- Email: [email protected]
Bleomycin Foam Treatment of Venous Malformations: A Promising Agent for Effective Treatment with Minimal Swelling
CLINICAL STUDY
Bleomycin Foam Treatment of Venous Malformations: A Promising Agent for Effective Treatment with Minimal Swelling Faheem Ul Haq, MD, Sally E. Mitchell, MD, Aylin Tekes, MD, and Clifford R. Weiss, MD
ABSTRACT Purpose: To report clinical and radiographic outcomes of patients with venous malformations (VMs) treated with bleomycin foam. Materials and Methods: Twenty patients (age, 2–68 y) presented with symptoms of swelling (n ¼ 19; 95%), pain (n ¼ 14; 70%), and bleeding (n ¼ 4; 20%). Lesions were located in the head and neck in 17 patients (85%), extremities in two (10%), and mediastinum in one (5%). Twenty-seven embolizations were performed, with a mean of 1.7 ⫾ 1.0 treatments per patient (range, 1–4). An average of 0.45 ⫾ 0.4 U/kg of bleomycin foam was used per procedure, with a range of 0.1–2.3 U/kg. Results: All procedures were technically successful with no intraprocedural complications. Mean follow-up was 66 days ⫾ 80, with a range of 4–403 days. Postprocedure complications were minor in 6 of 27 procedures (22%) and major in 2 of 27 procedures (7%). All 20 patients (100%) reported improvement in their symptoms after a single treatment session. Postprocedural magnetic resonance (MR) imaging demonstrated volume reduction of treated lesions in 13 of 14 patients (93%), with a mean lesion volume reduction of 66% ⫾ 21. Enhancement on MR imaging after treatment was decreased in 11 of 14 patients (79%), increased in two (14%), and stable in one (7%). T2 signal intensity on MR imaging after treatment was decreased in 12 of 14 patients (86%) and stable in two (14%). Conclusions: The use of bleomycin foam for the percutaneous treatment of VMs is safe and effective. Foaming bleomycin may be used to address the dose limitations of the liquid.
ABBREVIATIONS STS = sodium tetradecyl sulfate, VM = venous malformation
Venous malformations (VMs) are the most common type of vascular malformation, with an incidence of one to two per 10,000 births and a prevalence as high as 1% in the general population (1,2). Current mainstay treatment for VM is percutaneous sclerotherapy (3–6). Historically, a variety of sclerosant agents have been used to treat From the Divisions of Cardiovascular and Interventional Radiology (F.U.H., S.E.M., C.R.W.) and Pediatric Radiology and Pediatric Neuroradiology (A.T.), Russell H. Morgan Department of Radiology and Radiologic Science, The Johns Hopkins University School of Medicine, 601 N. Caroline St., Room 3125, Baltimore, MD 21287. Received August 22, 2014; final revision received April 27, 2015; accepted May 2, 2015. Address correspondence to S.E.M.; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.05.007
these lesions, but there is currently no consensus as to the best agent. Most agents cause destruction of endothelium lining the VM and lead to thrombosis, induce scarring, and decrease the size of the lesion (7–9). Recently, bleomycin liquid has emerged as a promising sclerosant agent for VMs and lymphatic malformations, particularly in areas where swelling needs to be minimized (10). Compared with 100% alcohol and sodium tetradecyl sulfate (STS; Sotradecol; AngioDynamics, Latham, New York), bleomycin does not lead to skin and mucosal necrosis and can be used without general anesthesia (11– 13). Minimal side effects, fast recovery, good outcomes, and low cost have led to increased interest in bleomycin as a sclerosant agent. In China, a variant of bleomycin, pingyangmycin (bleomycin A5 hydrochloride; Tianjin Taihe, Tianjin, China), is commonly used for sclerotherapy (14). However, lifetime dose is a limitation when using bleomycin to treat large VMs. The current recommended
2
’
Bleomycin Foam Treatment of Venous Malformations
lifetime dose is less than 400 mg or 5 mg/kg to reduce the risk of pulmonary fibrosis (11). Because of this dose limitation, bleomycin foam has been proposed to allow treatment of larger lesions while minimizing the total dose of bleomycin. The purpose of the present study is to report clinical and radiographic outcomes in patients with VMs treated with bleomycin foam.
MATERIALS AND METHODS After institutional review board approval, we performed retrospective review of electronic medical records of 20
Ul Haq et al
’
JVIR
patients (14 female and six male) with VMs treated with bleomycin foam at our institution between June 2012 and October 2013 (Table 1). Mean age of patients was 32 years ⫾ 16 (range, 2–68 y). Presenting symptoms were swelling (n ¼ 19; 95%), pain (n ¼ 14; 70%), and bleeding (n ¼ 4; 20%). Pain was assessed on a 10-point visual analog scale. Seventeen patients (85%) had lesions in the head and neck region (in or adjacent to the airway or tongue) causing symptoms such as dysphagia (n ¼ 10; 50%), dyspnea (n ¼ 4; 20%), and dysarthria (n ¼ 4; 20%) in addition to pain and swelling. Two patients (10%) had lesions in the extremities, and one (5%) presented with a mediastinal lesion that caused chest pain and swelling.
Table 1 . Patient Demographics, Venous Malformation Location, Symptoms, and Previous Treatments Pt. No./Sex/ Age (y) 1/F/23
Primary
Additional
Previous
Previous
Location
Symptoms
Symptoms
Treatments
Sclerosant Used
Head and neck
Pain, swelling
–
7 Sessions
Ethanol/ethiodol mixture (80%/ 20%), 50%, 75%, 95%, 100% alcohol
2/M/36
Head and neck
Swelling, bleeding
Dysphagia
11 Sessions
4:1 Dehydrated alcohol/ethiodol mixture, 100%
3/M/68
Head and neck
Swelling
Dysphagia,
3 Sessions
alcohol 100% Alcohol
dysarthria, airway 4/M/47
Chest
Pain, swelling
obstruction Chest tightness
No
None
5/F/18
Head and neck
Swelling
Dysphagia,
No
None
dysarthria, airway obstruction 6/F/25
Head and neck
Swelling
–
No
None
7/F/21 8/F/2
RLE Head and neck
Pain Swelling
– –
No No
None None
9/F/57
Head and neck
Pain, swelling
Dysphagia, airway
1 Session
STS foam
obstruction 10/F/20
RLE
Pain, swelling
No
None
11/M/33
Head and neck
Pain, swelling,
Dysphagia, airway
No
None
12/F/50
Head and neck
bleeding Pain, swelling
obstruction Dysphagia,
No
None
Loosening of teeth Dysphagia
No No
None None
dysarthria 13/F/30 14/F/39
Head and neck Head and neck
Swelling Pain, swelling
15/F/49
Head and neck
Pain, swelling
–
No
None
16/F/29
Head and neck
Pain, swelling
Dysphagia, dysarthria
No
None
17/F/18
Head and neck
Pain, swelling
–
No
None
18/M/17
Head and neck
Pain, swelling
–
10 Sessions
100% Alcohol, STS, and Onyx
19/F/41
Head and neck
Pain, swelling,
Dysphagia
2 Sessions
75% Ethanol, 100%
Head and neck
bleeding Pain, swelling,
Dysphagia
No
ethanol None
20/M/12
bleeding RLE ¼ right lower extremity, STS ¼ sodium tetradecyl sulfate.
