-
failure to record recurrent varicose with the REVAS classification. nous ablation (RFA and EVLT) and foam sclerotherapy versus
This could well have affected their recurrence figures. conventional surgery (Eklof 2009).
Darwood 2008 declared recruitment problems; they were unable Regarding EVLT versus surgery, they concur with the finding sum-
to include sufficient numbers of patients to perform equivalence marised above and considered that RCTs did not report bias pro-
analysis and this limited the significance of their results. Anaes- tection measures, and therefore that the available evidence is of
thetic and additional phlebectomies were also potential confound- low quality. With regard to RFA versus surgery, they conclude
ing variables which were not controlled. All patients undergoing that none of the RCTs reported allocation concealment or blinded
EVLT were operated on under local anaesthetic and no mini-phle- outcomes and hence quality of evidence is considered low.
bectomies were performed until six weeks post-procedure. Surgi- In the current review CN contacted Rasmussen et al to ensure that
cal patients all had a general anaesthetic and additional mini-phle- adequate details were provided demonstrating allocation conceal-
bectomies, which may have affected post-operative recovery. De- ment.
spite adequate randomisation there were differences in the baseline Carradice 2008 is a large RCT which, from its two published
characteristics between the groups as surgical patients were older abstracts to date, appears to demonstrate early benefits of EVLT
and had a higher baseline AVVSS. Again bias could have been and long-term efficacy equivalent to surgery. The author was con-
reduced by blinding follow-up observers but this was not done for tacted and has confirmed that the trial is almost published. Unfor-
practical issues. tunately without full data it is not possible to include the results
EVOLVeS Study did not blind their post-operative observers. They in this present review. The results of Carradice et al and the other
failed to accurately compare treatment times due to problems in ongoing studies will provide exciting additions to future updates
collating data in a uniform fashion. Inter-operator variations seen of this Cochrane review.
between the participating centres, although declared as real world
comparisons, do introduce an element of bias in the absence of a
truly standard technique. There was also considerable variability
between the two groups and between centres with regards to the
anaesthetic used. This was also the only trial to recall their patients AUTHORS CONCLUSIONS
at 72 hours and one week post-operatively, and this could initiate
recall bias. Selection of patients for inclusion into this trial was Implications for practice
based on suitability for RFA. Since the year 2000 these indica- This review lacks sufficient data to provide any robust recommen-
tions have broadened significantly and this could well affect the dations for practice. However, a summary of the available evidence
relevance of these results to the operators of today. suggests the following points.
The biggest limitation of Rautio 2002 is the small numbers of
patients (only 28). This limited the statistical significance of any Patients should be advised that early recurrence and recanalisation
of their findings. Indeed Rautio 2002 commented that despite of GSV varices appears to be similar whether treated by conven-
their results a non-radical treatment of SFJ reflux should still be tional surgery or endovenous ablation and the risk of recanalisa-
considered a main source of recurrent varicose veins. The authors tion in the ablation group appears to increase with time post-ab-
also did not blind their post-operative assessors. In the publication lation. Given this finding, the need for re-intervention is similar
of their three-year outcomes the authors admitted that a lack of in both procedures.
blinding could well influence patients subjective evaluation of
Patients should be advised that neovascularisation can occur in
postoperative pain.
both techniques, however the risk appears to be reduced following
Subramonia 2010 did not blind their post-operative observers.
endovenous ablation by both laser and radiofrequency compared
They also included patients with recurrent varicose veins. Al-
with conventional surgery, although this observed trend was not
though numbers were small this remains a potential source of inac-
statistically significant.
curacy. In their randomisation process they used two stratification
variables (age and sex) which they judged most likely to influence There appear to be no differences in patient satisfaction and quality
outcome in the two groups. No details on how this was achieved of life measures between endovenous and conventional treatments,
were published. This could introduce a potential bias. however the return to normal activities and work may be sooner
following RFA compared with conventional surgery.
Patients should be advised that the risks of both minor and major
Agreements and disagreements with other complications following conventional surgery appear greater than
studies or reviews following RFA and EVLT.
(Review)
No conclusions can be drawn on the pros or cons of general versus In addition, an RCT comparing endovenous ablation (RFA and
local or regional anaesthetic. Both techniques were employed and EVLT) and foam sclerotherapy versus conventional surgery for
are dependent on operator preference and local resources. great saphenous varicose veins needs to be initiated with sufficient
numbers to attain statistical significance. Future trials should aim
Both conventional surgery and EVLT can be performed on a day-
to standardise their follow-up schedule to allow for more accurate
case basis. The same appears to be true for RFA, with only a few
meta-analysis.
patients requiring admission overnight for non-surgical reasons.
This review highlights a need to standardise outcome measures
There are no complete or reliable data to suggest a definite cost of quality of life. The V-Q/SymQ or AVVSS seem a reasonable
benefit of endovenous ablation over conventional surgery. standard to adopt. In addition, in future trials post-operative pro-
Operating theatre time for RFA appears to be longer than for cedure assessors should be blinded to increase the quality of the
conventional surgery. No data are available for EVLT or foam related outcome measures.
sclerotherapy. Although excluded, the RECOVERY Trial compared RFA (Clo-
It must be borne in mind that with any of the novel endove- sureFAST) versus EVLT. In the absence (at this stage) of any tri-
nous techniques, early recanalisation may reflect operator failures als comparing these treatments to surgery, this trial does offer an
rather than true failures of the procedure. The patterns of early interesting insight into RFA versus EVLT. Sixty-nine patients (87
recurrence seen in this meta-analysis may not be a true representa- limbs) were randomised to receive RFA or EVLT in a multi-cen-
tion of treatment failure. Indeed, in subsequent trials we may see tre RCT. Pain, bruising, adverse events and global quality of life
improving patterns of recanalisation that reflect improved opera- scores were significantly better in patients undergoing RFA (Clo-
tor performance. sure FAST) at 48 hours and one, two and four weeks post-pro-
cedure. It will be interesting to see if any of the future trials that
All conclusions are based on a maximum included trial follow-up compare these modalities to conventional surgery confirm their
duration of three years (range five weeks to three years). conclusions.
We can conclude from the current evidence that the minimally Encouragingly, a number of trials that will be suitable for this re-
invasive techniques RFA and EVLT seem to be at least as effective view are due for publication in the near future. Carradice 2008
as surgery in the treatment of great saphenous varicose veins. Un- randomised 267 patients to compare EVLT and HL/S. Rasmussen
fortunately, however, there are insufficient published data on foam 2010a, CLASS, RAFPELS Trial and Magna 2007 are all well con-
sclerotherapy to make any practical conclusions on this technique. structed RCTs comparing surgery to the novel techniques of vari-
cose vein management. What is especially important about these
Implications for research trials is their comparison of all three novel therapies to conven-
tional surgery in the same RCT. With large projected patient num-
There is a clear need for an RCT comparing foam sclerother- bers these could well prove to be the trials to shape the treatment
apy (without surgical intervention in the groin) with conventional of varicose veins in the future.
surgery.
As well as data on primary GSV varicosities, further trial evidence is
The long term (> 5 year) rates of recanalisation following RFA and required regarding the treatment of both small saphenous varicose
EVLT need to be assessed as there appears to be more recanalisation veins and recurrent varicose veins (both SSV and GSV).
in the second and third years of follow up in the included studies.
The question which still remains unanswered with regards to RFA
and EVLT is the longer-term effect (five years) of leaving a proximal
segment of GSV open and leaving the GSV tributaries intact.
ACKNOWLEDGEMENTS
Many studies were undertaken in the earlier days of these novel
CN and RI would like to thank the authors of all included and
techniques. Authors admit to a learning curve with newer tech-
many excluded papers who, when contacted, kindly gave up their
niques. These techniques and the procedural equipment have been
time to provide details to facilitate this review. All authors would
modernised since these early days. Indications for patients who
like to thank the staff from the Cochrane Peripheral Vascular Dis-
are suitable for endovascular intervention are now also broader.
ease group for their support.
There is a lack of robust RCT data comparing these novel thera-
pies with newer equipment and more experienced operators. Such CN would especially like to thank Heather Maxwell as without
trials may reduce the rates of technical error and adverse events her patient help this review would never have been finished; may
demonstrated in earlier trials. we wish her well in her retirement!
(Review)
REFERENCES
References to studies included in this review saphenous vein. Recurrence rate and pattern after two years.
Phlebology 2009; Vol. 24:231.
Darwood 2008 {published data only} Rautio 2002 {published data only}
Beale R, Theivacumar N, Mavor AID, Gough MJ. Perala J, Rautio T, Biancari F, Ohtonen P, Wiik H,
Endovenous Laser Treatment (EVLT) or surgery of Heikkinen T, et al.Radiofrequency endovenous obliteration
varicose veins? A randomised controlled trial in patients versus stripping of the long saphenous vein in the
with saphenofemoral and long saphenous incompetence. management of primary varicose veins: 3-year outcome of a
Yearbook 2005, The Vascular Society of Great Britain and randomized study. Annals of Vascular Surgery 2005;19(5):
Ireland. 2005:Abstract 77. 66972.

Darwood RJ, Theivacumar N, Dellagrammaticas D, Rautio T, Ohinmaa A, Perala J, Ohtonen P, Heikkinen T,
Mavor AI, Gough MJ. Randomized clinical trial comparing Wiik H, et al.Endovenous obliteration versus conventional
endovenous laser ablation with surgery for the treatment of stripping operation in the treatment of primary varicose
primary great saphenous varicose veins. British Journal of veins: a randomized controlled trial with comparison of the
Surgery 2008;95(3):294301. costs. Journal of Vascular Surgery 2002;35(5):95865.
EVOLVeS Study {published data only} Subramonia 2010 {published data only}
Lurie F, Creton D, Eklof B, et al.Prospective randomised Balakrishnan A, Mylankal K, Nalachandran S, Subramonia
study of Endovenous Radiofrequency Obliteration S, Lees T. A randomized controlled trial of radiofrequency
(Closure) Versus Ligation and Vein Stripping (EVOLVeS): ablation and conventional surgery for primary long
2-year follow-up. Journal of Vascular Surgery 2005;42(1): saphenous varicose veins. Phlebology 2008;23(4):198.

178. Subramonia S, Lees T. Randomized clinical trial of
Lurie F, Creton D, Eklof B, Kabnick L, Kistner R, Pichot radiofrequency ablation or conventional high ligation and
O, et al.Prospective randomised study of endovenous stripping for great saphenous varicose veins. British Journal
radiofrequency obliteration (closure) versus ligation and of Surgery 2010;97(3):32836.
stripping (evolves study): early results and one year follow-
up. European Society for Vascular Surgery, Programme References to studies excluded from this review
and Abstract Book, XVII Annual Meeting and Course on
Abela 2008 {published data only}
Vascular Surgical Techniques. 2003:745.
Abela R, Liamis A, Prionidis I, Mathai J, Gorton L,
Lurie F, Creton D, Eklof B, Kabnick LS, Kistner RL,
Browne T, et al.Reverse foam sclerotherapy of the great
Pichot O, et al.Prospective randomised study of endovenous
saphenous vein with sapheno-femoral ligation compared to
radiofrequency obliteration (closure) versus ligation and
standard and invagination stripping: a prospective clinical
vein stripping (EVOLVeS): two-year follow-up. European
series. European Journal of Vascular and Endovascular Surgery
Journal of Vascular and Endovascular Surgery 2005;29(1):
2008;36(4):48590.
6773.
Liamis A, Prionidis I, Mathai J, Gorton L, Browne T,
Panayiotopoulos YP. Long saphenous vein reverse foam
Pichot O, et al.Prospective randomized study of endovenous
sclerotherapy with saphenofemoral junction ligation
radiofrequency obliteration (closure procedure) versus
compared with head and invagination stripping: a
ligation and stripping in a selected patient population
prospective randomized trial. Phlebology. 20 2005; Vol.
(EVOLVeS Study). Journal of Vascular Surgery 2003;38(2):
20, issue 3:149.
20714.
Almeida 2007 {published data only}
Rasmussen 2007 {published data only}
Almeida J. Weve Got Plenty of Data to Show that
Rasmussen LH, Bjoern L, Lawaetz B, Blemings A, Eklof
Endovenous Thermal Ablation is Superior to Open Surgery.
B. Randomised clinical trial comparing endovenous laser
http://www.prolibraries.com/veithsymposium/?select=
ablation with striping of the great saphenous vein: clinical
error&conferenceID=7&sessionID=308. 2007.
outcome and recurrence after 2 years. European Journal of
Vascular and Endovascular Surgery 2010;39(5):6305. Alos 2006 {published data only}

Rasmussen LH, Bjoern L, Lawaetz M, Blemings Alos J, Carreno P, Lopez JA, Estadella B, Serra-Prat M,
A, Lawaetz B, Eklof B. Randomized trial comparing Marinel-Lo J. Efficacy and safety of sclerotherapy using
endovenous laser ablation of the great saphenous vein with polidocanol foam: a controlled clinical trial. European
high ligation and stripping in patients with varicose veins: Journal of Vascular and Endovascular Surgery 2006;31(1):
short-term results. Journal of Vascular Surgery 2007;46(2): 1017.
30815. Anon 2008 {published data only}
Rasmussen LH, Lawaetz M, Bjoern L, Lawaetz B, Blemings Anon. The BRILLIANT Study. http://clinicaltrials.gov/
A, Eklof B. Medium-term follow-up of a randomised ct2/show/NCT00618514?term=endovenous&rank=5
trial comparing laser ablation with stripping of the great 2008.
(Review)
Belcaro 2000 {published data only} results of a randomized controlled trial with 1-year follow-
Belcaro G, Nicolaides AN, Ricci A, Dugall M, Errichi up. Dermatologic Surgery 2007;33(3):27681.
BM, Vasdekis S, et al.Endovascular sclerotherapy, surgery,
Chant 1972 {published data only}
and surgery plus sclerotherapy in superficial venous
Beresford SA, Chant AD, Jones HO, Piachaud D, Weddell
incompetence: a randomized, 10-year follow-up trial--final
JM. Varicose veins: A comparison of surgery and infection/
results. Angiology 2000;51(7):52934.
compression sclerotherapy. Five-year follow-up. Lancet
Blaise 2010 {published data only} 1978;1(8070):9214.

