Predictors of response to percutaneous ethanol sclerotherapy (PES) in patients with venous malformations: Analysis of patient self-assessment and imaging

Article Outline

• Abstract
• Methods
  • Patients
  • Preprocedural considerations
  • Procedural details of PES
  • PES outcome assessment
  • Statistical analyses
• Results
• Discussion
• Author contributions
• Appendix, online only. Patients self assessment questionnaires
• References
• Copyright

Background

Percutaneous ethanol sclerotherapy (PES) is the primary tool in the treatment of venous malformations (VM). However, PES has known serious complications. This study is aimed at identifying predictors of good response to PES in patients with VM to improve patient selection.

Methods

We performed a retrospective, cross-sectional study of 158 VM patients (mean age, 14.3 years, male 42%) who underwent ethanol sclerotherapy at a specialized vascular malformation center. For clinical result assessment, patients or parents in pediatric patients answered questions on symptomatic, functional, and cosmetic improvement after PES. In each category, the possible choices were markedly improved, moderately improved, no change, moderately worse, or markedly worse compared with pretreatment status. A “good response” was defined as one or more areas of marked improvement on the self-assessment in conjunction with marked improvement on post-treatment images (≥30% decrease in maximal diameter of VM on magnetic resonance imaging [MRI] or ≥50% decrease in abnormal blood pool ratio on whole body blood pool scintigraphy [WBBPS] compared with pretreatment images). To determine predictors of a good response to PES, uni- and multivariate analysis were conducted on demographics (age, gender), clinical features of VM (location, size, depth of involved tissue, presence of associated lymphatic malformation, MRI findings; well-defined vs ill-defined margin, characteristics of venous drainage during PES) and treatment variables (number of PES sessions, maximal concentration and dosage of ethanol used in PES, adjuvant therapy).

Results

Symptomatic, functional, and cosmetic improvement was 28%, 27%, and 34%, respectively, based on patient questionnaires. Based on imaging studies, 42 patients (27%) had markedly improvement. Composite outcome combining questionnaire results and imaging study showed that 16% of patients had a “good response”. On multivariate analysis, female gender (odds ratio [OR]: 4.49, 95% confidence interval [CI]: 1.24-16.28), no or delayed visualization of drainage vein (OR: 9.22, 95% CI: 1.79-47.51), and a well-defined margin on MRI (OR: 13.38, 95% CI: 2.84-63.12) were independent predictors of “good response” to PES.

Conclusions

PES should be performed in selected patients in order to obtain the best outcomes and minimize complications. No or delayed visualization of drainage vein on initial direct puncture venogram, a well-defined margin on MRI, and female gender were statistically significant predictors of a “good response” to PES and may be useful in selecting patients.

Venous malformations (VM) are the most common type of vascular malformation and account for 44% to 64% of all vascular malformation.¹, ² The clinical presentation of venous malformations (VM) vary in symptom, site, depth of involvement, and can be combined with other type of vascular malformations such as lymphatic malformation (LM) and capillary malformation (CM). For treatment of VM, sclerotherapy, embolization, and surgical excision have been reported.³ Sclerotherapy has become the mainstay of treatment while surgical excision alone has a limited role in most patients due to the potential for functional and cosmetic problems.¹, ⁴, ⁵, ⁶ With sclerotherapy, several agents are available. Of these, ethanol is the most common agent and its efficacy and potential complications are relatively well described.⁷, ⁸, ⁹

It is apparent that some VM lesion responds well to sclerotherapy while others do not. Reported outcomes of percutaneous ethanol sclerotherapy (PES) vary and are dependent on proper patient selection. Previous authors often used absolute VM size reduction or patient self-assessment to document success of VM treatment.¹⁰, ¹¹, ¹² There have been reports describing the use of magnetic resonance imaging (MRI) or phlebographic findings to assess treatment success.¹¹, ¹³ Only a few reports¹⁰ perform multivariate analysis to determine predictors of good sclerotherapy response. Such predictors would be important in helping to select the proper patients and minimize the side effects of unfruitful PES. This article attempts to determine predictors of good response to PES in patients with VM.