Volume XX
’
Number X
’
Month
’
2015
3
Six patients (30%) had a history of previous embolizations of the same lesion with agents other than bleomycin. Patients underwent one (n ¼ 15; 75%), two (n ¼ 4; 20%), or four (n ¼ 1; 5%) treatment sessions of bleomycin foam embolization. Each patient was evaluated by the vascular anomalies team. Preprocedural imaging included magnetic resonance (MR) imaging of the involved body part per our institutional vascular anomalies protocol on 1.5- or 3.0T scanners (Avanto/Verio; Siemens, Erlangen, Germany). The following sequences were obtained: triplanar T2-weighted imaging with fat saturation, precontrast axial T1-weighted imaging, time-resolved contrastenhanced MR angiography, and postcontrast triplanar T1-weighted imaging with fat saturation. Time-resolved contrast-enhanced MR angiography was performed in the coronal plane following intravenous administration of contrast material at 0.03 mmol/kg body weight (ABLAVAR; Lantheus Medical Imaging, North Billerica, Massachusetts). Contrast agent was administered at 2 mL/s, followed by a 10-mL saline solution flush at a rate of 1 mL/s. Nineteen of 20 patients (95%) had MR imaging before embolization and one underwent computed tomography. Posttreatment MR imaging was available in 14 patients (74%). Preprocedural blood tests included fibrinogen and Ddimer measurements, prothrombin time, International Normalized Ratio, and activated partial thromboplastin time to assess the possibility of localized intravascular coagulopathy, which is a concern in patients with VMs because it predisposes the patient to peritreatment disseminated intravascular coagulopathy (15). If localized intravascular coagulopathy was found, the patient was administered anticoagulation, typically enoxaparin sodium (Lovenox; Sanofi, Bridgewater, New Jersey), for 10 days before and after the procedure to prevent disseminated intravascular coagulopathy. All embolization procedures were performed under general anesthesia. Ultrasound (US) in combination with direct vision was used to access the VMs in 12 of 27 embolizations (45%), US alone was used in six (22%), and direct vision alone was used in nine (33%). Access was obtained with a 21-gauge needle (Cook, Bloomington, Indiana) connected to microtubing with 0.2 mL dead space (Plexufix; B. Braun, Bethlehem, Pennsylvania) connected to a 3-mL syringe. Digital subtraction venography was performed at one film per second (Allura; Philips, Best, The Netherlands) before
embolization to assess the size of the accessed portion(s) of the VM and anatomy of draining veins. The contrast agents used were iohexol (Omnipaque; GE Healthcare, Princeton, New Jersey) for children and ioxilan (Oxilan; Guerbet, Villepinte, France) for adults. Bleomycin foam was prepared by using a ratio of 6 U bleomycin suspended in 1 mL normal saline solution plus 1 mL human 25% serum albumin. We used a 2:1 air:liquid ratio to create foam (Table 2). Bleomycin foam was injected into each site under digital subtraction venography or US guidance. Volume of injected foam was determined by contrast agent displacement technique, during which foam displaces previously injected contrast agent. Manual compression was applied for 3–5 minutes to compress venous outflow when present. After a 20-minute dwell time, sites with residual blood flow were checked with digital subtraction venography to determine the size of residual VMs and were treated again as needed. After a second 20-minute dwell time, all needles were removed. Intraprocedural nerve monitoring was used during embolizations when there was a potential risk of nerve damage, defined as a VM o 1 cm from a major nerve on MR imaging. This was performed by using custom neurophysiologic protocols based on combinations of electromyography, motor and sensory nerve conduction studies, somatosensory-evoked potentials, and motor-evoked potentials. Intraoperative nerve monitoring was performed in nine of 27 procedures (45%; eight head and neck VMs and one right lowerextremity VM). Posttreatment clinical outcomes were classified as improved, unchanged, or worse. To assess treatment response on MR imaging, the treated portion of the VM was measured in three dimensions on pre- and postprocedural MR imaging by three radiologists (A.T., S.E. M., and C.R.W.). The volume of each lesion was calculated before and after embolization. Because VMs have complex shapes, each lesion was assumed as ellipsoid and was calculated based on the ellipsoid volume formula 4/3πabc (with a as anteroposterior axis, b as transverse axis, and c as craniocaudal axis). Treatment response on MR imaging was classified as complete reduction (4 90%), considerable reduction (60%– 90%), partial reduction (20%–60%), unchanged (o 20%), or increased. Subjective analysis of T2 signal intensity and enhancement strength was also assessed for each treated VM before and after treatment.
Table 2 . Bleomycin Foam Preparation Bleomycin Dose (U)
Saline Solution (mL)
Human Serum Albumin 25% (mL)
Bleomycin Liquid Volume (mL)
Air (mL)
Total Bleomycin Foam Volume (mL)
15 30
2.5 5
2.5 5
5 10
10 20
15 30
45
7.5
7.5
15
30
45
Note–Bleomycin liquid was prepared by the hospital pharmacy in three doses meant for small, medium, and large venous malformations. The albumin is what catalyzes the foaming process. Bleomycin mix was foamed with air in a 1:2 ratio.
4
1 2 3
No.
Location
Bleomycin
Other Agents, Dose
Nerve
Procedural
Discharged
Clinical
Radiographic
Dose (U)
(mL)
Monitoring
Complications
After
Outcome
Reduction (%)
Improved
55, Partial 76, Considerable NA
1
Head and neck
30
–
No
No
1d
2
Head and neck
60
–
Yes
No
Id
1 1
Head and neck Head and neck
38 30
– –
No No
No No
4d 1d
Improved Improved
4
1
Mediastinum
18
–
No
No
1d
Improved
29, Partial
5
1
Head and neck
9
Alcohol at front of tongue, bleomycin
No
No
15 d
Improved
81, Considerable
2 3
Head and neck Head and neck
14 15
No No
No No
1d 1d
at back of tongue (7) – –
4
Head and neck
60
–
No
No
1d
6
1
Head and neck
15
Alcohol on lip, bleomycin on eyelid
No
No
2d
Improved
NA
7 8
1 1
RLE Head and neck
45 30
– –
No Yes
No No
1d 1d
Improved Improved
61, Considerable 98, Complete
9
1
Head and neck
15
Alcohol on face and
No
No
1d
Improved
100, Complete
No
No
1d
(7.5)
Bleomycin Foam Treatment of Venous Malformations
Procedure Pt. No.
’
Table 3 . Clinical and Radiographic Outcomes of Bleomycin Embolization
neck, bleomycin on tongue and tonsil (15.5) 2
Head and neck
22
Alcohol on face and neck, bleomycin on tongue and tonsil
10
1
RLE/bottom of
15
(22.5) –
Yes
No
1d
Improved*
NA
15 30
– STS on neck,
No No
No No
1d 1d
Improved
81, Considerable
No
No
1d
Improved
51, Partial
No
No
1d
Improved
75, Considerable
Improved*
NA
foot 11
1 2
Head and neck Head and neck
bleomycin on 1
Head and neck
30
13
1
Head and neck
15
STS and bleomycin
2
Head and neck
30
(4) –
No
No
1d
1
Head and neck
12
–
Yes
No
1d
14
’
(Continued)
Ul Haq et al
12
oropharynx (24) –
JVIR
NA
RESULTS
NA ¼ not applicable, RLE ¼ right lower extremity, STS ¼ sodium tetradecyl sulfate. Long-term follow-up performed by phone call. n
44, Partial
No
5
Access 2007 (Microsoft, Redmond, Washington) was used to enter the data. Most data reported are frequencies. Stata/IC 12.0 (STATA, College Station, Texas) for Windows (Microsoft) was used to calculate means and standard deviations.
Improved
Improved
2015
1d
1d
’
No
Yes
– 15 1 20
Head and neck
1 19
Head and neck
38
–
No
41, Partial
NA 64, Considerable Improved* Improved 1d 1d Yes Yes – – 1 1 17 18
Chest Head and neck
15 24
Yes
No No
3, Increased Improved
Improved 1d
1d Yes
30 1
Head and neck
1
16
Head and neck
30
–
No
Month
15
Radiographic
’
–
No
Outcome
Clinical Discharged
After Complications
Procedural Nerve
Number X
Monitoring
Other Agents, Dose
’
(mL) Dose (U)
Bleomycin
Location Procedure
No. Pt. No.
