Comparisons of side effects using air and carbondioxide foam for endovenous chemical ablationNick Morrison, MD,a Diana L. Neuhardt, RVT,b Charles R. Rogers, MD,a

James McEown, MD,a Terri Morrison, RN, BSED,a Elaine Johnson, RN, BSN, CNOR,a andSergio X. Salles-Cunha, PhD,b Scottsdale, Ariz

Objective: This clinical study evaluated prospectively adverse events immediately following ultrasound-guided foamsclerotherapy (UGFS) for the treatment of lower extremity venous valvular insufficiency. Incidence of side effectsassociated with carbon dioxide (CO2) foam was compared with a historical control using air-based foam. The literatureon the subject was reviewed.Methods: Vital signs were monitored during and immediately after UGFS, and adverse events were recorded for 24 hoursfollowing the procedure. The air-based foam group had 49 patients: 44 women and 5 men. The CO2-based foam grouphad 128 patients: 115 women and 13 men. CEAP class was C2EpAsPr, describing varicose veins, primary etiology, andsaphenous reflux. UGFS followed thermal ablation of the great saphenous vein. Foam was prepared using the three-waytap technique to mix gas with 1% polidocanol in a 4:1 ratio. Segments of the great and small saphenous veins and theirtributaries were treated with UGFS. Foam volumes injected were 27 � 10 (SD) (6-46 range) and 25 � 12 (6-57 range)mL for air- and CO2-based foams respectively (P � .39). Incidence of adverse events was compared by �2 statistics. Vitalsigns were compared by paired t test.Results: During the procedure, the average heart rate decreased by less than 5 bpm for both groups (P < .001), and bloodpressure decreased by less than 3 mm Hg in the CO2 group (P < .02). Respiratory rate, electrocardiogram, and pulseoxymetry did not change significantly in both air- and CO2-foam series (P > .05). Visual disturbances were experiencedby 3.1% (4/128) and 8.2% (4/49) patients in the CO2 and air groups respectively (P � .15). Respiratory difficulties orcircumoral paresthesia each occurred in 0.8% (n � 1) of the CO2 patients. Incidence of chest tightness (3.1% vs 18%), drycough (1.6% vs 16%), or dizziness (3.1% vs 12%) were significantly lower in the CO2 vs air groups (P < .02). Nauseaoccurred in 2% and 4% of the CO2 and air-based foam groups (P � .53). Overall, the proportion of patients describingside effects decreased from 39% (19/49) to 11% (14/128) as CO2 replaced air for foam preparation (P < .001). Similarfindings were described in the literature of air-based foam but data on the use of physiological gas were rare.Conclusions: Side effects decreased significantly if CO rather than air was employed to make the sclerosing foam for

2

chemical ablation of superficial veins of the lower extremity. ( J Vasc Surg 2008;47:830-6.)

Endovenous chemical ablation, ultrasound-guidedfoam sclerotherapy (UGFS) in particular, has become acommon treatment of patients with chronic venous valvularinsufficiency.1,2 The foam is produced by agitating a gas witha detergent, liquid sclerosant. Air has been commonly em-ployed but a more physiological gas such as carbon dioxide(CO2) may have certain advantages.1-7 Foam can be preparedin a variety of ways to treat small or large veins,2,8 improvesultrasound (US) visualization of the venous lumen, i s moreeffective than liquid at lower dosages of sclerosant, and hasbecome an efficient alternative to venous phlebectomy orstripping.4,6,7,9-16

Patients of all abnormal, clinical, CEAP classes havebeen treated with UGFS.1,2,4,6,7,11-14,16-28 Foam sclero-

From the Morrison Vein Institutea and Compudiagnostics, Inc.b

Competition of interest: Dr Morrison has received fees for consulting fromProvensis (sponsors of the work described in reference 7). The presentstudy was sponsored by the authors.

Reprint requests: Nick Morrison, MD, Morrison Vein Institute, 8575 EastPrincess Dr Suite 223, Scottsdale, AZ 85255 (e-mail: nickmorrison@morrisonveininstitute.com).