Anon. ESGVS: Sclerotherapy With Lauromacrogol. Study Chant AD, Jones HO, Weddell JM. Varicose veins:
NCT00348764 Information provided by University a comparison of surgery and injection/compression
Hospital, Grenoble 2006. sclerotherapy. Lancet 1972;2(7788):118891.

Blaise S, Bosson JL, Diamand JM. Ultrasound-guided Christenson 2010 {published data only}
sclerotherapy of the great saphenous vein with 1% vs. 3% Christensen JT, Gueddi S, Gemayel G, Bounameaux
polidocanol foam: a multicentre double-blind randomised H. Prospective randomized trial comparing endovenous
trial with 3-Year follow-up. European Journal of Vascular laser ablation and surgery for treatment of primary great
and Endovascular Surgery 2010;39(6):77986. saphenous varicose veins with a 2-year follow-up. Journal of
BLARA Trial {published data only} Vascular Surgery 2010;52(5):123441.
Goode S. A Randomised Study Comparing Endovenous De Medeiros 2006 {published data only}
Laser Ablation and Radiofrequency Ablation for the
De Medeiros CA, Luccas GC. Comparison of endovenous
Treatment of Varicose Veins. European Society for Vascular treatment with an 810 nm laser versus conventional
Surgery Annual Meeting; 2008 Sep 4-7; Nice, France. 208: stripping of the great saphenous vein in patients with
65. primary varicose veins. Dermatologic Surgery 2005;31(12):
Goode S, Crockett M, Richards T, Braithwaite B. 168594.
Preliminary results for the bilateral laser and radiofrequency De Medeiros CAF. Comparison of endovenous laser therapy
ablation study. Phlebology 2008;23(4):1967. vs. conventional stripping of the great saphenous vein:

Goode S, Crockett M, Richards T, Braithwaite B, Midterm results. Jornal Vascular Brasileiro 2006;5(4):
Chowdhury A. The bilateral laser and radiofrequency 27787.
ablation (BLARA) trial. Phlebology 2008;23(5):241.
Demagny 2002 {published data only}
Bountouroglou 2006 {published data only} Demagny A. Comparative study into the efficacy of
Bountouroglou DG, Azzam M, Kakkos SK, Pathmarajah a sclerosant product in the form of liquid or foam in
M, Young P, Geroulakos G. Prospective randomized study echo-guided sclerosis of the arches of the long and short
of ultrasound-guided foam sclerotherapy and adjuvant high saphenous veins. Phlbologie 2002;55(2):1337.
tie under local anaesthesia versus conventional surgery for Desmyttere 2005 {published data only}
primary varicose veins: early results. Phlebology. 19 2004; Desmyttere J, Grard C, Mordon S. A 2 years follow-up
Vol. 19, issue 3:151. study of endovenous 980 nm laser treatment of the great

Bountouroglou DG, Azzam M, Kakkos SK, Pathmarajah saphenous vein: Role of the blood content in the GSV.
M, Young P, Geroulakos G. Ultrasound-guided foam Medical Laser Application 2005;20(4):2839.
sclerotherapy combined with sapheno-femoral ligation
compared to surgical treatment of varicose veins: early Disselhoff 2008 {published data only}
results of a randomised controlled trial. European Journal of Disselhoff BC, Buskens E, Kelder JC, der Kinderen DJ,
Vascular and Endovascular Surgery 2006;31(1):93100. Moll FL. Randomised comparison of costs and cost-
Kalodiki E, Azzam M, Kakkos SK, Zambas N, effectiveness of cryostripping and endovenous laser ablation
Bountouroglou D, Geroulakos G. Ultrasound-guided foam for varicose veins: 2-year results. European Journal of
sclerotherapy combined with saphenofemoral ligation versus Vascular and Endovascular Surgery 2009;37(3):35763.
surgical treatment of varicose veins: intermediate results of a Disselhoff BC, der Kinderen DJ, Kelder JC, Moll FL.
randomized controlled trial. Phlebology 2008;23(5):2423. Randomized clinical trial comparing endovenous laser
ablation of the great saphenous vein with and without
Bush 2008 {published data only} ligation of the sapheno-femoral junction: 2-year results.
Bush RG, Shamma HN, Hammond K. Histological changes European Journal of Vascular and Endovascular Surgery 2008;
occurring after endoluminal ablation with two diode lasers 36(6):7138.
(940 and 1319 nm) from acute changes to 4 months. Lasers
Disselhoff BC, der-Kinderen DJ, Kelder JC, Moll FL.
in Surgery and Medicine 2008;40(10):6769. Randomized clinical trial comparing endovenous laser with
cryostripping for great saphenous varicose veins. British
Ceulen 2007 {published data only}
Journal of Surgery 2008;95(10):12328.
Ceulen RP, Bullens-Goessens YI, Pi-VAN DE Venne SJ,
Nelemans PJ, Veraart JC, Sommer A. Outcomes and side Doran 1975 {published data only}
effects of duplex-guided sclerotherapy in the treatment of Doran FS, White M. A clinical trial designed to discover if
great saphenous veins with 1% versus 3% polidocanol foam: the primary treatment of varicose veins should be by Fegans
(Review)
method or by an operation. British Journal of Surgery 1975; on circulating blood - a controlled randomized study.
62(1):726. Phlebology 2008;23(5):2456.
Duffy 2005 {published data only} Hamel-Desnos 2009 {published data only}
Duffy D. Double-blind prospective comparative trial Hamel-Desnos CM, Gillet JL, Desnos PR, Allaert FA.
between foamed and liquid polidocanol and sodium Sclerotherapy of varicose veins in patients with documented
tetradecyl sulfate in the treatment of varicose and thrombophilia: a prospective controlled randomized study
telangiectatic leg veins: Commentary. Dermatologic Surgery of 105 cases. Phlebology 2009;24(4):17682.
2005;31(6):635. Hayes 2008 {published data only}
Einarsson 1993 {published data only} Hayes CA, Kingsley JR, Hamby KR, Carlow J. The effect
Einarsson E, Eklof B, Neglen P. Sclerotherapy or surgery of endovenous laser ablation on restless legs syndrome.
as treatment for varicose veins: a prospective randomized Phlebology 2008;23(3):1127.
study. Phlebology 1993;8:226. Hinchliffe 2006 {published data only}
Hinchliffe RJ, Ubhi J, Beech A, Ellison J, Braithwaite
Figueiredo 2009 {published data only}
BD. A prospective randomised controlled trial of VNUS
Figueiredo M, Araujo S, Barros N Jr, Miranda F Jr. Results
closure versus surgery for the treatment of recurrent long
of surgical treatment compared with ultrasound-guided
saphenous varicose veins. European Journal of Vascular and
foam sclerotherapy in patients with varicose veins: a
Endovascular Surgery 2006 Sept;31(2):2128.
prospective randomised study. European Journal of Vascular
and Endovascular Surgery 2009 Dec;38(6):75863. Hobbs 1968 {published data only}
Hobbs JT. Surgery and sclerotherapy in the treatment of
Gale 2009 {published data only} varicose veins. A random trial. Archives of Surgery 1974;
Gale SS, Lee JN, Walsh ME, Wojnarowksi DL, Comerota 109:7936.
AJ. A randomized trial of endovenous thermal ablation for Hobbs JT. Surgery or sclerotherapy for varicose veins. 10
superficial venous insufficiency of the great saphenous vein: year results of a random trial. In: Tesi M, Dormandy JA
radiofreqeuncy ablation versus endovenous laser therapy. editor(s). Superficial and Deep Venous Diseases of the Lower
Phlebology 2009;24:231. Limbs/ Surgical Therapy. Edizioni Panminerva Medica,
Hamel-Desnos 2003 {published data only} 1984:2438.

Hamel-Desnos C, Desnos P, Wollmann JC, Ouvry Hobbs JT. The treatment of varicose veins. A random
P, Mako S, Allaert FA. Evaluation of the efficacy of trial of injection-compression therapy versus surgery. British
polidocanol in the form of foam compared with liquid Journal of Surgery 1968;55(10):77780.
form in sclerotherapy of the greater saphenous vein: initial Kabnick 2003 {published data only}
results. Dermatologic Surgery 2003;29(12):11705. Kabnick L. Comparison of 980nm and 810nm diode
Hamel-Desnos C, Ouvry P, Desnos P, Mako S. Evaluation of endovenous ablation lasers for treatment of the great
the efficacy of polidocanol in the form of foam versus liquid saphenous vein. Abstracts of the UIP World Congress
form in sclerotherapy of the long saphenous vein. 16th Chapter Meeting. 2003.

Annual Congress of the American College of Phlebology; Kabnick LS. Outcome of different endovenous laser
2002 Nov 7-10; Fort Lauderdale, Florida. 2002. wavelengths for great saphenous vein ablation. Journal of
Hamel-Desnos 2005 {published data only} Vascular Surgery 2006;43(1):8893.

Hamel-Desnos C, Allaert FA, Benigni JP, Boitelle G, Kalteis 2008 {published data only}
Chleir F, Ouvry P, et al.Study 3/1. Polidocanol foam Kalteis M, Berger I, Messie-Werndl S, Pistrich R, Schimetta
3% versus 1% in the great saphenous vein: early results. W, Polz W, et al.High ligation combined with stripping and
Phlbologie 2005;58(2):16573. endovenous laser ablation of the great saphenous vein: Early
Hamel-Desnos C, Ouvry P, Benigni J-P, Boitelle G, results of a randomized controlled study. Journal of Vascular
Schadeck M, Desnos P, et al.Comparison of 1% and 3% Surgery 2008;47(4):822-9, discussion 829.
polidocanol foam in ultrasound guided sclerotherapy of the Kern 2005 {published data only}
great saphenous vein: A randomised, double-blind trial Kern P. Foam sclerotherapy of telangiectases: a controlled
with 2 year-follow-up. The 3/1 Study. European Journal study. Phlbologie 2005;58(2):1837.
of Vascular and Endovascular Surgery 2007;34(6):7239.
Kern P, Ramelet AA, Wutschert R, Bounameaux H, Hayoz
Hamel-Desnos C, Ouvry P, Desnos P, Allaert FA. 3% D. Single-blind, randomized study comparing chromated
versus 1% polidocanol foam in sclerotherapy of the great glycerin, polidocanol solution, and polidocanol foam for
saphenous vein: randomised double-blind controlled trial treatment of telangiectatic leg veins. Dermatologic Surgery
with a 2-year follow-up. Study 3/1. Phlbologie 2008;61(1): 2004;30(3):36772.
1039. Kuznetsov 2005 {published data only}
Hamel-Desnos 2008 {published data only} Kuznetsov NA, Barinov VE, Teleshov BV, Trepilets VE,
Hamel-Desnos CM, Guias B, Quehe P, Mesgard A, Desnos Zheltikov AN. [Surgical technologies in the treatment of
P, Ferre B, et al.Foam sclerotherapy of saphenous veins with varicose veins of the lower extremities complicated with
or without compression stockings. Biological consequences trophic ulcer]. Khirurgiia 2005, (4):3741.
(Review)
Lin 2007 {published data only} Neglen 1993 {published data only}
Lin Y, Ye CS, Huang XL, Ye JL, Yin HH, Wang SM. A Neglen P, Einarsson E, Eklof B. The functional long-term
random, comparative study on endovenous laser therapy value of different types of treatment for saphenous vein
and saphenous veins stripping for the treatment of great incompetence. Journal of Cardiovascular Surgery 1993;34:
saphenous vein incompetence. Chung-Hua i Hsueh Tsa 295301.
Chih/Chinese Medical Journal 2007;87(43):30436. Ogawa 2008 {published data only}
Lugli 2009 {published data only} Ogawa T, Hoshino S, Makimura S, Shigematsu H, Azuma
Lugli M, Cogo A, Guerzoni S, Petti A, Maleti O. Effects N, Sasajima T, et al.Endovenous laser ablation compared
of eccentric compression by a crossed-tape technique after with stripping - multicenter RCT in Japan. Presented at
endovenous laser ablation of the great saphenous vein: a American Venous Forums annual meeting in Charleston, SC,
randomized study. Phlebology 2009;24(4):1516. USA February 2008.
Lupton 2002 {published data only} Ouvry 2008 {published data only}
Lupton JR, Alster TS, Romero P, Lupton JR, Alster TS, Ouvry P, Allaert FA, Desnos P, Hamel-Desnos C. Efficacy of
Romero P. Clinical comparison of sclerotherapy versus polidocanol foam versus liquid in sclerotherapy of the great
long-pulsed Nd:YAG laser treatment for lower extremity saphenous vein: a multicentre randomised controlled trial
telangiectases. Dermatologic Surgery 2002;28(8):6947. with a 2-year follow-up. European Journal of Vascular and
Martimbeau 2003 {published data only} Endovascular Surgery 2008;36(3):36670.
Martimbeau PR. A randomized clinical trial comparing the Rabe 2008 {published data only}
sclerosing and side effects of foam vs. liquid formula for Rabe E, Otto J, Schliephake D, Pannier F. Efficacy
sclerotherapy of primary varicose veins. 2003 UIP World and safety of great saphenous vein sclerotherapy using
Congress Chapter Meeting; 2003 Aug 27-31; San Diego, standardised polidocanol foam (ESAF): a randomised
California. 2003. controlled multicentre clinical trial. European Journal of
Maurins 2009 {published data only} Vascular and Endovascular Surgery 2008;35(2):23845.
Maurins U, Rabe E, Pannier F. Does laser power influence Rao 2005 {published data only}
the results of endovenous laser ablation (EVLA) of Rao J, Wildemore JK, Goldman MP. Double-blind
incompetent saphenous veins with the 1 470-nm diode prospective comparative trial between foamed and liquid
laser? A prospective randomized study comparing 15 and polidocanol and sodium tetradecyl sulfate in the treatment
25 W. International Angiology 2009;28(1):327. of varicose and telangiectatic leg veins. Dermatologic Surgery
McDaniel 1999 {published data only} 2005;31(6):6315.
McDaniel DH, Ash K, Lord J, Newman J, Adrian REACTIV {published data only}
RM, Zukowski M. Laser therapy of spider leg veins: Michaels JA, Campbell WB, Brazier JE, MacIntyre JB,
clinical evaluation of a new long pulsed alexandrite laser. Palfreyman SJ, Ratcliffe J, et al.Randomised clinical trial,
Dermatologic Surgery 1999;25(1):528. observational study and assessment of cost-effectiveness of
Mekako 2006 {published data only} the treatment of varicose veins (REACTIV trial). [Review]
Mekako A. EVLT for Sapheno-Popliteal incompetence and [220 refs]. Health Technology Assessment (Winchester,
SS Reflux: A RCT. ISRCTN Register 2006. England) 2006;10(13):1-196, iii-iv.
Mekako 2007 {published data only} RECOVERY Trial {published data only}