Table 3. Clinical and Radiographic Outcomes of Bleomycin Embolization (continued )
Reduction (%)
Volume XX
A total of 27 embolizations were performed, with a mean of 1.7 ⫾ 1.0 and a range of one to four treatments per patient (Table 3). Twenty-four embolizations (89%) were in the head and neck, two (7%) were in the right lower extremity, and one (4%) was in the chest. Bleomycin foam was the only sclerosant agent in 21 procedures. In five procedures, an additional sclerosant agent was used at a separate location from the one treated with bleomycin: four used 100% alcohol and one used 3% foamed STS. Finally, in one procedure, bleomycin foam and STS foam were used to treat the same lesion. It was necessary to add these secondary sclerosant agents because the preordered bleomycin dose was depleted before the embolization was complete. An average of 0.45 U/kg ⫾ 0.4 of bleomycin foam (26 U ⫾ 14) was used per procedure, with a range of 0.1–2.3 U/kg (9–60 U; Fig 1). All procedures were technically successful. There were no intraprocedural complications. Postoperative hospital stay was overnight in 24 of 27 procedures (89%). After one procedure in which only bleomycin foam was used, there was a 4-day hospital stay as a result of swelling that requiring extended intubation. The longest postoperative stay was 15 days after one procedure in which alcohol and bleomycin foam were used to embolize a tongue lesion. Swelling was seen only in the area treated with alcohol, not in the area treated with bleomycin foam. However, this alcohol-related swelling did lead to prolonged intubation. To facilitate safe treatment, three of 20 patients (15%) underwent preembolization tracheostomies in the presence of a large lesion and risk to airway. One patient had already undergone a tracheostomy before embolization as a result of obstructive symptoms. Treated patients were followed up in clinic 1 and 6 weeks after embolization. Mean duration of clinic follow-up was 66 days ⫾ 80, with a range of 4–403 days after embolization. Three of 20 patients (15%) did not return for 6-week follow-up and instead were contacted by phone at the time of manuscript preparation. Their follow-up averaged 753 days ⫾ 130, with a range of 604– 838 days. On follow-up, all 20 patients (100%) reported improvement in their symptoms after one session of embolization with bleomycin foam. All 14 patients (100%) who presented with pain noticed clinical improvement in pain score, decreasing from a highest score of 10 to a lowest score of 0. All 10 patients (100%) with obstructive symptoms and swallowing issues reported significant improvement in symptoms.
6
’
Bleomycin Foam Treatment of Venous Malformations
Ul Haq et al
Flagellate hyperpigmentation of the chest, back, ankles, and wrist developed in one patient treated with 22 U of bleomycin foam. It was partially resolved but remained visible at follow-up 40 days later. According to Society of Interventional Radiology (SIR) classifications for complications, there were minor complications in six of 27 procedures (22%) and six of 20 patients (30%) and major complications in two of 27 procedures (7%) and two of 20 patients (10%; Table 4). Of note, only one case of swelling associated with bleomycin alone was deemed significant enough to warrant treatment with steroids. Postprocedural MR imaging 6 weeks after embolization was available in 14 of 19 patients (74%). Lesion volume reduction on MR imaging was noted in 13 of 14 patients (93%). Percentages of lesion volume reduction ranged between 29% and 100%, with a mean of 66% ⫾ 21. Five of 14 patients (36%) had partial radiographic reduction, six (43%) had considerable reduction, and two (14%) had complete reduction. One patient had a 3% increase in lesion size on posttreatment MR imaging despite improvement in clinical symptoms. Enhancement on MR imaging after treatment was decreased in 11 of
’
JVIR
14 patients (79%), increased in two (14%), and stable in one (7%). T2 signal on MR imaging after embolization was decreased in 12 of 14 patients (86%) and stable in two (14%; Figs 2, 3).
DISCUSSION Bleomycin liquid has proven to be effective and safe in the treatment of VMs and lymphatic malformations and to produce minimal swelling and postprocedural pain. Because of this, it has been used to treat low-flow malformations in sensitive locations such as the orbit, airway, and face, where nerves are at risk from compressive injury (10,11,14,16–18). Even though bleomycin is commonly used in liquid form to treat low-flow malformations, we are aware of no published literature describing foamed bleomycin as a therapeutic agent. For the treatment of VMs, we use ethanol in locations distant from major nerves and where muscle contractures are unlikely. Wide availability, long shelf life of 36 months, ease of administration, comparatively low cost,
Figure 1. Bleomycin dose per treatment session.
Table 4 . Complications: Time of Presentation and Management Complication Minor Blisters on treated area
Time of Presentation
Treatment
1 wk after treatment
Triamcinolone acetonide dental paste
5 d after treatment
Triamcinolone acetonide dental paste
Pruritic rash on face shoulder and neck Flagellate erythema on chest and back
1 d after treatment 1–2 d after treatment
Diphenhydramine Topical moisturizer
Swelling of treated area (eyelid)
Blisters on treated area
1–2 d after treatment
None
Swelling and pain of treated area (right calf) Major
Day of treatment
Oxycodone/acetaminophen
Swelling of treated area (tongue, pharynx)
2 d after treatment
i.v. Steroids in ER and oral steroids for 2 wk
Day of treatment
i.v. Steroids (prolonged intubation)
Ethanol-related swelling of treated area (tongue) Note–Each complication occurred in a single patient. i.v. ¼ intravenous, ER ¼ emergency room.
Volume XX
’
Number X
’
Month
’
2015
7
Figure 2. Images of a tongue VM treated with bleomycin foam in a 16-year-old girl who presented with dysphagia, dysarthria, and airway obstruction. (a) Clinical image shows VM of the tongue (arrows), and (b) axial, (c) coronal, and (d) sagittal T2-weighted MR images with fat saturation demonstrate infiltration of the intrinsic muscles of the tongue with a T2-bright VM (arrows). After bleomycin foam embolization, the patient was tolerating solid foods and had no dysarthria or airway obstruction. (e) Clinical image shows reduced size of the VM (arrows), and (f) axial, (g) coronal, and (h) sagittal T2-weighted MR images with fat saturation demonstrate volume reduction in all axes (arrows).
and known metabolism and widely used sclerosant agent effective agent that directly denatures proteins, and leads
excretion makes alcohol a (7,8). Alcohol is a highly damages endothelial cells, to thrombosis and fibrosis,
ultimately obliterating VMs. However, because of its toxicity to cells, it leads to extreme pain and edema immediately after embolization. Time for resolution of edema is as long as 48 hours after the procedure, and
8
’
Bleomycin Foam Treatment of Venous Malformations
Ul Haq et al
’
JVIR
Figure 3. Images of a VM of the tongue and right-sided lower lip treated with bleomycin foam in a 36-year-old man who presented with frequent biting of swollen tongue and lip leading to dysphagia and bleeding. (a) Clinical image shows VM of lower lip (arrows), and (b) axial T2-weighted MR image with fat saturation demonstrates infiltration of the intrinsic muscles of the tongue with T2-bright VM (arrows). After one bleomycin foam embolization, the patient did not report any biting of the tongue or lip and also started tolerating solid food. (c) Clinical image shows reduced size of the lip component of the VM (arrows). (d) Axial T2-weighted images with fat saturation demonstrate significant volume reduction in the size and T2 signal of the tongue component of the lesion (arrows).
