The n e w e ng l a n d j o u r na l of m e dic i n e

Original Article

A Randomized Trial Comparing Treatments
for Varicose Veins
Julie Brittenden, M.D., Seonaidh C. Cotton, Ph.D., Andrew Elders, M.Sc.,
Craig R. Ramsay, Ph.D., John Norrie, M.Sc., Jennifer Burr, M.D.,
Bruce Campbell, M.B., B.S., Paul Bachoo, M.B., Ch.B.,
Ian Chetter, M.B., Ch.B., M.D., Michael Gough, M.B., Ch.B.,
Jonothan Earnshaw, D.M., Tim Lees, M.B., Ch.B., M.D.,
Julian Scott, M.B., Ch.B., M.D., Sara A. Baker, M.Sc., Jill Francis, Ph.D.,
Emma Tassie, M.Sc., Graham Scotland, Ph.D., Samantha Wileman, Ph.D.,
and Marion K. Campbell, Ph.D.

A BS T R AC T

BACKGROUND
From the Division of Applied Medicine Ultrasound-guided foam sclerotherapy and endovenous laser ablation are widely used
(J. Brittenden), the Health Services Re- alternatives to surgery for the treatment of varicose veins, but their comparative
search Unit (S.C.C., A.E., C.R.R., J.N.,
G.S., S.W., M.K.C.), and the Health Eco- effectiveness and safety remain uncertain.
nomics Research Unit (E.T., G.S.), Uni-
versity of Aberdeen, and the Department METHODS
of Vascular Surgery, NHS Grampian, Ab-
erdeen Royal Infirmary (P.B.), Aberdeen;
In a randomized trial involving 798 participants with primary varicose veins at 11
the School of Medicine, Medical and Bio- centers in the United Kingdom, we compared the outcomes of foam, laser, and surgi-
logical Sciences, University of St. An- cal treatments. Primary outcomes at 6 months were disease-specific quality of life
drews, St. Andrews (J. Burr); the Depart-
ment of Vascular Surgery, Royal Devon
and generic quality of life, as measured on several scales. Secondary outcomes
and Exeter Hospital, Exeter (B.C.); the included complications and measures of clinical success.
Department of Vascular Surgery, Hull
Royal Infirmary, Hull (I.C.); the School of
RESULTS
Surgery, University of Leeds (M.G.), and
Vascular Surgery, St. James University After adjustment for baseline scores and other covariates, the mean disease-spe-
Hospital (J.S.), Leeds; Vascular Surgery, cific quality of life was slightly worse after treatment with foam than after surgery
Gloucestershire Royal Hospital, Glouces-
(P = 0.006) but was similar in the laser and surgery groups. There were no signifi-
ter (J.E.); Vascular Surgery, Freeman
Hospital, Newcastle upon Tyne (T.L.); cant differences between the surgery group and the foam or the laser group in
the Vascular Surgical Unit, Royal Bour- measures of generic quality of life. The frequency of procedural complications was
nemouth Hospital, Bournemouth (S.A.B.);
similar in the foam group (6%) and the surgery group (7%) but was lower in the
and the School of Health Sciences, City
University London, London (J.F.) — all in laser group (1%) than in the surgery group (P<0.001); the frequency of serious ad-
the United Kingdom. Address reprint re- verse events (approximately 3%) was similar among the groups. Measures of clinical
quests to Prof. Brittenden at the Univer-
success were similar among the groups, but successful ablation of the main trunks
sity of Aberdeen, Division of Applied Medi-
cine, University of Aberdeen, Polwarth of the saphenous vein was less common in the foam group than in the surgery
Bldg., Foresterhill, Aberdeen AB25 2ZN, group (P<0.001).
United Kingdom, or at ­j​.­brittenden@​­abdn​
.­ac​.­uk.
CONCLUSIONS
N Engl J Med 2014;371:1218-27. Quality-of-life measures were generally similar among the study groups, with the
DOI: 10.1056/NEJMoa1400781
Copyright © 2014 Massachusetts Medical Society. exception of a slightly worse disease-specific quality of life in the foam group than
in the surgery group. All treatments had similar clinical efficacy, but complica-
tions were less frequent after laser treatment and ablation rates were lower after
foam treatment. (Funded by the Health Technology Assessment Programme of
the National Institute for Health Research; Current Controlled Trials number,
ISRCTN51995477.)

1218 n engl j med 371;13 nejm.org  September 25, 2014

The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2015. For personal use only. No other uses without permission.
Copyright © 2014 Massachusetts Medical Society. All rights reserved.