0741-5214/$34.00Copyright © 2008 by The Society for Vascular Surgery.

doi:10.1016/j.jvs.2007.11.020

830

therapy has been used to successfully treat: (1) saphenousveins primarily, (2) to complement, complete, or extendstripping or thermal ablation by laser or radiofrequency, (3)recurrent veins, (4) venous malformations, (5) venous an-eurysms, (6) sciatic nerve varices, and (7) saphenous ornonsaphenous perforating veins.1-31

Side effects and adverse events have been described notonly for UGFS but for all methods of treating the superfi-cial veins of the lower extremity, surgical or nonsurgi-cal.1-35 The gas in the foam may have the potential for local,pulmonary, visual and/or cerebral effects, particularly in apatient with a patent foramen ovale or other, right-to-leftshunt. W e have tracked foam, or gas bubbles, to the rightand left chambers of the heart and to the middle cerebralartery using transcutaneous, transthoracic echo (TTE),and transcranial Doppler (TCD).36 In the initial phase ofour experience, the safety profile of air-based polidoca-nol foam was evaluated in-house. Contemplating the useof a more physiological gas, CO2, an investigation wasrepeated for CO2-based foam. Therefore, this cohortstudy, with comparison with historical control, focusedon two primary objectives: (1) to evaluate possible tox-icity and complications of CO2-based foam; and (2) tocompare the safety profiles obtained for CO2 vs air as two

distinct gases used in UGFS.

JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Morrison et al 831

PATIENTS AND METHODS

A prospective study was designed to investigate theincidence of side effects and adverse reactions of a carbondioxide (CO2)-based foam for chemical ablation in thetreatment of chronic venous valvular insufficiency. Theresults were compared with a similar prospective investiga-tion previously performed to evaluate air-based foam.

Patient population. Subjects who were candidates forchemical ablation received instructions about polidocanol,its use in Europe, Canada, and the United States, its statuswith the Federal and Drug Administration (FDA), andpotential alternative treatments. Patients signed informedconsents for treatment with polidocanol and for participa-tion in the quality assurance research program.

Subjects entered either study after thermal ablation ofthe great saphenous vein (GSV). Thermal ablation wasperformed either with laser or radio-frequency. The GSVwas occluded to the saphenofemoral junction (SFJ). Con-current ambulatory phlebectomy was performed as indi-cated for the treatment of varicose veins. Tributaries of theSFJ were not treated. Other abnormal, superficial veinsremained patent after the initial GSV ablation. The thermalablation procedure was followed by chemical ablation atone-week to eliminate (1) the GSV distal to the ablatedsegment, (2) saphenous tributaries, (3) a refluxing smallsaphenous vein, (4) other superficial veins with reflux orwith potential to be part of a refluxing network, and/or (5)some unablated or recanalized segment of the GSV previ-ously treated.

Investigations of air- and CO2-based foams included49 and 128 patients: 44 women and 5 men and 115 womenand 13 men, respectively. Predominance of female genderhad a 9:1 ratio. The subjects evaluated represented com-mon patients seen in a phlebology outpatient clinic of athriving urban community. Young adults to active retiredpeople were treated. Average age of the air-based foamgroup was 52 � 12 (SD), ranging from 23 to 80 years.Average age of the CO2-based foam group was 52 � 11(SD), ranging from 29 to 89 years. The CEAP classificationof both groups was C2EpAsPr describing varicose veins,primary etiology, and saphenous anatomy with reflux last-ing longer than 500 msec.

Foam preparation. A liquid sclerosant, polidocanol,was mixed with gas in a 4:1 gas/liquid ratio. Liquid poli-docanol was diluted to 1%. In the initial safety profile study,the gas employed was air. In this update, carbon dioxide(CO2) was mixed with liquid polidocanol. The three-waytap technique described by Tessari was employed to pre-pare the foam.17

Foam injection. Foam was injected under ultrasound-guidance. The first injection was performed distally. Foamprogression was followed by ultrasound to determine thesite of the following injection. Injections were stoppedonce foam approached a connection to the deep venoussystem.

Foam volumes. The average volume of air-based foam

injected was 27 � 10 (SD) mL, ranging from 6 to 46 mL.

The average volume of CO2-based foam injected was 25 �12 (SD) mL, ranging from 6 to 57 mL. The end point oftreatment was the perceived injection of foam by ultra-sound in the various venous segments selected for treat-ment. Total volume injected became a theoretical end-point since the volumes injected in these studies exceededthe maximum volumes described by other investigators.11

Patient monitoring. A registered nurse (RN) moni-tored vital signs at various times: (1) preinjection; (2) every15 minutes during the injection procedure; and (3) 30 and60 minutes after the procedure was terminated. The fol-lowing vital signs were monitored: systolic and diastolicblood pressures, heart rate, respiratory rate, electrocardio-gram (ECG), and pulse oximetry.