Back to Article Outline

Methods 

Patients 

We retrospectively reviewed a patient database in a specialized vascular malformation clinic from the period January 1995 to June 2008. Before conducting this study, Institutional Review Board approval was obtained.

Venous malformation (VM) was diagnosed based on clinical findings and imaging studies such as MRI, whole body blood pool scintigraphy (WBBPS) using Tc-99m red blood cells (Biad, Trionix Research Laboratory, Twinsburg, Ohio), and/or duplex ultrasonography. In addition to pure VM, predominantly venous defects, which were combined with LM or CM, were also included. Patients with an arterio-venous fistula component, whether gross or micro, were excluded from the study. Figure 1 demonstrates our treatment of VM patients. Among 1045 patients diagnosed with VM, 158 patients with data of treatment outcome assessment available were included in this study.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Treatment of patients with venous malformations (VM).

Preprocedural considerations 

In selecting candidates for PES, patients' age (pediatric or adult), body weight (BW), clinical symptoms and signs, location of VM (proximity to airway, eye, ear, joint), depth from skin (subcutaneous vs deep), and coexisting deep venous anomalies were considered. The indication for PES have previously been described elsewhere.¹ Our indications for PES in patients with VM were various according to its location of VM lesion. Patients often came to the hospital on a cosmetic reason when lesions involve the head and neck. Patients with VM involving the extremity or trunk sought medical care with symptom of pain or severe discomfort due to acute phlebitis, joint involvement, or chronic venous hypertension. Sometimes patients were treated for the bloody or serous discharge from the cutaneous lesion.

In patients with head and neck VM, an MRI was thoroughly reviewed to avoid airway obstruction or cranial nerve palsy after PES. For those patients with head and neck VM close to the upper airway, endotracheal intubation was maintained after PES treatment until the risk of airway obstruction safely diminished. Patients with VM at the angle of the mandible underwent facial nerve branch mapping on the skin using electric stimulation in order to avoid needle injury or direct ethanol injection into the facial nerve branches during PES. In patients older than 15 years of age, with large VM requiring large amount of ethanol (>0.5 mL/Kg of BW) or those with cardiopulmonary depression during previous PES, a pulmonary artery catheter was utilized to monitor pulmonary artery pressure (PAP).

Procedural details of PES 

All PES was performed under general anesthesia and fluoroscopy. A 21 gauge needle was used to percutaneously access the VM lesion. A measurement of 5 mL of contrast (iobitridol, Xenet 300; Guerbet, Cedex, France) was injected to observe the characteristics of the drainage vein. Some VM lesions showed immediate drainage of contrast while others had no or delayed drainage after the initial contrast injection (Fig 2). When drainage vein was immediately visualized, drainage vein compression was attempted manually or with a pneumatic cuff to increase exposure time of the sclerosing agent in the VM lesion. When manual compression was performed, direct compression over the drainage vein was attempted under fluoroscopy. When the pneumatic cuff compression was applied on the purpose of drainage vein blockage, a tourniquet was inflated to a pressure of 50 to 60 mm Hg. A proximal tourniquet was not routinely used due to the concern for pulmonary vasospasm or pulmonary embolism due to a sudden influx of ethanol into the pulmonary circulation upon sudden deflation of tourniquet.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Characteristics of venous drainage from venous malformations (VM) on a direct puncture venogram during PES. A and B, No or delayed visualization of drainage vein. C and D, Immediate visualization of drainage vein.

The sclerosing agent was 60% to 100% ethanol (upper limit of 100% ethanol <1 mL/Kg of BW/session). Pure ethanol was routinely used, except in cases where the lesion involved skin or located near a nerve or major vascular structure. In these cases, 60% to 90% ethanol was used to prevent skin necrosis, nerve damage, or deep vein thrombosis. Five to 10 minutes after ethanol injection, a venogram was performed to check for residual filling of the VM lesion. If present, a repeat session of PES was performed at least 6 weeks later to allow for healing and function of the local tissue.