2015. a diameter of the main truncal Survey (SF-36).9 but to our knowledge. and Foam Sclerotherapy (CLASS) trial. with C0 indi. with the use of a minimization algorithm that dovenous laser ablation. saphenous veins performed while the patient was cular surgeon and underwent initial duplex ultra. with the exception that one ria were an age of 18 years or older.21 of the Tessari technique21 at a ratio of 0. and patient-reported generic (i. to assess quality of life and other out. treatment options available at each investigating nical success and clinician-reported outcomes. randomized. age (<50 years or ≥50 years). The primary out- cating no signs of venous disease. C3 the presence of edema. under local anesthesia was followed by foam sonographic scanning to assess suitability for sclerotherapy to residual varicosities at the 6-week treatment and entry into the study. Treatments were assigned randomized trials that compared surgery and en. Outcomes were assessed at baseline and at 6 weeks logical [CEAP] classification system. and reflux of the great ity of life. unilateral or bilateral varicose veins. Inclusion crite. measured as cost per quality-adjusted for laser treatment. The use of sodium tetradecyl We recruited patients requiring treatment of vari. Foam was produced with the use cosities. if required). Previous randomized Aberdeen. All rights reserved. and comes of treatment of varicose veins. All patients were assessed by a vas. No other uses without permission. C1 telangiecta.22). and contraindications to the life-year gained — is not reported here. laser therapy (with the great saphenous vein only) and concurrent subsequent foam sclerotherapy for residual vari. 2014 1219 The New England Journal of Medicine Downloaded from nejm.5 ml of air (3% so- dium tetradecyl sulfate for saphenous veins and Me thods 1% for varicosities. University of Aberdeen. consisted of proximal ligation and stripping (of pared foam sclerotherapy. come measures were patient-reported disease- ses or veins ≤3 mm in diameter.org on January 4. tortuous veins considered to be unsuitable tiveness. United Kingdom. Exclusion criteria were Self-Report Questionnaire (EQ-5D) and the Med- current deep-vein thrombosis. comparative-effective. phlebectomies. gery. follow-up if required. and 6 months after treatment.5. C4 the Aberdeen Varicose Veins Questionnaire (AVVQ). three hospitals offering treatment with life was a primary outcome measure in two small only foam or surgery). Sur. Briefly. surgery comes of treatment of varicose veins. hospitals offering all three treatment options. anatomical. use of foam or to general or regional anesthesia.13 nejm. .e. Participants under- trials and meta-analyses have shown these treat. We com. a large. Laser ablation of truncal October 2012.5 ml of sodium tetradecyl sulfate to 1. sex. skin and subcutaneous changes. C2 varicose veins specific quality of life. of unilateral or bilateral primary symptomatic varicose veins (grade C2 or higher according to the Outcome Measures clinical. sulfate is licensed.org). The AVVQ is an internationally validated n engl j med 371. Copyright © 2014 Massachusetts Medical Society.. text of this article at NEJM. went randomization with even assignments to all ments to be effective in terms of short-term tech. measured at 6 months after treatment or small saphenous veins of more than 1 second with the use of the EuroQoL Group 5-Dimension on duplex ultrasonography.org  September 25. ical Outcomes Study 36-Item Short-Form Health vein thrombosis. Randomization and Study Treatment therapy and thermal ablation techniques A computer-generated randomization system was such as endovenous laser ablation have used and was managed by the Centre for Health- become widely used alternatives to surgery for the care Randomised Trials. veins (or both). eight patient-reported quality of life to assess the out. general) qual- and C6 active ulceration. but the trial involved its off- cose veins in 11 vascular surgery departments in license use as a foam rather than as its manu- the United Kingdom between November 2008 and factured liquid form. and the presence or absence of We performed the Comparison of Laser. included center. Details of treatment methods are described in multicenter. and pathophysio.1-19 center and with stratification according to the Clinical practice guidelines recommend the use of number of available options (stratum A. Another prespecified primary out- saphenous vein of less than 3 mm or more than come for this trial — 5-year estimated cost-effec- 15 mm.20 Quality of stratum B. Comparing Treatments for Varicose Veins U ltrasound-guided foam sclero. For personal use only. the published protocol21 (available with the full ness trial. acute superficial. treatment of varicose veins. with a maximum of 12 ml of Patients foam per session). the presence center performed concurrent phlebectomies. it has not been assessed as a primary outcome in reflux of either the great or the small saphenous randomized trials involving foam sclerotherapy. etiologic. and surgery. measured with the use of >3 mm in diameter. C5 healed ulcers.

29 The initial planned sample size was 1015 pa- Study Oversight tients.20. version 9. However. Sensi- ter where scanning was performed by a surgeon tivity analyses (see Table S1 in the Supplemen- who had not performed the treatment). The n e w e ng l a n d j o u r na l of m e dic i n e change-responsive tool for the assessment of qual.would provide more than 90% power to detect a mittee and the Medicines and Healthcare Prod. 2014 The New England Journal of Medicine Downloaded from nejm. 2015. scores (AVVQ. quality of life according to the consider differences to be significant only for AVVQ.org) with the use of SAS soft- success at 6 weeks and 6 months. and anxiety ses. . Data analysis was performed by statisti- veins. we cose veins (assessed by the participant and the used general linear models with adjustments for covariates used in the minimization algorithm nurse). the EQ-5D P values of less than 0.comparison of foam sclerotherapy with surgery. assessed dom effect. Blinding tary Appendix.23-26 It was overseen by a trial steering committee and consists of 12 questions and a set of mannequin an independent data and safety monitoring com- legs on which participants are asked to draw their mittee. venous clinical severity score (a score com- posed of nine categories relating to symptoms or and. and complications (assessed by the sur. for the sec- disease]). The project management group (the first six is a standardized index valuation for health sta. SF-36.difference of 0.org  September 25. No adjustment was prespecified 0 [no venous disease] to 30 [most severe venous for multiple comparisons.30.27 The SF-36 is a vali.org on January 4. Written informed con. scores range from severity scores). covariance matrix and with center fitted as a ran- sonography at 6 weeks and 6 months. In addition.594 to 1.000. and the last four authors) takes responsibility for tus. at 6 months. pain or discomfort. laser therapy versus foam sclerotherapy. sion. the prespecified comparisons of treatment with dated and reliable assessment of quality of life foam versus surgery. with higher cians at the Centre for Healthcare Randomised scores indicating a worse quality of life. the accuracy and completeness of the data. The trial was approved by the research ethics com. Saphenous-vein ablation rates were with the use of a standardized technique22 and analyzed with the use of ordinal logistic regres- reporting tool by independent. self-care. and treatment with laser versus tions. where possible. with a compound-symmetry of the saphenous vein according to duplex ultra. and SF-36 at 6 weeks. at a two-sided 5% significance level. to the protocol. and reporting and for the fidelity of the study or depression. The EQ-5D Trials. and rates of complications were analyzed lar technologists (with the exception of one cen. For personal use only.31 1220 n engl j med 371. EQ-5D. we performed a post hoc analysis of nent and the mental component). For all theseanalysis of the trial was performed when all par- measures of quality of life. we geon or nurse). The trial ity of life in patients with varicose veins. Scores range from 0 to 100. it includes five dimensions (mobility. with each sum. scores range from −0. All rights reserved. and venous clinical ing to the use of compression. accredited vascu. as measured ware.3 (SAS Institute). by the proportion of patients with residual vari. with higher scores An intention-to-treat analysis was performed for indicating better health). minimal clinically im-ticipants had completed the 6-month follow-up. We analyzed the con- VAS and the eight SF-36 domains at 6 weeks and tinuous outcomes with mixed-model repeated- 6 months. with the use of binary logistic regression. To analyze comparisons between groups. portant differences after treatment for varicose Study analyses were conducted according to a veins are not known.005.13 nejm.org) were car- with respect to the treatment used was not fea.25 SD in the AVVQ score for the ucts Regulatory Authority. The principal scores indicating greater well-being.28 The 36 questions assess eight domains and surgery. ried out in the case of missing AVVQ responses sible. analy- usual activities. available at NEJM. yield two summary scores (the physical compo. with higher scores indicating a better quality of life) and a visual analogue scale (EQ-5D VAS. Statistical Analysis scores range from 0 to 100. involving participants from and is widely used for a variety of clinical condi-strata A and B. involving participants from stratum A. prespecified statistical plan (available with the Secondary outcomes were as follows: clinical protocol at NEJM. Copyright © 2014 Massachusetts Medical Society. involv- mary score ranging from 0 to 100 and higher ing participants from stratum A. which. No other uses without permission. sent was obtained from all participants. adjustments for baseline signs of venous disease and one category relat. and ablation rates of the main trunks measures analysis. EQ-5D.ondary outcome measures presented here.