The RN also monitored and registered sequelae, sideeffects, complications and/or patients statements of un-usual sensations every 15 minutes during the procedureand ½, 1, 2, 4-6, and 24 hours after the procedure. Theexpected sequelae were: (1) injection site itching; and (2)localized leg pain. Side effects or complications specificallymonitored for were: (1) chest tightness; (2) dry cough; (3)metallic/medicinal taste; (4) nausea; (5) dizziness; (6)circumoral paresthesia, (7) respiratory difficulty; and (8)visual disturbance. All nurses involved with the study hadup to 10 years of extensive experience dealing with patientsbeing treated for chronic venous insufficiency in an outpa-tient setting. A patient-supporting role was an intrinsic partof their participation. Besides being available to counsel thepatients and address their questions and concerns about thetreatment being performed, the nurses were instructed toask the patients to describe their condition at set intervals;they did not volunteer an answer or ask specific questionsabout the items mentioned above. Suggestions wereavoided but, in contrast, the patients had freedom to ex-press their concerns and unusual sensations. The phrasesused by the patients were then recorded and compared withthe closest item in the list. For example, aching was classi-fied as leg pain. The 2-hour, 4-6-hour, and 24-hour inter-views were conducted by phone. Although not “blind” tothe gas used, the RNs in the clinic had no apparent reasonto favor one type of gas over another.

Patients wore loose clothing, slacks, or shorts andbrought two pairs of support hose to the injection sessions.Patients were asked not to shave or apply lotion to their legsthe morning before the injections. Emphasis was placed onpostinjection compression. The patients were instructed towear two pairs of compression hose during the day for thefirst 48 hours and to sleep with one pair only for those twonights. Shower was allowed but hose had to be reappliedpromptly. Patients had to take, at a minimum, a brisk30-minute walk every day in their support hose. Normalactivities were resumed immediately after injection with thesupport hose in place. Strenuous activity had to be avoidedfor 2 days. Weight lifting had to be avoided for 2 weeks, andhot baths, hot showers, saunas, steam rooms, and whirlpooltype tubs for 2 to 3 weeks. Prescribed baby aspirin weretaken and patients could take acetaminophen for pain or

discomfort. A list of anti-inflammatory medicines not to be

JOURNAL OF VASCULAR SURGERYApril 2008832 Morrison et al

taken for 1 week was provided to the patients. Iron orvitamin E supplements were to be avoided for 2 days beforeand 2 weeks after injections. The patient spent most of thefirst hour after treatment getting ready with support hose,dressings, and performing mild activity in the clinic whilereceiving detailed instructions. This basic protocol wassimilar to all patients in the air- and CO2-based foamgroups.

Statistical analysis. Vital signs comparisons were per-formed using the paired t test provided with the Excelprogram (Microsoft, Inc, Seattle, Wash). Incidence of sideeffects and adverse events observed for the air- or CO2-

based foam groups were compared using the �2 programalso provided with the Excel program.

RESULTS

After injection of air- or CO2-based foam, there wereno physiologically significant changes in vital signs or in theelectrocardiogram (Table I). However, statistically signifi-cant differences were noted in the following vital signsmeasured at base line and during foam injection:

(1) during air-based injection, the average decrease of 4bpm in heart rate was significant (P � .001 by pairedt test); and

(2) during CO2-based injection, the average decreases of 3bpm in heart rate (P � .001), 4 mm Hg in systolicblood pressure (P � .001), and 2 mm Hg in diastolicblood pressure (P � .03) were significant by pairedt test.

Localized itching was reported by 6% (3/49) and by15% (19/128) of the patients receiving air- or CO2-basedfoam injections, respectively (P � .12). Localized pain wasreported by 22% (11/49) and by 20% (26/128) of thepatients in the air- and CO2-based foam groups, respec-tively (P � .75).

Table II presents the incidence of side effects andcomplications following air- or CO2-based foam sclero-

Table I. Vital signs associated with chemical ablation

Ultrasound-guidLiquid: 1

Gas: air or car

Preprocedure

Air-based foamHeart rate (bpm) 77 � 10 (SD)Respiratory rate (rpm) 19 � 3PO2 (%) 98 � 2Systolic pressure (mm Hg) 127 � 16Diastolic pressure (mm Hg) 79 � 11

CO2-based foamHeart rate (bpm) 75 � 12Respiratory rate (rpm) 14 � 4PO2 (%) 98 � 2Systolic pressure (mm Hg) 128 � 16Diastolic pressure (mm Hg) 76 � 11

therapy. Chest tightness, dry cough, and dizziness de-

creased significantly by substituting CO2 for air as a gas toform the sclerosing foam.

Average air-based foam volumes injected in patientsreporting vs patients not reporting (1) chest tightness,29 � 11 (SD) vs 26 � 10 mL, (2) dizziness, 26 � 11 vs26 � 10 mL, or (3) leg pain, 28 � 10 vs 26 � 10 mL, werenot statistically significantly different (P � .37). A statisti-cally significant higher volume was injected in patientsreporting vs not reporting dry cough, 35 � 9 vs 24 � 9 mL(P � .005). No side effects or complications listed in TableII were reported by five patients receiving less than 15 mLof air-based foam. The four patients reporting visual distur-bances received 27, 17, 15, and 15 mL of air-based foam.