ethanol can take 2–6 months to reduce the size of the VM. Serious complications include nerve damage, ulceration, and cardiopulmonary collapse. Even though ethanol is a highly effective sclerosant agent (75%–95%), complication rates as high as 41.2% have been reported (17), with others reporting a lower complication rate of 27.9% (19). Another common agent used for sclerotherapy is STS. It is a detergent that acts on the lipid molecules in the cells of the VM wall, causing inflammatory destruction, thrombus formation, and eventually sclerosis. Complications reported in a prospective case series of 11 patients treated with intralesional STS (20) include cutaneous blistering, erosion, and crusting in seven patients (53.8%) and atrophic scarring in four (30.7%). Other complications
include temporary nerve damage, allergic reaction (21), and one case of monocular blindness (22). Tan et al (21) reported 70% of patients with symptomatic improvement, no major complications, and minor complications such as skin necrosis/ulcer, temporary paresthesia, and allergic reaction in 3.1% of sessions and 9.7% of patients. A prospective case series of 15 patients treated with STS percutaneous sclerotherapy by O’Donovan et al (21,23) reported a beneficial result in 13 patients (87%), with minor complications. Although STS is considered less effective than alcohol, it is considered to have a better safety profile (23). The mechanism of action of bleomycin is inhibition of DNA synthesis leading to cell death. The sclerosing
Volume XX
’
Number X
’
Month
’
2015
mechanism of bleomycin involves damage to endothelial cells with nonspecific inflammatory reaction and occlusion of vessels (24). Typically used to treat many kinds of cancer, bleomycin was first used by Yura et al (25) as a sclerosing agent for the treatment of lymphatic malformations in 1977. Bleomycin was later injected in aspirated cyst cavities of cervical lymphatic malformations by Tanigawa et al in 1987 (26). Pingyangmycin (bleomycin A5 hydrochloride) liquid has been used for the treatment of vascular malformations in China since 1991 (9,12,13). Bleomycin has also shown promising results for the treatment of micro- and macrocystic lymphatic malformations (10). Cutaneous manifestations from bleomycin such as hyperpigmentation and flagellate erythema can occur at total doses of 200–300 mg but have also been reported with doses as low as 15 mg (11,27). As an antineoplastic drug, bleomycin’s most severe complication is pulmonary fibrosis, which can occur with total lifetime doses 4 400 mg or 5 mg/kg. Therefore, during embolization, the concentration and dosage of drug are very important (11,12,27). Of note, to date, we are aware of no reported case of pulmonary fibrosis secondary to the use of bleomycin as a sclerosant agent. Toxicity risk with bleomycin is considered to be minimal if the dose is kept at or lower than 1 mg/kg per session, at an interval of not less than 2 weeks, with total dose limited to 5 mg/kg (13). In comparison with alcohol’s shelf life of 36 months, bleomycin has a longer shelf life of 42 months and is relatively inexpensive. It is very well tolerated, with minimal swelling and pain after sclerotherapy (14,18). A prospective case series of 82 patients with craniofacial VMs treated with intralesional pingyangmycin (13) showed 76%–100% regression in 81.7% of cases and 51%–75% regression in 14.6%. A previously published retrospective case series (17) compared bleomycin and alcohol sclerotherapy based on subjective clinical evaluation (unchanged/worse/better) recorded by the patient and physician, whereas the present study outcomes included clinical and radiographic categories. Interestingly, Spence et al (17) reported improvement in all patients treated with alcohol, compared with 82% of patients treated with bleomycin. The present results showed positive clinical outcomes in all 20 patients (100%) treated with bleomycin foam. We also evaluated pre- and postoperative MR imaging, which showed decreased lesion size in 13 of 14 patients (93%), decreased enhancement in 11 patients (79%), and decreased T2 signal intensity in 12 patients (86%). The present sample size was too small to allow correlation of clinical outcomes with these three changes on MR imaging. Also, of note, VM size on MR imaging does not always correlate with symptomatic improvement (28). Although not reported according to SIR guidelines, the study by Spence et al (17) reported that complications developed in seven of 17 patients (41.2%) treated with alcohol, as opposed to none in the matched
9
bleomycin pair. Among the 27 treatments in the present study, there were six minor complications (22%) and two major complications (7%). Despite the small sample size, our complication rates with the use of bleomycin foam were low compared with reported complications with alcohol (27.9%–41.2%) (17,19). Limitations of the present study include that it is a retrospective study with a limited number of patients. Clinical outcomes were derived from electronic medical records, and, with the exception of pain scoring, objective/quantifiable measurements of clinical outcomes were not possible. In addition, there was no direct comparison of bleomycin foam versus other sclerosing agents. A direct comparison would allow a better assessment of clinical and radiographic outcomes and complication rates, and would eliminate possible selection bias. Also, posttreatment MR imaging was not available in all patients, further weakening the imaging-based outcome assessment. In addition, MR imaging assessment was not fully quantitative, as enhancement and T2 signal intensity were assessed subjectively. In conclusion, the use of foamed bleomycin is safe and effective for the percutaneous treatment of VMs. A secondary benefit is that the foam may help solve the problem of bleomycin dose limitation because it provides more volume per unit of drug, which is helpful to cover large vascular malformations. This could be further evaluated in a side-by-side study between bleomycin liquid and foam. To avoid lifetime dose limitations, the use of bleomycin should be reserved for locations where postprocedural swelling would be dangerous, such as near the facial nerve, in or around the airway, in the orbit, or in closed compartments such as the forearm.
ACKNOWLEDGMENT The authors thank Dr. William Sheils for his development work on bleomycin foam presented at the International Society for the Study of Vascular Anomalies 2012.
REFERENCES 1. Lowe LH, Marchant TC, Rivard DC, Scherbel AJ. Vascular malformations: classification and terminology the radiologist needs to know. Semin Roentgenol 2012; 47:106–117; http://dx.doi.org/10.1053/j.ro.2011. 11.002, PubMed PMID: 22370189. 2. Vikkula M, Boon LM, Mulliken JB. Molecular genetics of vascular malformations. Matrix Biol 2001; 20:327–335, PubMed PMID: 11566267. 3. Greene AK, Alomari AI. Management of venous malformations. Clin Plast Surg 2011; 38:83–93; http://dx.doi.org/10.1016/j.cps.2010. 08.003, PubMed PMID: 21095474. 4. 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; http://dx.doi.org/10.1007/s00247-002-08 38-9, PubMed PMID: 12557065. 5. Sachin K, Rashmi S, Manish S, Siddhartha W, Uday L. Haemangiomas and venous malformations of the head and neck: A retrospective analysis of endovascular management in 358 patients. Indian J Plast Surg 2013109–116; http://dx.doi.org/10.4103/0970-0358.113727, PubMed PMID: 23960315; PubMed Central PMCID: PMC3745094.
10
’
Bleomycin Foam Treatment of Venous Malformations
6. Thawait SK, Puttgen K, Carrino JA, et al. MR imaging characteristics of soft tissue vascular anomalies in children. Eur J Pediatr 2013; 172:591– 600; http://dx.doi.org/10.1007/s00431-012-1828-z, PubMed PMID: 22986771. 7. Legiehn GM, Heran MK. Classification, diagnosis, and interventional radiologic management of vascular malformations. Orthop Clin North Am 2006; 37:435–474, vii-viii; http://dx.doi.org/10.1016/j.ocl.2006.04.005, Pub Med PMID: 16846771. 8. Legiehn GM, Heran MK. Venous malformations: classification, development, diagnosis, and interventional radiologic management. Radiol Clin North Am 2008; 46:545–597; http://dx.doi.org/10.1016/j.rcl.2008. 02.008, PubMed PMID: 18707962. 9. Zheng JW, Yang XJ, Wang YA, He Y, Ye WM, Zhang ZY. Intralesional injection of Pingyangmycin for vascular malformations in oral and maxillofacial regions: an evaluation of 297 consecutive patients. Oral Oncol 2009; 45:872–876; http://dx.doi.org/10.1016/j.oraloncology.2009. 02.011, PubMed PMID: 19628423. 10. Yang Y, Sun M, Ma Q, et al. Bleomycin A5 sclerotherapy for cervicofacial lymphatic malformations. J Vasc Surgery 2011; 53:150 155; http://dx.doi.org/10.1016/j.jvs.2010.07.019, PubMed PMID: 20843632. 11. Muir T, Kirsten M, Fourie P, Dippenaar N, Ionescu GO. Intralesional bleomycin injection (IBI) treatment for haemangiomas and congenital vascular malformations. Pediatr Surg Int 2004; 19:766–773; http://dx.doi. org/10.1007/s00383-003-1058-6, PubMed PMID: 14740248. 12. Zhao JH, Zhang WF, Zhao YF. Sclerotherapy of oral and facial venous malformations with use of pingyangmycin and/or sodium morrhuate. Int J Oral Maxillofac Surg 2004; 33:463–466; http://dx.doi.org/10.1016/j.ijom. 2003.10.003, PubMed PMID: 15183410. 13. Zhi K, Wen Y, Li L, Ren W. The role of intralesional Pingyangmycin in the treatment of venous malformation of facial and maxillary region. Int J Pediatr Otorhinolaryngol 2008; 72:593–597; http://dx.doi.org/10.1016/ j.ijporl.2008.01.007, PubMed PMID: 18294701. 14. Hou R, Guo J, Hu K, et al. A clinical study of ultrasound-guided intralesional injection of bleomycin A5 on venous malformation in cervical-facial region in China. J Vasc Surg 2010; 51:940–945; http://dx. doi.org/10.1016/j.jvs.2009.11.038, PubMed PMID: 20347690. 15. Dompmartin A, Acher A, Thibon P, et al. Association of localized intravascular coagulopathy with venous malformations. Arch Dermatol 2008; 144:873–877; http://dx.doi.org/10.1001/archderm.144.7.873 PubMed PMID: 18645138. 16. Bajpai H, Bajpai S. Comparative analysis of intralesional sclerotherapy with sodium tetradecyl sulfate versus bleomycin in the management of low flow craniofacial soft tissue vascular lesions. J Maxillofac Oral Surg 2012; 11:13– 20; http://dx.doi.org/10.1007/s12663-011-0325-7, PubMed PMID: 23449774; PubMed Central PMCID: PMC3319813.