8). 3369 (51. −1.8). Comparing Treatments for Varicose Veins and 80% power to detect a difference of 0.7.8). the AVVQ score in the AVVQ score at baseline and at 6 months was the foam group was significantly higher (indicat- better than originally assumed.13 nejm. vitality (effect size. .25 SD performed in 38% of participants in the foam in the AVVQ score for the comparison of laser group and 31% of participants in the laser group. role limitations due to emotional square centimeter was achieved per laser pulse. which excluded the data from the truncal saphenous veins (i. pants had been recruited. The improve- ment in the AVVQ score in the laser group did not differ significantly from that in the surgery R e sult s group.97 to −0. 95% CI.2 to −1.54. and the results for secondary quality-of-life outcomes.. Most in.02). Baseline characteristics were group (P = 0. but the difference was outcome data according to group assignment dur. All rights reserved. laser group. terval [CI]. ment with foam or laser had the option of treat. lowing exceptions: the diameter of the great sa- phenous vein was larger in the laser group than Secondary Outcomes in the foam group (P = 0. significant between-group differences surgery group than in the foam group (P = 0. role limitations due to physical health (effect size. except that the AVVQ score was significantly ticipants who declined to undergo randomization worse in the foam group than in the laser group had a preference for a specific treatment. −4. There were no significant differences be- Patients and Treatment tween the groups in the EQ-5D score or the SF-36 Of 6592 participants referred from primary care physical component score. 2015. the only the eligibility criteria.7 to −0. procedures.9 to −0. −2.0 to −0. 95% CI. ble S2 in the Supplementary Appendix. health (effect size. which was slightly higher (bet- 785 were included in the trial. −3.006). 2014 1221 The New England Journal of Medicine Downloaded from nejm. Only the data and ing a worse disease-specific quality of life) than safety monitoring committee was aware of the that in the surgery group. The reasons for ter generic quality of life) in the laser group than ineligibility and the reasons that eligible patients in the foam group (effect size. The data and safety monitoring com- mittee and trial steering committee approved a Primary Outcomes revised recruitment target of 779 patients on the Table 2 summarizes results for the primary out- basis of data showing that the correlation between come measures. For the post hoc analy- for consideration of treatment. and 87% in −2. the patients were the center where the largest number of partici- asymptomatic or did not have reflux) or had re. 93% in the foam group. 77% of surgery group than in the laser group for the do- the foam procedures.005) included a lower AVVQ score (indicat- The proportion of treated participants who re. 95% CI.9) and lower SF-36 scores the surgery group.06. 95% CI. formed by surgical trainees. with the fol.9). Fully trained staff surgeons (indicating a worse generic quality of life) in the performed 73% of the laser procedures. For personal use only. 1) and component score.74. with surgery. −2. No other uses without permission.005 considered to indicate statistical signifi- n engl j med 371. the incidence of Quality of Life reflux below the knee was greater in the foam Table S4 in the Supplementary Appendix shows group than in the laser group (P = 0. P = 0. the remaining procedures were per. to 3. −2.50. and 59% of the surgical mains of bodily pain (effect size.3. Baseline (P = 0.3.1%) met sis of treatment with laser versus foam. −2.4. Results of a post hoc sensitivity analysis of the eligible participants did not require treatment of primary outcomes. of whom 798 (23. 1. −3. ing a better disease-specific quality of life) in the ceived their assigned treatment was 97% in the surgery group than in the foam group (effect size. 95% confidence in- ing the trial.org on January 4.7%) con.03).006). P = 0. 0. significant difference was in the SF-36 mental sented to participation in the trial (Fig.01 declined randomization are summarized in Ta. such treatment was P<0. Most of the eligible par.5.4 to −0.e. −5. A mean of 83 J per −3. These four SF-36 ment with foam for any residual varicosities at the domain scores did not differ significantly (with 6-week follow-up assessment. were generally similar. At incidence of deep venous reflux was less in the 6 weeks.org  September 25. At 6 months. moderate (effect size. 95% CI. (P<0. 95% CI. tary Appendix). −4. Copyright © 2014 Massachusetts Medical Society.04) (see Table S3 in the Supplemen- generally similar among the groups. and the SF-36 mental component score demographic and clinical data for the three groups was worse in the surgery group than in the laser are shown in Table 1.048).008). and Participants who were randomly assigned to treat. current varicose veins.

