- Email: [email protected]
Sclerotherapy after embolization of draining vein: A safe treatment method for venous malformations
Sclerotherapy after embolization of draining vein: A safe treatment method for venous malformations Yunbo Jin, MD, Xiaoxi Lin, MD, Wei Li, MD, Xiaojie Hu, MD, Gang Ma, MD, and Wei Wang, MD, Shanghai, People’s Republic of China Background: Treatment of congenital venous malformations poses a major clinical challenge. Great successes have been achieved with ethanol sclerotherapy in most lesions; however, severe complications are more likely to occur when more ethanol is used. Objectives: This study evaluated the safety and efficacy of a new sclerotherapy treatment that uses absolute alcohol and bleomycin A5. We evaluated our experience to present a safe treatment method for venous malformations. Methods: The study population comprised 201 patients. Before treatment, lesions were categorized according to magnetic resonance imaging features: 120 patients had limited venous malformations, and 81 had infiltrating venous malformations. Percutaneous sclerotherapy was performed by direct injection of absolute alcohol and bleomycin A5 under fluoroscopy. The mean dose of the two medicines was 3.4 mL and 3.7 mg, respectively. We investigated 592 sclerotherapy sessions. A retrospective study was performed of medical records and color photographs and magnetic resonance imaging before and after treatment The mean follow-up was 29 months (range, 12-65 months). Results: After embolization of the drainage vein with ethanol, a subsequent sclerosis therapy with bleomycin A5 was judged beneficial in 196 of 201 patients. Among the 196 responders, 56 showed disappearance of their chief symptoms, 42 showed improvement to nearly normal, and 62 showed marked improvement. The best response to our treatment was among the patients with limited venous malformations, with 114 (95%) showing at least marked improvement. Complications were tissue necrosis in 6 sessions, peripheral nerve palsy in 5, pigmentation in 10, blistering in 5, and 87 sessions resulted in transient side effects related to bleomycin A5, such as fever and gastrointestinal irritation. All were self-limited. No major complications such as cardiopulmonary collapse or pulmonary fibrosis were observed. Conclusion: Percutaneous sclerotherapy of venous malformations using absolute ethanol and bleomycin A5 is safe and effective. The simplicity, speed, and safety of sclerotherapy, combined with the quality and stability of the outcome achieved with our novel procedure, may make the technique the choice for treatment of venous malformations. ( J Vasc Surg 2008;47:1292-9.)
Venous malformations (VMs) are congenital, they grow proportionately with the patient, and they do not regress.1,2 Furthermore, clinical presentations of VMs are variable, ranging from asymptomatic birthmarks to lifethreatening conditions.3,4 The most accurate radiologic techniques for delineating VMs are magnetic resonance imaging (MRI) and direct injection venography, and these anomalies exhibit high signal intensity on spin-echo T2weighted MRI sequence.5 Treatment of VMs is difficult. Surgical resection may be hazardous, leading to major blood loss and incomplete resection. Recurrence and cosmetic problems are also common after surgical resections.6 Because of this, percutaneous sclerotherapy, most commonly with ethanol, is now advocated as the treatment of choice.3,5,7-9 This approach can be extremely dangerous, however, because of various major and minor ethanol toxicity-related complications that are associated with substantial acute or chronic morbidity, even when precautions are taken.10 The correlation
of the amount of ethanol administered with the serum ethanol level indicates the need to carefully monitor the ethanol dose during therapeutic procedures.11 Bleomycin is an antitumor agent that is also known to produce a sclerosing effect on endothelial cells. This drug has been used routinely in the treatment of lymphangiomas because of its safety and efficacy.12,13 The VM lesions cannot confine bleomycin as well as lymphangiomas, however, and treatment of VMs with bleomycin showed higher incidence of residual disease.13 We used a small dose of ethanol to sclerose the draining vein of VMs and then injected bleomycin A5 as a sclerosant into the lesion. Thus, the dose of ethanol used was reduced and the resistance time of bleomycin A5 to shrink the VM was prolonged. This study was done to evaluate the safety and efficacy of this treatment. We retrospectively evaluated our experience to present results of this treatment method for VMs. MATERIAL AND METHODS
From the Department of Plastic Surgery, Shanghai 9th People’s Hospital, School of Medicine, Shanghai Jiaotong University. Competition of interest: none. Correspondence: Xiaoxi Lin, MD, Plastic Surgery Department, Shanghai 9th People’s Hospital, No. 639 Zhizaoju Rd, Huangpu Region, Shanghai, PR China (e-mail: [email protected]). 0741-5214/$34.00 Copyright © 2008 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2008.01.010
1292
Patients. Between January 2002 and July 2006, 592 sclerotherapy sessions were performed in 201 patients (99 females and 102 males) who were a mean age of 14.1 years. All patients had been diagnosed with common sporadic VMs by the referring center and had had an MRI before sclerotherapy to investigate the depth of the malformation and tissue involvement. Patients with VMs due to Bean
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
syndrome, glomuvenous malformation, Maffucci syndrome, and Klippel-Trenaunay syndrome were not included in the series. Thirty-eight patients had undergone previous surgery elsewhere and had experienced recurrences. All patients had symptoms caused by the malformation, and some had two or more symptoms. The study did not include asymptomatic patients, who were observed only. In addition, only patients who were followed up ⬎1 year after the treatment were included in the study. The sex and age of the patients, location of the lesions, clinical findings, and previous treatments were noted. The number of sclerotherapy sessions and dosages of the two sclerosing agents were documented, as were any complications. The clinical characteristics of the patients are reported in Table I. Lesions were categorized using MRI on the basis of lesion margins. We divided all VMs into two categories: (1) 120 VMs had a well-defined or circumscribed margin with a sharp abrupt transition from the surrounding tissue, which remained confined to its tissue and fascial plane, and were defined as limited VMs; and (2) 81 VMs had an ill-defined margin, with an irregular interface with the surrounding tissue that appeared to cross tissue and fascial planes, and were defined as infiltrating VMs. Techniques. Percutaneous sclerotherapy was performed under fluoroscopy, without general anesthesia. Sedation was necessary for 78 pediatric patients to ensure adequate compliance. A venous pressure tourniquet was applied if anatomically possible. A sterile field was created, and a direct puncture was performed through normal skin with a 21-gauge intravenous cannula. Once venous blood returned, contrast (UItravist 300, Schering, Germany) was injected to opacify the malformation and to evaluate the resistance time of the contrast and the connection