Carradice D, Mekako A, McCollum C, Chetter I. Almeida JI, Kaufman J, Gockeritz O, Chopra P, Evans
Endovenous Laser Therapy With Concomitant Or MT, Hoheim DF, et al.Radiofrequency endovenous
Sequential Phlebectomy: A Randomised Controlled Trial. ClosureFAST versus laser ablation for the treatment
European Society for Vascular Surgery Annual Meeting; of great saphenous reflux: a multicenter, single-blinded,
2008 Sep 4-7; Nice, France. 2008:67. randomized study (RECOVERY study). Journal of Vascular
Carradice D, Mekako AI, Hatfield J, Chetter IC. and Interventional Radiology 2009;20(6):7529.
Randomized clinical trial of concomitant or sequential Anon. RECOVERY Trial; Final Results. http://vnus.fr/pdf/
phlebectomy after endovenous laser therapy for varicose Recovery-Data.pdf 2009.
veins. British Journal of Surgery 2009;96(4):36975. Kaufman JA, Williamson JA. Radiofrequency Endovenous

Mekako AI, Hatfield J, Abdul Rahman MNA, Gulati S, ClosureFAST versus Laser Ablation for the Treatment of
McCollum PT, Chetter IC. Endovenous laser therapy with Great Saphenous Reflux: A Multicenter, Single-blinded,
concomitant or sequential phlebectomy: a randomized Randomized Study (RECOVERY). 34th Annual Scientific
controlled trial. The Vascular Society of Great Britain and Meeting of the Society of Interventional Radiology; 2009
Ireland Yearbook. 2007:72. Mar 7-12; San Diego, USA. 2009:Abstract no: 22.
NCT00841178 {published data only} Rutgers 1994 {published data only}
NCT00841178. Endovenous Laser Therapy (EVLT) for Rutgers PH, Kistlaar PJ. Randomized trial of stripping
Sapheno-Popliteal Incompetence and Short Saphenous Vein versus high ligation combined with sclerotherapy in the
(SSV) Reflux: A RCT. http://clinicaltrials.gov/ct2/show/ treatment of the incompetent greater saphenous vein.
NCT00841178 (accessed 25 September 2010). American Journal of Surgery 1994;168(4):3115.
(Review)
Rybak 2003 {published data only} Surgery, sclerotherapy, foamsclerotherapy and combined
Rybak Z, Garcarek J. Aethoxysclerol foam obliteration of (surgery+sclerotherapy) options. A 10-year, prospective,
insufficient perforating veins in patients suffering from leg randomised, controlled, follow-up study. The VEDICO*
ulcers: A clinical recommendation. Abstracts from the UIP trial and EST (European Sclerotherapy Trial). Angeiologie
World Congress Chapter Meeting, San Diego Aug 27-31. 2003;55(1):2936.

2003. Belcaro G, Cesarone MR, Di Renzo A, Brandolini R,
Sadoun 2003 {published data only} Coen L, Acerbi G, et al.Foam-sclerotherapy, surgery,
Sadoun S. Long-term follow-up study of ultrasound sclerotherapy, and combined treatment for varicose veins:
findings in varicose greater saphenous veins treated with a 10-year, prospective, randomized, controlled, trial
foam. Abstracts from the 2003 UIP World Congress (VEDICO trial). Angiology 2003;54(3):30715.
Chapter Meeting, San Diego Aug 27-31. 2003. Viarengo 2007 {published data only}
Seddon 1973 {published data only} Viarengo LM, Potrio-Filho J, Potrio GM, Menezes FH,
Seddon J. The management of varicose veins. British Meirelles GV. Endovenous laser treatment for varicose veins
Journal of Surgery 1973;60(5):3457. in patients with active ulcers: measurement of intravenous
and perivenous temperatures during the procedure.
Selles 2008 {published data only}
Dermatologic Surgery 2007;33(10):1234-42, discussion
Selles DR, Arenas RJ, Ballester IC, Perez MJ, Gonzalez VS,
1241-2.
Ruiz del CJ. [Phlebectomies or foam sclerosis for treatment
of the distal venous segment in the 3-S saphenectomy Vuylsteke 2006 {published data only}
technique]. [Spanish]. Cirugia Espanola 2008;84(2):929. Vuylsteke M, Van den Bussche D, Audenaert E A, Lissens P.
Endovenous laser obliteration for the treatment of primary
Sica 2006 {published data only}
varicose veins. Phlebology 2006;21:8087.
Sica M, Biasi G, George E. A simplified method of
endovenous injection of 1% trombovar microfoam via Wright 2006 {published data only}
a short catheter (m.s. method): Lasting effectiveness Wright D. European randomized controlled trial of
confirmed by 2 year results [Mthode simplifie dinjection Varisolve PD microfoam compared with alternative
endoveineuse de micromousee de Trombovar 1% therapy in management of moderate-to-severe varicose
par cathter court (mthode MS): une efficacit durable veins: preliminary results. 2003 UIP World Congress
confirme par les rsultats 2 ans]. Phlbologie 2006;59(4): Chapter Meeting; 2003 Aug 27-31; San Diego, California.
33942. 2003.

Wright D, Gobin JP, Bradbury AW, Coleridge-Smith
Stotter 2005 {published data only}
P, Spoelstra H, Berridge D, et al.Varisolve polidocanol
Stotter L, Schaaf I, Bockelbrink A, Baurecht HJ.
microfoam compared with surgery or sclerotherapy in the
Radiofrequency obliteration, invagination or cryostripping:
management of varicose veins in the presence of trunk
Which is the best tolerated treatment by the patients?.
vein incompetence: European randomized controlled trial.
Phlbologie 2005;34(1):1924.
Phlebology 2006;21(4):18090.
Theivacumar 2008 {published data only}

Theivacumar NS, Dellagrammaticas D, Mavor AI, Gough Yamaki 2008 {published data only}
MJ. Endovenous laser ablation: does standard above-knee Yamaki T, Nozaki M, Sakurai H, Takeuchi M, Soejima
great saphenous vein ablation provide optimum results K, Kono T. Multiple small-dose injections can reduce the
in patients with both above- and below-knee reflux? A passage of sclerosant foam into deep veins during foam
randomized controlled trial. Journal of Vascular Surgery sclerotherapy for varicose veins. European Journal of Vascular
2008;48(1):1738. and Endovascular Surgery 2009;37(3):3438.

Theivacumar SN, Dellagrammaticas D, Darwood RJ, Yamaki T, Nozaki M, Sakurai H, Takeuchi M, Soejima
Mavor AI, Gough MJ. Endovenous laser ablation (EVLA): K, Kono T. Prospective randomized efficacy of ultrasound-
is standard above-knee great saphenous vein (AK-GSV) guided foam sclerotherapy compared with ultrasound-
ablation sufficient? A randomised controlled trial. The guided liquid sclerotherapy in the treatment of symptomatic
Vascular Society of Great Britain and Ireland Yearbook. venous malformations. Journal of Vascular Surgery 2008;47
2007:71. (3):57884.
Theivacumar 2009 {published data only} Zeh 2003 {published data only}
Theivacumar NS, Darwood R, Gough MJ. Zeh RG. Expanding sclerosing properties of polidocanol
Neovascularisation and recurrence 2 years after varicose foam with Gelofusine. 2003 UIP World Congress
vein treatment for sapheno-femoral and great saphenous Chapter Meeting; 2003 Aug 27-31; San Diego, California
vein reflux: a comparison of surgery and endovenous laser 2003.
ablation. European Journal of Vascular and Endovascular References to ongoing studies
Surgery 2009;38(2):2037.
VEDICO Trial {published data only} Carradice 2008 {published data only}
Belcaro G, Cesarone M, Di Renzo A, Brandolini R, Carradice D, Mekako A, Hatfield J, Chetter I. Recurrent
Coen L, Acerbi G, et al.Treatments for varicose veins: varicose veins are more common following surgery than
(Review)
EVLT - results of a randomised controlled trial. The Fischer 2002
Vascular Society of Great Britain and Ireland Yearbook 2009 Fischer R, Chandler JG, De Maeseneer MG, Frings N,
2009:86. Lefebvre-Vilarbedo M, Earnshaw JJ, et al.The unresolved
Carradice D, Mekako AI, Hatfield J, Chetter IC. A problem of recurrent saphenofemoral reflux. Journal of the
randomised trial of EVLT vs surgery for varicose veins. The American College of Surgeons 2002;195(1):8094.
Vascular Society of Great Britain and Ireland Yearbook Frullini 2002
2008. 2008:91. Frullini A, Cavezzi A. Sclerosing foam in the treatment of
CLASS {published data only} varicose veins and telangiectases: history and analysis of
Brittenden J. Randomised controlled trial comparing foam safety and complications. Dermatologic Surgery 2002;28(1):
sclerotherapy, alone or in combination with endovenous 115.
laser therapy, with conventional surgery as a treatment for
Hayden 2001
varicose veins. CLASS (Comparison of LAser, Surgery and
Hayden A, Holdsworth J. Complications following re-
foam Sclerotherapy). http://www.controlled-trials.com/
exploration of the groin for recurrent varicose veins. Annals
ISRCTN51995477.
of the Royal college of Surgeons of England 2001;83(4):2723.
Magna 2007 {published data only}
Anon. Surgery or Noninvasive Therapy for Varicose Higgins 2011
Veins (Magna). available from www.clinicaltrials.gov Higgins JPT, Green S, editors. Cochrane Handbook for
NCT00529672 2007. Systematic Reviews of Interventions Version 5.1.0 [updated
March 2011]. The Cochrane Collaboration, 2011.
RAFPELS Trial {published data only}
Available from www.cochrane-handbook.org.
NCT00621062. Prospective Randomized Trial Comparing
the New Endovenous Procedures Versus Conventional Jones 1996
Surgery for Varicose Veins Due to Great Saphenous Vein Jones L, Braithwaite BD, Selwyn D, Cooke S, Earnshaw
Incompetence (RAFPELS). http://clinicaltrials.gov/ct2/ JJ. Neovascularisation is the principle cause of varicose
show/NCT00621062 (assessed 25 September 2010). vein recurrence: results of a randomised trial of stripping
the long saphenous vein. European Journal of Vascular and
Rasmussen 2010a {published data only}
Endovascular Surgery 1996;12(4):4425.
Rasmussen L. Randomised trial shows higher technical
failure rate with foam. VascularNews 2010;May:33. Lurie 2003
Additional references Pichot O, et al.Prospective randomised study of endovenous
Baccaglini 1997 radiofrequency obliteration (closure procedure) versus
Baccaglini U, Spreafico G, Castoro C, Sorrentino P. ligation and stripping in a selected patient population
Consensus conference on sclerotherapy or varicose veins of (EVOLVeS Study). Journal of Vascular Surgery 2003;38(2):
the lower limbs. Phlebology 1997;12(1):216. 20714.
Lurie 2005
Bergan 2002
Lurie F, Creton D, Eklof B, et al.Prospective randomised
Bergan JJ, Kumins NH, Owens EL, Sparks SR. Surgical
study of Endovenous Radiofrequency Obliteration (closure)
and endovascular treatment of lower extremity venous
Versus Ligation and Vein Stripping (EVOLVeS): 2-year
insufficiency. Journal of Vascular and Interventional
follow-up. Journal of Vascular Surgery 2005;42(1):178.
Radiology 2002;13(6):5638.
Critchley 1997 Merchant 2002
Critchley G, Handa A, Maw A, Harvey A, Harvey MR, Merchant RF, DePalma RG, Kabnick LS. Endovascular
Corbett CR. Complications of varicose vein surgery. Annals obliteration of saphenous reflux: a multicentre study.
of the Royal College of Surgeons of England 1997;79(2): Journal of Vascular Surgery 2002;35(6):11906.
10510. Min 2003
Darke 1992 Min RJ, Khilnani N, Zimmet SE. Endovenous laser
Darke SG. The morphology of recurrent varicose veins. treatment of saphenous vein reflux: long-term results.
European Journal of Vascular Surgery 1992;6(5):5127. Journal of Vascular and Interventional Radiology 2003;14(8):
9916.
Dwerryhouse 1999
Dwerryhouse S, Davies B, Harradine K, Earnshaw JJ. Navarro 2001
Stripping the long saphenous vein reduces the rate of Navarro L, Min RJ, Bone C. Endovenous laser: a new
reoperation for recurrent varicose veins: five-year results of a minimally invasive method of treatment for varicose veins-
randomized trial. Journal of Vascular Surgery 1999;29(4): -preliminary observations using an 810 nm diode laser.
58992. Dermatologic Surgery 2001;27(2):11722.
Eklof 2009 Nicolini 2005
Eklof B. Fire, Foam, and Knife for Varicose Veins: What Nicolini P. Closure Group. Treatment of primary varicose
Have Prospective Randomized Clinical Trials Taught Us?. veins by endovenous obliteration with the VNUS closure
VEITHsymposium 2009. system: results of a prospective multicentre study. European
(Review)
Journal of Vascular and Endovascular Surgery 2005;29(4): Rautio 2002a
4339. Rautio T, Ohinmaa A, Perala J, Ohtonen P, Heikkenen T,
Perala 2005 Wiik H, et al.Endovenous obliteration versus conventional
Perala J, Rautio T, Biancari F, Ohtonen P, Wiik H, stripping operation in the treatment of primary varicose
Heikkinen T, et al.Radiofrequency endovenous obliteration veins: a randomized controlled trial with comparison of the
versus stripping of the long saphenous vein in the costs. Journal of Vascular Surgery 2002;35(5):95865.
management of primary varicose veins: 3-year outcome of a
Sarin 1992
randomized study. Annals of Vascular Surgery 2005;19(5):
Sarin S, Scurr JH, Coleridge Smith PD. Assessment of
66972.
stripping the long saphenous vein in treatment of primary
Pichot 2004 varicose veins. British Journal of Surgery 1992;79(9):
Pichot O, Kabnick LS, Creton D, Merchant RF, Schuller- 88993.
Petroviae S, Chandler JG. Duplex ultrasound scan findings
two years after great saphenous vein radiofrequency Tessari 2001
endovenous ablation. Journal of Vascular Surgery 2004;39 Tessari L, Cavezzi A, Frullini A. Preliminary experience with
(1):18995. a new sclerosing foam in the treatment of varicose veins.
Dermatologic Surgery 2001;27(1):5860.
Rasmussen 2010
Rasmussen LH, Bjoern L, Lawaetz B, Blemings A, Eklof Wigger 1998
B. Randomised clinical trial comparing endovenous laser Wigger P. Surgical therapy of primary varicose veins.
ablation with striping of the great saphenous vein: clinical Schweizerische Medizinische Wochenschrift. Journal Suisse de
outcome and recurrence after 2 years. European Journal of Medecine 1998;128(45):17818.