recurrent varicose veins (3 participants).044. 2015. The venous clinical severity score was moderately There were fewer residual varicose veins in the but significantly lower (indicating less residual surgery group than in the laser group at 6 weeks. The n e w e ng l a n d j o u r na l of m e dic i n e 6592 Patients were assessed for eligibility 2847 Were ineligible 376 Had unknown eligibility status 3369 Were eligible 2571 Declined to participate 798 Underwent randomization 292 Were assigned to undergo 294 Were assigned to undergo 212 Were assigned to undergo foam sclerotherapy surgery endovenous laser ablation 286 Were included in trial 289 Were included in trial 210 Were included in trial 6 Were excluded after randomization 5 Were excluded after randomization 2 Were excluded after randomization 6-Wk follow-up 6-Wk follow-up 6-Wk follow-up 265 (93%) Completed clinical 251 (87%) Completed clinical 193 (92%) Completed clinical assessment assessment assessment 247 (86%) Completed questionnaire 237 (82%) Completed questionnaire 186 (89%) Completed questionnaire 12 (4%) Were withdrawn 29 (10%) Were withdrawn 6 (3%) Were withdrawn 6-Mo follow-up 6-Mo follow-up 6-Mo follow-up 251 (88%) Completed clinical 236 (82%) Completed clinical 183 (87%) Completed clinical assessment assessment assessment 238 (83%) Completed questionnaire 214 (74%) Completed questionnaire 175 (83%) Completed questionnaire 16 (6%) Were withdrawn 33 (11%) Were withdrawn 9 (4%) Were withdrawn Figure 1. or coexisting conditions or lack of reflux (5 par- ticipants).014 to 0.074). and Outcomes. in the surgery group than Clinical Outcomes in the foam group. in the foam group at 6 weeks (0.org on January 4. but the differences were small. but not the participants.org  September 25. No other uses without permission. as reported by the nurses foam group after 6 weeks. 95% CI.13 nejm. there were 0. 2014 The New England Journal of Medicine Downloaded from nejm. only the EQ-5D score was significantly ser and the surgery groups or between the laser lower (indicating a worse generic quality of life) and the foam groups at either time. cance) between groups at 6 months. A total of 13 participants (1. fewer residual varicose veins. as assessed by both participants and nurses. For personal use only. There was no sig- hoc comparisons of laser treatment versus foam nificant difference in this score between the la- treatment. For the post the Supplementary Appendix). . There were no signifi- nificant difference at 6 months (see Table S5 in cant differences between the laser and the foam 1222 n engl j med 371. Copyright © 2014 Massachusetts Medical Society. venous disease) in the surgery group than the but not at 6 months.6%) were excluded after randomization because they had veins greater than 15 mm in diameter (5 participants). Randomization. Assessment for Eligibility. but there was no sig. At both 6 weeks and 6 months. All rights reserved.

(%) 120 (57..7) 47 (16.8 Reflux above the knee only — no.006.org  September 25.4±2. varicose veins >3 mm in diameter 113 (54.8) 74 (25. However. anatomical. CEAP denotes clinical.1) Saphenous-vein involvement Great saphenous vein — no. No other uses without permission.2) C3.2%) or the surgery group (7.7 Female sex — no.6) C4.4%) or laser treatment (83. tially between groups.8) * Plus–minus values are means ±SD.8) 11 (3.2±13.001 for both comparisons).1) 147 (51.0%) than in the at 6 months.* Laser Group Foam Group Surgery Group Characteristic (N = 210) (N = 286) (N = 289) Age — yr 49.7 Unilateral involvement — no.1 8. edema 28 (13. † The body-mass index is the weight in kilograms divided by the square of the height in meters.1±3.7 8.7±2.6±2. complications occurring during treatment) veins in the laser group than in the foam group was lower in the laser group (1.1) 3 (1.7±4.1) 239 (82.0±4. Baseline Characteristics of the Study Participants According to Treatment Group.3 49. (%)§ 6 (3.4 27.3) 21 (7. (%) 182 (86. P = 0.3) Widest diameter — mm 7. surgery and the laser groups (Table 3). quality-of- verse events (see Table S6 in the Supplementary life measures at 6 months did not differ substan- Appendix for details of serious adverse events). (P<0.0%) than group than in the surgery group or the laser among those assigned to foam treatment (54. At 6 weeks and tary Appendix). .1±4.9) CEAP classification — no.2) 39 (13.8) 25 (8. Copyright © 2014 Massachusetts Medical Society. (%) 6 (2. (%)¶ 28 (13. (%) 153 (72. 6 months.7) 21 (7. P = 0. All rights reserved.008. in the foam group than in the surgery group at nificant difference in success rates between the 6 months. 2015.7 7. etiologic.6) 163 (56.4) 35 (12.02.7±14.1) 169 (59.13 nejm.8) 11 (6.2 7.8) Previous deep-vein thrombosis — no.6%.7) 8 (2.7) 232 (81. Comparing Treatments for Varicose Veins Table 1. the frequency of overall complications The frequency of completely successful ablation (predominantly lumpiness and skin staining) of great saphenous veins was significantly higher was greatest in the foam group.org on January 4. (%) 151 (96. There were no significant differences therapy or laser treatment with surgery for the between groups in the number of serious ad. treatment of primary varicose veins. and pathophysiological.4) Body-mass index† 27.9) 215 (75. There was no sig. patients treated The frequency of any procedural complications with foam had moderately worse outcomes on a n engl j med 371.7) Widest diameter — mm‡ 9.4) Reflux above and below the knee — no. skin changes with healed or active venous ulcer 12 (5. § For the comparison between the foam and surgery groups.7±3.2) 196 (67.0±13. but there were fewer residual (i. For personal use only.6) Small saphenous vein — no. group at 6 weeks and was significantly greater P<0. ¶ For the comparison between the foam and surgery groups.e.001 for both comparisons). P = 0.4) 9 (3.1%) not the nurses) (see Table S5 in the Supplemen.4) C5 or C6. Discussion Complications Serious and nonserious adverse events are shown In this multicenter trial comparing foam sclero- in Table 4. 2014 1223 The New England Journal of Medicine Downloaded from nejm.1±1. groups at 6 weeks.9) 90 (31.9) 4 (1. (%) 14 (6.4 49. the complica- among participants who were randomly assigned tion rate was significantly greater in the foam to surgery (84. skin and subcutaneous changes 56 (26.3 Deep venous reflux — no.9) 180 (98.6 27.1) 162 (56. ‡ For the comparison between the laser and foam groups. as reported by the participants (but foam group (6.2) 169 (93. (%) C2.