between the VM and the systemic circulation. If the extent of the lesion shown under fluoroscopy was significantly smaller than that demonstrated on MRI, more punctures were required in the same session. After appropriate imaging, absolute alcohol was the first used and 0.5 to 8 mL was injected. Compression was maintained for 5 minutes to retain the sclerosant until the draining vein of VMs was occluded (Fig 1). A maximum of 8 mg of bleomycin A5 was then injected. The maximum dose in one treatment session was 0.2 mL/kg for absolute alcohol and 0.2 mg/kg for bleomycin A5. Repeated courses of sclerotherapy were administered when remaining VMs were confirmed by a direct puncture test. We prefer to wait at least 4 to 6 weeks between sessions to let the patients recover from the previous therapy. The delay also allows the local tissue reaction to subside. Each patient had a mean of three sessions (range, 1-8 sessions). Evaluation of outcome. For patients with a chief complaint of a disfiguring VM, treatment success was defined by a reduction of the mass. Treatment success for patients with chief complaints of pain, bleeding, or functional limitation was defined by resolution or reduction of these presenting symptoms. Treatment outcomes were
Jin et al 1293
Table I. Clinical characteristics of patients with vascular malformation Vascular malformation type Characteristic Sex, No. (%) Male Female Age, mean y Site of VM, No. (%) Head Neck Trunk Upper extremity Lower extremity Symptoms, No. (%) Disfigurement Pain Bleeding Functional limitation Prior treatment, No. (%) Surgery Others Total patients, No. (%)
Limited
Infiltrating
Total
62 (30.8) 58 (28.9) 14.9
40 (19.9) 41 (20.4) 13.1
102 (50.7) 99 (49.3) 14.1
94 (46.8) 4 (2.0) 8 (4.0) 8 (4.0) 6 (3.0)
14 (7.0) 2 (1.0) 8 (4.0) 22 (11.0) 35 (17.4)
108 (53.7) 6 (3.0) 16 (8.0) 30 (15.0) 41 (20.3)
116 (57.7) 16 (8.0) 2 (1.0) 4 (2.0)
58 (28.9) 46 (22.9) 4 (2.0) 7 (3.5)
174 (86.6) 62 (30.9) 6 (3.0) 11 (5.5)
26 (12.9) 16 (8.0) 120 (59.7)
12 (6.0) 10 (5.0) 81(40.3)
38 (18.9) 26 (12.9) 201 (100)
confirmed with follow-up clinical evaluation or the patient’s report. A panel study was also conducted to evaluate therapeutic outcome. The three raters were physician-members of the plastic surgery department who were not involved in the treatment of these patients. Two pretreatment and two post-treatment colored photographs were rated only for patients with a complaint of disfiguring VMs. For some patients, MRI follow-up was available but was not included in the evaluation of clinical results. Patients were followed up from 1 to 5 years. The post-treatment photograph from 6 weeks to 5 years after the last procedure was used. The results were graded on a 5-point scale: 1, no change or worse; 2, minor improvement; 3, marked improvement but malformation still obvious; 4, nearly normal appearance; 5, apparently cured. A questionnaire was also sent to adult patients and the parents of children with treated VMs. Questions were asked about symptoms, complications, and recurrences. The respondents with complaints of pain, bleeding, or functional limitation were asked to grade the outcome of treatment on the same 5-point scale. The questionnaire was mailed between 1 and 5 years after the last procedure. Total outcome was determined according to change of chief complaint. Complications were defined as minor or major. Minor complications were defined as no therapy required and no consequences. Major complications included therapy required, permanent adverse sequelae, or death. Statistical analysis. Kruskal-Wallis test was used to evaluate the difference of clinical outcomes and the number of sessions required between limited VMs and infiltrating VMs. Two-tailed values of P ⬍ .05 were considered statistically significant. Data analysis was conducted using the SAS 6.12 software (SAS Institute, Cary, NC).
1294 Jin et al
JOURNAL OF VASCULAR SURGERY June 2008
Fig 1. Digital subtraction image shows (A) the image right after contrast injection into the main component of the lesion without absolute alcohol injection. After 2 seconds of contrast developing, (B) significant venous egress is shown. Radiography after a 3-mL absolute alcohol injection shows lesion without venous egress, even after 12 seconds of contrast developing; (C, D) the draining vein has been embolized.
RESULTS Table I reports the clinical characteristics of the patients. We treated 120 patients with limited VMs, of which 94 (78.3%) were located in the head. In the 81 patients with infiltrating VMs, the most frequent location was in the extremities (70.4%). When the two groups were considered together, the most frequent symptom of VMs was cosmetic deformity. Table II summarizes the outcome of the treatment in the 201 patients. Patients underwent a mean of three
sessions (range, 1-8 sessions). Of the patients with infiltrating VMs, 70.4% underwent more than three sessions, while only 35% of the patients with limited VMs needed more than three sessions. Mean amounts of absolute alcohol and bleomycin A5 were 3.4 mL and 3.7 mg, respectively. In 196 patients (97.5%), an improvement was shown on the photographs reviewed by the panel (Figs 2 and 3) or was reported by patients on the questionnaire. Only five patients were graded as “no change or worse.” Among them, three patients complained of both disfigurement and
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
Jin et al 1295
Table II. Outcome and results of the treatment in 201 patients with vascular malformation Vascular malformation type Outcome Treatments, No. (%) ⱕ2 3-5 ⱖ6 Dosage, mean ⫾ SD Absolute alcohol, mL Bleomycin A5, mg Follow-up, median (range) mon Symptom, No. (%) Disfiguring lesion Disappeared Decreased Pain Disappeared Decreased Bleeding Disappeared Decreased Functional limitation Disappeared Decreased Rating, No. (%) Cure Nearly normal Marked improvement Minor improvement No change or worse
Limited
Infiltrating
78 (38.8) 38 (18.9) 4 (2.0)
24 (11.9) 41 (20.4) 16 (8.0)
2.8 ⫾ 0.3 3.6 ⫾ 0.3
4.3 ⫾ 0.3 3.8 ⫾ 0.4
31 (12-65) 25 (12-63)
Total 102 (50.7) 79 (39.3) 20 (10.0) 3.4 ⫾ 0.2 3.7 ⫾ 0.2 29 (12-65)
56 (32.2) 60 (34.5)
0 54 (31.0)
56 (32.2) 114 (65.5)
12 (19.4) 4 (6.4)
28 (45.1) 14 (22.6)
40 (64.5) 18 (29.0)
2 (33.3) 0
4 (66.7) 0
6 (100) 0
2 (18.2) 2 (18.2)
2 (18.2) 5 (45.4)
4 (36.4) 7 (63.6)
56 (27.9) 34 (16.9) 24 (11.9) 6 (3.0) 0
0 8 (4.0) 38 (18.9) 30 (14.9) 5 (2.5)
56 (27.9) 42 (20.9) 62 (30.8) 36 (17.9) 5 (2.5)
pain, with disfigurement as their chief complaint. One patient complained only about disfigurement. All of them were judged as “no change” by the physician panel that evaluated the photographs. Another patient complained about pain in his chin due to the VM. The pain was not relieved after the procedure. Skin necroses occurred and left a small scar, and the patient reported a “worse” outcome in the questionnaire. No patient had a relapse after the initial response. Among the 196 responders to the treatment, 56 showed a total disappearance of their chief presenting symptoms, 42 improved to nearly normal, and 62 showed marked improvement. Of the 174 patients who presented with disfigurement, the disfiguring lesion disappeared in 56 and was reduced in 114. Of the 62 patients who presented with pain, this disappeared in 40 and decreased in 18. The six patients who presented with bleeding were all completely controlled. Eleven patients had functional limitation (Table II); visual obstruction disturbed six patients with periorbital VMs, and intraoral VMs of two patients caused speaking problems. Venous malformations located in the sole of the foot of the other three patients resulted in pain with ambulation. All three patients with claudication and one patient with visual obstruction reported that the functional limitation disappeared. The other patients reported improvement. Sixty-four patients had been previously