Vascular and Endovascular Surgery 2010;38(5):6305. Indicates the major publication for the study
(Review)
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Darwood 2008
Methods Study design: Prospective RCT

Method of randomisation: Block randomisation using sealed envelopes. Randomisation
was stratified according to consultant to allow for variations in technique. Bilateral
varicose veins were randomised once and had the same treatment on both legs
Concealment of allocation: Sealed envelopes
Blinded: Not for operator or patient at treatment. No mention of blinding at follow up
No of patients randomised: 118 patients (136 legs)
No of patients analysed: 95 patients (114 legs)
Exclusions post randomisation: Seven patients (11 legs) withdrew from the study as
not happy with their treatment allocation. Six patients were treated outside the study
interval and were also excluded
Losses to follow up: 11 patients
Intention-to-treat analysis: no (one surgery patient had EVLT 1 and was followed up
in the laser cohort)
Source of funding: Promed (Bluntisham, UK) - sponsor had no input in study design,
data collection, data analysis/interpretation or preparation of the manuscript
Participants Country: UK
Setting/Location: Hospital
Number: 534 legs assessed for eligibility. 136 randomised. 95 patients entered study
(114 legs). 103 legs analysed
Patients with bilateral veins received the same treatment simultaneously on each leg
Age: Three treatment groups [median (IQ range)]; EVLT1: 42 (30.5 - 54.5) / EVLT2:
52 (35 - 59) / Surgery: 49 (38.5 - 57.5)
Sex: Three treatment groups sex ratio (F:M); EVLT1: 22:16 / EVLT2: 16:11 / Surgery:
16:14
Inclusion criteria: >18 years of age. Symptomatic varicose veins and primary SFJ in-
competence (confirmed on duplex ultrasonography)
Exclusion criteria: On warfarin. Unsuitable for EVLT (tortuous GSV, large incompetent
anterior accessory saphenous vein)
Interventions Treatment(s): EVLT versus surgery. 2 EVLT techniques: 12W power with 1s laser pulses
and 1s intervals between pulses. Laser fibre withdrawn 2-3mm during intervals (EVLT1).
14W continuous power and continuous laser withdrawal (EVLT2). Surgery: SFJ ligation,
GSV stripping to knee level + multiple phlebectomies of varicosities
Control/Placebo: Control = conventional surgery
Duration: Follow up at 1, 6, 12 weeks and 1 year after treatment.
Outcomes Primary outcomes: Abolition of reflux in the treated segment of GSV and improvement
in disease-specific quality of life three months after treatment
Secondary outcomes: Post-procedure pain. Time to return to normal activity + work.
Cosmesis. Overall satisfaction at three months
(Review)
Darwood 2008 (Continued)
Notes Authors reported difficulty recruiting patients to the study. They did not meet the sample
sizes for their study groups to make their desired power calculations. Statistical tests for
equivalence were therefore not performed
Risk of bias
Bias Authors judgement Support for judgement
Random sequence generation (selection Low risk Block randomisation using sealed en-
bias) velopes.
Randomisation was stratified by consultant
to allow for any minor variations in tech-
nique. No clear details on how this strati-
fication was achieved
Allocation concealment (selection bias) Unclear risk No details of adequate safeguards. No men-
tion that envelopes were numbered, sealed
or opaque
Blinding (performance bias and detection High risk Not possible to blind investigators or pa-
bias) tients. Assessors at follow up were also not
All outcomes blinded
Incomplete outcome data (attrition bias) Low risk Missing outcome data was documented (23
Outcomes in total). 11 legs withdrew owing to dissat-
isfaction with allocated treatment, (6 laser/
5 surgery), 11 limbs were treated outside of
the RCT trial period and were subsequently
excluded. One limb was excluded from the
surgery group because they were switched
over to receive laser (EVLT1)
Incomplete outcome data (attrition bias) Low risk Missing data is balanced across the groups,
All outcomes with similar reasons given for the missing
data
Selective reporting (reporting bias) Low risk The pre-specified outcomes in the study
protocol were reported in the pre-specified
way
Authors declared that their sample sizes
were insufficient to permit statistical test-
ing for equivalence
Analysis was not on an intention to treat
basis.
Other bias High risk Authors reported difficulty recruiting pa-

tients to the study. They declared that their
sample sizes were insufficient to permit sta-
tistical testing for equivalence
(Review)
Darwood 2008 (Continued)
One patient randomised to surgery under-

went laser, and was followed up in the laser
cohort showing no analysis of intention to
treat
Some patients received bilateral treatment.
Authors stated these patient would receive
the same treatment on both limbs. These
patients who received bilateral treatment
were not stratified in the results. Some out-
come measures can be affected by bilateral
treatment e.g. QoL, pain, time to return to
work etc. therefore introducing a bias
Patients who underwent laser did not have
concomitant mini-phlebectomies. This
adds a potential confounding bias when
analysing results of post-operative pain,
time to return to work etc
EVOLVeS Study
Methods Study design: Multi-centre prospective RCT

Method of randomisation: Randomisation was allocated via an Internet based central
system
Concealment of allocation: Not clear
Blinded: not possible to blind operators or patients in this trial. Assessors at follow up
would also be aware which procedure the patients had undergone
No of patients randomised: 85 patients (86 limbs)
No of patients analysed: 79 patients (80 limbs)
Exclusions post randomisation: 3 patients refused surgery, 1 patient repeatedly DNAd,
2 patients excluded from RFA due to inclusion criteria violation
Losses to follow up: Yes: 1) Clinical Ex: 2 surgery + 1 RFA no follow up at 4 months /
2) QoL questionnaires: Surgery: 1 at 72 hours, 4 at 4 months not completed. RFA: 1 at
72hr, 1 wk, 3 wk + 4 months not completed / At 1 year 19 limbs in RFA and 16 limbs
in surgery were lost but at 2 years it improved with only 8 RFA and 7 surgery losses
Intention-to-treat analysis: No
Source of funding: VNUS medical technologies provided financial support for data
collection, clinical monitors and disposable catheters (RFA) free of charge
Participants Country: France, Austria and USA

Setting/Location: Hospitals
Number: 85 patients
Age: [mean SD] Surgery 47 4 / RFA 49 4
Sex: RFA F 32 M 12 / Surgery F 26 M 10
Inclusion criteria: Reverse flow in GSV lasting > 0.5s in standing position / Age 21 - 80
/ CEAP classification C2, C3, C4 / Ambulatory status / Segmental deep reflux allowable
/ Saphenous vein diameter 1.2 cm in supine position / Availability for follow up visits
- 72 hours, 1 wk, 3 wk, 4 months
Exclusion criteria: Saphenous vein diameter > 1.2 cm or < 0.2 cm / Duplication of
saphenous trunk or incompetent accessory branch / Small saphenous vein reflux / Varices
(Review)
EVOLVeS Study (Continued)
of the thigh / Previous DVT / ABPI < 0.9 / Axial deep venous reflux from groin through
popliteal vein / Tortuosity of GSV segment to be treated on basis of appearance and USS
as unsuitable for catheterisation
Interventions Treatment(s): Conventional vein stripping (from knee or upper calf to the SFJ) with
high ligation of SFJ versus GSV obliteration with RFA without high ligation of SFJ
Control/Placebo: Control = surgery
Duration: Follow up was at 72 hours, 1 and 3 weeks, 4 months, 1 and 2 years
Outcomes Primary outcomes: It is not clear from the paper what the specific primary or secondary
measures were. EVOLVeS was designed to compare procedure-related complications,
patient recuperation and QoL outcomes
Secondary outcomes: Although it was not initially declared, the EVOLVeS trials later
presented rates of neovascularisation in the groin and recurrence at 2 years
Notes Two investigators audited the studys raw data handling and storage methods, data pro-
cessing accuracy, and presentation of specific results. They reported all was in order and
that the raw data reflected the results accurately. This was done at four months and two
years post data collection
Risk of bias
Random sequence generation (selection Unclear risk Randomisation was allocated via Internet
bias) - no further details were given
Allocation concealment (selection bias) Unclear risk No details on adequate concealment.
Blinding (performance bias and detection High risk Not possible to blind patients or operators.
bias) Assessors were also not blinded, not appli-
All outcomes cable in this study
Incomplete outcome data (attrition bias) Low risk Details were provided on all missing out-
Outcomes come data in the early EVOLVeS trial.
However it led to an imbalance in the study
group cohorts
86 limbs randomised (46 RFA, 40 surgery)
. 4 limbs excluded in surgery (3 ex-
cluded themselves after randomisation, 1
was treated outside the study period) 1 in
RFA (when surgeon learned of previous
surgical intervention in the target vein).
The final result was 45 RFA, 36 surgery
There is also discrepancy with the missing
outcome data and explanations of this miss-
ing data compared to the published two
year follow up:
(Review)
EVOLVeS Study (Continued)
44 limbs RFA, 36 limbs surgery. 2 patients

excluded from RFA because of inclusion
criteria violation, 4 patients excluded from
surgery because they refused their allocated
treatment
Incomplete outcome data (attrition bias) High risk Missing data led to an imbalance in the
All outcomes groups; 45 RFA, 36 surgery
Discrepancy between initial and 2 year re-
ports.
way
Other bias High risk The RFA treatment cohort included one
patient who underwent treatment of both
limbs with a three month gap between
treatments. The patient was only ran-
domised once and each limb treated as a
separate episode
All centres were established centres in the
use of RFA and the company funded the
research. No subjective data was reported.
However as in all of these studies surgical
technique and ultrasonographic results are
operator dependent
Rasmussen 2007
Methods Study design: RCT

Method of randomisation: blocks of 10 concealed envelopes. Groups were well
matched for demographic data, CEAP classification and GSV details
Concealment of allocation: envelopes
Blinded: not clear
No of patients randomised: 121 patients (137 legs)
No of patients analysed: see losses to follow up
Exclusions post randomisation: none
Losses to follow up: 12 days - EVLT 2, HL/S 0; 1 month - EVLT 4, HL/S 2; 3 months
- EVLT 6, HL/S 5; 6 months - EVLT 15, HL/S 18
Intention-to-treat analysis: Yes
Source of funding: grant from the Public Health Insurance Research Foundation of
Denmark. EVLT catheters provided, in part, by Biolitec AG (Bonn, Germany) and
Micronmed (Kristianstad, Sweden)
Participants Country: Denmark

Setting/Location: office based setting, private clinic
Number: Overall 121 patients (137 legs). 59 patients (68 legs) - HL/S. 62 patients (69
legs) - EVLT
(Review)
Rasmussen 2007 (Continued)
Patients with bilateral veins received the same treatment simultaneously on each leg
Age: [mean (range)] HL/S 54 (22 - 78), EVLT 53 (26 - 79)
Sex: HL/S male 16:female 43, EVLT male 21:female 41
Inclusion criteria: CEAP C2 - 4, Ep, As, Pr / Informed consent / Age 18 - 80 / GSV
incompetence confirmed by > 0.5s reflux on duplex imaging
Exclusion criteria: Duplication of GSV or incompetent anterior accessory GSV / SSV
reflux (or < 3 months since surgery for SSV incompetence) / Previous DVT / ABPI < 0.
9 or Hx arterial disease / Fem or pop insufficiency / tortuous GSV
Interventions Treatment(s): EVLT (duplex guided) 980 nm diode laser, 1.5 sec pulses, 1.5 sec pause,
12 W energy
Control/Placebo: High tie strip and multiple stab avulsion (HL/S)
Duration: Follow up 12 days, 1, 3, and 6 months, 2 years post procedure
Outcomes Primary outcomes: It is not clear what their specific primary or secondary measures
were. Rasmussen et al set out to assess safety, efficacy, post-operative morbidity, sick leave,
QoL and costs. They reported results on:
Absence from work and normal activity / AVVSS / Short Form-36 (SF-36 score) / Varicose
Vein Severity Score (VVSS) / pain (visual analogue score)/ complications (minor e.g.
required no treatment versus major e.g. required treatment, hospitalisation, permanent
sequelae or death) / Cost (procedure and days off sick from work)
Secondary outcomes:
Notes Eight patients in each group had previous high ligation i.e. were recurrent. They were
permitted as they had a patent refluxing SFJ and GSV
Author contacted and further information on randomisation process given: A block of
10 envelopes would ensure that a sufficient number of each treatments were available, i.
e. 5 of each all the time. This is like tossing a coin but easier to document. The envelopes
were kept in a basket, but the basket was filled by a research nurse when the patients
were not present.
All envelopes were alike. There was no chance of bias.
Risk of bias
Random sequence generation (selection Low risk After contacting the author, further details
bias) on the random sequence generation were
confirmed:
Blocks of 10 envelopes kept in a basket, the
basket was filled by a research nurse when
the patients were not present. All envelopes
were alike.
Allocation concealment (selection bias) Low risk After contacting the author, further details
on allocation concealment were confirmed:
The envelopes were kept in a basket, but
the basket was filled by a research nurse
when the patient were not present. All en-
(Review)
Rasmussen 2007 (Continued)
velopes were alike.
Blinding (performance bias and detection High risk Impossible to blind operator or patients to
bias) treatment. No mention that assessors post-
All outcomes operatively were blinded
Incomplete outcome data (attrition bias) Low risk 121 patients (137 limbs) were randomised;
Outcomes 68 surgery, (HL/S) and 69 laser (EVL)
There were no exclusions from the study
post randomisation.
Losses to follow up were documented:
At 12 days - EVL 2, HL/S 0; 1 month -
EVL 4, HL/S 2; 3 months EVL 6, HL/S 5;
6 months EVL 15, HL/S 18
These losses to follow up did not have an
impact on the outcome measures. The two
treatment groups remained similar in num-
bers despite these losses
Incomplete outcome data (attrition bias) Low risk As above