3 52. ‡‡ P = 0.9) −1.0 48.6±7.1)‡‡ * Plus–minus values are means ±SD.13 nejm. after treatment with laser than after treatment However.5)‖ −0. ¶ The Aberdeen Varicose Veins Questionnaire (AVVQ) consists of 12 questions and a leg diagram assessing disease-specific quality of life. these differences are nous veins at 6 weeks occurred significantly less likely to have affected quality of life. with scores ranging from 0 (worst) to 100 (best).6.6) −1. The three groups had similar improvements ness. 83%. which used less strict in quality of life that we observed were similar definitions of success.3) 0.9 to 0.177 0. partial ablation with a patent segment and cosities in a third of the participants in the laser no reflux.2±9.org on January 4.7 52.6±7.32 The nature and frequency of differences were small.8-10.5 (0.006. 23%) than after either surgery (com- group at 6 weeks may have attenuated early plete ablation.1 to 1. skin staining.1 (−2. 55%.* Laser Foam Surgery Group Group Group Surgery Surgery Laser Measure (N = 210) (N = 286) (N = 289) vs. were lower than the rates The baseline scores and overall improvements in some previous studies.1±8. § The post hoc comparison of laser versus foam includes participants from stratum A only.051 to 0. No other uses without permission.784±0.015 0. Laser‡ vs.g.1±7.7 (−0.8 to 2.5±7.6 to 0.881±0. complications were also similar to those previ- tance is uncertain.3±8.6.7 51. with scores ranging from −0.202 0. often after foam treatment (complete ablation.9 1. 8%). and their clinical impor. For personal use only.035) (−0.903±0.14 but were similar to to those reported in smaller European random. Observed ablation rates.1 52. †† The Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) comprises a physical component summary score and a mental component summary score.5 52.792±0. ‡ The comparison includes participants from stratum A only.1-16 The frequency of complications (e.4 to 1.2 (−1. 84%. Foam§ score effect size (95% CI) AVVQ¶ Baseline 17.2) 1.4 9.9 7.174 0.168 0. and numbness) was lower in the venous clinical severity score at 6 months.9±8.803±0. Estimates of Treatment Effect on Primary Quality-of-Life Outcomes.12.0 to 3.6 6 Mo after treatment 52. ously reported in the literature.7 (−3. the rates reported in two randomized trials that 1224 n engl j med 371.4) EQ-5D** Baseline 0. The n e w e ng l a n d j o u r na l of m e dic i n e Table 2. These findings. Foam† vs.025 to 0. with scores ranging from 0 (best) to 100 (worst).2.9 18.2±9.594 (worst) to 1 (best).31 for foam sclerotherapy.171 0.895±0.5 −1. ‖ P = 0. supporting the generalizability of our than did those who underwent surgery.1 6 Mo after treatment 7.0 to −0.2±8. successful ablation of the great saphe- with either foam or surgery. treatment (complete ablation.048.025 (−0. in particular surgery. However.021) (−0.0 52. † The comparison includes participants from stratum A (eight hospitals offering all three treatment options) and stratum B (three hospitals offering treatment with only foam or surgery).6) 0.6 0. 2015.1 (−1.4±8.8±9.1 17. each consisting of the sum of four individual physical domains.1.org  September 25. 6%) or laser quality-of-life benefits associated with this treat.0 (−0. measure of disease-specific quality of life (AVVQ) ized trials.4 6 Mo after treatment 53. each with three response options. lumpi.9±8..4±8.8±7.10.3 (−2. 2014 The New England Journal of Medicine Downloaded from nejm. partial abla- ment as compared with treatment with foam or tion.2±9.5.010 to 0. .059) SF-36 physical component†† Baseline 48.6±9. partial ablation. the use of foam sclerotherapy for residual vari.2 to 2.9±9.2 to 0. All rights reserved.005 −0. Copyright © 2014 Massachusetts Medical Society.6 (−2.8 48.1) SF-36 mental component†† Baseline 51. ** The EuroQoL Group 5-Dimension Self-Report Questionnaire (EQ-5D) consists of five questions.175 6 Mo after treatment 0.