After PES, patients who received large doses of ethanol (>0.5 mL/Kg of BW) were monitored in the ICU until PAP and other vital signs returned to pretreatment levels. For patients with VM located near the airway, tracheal intubation was our routine, and flexible laryngoscopy was used to assess the degree of edema prior to extubation. For patients who had severe swelling or tenderness, anti-inflammatory agents, steroids, and ice packs were used. In limbs lesions, a venous duplex scan was routinely performed the day after treatment to rule out deep venous thrombosis.

We made a protocol to check WBBPS on yearly basis for all patients whether the patients were treated or not in order to assess volume changes of VM lesion. MRI was followed when more detailed anatomic information was required before or after PES to establish a plan for the subsequent treatment.

PES outcome assessment 

To assess treatment outcomes of VM, we used postprocedure imaging (MRI and/or WBBPS) and patient self-assessment questionnaires that evaluated three areas: symptomatic, functional, and cosmetic outcomes (Appendix, online only). In each category, possible answer choices were markedly worse, moderately worse, no change, moderately improved, and markedly improved. In pediatric patients, parents were asked to complete the questionnaire.

Pre- and post-treatment image MRI and/or WBBPS were compared. Maximal VM diameter was measured with electronic calipers, and lesion-to-whole body blood pool count ratio was measured by a semiquantitative method on WBBPS. Marked improvement by imaging was defined as ≥30% decrease in maximal MRI diameter of VM or ≥50% decrease in abnormal blood pool ratio on WBBPS compared to pretreatment images (Fig 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Imaging studies showing “good response” to percutaneous ethanol sclerotherapy (PES). A and B, Pre- and post-PES MRIs showing diminished size of venous malformations (VM) lesion after PES. C and D, Pre- and post-PES WBBPSs showing diminished abnormal blood pool in the neck after PES.

A “good response” to PES was then defined as at least one area of marked improvement on the clinical assessment (symptomatic, cosmetic, or functional) and marked improvement on post-treatment images. Patients who did not have a “good response” were considered to have a “no-good response”.

Statistical analyses 

To determine predictors of good response to PES, uni- and multivariate analyses were conducted on demographic (age, gender) and clinical variables (location, size of VM, depth of tissue involve, presence of associated LM or CM, MRI findings, well-defined vs ill-defined margin, and characteristics of venous drainage during PES) and treatment variables (number of PES sessions, maximal concentration and total dosage of ethanol injected, and adjuvant therapy).

For univariate analysis, χ² test and Mann-Whitney test were used to compare variables between “good response” group and “no-good response” group. Multivariate analysis was performed with binary logistic regression model.

P values < .05 were considered to be statistically significant. Statistical analysis was performed with the SPSS software (SPSS version 12.0.1; SPSS Inc, Chicago, Ill).

Back to Article Outline

Results 

Patient demographic and clinical data is shown in Table I. A total of 654 PES sessions were performed in 158 patients with VM (median, 3.0 sessions/patient). Eighty percent of patients (127/158) required more than one PES session. The details of PES are shown in Table II. Regarding adjuvant treatment, 13 patients underwent local excision and five patients underwent surgical resection of a lateral embryonal vein after PES. In superficial VM lesions, there were 11 ethanolamine oleate sclerotherapy and 14 foam sclerotherapy with sodium tetradecil sulfate following PES.

Table I. Demographic and clinical data of VM patients

	N = 158 patients
Age at first PES, mean (range)	14.3±11.5(1-58)years
Gender, male	67(42%)
Type of VM
Pure VM	122(77%)
VM combined with LM	16(10%)
VM combined with CM	17(11%)
VM combined with LM and CM	3(2%)
Location of VM treated
Head and neck	66(42%)
Trunk	19(12%)
Extremity	73(46%)
Upper extremity	14(9%)
Lower extremity	59(37%)
Clinical presentation
Swelling or mass	93(59%)
Episode of pain or tenderness	94(60%)
Varicosity	75(48%)
Lateral embryonal vein	9(6%)
Skin discharge	3(2%)
Limb length discrepancy (>2 cm)	4(3%)
Size of VM on MRI:
Maximal diameter <5 cm	48(30%)
Maximal diameter ≥5 cm	110(70%)
Tissue involvement of VM
Only skin or subcutaneous tissue	29(18%)
To the skeletal muscle	105(66%)
To the bone or joint	24(15%)
MRI finding
Well-defined margin	29(18%)
Poorly-defined margin	129(82%)
Visualization of drainage vein on a direct puncture venogram
No or delayed	14(9%)
Immediate	144(91%)