All outcomes
way
Additional outcome measures were re-
ported in a subsequent publication (2 year
results) reporting recurrence rates, which
were not a pre-specified outcome measure.
However this does not introduce any bias
or inaccuracy into the trial
Other bias High risk 121 patients (137 limbs): Included 16 pa-
tients with bilateral varicose veins. No strat-
ification of these patients in the results.
Could introduce a potential bias into re-
sults such as pain, time to return to normal
activities, QoL etc
(Review)
Rautio 2002

Method of randomisation: The sealed envelope method
Concealment of allocation: From the author: 36 named tags to identical envelopes,
which were sealed. After shuffling the envelopes I numbered them randomly. List of
numbers for randomisation was done earlier according to instructions of the biostatisti-
cian of our department. I opened the envelopes in numerical order.
Blinded: Not possible to blind patients or operators. No subjective measurements were
made post-operatively so assessors were not blinded
No of patients randomised: 36
No of patients analysed: 28 (15 surgery, 13 RFA)
Exclusions post randomisation: three patients left as found schedule unsuitable, further
four refused to chosen treatment and one excluded due to pregnancy
Losses to follow up: No
Intention-to-treat analysis: One patient was excluded after randomisation but not
withdrawn from the study indicating some intention to treat process, but which group
this patient retired from is not made clear
Source of funding: Grant from University of Oulu, Finland
Participants Country: Finland

Setting/Location: University of Oulu
Number: 36 patients (8 subsequently excluded/left) 33 patients randomised
Age: RFA 33 mean (SD 6.7) / Surgery 38 (SD 6.8)
Sex: male: female - RFA 1:14 / Surgery 1:12
Inclusion criteria: Confirmed reflux (USS > 2 sec GSV reflux) / pt suitable for day
case / symptomatic previously untreated uncomplicated GSV tributary varicosities and
isolated unilateral SFJ incompetence
Exclusion criteria: Coagulopathies / pregnancy / multiple, tortuous (> 90 degree bend)
large-diameter GSV trunks / bilateral varicose veins / concomitant SSV varicosities
Interventions Treatment(s): Radiofrequency ablation - RFA Closure system, inserted into GSV at
ankle level. No ligation of SFJ
Control/Placebo: Conventional surgery: SFJ ligation of all tributaries and stripping of
GSV to just below knee
Duration: Follow up for three years.
Outcomes Primary outcomes: It is not clear from Rautio et al what their specific primary or
secondary measures were. Rautio et al aimed to evaluate outcome in terms of pain, sick
leave health related QoL and cost
Secondary outcomes: Rautio et al assessed further outcomes at three years including
recurrence, satisfaction, VCSS, VSDS and the VDS. Patency of GSV and presence of
neovascularisation was also assessed
Notes Author contacted 2/2/2010.

Replied 8/2/2010:
The 36 patients had their preoperative diagnostic done in an earlier trial (Accuracy of
HHD in planning the operating for primary varicose veins. Eur J Vasc Endovasc Surg
2002). After examining these patients and ensuring their suitability to the study they
were included.
The patients were given the study information and after getting informed consent from
all of them, I put 36 named tags to identical envelopes, which were sealed. After shuffling
(Review)
Rautio 2002 (Continued)
the envelopes I numbered them randomly. List of numbers for randomization was done
earlier according to instructions of the biostatistician of our department. I opened the
envelopes in numerical order.
Randomisation was done this way, because our strict schedule. Resource allocations
(operating theatres, angiography suites etc.) forced us to perform the operations and
procedures during a period of two weeks.
We also thought, that it was better to inform the result of randomisation to patients in
good time beforehand. Four patients withdrew because of the disappointment of having
been assigned to the stripping group. Three patients discontinued the study because of
an unsuitable schedule. One patient was excluded because of pregnancy. As a result we
missed three patients from the RFA group and five patients from the stripping group.
I do not see any chance of bias because of selection process itself. Withdrawal of eight
patients might have had some influence to the results.
Risk of bias
Random sequence generation (selection Low risk After contacting the author, the sequence
bias) generation details were clarified:
36 named tags to identical envelopes,
which were sealed. After shuffling the en-
velopes they were numbered randomly. List
of numbers for randomization was done
earlier according to instructions of the bio-
statistician of our department. The en-
velopes were opened in numerical order
Allocation concealment (selection bias) Low risk as above
Blinding (performance bias and detection High risk Impossible to blind operator or patients to
bias) treatment. No mention that assessors post-
All outcomes operatively were blinded
Incomplete outcome data (attrition bias) Low risk Full details on excluded patient:
Outcomes 33 patients (33 limbs) were randomised, 4
refused when they were allocated to surgery,
1 patient was excluded from the RFA group
due to pregnancy (they were not addressed
in the outcome analysis)
Final groups included 15 surgery and 13
RFA.
Incomplete outcome data (attrition bias) Low risk In the later publication (three year outcome
All outcomes measures), the authors claim all patients
also underwent 3 year follow up. They re-
port no long term losses to follow up
(Review)
Rautio 2002 (Continued)
way
Additional outcome measures were re-
ported in a subsequent publication (three
year results), recurrence rates, an additional
outcome which was not a pre-specified out-
come measure. However this does not in-
troduce any bias or inaccuracy into the trial
Other bias Low risk States competition of interest: nil

Small study groups.
Subramonia 2010

Method of randomisation: web-based randomisation method. Two stratification vari-
ables (age and sex) were controlled
Concealment of allocation: not clear
Blinded: not clear
No of patients randomised: 93
No of patients analysed: 76 patients (88 limbs)
Exclusions post randomisation: 2 RFA patients (1 taken off waiting list, 1 did not
receive any treatment) 4 surgery (1 taken off waiting list, 1 developed artrial fibrillation,
1 developed hypertension, 1 operated on a non-trial list)
Losses to follow up: none at 6 weeks. 53 patients (61 limbs) available at 20 months
Intention-to-treat analysis: no cross-over occurred
Source of funding: VNUS Medical Technologies funded some of the Closure PLUS
radiofrequency ablation catheters used in the trial. They were not involved in the running
of the trial, data collection, interpretation or analyses
Participants Country: UK
Setting/Location: Hospital
Number: 76 patients (88 limbs)
Age: [median, (IQR)] RFA: 47 (38 - 58); surgery: 45 (37 - 53)
Sex: RFA: 13 men, 34 women; surgery: 14 men, 27 women.
Inclusion criteria: Age 18 - 70 years / primary or recurrent GSV reflux on duplex
imaging / duplex confirmed suitable for RFA / pt fit for GA / physical condition allowing
ambulation after surgery / pt give informed consent / pt and surgeon agree intervention
is required / availability for follow up
Exclusion criteria: Varicose veins without GSV incompetence on duplex / associated
small saphenous or deep venous incompetence / tortuous GSV unsuitable for RFA / GSV
diameter < 3 mm > 12 mm in supine position / GSV thrombus / patients with permanent
pacemaker or internal defibrillator / concomitant PVD (ABPI < 0.9) / pregnancy / unable
to complete QoL questionnaire due to poor English language skills
(Review)
Subramonia 2010 (Continued)
Interventions Treatment(s): Radiofrequency ablation

Control/Placebo: Conventional surgery
Duration: 1 and 5 week follow up
Outcomes Primary outcomes: time taken to return to normal household activities

Secondary outcomes: Intraoperative complications / duration of the procedure / post
op morbidity (pain, analgesic requirements, sensory abnormalities, wound problems,
phlebitis, skin burns, pigmentation) / time to return to driving and patient satisfaction
and QoL
Notes Age and sex were judged most likely to influence outcome in the two groups. Author
contacted for further details:
A web-based randomisation method was used (with assistance from the Institute of
Health and Society, Newcastle University, UK) with stratification to ensure appropriate
balance between the arms with respect to variables that might influence outcome in the
two groups and to minimise the risk of confounding. The method used two stratification
variables, age and sex, that were judged most likely to influence the outcome in the two
groups. Two levels of each stratification variable were employed:
Age - 50 years and > 50 years
Sex - male or female
Simple randomisation without stratification does not guarantee equivalence between the
two groups and several levels of stratification can make the randomisation system more
complicated and also result in some small strata. The same procedure was allocated to
those with bilateral varicose veins both of which were suitable for the trial with a minimum
period of 3 months between the procedures. Access to the website was protected by
password and the file server maintained by the University of Newcastle had high security
protocols. The researcher alone had knowledge of the password to access the website. No
problems were encountered either in accessing the website or in randomising patients
during the trial
Risk of bias
Random sequence generation (selection High risk A web-based randomisation method that
bias) used two stratification variables (age and
sex) was used
Age and sex were judged most likely to in-
fluence outcome in the two groups. Au-
thor contacted for further details
Allocation concealment (selection bias) Unclear risk No exact details on how exactly the web
based randomisation process worked
Blinding (performance bias and detection High risk Unable to blind surgeon or patient to treat-
bias) ment. No mention that assessors post-op-
All outcomes eratively were blinded
(Review)
Subramonia 2010 (Continued)
Incomplete outcome data (attrition bias) Low risk Missing outcome data fully reported and
Outcomes balanced in numbers across intervention
groups:
93 randomised with 5 exclusions prior to
treatment; 1 RFA (taken off waiting list)
and 4 surgery (1 taken off waiting list, 2
medically unsuitable for surgery, 1 oper-
ated on a non trial list). 47 patients under-
went RFA and 41 underwent surgery. All
patients were analysed at 5 weeks
Incomplete outcome data (attrition bias) Low risk All patients were followed up at 5 weeks.
All outcomes
way
Other bias High risk Article was written and designed by two
vascular surgeons who perform both pro-
cedures regularly and both authors declare
no personal conflict of interests in either
treatment
Included five patients with recurrent vari-
cose veins. No stratification of these pa-
tients in the results. This could introduce a
potential bias into results such as pain, time
to return to normal activities, QoL etc
The authors standardised their anaesthetic,
and inter-operator variability thus reducing
bias
Age and sex variables were controlled in the
randomisation process thus reducing po-
tential confounding
Included 12 patients with bilateral varicose
veins (randomised on one occasion to the
same treatment, but had their limbs treated
with a minimum of 6 weeks in between
treatments, thus treating each limb as a sep-
arate case)
ABPI: ankle brachial pressure index