3) Partial success with reflux — no.1 (3. we did not bectomy concomitantly with laser therapy. (%) 3 (2.4–1. of patients 153 205 192 Complete success — no. involved primary outcomes. include this method in our trial. First.1) 9 (4. we did not include a basis of duplex ultrasound scans obtained by in. Copyright © 2014 Massachusetts Medical Society. 2015. (%) 9 (6. and surgery for the cant differences in the AVVQ score among patients treatment of varicose veins showed no clinically undergoing laser treatment versus surgery at substantial between-group differences in quality 6 weeks. Foam† vs. quality of life favored surgery over treatment n engl j med 371. ‡ The comparison includes participants from stratum A only. ¶ P≤0.2) 35 (19.13 nejm. For personal use only.9 (3. foam sclerotherapy.005.1–8. tion is also used to treat varicose veins.4) 14 (8.1–7.5) 9 (4. and failure as more than one patent segment.33 In our study. our multicenter trial compar- saphenous vein. Finally. single-center trial showed ger follow-up is needed to assess the durability significant improvements in the AVVQ score at of effects. § The post hoc comparison of laser versus foam includes participants from stratum A only.5)¶ 1.8 (2. used definitions of technical success that were The limitations of our study should be ac- similar to ours.01. lon- versial.8–5. (%) 3 (2. Second. (%) 10 (6.org on January 4. (%) 13 (9. it is likely that some differences may have trials of treatment for varicose veins. we determined ticipants or assessors to be kept unaware of the whether ablation was complete or partial on the assigned treatment.3) 79 (43.9)¶ 0. of life.9) 12 (6. Third.2 (0. (%) 13 (8.8 (0.* Laser Foam Surgery Surgery Surgery Laser Variable Group Group Group vs.0) 37 (18.6) Failure — no. without concomitant phlebecto. Moderate differences in disease-specific tomies in the surgery group. 2014 1225 The New England Journal of Medicine Downloaded from nejm.3) Partial success with reflux — no. ablation.4) 9 (4. partial success as a patent segment of treated vein with or without reflux. We found no signifi.5) 47 (22.3) 3. and thus cannot assess the effect of treatment er than by the surgeons who had performed the relative to such a control.6) 162 (84. All rights reserved.4) Partial success without reflux — no.4) 59 (32. con.8–8. Comparing Treatments for Varicose Veins Table 3.19 occurred by chance.5) 4.1) 4.12.6–2. and a large number sures of success and technical success observed of comparisons involved secondary outcomes.7) 5. The disparity between clinical mea.9) 20 (11. of patients 141 182 173 Complete success — no. Estimates of the Success of Ablation of the Great Saphenous Vein According to Treatment Group. in this study has also been observed in other thus.0) 9 (4.org  September 25. eight comparisons treatment. group of patients assigned to a sham procedure dependent.16 A previous. secondary outcome measures to be significant current treatment of varicosities remains contro. although radiofrequency abla- 6 weeks among patients who underwent phle. † The comparison includes participants from strata A and B.16 group received treatment directed to the main In conclusion. it was not feasible for the par- of ablation in previous trials. ing the clinical effectiveness of endovenous laser mies (except at one center). Laser‡ vs.0) 112 (54. as com.0) Partial success without reflux — no.5)¶ * Complete success was defined as complete occlusion at the treated vein. Foam§ odds ratio (95% CI) At 6 wk No. (%) 116 (82. despite the use of concomitant phlebec. . No other uses without permission.1 (1. other studies pared with those who underwent laser therapy have suggested that it results in clinical outcomes alone. (%) 127 (83.4) 135 (78.4)¶ At 6 mo No. We considered differences for For patients undergoing laser treatment. participants in the laser similar to those associated with laser treatment.7) Failure — no. only for P values of less than 0.15 In contrast to the assessment knowledged.2) 4 (2.18. Fourth. accredited vascular technologists rath.