Ul Haq et al
’
JVIR
17. Spence J, Krings T, TerBrugge KG, Agid R. Percutaneous treatment of facial venous malformations: a matched comparison of alcohol and bleomycin sclerotherapy. Head Neck 2011; 33:125–130; http://dx.doi.org/ 10.1002/hed.21410, PubMed PMID: 20848431. 18. Zhang J, Li HB, Zhou SY, et al. Comparison between absolute ethanol and bleomycin for the treatment of venous malformation in children. Experimental and therapeutic medicine 2013; 6:305–309; http://dx.doi.org/ 10.3892/etm.2013.1144, PubMed PMID: 24137179; PubMed Central PMCID: PMC3786900. 19. Berenguer B, Burrows PE, Zurakowski D, Mulliken JB. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg 1999; 104:1–11, discussion 2-5. PubMed PMID: 10597669. 20. Khandpur S, Sharma VK. Utility of intralesional sclerotherapy with 3% sodium tetradecyl sulphate in cutaneous vascular malformations. Dermatol Surg 2010; 36:340–346; http://dx.doi.org/10.1111/j.1524-4725. 2009.01440.x, PubMed PMID: 20100267. 21. 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; http://dx.doi.org/10.1016/j.jvir.2006.12. 735, PubMed PMID: 17377179. 22. Siniluoto TM, Svendsen PA, Wikholm GM, Fogdestam I, Edstrom S. Percutaneous sclerotherapy of venous malformations of the head and neck using sodium tetradecyl sulphate (sotradecol). Scand J Plast Reconstr Surg Hand Surg 1997; 31:145–150, PubMed PMID: 9232699. 23. O’Donovan JC, Donaldson JS, Morello FP, Pensler JM, Vogelzang RL, Bauer B. Symptomatic hemangiomas and venous malformations in infants, children, and young adults: treatment with percutaneous injection of sodium tetradecyl sulfate. AJR Am J Roentgenol 1997; 169:723–729; http://dx.doi.org/10.2214/ajr.169.3.9275886, PubMed PMID: 9275886. 24. Umezawa H. Recent study on biochemistry and action of Bleomycin. In: Carter SK, Crook ST, Umezawa H, editors. Bleomycin: Current Status and New Developments. New York, NY: Academic Press; 1978. p. 15–20. 25. Yura J, Hashimoto T, Tsuruga N, Shibata K. Bleomycin treatment for cystic hygroma in children. Nippon Geka Hokan 1977; 46:607–614, PubMed PMID: 73364. 26. Tanigawa N, Shimomatsuya T, Takahashi K, et al. Treatment of cystic hygroma and lymphangioma with the use of bleomycin fat emulsion. Cancer 1987; 60:741–749, PubMed PMID: 2439189. 27. Mowad CM, Nguyen TV, Elenitsas R, Leyden JJ. Bleomycin-induced flagellate dermatitis: a clinical and histopathological review. Br J Dermatol 1994; 131:700–702, PubMed PMID: 7528043. 28. Vogelzang RL, Atassi R, Vouche M, Resnick S, Salem R. Ethanol embolotherapy of vascular malformations: clinical outcomes at a single center. J Vasc Interv Radiol 2014; 25:206–213; http://dx.doi.org/10.1016/ j.jvir.2013.10.055, PubMed PMID: 24461130.
Bleomycin Foam Treatment of Venous Malformations: A Promising Agent for Effective Treatment with Minimal Swelling Faheem Ul Haq, MD, Sally E. Mitchell, MD, Aylin Tekes, MD, and Clifford R. Weiss, MD
ABSTRACT Purpose: To report clinical and radiographic outcomes of patients with venous malformations (VMs) treated with bleomycin foam. Materials and Methods: Twenty patients (age, 2–68 y) presented with symptoms of swelling (n ¼ 19; 95%), pain (n ¼ 14; 70%), and bleeding (n ¼ 4; 20%). Lesions were located in the head and neck in 17 patients (85%), extremities in two (10%), and mediastinum in one (5%). Twenty-seven embolizations were performed, with a mean of 1.7 ⫾ 1.0 treatments per patient (range, 1–4). An average of 0.45 ⫾ 0.4 U/kg of bleomycin foam was used per procedure, with a range of 0.1–2.3 U/kg. Results: All procedures were technically successful with no intraprocedural complications. Mean follow-up was 66 days ⫾ 80, with a range of 4–403 days. Postprocedure complications were minor in 6 of 27 procedures (22%) and major in 2 of 27 procedures (7%). All 20 patients (100%) reported improvement in their symptoms after a single treatment session. Postprocedural magnetic resonance (MR) imaging demonstrated volume reduction of treated lesions in 13 of 14 patients (93%), with a mean lesion volume reduction of 66% ⫾ 21. Enhancement on MR imaging after treatment was decreased in 11 of 14 patients (79%), increased in two (14%), and stable in one (7%). T2 signal intensity on MR imaging after treatment was decreased in 12 of 14 patients (86%) and stable in two (14%). Conclusions: The use of bleomycin foam for the percutaneous treatment of VMs is safe and effective. Foaming bleomycin may be used to address the dose limitations of the liquid.
ABBREVIATIONS STS = sodium tetradecyl sulfate, VM = venous malformation
Venous malformations (VMs) are the most common type of vascular malformation, with an incidence of one to two per 10,000 births and a prevalence as high as 1% in the general population (1,2). Current mainstay treatment for VM is percutaneous sclerotherapy (3–6). Historically, a variety of sclerosant agents have been used to treat From the Divisions of Cardiovascular and Interventional Radiology (F.U.H., S.E.M., C.R.W.) and Pediatric Radiology and Pediatric Neuroradiology (A.T.), Russell H. Morgan Department of Radiology and Radiologic Science, The Johns Hopkins University School of Medicine, 601 N. Caroline St., Room 3125, Baltimore, MD 21287. Received August 22, 2014; final revision received April 27, 2015; accepted May 2, 2015. Address correspondence to S.E.M.; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.05.007
these lesions, but there is currently no consensus as to the best agent. Most agents cause destruction of endothelium lining the VM and lead to thrombosis, induce scarring, and decrease the size of the lesion (7–9). Recently, bleomycin liquid has emerged as a promising sclerosant agent for VMs and lymphatic malformations, particularly in areas where swelling needs to be minimized (10). Compared with 100% alcohol and sodium tetradecyl sulfate (STS; Sotradecol; AngioDynamics, Latham, New York), bleomycin does not lead to skin and mucosal necrosis and can be used without general anesthesia (11– 13). Minimal side effects, fast recovery, good outcomes, and low cost have led to increased interest in bleomycin as a sclerosant agent. In China, a variant of bleomycin, pingyangmycin (bleomycin A5 hydrochloride; Tianjin Taihe, Tianjin, China), is commonly used for sclerotherapy (14). However, lifetime dose is a limitation when using bleomycin to treat large VMs. The current recommended
2
’
Bleomycin Foam Treatment of Venous Malformations
lifetime dose is less than 400 mg or 5 mg/kg to reduce the risk of pulmonary fibrosis (11). Because of this dose limitation, bleomycin foam has been proposed to allow treatment of larger lesions while minimizing the total dose of bleomycin. The purpose of the present study is to report clinical and radiographic outcomes in patients with VMs treated with bleomycin foam.
MATERIALS AND METHODS After institutional review board approval, we performed retrospective review of electronic medical records of 20
Ul Haq et al
’
JVIR
patients (14 female and six male) with VMs treated with bleomycin foam at our institution between June 2012 and October 2013 (Table 1). Mean age of patients was 32 years ⫾ 16 (range, 2–68 y). Presenting symptoms were swelling (n ¼ 19; 95%), pain (n ¼ 14; 70%), and bleeding (n ¼ 4; 20%). Pain was assessed on a 10-point visual analog scale. Seventeen patients (85%) had lesions in the head and neck region (in or adjacent to the airway or tongue) causing symptoms such as dysphagia (n ¼ 10; 50%), dyspnea (n ¼ 4; 20%), and dysarthria (n ¼ 4; 20%) in addition to pain and swelling. Two patients (10%) had lesions in the extremities, and one (5%) presented with a mediastinal lesion that caused chest pain and swelling.