Average CO2-based foam volumes injected in patientsreporting vs patients not reporting (1) events listed in TableII, 27 � 14 (SD) vs 24 � 11 mL, or (2) leg pain, 23 � 9(SD) vs 25 � 12 mL, were not statistically significantlydifferent (P � .36). The four patients reporting visualdisturbances received 8, 16, 18, and 40 mL of CO2-basedfoam.

The figure reports and compares the probability of

m sclerotherapyidocanolioxide (CO2)

15 min into procedure 60 min postprocedure

73 � 8 68 � 1019 � 4 19 � 397 � 2 98 � 1

129 � 15 124 � 1580 � 12 74 � 9

72 � 10 71 � 1114 � 4 14 � 498 � 2 98 � 2

124 � 16 125 � 1674 � 12 72 � 12

Table II. Side effects following chemical ablation

Ultrasound-guided foam sclerotherapyLiquid: 1% polidocanol

Gas: air or carbon dioxide (CO2)

Condition

Gas

�2Air CO2

Chest tightness 9 (18%) 4 (3.1%) P � .001Dry cough 8 (16%) 2 (1.6%) P � .001Dizziness 6 (12%) 4 (3.1%) P � .019Visual disturbance 4 (8.2%) 4 (3.1%) P � .15Metallic/medicine taste 0 (0%) 2 (1.6%) P � .39Nausea 2 (4%) 3 (2%) P � .53Circumoral paresthesia 0 (0%) 1 (0.8%) P � .53Respiratory difficulty 0 (0%) 1 (0.8%) P � .53Total 49 (100%) 128 (100%)

ed foa% polbon d

occurrence of chest tightness, dry cough, and/or visual

JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Morrison et al 833

disturbance for air- and CO2-based foam. These events didnot occur if the volume of foam injected was less than15-16 mL. Above these values, the event probability wasnot related to volume injected. In these studies, however,only five (10%) and 15 (12%) patients in the air- andCO2-based groups had such low volumes injected. A 30%to 40% incidence of such events with air-based foam de-creased to less than 10% if CO2 were used rather than air.

In summary, 39% (19/49) and 11% (14/128) of thepatients in the air- and CO2-based foam groups, respec-tively, reported events listed in Table II (P � .001). Withthe exception of one patient in the CO2 group reportingnausea, all these events cleared within 24 hours.

DISCUSSION

The evolution from visual liquid sclerotherapy to (1)ultrasound-guided liquid sclerotherapy, (2) air-based foam,and, presently, (3) a more physiological gas, eg, CO2- basedfoam has growing evidence presented in the literature. Thesafety profile studies performed in our own clinic alsofavored the use of a physiologic gas, CO2, to create thefoam needed for chemical ablation of insufficient superficialveins of the lower extremity.

In our practice, transitions occurring in chemical abla-tion have been monitored clinically, physiologically, andultrasonographically. The nurses involved in the recordingof events had several years of experience in dealing withpatients being treated in an outpatient phlebology clinic.Of interest in the methodology employed, the patients didnot receive suggestions of signs, or symptoms being inves-tigated. Any description of side effects had to be volun-teered by the patient. Some events, specifically dry cough,were observed directly while the patients were still in theclinic. Transthoracic echo (TTE) and transcranial Doppler(TCD) have been employed to demonstrate if a foramenovale is patent and if bubbles can reach the brain.36 Thisstudy with CO2-based foam, as well as its predecessor withair-based foam, was designed to record the patients’ vitalsigns, side effects, and adverse reactions following a pro-

Fig. Endovenous chemical ablation: event probability as a func-tion of air- or carbon dioxide (CO2)-based foam sclerotherapywith volumes less than indicated in the horizontal axis.

spective, monitored protocol.

The incorporation of foam sclerotherapy was based onreports describing a broad international experience. Feasi-bility, effectiveness, efficacy4,6,17-19,21,23-29,31,34,37-39 andcomparative5,7,10,12-16,20,22,40-41 studies have been con-ducted in many countries and distinct continents. Thetransition from air to a physiologic gas has been supportedby the experience of Cabrera in Spain and by the interna-tional trials conducted by Provensis to test the physiologicmix Varisolve.3-7,18,23,29 In particular, Cabrera et al haveinjected larger amounts of physiological foam than thoserecommended by consensus based primarily on air-basedfoams.11 This study and our general experience are inaccordance with the injection of large volumes of foam.The events described were not directly related to the vol-ume of foam used. Only dry cough with air-based foam hada relation to foam dosage.