PES, Percutaneous ethanol sclerotherapy; VM, venous malformations; LM, lymphatic malformation; CM, capillary malformation; MRI, magnetic resonance imaging.

Table II. Details of percutaneous ethanol sclerotherapy (PES) for VM patients

Details	No (%)
Number of PES/patient, mean	4.1(1-22)
≤5 sessions	123(78)
6-10 sessions	25(16)
>10 sessions	10(6)
Maximal concentration of ethanol used
< 80%	5(3)
80%-99%	19(12)
100%	134(85)
Amount of ethanol used/session, mL of 100% ethanol/Kg of BW, median/IQR	0.3/0.3
Adjuvant treatment with PES
Local excision of VM	13(8)
Surgical resection of lateral embryonal vein	5(3)
Use of ethanolamine oleate for skin VM lesion	11(7)
Foam sclerotherapy with sodium tetradecil sulfate	14(9)

PES, Percutaneous ethanol sclerotherapy; VM, venous malformations; BW, body weight; IQR, interquartile range.

Table III shows PES complications. There were a total of 20 patients with nerve palsies. By location, the most common site of nerve palsy was in the upper extremity. Five nerve palsies occurred in the 14 upper extremity VMs for a 36% palsy rate in the upper extremity, a significantly greater rate compared with head and neck, lower extremity or trunk VMs (P = .037, Fisher exact test): head and neck VMs, 9% (6/9); lower extremity VMs, 14% (8/59); trunk VMs, 5% (1/19). There were three incidences of permanent peroneal nerve palsy and 17 incidences of temporary palsy with a mean recovery time of 5.4 months (range 1 to 24 months). An 11-year-old female patient with a 5 cm VM lesion of the back in close proximity to the sixth to eighth thoracic spine (Fig 4) had a complication of transient spinal cord ischemia following PES with 13 mL of 100% ethanol. Twelve hours following PES injection, the patient developed bilateral lower extremity weakness, urinary incontinence, and paresthesia of the left medial foot. MRI showed high signal intensity and discontinuous enhancement in spinal cord at the level of T6 to T12 consistent with acute or subacute spinal cord ischemia. She was placed on steroids and symptoms resolved in 7 days.

Table III. Complications of percutaneous ethanol sclerotherapy (PES)

Complications	No (%)
Local complications
Skin necrosis	12(8)
Deep vein thrombosis	0
Nerve palsy	20(13)
Head and neck nerve	6(4)
Sensory	4
Motor and sensory nerve	2
Lower extremity nerve	8(5)
Motor nerve	1
Sensory nerve	2
Motor and sensory nerve	5
Upper extremity or finger nerve	5(3)
Motor nerve	0
Sensory nerve	2
Motor and sensory nerve	3
Spinal cord	1(1)
Systemic complications
Transient pulmonary hypertensiona	2(1)
Elevated liver enzyme	3(2)
Cardiac arrhythmia	1(1)
Hemoglobinuria	12(8)

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Pretreatment magnetic resonance imaging (MRI) (A), (B), and (C) and direct puncture venogram (D) during percutaneous ethanol sclerotherapy (PES) in a patient who developed transient spinal cord ischemic symptom.

Patient self-assessment questionnaire results are given in Table IV. Patients reported “marked improvement” in symptoms 28% (26/94), function 27% (19/71), and cosmesis 34% (53/158). “Marked improvement” was noted in at least one category in 41% (65/158) (Table IV).