Table 1 . Patient Demographics, Venous Malformation Location, Symptoms, and Previous Treatments Pt. No./Sex/ Age (y) 1/F/23
Primary
Additional
Previous
Previous
Location
Symptoms
Symptoms
Treatments
Sclerosant Used
Head and neck
Pain, swelling
–
7 Sessions
Ethanol/ethiodol mixture (80%/ 20%), 50%, 75%, 95%, 100% alcohol
2/M/36
Head and neck
Swelling, bleeding
Dysphagia
11 Sessions
4:1 Dehydrated alcohol/ethiodol mixture, 100%
3/M/68
Head and neck
Swelling
Dysphagia,
3 Sessions
alcohol 100% Alcohol
dysarthria, airway 4/M/47
Chest
Pain, swelling
obstruction Chest tightness
No
None
5/F/18
Head and neck
Swelling
Dysphagia,
No
None
dysarthria, airway obstruction 6/F/25
Head and neck
Swelling
–
No
None
7/F/21 8/F/2
RLE Head and neck
Pain Swelling
– –
No No
None None
9/F/57
Head and neck
Pain, swelling
Dysphagia, airway
1 Session
STS foam
obstruction 10/F/20
RLE
Pain, swelling
No
None
11/M/33
Head and neck
Pain, swelling,
Dysphagia, airway
No
None
12/F/50
Head and neck
bleeding Pain, swelling
obstruction Dysphagia,
No
None
Loosening of teeth Dysphagia
No No
None None
dysarthria 13/F/30 14/F/39
Head and neck Head and neck
Swelling Pain, swelling
15/F/49
Head and neck
Pain, swelling
–
No
None
16/F/29
Head and neck
Pain, swelling
Dysphagia, dysarthria
No
None
17/F/18
Head and neck
Pain, swelling
–
No
None
18/M/17
Head and neck
Pain, swelling
–
10 Sessions
100% Alcohol, STS, and Onyx
19/F/41
Head and neck
Pain, swelling,
Dysphagia
2 Sessions
75% Ethanol, 100%
Head and neck
bleeding Pain, swelling,
Dysphagia
No
ethanol None
20/M/12
bleeding RLE ¼ right lower extremity, STS ¼ sodium tetradecyl sulfate.
Volume XX
’
Number X
’
Month
’
2015
3
Six patients (30%) had a history of previous embolizations of the same lesion with agents other than bleomycin. Patients underwent one (n ¼ 15; 75%), two (n ¼ 4; 20%), or four (n ¼ 1; 5%) treatment sessions of bleomycin foam embolization. Each patient was evaluated by the vascular anomalies team. Preprocedural imaging included magnetic resonance (MR) imaging of the involved body part per our institutional vascular anomalies protocol on 1.5- or 3.0T scanners (Avanto/Verio; Siemens, Erlangen, Germany). The following sequences were obtained: triplanar T2-weighted imaging with fat saturation, precontrast axial T1-weighted imaging, time-resolved contrastenhanced MR angiography, and postcontrast triplanar T1-weighted imaging with fat saturation. Time-resolved contrast-enhanced MR angiography was performed in the coronal plane following intravenous administration of contrast material at 0.03 mmol/kg body weight (ABLAVAR; Lantheus Medical Imaging, North Billerica, Massachusetts). Contrast agent was administered at 2 mL/s, followed by a 10-mL saline solution flush at a rate of 1 mL/s. Nineteen of 20 patients (95%) had MR imaging before embolization and one underwent computed tomography. Posttreatment MR imaging was available in 14 patients (74%). Preprocedural blood tests included fibrinogen and Ddimer measurements, prothrombin time, International Normalized Ratio, and activated partial thromboplastin time to assess the possibility of localized intravascular coagulopathy, which is a concern in patients with VMs because it predisposes the patient to peritreatment disseminated intravascular coagulopathy (15). If localized intravascular coagulopathy was found, the patient was administered anticoagulation, typically enoxaparin sodium (Lovenox; Sanofi, Bridgewater, New Jersey), for 10 days before and after the procedure to prevent disseminated intravascular coagulopathy. All embolization procedures were performed under general anesthesia. Ultrasound (US) in combination with direct vision was used to access the VMs in 12 of 27 embolizations (45%), US alone was used in six (22%), and direct vision alone was used in nine (33%). Access was obtained with a 21-gauge needle (Cook, Bloomington, Indiana) connected to microtubing with 0.2 mL dead space (Plexufix; B. Braun, Bethlehem, Pennsylvania) connected to a 3-mL syringe. Digital subtraction venography was performed at one film per second (Allura; Philips, Best, The Netherlands) before
embolization to assess the size of the accessed portion(s) of the VM and anatomy of draining veins. The contrast agents used were iohexol (Omnipaque; GE Healthcare, Princeton, New Jersey) for children and ioxilan (Oxilan; Guerbet, Villepinte, France) for adults. Bleomycin foam was prepared by using a ratio of 6 U bleomycin suspended in 1 mL normal saline solution plus 1 mL human 25% serum albumin. We used a 2:1 air:liquid ratio to create foam (Table 2). Bleomycin foam was injected into each site under digital subtraction venography or US guidance. Volume of injected foam was determined by contrast agent displacement technique, during which foam displaces previously injected contrast agent. Manual compression was applied for 3–5 minutes to compress venous outflow when present. After a 20-minute dwell time, sites with residual blood flow were checked with digital subtraction venography to determine the size of residual VMs and were treated again as needed. After a second 20-minute dwell time, all needles were removed. Intraprocedural nerve monitoring was used during embolizations when there was a potential risk of nerve damage, defined as a VM o 1 cm from a major nerve on MR imaging. This was performed by using custom neurophysiologic protocols based on combinations of electromyography, motor and sensory nerve conduction studies, somatosensory-evoked potentials, and motor-evoked potentials. Intraoperative nerve monitoring was performed in nine of 27 procedures (45%; eight head and neck VMs and one right lowerextremity VM). Posttreatment clinical outcomes were classified as improved, unchanged, or worse. To assess treatment response on MR imaging, the treated portion of the VM was measured in three dimensions on pre- and postprocedural MR imaging by three radiologists (A.T., S.E. M., and C.R.W.). The volume of each lesion was calculated before and after embolization. Because VMs have complex shapes, each lesion was assumed as ellipsoid and was calculated based on the ellipsoid volume formula 4/3πabc (with a as anteroposterior axis, b as transverse axis, and c as craniocaudal axis). Treatment response on MR imaging was classified as complete reduction (4 90%), considerable reduction (60%– 90%), partial reduction (20%–60%), unchanged (o 20%), or increased. Subjective analysis of T2 signal intensity and enhancement strength was also assessed for each treated VM before and after treatment.
Table 2 . Bleomycin Foam Preparation Bleomycin Dose (U)
Saline Solution (mL)
Human Serum Albumin 25% (mL)
Bleomycin Liquid Volume (mL)
Air (mL)
Total Bleomycin Foam Volume (mL)
15 30
2.5 5
2.5 5
5 10
10 20
15 30
45
7.5
7.5
15
30
45
Note–Bleomycin liquid was prepared by the hospital pharmacy in three doses meant for small, medium, and large venous malformations. The albumin is what catalyzes the foaming process. Bleomycin mix was foamed with air in a 1:2 ratio.
4
1 2 3
No.
Location
Bleomycin
Other Agents, Dose
Nerve
Procedural
Discharged
Clinical
Radiographic
Dose (U)
(mL)
Monitoring
Complications
After
Outcome
Reduction (%)
Improved
55, Partial 76, Considerable NA
1
Head and neck
30
–
No
No
1d
2
Head and neck
60
–
Yes
No
Id
1 1
Head and neck Head and neck
38 30
– –
No No
No No
4d 1d
Improved Improved
4
1
Mediastinum
18
–
No
No
1d
Improved
29, Partial
5
1
Head and neck
9
Alcohol at front of tongue, bleomycin
No
No
15 d
Improved
81, Considerable
2 3
Head and neck Head and neck
14 15
No No
No No
1d 1d
at back of tongue (7) – –
4
Head and neck
60
–
No
No
1d
6
1
Head and neck
15
Alcohol on lip, bleomycin on eyelid
No
No
2d
Improved
NA
7 8
1 1
RLE Head and neck
45 30
– –
No Yes
No No
1d 1d
Improved Improved
61, Considerable 98, Complete
9
1
Head and neck
15
Alcohol on face and
No
No
1d
Improved
100, Complete
No
No
1d
(7.5)
Bleomycin Foam Treatment of Venous Malformations
Procedure Pt. No.