Side effects, adverse reactions, and minor or majorcomplications following foam sclerotherapy have been lim-ited in incidence and in duration. Different objectives,varying descriptions, and low incidence make it difficult toanalyze the information presented in the literature. Mostpublished reports described experiences with air-basedfoam. The following is a summary of observations similar tothe ones described in this work relating adverse events.

Adverse events: air-based foam. Visual disturbancesor scotomas have been a major focus of attention whenevaluating adverse events following foam sclerotherapy.Frequencies of occurrence, as mentioned in the litera-ture, varied from 0% to 6% following air-based foaminjection.15-17,19,21-22,24,26,34,37-38 Most publicationsindicated a frequency of less than 1%. Our prospectiveevaluation showed frequencies of occurrence of 8% forair- and 3% for CO2-based foams. The prospective natureof this study may have resulted in slightly higher find-ings.

Respiratory difficulties were not reported by our pa-tients in the air-based foam series, with only one patienthaving such problems after CO2-based foam injection.Pulmonary embolism is a concern, but its occurrence is rareand may relate more to other patient conditions than to theprocedure itself. Pascarella et al reported one case of pul-monary embolism occurring 4 months after the proce-dure.25

Forlee et al reported one case of stroke following vari-cose vein foam injection sclerotherapy.35 The patient hadan undiagnosed, very large, patent foramen ovale. Theopinion of international experts was that this single episodeshould not justify presclerotherapy clearance for patentforamen ovale or right-to-left shunts.42 Thousands of pa-tients with patent foramen ovale have been treated withfoam sclerotherapy without adverse events.34 Serious inci-dents may occur not only after foam sclerotherapy but alsoafter liquid sclerotherapy or surgical stripping as attested bythe reports on cardiac arrest and brain infarct.32,33 Thepracticing phlebologist should have protocols to deal withsuch a serious adverse event if one occurs.

Other events included in this investigation have been

rarely reported. A prospective study like ours, focusing on

JOURNAL OF VASCULAR SURGERYApril 2008834 Morrison et al

such events, may have frequencies higher than retrospectiveevaluations. The following are some comments about theevents listed in Table I.

Chest thightness was mentioned by Bergan et al as anadverse event; only a few of their patients described thiscondition.26 In our studies, chest thightness occurred witha significant frequency when air-based foam was used, 18%,and decreased significantly to 3% when CO2-based foamwas administered.

Dry cough was also mentioned by Bergan et al as anadverse event that occurred infrequently.26 Our study alsohad a higher frequency with air-based foam, 16%, decreas-ing to less than 2% with CO2-based foam.

Dizziness apparently has not been formally reported orinvestigated in a prospective manner. Alos et al indicatedthat none of their patients had dizziness.15 We found asignificant frequency of 12% with air-based foam that de-creased to 3% with CO2-based foam.

Nausea also has not been formally reported after aprospective evaluation. In our studies, the frequency ofnausea was relatively small and the decrease with the changein gas was not statistically significant.

Metalic or medicinal taste has not been formally evalu-ated or described. This event occurred twice after CO2-based foam injection, and, perhaps, should not influencedaily practice.

Paresthesia was also mentioned by Bergan et al asoccurring infrequently.26 In our evaluation, no patientreported paresthesia. Minimizing this event seems to be amajor advantage of foam-based chemical ablation whencompared with surgical stripping.

Other events have been described in the literature.True migraine or occular migraine was experienced by acouple of patients as described by Bergan et al.26 Transientconfusion had a frequency of about 1% according to Frulliniet al.19 Neurasthenia has been reported but with an insig-nificant frequency.21,24

The frequencies of occurrence of tissue necrosis havebeen reported as less than 2%.19,26,31 Tissue necrosis can beinsignificant in general, but it can have a significant inci-dence when varicose vein patients are excluded from thecalculation.26-30 Tissue necrosis occurred in about 1% ofpatients with skin changes and ulcers (CEAP clinical classi-fication C4 to C6), perhaps between 1% and 10% for a smallgroup of patients with open ulcers, or in close to 17% (1/6)in patients with angiomata. The later statistics are affectedby the small number of cases but raise the concept thatadverse event frequencies are dependent on the conditionsof the patients. Improved care must take into considerationthe frequency of events in very specific populations.

Venous thrombosis has been reported as expected. Thisadverse event can be separated in four major groups: (1)extensive deep venous thrombosis involving the femoro-popliteal veins; (2) thrombus extension into a small seg-ment of the common femoral or popliteal vein; (3) deepvenous thrombosis limited to calf veins; and (4) superficial

thrombophlebitis.

