Article Text

Download PDFPDF

Original research
Percutaneous sclerotherapy with ethanolamine oleate for venous malformations of the head and neck
  1. Matthew David Alexander1,
  2. Ryan A McTaggart2,
  3. Omar A Choudhri2,
  4. Mary L Marcellus2,
  5. Huy M Do2
  1. 1Department of Radiology, Santa Clara Valley Medical Center, San Jose, California, USA
  2. 2Departments of Radiology and Neurosurgery, Stanford University Medical Center, Stanford, California, USA
  1. Correspondence to Dr H M Do, Departments of Radiology and Neurosurgery, Stanford University Medical Center, 300 Pasteur Dr, S047 MC 5105, Stanford, CA 94305, USA; huymdo{at}


Introduction Venous malformations frequently occur in the head and neck, and they can require treatment for a variety of reasons. Among multiple therapeutic approaches employed, percutaneous sclerotherapy has become one of the most commonly used treatments, with numerous sclerosants successfully utilized. Ethanolamine oleate has approval from the Food and Drug Administration for sclerosis of esophageal varices, and is used by some practitioners for the treatment of venous malformations. This study reports single center results of percutaneous sclerotherapy with ethanolamine oleate to treat venous malformations of the head and neck.

Materials and methods Prospectively maintained procedural records were retrospectively reviewed to identify all patients with venous malformations who underwent percutaneous sclerotherapy. The Mulliken and Glowacki classification was used to diagnose venous malformations. Medical records and images were reviewed to record demographic information, lesion characteristics, treatment sessions, and clinical and imaging response. Quantitative volumetric analysis was conducted to augment commonly used poorly reproducible subjective outcome measures. Response was assessed after each session and completion of all percutaneous treatment. A χ2 analysis was performed to evaluate the effects of the above described characteristics on outcomes.

Results 52 interventions were performed for lesions in 26 patients. No complications occurred following any procedures. Response to individual sessions was categorized as excellent following two (3.8%) sessions, good following 45 (86.5%), and fair following four (7.7%) session. No sessions resulted in poor responses. Final results were excellent in two patients (7.7%), good in 22 (84.6%), and fair in two (7.7%). Average lesion volume reduction was 39% following each session, and 61% after treatment completion. Periorbital lesions were significantly less likely than lesions located elsewhere to have good or excellent outcomes. No other lesion or demographic features affected outcomes.

Conclusions Percutaneous sclerotherapy with ethanolamine oleate appears to be safe and effective for the treatment of venous malformations and should be considered when treating these complex lesions. The efficacy of this agent appears to match or exceed that of other sclerosants used for such treatment, and further investigation in prospective controlled research is warranted.

  • Vascular Malformation
  • Technique
View Full Text

Statistics from


Venous malformations are the most common vascular malformations and frequently affect the head and neck.1 These lesions often require treatment to address airway compromise, infection risk, bleeding, or cosmesis.2–4 Treatment of these lesions can be performed with numerous modalities, most frequently surgery, percutaneous sclerotherapy, or both.3 ,5 Use of numerous sclerosants has been reported in clinical studies, yet no clearly superior agent has been identified.6 No perfect sclerosant exists, and those compounds yielding better results often have higher rates of side effects and complications. Ethanolamine oleate has an established role as a sclerosant for lesions elsewhere in the body, most notably for esophageal varices requiring sclerotherapy, for which it is the only agent with Food and Drug Administration (FDA) approval. However, little research has been performed to assess the use of this agent for the treatment of venous malformations of the head and neck.7 This current study presents a single center experience using ethanolamine oleate for such lesions.

Materials and methods

Prospectively maintained procedure records were retrospectively reviewed to identify all patients with vascular malformations who underwent percutaneous sclerotherapy with ethanolamine oleate in the interventional neuroradiology department of a major academic medical center. Medical records and images were reviewed to record demographic information, clinical and imaging characteristics of the lesions, treatment sessions, and clinical response.

All interventions were performed according to the same departmental protocol. All patients were treated for pain or discomfort, although many patients also expressed cosmetic concerns. Treatment generally occurred under moderate sedation. General endotracheal anesthesia was performed for infants and young children who could not tolerate the procedure under moderate sedation. The lesion was palpated with the aim of selecting a dominant pocket seen on pretreatment imaging. The sclerosant was mixed with an equal volume of contrast for intralesional instillation. Lesions were cannulated with a 21 gauge butterfly needle and advanced until bloody aspirate was noted. The lesion was then gently infiltrated with the sclerosant under fluoroscopy. Sclerosant was injected until the lesion became less compressible and resistance was felt in the syringe. The needle was removed, and a bandage was left on the skin. If additional pockets remained that were amenable to treatment, the process was repeated. Patients were admitted for observation and pain control overnight following the procedures (figure 1).