CEAP C2 - 4, Ep, As, Pr: CEAP classification of venous disease class C2 to C4, primary superficial veins reflux
DNA: did not attend
DVT: deep vein thrombosis
GA: general anaesthetic
pt: patient
PVD: peripheral vascular disease
(Review)
USS: ultrasound scan
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Abela 2008 This paper compared foam sclerotherapy plus SFJ ligation to conventional surgery. Foam plus SFJ ligation
does not represent standard foam sclerotherapy
Almeida 2007 Based on a talk given to the 2007 VEITHsymposium, entitled Weve Got Plenty of Data to Show that
Endovenous Thermal Ablation is Superior to Open Surgery. Not a report of any original work by himself
but a report of existing RCTs
Alos 2006 Treatment of reticular or postoperative varices. Foam only.
Anon 2008 This study compares two laser techniques.
Belcaro 2000 Trial of sclerotherapy versus surgery, hwr surgery = ligation of SFJ only NOT stripping of GSV + multiple
stab avulsions
Blaise 2010 Foam only, does not include comparisons to patients undergoing surgery
BLARA Trial Bilateral veins. Dual therapy. No surgery.
Bountouroglou 2006 This paper compared foam sclerotherapy plus SFJ ligation to conventional surgery. Foam plus SFJ ligation
does not represent standard foam sclerotherapy
Bush 2008 EVLT only. No comparison made to alternative treatment.
Ceulen 2007 Foam only, no comparison made to alternative therapy.
Chant 1972 Not foam sclerotherapy.
Christenson 2010 Christenson 2010 included the treatment of 200 limbs, randomised to receive conventional surgery or
endovenous laser ablation
After contacting the author it was confirmed that 40 patients underwent bilateral varicose vein treatment.
It was also confirmed that patients limbs were randomised, not patients. In fact, eight patients underwent
surgery on one limb and laser on the other. All patients with bilateral varicose veins were treated on the same
day
This clearly biases any results regarding post-operative quality of life scores. The high proportion of bilaterally
treated patients also affects pain scores. Time to return to work is also published, but limbs cannot return to
work independently of one another and subsequently these results are not suitable for our Cochrane review
Trials randomising and analysing results according to number of limbs rather than number of patients as the
unit of analysis means that the standard error of the treatment effect is much smaller than it should be. These
results have a much tighter confidence interval. Subsequently the results of recurrence from this trial cannot
be included in the meta-analysis of our review
(Review)
(Continued)
De Medeiros 2006 Compared laser versus conventional surgery. However the patients undergoing laser also had a high tie of
SFJ. This does not constitute standard laser treatment and subsequently the trial was excluded
Demagny 2002 Sclerotherapy only: foam versus liquid.
Desmyttere 2005 Only evaluating EVLT, no comparison made to alternate therapies
Disselhoff 2008 Not including cryostripping.
Doran 1975 Fegans method; included liquid sclerotherapy.
Duffy 2005 Study only included patients without SFJ incompetence.
Einarsson 1993 Surgery versus liquid sclerotherapy - not foam.
Figueiredo 2009 Figueiredo 2009 randomised 60 patients to receive conventional surgery or foam sclerotherapy. They included
treatment of both small and great saphenous varicose veins in both the thigh and lower leg. They also treated
accessory varicose veins in the thigh and lower leg, and perforating veins of the thigh and lower leg. In their
original trial they did not include any stratification in the results of who underwent treatment of GSV varicose
veins alone
The author was contacted to enquire about any stratified results
The author provided a table which showed treatment success versus treated vein segment. Treated segments
included (as above):
great saphenous vein (thigh, great saphenous vein (lower leg), small saphenous vein, accessory vein (thigh),
accessory vein (lower leg), perforating vein (thigh), perforating vein (lower leg)
After further contact the author clarified success as: total occlusion and partial recanalisation without reflux
A total of 72 (surgery) and 74 (foam sclerotherapy) treatments were carried out. Yet only 60 patients were
randomised. It is not clear if patients underwent multiple treatments or bilateral treatments
The author presents results differentiating between the treatment of the great saphenous vein in the thigh
and the lower limb. It is unclear how or why this was so
Many aspects of this trials results remain unclear. It was the joint decision of CN, RE and GS that this trial be
excluded on the grounds that its results could not be adequately analysed within the inclusion and exclusion
criteria of this review
Gale 2009 This study compares RFA and EVLA. There is no comparison made to conventional surgery
Hamel-Desnos 2003 No comparison made to alternative therapy.
Hamel-Desnos 2005 Foam only and does not include comparisons to patients undergoing surgery
Hamel-Desnos 2008 No comparison made to alternative therapy.
Hamel-Desnos 2009 Sclerotherapy only.
Hayes 2008 EVLT only versus no treatment.
(Review)
(Continued)
Hinchliffe 2006 This trial specifically looked at recurrent varicose veins. Re-do groin surgery involved exposing 5cm of femoral
vein above and below the SFJ. This does not constitute conventional varicose vein surgery. Subsequently
this paper has been excluded
Hobbs 1968 Used liquid sclerotherapy, not foam.
Kabnick 2003 Only evaluating laser. No comparison made to other treatments
Kalteis 2008 This paper compared laser plus SFJ ligation with conventional surgery. This is not standard laser therapy,
hence the study was excluded
Kern 2005 Not treatment of GSV. No comparison made to other treatments - foam only
Kuznetsov 2005 Treatment of varicose veins complicated by trophic ulcers.
Lin 2007 This paper was written in Chinese. Despite complete translation we were not able to extrapolate any mean-
ingful data that would enhance this Cochrane review
Lugli 2009 Sclerotherapy only.
Lupton 2002 Not treatment of GSV.
Martimbeau 2003 Compares different forms of foam, not two different treatments of varicose veins
Maurins 2009 Only includes patients undergoing endovenous laser ablation.
McDaniel 1999 Study does not include patients with varicose veins of the great saphenous vein
Mekako 2006 This study includes varicose veins of other sites as well as the great saphenous veins
Mekako 2007 Only evaluating EVLT. No comparison to alternate treatments.
NCT00841178 This study includes varicose veins of other sites as well as the great saphenous veins
Neglen 1993 Compared surgery to liquid sclerotherapy, not foam sclerotherapy
Ogawa 2008 Eight of the 92 patients had treatment of short saphenous vein reflux. There was no stratification in the
results between small and great saphenous vein reflux
Ouvry 2008 Compares two different types of foam, no comparison to different treatment techniques
Rabe 2008 Sclerotherapy treatment only (liquid versus foam).
Rao 2005 Sclerotherapy treatment only (foam versus liquid).
(Review)
(Continued)
REACTIV Excluded for two reasons. 1) intervention 2 (surgery versus sclerotherapy) included both saphenopopliteal
junction (SPJ) and SFJ incompetence, with no stratification in the analysis of the numbers of each group. 2)
sclerotherapy was not foam sclerotherapy
RECOVERY Trial No comparison made to conventional surgery. RFA versus laser only
Rutgers 1994 Liquid not foam sclerotherapy.
Rybak 2003 Foam treatment only.
Sadoun 2003 Foam treatment only.
Seddon 1973 Quasi randomised. Also, used liquid, not foam, sclerotherapy
Selles 2008 Comparing two different foam techniques only.
Sica 2006 Foam only, no comparison to alternative treatments.
Stotter 2005 Cryostripping. Not included.
Theivacumar 2008 Only compares EVLT (two different techniques). No comparison to alternative treatments
Theivacumar 2009 This paper presents the two year results of both neovascularisation and recurrence for the Darwood 2008
RCT series. However they also include all other patients who were treated (non-randomly) at the same time of
the original RCT. There is no stratification in these results of the randomised and non-randomised patients.
These results could therefore not be utilised and the study was excluded
VEDICO Trial This study compared six treatment options. 1) standard sclerotherapy; 2) high dose sclerotherapy; 3) multiple
vein ligation; 4) stab avulsion; 5) foam-sclerotherapy; 6) surgery (ligation with sclerotherapy)
Therefore the study does not include a true conventional surgery group, i.e. SFJ ligation + stripping group
NB: a stripping group was included but the authors declared that this was a non-randomised reference group
only
Viarengo 2007 EVLT only, no comparison made to alternative treatments.
Vuylsteke 2006 This is a controlled trial, no randomisation was used.
Wright 2006 In this study patients were divided into two groups initially by physicians choice based on the extent of their
disease into a surgery or sclerotherapy group. These two groups were then randomised. This introduces a
clear bias from the outset. In the varisolve versus surgery group, surgery was not standardised. It included
stripping in only 88.3% of cases. Their inclusion criteria for surgery included GSV and SSV and this was
not stratified in the results
Yamaki 2008 Foam only: no other treatment compared.
Zeh 2003 Foam only: no other treatment compared.
(Review)
Characteristics of ongoing studies [ordered by study ID]
Carradice 2008
Trial name or title A randomised trial of EVLT versus surgery for varicose veins. HELP-1
Methods RCT
Participants 267 limbs
Interventions Conventional surgery versus EVLT
Outcomes VCSS, AVVQ, SF-36, Euro QoL, post-operative pain, return to work/normal activities
Starting date Not known
Contact information dan1@doctors.org.uk
Notes Due for publication late 2010

An abstract of the interim results for this upcoming trial was published in Vascular Society of Great Britain &
Ireland Yearbook, 2009 . CN contacted the author who confirmed publication of the main trial is expected
soon
CLASS
Trial name or title CLASS (Comparison of LAser, Surgery and foam Sclerotherapy)
Methods Randomised controlled trial comparing foam sclerotherapy, alone or in combination with endovenous laser
therapy, with conventional surgery as a treatment for varicose veins
Participants Aim: 1016
Interventions Conventional surgery for the treatment of varicose veins compared with foam sclerotherapy alone of main
great or small saphenous trunk and non-trunk varicosities and endovenous laser ablation (EVLA) of main
trunk including foam sclerotherapy of non-trunk varicosities performed under local anaesthetic
Outcomes The primary patient outcomes are disease specific (Aberdeen Varicose Vein Questionnaire) and generic quality
of life (EQ-5D, SF-36) at 6 months. The primary economic outcome is the incremental cost per quality
adjusted life years (QALY) at 6 months. The secondary outcomes include (a) costs to the health service and
patients and any subsequent care at 6 months; (b) technical success of venous intervention at 6 weeks and 6
months; (c) clinical success of venous intervention at 6 weeks and 6 months; (d) disease specific and generic
quality of life at 6 weeks; and (e) behavioural recovery at 6 weeks
Starting date
Contact information Dr Seonaidh Cotton
University of Aberdeen
Centre for Healthcare Randomised Trials (CHaRT)
Health Services Research Unit
(Review)
CLASS (Continued)
3rd Floor, Health Sciences Building

Polwarth Building
Foresterhill
Aberdeen
Aberdeenshire
AB25 2ZD
UNITED KINGDOM
Tel: 01224 551126

s.c.cotton@abdn.ac.uk
Notes
Magna 2007
Trial name or title Surgery or Noninvasive Therapy for Varicose Veins (Magna)
Methods RCT
Participants Estimated enrolment: 240 patients
Interventions Procedure 1: cross-sectomy and short stripping

Procedure 2: ultrasound guided sclerocompression therapy
Procedure 3: endovenous laser therapy
Outcomes Primary Outcome Measures:

anatomical success rate (absence or obliteration of GSV on US examination) [Time Frame: 3 months, 1 and
5 years]
Secondary Outcome Measures:

treatment related adverse events and complications [Time Frame: 3 months, 1 and 5 years]
patient reported outcomes (HRQOL and treatment satisfaction) [Time Frame: 3 months, 1 and 5 years]
cost effectiveness analysis [Time Frame: 3 months, 1 and 5 years]
Starting date Study start date: May 2007
Contact information Michael Kockaert, MD email: m.kockaert@erasmusmc.nl

Tamar Nijsten, MD, PhD email: t.nijsten@erasmusmc.nl
Notes This study is still recruiting
(Review)
RAFPELS Trial
Trial name or title Prospective Randomised Trial Comparing the New Endovenous Procedures Versus Conventional Surgery for
Varicose Veins Due to Great Saphenous Vein Incompetence (RAFPELS)
Methods RCT
Participants Estimated enrolment: 1000
Interventions Procedure: High ligation of the great saphenous vein + stripping

Procedure: Endovenous laser ablation
Procedure: Radiofrequency ablation
Procedure: Foam sclerotherapy
Outcomes Primary Outcome Measures:

Recurrence rate and complications after venous surgery. Duplex ultrasound and clinical evaluation (Time
Frame: 3 years).
Secondary Outcome Measures:

Quality of life with SF36, Aberdeen Vein Score, VCSS and Visual Analogue Scale Scoring (Time Frame: 3
years)
Starting date Study start date: January 2008
Contact information Anders Hellberg, MD, PhD Tel +4621173000, email anders.hellberg@ltv.se
Achilleas Karkamanis, MD Tel +4621173000, email achilleas.karkamanis@ltv.se
Notes Estimated study completion date: August 2011

Estimated primary completion date: January 2011 (Final data collection date for primary outcome measure)
Rasmussen 2010a
Trial name or title Randomised clinical trial comparing radiofrequency ablation, endovenous laser ablation, ultrasound guided
foam sclerotherapy versus conventional surgery for the treatment for great saphenous varicose veins
Methods RCT
Participants 500 patients (575 limbs)
Interventions Radiofrequency ablation, endovenous laser ablation, ultrasound guided foam sclerotherapy versus conven-
tional surgery
Outcomes Failure rates at 1 year. 5.7% (radiofrequency ablation), 5.4% (endovenous laser ablation), 15% (ultrasound
guided foam sclerotherapy), 4% (conventional surgery)
Starting date Not known
Contact information larshrasmussen@yahoo.com
(Review)
Rasmussen 2010a (Continued)
Notes Preliminary results of this trial were presented to the Charing Cross Symposium 2010 and published in
Venous News, May 2010
(Review)
DATA AND ANALYSES
Comparison 1. Laser ablation versus surgery
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Recanalisation - (number of 2 447 Odds Ratio (M-H, Fixed, 95% CI) 3.30 [0.85, 12.74]
limbs)
1.1 Early (< 4 months) 2 249 Odds Ratio (M-H, Fixed, 95% CI) 3.83 [0.45, 32.64]
1.2 Late (> 4 months) 2 198 Odds Ratio (M-H, Fixed, 95% CI) 2.97 [0.52, 16.98]
2 Technical failure - (number of 2 249 Odds Ratio (M-H, Fixed, 95% CI) 0.12 [0.02, 0.75]
limbs)
Comparison 2. Radiofrequency ablation versus surgery
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Recurrence 2 93 Odds Ratio (M-H, Fixed, 95% CI) 1.04 [0.37, 2.91]
2 Recanalisation 3 286 Odds Ratio (M-H, Fixed, 95% CI) 1.79 [0.48, 6.64]
2.1 Early (< 4 months) 3 193 Odds Ratio (M-H, Fixed, 95% CI) 7.86 [0.41, 151.28]
2.2 Late ( > 4months) 2 93 Odds Ratio (M-H, Fixed, 95% CI) 0.79 [0.15, 4.23]
3 Neovascularisation 2 93 Odds Ratio (M-H, Fixed, 95% CI) 0.39 [0.09, 1.63]
4 Technical failure 3 202 Odds Ratio (M-H, Random, 95% CI) 0.48 [0.01, 34.25]
(Review)
Analysis 1.1. Comparison 1 Laser ablation versus surgery, Outcome 1 Recanalisation - (number of limbs).
Review: Endovenous ablation (radiofrequency and laser) and foam sclerotherapy versus conventional surgery for great saphenous vein varices
Comparison: 1 Laser ablation versus surgery
Outcome: 1 Recanalisation - (number of limbs)
Study or subgroup Laser Surgery Odds Ratio Weight Odds Ratio

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
1 Early (< 4 months)

Darwood 2008 3/80 0/32 22.3 % 2.94 [ 0.15, 58.45 ]
Rasmussen 2007 2/69 0/68 15.9 % 5.07 [ 0.24, 107.67 ]
Subtotal (95% CI) 149 100 38.2 % 3.83 [ 0.45, 32.64 ]