Table IV. Outcomes of PES

Clinical outcomes assessed by questionnaire	Frequency (%)
Symptomatic improvement (n = 94)
Markedly worse	2(2)
Moderately worse	3(3)
No change	44(47)
Moderately improved	19(20)
Markedly improved	26(28)
Cosmetic improvement (n = 158)
Markedly worse	2(1)
Moderately worse	3(2)
No change	47(30)
Moderately improved	53(34)
Markedly improved	53(34)
Functional improvement (n = 71)
Markedly worse	2(3)
Moderately worse	5(7)
No change	25(35)
Moderately improved	20(28)
Markedly improved	19(27)

Post-treatment image compared with pretreatment image
Lesion-to-whole body blood pool count ratio with semi-quantitative method on WBBPS (n = 132)
Increase ≥10%	19(14)
No or minor (<10%) change	31(23)
Diminished ≥10% and <50%	55(42)
Diminished ≥50%	27(20)
Maximal diameter of VM on MRI (n = 123)
Increased ≥10%	22(18)
No or minor (<10%) change	57(46)
Decreased ≥10% and <30%	18(15)
Decreased ≥30%	26(21)

PES, Percutaneous ethanol sclerotherapy; VM, venous malformations; WBBPS, whole body blood pool scintigraphy; MRI, magnetic resonance imaging.

Post-treatment WBBPSs were available in 132 (84%) patients and 20% of patients showed “marked improvement”. Follow-up MRIs were available in 123 (78%) patients, and 21% of them revealed “marked improvement”. In total, 42 (27%) of 158 patients revealed “marked improvement” on follow-up imaging study. Combining questionnaire and imaging results, a “good response” was achieved in 16% (25/158) of patients after PES.

Table V demonstrates results of univariate analysis for “good response” to PES. Female gender, non-extremity VM, small size (<5 cm in diameter), poor drainage from VM during PES, and well-defined margins on MRI were significantly associated with a good response to PES.

Table V. Univariate analysis of variables to predict “good response” to PES for VM patients

Variables	Response to PES		P
	Good (n = 25, 16%)	No-good (n = 133, 84%)
Age			.554a
<20 years	20(80)	99(74)
≥20 years	5(20)	34(26)
Gender			.042a
Male	6(24)	61(46)
Female	19(76)	72(54)
Type of VM			.875a
Pure VM	19(76)	103(77)
Combined with LM or CM	6(24)	30(23)
Location of VM			.001a
Non-extremity	21(84)	64(48)
Extremities	4(16)	69(52)
Size of VM			.037a
<5 cm	12(48)	36(27)
≤5 cm	13(52)	97(73)
Visualization of drainage vein			.002b
No or delayed	7(28)	7(5)
Immediate	18(72)	126(95)
VM margin on MRI			<.001a
Well-defined	13(52)	16(12)
Poorly-defined	12(48)	117(88)
Involved tissue			1.000b
Skin or subcutaneous tissue only	4(16)	25(19)
Muscle, bone, or joint	21(84)	108(81)
Total session number of PES			.515b
1-5 sessions	21(84)	102(77)
6-10 sessions	4(16)	21(16)
11 or more sessions	0	10(7)
Maximal concentration of ethanol used during PES			.892b
<80%	1(4)	4(3)
80%-99%	3(12)	16(12)
100%	21(84)	113(85)
Amount of ethanol used/session, median/IQR			.904c
mL of 100% ethanol/Kg of BW	0.3/0.4	0.3/0.3
Adjuvant treatment with PES			.142a
Yes	3(12)	34(26)
No	22(88)	99(74)

PES, Percutaneous ethanol sclerotherapy; VM, venous malformations; LM, lymphatic malformation; CM, capillary malformation; BW, body weight; IQR, interquartile range; MRI, magnetic resonance imaging.

On multivariate analysis, female gender (odds ratio [OR]: 4.49, 95% confidence interval [CI]: 1.24-16.28), no or delayed visualization of drainage vein (OR: 9.22, 95% CI: 1.79-41.51), and a well-defined margin on MRI (OR: 13.38, 95% CI: 2.84-63.12) were independent predictors of good response to PES (Table VI).