Luke Vale and Laura Ternent.2) 34/251 (13.4) Other complications Any procedural complication during treatment 2/205 (1. members of the Project Management Group for their and do not necessarily reflect the views of the National Institute ongoing advice and support of the trial. naire. ¶ P<0. treatment. For personal use only. ** P<0.7) 45/251 (17.4) Skin loss or ulceration at 6 mo 1/183 (0.0) 4/289 (1. trial.1)‡ Lumpiness at 6 mo 25/183 (13. We thank Janice Cruden for her secretarial support and data All treatments had similar clinical efficacy. Jackie Ellington.6)‖ Persistent bruising at 6 wk 10/193 (5. 2015. and ablation rates were lower after treat. Gladys McPherson and the programming team at there were fewer complications after laser treat.2)¶ 10/251 (4.4)¶ 15/265 (5.0)† 79/251 (31. 31% of the patients in the laser group also underwent foam treatment of varicosities.5)†¶ Skin loss or ulceration at 6 wk 0/193 2/265 (0. The post hoc anal- ysis of the laser group versus the foam group includes participants from stratum A only. nie for help with scoring the Aberdeen Varicose Vein Question- ment with foam. members of the study for Health Research.4)§ 109/236 (46.8) 0/236 Lumpiness at 6 wk 36/193 (18.9)‖ Numbness at 6 mo 17/183 (9. Maria Prior. All rights reserved. the Depart.001 for the comparison with the surgery group.4) Development of thread vein at 6 mo 24/183 (13. and follow-up of trial participants. team (Graeme MacLennan. † P<0.2) 40/236 (17. Ian Loftus.6) 38/251 (15. ed recruitment. 1226 n engl j med 371.001 for the comparison with the foam group. Thread veins are small clusters of blue or red veins. 2014 The New England Journal of Medicine Downloaded from nejm.1)† Any complication at 6 wk 103/193 (53. Serious Adverse Events and Other Complications According to Treatment Group.6)§ 67/251 (26.5)† 32/251 (12. § P<0.org on January 4.6)†** 24/236 (10.6) 2/251 (0.2) 27/265 (10.0) 17/275 (6.7) 171/265 (64.2) * The comparison between the foam group and the surgery group includes participants from strata A and B.05 for the comparison with the laser group.05 for the comparison with the foam group.0) Skin staining at 6 mo 32/183 (17. and Marcus Flather). ‖ P<0.6) 144/251 (57. the National Health Service.5) Serious adverse events related to treatment 2/210 (1.0) 3/286 (1.8) 1/251 (0. The Health Services Research Unit is funded by the ston Banya.2) Numbness at 6 wk 22/193 (11. our original economists in The views expressed in this article are those of the authors the group.0) Skin staining at 6 wk 18/193 (9. Copyright © 2014 Massachusetts Medical Society. the Centre for Healthcare Randomised Trials. quality of life favored laser treatment over foam. No other uses without permission. or the funders that provided institutional sup.6) 26/236 (11.9)‡ Any complication at 6 mo 89/183 (48.3) 105/265 (39.0) Persistent tenderness and discomfort at 6 wk 41/193 (21.5)†** 83/251 (33.* Event Laser Group Foam Group Surgery Group number/total number (percent) Serious adverse events overall 7/210 (3. cruitment sites (see the Supplementary Appendix) who facilitat- torate.8) 10/289 (3. The n e w e ng l a n d j o u r na l of m e dic i n e Table 4. . and Denise Bolsover) ment of Health. but management.4)§ 92/251 (36. Win- (06/45/02). with foam.13 nejm.7)‡ Persistent bruising at 6 mo 25/183 (13.3) 11/286 (3.6)† 168/251 (66.6)‡** 17/236 (7.2) 49/265 (18. and Yvonne Fer- ment. and moderate differences in generic Disclosure forms provided by the authors are available with the full text of this article at NEJM. Tracey Davidson.0) 37/236 (15. Lynda Constable.4) 219/265 (82.2)‡ Development of thread vein at 6 wk 10/193 (5. and the staff members at re- Chief Scientist Office of the Scottish Government Health Direc. Laura Elliott.6)†** 20/251 (8. the com- parison between the surgery group and the laser group includes participants from stratum A only.org.2)‡ 21/251 (8.05 for the comparison with the surgery group. who contributed to the behavioral recovery component of the port for this study.2)† 19/267 (7.2) 122/265 (46. the independent members of the trial steering committee Supported by a grant from the Health Technology Assessment (Alun Davies [chair].org  September 25. ‡ P<0.001 for the comparison with the laser group. and Jane Nixon) and the data Programme of the National Institute for Health Research and safety monitoring committee (Gerry Stansby [chair].