Total events: 5 (Laser), 0 (Surgery)
Heterogeneity: Chi2 = 0.06, df = 1 (P = 0.80); I2 =0.0%
Test for overall effect: Z = 1.23 (P = 0.22)
2 Late (> 4 months)
Darwood 2008 6/49 1/12 46.2 % 1.53 [ 0.17, 14.11 ]
Rasmussen 2007 3/69 0/68 15.7 % 7.21 [ 0.37, 142.29 ]
Subtotal (95% CI) 118 80 61.8 % 2.97 [ 0.52, 16.98 ]

Total (95% CI) 267 180 100.0 % 3.30 [ 0.85, 12.74 ]
Test for subgroup differences: Chi2 = 0.03, df = 1 (P = 0.86), I2 =0.0%
0.02 0.1 1 10 50
Favours Laser Favours Surgery
(Review)
Analysis 1.2. Comparison 1 Laser ablation versus surgery, Outcome 2 Technical failure - (number of limbs).
Comparison: 1 Laser ablation versus surgery
Outcome: 2 Technical failure - (number of limbs)
Study or subgroup Laser Surgery Odds Ratio Weight Odds Ratio

Darwood 2008 1/80 4/32 69.3 % 0.09 [ 0.01, 0.83 ]
Rasmussen 2007 0/69 2/68 30.7 % 0.19 [ 0.01, 4.06 ]
Total (95% CI) 149 100 100.0 % 0.12 [ 0.02, 0.75 ]

Test for subgroup differences: Not applicable
0.01 0.1 1 10 100

Favours Laser Favours Surgery
Analysis 2.1. Comparison 2 Radiofrequency ablation versus surgery, Outcome 1 Recurrence.
Comparison: 2 Radiofrequency ablation versus surgery
Outcome: 1 Recurrence
Study or subgroup RFA Surgery Odds Ratio Weight Odds Ratio

EVOLVeS Study 5/36 6/29 80.0 % 0.62 [ 0.17, 2.28 ]
Rautio 2002 5/15 2/13 20.0 % 2.75 [ 0.43, 17.49 ]
Total (95% CI) 51 42 100.0 % 1.04 [ 0.37, 2.91 ]

Total events: 10 (RFA), 8 (Surgery)
Heterogeneity: Chi2 = 1.67, df = 1 (P = 0.20); I2 =40%
0.01 0.1 1 10 100

Favours RFA Favours Surgery
(Review)
Analysis 2.2. Comparison 2 Radiofrequency ablation versus surgery, Outcome 2 Recanalisation.
Outcome: 2 Recanalisation
Study or subgroup RFA Surgery Odds Ratio Odds Ratio

1 Early (< 4 months)

EVOLVeS Study 4/43 0/34 7.86 [ 0.41, 151.28 ]
Rautio 2002 0/15 0/13 0.0 [ 0.0, 0.0 ]
Subramonia 2010 0/47 0/41 0.0 [ 0.0, 0.0 ]
Subtotal (95% CI) 105 88 7.86 [ 0.41, 151.28 ]

2 Late ( > 4months)
EVOLVeS Study 3/36 3/29 0.79 [ 0.15, 4.23 ]
Rautio 2002 0/15 0/13 0.0 [ 0.0, 0.0 ]
Subtotal (95% CI) 51 42 0.79 [ 0.15, 4.23 ]

Total (95% CI) 156 130 1.79 [ 0.48, 6.64 ]
Heterogeneity: Chi2 = 1.87, df = 1 (P = 0.17); I2 =47%
Test for subgroup differences: Chi2 = 1.76, df = 1 (P = 0.19), I2 =43%
0.01 0.1 1 10 100

(Review)
Analysis 2.3. Comparison 2 Radiofrequency ablation versus surgery, Outcome 3 Neovascularisation.
Outcome: 3 Neovascularisation
Study or subgroup RFA Surgery Odds Ratio Weight Odds Ratio

EVOLVeS Study 1/29 4/36 52.3 % 0.29 [ 0.03, 2.71 ]
Rautio 2002 2/13 4/15 47.7 % 0.50 [ 0.08, 3.32 ]
Total (95% CI) 42 51 100.0 % 0.39 [ 0.09, 1.63 ]

0.01 0.1 1 10 100

Analysis 2.4. Comparison 2 Radiofrequency ablation versus surgery, Outcome 4 Technical failure.
Outcome: 4 Technical failure
Study or subgroup RFA Surgery Odds Ratio Odds Ratio

M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
EVOLVeS Study 2/44 0/36 4.29 [ 0.20, 92.36 ]
Rautio 2002 0/15 0/13 0.0 [ 0.0, 0.0 ]
Subramonia 2010 0/47 7/47 0.06 [ 0.00, 1.03 ]
Total (95% CI) 106 96 0.48 [ 0.01, 34.25 ]

Heterogeneity: Tau2 = 7.18; Chi2 = 4.10, df = 1 (P = 0.04); I2 =76%
0.01 0.1 1 10 100

(Review)
ADDITIONAL TABLES
Table 1. Study Sample Sizes
Technique Study Patients Randomised Patients Analysed
Overall Overall Surgery Alternative Rx
Foam vs surgery no data
Laser vs surgery Rasmussen 121 121 59 62
Darwood 118 95 30 65
Total 239 216 89 127
RFA vs surgery Subramonia 93 88 41 47
EVOLVeS 85 79 36 43
Rautio 33 28 13 15
Total 211 195 90 105

Rx = treatment
Table 2. Age and Sex of Participants
Technique Study Age (units specified) Sex (F:M)
Surgery Alternative Rx Surgery Alternative Rx
Foam vs surgery No data
Laser vs surgery Darwood 49 (38.5 - 57.5) EVLT1 42 (30.5 - 54. 16:14 EVLT1 22:16
median (IQR) 5) EVLT2 16:11
EVLT2 52 (35 - 59)
median (IQR)
Rasmussen 54 (22 - 78) 53 (26 - 79) 43:16 41:21

mean (range) mean (range)
RFA vs surgery Subramonia 45 (37 - 53) 47 (38 - 58) 27:14 34:13

median (IQR) median (IQR)
(Review)
Table 2. Age and Sex of Participants (Continued)
EVOLVeS 47 (4) 49 (4) 26:10 32:12

mean (SD) mean (SD)
Rautio 38 (6.8) 33 (6.7) 12:1 14:1

mean (SD) mean (SD)
IQR = inter quartile range
Rx = treatment
SD = standard deviation
Table 3. Laser technique used
Study Laser Pulse/Continuous Energy Technique
Darwood 810nm diode 1) Pulsed 12W 1 sec pulses, 1 sec intervals
2) Continuous 14W withdrawn 2 - 3 mm/sec
Rasmussen 980nm diode Pulsed 12W 1.5 sec pulses, 1.5 sec intervals
Darwood: Laser 1 = pulsed. Laser 2 = continuous
Table 4. Additional Phlebectomies
Technique Study Additional phlebectomies
Surgery Alternative Rx
Laser vs surgery Darwood yes no
Rasmussen yes yes
RFA vs surgery Subramonia yes yes
EVOLVeS yes yes
Rautio yes yes

Rx = treatment
(Review)
Table 5. Outcome Measures
Tech- Study Outcome Measure

nique
Techni- Time Pain QoL/ Compli- Cost Neo Recur- Satisfac- Duration
cal fail- to rtn to Severity cations vasculari- rence tion/ of
ure work/ score sation cosmesis proce-
N activ- dure
ities/
driving
Foam vs No data
surgery

Laser vs Dar-
surgery wood

Ras-
mussen

RFA vs Subra-
surgery monia

EVOLVeS

Rautio
N = number
rtn = return
Table 6. Recurrence and Neovascularisation
Technique Study Neovascularisation (%) Recurrence noted Recurrence noted

by clinician at 2 yrs (%) by patients at 2 yrs (%)
Surgery Alternative Surgery Alternative Surgery Alternative

Rx Rx Rx
Foam vs No data
surgery
Laser vs Darwood** 1/12 (8.3)* 0/49 (0) no results no results

surgery
Rasmussen** no results 25/68 (37) 18/69 (26) 9/68 (13) 8/69 (12)
RFA vs Subramonia no results no results no results

surgery
(Review)
Table 6. Recurrence and Neovascularisation (Continued)
EVOLVeS** 4/29 (13.8) 1/36 (2.8) 6/29 (20.9) 5/36 (14.3) no data
Rautio 1/13 (8) 1/15 (7) 2/13 (15.4) 5/15 (33.3) 2/13 (15.4) 4/15 (26.7)
* Only 12 limbs available for follow-up at 1 year
** Results only available per limb, not per patient.
Rx = treatment
Table 7. Re-intervention rates
Technique Study Re-intervention (%)
Laser vs surgery Rasmussen* (within 2 yrs) 6/68 (8.8) 9/69 (13)
Darwood No data No data
RFA vs surgery Subramonia No data No data
EVOLVeS No data No data
Rautio (within 3 yrs) 2/13 (15.4) 2/15 (13.3)

* results only available per limb, not per patient
Rx = treatment
Table 8. Quality of Life and Venous Severity Scores
Tech- Study Quality of Life Severity Score

nique
V-Q/ AVVSS CIVIQ2 SF-36 RAND- VCSS TCSS CEAP VDS VSDS
SymQ 36
Foam vs No data
surgery

Laser vs Dar-
surgery wood

Ras-
mussen
(Review)
Table 8. Quality of Life and Venous Severity Scores (Continued)

RFA vs Subra-
surgery monia

EVOLVeS

Rautio
AVVSS: Aberdeen Varicose Vein Symptom Severity Score
CEAP: Clinical severity, Etiology, Anatomy, Pathophysiology
CIVIQ2: ChronIc Venous Insufficiency Quality of Life Questionnaire
RAND-36: Short term RAND-36 (validated for Finland)
SF-36: Medical Outcomes Study Short Form 36
TCSS: Total Clinical Severity Score
VCSS: Venous Clinical Severity Score
VDS: Venous Disability Score
V-Q/SymQ: VEINES-QoL/Sym questionnaire
VSDS: Venous Segmental Disability Score
Table 9. Post operative complications
Early (within 3 months)
Minor (not requiring intervention) (%) Major (requiring inter-

Study vention) (%)
Adverse Saphe- Thermal in- Wound Bruis- Phlebitis Wound Other

event Haematoma nous nerve jury/in- problems ing and pig- Problems
(wound or injury flammation (groin/ mentation
thigh) stab)
Technique Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt
Rx Rx Rx Rx Rx Rx Rx Rx
Laser vs surgery
Darwood 0/32 0/80 4/32 1/80 0/32 0/80 1/32 1/80 2/32 1/80 0/32 9/80 2/32 0/80 1/32 0/80
***** (0) (0) (13) (1) (0) (0) (3) (1) (6) (1) (0) (11) (6) (0) (3)** (0)
Ras- 5/86 3/69 1/68 1/69 0/68 0/69 1/68 0/69 15/ 7/69 2/68 2/69 1/68 0/69 0/68 0/69
mussen***** (8) (5)* (2) (2) (0) (0) (2) (0) 68 (11) (3) (3) (2) (0) (0) (0)
(25) ****
Over- 5/ 3/ 5/ 2/ 0/ 0/ 2/ 1/ 17/ 8/ 2/ 11/ 3/ 0/ 1/ 0/

all (Laser vs 102 149 102 149 102 149 102 149 102 149 102 149 102 149 102 149
surgery) (5) (2) (5) (1) (0) (0) (2) (1) (17) (5) (2) (7) (2) (0) (1) (0)
RFA vs surgery
(Review)
Table 9. Post operative complications (Continued)
EVOLVeS 20/ 9/44 5/36 10/ 3/36 6/44 16/ 8/44 23/ 15/ 3/36 6/44 2/36 0/44 0/36 0/44
36 (20) (14) 44 (8) (14) 36 (18) 36 44 (8) (14) (6) (0) (0) (0)
(56) (23) (44) (64) (34) ***
Rautio 4/13 1/15 3/13 2/15 0/13 2/15 0/13 0/15 0/13 0/15 0/13 3/15 0/13 0/15 0/13 0/15
(31) (7) (23) (13) (0) (13) (0) (0) (0) (0) (0) (20) (0) (0) (0) (0)
Subramonia 1/41 0/47 20/ 9/47 0/41 0/47 6/41 0/47 0/41 5/47 0/41 0/47 0/41 0/47 0/41 0/47
(2) (0) 41 (19) (0) (0) (15) (0) (0) (11) (0) (0) (0) (0) (0) (0)
(49)
Over- 25/ 10/ 28/ 21/ 3/90 8/ 22/ 8/ 23/ 20/ 3/90 9/ 2/90 0/ 0/90 0/
all (RFA vs 90 106 90 106 (3) 106 90 106 90 106 (3) 106 (2) 106 (0) 106
surgery) (28) (9) (31) (20) (8) (24) (8) (26) (19) (8) (0) (0)
Over- 30/ 13/ 33/ 23/ 3/ 8/ 24/ 9/ 40/ 28/ 5/ 20/ 5/ 0/ 1/ 0/

all total (< 3 192 255 192 255 192 255 192 255 192 255 192 255 192 255 192 255
months) (16) (6) (17) (9) (2) (3) (13) (4) (21) (11) (3) (8) (3) (0) (1) (0)
Late (beyond 3 months)
Adverse Saphe- Thermal in- Wound Bruis- Phlebitis Wound Other

event Haematoma nous nerve jury/ problems ing and pig- Problems
(wound or injury inflamma- (groin/ mentation
thigh) tion stab)
Technique Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt Surg Alt
Rx Rx Rx Rx Rx Rx Rx Rx
Laser vs surgery
Dar- 0/34 0/80 1/34 0/80 0/34 0/80 0/34 0/80 0/34 0/80 0/34 0/80 0/34 0/80 0/34 0/80
wood***** (0) (0) (3) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0)
Rasmussen 0/68 0/96 1/68 0/96 0/68 0/96 0/68 0/96 0/68 0/96 0/68 0/96 0/68 0/96 0/68 0/96
***** (0) (0) (2) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0)
Over- 0/ 0/ 2/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/ 0/
all (Laser vs 102 149 102 149 102 149 102 149 102 149 102 149 102 149 102 149
surgery) (0) (0) (2) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0)
RFA vs surgery
EVOLVeS 3/34 0/43 0/34 0/43 0/34 0/43 0/34 0/43 1/34 0/43 2/34 0/43 0/34 0/43 0/34 0/43
(9) (0) (0) (0) (0) (0) (0) (0) (3) (0) (6) (0) (0) (0) (0) (0)
Rautio 0/13 0/15 5/13 1/15 0/13 0/15 0/13 0/15 0/13 0/15 0/13 0/15 0/13 0/15 0/13 0/15
(0) (0) (38) (3) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0) (0)
(Review)
Table 9. Post operative complications (Continued)
Subramonia No follow-up beyond 5 weeks
Over- 3/47 0/58 5/47 1/58 0/47 0/58 0/47 0/58 1/47 0/58 2/47 0/58 0/47 0/58 0/47 0/58
all (RFA vs (6) (0) (11) (2) (0) (0) (0) (0) (2) (0) (4) (0) (0) (0) (0) (0)
surgery)
Over- 3/ 0/ 7/ 1/ 0/ 0/ 0/ 0/ 1/ 0/ 2/ 0/ 0/ 0/ 0/ 0/
all total (> 3 149 207 149 207 149 207 149 207 149 207 149 207 149 207 149 207
months) (2) (0) (5) (0.5) (0) (0) (0) (0) (1) (0) (1) (0) (0) (0) (0) (0)
* in one patient the saphenous thrombus extended into the femoral vein, it resolved without intervention
** post operative acute respiratory distress syndrome (requiring seven days ITU support) following aspiration post operatively
*** includes one patient that required debridement and IV antibiotics for a thigh and calf infection
**** groin infection requiring antibiotics
***** results only available per limb, not per patient
Surg = surgery
Alt Rx = alternative treatment
Table 10. Type of anaesthetic used
Technique Study General anaesthesia Tumescent anaesthesia Local anaesthesia
Surgery Alternative Surgery Alternative Surgery Alternative