Table VI. Multivariate analysis^a of variables to predict “good response” to PES for VM patients

PES, Percutaneous ethanol sclerotherapy; VM, venous malformations; LM, lymphatic malformation; CM, capillary malformation; BW, body weight; MRI, magnetic resonance imaging.

Back to Article Outline

Discussion 

Clinical features of VM ranges from asymptomatic varicosities or vascular lesions to symptomatic lesions causing episodic pain or swelling from thrombophlebitis, bleeding, or even significant body disfigurement. The severity of the clinical features is related to the location and extension of lesion, and coexistence of other type vascular malformation (capillary malformation or lymphatic malformation) and/or tissue hypertrophy. Head and neck VM usually cause cosmetic problems, airway obstruction, oral bleeding, or tongue motion abnormalities. Extremity or trunk VMs usually present with pain due to thrombophlebitis, limb swelling and heaviness, bleeding, or limb length or size discrepancy.

The complex morphologic and hemodynamic characteristics of VM make patient selection difficult. Sclerotherapy, the mainstay of treatment, is aimed at eradicating or reducing abnormal blood vessels by endothelial damage, inflammation, thrombosis, fibrosis, and resultant obliteration of the VM lesion. Various sclerosing agents have been used including ethanol, ethibloc (radiopaque chemical sclerosant based on ethanol, corn protein, oleum papaveris, and propylene glycol), sodium tetradecil sulphate, polydocanol, ethanolamine oleate, and bleomycin.⁷, ⁸, ⁹, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹

Among these, pure ethanol is the most powerful sclerosant. The basic mechanism of pure ethanol is cellular dehydration and vascular wall protein denaturation.⁷ Use of ethanol is limited due to its local (pain, cellulitis, compartment syndrome, perivascular or local tissue damage such as skin necrosis, nerve injury, and deep venous thrombosis) and systemic (central nervous system depression, hypoglycemia, hypertension, distal embolization, remote organ damage, pulmonary vasospasm, and blood cell damage) side effects.⁷, ¹⁰, ²⁰, ²¹ In our series, 27% of patients who underwent PES had one or more complications. There was one mortality after PES secondary to cardiovascular collapse. This patient case was not included in this series owing to a lack of follow-up imaging.

Considering the potential for local and systemic toxicity of ethanol, proper patient selection for PES is critical to achieve the best results and minimize complications. There is no standardized metric to assess treatment outcome after sclerotherapy for VM. Some authors have relied on questionnaire based on patient assessment ¹¹ while others have compared pre- and post-treatment photography.¹⁰, ¹⁵ Comparison of radiographic imaging studies have rarely been used to assess treatment.

In our clinical practice, we have observed clinical improvement after sclerotherapy for VM despite a lack of improvement on follow-up imaging study. We have also noted the converse. There have been marked reduction of VM mass on imaging without improvement clinically. This discrepancy between clinical and morphologic response make assessment of VM treatment difficult.¹² Therefore, we used both clinical and radiologic criteria to assess response to PES. Clinical response assessment is based on the patient self-assessment and not on the health care provider assessment. In our series, a “good response” was obtained only in 16% of patients, lower than in other reported series. Our definition of a “good response” is stricter than other reports.

To define good response to PES, we used strict criteria to obtain more powerful evidence on a statistical analysis. There has been reported no image criteria of good response in treatment of VM. We were concerned about the potential bias from the questionnaire survey when we used more generous criteria of good response. Therefore, we used only marked improvement as a criterion of good response. In determining image criteria of good response, we arbitrary defined marked improvement as ≥50% reduction of lesion volume (≥50% reduction on WBBPS or ≥30% decrease of MRI diameter). In measuring the volume change of VM lesion, there are several weak points in MRI and WBBPS. Though “lesion to WBBP ratio” on WBBPS is a semiquantitative method, it can be affected by increase of whole body blood pool in patients in their growing age. With MRI, it is difficult to measure volume of VM lesion and diameters are not always change equally in all direction. Despite of these limitations, we think those two are the best tools currently available to measure volume change of VM lesion.