treated: 38 had undergone surgery with subsequent recurrence, and 26 had had other therapy, with scant improvement. All of these patients improved after our treatment, and their satisfaction level after the treatment was much higher than after previous treatment. The best response to treatment was among the patients with limited VMs, because these patients were likely to have a better clinical outcome and required fewer sessions than those patients with infiltrating VMs (P ⬍ .001 for both comparisons, Kruskal-Wallis). After 592 treatment sessions, 21 sessions (3.6%) resulted in minor complications that were related to ethanol in 17 of the 201 patients (8.5%). The patients developed fingertip necrosis after five sessions and had skin necrosis in the chin area after one session. All of these healed spontaneously (Table III). Five treatment sessions caused transient peripheral nerve paresis (3 were transient facial nerve paresis) that completely resolved ⱕ3 weeks after the treatment. Five sessions evoked acute blistering, which disappeared spontaneously 1 week after the treatment. Skin pigmentation appeared after 10 sessions after superficial sclerosis and disappeared spontaneously. Five sessions caused two kinds of the above-mentioned complications, and four patients sustained two complications in two separate sessions. Some cases of transient side effects also occurred related to the bleomycin A5. Transient fever occurred in 35 sessions (5.9%) and gastrointestinal irritation in 76 (12.8%); both fever and gastrointestinal irritation occurred in 24 sessions. All patients with these complications recovered spontaneously ⱕ2 days. No patient had thrombosis of the deep venous system, pulmonary fibrosis, or cardiopulmonary collapse. DISCUSSION The treatment of VMs is complicated. Throughout the centuries, many attempts have been made to control this ever-challenging problem, mostly led by surgeons, especially in last century.14 The results of these approaches were often disastrous, however, mainly because of poorly planned and overaggressive surgical treatment on the basis of limited knowledge of the complex nature of VMs.3 According to the biologic classification of Mulliken and Glowacki proposed in 1982,1 VMs are vascular malformations. In 1988, at the Seventh Meeting of the International Workshop on Vascular Malformations in Hamburg, a new classification system, the “Hamburg Classification,” was proposed.15,16 This classification has excellent clinical applicability, with minimum confusion, because the terminology itself provides substantial information about the anatomic and pathophysiologic status of a VM on the basis of embryologic development.17 Because of the pathognomonic characteristics of vascular defects originating from mesenchymal cells, defects at the earlier stage of organogenesis especially have been known for their notorious reputation of recurrence caused by minimal stimulation and change of milieu.18
1296 Jin et al
JOURNAL OF VASCULAR SURGERY June 2008
Fig 2. A 28-year-old man with infiltrating venous malformation treated with seven sessions is shown (a) before the start of treatment and (b) 13 months after the end of treatment. The magnetic resonance imaging (c) shows that the lesion involves subcutaneous tissue and muscle levels of the cheek (infiltrating venous malformation), and (d) that the lesion improved significantly after the treatment.
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
Jin et al 1297
Fig 3. A 32-year-old man with venous malformation of the cheek treated with 6 sessions, (a) before the start of treatment and (b) 6 months after the end of the treatment. The magnetic resonance imaging (c) shows an ill-defined margin that has an irregular interface with the surrounding tissue (infiltrating venous malformation) and that lesion infiltrates subcutaneous tissue, masseter muscle, and the tongue.
Table III. Complications in 201 patients with vascular malformation Vascular malformation type Complication
Limited Infiltrating
Ethanol related, No. (%) Tissue necrosis 2 (0.3) Peripheral nerve palsy 2 (0.3) Acute blistering 0 Skin pigmentation 2 (0.3) Bleomycin A5 related, No. (%) Fever 10 (1.7) Gastrointestinal irritation 42 (7.1)
Total
4 (0.7) 3 (0.5) 5 (0.8) 8 (1.4)
6 (1.0) 5 (0.8) 5 (0.8) 10 (1.7)
25 (4.2) 34 (5.7)
35 (5.9) 76 (12.8)
Percutaneous sclerotherapy with ethanol has been demonstrated to be associated with limited to no recurrence and is thus becoming the treatment of choice for VMs rather than surgery.10 Absolute ethanol induces denaturation of tissue protein, precipitation of protoplasm, and subsequent permanent obliteration of the vessel lumen. Ethanol is the only known and clinically tested effective sclerosant to be able to provide no chance of regeneration of endothelial cells, with the prospect of a “cure” in selected cases.3,10,17 A 74% to 91% response rate, observed as an improvement in symptoms or a reduction of the lesion, has been reported for ethanol sclerotherapy of VMs.3,18,19 Most adverse effects of treatment are manageable, but severe complications can result from the administration of ethanol.20 Its use can also be extremely dangerous because of various toxicity-related complications, including tissue necrosis requiring repair, permanent peripheral nerve injury, central nervous system depression, hypoglycemia, hypertension, hemolysis, pulmonary embolism, pulmonary
vasospasm, cardiac arrhythmias, and electromechanical disassociation.21,22 Manson et al11 have demonstrated a strong correlation between the serum ethanol level and the dose of ethanol administered during interventional procedures. Patients who receive a dose of ethanol up to 1.0 mL/kg during sclerotherapeutic procedures may have elevated serum ethanol levels that could put them at risk of respiratory depression, cardiac arrhythmias, seizures, rhabdomyolysis, and hypoglycemia. The total dose of 1 mL/kg per session should never be exceeded. Acute alcohol administration is also associated with myocardial depression and a risk of developing angina pectoris.23 Thus, the safety issue of the clinical use of absolute ethanol in VMs still remains unclear, and absolute ethanol seems too dangerous to be accepted as a “routine” sclerosant to treat VMs.10 Bleomycin was developed by Umezawa as an antitumor agent in 1966, and its mechanism of action was by inhibition of DNA synthesis. Researchers later discovered that this drug has a sclerosing effect on the endothelial cells. In 1977, Yura et al24 were the first to use bleomycin as a sclerosing agent in lymphangiomas, and good results from treatment by bleomycin sclerotherapy of lymphangiomas have recently been reported.25 Mathur et al13 also reported bleomycin sclerotherapy of VMs in three patients, achieving volume reduction of at least 50% of the lesions. In addition, they encountered no major complications with this therapy; however, therapy with bleomycin alone was associated with a high incidence of residual disease. The side effects of bleomycin A5 sclerotherapy are generally minimal. The most feared complication of this medicine is pulmonary fibrosis, which was reported when it was administered during cancer therapy. For sclerotherapy in cystic hygroma, Sung et al12 suggested the bleomycin A5