Rx Rx Rx
Foam vs No data
surgery
Laser vs Darwood 30/30 0/65 0/30 65/65 30/30 65/65

surgery
Rasmussen 0/59 0/62 59/59 62/62 59/59 62/62
RFA vs EVOLVeS 19/36 12/44 17/36 32/44 17/36 32/44

surgery
Rautio 13/13 15/15 0/13 0/15 0/13 0/15
Subramonia 41/41 47/47 0/41 47/47 41/41 0/47

Rx = treatment
Table 11. Length of procedure/Operative time
Technique Study Time mins
(Review)
Table 11. Length of procedure/Operative time (Continued)
Laser vs surgery No data
RFA vs surgery Rautio Operating time: 57 11 Operating time: 75 16.6

Operating room time: 99 12.9 Operating room time: 115 18.3
Recovery room time: 198 40.7 Recovery room time: 227 57.6
(mean SD) (mean SD)
EVOLVeS Overall: 89 12 Overall: 74 10

(mean SD) (mean SD)
Subramonia Theatre time: 55 (48 - 63) Theatre time: 82 (73 - 91)

Procedure time: 48 (39 - 54) Procedure time: 76 (67 - 84)
(median (range)) (median (range))
mins = minutes
Rx = treatment
Table 12. Duration of hospital stay
Technique Study Length of hospital stay

% day case
Laser vs surgery Darwood 100 100
Rasmussen 100 100
RFA vs surgery Subramonia No data on length of hospital stay
EVOLVeS 86* 95**
Rautio No data on length of hospital stay

* 5 patients kept overnight
** 2 patients kept overnight
Rx = treatment
(Review)
Table 13. Operative costs
Technique Study Cost () Apparent saving of alt Rx
Laser vs surgery Rasmussen 2,066.14 2,275.06 -208.92
RFA vs surgery Rautio 1,339.73 974.54 365.19

alt Rx = alternative treatment
Rx = treatment
Table 14. Time to Return to Work/Normal activities
Technique Study Time to return to work (days) Time to Return Normal Activities
Surgery Alternative Rx Surgery Alternative Rx
Laser vs surgery Darwood* 17 (7.25 - 33.25) laser 1: 4 (2.5 - 7) 7 (2 - 26) laser 1: 2 (0 - 7)

median (IQR) laser 2: 4 (1 - 12) median (IQR) laser 2: 2 (0 - 7)
median (IQR) median (IQR)
Rasmussen 7.6 4.9 (1 - 28) 7 6 (1 - 31) 7.7 6.1 (0 - 29) 6.9 7 (0 - 29)
mean SD (range) mean SD (range) mean SD (range) mean SD (range)
RFA vs surgery Rautio** actual: 15.6 6 actual: 6.5 3.3 no data no data
perceived: 19.2 10 perceived: 6.1 4.4
mean SD mean SD
EVOLVeS*** 12.4 mean 4.7 mean 3.89 mean 1.15 mean
Subramonia 18.5 (11 - 28) 10 (4 - 13) 12.5 (4 - 21) 3 (0 - 7)

median (IQR) median (IQR) median (IQR) median (IQR)
*presented both laser techniques separately
**sick leave days taken and patients own perception of required sick leave
***adjusted according to the number of phlebectomies performed, and the type of anaesthetic used
Rx = treatment
(Review)
Table 15. EVOLVeS Study: Effect of pain on daily activities
Time post-procedure Differences in pain score from pre-treatment

score
mean (standard error)
Surgery RFA P value
72 hrs 2.9 (0.7) -1.77 (0.6) <0.0001
1 week 1.2 (0.7 -2.4 (0.6) <0.0001
Table 16. Subramonia 2010: Severity grading of pain in the first week after intervention
Surgery (No. of patients) RFA (No. of patients)
None 0 7
Very Mild 6 19
Mild 9 10
Moderate 17 9
Severe 9 2
Very Severe 0 0
Number of patients who reported pain
APPENDICES
Appendix 1. CENTRAL search strategy
#1 MeSH descriptor Sclerotherapy explode all trees 414
#2 MeSH descriptor Sclerosing Solutions explode all trees 334
#3 (sclero*):ti,ab,kw 4916
#4 (tetradecyl near2 (sulfate or sulphate)):ti,ab,kw 52
(Review)
(Continued)
#5 MeSH descriptor Sodium Tetradecyl Sulfate explode all trees 34
#6 MeSH descriptor Saline Solution, Hypertonic explode all trees 294
#7 MeSH descriptor Ethanolamines, this term only 1173
#8 (polydocanol or polidocanol):ti,ab,kw 141
#9 (saline):ti,ab,kw 12121
#10 (ethanolamine near2 oleate):ti,ab,kw 53
#11 (sodium near2 morrhuate):ti,ab,kw 16
#12 (sotradecol):ti,ab,kw 6
#13 (aetoxisclerol or aethoxysclerol):ti,ab,kw 14
#14 (aetoxiskerol or aethoxyskerol):ti,ab,kw 1
#15 (Turbofoam):ti,ab,kw 1
#16 (foam* or microfoam*):ti,ab,kw 783
#17 (varisolve):ti,ab,kw 2
#18 MeSH descriptor Laser Therapy explode all trees 2474
#19 (endovenous or EVLA or EVLT or radiofrequency or laser* or 8766

ablation* or obliteration*):ti,ab,kw
#20 MeSH descriptor Catheter Ablation explode all trees 840
#21 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR 27368

#9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR
#16 OR #17 OR #18 OR #19 OR #20)
#22 MeSH descriptor Vascular Surgical Procedures explode all trees 5487
#23 MeSH descriptor Ligation explode all trees 454
#24 (surg* or ligat* or strip* or phlebectomy):ti,ab,kw 83770
#25 (#22 OR #23 OR #24) 86629
#26 (#21 AND #25) 8155
(Review)
(Continued)
#27 MeSH descriptor Varicose Veins explode all trees 706
#28 MeSH descriptor Venous Insufficiency explode all trees 303
#29 MeSH descriptor Saphenous Vein explode all trees with qual- 139
ifier: SU
#30 (varic* or incomp* or insuffici* or tortuous or sapheno* or 13735

GSV):ti,ab,kw
#31 (#27 OR #28 OR #29 OR #30) 13755
#32 (#26 AND #31) 500
Appendix 2. Authors EMBASE search strategy

Ovid EMBASE 1947 - 2010. Last searched July 2010.
Strategy of EMBASE searches are shown below, number of articles for each search is shown in brackets:
1) varicose AND vein .ti,ab (2980)

2) exp VEIN DISEASE (75685)
3) 1 OR 2 (75992)
4) long AND saphenous AND vein .ti,ab (1866)
5) exp SAPHENOUS VEIN/ OR exp LEG VARICOSIS/ (5531)
6) 4 OR 5 (6535)
7) sclerotherapy .ti,ab (4249)
8) exp SCLEROTHERAPY (5939)
9) 7 OR 8 (6757)
10) 3 AND 6 AND 9 (364)
11) exp RADIOFREQUENCY ABLATION/ OR exp CATHETER ABLATION (13209)
12) radiofrequency AND ablation .ti,ab (7073)
13) 11 OR 12 (14168)
14) 3 AND 6 AND 13 (125)
15) laser.ti,ab (84909)
16) exp LASER SURGERY (28042)
17) exp LASER/ OR exp DIODE LASER/ OR exp LASER SURGERY (55957)
18) 15 OR 16 OR 17 (99370)
19) 3 AND 6 AND 18 (248)
20) 10 OR 14 OR 19 (584)
(Review)
HISTORY
Protocol first published: Issue 1, 2006
Review first published: Issue 10, 2011
CONTRIBUTIONS OF AUTHORS
CN, PC, HB and RE selected and assessed the methodological quality of trials and extracted data. CN and RE analysed the data and
were the main authors on the review. GS resolved any disagreements regarding methodological quality of trials. The final contents of
the review were agreed by GS and VB.
DECLARATIONS OF INTEREST
VB declares he is entitled to royalties because he is a co-editor of the text book entitled Postgraduate Vascular Surgery - Candidates
guide to the FRCS published by Cambridge University Press.
SOURCES OF SUPPORT
Internal sources
No sources of support supplied
External sources
Chief Scientist Office, Scottish Government Health Directorates, The Scottish Government, UK.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Additional outcome data were collected on post-operative pain and time to return to work and normal activities.
INDEX TERMS
Medical Subject Headings (MeSH)
Saphenous Vein; Catheter Ablation [ methods]; Laser Therapy [ methods]; Randomized Controlled Trials as Topic; Recurrence;
Sclerotherapy [ methods]; Varicose Veins [ therapy]
(Review)
MeSH check words
Humans
(Review)

Endovenous Ablation (Radiofrequency and Laser) and Foam Sclerotherapy Versus Conventional Surgery For Great Saphenous Vein Varices (Review)

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Endovenous Ablation (Radiofrequency and Laser) and Foam Sclerotherapy Versus Conventional Surgery For Great Saphenous Vein Varices (Review)

Încărcat de

Drepturi de autor:

Formate disponibile

Endovenous ablation (radiofrequency and laser) and foam

sclerotherapy versus conventional surgery for great

Nesbitt C, Eifell RKG, Coyne P, Badri H, Bhattacharya V, Stansby G

Endovenous ablation (radiofrequency and laser) and foam

Editorial group: Cochrane Peripheral Vascular Diseases Group.

PLAIN LANGUAGE SUMMARY

Description of the condition

Search methods for identification of studies

Searching other resources

Re-intervention or technical failure Post-operative complications

All studies declared and reported post-operative complications.

DISCUSSION (ii) Radiofrequency ablation versus surgery

Characteristics of included studies [ordered by study ID]

Methods Study design: Prospective RCT

Bias Authors judgement Support for judgement

Other bias High risk Authors reported difficulty recruiting pa-

One patient randomised to surgery under-

Methods Study design: Multi-centre prospective RCT

Participants Country: France, Austria and USA

Bias Authors judgement Support for judgement

Allocation concealment (selection bias) Unclear risk No details on adequate concealment.

44 limbs RFA, 36 limbs surgery. 2 patients

Methods Study design: RCT

Participants Country: Denmark

Bias Authors judgement Support for judgement

velopes were alike.

Incomplete outcome data (attrition bias) Low risk As above

Methods Study design: RCT

Participants Country: Finland

Notes Author contacted 2/2/2010.

Bias Authors judgement Support for judgement

Allocation concealment (selection bias) Low risk as above

Other bias Low risk States competition of interest: nil

Methods Study design: RCT

Interventions Treatment(s): Radiofrequency ablation

Outcomes Primary outcomes: time taken to return to normal household activities

Bias Authors judgement Support for judgement

ABPI: ankle brachial pressure index

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Alos 2006 Treatment of reticular or postoperative varices. Foam only.

Anon 2008 This study compares two laser techniques.

BLARA Trial Bilateral veins. Dual therapy. No surgery.

Bush 2008 EVLT only. No comparison made to alternative treatment.

Ceulen 2007 Foam only, no comparison made to alternative therapy.

Chant 1972 Not foam sclerotherapy.

Demagny 2002 Sclerotherapy only: foam versus liquid.

Desmyttere 2005 Only evaluating EVLT, no comparison made to alternate therapies

Disselhoff 2008 Not including cryostripping.

Doran 1975 Fegans method; included liquid sclerotherapy.

Duffy 2005 Study only included patients without SFJ incompetence.

Einarsson 1993 Surgery versus liquid sclerotherapy - not foam.

Hamel-Desnos 2003 No comparison made to alternative therapy.

Hamel-Desnos 2008 No comparison made to alternative therapy.

Hamel-Desnos 2009 Sclerotherapy only.

Hayes 2008 EVLT only versus no treatment.

Hobbs 1968 Used liquid sclerotherapy, not foam.

Kabnick 2003 Only evaluating laser. No comparison made to other treatments

Kuznetsov 2005 Treatment of varicose veins complicated by trophic ulcers.

Lugli 2009 Sclerotherapy only.

Lupton 2002 Not treatment of GSV.

Maurins 2009 Only includes patients undergoing endovenous laser ablation.

Mekako 2007 Only evaluating EVLT. No comparison to alternate treatments.