Ann Surg 12. Eur J Vasc 22. Ran. For personal use only. Carradice D. Della. Nesbitt C. Br J Surg 2009. 51995477. 1988. Endovenous ab. Horton NJ. pean randomised controlled trial. Qual Life domized clinical trial of endovenous laser 15. Carradice D. sdale. Carradice D. NJ:​Lawrence Erlbaum Associates. 2015. high ligation and stripping in patients 53:​Suppl:​2S-48S. Badri H. Reliability and va- Br J Surg 2011. Stenger al. et al. Lattimer CR. The care of patients with vari.uk/​dataset/​provisional-monthly-patient dovenous laser ablation with surgery for Database Syst Rev 2011. nous laser ablation (980 nm) using local analysis.​46:​582-9.​5:​223-34. LAser.sf-36​. et al. nous vein. for patients with varicose veins. Figueiredo M. Wallace T.shtml). BMJ 2011. SF-36 Health Survey update for great saphenous varicose veins. Rasmussen LH. Kalodiki E. Randomized clinical trial of Blemings A. Brazier JE. Cochrane .​Lancet protocol 09PRT/2275 2. Comparing endovenous la. PROMs outcomes summary by provid- grammaticas D. and recurrence after 2 years. life in patients with varicose veins. Christenson JT. Biemans AAM. Nijsten T. J Vasc Surg 2011. J Vasc Surg 2013. van den Bos R. the great saphenous vein with endove. Eklof B. management of varicose veins in the pres. Kockaert M. Random.​21:​180-90. Azzam M. Ruta DA. nous laser ablation with stripping of the (http://www​. All rights reserved. 13. April 2010–March 2011 (http://data​. Bjoern L. Comparing Treatments for Varicose Veins References 1. year results of a randomized clinical trial trial. Eklof B. Wright D. White IR. Phle. Akkers. J Vasc Surg 2007. Kundu S. London:​National Institute for Health (http://www​. Blemings A. early results of a randomised controlled 23. Shawish E.​257:​419-26.​52:​1234-41. 33.com/​protocol Mekako A. et al. ciety of Interventional Radiology. Ran. Rasmussen LH. et al.org/​tools/​sf36​. Mekako AI.​39:​630-5. Endovenous laser ablation with sound-guided foam sclerotherapy versus 24. Gough MJ. Eklöf B. Bradbury A.​30:​710-9. Campbell WB. Eur J Vasc Vennits B. 2014 1227 The New England Journal of Medicine Downloaded from nejm.​28:​ compared with ultrasound-guided foam 25. Blemings A. Cost-effectiveness analysis of surgery ver- Eur J Vasc Endovasc Surg 2010.​95:​294-301. Flessenkämper I. Random. Garratt AM. Gauw SA. Responsiveness of the SF-36 5. Buckingham JK. 10. great saphenous vein: clinical outcome -reviews/​09PRT-2275). Lurie F.euroqol​. Surg 2011.​38:​758-63. J Vasc initial results of a multicentre random. stripping of the great saphenous vein: two.org on January 4. venous insufficiency: joint statement of 3. for great saphenous vein varices. Recommended reporting standards for small saphenous varicose veins. a treatment for varicose veins (ISRCTN ventional surgery for great saphenous Lawaetz B. J Vasc Surg 2009.​58:​727-34.​40:​649-56. Lancet 2010. ZH. Hatfield J. Phlebology 2013. Hartmann M. Nelemans P. Res 1996. Kockaert M. Br J veins. Statistical power analysis and surgery for treatment of primary great 2008. foam sclerotherapy and surgical stripping diagnosis and management of varicose 28. Copyright © 2014 Massachusetts Medical Society. foam sclerotherapy. Varicose Vein Questionnaire. ing outcome data.​46:​308-15. Br J Surg 2012. Palfreyman S. Results of surgical treatment Surg 1999.​2:​5-10. Br J Surg 2006. lation versus conventional surgery for Endovasc Surg 2010. Macdonald LM. Br J Surg 2008.​EudraCT 2008-001069-26). Hatfield J. Eur J Vasc Endovasc Surg 2012. and con. Hill- saphenous varicose veins with a 2-year 19. Health Care 1993. the Netherlands:​ laser ablation. clinical trial. with varicose veins: short-term results. Guest M. trial comparing endovenous laser ablation mary varicosis. and foam sclerotherapy for pri. Gemayel G. Frings N. Samuel N. The CLASS Study:​Comparison of dijk GP.13 nejm.​148:​ 594-600. Greenhalgh RM.uk/​CG168). sclerotherapy in patients with varicose Russell IT. Krukowski (RELACS study). 11.​ cose veins. pared with surgery or sclerotherapy in the CHA. and improving health-related quality of ment of great saphenous varicose veins: 14. Systematic re. . lidity of the Dutch translated Aberdeen 6. after endovenous laser therapy for vari- ablation of the great saphenous vein with ican Venous Forum.​49:​230-9. ence of trunk vein incompetence: Euro. et al. Random. Miranda FJ Jr. Michaels JA. Garratt AM. EQ-5D. Trueman P. Cohen J.​99:​1062-70. MC.org  September 25. Dalsing -england-april-2010-to-march-2011).​96:​369-75.​342:​d40. Bhattacharya V. Comerota AJ. Mazari FAK. Surgery and foam Sclerotherapy as ser ablation. 26.gov​ Randomized clinical trial comparing en. Gobin JP. J Vasc laser ablation and high ligation with therapy in superficial venous insufficiency: Surg 2007. and a condition-specific measure of health Samuel N. ized clinical trial of endovenous laser ab. Smith JJ. ized clinical trial comparing endovenous bology 2006. diofrequency obliteration. Geroulakos G. Po- 7. Varicose veins in the legs CG168:​the EuroQol (http://www​. J Cardiovasc Surg (Torino) 31. Endovenous thera. lation (radiofrequency and laser) and foam 32. Rasmussen LH. endovenous ablation for the treatment of 2013. 21. Ratcliffe J. Rotterdam. Shadid N. No other uses without permission. ized controlled trial. Hatfield J. n engl j med 371. endovenous laser cated varicose veins in a randomized Bounameaux H. follow-up.​93:​182-6. Qual 4. diseases: clinical practice guidelines of the concomitant or sequential phlebectomy ized trial comparing endovenous laser Society for Vascular Surgery and the Amer. et al. cock SJ. Bjoern L.​98:​1079-87. Mavor AI. Ceulen R. Ruta DA.​10:​CD005624. Gloviczki P. veins: a prospective randomised study.org).​49:​213-33. analysis in randomised trials with miss- sapheno-femoral ligation and stripping of Neumann M. Sybrandy JEM. Klem TMAL. 27. Mekako AI. Brunkwall J. Arch Dermatol 2012. Garratt AM.thelancet​. Gueddi S. Lawaetz M. Arends L. Lawaetz M. Endovasc Surg 2009. cose veins and associated chronic venous Chetter IC. domised clinical trial comparing endove. 2nd ed. Darwood RJ. Stansby G. 16. Lawaetz M. Prospective randomized therapy. sus conservative treatment for uncompli- 8.org​. Pronk P. Carpenter J. Chetter IC. ablation compared with conventional sur. radiofrequency ablation. Lawaetz B. Towards measurement of outcome 49-58. Coyne P. Essink Bot ML. Cost and effectiveness of laser with phle. varicose veins. J Vasc Surg 2010. Chetter I. Copyright © 2014 Massachusetts Medical Society. and Care Excellence (http://guidance​. 16-23. pies of lower extremity varicosities: a meta. D. Glowacki P. Strategy for intention to treat Randomised controlled trial comparing 17. Randomized clinical trial of ultra. The American Venous Forum and The So- Comparable effectiveness of endovenous bectomies compared with foam sclero. MacIntyre JB. 30. view and meta-analysis of endovenous ra. 9. Araújo S. tumescent anaesthesia: one year results. Millward SF. -reported-outcome-measures-proms-in the treatment of primary great saphenous 20. varicose veins. Theivacumar N. Varisolve polidocanol microfoam com.​43:​ Russell IT.nice​ 29. sclerotherapy versus conventional surgery er. . for the behavioral sciences. et for patients with varicose veins. Wittens gery for great saphenous varicose veins. Rass K. Evaluating high ligation and stripping in the treat.​98:​501-10. Ware JE Jr. Roll S. and without high ligation compared with surgery for the incompetent great saphe. Eifell RK. Luebke T. 18.​37:​232-8. Davies AH. Barros NJ Jr. Eur J Vasc Endovasc Surg 2009. Mooij MC. Bjoern L. Abdalla MI.