JOURNAL OF VASCULAR SURGERY June 2008
1298 Jin et al
dose should be ⬍1 mg/kg, using intervals of ⬎2 weeks, with the total dose limited to 5 mg/kg. Ethanol was used in our study only for sclerosis of the draining vein of VMs (Fig 1). The thrombosis induced by ethanol resulted in occlusion of the draining vein. This approach allowed us to reduce the dosage of ethanol significantly. In the 201 cases in our study, the mean amount of ethanol used per session was 3.4 mL (range, 0.5-8 mL), and the maximum dose was 0.2 mL/kg, which was significantly lower than the maximum dose with potential risk (1.0 mL/kg). Using this approach, we encountered 21 episodes (3.6%) of transient minor complications, which was lower than the reported range of 7.5% to 26.7%,3,17,19 and no major complications related to ethanol such as permanent nerve injury or cardiovascular collapse. Sclerosis of the drainage vein in our treatment then minimized the passage of bleomycin A5 into the systemic circulation (Fig 1). The resistance time of bleomycin A5 in the VM was also prolonged, inducing a more sufficient sclerosing effect. We used the similar regimen suggested by Sung.12 The mean amount of bleomycin A5 per session was 3.7 mg (range, 0.5-8 mg), with the maximum dose limited to 0.2 mg/kg, which was far below 1 mg/kg maximum recommended dose. In the 592 treatment sessions in our study, transient fever occurred in 35, gastrointestinal irritation occurred in 76, and both fever and gastrointestinal irritation occurred in 24 sessions. Thus, the complication rate related to bleomycin A5 was 14.7% (87 of 592). All patients recovered spontaneously ⱕ2 days, and no pulmonary fibrosis was reported. Besides the satisfactory safety from the treatment, the efficacy of this combined approach is encouraging. The treatment showed a high degree of success: 97.5% of the patients reported an improvement in symptoms or a reduction of the lesion. Among the patients whose treatment was completed, 90 (75%) of the limited VMs disappeared or became nearly normal. Furthermore, although none of the infiltrating VMs had an excellent response, the treatment proved beneficial in 46 of the 81 patients (56.8%), who had markedly improved symptoms. We therefore believe sclerotherapy with ethanol and bleomycin A5 is a worthwhile treatment option, even in patients with infiltrating VMs. Patients with limited VMs were likely to have a better clinical outcome and required fewer sessions than those with infiltrating VMs (P ⬍ .001 for both comparisons, Kruskal-Wallis). Magnetic resonance imaging may be potentially useful to help predict outcome of the treatment and may prove useful in the choice of appropriate patients for the treatment. Detergent sclerosant agents, such as sodium tetradecyl sulfate, polydocanol, sodium morrhuate, and ethanolamine, might give the same positive effect when combined with ethanol, but we did not include these in our series. Producing a foaming sclerosant by mixing the drug with air has become popular.26 A prolonged contact with the inner venous wall could explain why a foam sclerosant is more efficient compared with a liquid sclerosant,27 but further study is needed.
CONCLUSION Sclerotherapy using absolute ethanol and bleomycin A5 for VMs is safe and effective. Our experience using this new combined form of sclerotherapy on patients with VMs showed a high degree of success. Moreover, no major complications such as cardiovascular collapse or pulmonary fibrosis occurred. The simplicity, speed, and safety of sclerotherapy, combined with the quality and stability of the outcome achieved with our novel procedure, may make sclerotherapy with absolute ethanol and bleomycin A5 the choice for treatment of VMs. AUTHOR CONTRIBUTIONS Conception and design: XL Analysis and interpretation: XL, YJ Data collection: XL, YJ, WL, XH, GM, WW Writing the article: YJ Critical revision of the article: XL, YJ, WL, XH, WW Final approval of the article: XL Statistical analysis: YJ, GM Obtained funding: Not applicable Overall responsibility: XL REFERENCES 1. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412-22. 2. Finn MC, Glowacki J, Mulliken JB. Congenital vascular lesions: clinical application of a new classification. J Pediatr Surg 1983;18:894-900. 3. 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; 33:764-72. 4. Fishman SJ, Mulliken JB. Vascular anomalies. A primer for pediatricians. Pediatr Clin North Am 1998;45:1455-77. 5. Berenguer B, Burrows PE, Zurakowski D, Mulliken JB. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg 1999;104:1-11. 6. Trout HH, McAllister HA, Giordano JM, Rich NM. Vascular malformations. Surgery 1985;97:36-41. 7. Yakes WF, Luethke JM, Parker SH, Stavros AT, Rak KM, Hopper KD, et al. Ethanol embolization of vascular malformations. Radiographics 1990;10:787-96. 8. Rautio R, Laranne J, Kähärä 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;45:738-45. 9. Lee CH, Chen SG. Direct percutaneous ethanol instillation for treatment of venous malformation in the face and neck. Br J Plast Surg 2005;58:1073-8. 10. Lee BB, Do YS, Byun HS, Choo IW, Kim DI, Huh SH. Advanced management of venous malformation with ethanol sclerotherapy: midterm results. J Vasc Surg 2003;37:533-8. 11. Mason KP, Michna E, Zurakowski D, Koka BV, Burrows PE. Serum ethanol levels in children and adults after ethanol embolization or sclerotherapy for vascular anomalies. Radiology 2000;217:127-32. 12. Sung MW, Chang SO, Choi JH, Kim JY. Bleomycin sclerotherapy in patients with congenital lymphatic malformation in the head and neck. Am J Otolaryngol 1995;16:236-41. 13. Mathur NN, Rana I, Bothra R, Dhawan R, Kathuria G, Pradhan T. Bleomycin sclerotherapy in congenital lymphatic and vascular malformations of head and neck. Int J Pediatr Otorhinolaryngol 2005;69:75-80. 14. Malan E. Vascular malformations (angiodysplasias). Milan, Italy: Carlo Erba Foundation; 1974. p. 17.
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
15. Bastide G, Lefebvre D. Anatomy and organogenesis and vascular malformations. In: Belov ST, Loose DA, Weber J, editors. Vascular malformations. Reinbek, Germany: Einhorn-Presse Verlag GmbH; 1989. p. 20-2. 16. Belov ST. Anatomopathological classification of congenital vascular defects. Semin Vasc Surg 1993;6:219-24. 17. Yakes WF, Haas DK, Parker SH, Gibson MD, Hopper KD, Mulligan JS, et al. Symptomatic vascular malformations: ethanol embolotherapy. Radiology 1989;170:1059-66. 18. 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;223:639-44. 19. Hammer FD, Boom LM, Mathurin P, Vanwijck RR. Ethanol sclerotherapy of venous malformations: evaluation of systemic ethanol contamination. J Vasc Interv Radiol 2001;12:595-600. 20. Burrows PE, Mason KP. Percutaneous treatment of low flow vascular malformations. Vasc Interv Radiol 2004;15:431-45. 21. Gelczer RK, Charboneau JW, Hussain S, Brown DL. Complications of percutaneous ethanol ablation. J Ultrasound Med 1998;17:531-3.