’
Table 3 . Clinical and Radiographic Outcomes of Bleomycin Embolization
neck, bleomycin on tongue and tonsil (15.5) 2
Head and neck
22
Alcohol on face and neck, bleomycin on tongue and tonsil
10
1
RLE/bottom of
15
(22.5) –
Yes
No
1d
Improved*
NA
15 30
– STS on neck,
No No
No No
1d 1d
Improved
81, Considerable
No
No
1d
Improved
51, Partial
No
No
1d
Improved
75, Considerable
Improved*
NA
foot 11
1 2
Head and neck Head and neck
bleomycin on 1
Head and neck
30
13
1
Head and neck
15
STS and bleomycin
2
Head and neck
30
(4) –
No
No
1d
1
Head and neck
12
–
Yes
No
1d
14
’
(Continued)
Ul Haq et al
12
oropharynx (24) –
JVIR
NA
RESULTS
NA ¼ not applicable, RLE ¼ right lower extremity, STS ¼ sodium tetradecyl sulfate. Long-term follow-up performed by phone call. n
44, Partial
No
5
Access 2007 (Microsoft, Redmond, Washington) was used to enter the data. Most data reported are frequencies. Stata/IC 12.0 (STATA, College Station, Texas) for Windows (Microsoft) was used to calculate means and standard deviations.
Improved
Improved
2015
1d
1d
’
No
Yes
– 15 1 20
Head and neck
1 19
Head and neck
38
–
No
41, Partial
NA 64, Considerable Improved* Improved 1d 1d Yes Yes – – 1 1 17 18
Chest Head and neck
15 24
Yes
No No
3, Increased Improved
Improved 1d
1d Yes
30 1
Head and neck
1
16
Head and neck
30
–
No
Month
15
Radiographic
’
–
No
Outcome
Clinical Discharged
After Complications
Procedural Nerve
Number X
Monitoring
Other Agents, Dose
’
(mL) Dose (U)
Bleomycin
Location Procedure
No. Pt. No.
Table 3. Clinical and Radiographic Outcomes of Bleomycin Embolization (continued )
Reduction (%)
Volume XX
A total of 27 embolizations were performed, with a mean of 1.7 ⫾ 1.0 and a range of one to four treatments per patient (Table 3). Twenty-four embolizations (89%) were in the head and neck, two (7%) were in the right lower extremity, and one (4%) was in the chest. Bleomycin foam was the only sclerosant agent in 21 procedures. In five procedures, an additional sclerosant agent was used at a separate location from the one treated with bleomycin: four used 100% alcohol and one used 3% foamed STS. Finally, in one procedure, bleomycin foam and STS foam were used to treat the same lesion. It was necessary to add these secondary sclerosant agents because the preordered bleomycin dose was depleted before the embolization was complete. An average of 0.45 U/kg ⫾ 0.4 of bleomycin foam (26 U ⫾ 14) was used per procedure, with a range of 0.1–2.3 U/kg (9–60 U; Fig 1). All procedures were technically successful. There were no intraprocedural complications. Postoperative hospital stay was overnight in 24 of 27 procedures (89%). After one procedure in which only bleomycin foam was used, there was a 4-day hospital stay as a result of swelling that requiring extended intubation. The longest postoperative stay was 15 days after one procedure in which alcohol and bleomycin foam were used to embolize a tongue lesion. Swelling was seen only in the area treated with alcohol, not in the area treated with bleomycin foam. However, this alcohol-related swelling did lead to prolonged intubation. To facilitate safe treatment, three of 20 patients (15%) underwent preembolization tracheostomies in the presence of a large lesion and risk to airway. One patient had already undergone a tracheostomy before embolization as a result of obstructive symptoms. Treated patients were followed up in clinic 1 and 6 weeks after embolization. Mean duration of clinic follow-up was 66 days ⫾ 80, with a range of 4–403 days after embolization. Three of 20 patients (15%) did not return for 6-week follow-up and instead were contacted by phone at the time of manuscript preparation. Their follow-up averaged 753 days ⫾ 130, with a range of 604– 838 days. On follow-up, all 20 patients (100%) reported improvement in their symptoms after one session of embolization with bleomycin foam. All 14 patients (100%) who presented with pain noticed clinical improvement in pain score, decreasing from a highest score of 10 to a lowest score of 0. All 10 patients (100%) with obstructive symptoms and swallowing issues reported significant improvement in symptoms.
6
’
Bleomycin Foam Treatment of Venous Malformations
Ul Haq et al
Flagellate hyperpigmentation of the chest, back, ankles, and wrist developed in one patient treated with 22 U of bleomycin foam. It was partially resolved but remained visible at follow-up 40 days later. According to Society of Interventional Radiology (SIR) classifications for complications, there were minor complications in six of 27 procedures (22%) and six of 20 patients (30%) and major complications in two of 27 procedures (7%) and two of 20 patients (10%; Table 4). Of note, only one case of swelling associated with bleomycin alone was deemed significant enough to warrant treatment with steroids. Postprocedural MR imaging 6 weeks after embolization was available in 14 of 19 patients (74%). Lesion volume reduction on MR imaging was noted in 13 of 14 patients (93%). Percentages of lesion volume reduction ranged between 29% and 100%, with a mean of 66% ⫾ 21. Five of 14 patients (36%) had partial radiographic reduction, six (43%) had considerable reduction, and two (14%) had complete reduction. One patient had a 3% increase in lesion size on posttreatment MR imaging despite improvement in clinical symptoms. Enhancement on MR imaging after treatment was decreased in 11 of
’
JVIR
14 patients (79%), increased in two (14%), and stable in one (7%). T2 signal on MR imaging after embolization was decreased in 12 of 14 patients (86%) and stable in two (14%; Figs 2, 3).
DISCUSSION Bleomycin liquid has proven to be effective and safe in the treatment of VMs and lymphatic malformations and to produce minimal swelling and postprocedural pain. Because of this, it has been used to treat low-flow malformations in sensitive locations such as the orbit, airway, and face, where nerves are at risk from compressive injury (10,11,14,16–18). Even though bleomycin is commonly used in liquid form to treat low-flow malformations, we are aware of no published literature describing foamed bleomycin as a therapeutic agent. For the treatment of VMs, we use ethanol in locations distant from major nerves and where muscle contractures are unlikely. Wide availability, long shelf life of 36 months, ease of administration, comparatively low cost,
Figure 1. Bleomycin dose per treatment session.
Table 4 . Complications: Time of Presentation and Management Complication Minor Blisters on treated area
Time of Presentation
Treatment
1 wk after treatment
Triamcinolone acetonide dental paste
5 d after treatment
Triamcinolone acetonide dental paste
Pruritic rash on face shoulder and neck Flagellate erythema on chest and back
1 d after treatment 1–2 d after treatment
Diphenhydramine Topical moisturizer
Swelling of treated area (eyelid)
Blisters on treated area
1–2 d after treatment
None
Swelling and pain of treated area (right calf) Major
Day of treatment
Oxycodone/acetaminophen
Swelling of treated area (tongue, pharynx)
2 d after treatment
i.v. Steroids in ER and oral steroids for 2 wk
Day of treatment
i.v. Steroids (prolonged intubation)
Ethanol-related swelling of treated area (tongue) Note–Each complication occurred in a single patient. i.v. ¼ intravenous, ER ¼ emergency room.
Volume XX
’
Number X
’
Month
’
2015
7
Figure 2. Images of a tongue VM treated with bleomycin foam in a 16-year-old girl who presented with dysphagia, dysarthria, and airway obstruction. (a) Clinical image shows VM of the tongue (arrows), and (b) axial, (c) coronal, and (d) sagittal T2-weighted MR images with fat saturation demonstrate infiltration of the intrinsic muscles of the tongue with a T2-bright VM (arrows). After bleomycin foam embolization, the patient was tolerating solid foods and had no dysarthria or airway obstruction. (e) Clinical image shows reduced size of the VM (arrows), and (f) axial, (g) coronal, and (h) sagittal T2-weighted MR images with fat saturation demonstrate volume reduction in all axes (arrows).
and known metabolism and widely used sclerosant agent effective agent that directly denatures proteins, and leads
excretion makes alcohol a (7,8). Alcohol is a highly damages endothelial cells, to thrombosis and fibrosis,
ultimately obliterating VMs. However, because of its toxicity to cells, it leads to extreme pain and edema immediately after embolization. Time for resolution of edema is as long as 48 hours after the procedure, and
8
’
Bleomycin Foam Treatment of Venous Malformations
Ul Haq et al
’
JVIR
Figure 3. Images of a VM of the tongue and right-sided lower lip treated with bleomycin foam in a 36-year-old man who presented with frequent biting of swollen tongue and lip leading to dysphagia and bleeding. (a) Clinical image shows VM of lower lip (arrows), and (b) axial T2-weighted MR image with fat saturation demonstrates infiltration of the intrinsic muscles of the tongue with T2-bright VM (arrows). After one bleomycin foam embolization, the patient did not report any biting of the tongue or lip and also started tolerating solid food. (c) Clinical image shows reduced size of the lip component of the VM (arrows). (d) Axial T2-weighted images with fat saturation demonstrate significant volume reduction in the size and T2 signal of the tongue component of the lesion (arrows).