Figure 1

Pretreatment T1 weighted MRI with fat saturation (A) and T2 weighted MRI (B) of a patient in his/her forties demonstrating a right premaxillary venous malformation. Fluoroscopy (C) demonstrates positioning of a needle for percutaneous sclerotherapy. Digital subtraction angiography (D) demonstrates sclerosant and contrast filling the lesion. CT reformatted images (E) at completion of the procedure demonstrate the sclerosant and fluid in the lesion, the size and shape of which matches that seen on pretreatment imaging. Six months after treatment, T1 weighted MRI with fat saturation (F) and T2 weighted MRI (G) demonstrate near resolution of the lesion. The patient was pain free and happy with the cosmetic result.

All vascular lesions were assessed according to the classification scheme of Mulliken and Glowacki.8 Locations were classified as neck, oral, periorbital, or elsewhere in the face. Response to treatment was categorized as excellent, good, fair, or poor, based on physical examination. This approach was modeled after the most commonly used measures used in prior series evaluating percutaneous sclerotherapy.1 ,7 ,9 ,10 Lesions with no remaining visible abnormality were considered to have had an excellent outcome. Lesions that were visibly smaller and subjectively less than or greater than half their original sizes were considered to have had good or fair outcomes, respectively. A poor outcome was designated following no change or lesion growth. Values were assigned based on reviews of history and physical examination reports, procedure reports, and clinic follow-up notes. All patients were seen by the same attending interventionalist who performed all treatment, as well as several interventional neuroradiology fellows and a nurse practitioner participating in patient care. Thus the same practitioner participated and reported finding prior to, during, and following treatment.

All patients were evaluated with contrast enhanced MRI before and after all treatment sessions. Lesion volume was calculated using STIR MR images. If STIR sequences were not obtained, calculation was based on T2 weighted MR images. Calculation was performed with Aquarius iNtuition V.4.4.6 (Terarecon, Foster City, California, USA) using methods described elsewhere.11 Any complications were noted. Treatment related effects, such as swelling, pain, tenderness, and low grade fever, were considered sequelae of the intervention rather than frank complications like blistering, ulceration, nerve injury, or cellulitis. A χ2 analysis was performed to evaluate effects of the above described characteristics on outcomes.


Fifty-two interventions were performed for lesions in 26 patients. The same attending interventionalist performed all sessions. Demographics and lesion characteristics are summarized in table 1. Mean number of sessions was 2.0, with a single session performed in 11 (42.3%) patients. Mean follow-up was 21.8 months (range 29 days to 6.7 years). No complications occurred following any procedures. Response to individual sessions was categorized as excellent following two (3.8%) sessions, good following 45 (86.5%), and fair following four (7.7%) sessions. No sessions resulted in poor responses. Final results were excellent in two patients (7.7%), good in 22 (84.6%), and fair in two (7.7%). Average lesion volume reduction was 39% following each session, and 61% after treatment completion. Periorbital lesions were less likely than lesions located elsewhere to have good or excellent outcomes (χ2 (3, n=47)=9.730; p=0.021). No other demographic variable or lesion characteristic significantly affected outcome, including patient age, lesion location, or lesion volume. Lesion volumes, patient age, and response to completed treatment are summarized by lesion location in table 2.

Table 1
Table 2


Venous malformations are the most common type of vascular malformation and have a propensity for the head and neck.1 These lesions can be treated in many ways, including surgery, cryoablation, electrocautery, laser photocoagulation, and compressive wrapping.1 These lesions contain thin walls lined with endothelium, a feature that makes them amenable to sclerotherapy. This minimally invasive procedure has become more widely used in the last several decades. At most high volume centers, current practice typically includes primary excision for small lesions and sclerotherapy for larger lesions, after which excision may be performed if appropriate.1

When percutaneous sclerotherapy is performed, there is much variability between practitioners and medical centers with respect to the agents employed. In the USA there are no compounds with FDA approval for such treatment, so all administration is performed off label.6 Additionally, other agents are used outside the USA that do not have FDA approval for any use.6

Ethanol has earned widespread use for sclerotherapy due to its pronounced effect and good results following a small number of treatment sessions.1 It causes dehydration and eventual sloughing of endothelium.12 This very powerful sclerosant necessitates general endotracheal anesthesia during treatment due to intense pain caused by injection.6

Sotradecol is a detergent that is most commonly used for lower extremity varices although it has also been used for vascular malformations of the head and neck.6 Some interventionalists select this agent over ethanol due to its milder activity.1 While it induces less robust responses from treated lesions and therefore tends to cause fewer complications, it is widely considered less efficacious and typically requires more treatment sessions than ethanol.1 Its mechanism of action is believed to involve decreased surface tension via displacement of membrane lipoproteins that in turn leads to increased permeability.6 ,13 This effect has been exploited by some practitioners who inject ethanol following sotradecol in order to potentiate the effect of the former.6

Doxycycline is an antibiotic that can also be used as a sclerosant, most likely due to inhibition of metalloproteases and vascular endothelial growth factor.6 It has demonstrated benefit as a sclerosant for vascular malformations, particularly lymphatic malformations with microcystic components.14 Bleomycin is a chemotherapeutic antibiotic that can be used for sclerotherapy. Its mechanism of action is uncertain but most likely involves induction of exuberant inflammation by endothelial cells.6