Jin et al 1299
22. Villavicencio JL. Primum non nocere: Is it always true? The use of absolute ethanol in the management of congenital vascular malformations. J Vasc Surg 2001;33:904-6. 23. Piano MR, Schwertz DW. Alcoholic heart disease: a review. Heart Lung 1994;23:3-17; quiz 18-20. 24. Yura J, Hashimoto T, Tsuruga N, Shibata K. Bleomycin treatment for cystic hygroma in children. Nippon Geka Hokan 1977;46:607-14. 25. Kim KH, Sung MW, Roh JL, Han MH. Sclerotherapy for congenital lesions in the head and neck. Otolaryngol Head Neck Surg 2004;131: 307-16. 26. Cabrera J, Cabrera J Jr, Garcia-Olmedo MA, Redondo P. Treatment of venous malformations with sclerosant in microfoam form. Arch Dermatol 2003;139:1409-16. 27. Hamel-Desnos C, Desnos P, Wollmann JC, Ouvry P, Mako S, Allaert FA. Evaluation of the efficacy of polidocanol in the form of foam compared with liquid foam in sclerotherapy of the greater saphenous vein: initial results. Dermatol Surg 2003;29:1170-5. Submitted Sep 7, 2007; accepted Jan 7, 2008.
CME Credit Now Available to JVS Readers Readers can now obtain CME credits by reading selected articles and correctly answering multiple choice questions on the Journal website (www.jvascsurg.org). Four articles are identified in the Table of Contents of each issue and 2 questions for each are posted on the website. After correctly answering the 8 questions, readers will be awarded 2 hours of Category I CME credit.
Venous malformations (VMs) are congenital, they grow proportionately with the patient, and they do not regress.1,2 Furthermore, clinical presentations of VMs are variable, ranging from asymptomatic birthmarks to lifethreatening conditions.3,4 The most accurate radiologic techniques for delineating VMs are magnetic resonance imaging (MRI) and direct injection venography, and these anomalies exhibit high signal intensity on spin-echo T2weighted MRI sequence.5 Treatment of VMs is difficult. Surgical resection may be hazardous, leading to major blood loss and incomplete resection. Recurrence and cosmetic problems are also common after surgical resections.6 Because of this, percutaneous sclerotherapy, most commonly with ethanol, is now advocated as the treatment of choice.3,5,7-9 This approach can be extremely dangerous, however, because of various major and minor ethanol toxicity-related complications that are associated with substantial acute or chronic morbidity, even when precautions are taken.10 The correlation
of the amount of ethanol administered with the serum ethanol level indicates the need to carefully monitor the ethanol dose during therapeutic procedures.11 Bleomycin is an antitumor agent that is also known to produce a sclerosing effect on endothelial cells. This drug has been used routinely in the treatment of lymphangiomas because of its safety and efficacy.12,13 The VM lesions cannot confine bleomycin as well as lymphangiomas, however, and treatment of VMs with bleomycin showed higher incidence of residual disease.13 We used a small dose of ethanol to sclerose the draining vein of VMs and then injected bleomycin A5 as a sclerosant into the lesion. Thus, the dose of ethanol used was reduced and the resistance time of bleomycin A5 to shrink the VM was prolonged. This study was done to evaluate the safety and efficacy of this treatment. We retrospectively evaluated our experience to present results of this treatment method for VMs. MATERIAL AND METHODS
From the Department of Plastic Surgery, Shanghai 9th People’s Hospital, School of Medicine, Shanghai Jiaotong University. Competition of interest: none. Correspondence: Xiaoxi Lin, MD, Plastic Surgery Department, Shanghai 9th People’s Hospital, No. 639 Zhizaoju Rd, Huangpu Region, Shanghai, PR China (e-mail: [email protected]). 0741-5214/$34.00 Copyright © 2008 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2008.01.010
1292
Patients. Between January 2002 and July 2006, 592 sclerotherapy sessions were performed in 201 patients (99 females and 102 males) who were a mean age of 14.1 years. All patients had been diagnosed with common sporadic VMs by the referring center and had had an MRI before sclerotherapy to investigate the depth of the malformation and tissue involvement. Patients with VMs due to Bean
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
syndrome, glomuvenous malformation, Maffucci syndrome, and Klippel-Trenaunay syndrome were not included in the series. Thirty-eight patients had undergone previous surgery elsewhere and had experienced recurrences. All patients had symptoms caused by the malformation, and some had two or more symptoms. The study did not include asymptomatic patients, who were observed only. In addition, only patients who were followed up ⬎1 year after the treatment were included in the study. The sex and age of the patients, location of the lesions, clinical findings, and previous treatments were noted. The number of sclerotherapy sessions and dosages of the two sclerosing agents were documented, as were any complications. The clinical characteristics of the patients are reported in Table I. Lesions were categorized using MRI on the basis of lesion margins. We divided all VMs into two categories: (1) 120 VMs had a well-defined or circumscribed margin with a sharp abrupt transition from the surrounding tissue, which remained confined to its tissue and fascial plane, and were defined as limited VMs; and (2) 81 VMs had an ill-defined margin, with an irregular interface with the surrounding tissue that appeared to cross tissue and fascial planes, and were defined as infiltrating VMs. Techniques. Percutaneous sclerotherapy was performed under fluoroscopy, without general anesthesia. Sedation was necessary for 78 pediatric patients to ensure adequate compliance. A venous pressure tourniquet was applied if anatomically possible. A sterile field was created, and a direct puncture was performed through normal skin with a 21-gauge intravenous cannula. Once venous blood returned, contrast (UItravist 300, Schering, Germany) was injected to opacify the malformation and to evaluate the resistance time of the contrast and the connection between the VM and the systemic circulation. If the extent of the lesion shown under fluoroscopy was significantly smaller than that demonstrated on MRI, more punctures were required in the same session. After appropriate imaging, absolute alcohol was the first used and 0.5 to 8 mL was injected. Compression was maintained for 5 minutes to retain the sclerosant until the draining vein of VMs was occluded (Fig 1). A maximum of 8 mg of bleomycin A5 was then injected. The maximum dose in one treatment session was 0.2 mL/kg for absolute alcohol and 0.2 mg/kg for bleomycin A5. Repeated courses of sclerotherapy were administered when remaining VMs were confirmed by a direct puncture test. We prefer to wait at least 4 to 6 weeks between sessions to let the patients recover from the previous therapy. The delay also allows the local tissue reaction to subside. Each patient had a mean of three sessions (range, 1-8 sessions). Evaluation of outcome. For patients with a chief complaint of a disfiguring VM, treatment success was defined by a reduction of the mass. Treatment success for patients with chief complaints of pain, bleeding, or functional limitation was defined by resolution or reduction of these presenting symptoms. Treatment outcomes were
Jin et al 1293
Table I. Clinical characteristics of patients with vascular malformation Vascular malformation type Characteristic Sex, No. (%) Male Female Age, mean y Site of VM, No. (%) Head Neck Trunk Upper extremity Lower extremity Symptoms, No. (%) Disfigurement Pain Bleeding Functional limitation Prior treatment, No. (%) Surgery Others Total patients, No. (%)
Limited
Infiltrating
Total
62 (30.8) 58 (28.9) 14.9
40 (19.9) 41 (20.4) 13.1
102 (50.7) 99 (49.3) 14.1
94 (46.8) 4 (2.0) 8 (4.0) 8 (4.0) 6 (3.0)
14 (7.0) 2 (1.0) 8 (4.0) 22 (11.0) 35 (17.4)
108 (53.7) 6 (3.0) 16 (8.0) 30 (15.0) 41 (20.3)
116 (57.7) 16 (8.0) 2 (1.0) 4 (2.0)
58 (28.9) 46 (22.9) 4 (2.0) 7 (3.5)
174 (86.6) 62 (30.9) 6 (3.0) 11 (5.5)
26 (12.9) 16 (8.0) 120 (59.7)
12 (6.0) 10 (5.0) 81(40.3)
38 (18.9) 26 (12.9) 201 (100)
confirmed with follow-up clinical evaluation or the patient’s report. A panel study was also conducted to evaluate therapeutic outcome. The three raters were physician-members of the plastic surgery department who were not involved in the treatment of these patients. Two pretreatment and two post-treatment colored photographs were rated only for patients with a complaint of disfiguring VMs. For some patients, MRI follow-up was available but was not included in the evaluation of clinical results. Patients were followed up from 1 to 5 years. The post-treatment photograph from 6 weeks to 5 years after the last procedure was used. The results were graded on a 5-point scale: 1, no change or worse; 2, minor improvement; 3, marked improvement but malformation still obvious; 4, nearly normal appearance; 5, apparently cured. A questionnaire was also sent to adult patients and the parents of children with treated VMs. Questions were asked about symptoms, complications, and recurrences. The respondents with complaints of pain, bleeding, or functional limitation were asked to grade the outcome of treatment on the same 5-point scale. The questionnaire was mailed between 1 and 5 years after the last procedure. Total outcome was determined according to change of chief complaint. Complications were defined as minor or major. Minor complications were defined as no therapy required and no consequences. Major complications included therapy required, permanent adverse sequelae, or death. Statistical analysis. Kruskal-Wallis test was used to evaluate the difference of clinical outcomes and the number of sessions required between limited VMs and infiltrating VMs. Two-tailed values of P ⬍ .05 were considered statistically significant. Data analysis was conducted using the SAS 6.12 software (SAS Institute, Cary, NC).