There have been previous reports aimed at identifying predictors of response to sclerotherapy in VM patients. Puig et al¹³ proposed a classification scheme of VM based on the characteristics of venous drainage from the VM lesion. According to the classification, VMs were classified into one of four types: type I, isolated malformation without venous drainage; type II, malformation with drainage into normal veins; type III, malformation with drainage into dysplastic veins; and type IV, venous ectasia. According to Puig and colleagues, type III and IV lesions are more difficult to treat with sclerosing agents because of extensive drainage the normal venous circulation resulting in embolic complications. Therefore, those VM lesions are deemed untreatable. Goyal et al¹¹ categorized VM into three categories based on MRI findings of VM size and marginal morphology. They reported poor response to percutaneous sclerotherapy in patients with VM lesion greater than 5 cm or an ill-defined margin. We also found that no or delayed visualization of drainage vein on initial direct puncture venogram and a well-defined margin on MRI were independent predictors of good response to PES. However, in our series, VM size was not a predictor of good response on multivariate analysis.

To improve the results of sclerotherapy in VM patients, maintaining long sclerosant dwell time in the target VM lesion is crucial. Strategies to prolong contact time between sclerosant and vascular endothelium include drainage vein compression or the use of foam sclerosant. Drainage vein compression was achieved manually under fluoroscopy or utilizing a pneumatic cuff under fluoroscopy in the extremity.

Recently, the Shanghai group¹⁵ reported their 4-year experience of percutaneous sclerotherapy of venous malformations using embolization of the drainage vein with ethanol and a subsequent sclerosis therapy with bleomycin A5. Ethanol infusion into the drainage vein is a method to prolong dwell time of sclerosant in the target VM lesion.

Berenguer et al¹⁰ reported male gender and number of sclerotherapy sessions as independent predictors of good outcomes. The findings in our study were different. We found female gender to be a predictor of “good response” to PES, and number of PES sessions was not associated with a “good response”. In our series, 68% of patients with a good response underwent ≤2 sessions of PES while 10 patients who underwent >10 sessions of PES did not have a good response. This suggests that there is a specific group of patients who respond well to the PES. To minimize sclerotherapy-related complications, consideration should be given to whether to repeat PES in patients with a no-good response in the absence of these predictors of a good response.

When we compared the demographic, clinical, and therapeutic variables between female and male patients, we could not find any significant difference. Further investigation is needed to address why female patients have a better response to PES.

In summary, through a cross-sectional, retrospective analysis, we found that no or slow appearance of drainage vein on initial direct puncture venography, well-defined margins on MRI, and female gender were statistically significant, independent factors predictive of a good response to PES in patients with VM. These predictors are clinically useful in patient selection and can help improve treatment result and minimize complications. Inferring from our observation that the slow appearance of drainage vein is associated with good response to PES, adjuvant techniques to prolong sclerosant dwell time in the target lesion may further improve PES results.

The limitations of this study include retrospective design not covering all VM patients treated with PES and a lack of standardized assessment tool to compare our results with other reports.

Back to Article Outline

Author contributions 

Back to Article Outline

Appendix, online only Patients self assessment questionnaires 

This questionnaire asks for your view about outcome after treatment for venous malformation. For each of the following three questions, please select the one response that best describes your answer.