Extensive femoropopliteal thrombosis following chem-ical ablation has been rare. A preprocedure clinical historyfor familial and individual, hypercoagulable states, mostlikely, can minimize occurrences of this event.2 Extensionof thrombosis to the femoral or popliteal veins has beendescribed by several authors with an incidence of less than2%.16,19,37-39 Calf thrombosis limited to the posterior tibialor gastrocnemius veins has also been described with anincidence varying from near 0% to 4%.16,21,24-26,38 Suchfrequencies may also be dependent on the overall conditionof the patient and may be lower for patients with simplevaricose veins compared with patients with advancedchronic venous insufficiency.

Superficial thrombophlebitis, in principle, should beconsidered a direct consequence of the treatment. Thisfinding, however, is considered an adverse event if throm-bophlebitis extends beyond the region treated or if theinflammation is significantly worse than commonly ex-pected. Therefore, one should expect the reports to behighly variable and subject to individual interpretation.Incidences have been reported varying from 1% to 11%, oreven 33% for a specific population with reflux starting in theveins at the groin.14,17,19,31,37-38 Significant inflammationhas occurred with an incidence close to 5%.10,21,24

Hamel-Desnos described the incidence of hematomasat about 2%.10 Frequency of telangiectatic matting wasinsignificant according to Barrett et al.21,24

Adverse events – CO2-based foam. CO2 has notbeen used as much as air and the literature on adverseevents reflects such a pattern. The experience of Cabrera’sgroup and findings of the Provensis international researchare summarized below.

Transient visual disturbances and dry cough were de-scribed by two patients each for an incidence of less than2%.23 This report indicates usage of 20 to 30 mL of foam totreat the saphenous vein. The mixed gas used was primarilyCO2 and O2. Coughing, inflammation, and photopsia werereported in 2001.18

Treatment of venous malformations was performedwith volumes up to 80 mL.29 Skin necrosis was an adverseevent in 6% of the patients. An injection in a small arterialbranch was reported. Deep venous thrombosis, pulmonaryembolism, or neurological symptoms did not occur asmentioned in the 2000, 2001, and 2003 reports.4,18,29

Apparently CO2-based foam does not increase the riskof deep venous thrombosis. Anecdotal descriptions ofthrombosis in calf veins near treatment sites have beenmentioned. Superficial thrombosis has also been reported.The incidence of significant superficial thrombophlebitiswas 10%.23

The Varisolve report of 2006 described mild contusion,skin discoloration and limb pain.7 A very small proportionof transient paresthesias, or visual and speech disordersoccurred with Varisolve. Cardiopulmonary adverse eventssuggestive of gas embolism did not occur. Incidence ofheadache was similar for the Varisolve vs saphenous strip-ping comparison, or for the Varisolve vs other sclerotherapy

comparison. In total, however, the frequency of headaches

JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Morrison et al 835

was highest after surgery. Deep venous thrombosis oc-curred in 4.5% of the Varisolve patients in the surgerycomparison group and 0.8% in the sclerotherapy compari-son group. This adverse event did not occur after changesin protocol requiring less volume and compression whenfoam approached the saphenofemoral or popliteal junc-tions.

In summary, this cohort study evaluated a physiologicalgas alternative for endovenous chemical ablation. Compar-ison with historical control was performed in a stable clin-ical population treated with similar volumes of foam forsimilar varicose vein problems. Heart rate, respiratory rate,electrocardiogram, and pulse oximetry were not physiolog-ically affected by UGFS. This report confirmed the lowincidence of visual disturbances and respiratory distress thatmay occur after UGFS. It documented a significant de-crease in chest thightness, dry cough, and dizziness whencarbon dioxide replaced air. For volumes greater than 15mL, incidence of complications or side effects were notdirectly related to volume of foam injected; compared withair, use of CO2 resulted in a three-to fourfold decrease inchest tightness, dry cough, and/or visual disturbance. Useof a physiological gas can minimize short-term side effectsassociated with foam sclerotherapy of superficial veins ofthe lower extremity.

The authors thank Kathy Melfy, RN, BSN, ChristineHall, RN, BSN, Pam Norris RN, BSN, Sharon Olbert, RN,BSN, Barbara Deusterman, RN, BSN, Ellen Allen, RN,Janice Moreno, RN, BSN, CVN, Shannon Levin, RVT,and Kristin Hansen, RVT, RDCS.

AUTHOR CONTRIBUTIONS

Conception and design: NM, CR, JM, EJ, DN, TM, SS-CAnalysis and interpretation: NM, EJ, SS-CData collection: NM, DN, CR, JM, EJ, TMWriting the article: SS-CCritical revision of the article: NMFinal approval of the article: SS-C, NMStatistical analysis: SS-C (initially, a statistician was con-

sulted)Obtained funding: NM, TM, CR, JM, DNOverall responsibility: SS-C, NM

REFERENCES

1. Bergan JJ, editor. The vein book. Burlington, MA: Elsevier AcademicPress; 2007.

2. Goldman MP, Bergan JJ, Guex JJ, Partsch H, Ramelet AA. Sclerother-apy: Treatment of varicose and telangiectatic leg veins. 4th ed. Philadel-phia, PA: Mosby-Elsevier; 2007.