1294 Jin et al
JOURNAL OF VASCULAR SURGERY June 2008
Fig 1. Digital subtraction image shows (A) the image right after contrast injection into the main component of the lesion without absolute alcohol injection. After 2 seconds of contrast developing, (B) significant venous egress is shown. Radiography after a 3-mL absolute alcohol injection shows lesion without venous egress, even after 12 seconds of contrast developing; (C, D) the draining vein has been embolized.
RESULTS Table I reports the clinical characteristics of the patients. We treated 120 patients with limited VMs, of which 94 (78.3%) were located in the head. In the 81 patients with infiltrating VMs, the most frequent location was in the extremities (70.4%). When the two groups were considered together, the most frequent symptom of VMs was cosmetic deformity. Table II summarizes the outcome of the treatment in the 201 patients. Patients underwent a mean of three
sessions (range, 1-8 sessions). Of the patients with infiltrating VMs, 70.4% underwent more than three sessions, while only 35% of the patients with limited VMs needed more than three sessions. Mean amounts of absolute alcohol and bleomycin A5 were 3.4 mL and 3.7 mg, respectively. In 196 patients (97.5%), an improvement was shown on the photographs reviewed by the panel (Figs 2 and 3) or was reported by patients on the questionnaire. Only five patients were graded as “no change or worse.” Among them, three patients complained of both disfigurement and
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
Jin et al 1295
Table II. Outcome and results of the treatment in 201 patients with vascular malformation Vascular malformation type Outcome Treatments, No. (%) ⱕ2 3-5 ⱖ6 Dosage, mean ⫾ SD Absolute alcohol, mL Bleomycin A5, mg Follow-up, median (range) mon Symptom, No. (%) Disfiguring lesion Disappeared Decreased Pain Disappeared Decreased Bleeding Disappeared Decreased Functional limitation Disappeared Decreased Rating, No. (%) Cure Nearly normal Marked improvement Minor improvement No change or worse
Limited
Infiltrating
78 (38.8) 38 (18.9) 4 (2.0)
24 (11.9) 41 (20.4) 16 (8.0)
2.8 ⫾ 0.3 3.6 ⫾ 0.3
4.3 ⫾ 0.3 3.8 ⫾ 0.4
31 (12-65) 25 (12-63)
Total 102 (50.7) 79 (39.3) 20 (10.0) 3.4 ⫾ 0.2 3.7 ⫾ 0.2 29 (12-65)
56 (32.2) 60 (34.5)
0 54 (31.0)
56 (32.2) 114 (65.5)
12 (19.4) 4 (6.4)
28 (45.1) 14 (22.6)
40 (64.5) 18 (29.0)
2 (33.3) 0
4 (66.7) 0
6 (100) 0
2 (18.2) 2 (18.2)
2 (18.2) 5 (45.4)
4 (36.4) 7 (63.6)
56 (27.9) 34 (16.9) 24 (11.9) 6 (3.0) 0
0 8 (4.0) 38 (18.9) 30 (14.9) 5 (2.5)
56 (27.9) 42 (20.9) 62 (30.8) 36 (17.9) 5 (2.5)
pain, with disfigurement as their chief complaint. One patient complained only about disfigurement. All of them were judged as “no change” by the physician panel that evaluated the photographs. Another patient complained about pain in his chin due to the VM. The pain was not relieved after the procedure. Skin necroses occurred and left a small scar, and the patient reported a “worse” outcome in the questionnaire. No patient had a relapse after the initial response. Among the 196 responders to the treatment, 56 showed a total disappearance of their chief presenting symptoms, 42 improved to nearly normal, and 62 showed marked improvement. Of the 174 patients who presented with disfigurement, the disfiguring lesion disappeared in 56 and was reduced in 114. Of the 62 patients who presented with pain, this disappeared in 40 and decreased in 18. The six patients who presented with bleeding were all completely controlled. Eleven patients had functional limitation (Table II); visual obstruction disturbed six patients with periorbital VMs, and intraoral VMs of two patients caused speaking problems. Venous malformations located in the sole of the foot of the other three patients resulted in pain with ambulation. All three patients with claudication and one patient with visual obstruction reported that the functional limitation disappeared. The other patients reported improvement. Sixty-four patients had been previously
treated: 38 had undergone surgery with subsequent recurrence, and 26 had had other therapy, with scant improvement. All of these patients improved after our treatment, and their satisfaction level after the treatment was much higher than after previous treatment. The best response to treatment was among the patients with limited VMs, because these patients were likely to have a better clinical outcome and required fewer sessions than those patients with infiltrating VMs (P ⬍ .001 for both comparisons, Kruskal-Wallis). After 592 treatment sessions, 21 sessions (3.6%) resulted in minor complications that were related to ethanol in 17 of the 201 patients (8.5%). The patients developed fingertip necrosis after five sessions and had skin necrosis in the chin area after one session. All of these healed spontaneously (Table III). Five treatment sessions caused transient peripheral nerve paresis (3 were transient facial nerve paresis) that completely resolved ⱕ3 weeks after the treatment. Five sessions evoked acute blistering, which disappeared spontaneously 1 week after the treatment. Skin pigmentation appeared after 10 sessions after superficial sclerosis and disappeared spontaneously. Five sessions caused two kinds of the above-mentioned complications, and four patients sustained two complications in two separate sessions. Some cases of transient side effects also occurred related to the bleomycin A5. Transient fever occurred in 35 sessions (5.9%) and gastrointestinal irritation in 76 (12.8%); both fever and gastrointestinal irritation occurred in 24 sessions. All patients with these complications recovered spontaneously ⱕ2 days. No patient had thrombosis of the deep venous system, pulmonary fibrosis, or cardiopulmonary collapse. DISCUSSION The treatment of VMs is complicated. Throughout the centuries, many attempts have been made to control this ever-challenging problem, mostly led by surgeons, especially in last century.14 The results of these approaches were often disastrous, however, mainly because of poorly planned and overaggressive surgical treatment on the basis of limited knowledge of the complex nature of VMs.3 According to the biologic classification of Mulliken and Glowacki proposed in 1982,1 VMs are vascular malformations. In 1988, at the Seventh Meeting of the International Workshop on Vascular Malformations in Hamburg, a new classification system, the “Hamburg Classification,” was proposed.15,16 This classification has excellent clinical applicability, with minimum confusion, because the terminology itself provides substantial information about the anatomic and pathophysiologic status of a VM on the basis of embryologic development.17 Because of the pathognomonic characteristics of vascular defects originating from mesenchymal cells, defects at the earlier stage of organogenesis especially have been known for their notorious reputation of recurrence caused by minimal stimulation and change of milieu.18
1296 Jin et al
JOURNAL OF VASCULAR SURGERY June 2008
Fig 2. A 28-year-old man with infiltrating venous malformation treated with seven sessions is shown (a) before the start of treatment and (b) 13 months after the end of treatment. The magnetic resonance imaging (c) shows that the lesion involves subcutaneous tissue and muscle levels of the cheek (infiltrating venous malformation), and (d) that the lesion improved significantly after the treatment.