Sodium amidotrizoate (Ethibloc) is a combination agent that induces thrombosis, necrosis, and fibrosis.6 OK-432 (Picibanil) is an anticancer agent manufactured from a derivative of streptococcus pyogenes.6 Neither of these agents have FDA approval but are commonly used outside the USA. Less commonly used agents that have been described in sclerotherapy series include polidocinole, acetic acid, sodium morrhuate, and fibrin.6

Ethanolamine oleate is an emulsion of fatty acids that damages endothelium by inducing thrombosis.15 It is the only sclerosant granted FDA approval for sclerotherapy of esophageal varices. Given that varices and venous malformations are both lined with vascular endothelium, ethanolamine oleate has been used off label for sclerotherapy of the latter. It stores well with a long shelf life, causes little discomfort on injection, and results in few systemic complications.7 The maximum allowable dose of undiluted ethanolamine oleate is 20 mL for intravenous administration. Considering the extravascular distribution and one to one dilution with contrast, this limitation far exceeds any practical administration volume with the above described technique.

In the current series, ethanolamine oleate induced good or excellent results after completion of treatment in 92.3% of patients. By comparison, studies have demonstrated efficacy rates of 75–95% for treatment of head and neck venous malformations with ethanol.12 Such a response occurred in 75% of patients following treatment with sotradecol and 84% after treatment with bleomycin.1 ,16 In a head to head comparison of ethanol and bleomycin, ethanol was more effective, and ethanol achieved the desired results in fewer sessions.12

By nature of their intended activity and effects on endothelium, all sclerosants cause an inflammatory response in treated tissue. In the hours and days following treatment, swelling, pain, erythema, and low grade fevers are common.6 Indeed, more pronounced initial inflammation typically portends better response to treatment.12 These side effects of treatment should be distinguished from true complications, which can range from self-limited blistering, ulceration, or cranial nerve palsy to disfiguring scarring, drainage, and formation of permanent cutaneous fistulas.6 Additionally, intense pain from injection requiring deep sedation or general anesthesia occurs with injection of several agents, most notably ethanol and doxycycline.1 ,6 The risks inherent in general anesthesia must thus be considered when weighing the risks and benefits of such sclerosants. Additionally, systemic effects can occur due to intravasation of sclerosant, most commonly manifest as intoxication from systemic exposure to ethanol.6 ,16 This phenomenon is particularly important to prevent during treatment of very young children. Bleomycin also carries theoretical risk from systemic exposure, with well described risks of pulmonary fibrosis from the oncology literature. However, no cases of pulmonary fibrosis have been reported following sclerotherapy.6

In this current study no complications occurred. In the only large series examining ethanolamine oleate for sclerotherapy of vascular malformations, Kaji et al7 reported complications in 51% of cases, none of which was severe. However, the most frequent complication was hemolytic hemoglobinuria, a complication that was not encountered in this current series.7 The higher complication rate reported by Kaji et al7 may relate to the large number of patients treated with embolosclerotherapy that caused systemic exposure to the sclerosant. Additionally, the blistering and ulceration reported in that series did not occur in this current series.

While numerous sclerosants have demonstrated clinical efficacy, accurate comparison between agents proves difficult due to variability during assessment. Some studies have relied exclusively on clinical examination findings while others compile answers from patients or caretakers on subjective questionnaires.1 Such approaches have demonstrated only modest agreement with each other.1 Additionally, some studies emphasize cosmetic appearance while others emphasize changes in symptoms.1 ,7 ,9 ,10 Other investigators have attempted to incorporate quantitative data although methods have varied.7 Additionally, responses considered successful vary between studies. For instance, some research considers fair response grades when calculating total response rate, while other research only includes excellent response or complete resolution.1 ,7 ,9 ,10 While efforts were made in this current study to compare subjective measures in a way that most closely matched methods used in other studies, comparison is limited without standardized evaluation metrics. Elsewhere, we described a novel measurement approach using volumetric analysis based on three-dimensional image reformats.11 In this current series, this method demonstrated average lesion volume reduction by 39% after each session and 61% on completion of treatment. Unfortunately, these numbers are of limited value given the lack of direct comparisons with other series. Future investigations of sclerosants should employ similar methods to augment comparison.

This current study is limited by its retrospective design and lack of standardized clinical data collection at points of treatment and follow-up.


Percutaneous sclerotherapy of venous malformations of the head and neck can be safely performed using ethanolamine oleate. Treatment with this agent is well tolerated under moderate sedation and caused no complications in this study. However, its milder activity may reduce its effectiveness and necessitate an increased number of treatment sessions. Further investigation is needed for both ethanolamine oleate and other agents using standardized metrics in order to more effectively balance the risks and benefits when treating these lesions with percutaneous sclerotherapy.


View Abstract


  • This study was presented at the SNIS 2013 Annual Meeting.

  • Contributors MDA performed the data collection, data analysis, and authored the manuscript. RAM, OAC, and HMD participated in patient care, performed the data analysis, and edited the manuscript. MLM participated in patient care and data collection.

  • Competing interests None.

  • Ethics approval The study was approved by the institutional review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.