3. Cabrera J, Cabrera J Jr. Nuevo método de esclerosis en las varicestronculares. Patologia Vasculares 1995;4:55-73.

4. Cabrera J, Cabrera J Jr, Garcia-Olmedo MA. Treatment of varicose longsaphenous veins with sclerosant in microfoam form: long-term out-comes. Phlebology 2000;15:19-23.

5. Eckmann DM, Kobayashi S, Li M. Microvascular embolization follow-ing polidocanol microfoam sclerosant administration. Dermatol Surg2005;31:636-43.

6. Redondo P, Cabrera J. Microfoam sclerotherapy. Semin Cutan Med

Surg 2005;24:175-83.

7. Wright D, Gobin JP, Bradbury AW, Coleridge-Smith P, Spoelstra H,Berridge D, et al. VarisolveR polidocanol microfoam compared withsurgery or sclerotherapy in the management of varicose veins in thepresence of trunk vein incompetence: European randomized controlledtrial. Phlebology 2006;21:180-90.

8. Wollman JC. The history of sclerosing foams. Dermatol Surg 2004;30:694-703.

9. Hamel-Desnos C, Ouvry P, Benigni JP, Boitelle G, Schadek M, DesnosP, Allaert FA. Comparison of 1% and 3% polidocanol foam in ultrasoundguided sclerotherapy of the great saphenous vein: a randomized, dou-ble-blind trial with 2 year follow-up. “The 3/1 study.” Eur J VascEndovasc Surg 2007;34:723-9.

10. Hamel-Desnos C, Desnos P, Wollmann JC, Ouvry P, Mako S, AllaertFA. Evaluation of the efficacy of polidocanol in the form of foamcompared with liquid form in sclerotherapy of the greater saphenousvein: Initial results. Dermatol Surg 2003;29:1170-5.

11. Brew FX, Guggenbichler S. European consensus meeting on foamsclerotherapy, April, 4-6, 2003, Tegernsee, Germany. Dermatol Surg2004;30:709-17.

12. Yamaki T, Nozaki M, Iwasaka S. Comparative study of duplex-guidedfoam sclerotherapy and duplex-guided liquid sclerotherapy for the treat-ment of superficial venous insufficiency. Dermatol Surg 2004;30:718-22.

13. Coleridge Smith P. Saphenous ablation: sclerosant or sclerofoam? Se-min Vasc Surg 2005;18:19-24.

14. Bountouroglou DG, Azzam M, Kakkos SK, Pathmarajah M, Young P,Geroulakos G. Ultrasound-guided foam sclerotherapy combined withsapheno-femoral ligation compared to surgical treatment of varicoseveins: early results of a randomized controlled trial. Eur J Vasc EndovascSurg 2006;31:93-100.

15. Alòs J, Carreño P, López JA, Estadella B, Serra-Prat M, Marinel-Lo J.Efficacy and safety of sclerotherapy using polidocanol foam: a controlledclinical trial. Eur J Vasc Endovasc Surg 2006;31:101-7.

16. Myers KA, Jolley D, Clugh A, Kirwan J. Outcome of ultrasound-guidedsclerotherapy for varicose veins: medium-term results assessed by ultra-sound surveillance. Eur J Vasc Endovasc Surg 2007;33:116-21.

17. Tessari L, Cavezzi A, Frullini A. Preliminary experience with a newsclerosing foam in the treatment of varicose veins. Dermatol Surg2001;27:58-60.

18. Cabrera J, Cabrera J Jr, Garcia-Olmedo MA. Sclerosants in microfoam.A new approach in angiology. Int Angiol 2001;20:322-9.

19. Frullini A, Cavezzi A. Sclerosing foam in the treatment of varicose veinsand telangiectases: history and analysis of safety and complications.Dermatol Surg 2002;28:11-5.

20. Belcaro G, Cesarone MR, Di Renzo A, Brandolini R, Coen L, Acerbi G,et al. Foam-sclerotherapy, surgery, sclerotherapy, and combined treat-ment for varicose veins: a 10-year, prospective, randomized, controlledtrial (VEDIGO trial). Angiology 2003;54:307-15.

21. Barrett JM, Allen B, Ockelford A, Goldman MP. Microfoam ultrasound-guided sclerotherapy of varicose veins in 100 legs. Dermatol Surg 2004;30:6-12.