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
Jin et al 1297
Fig 3. A 32-year-old man with venous malformation of the cheek treated with 6 sessions, (a) before the start of treatment and (b) 6 months after the end of the treatment. The magnetic resonance imaging (c) shows an ill-defined margin that has an irregular interface with the surrounding tissue (infiltrating venous malformation) and that lesion infiltrates subcutaneous tissue, masseter muscle, and the tongue.
Table III. Complications in 201 patients with vascular malformation Vascular malformation type Complication
Limited Infiltrating
Ethanol related, No. (%) Tissue necrosis 2 (0.3) Peripheral nerve palsy 2 (0.3) Acute blistering 0 Skin pigmentation 2 (0.3) Bleomycin A5 related, No. (%) Fever 10 (1.7) Gastrointestinal irritation 42 (7.1)
Total
4 (0.7) 3 (0.5) 5 (0.8) 8 (1.4)
6 (1.0) 5 (0.8) 5 (0.8) 10 (1.7)
25 (4.2) 34 (5.7)
35 (5.9) 76 (12.8)
Percutaneous sclerotherapy with ethanol has been demonstrated to be associated with limited to no recurrence and is thus becoming the treatment of choice for VMs rather than surgery.10 Absolute ethanol induces denaturation of tissue protein, precipitation of protoplasm, and subsequent permanent obliteration of the vessel lumen. Ethanol is the only known and clinically tested effective sclerosant to be able to provide no chance of regeneration of endothelial cells, with the prospect of a “cure” in selected cases.3,10,17 A 74% to 91% response rate, observed as an improvement in symptoms or a reduction of the lesion, has been reported for ethanol sclerotherapy of VMs.3,18,19 Most adverse effects of treatment are manageable, but severe complications can result from the administration of ethanol.20 Its use can also be extremely dangerous because of various toxicity-related complications, including tissue necrosis requiring repair, permanent peripheral nerve injury, central nervous system depression, hypoglycemia, hypertension, hemolysis, pulmonary embolism, pulmonary
vasospasm, cardiac arrhythmias, and electromechanical disassociation.21,22 Manson et al11 have demonstrated a strong correlation between the serum ethanol level and the dose of ethanol administered during interventional procedures. Patients who receive a dose of ethanol up to 1.0 mL/kg during sclerotherapeutic procedures may have elevated serum ethanol levels that could put them at risk of respiratory depression, cardiac arrhythmias, seizures, rhabdomyolysis, and hypoglycemia. The total dose of 1 mL/kg per session should never be exceeded. Acute alcohol administration is also associated with myocardial depression and a risk of developing angina pectoris.23 Thus, the safety issue of the clinical use of absolute ethanol in VMs still remains unclear, and absolute ethanol seems too dangerous to be accepted as a “routine” sclerosant to treat VMs.10 Bleomycin was developed by Umezawa as an antitumor agent in 1966, and its mechanism of action was by inhibition of DNA synthesis. Researchers later discovered that this drug has a sclerosing effect on the endothelial cells. In 1977, Yura et al24 were the first to use bleomycin as a sclerosing agent in lymphangiomas, and good results from treatment by bleomycin sclerotherapy of lymphangiomas have recently been reported.25 Mathur et al13 also reported bleomycin sclerotherapy of VMs in three patients, achieving volume reduction of at least 50% of the lesions. In addition, they encountered no major complications with this therapy; however, therapy with bleomycin alone was associated with a high incidence of residual disease. The side effects of bleomycin A5 sclerotherapy are generally minimal. The most feared complication of this medicine is pulmonary fibrosis, which was reported when it was administered during cancer therapy. For sclerotherapy in cystic hygroma, Sung et al12 suggested the bleomycin A5
JOURNAL OF VASCULAR SURGERY June 2008
1298 Jin et al
dose should be ⬍1 mg/kg, using intervals of ⬎2 weeks, with the total dose limited to 5 mg/kg. Ethanol was used in our study only for sclerosis of the draining vein of VMs (Fig 1). The thrombosis induced by ethanol resulted in occlusion of the draining vein. This approach allowed us to reduce the dosage of ethanol significantly. In the 201 cases in our study, the mean amount of ethanol used per session was 3.4 mL (range, 0.5-8 mL), and the maximum dose was 0.2 mL/kg, which was significantly lower than the maximum dose with potential risk (1.0 mL/kg). Using this approach, we encountered 21 episodes (3.6%) of transient minor complications, which was lower than the reported range of 7.5% to 26.7%,3,17,19 and no major complications related to ethanol such as permanent nerve injury or cardiovascular collapse. Sclerosis of the drainage vein in our treatment then minimized the passage of bleomycin A5 into the systemic circulation (Fig 1). The resistance time of bleomycin A5 in the VM was also prolonged, inducing a more sufficient sclerosing effect. We used the similar regimen suggested by Sung.12 The mean amount of bleomycin A5 per session was 3.7 mg (range, 0.5-8 mg), with the maximum dose limited to 0.2 mg/kg, which was far below 1 mg/kg maximum recommended dose. In the 592 treatment sessions in our study, transient fever occurred in 35, gastrointestinal irritation occurred in 76, and both fever and gastrointestinal irritation occurred in 24 sessions. Thus, the complication rate related to bleomycin A5 was 14.7% (87 of 592). All patients recovered spontaneously ⱕ2 days, and no pulmonary fibrosis was reported. Besides the satisfactory safety from the treatment, the efficacy of this combined approach is encouraging. The treatment showed a high degree of success: 97.5% of the patients reported an improvement in symptoms or a reduction of the lesion. Among the patients whose treatment was completed, 90 (75%) of the limited VMs disappeared or became nearly normal. Furthermore, although none of the infiltrating VMs had an excellent response, the treatment proved beneficial in 46 of the 81 patients (56.8%), who had markedly improved symptoms. We therefore believe sclerotherapy with ethanol and bleomycin A5 is a worthwhile treatment option, even in patients with infiltrating VMs. Patients with limited VMs were likely to have a better clinical outcome and required fewer sessions than those with infiltrating VMs (P ⬍ .001 for both comparisons, Kruskal-Wallis). Magnetic resonance imaging may be potentially useful to help predict outcome of the treatment and may prove useful in the choice of appropriate patients for the treatment. Detergent sclerosant agents, such as sodium tetradecyl sulfate, polydocanol, sodium morrhuate, and ethanolamine, might give the same positive effect when combined with ethanol, but we did not include these in our series. Producing a foaming sclerosant by mixing the drug with air has become popular.26 A prolonged contact with the inner venous wall could explain why a foam sclerosant is more efficient compared with a liquid sclerosant,27 but further study is needed.