Back to Article Outline

References 

1. Lee BB, Do YS, Byun HS, Choo IW, Kim DI, Huh SH. Advanced management of venous malformation with ethanol sclerotherapy: mid-term results. J Vasc Surg. 2003;37:533–538
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (310 KB)
   • CrossRef
2. Loose DA. Surgical management of venous malformations. Phlebology. 2007;22:276–282
   • View In Article
   • CrossRef
3. Lee BB, Laredo J, Kim YW, Neville R. Congenital vascular malformations: general treatment principles. Phlebology. 2007;22:258–263
   • View In Article
   • CrossRef
4. Lee BB, Kim DI, Huh S, Kim HH, Choo IW, Byun HS, et al. New experiences with absolute ethanol sclerotherapy in the management of a complex form of congenital venous malformation. J Vasc Surg. 2001;33:764–772
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (340 KB)
   • CrossRef
5. Szilagyi DE, Smith RF, Elliott JP, Hageman JH. Congenital arteriovenous anomalies of the limbs. Arch Surg. 1976;111:423–429
   • View In Article
   • MEDLINE
6. Trout HH, McAllister HA, Giordano JM, Rich NM. Vascular malformations. Surgery. 1985;97:36–41
   • View In Article
   • MEDLINE
7. Yakes WF, Haas DK, Parker SH, Gibson MD, Hopper KD, Mulligan JS, et al. Symptomatic vascular malformations: ethanol embolotherapy. Radiology. 1989;170:1059–1066
   • View In Article
   • MEDLINE
8. Yakes WF, Luethke JM, Parker SH, Stavros AT, Rak KM, Hopper KD, et al. Ethanol embolization of vascular malformations. Radiographics. 1990;10:787–796
   • View In Article
   • MEDLINE
9. Burrows PE, Mason KP. Percutaneous treatment of low flow vascular malformations. J Vasc Intervent Radiol. 2004;15:431–445
   • View In Article
10. Berenguer B, Burrows PE, Zurakowski D, Mulliken JB. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg. 1999;104:1–11discussion 12-5
    • View In Article
    • MEDLINE
    • CrossRef
11. Goyal M, Causer PA, Armstrong D. Venous vascular malformations in pediatric patients: comparison of results of alcohol sclerotherapy with proposed MR imaging classification. Radiology. 2002;223:639–644
    • View In Article
    • MEDLINE
    • CrossRef
12. Rautio R, Laranne J, Kahara V, Saarinen J, Keski-Nisula L. Long-term results and quality of life after endovascular treatment of venous malformations in the face and neck. Acta Radiol. 2004;45:738–745
    • View In Article
    • MEDLINE
    • CrossRef
13. Puig S, Aref H, Chigot V, Bonin B, Brunelle F. Classification of venous malformations in children and implications for sclerotherapy. Pediatr Radiol. 2003;33:99–103
    • View In Article
    • MEDLINE
14. Rosenblatt M. Endovascular management of venous malformations. Phlebology. 2007;22:264–275
    • View In Article
    • CrossRef
15. Jin Y, Lin X, Li W, Hu X, Ma G, Wang W. Sclerotherapy after embolization of draining vein: a safe treatment method for venous malformations. J Vasc Surg. 2008;47:1292–1299
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (7519 KB)
    • CrossRef
16. Choi YH, Han MH KOK, Cha SH, Chang KH. Craniofacial cavernous venous malformations: percutaneous sclerotherapy with use of ethanolamine oleate. J Vasc Interv Radiol. 2002;13:475–482
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (497 KB)
    • CrossRef
17. Dubois JM, Sebag GH, De Prost Y, Teillac D, Chretien B, Brunelle FO. Soft-tissue venous malformations in children: percutaneous sclerotherapy with Ethibloc. Radiology. 1991;180:195–198
    • View In Article
    • MEDLINE
18. Pascarella L, Bergan JJ, Yamada C, Mekenas L. Venous angiomata: treatment with sclerosant foam. Ann Vasc Surg. 2005;19:457–464
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1251 KB)
    • CrossRef
19. Tan KT, Kirby J, Rajan DK, Hayeems E, Beecroft JR, Simons ME. Percutaneous sodium tetradecyl sulfate sclerotherapy for peripheral venous vascular malformations: a single-center experience. J Vasc Interv Radiol. 2007;18:343–351
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (766 KB)
    • CrossRef
20. Lee KB, Kim DI, Oh SK, Do YS, Kim KH, Kim YW. Incidence of soft tissue injury and neuropathy after embolo/sclerotherapy for congenital vascular malformation. J Vasc Surg. 2008;48:1286–1291
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (232 KB)
    • CrossRef
21. Rimon U, Garniek A, Galili Y, Golan G, Bensaid P, Morag B. Ethanol sclerotherapy of peripheral venous malformations. Eur J Radiol. 2004;52:283–287
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (157 KB)
    • CrossRef