22. Kern P, Ramelet AA, Wutschert R, Bounameaux H, Hayoz D. Single-blind, randomized study comparing chromated glycerin, polidocanolsolution, and polidocanol foam for treament of telangiectatic leg veins.Dermatol Surg 2004;30:367:72.

23. Cabrera J, Redondo P, Becerra A, Garrido C, Cabrera J Jr, Garcia-Olmedo MA, et al. Ultrasound guided injection of polidocanol micro-foam in the management of venous leg ulcers. Arch Dermatol 2004;140:667-73.

24. Barrett JM, Allen B, Ockelford A, Goldman MP. Microfoam ultrasound-guided sclerotherapy treatment for varicose veins in a subgroup withdiameters at the junction of 10 mm or greater compared with a subgroupof less than 10 mm. Dermatol Surg 2004;30:1386-90.

25. Pascarella L, Bergan JJ, Mekenas LV. Severe chronic venous insuffi-ciency treated by foamed sclerosant. Ann Vasc Surg 2006;20:83-91.

26. Bergan J, Pascarella L, Mekenas L. Venous disorders: treatment withsclerosant foam. J Cardiovasc Surg (Torino) 2006;47:9-18.

27. Darke SG, Baker SJ. Ultrasound-guided foam sclerotherapy for thetreatment of varicose veins. Br J Surg 2006;93:969-74.

28. Smith PC. Chronic venous disease treated by ultrasound guided foam

sclerotherapy. Eur J Vasc Endovasc Surg 2006;32:577-83.

JOURNAL OF VASCULAR SURGERYApril 2008836 Morrison et al

29. Cabrera J, Cabrera J Jr, Garcia-Olmedo MA, Redondo P. Treatment ofvenous malformations with sclerosant in microfoam form. Arch Derma-tol 2003;139:1409-16.

30. Pascarella L, Bergan JJ, Yamada C, Mekenas L. Venous angiomata:treatment with sclerosant foam. Ann Vasc Surg 2005;19:457-64.

31. Kakkos SK, Bountouroglou DG, Azzam M, Kalodiki E, DescalopoulosM, Geroulakos G. Effectiveness and safety of ultrasound-guided foamsclerotherapy for recurrent varicose veins: immediate results. J EndovascTher 2006;13:357-64.

32. Harzheim M, Becher H, Klockgether T. Brain infarct from a paradox-ical embolism following a varices operation. Dtsch Med Wochenschr2000;125:794-6.

33. Marrocco-Trischitta MM, Guerrini P, Abeni D, Stillo F. Reversiblecardiac arrest after polidocanol sclerotherapy of peripheral venous mal-formation. Dermatol Surg 2002;28:153-5.

34. Guex JJ, Allaert FA, Gillet JL, Chleir F. Immediate and midtermcomplications of sclerotherapy: report of a prospective multicenterregistry of 12,173 sclerotherapy sessions. Dermatol Surg 2005;31:123-8.

35. Forlee MV, Grouden M, Moore DJ, Shanik G. Stroke after varicose veinfoam injection sclerotherapy. J Vasc Surg 2006;43:162-4.

36. Hansen K, Morrison N, Neuhardt DL, Salles-Cunha SX. Transthoracicechocardiogram and transcranial doppler detection of emboli after foam

37. Cavezzi A, Frullini A. The role of sclerosant foam in ultrasound guidedsclerotherapy of the saphenous veins and of recurrent varicose veins.Aust NZ J Phlebol 1999;3:49-50.

38. Smith PC. Chronic venous disease treated by ultrasound guided foamsclerotherapy. Eur J Vasc Endovas Surg 2006;32:577-83.

39. Creton D, Uhl JF. Foam sclerotherapy combined with surgical treat-ment for recurrent varicose veins: short-term results. Eur J Vasc Endo-vasc Surg 2007;33:619-24.

40. Ceulen RP, Bullens-Goessens YI, Pi-Van de Venne SJ, Nelemans PJ,Veraart JC, Sommer A. Outcomes and side effects of duplex-guidedsclerotherapy in the treatment of great saphenous veins with 1% vs 3%polidocanol foam: results of a randomized controlled trial with 1-yearfollow-up. Dermatol Surg 2007;33:276-81.

41. Rao J, Wildemore JK, Goldman MP. Double-blind prospective com-parative trial between foamed and liquid polidocanol and sodiumtetradecyl sulfate in the treatment of varicose and telangiectatic legveins. Dermatol Surg 2005;31:631-5.

42. Morrison N, Cavezzi A, Bergan J, Partsch H. Letter to the editorregarding “Stroke after varicose vein foam injection sclerotherapy”.J Vasc Surg 2007;44:224-5.

sclerotherapy of leg veins. J Vasc Ultrasound 2007;31:213-6. Submitted Jul 31, 2007; accepted Nov 10, 2007.