CONCLUSION Sclerotherapy using absolute ethanol and bleomycin A5 for VMs is safe and effective. Our experience using this new combined form of sclerotherapy on patients with VMs showed a high degree of success. Moreover, no major complications such as cardiovascular collapse or pulmonary fibrosis occurred. The simplicity, speed, and safety of sclerotherapy, combined with the quality and stability of the outcome achieved with our novel procedure, may make sclerotherapy with absolute ethanol and bleomycin A5 the choice for treatment of VMs. AUTHOR CONTRIBUTIONS Conception and design: XL Analysis and interpretation: XL, YJ Data collection: XL, YJ, WL, XH, GM, WW Writing the article: YJ Critical revision of the article: XL, YJ, WL, XH, WW Final approval of the article: XL Statistical analysis: YJ, GM Obtained funding: Not applicable Overall responsibility: XL REFERENCES 1. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412-22. 2. Finn MC, Glowacki J, Mulliken JB. Congenital vascular lesions: clinical application of a new classification. J Pediatr Surg 1983;18:894-900. 3. 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; 33:764-72. 4. Fishman SJ, Mulliken JB. Vascular anomalies. A primer for pediatricians. Pediatr Clin North Am 1998;45:1455-77. 5. Berenguer B, Burrows PE, Zurakowski D, Mulliken JB. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg 1999;104:1-11. 6. Trout HH, McAllister HA, Giordano JM, Rich NM. Vascular malformations. Surgery 1985;97:36-41. 7. Yakes WF, Luethke JM, Parker SH, Stavros AT, Rak KM, Hopper KD, et al. Ethanol embolization of vascular malformations. Radiographics 1990;10:787-96. 8. Rautio R, Laranne J, Kähärä 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;45:738-45. 9. Lee CH, Chen SG. Direct percutaneous ethanol instillation for treatment of venous malformation in the face and neck. Br J Plast Surg 2005;58:1073-8. 10. Lee BB, Do YS, Byun HS, Choo IW, Kim DI, Huh SH. Advanced management of venous malformation with ethanol sclerotherapy: midterm results. J Vasc Surg 2003;37:533-8. 11. Mason KP, Michna E, Zurakowski D, Koka BV, Burrows PE. Serum ethanol levels in children and adults after ethanol embolization or sclerotherapy for vascular anomalies. Radiology 2000;217:127-32. 12. Sung MW, Chang SO, Choi JH, Kim JY. Bleomycin sclerotherapy in patients with congenital lymphatic malformation in the head and neck. Am J Otolaryngol 1995;16:236-41. 13. Mathur NN, Rana I, Bothra R, Dhawan R, Kathuria G, Pradhan T. Bleomycin sclerotherapy in congenital lymphatic and vascular malformations of head and neck. Int J Pediatr Otorhinolaryngol 2005;69:75-80. 14. Malan E. Vascular malformations (angiodysplasias). Milan, Italy: Carlo Erba Foundation; 1974. p. 17.
JOURNAL OF VASCULAR SURGERY Volume 47, Number 6
15. Bastide G, Lefebvre D. Anatomy and organogenesis and vascular malformations. In: Belov ST, Loose DA, Weber J, editors. Vascular malformations. Reinbek, Germany: Einhorn-Presse Verlag GmbH; 1989. p. 20-2. 16. Belov ST. Anatomopathological classification of congenital vascular defects. Semin Vasc Surg 1993;6:219-24. 17. Yakes WF, Haas DK, Parker SH, Gibson MD, Hopper KD, Mulligan JS, et al. Symptomatic vascular malformations: ethanol embolotherapy. Radiology 1989;170:1059-66. 18. 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;223:639-44. 19. Hammer FD, Boom LM, Mathurin P, Vanwijck RR. Ethanol sclerotherapy of venous malformations: evaluation of systemic ethanol contamination. J Vasc Interv Radiol 2001;12:595-600. 20. Burrows PE, Mason KP. Percutaneous treatment of low flow vascular malformations. Vasc Interv Radiol 2004;15:431-45. 21. Gelczer RK, Charboneau JW, Hussain S, Brown DL. Complications of percutaneous ethanol ablation. J Ultrasound Med 1998;17:531-3.
Jin et al 1299
22. Villavicencio JL. Primum non nocere: Is it always true? The use of absolute ethanol in the management of congenital vascular malformations. J Vasc Surg 2001;33:904-6. 23. Piano MR, Schwertz DW. Alcoholic heart disease: a review. Heart Lung 1994;23:3-17; quiz 18-20. 24. Yura J, Hashimoto T, Tsuruga N, Shibata K. Bleomycin treatment for cystic hygroma in children. Nippon Geka Hokan 1977;46:607-14. 25. Kim KH, Sung MW, Roh JL, Han MH. Sclerotherapy for congenital lesions in the head and neck. Otolaryngol Head Neck Surg 2004;131: 307-16. 26. Cabrera J, Cabrera J Jr, Garcia-Olmedo MA, Redondo P. Treatment of venous malformations with sclerosant in microfoam form. Arch Dermatol 2003;139:1409-16. 27. Hamel-Desnos C, Desnos P, Wollmann JC, Ouvry P, Mako S, Allaert FA. Evaluation of the efficacy of polidocanol in the form of foam compared with liquid foam in sclerotherapy of the greater saphenous vein: initial results. Dermatol Surg 2003;29:1170-5. Submitted Sep 7, 2007; accepted Jan 7, 2008.
CME Credit Now Available to JVS Readers Readers can now obtain CME credits by reading selected articles and correctly answering multiple choice questions on the Journal website (www.jvascsurg.org). Four articles are identified in the Table of Contents of each issue and 2 questions for each are posted on the website. After correctly answering the 8 questions, readers will be awarded 2 hours of Category I CME credit.