ethanol can take 2–6 months to reduce the size of the VM. Serious complications include nerve damage, ulceration, and cardiopulmonary collapse. Even though ethanol is a highly effective sclerosant agent (75%–95%), complication rates as high as 41.2% have been reported (17), with others reporting a lower complication rate of 27.9% (19). Another common agent used for sclerotherapy is STS. It is a detergent that acts on the lipid molecules in the cells of the VM wall, causing inflammatory destruction, thrombus formation, and eventually sclerosis. Complications reported in a prospective case series of 11 patients treated with intralesional STS (20) include cutaneous blistering, erosion, and crusting in seven patients (53.8%) and atrophic scarring in four (30.7%). Other complications
include temporary nerve damage, allergic reaction (21), and one case of monocular blindness (22). Tan et al (21) reported 70% of patients with symptomatic improvement, no major complications, and minor complications such as skin necrosis/ulcer, temporary paresthesia, and allergic reaction in 3.1% of sessions and 9.7% of patients. A prospective case series of 15 patients treated with STS percutaneous sclerotherapy by O’Donovan et al (21,23) reported a beneficial result in 13 patients (87%), with minor complications. Although STS is considered less effective than alcohol, it is considered to have a better safety profile (23). The mechanism of action of bleomycin is inhibition of DNA synthesis leading to cell death. The sclerosing
Volume XX
’
Number X
’
Month
’
2015
mechanism of bleomycin involves damage to endothelial cells with nonspecific inflammatory reaction and occlusion of vessels (24). Typically used to treat many kinds of cancer, bleomycin was first used by Yura et al (25) as a sclerosing agent for the treatment of lymphatic malformations in 1977. Bleomycin was later injected in aspirated cyst cavities of cervical lymphatic malformations by Tanigawa et al in 1987 (26). Pingyangmycin (bleomycin A5 hydrochloride) liquid has been used for the treatment of vascular malformations in China since 1991 (9,12,13). Bleomycin has also shown promising results for the treatment of micro- and macrocystic lymphatic malformations (10). Cutaneous manifestations from bleomycin such as hyperpigmentation and flagellate erythema can occur at total doses of 200–300 mg but have also been reported with doses as low as 15 mg (11,27). As an antineoplastic drug, bleomycin’s most severe complication is pulmonary fibrosis, which can occur with total lifetime doses 4 400 mg or 5 mg/kg. Therefore, during embolization, the concentration and dosage of drug are very important (11,12,27). Of note, to date, we are aware of no reported case of pulmonary fibrosis secondary to the use of bleomycin as a sclerosant agent. Toxicity risk with bleomycin is considered to be minimal if the dose is kept at or lower than 1 mg/kg per session, at an interval of not less than 2 weeks, with total dose limited to 5 mg/kg (13). In comparison with alcohol’s shelf life of 36 months, bleomycin has a longer shelf life of 42 months and is relatively inexpensive. It is very well tolerated, with minimal swelling and pain after sclerotherapy (14,18). A prospective case series of 82 patients with craniofacial VMs treated with intralesional pingyangmycin (13) showed 76%–100% regression in 81.7% of cases and 51%–75% regression in 14.6%. A previously published retrospective case series (17) compared bleomycin and alcohol sclerotherapy based on subjective clinical evaluation (unchanged/worse/better) recorded by the patient and physician, whereas the present study outcomes included clinical and radiographic categories. Interestingly, Spence et al (17) reported improvement in all patients treated with alcohol, compared with 82% of patients treated with bleomycin. The present results showed positive clinical outcomes in all 20 patients (100%) treated with bleomycin foam. We also evaluated pre- and postoperative MR imaging, which showed decreased lesion size in 13 of 14 patients (93%), decreased enhancement in 11 patients (79%), and decreased T2 signal intensity in 12 patients (86%). The present sample size was too small to allow correlation of clinical outcomes with these three changes on MR imaging. Also, of note, VM size on MR imaging does not always correlate with symptomatic improvement (28). Although not reported according to SIR guidelines, the study by Spence et al (17) reported that complications developed in seven of 17 patients (41.2%) treated with alcohol, as opposed to none in the matched
9
bleomycin pair. Among the 27 treatments in the present study, there were six minor complications (22%) and two major complications (7%). Despite the small sample size, our complication rates with the use of bleomycin foam were low compared with reported complications with alcohol (27.9%–41.2%) (17,19). Limitations of the present study include that it is a retrospective study with a limited number of patients. Clinical outcomes were derived from electronic medical records, and, with the exception of pain scoring, objective/quantifiable measurements of clinical outcomes were not possible. In addition, there was no direct comparison of bleomycin foam versus other sclerosing agents. A direct comparison would allow a better assessment of clinical and radiographic outcomes and complication rates, and would eliminate possible selection bias. Also, posttreatment MR imaging was not available in all patients, further weakening the imaging-based outcome assessment. In addition, MR imaging assessment was not fully quantitative, as enhancement and T2 signal intensity were assessed subjectively. In conclusion, the use of foamed bleomycin is safe and effective for the percutaneous treatment of VMs. A secondary benefit is that the foam may help solve the problem of bleomycin dose limitation because it provides more volume per unit of drug, which is helpful to cover large vascular malformations. This could be further evaluated in a side-by-side study between bleomycin liquid and foam. To avoid lifetime dose limitations, the use of bleomycin should be reserved for locations where postprocedural swelling would be dangerous, such as near the facial nerve, in or around the airway, in the orbit, or in closed compartments such as the forearm.
ACKNOWLEDGMENT The authors thank Dr. William Sheils for his development work on bleomycin foam presented at the International Society for the Study of Vascular Anomalies 2012.
REFERENCES 1. Lowe LH, Marchant TC, Rivard DC, Scherbel AJ. Vascular malformations: classification and terminology the radiologist needs to know. Semin Roentgenol 2012; 47:106–117; http://dx.doi.org/10.1053/j.ro.2011. 11.002, PubMed PMID: 22370189. 2. Vikkula M, Boon LM, Mulliken JB. Molecular genetics of vascular malformations. Matrix Biol 2001; 20:327–335, PubMed PMID: 11566267. 3. Greene AK, Alomari AI. Management of venous malformations. Clin Plast Surg 2011; 38:83–93; http://dx.doi.org/10.1016/j.cps.2010. 08.003, PubMed PMID: 21095474. 4. 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; http://dx.doi.org/10.1007/s00247-002-08 38-9, PubMed PMID: 12557065. 5. Sachin K, Rashmi S, Manish S, Siddhartha W, Uday L. Haemangiomas and venous malformations of the head and neck: A retrospective analysis of endovascular management in 358 patients. Indian J Plast Surg 2013109–116; http://dx.doi.org/10.4103/0970-0358.113727, PubMed PMID: 23960315; PubMed Central PMCID: PMC3745094.
10
’
Bleomycin Foam Treatment of Venous Malformations
6. Thawait SK, Puttgen K, Carrino JA, et al. MR imaging characteristics of soft tissue vascular anomalies in children. Eur J Pediatr 2013; 172:591– 600; http://dx.doi.org/10.1007/s00431-012-1828-z, PubMed PMID: 22986771. 7. Legiehn GM, Heran MK. Classification, diagnosis, and interventional radiologic management of vascular malformations. Orthop Clin North Am 2006; 37:435–474, vii-viii; http://dx.doi.org/10.1016/j.ocl.2006.04.005, Pub Med PMID: 16846771. 8. Legiehn GM, Heran MK. Venous malformations: classification, development, diagnosis, and interventional radiologic management. Radiol Clin North Am 2008; 46:545–597; http://dx.doi.org/10.1016/j.rcl.2008. 02.008, PubMed PMID: 18707962. 9. Zheng JW, Yang XJ, Wang YA, He Y, Ye WM, Zhang ZY. Intralesional injection of Pingyangmycin for vascular malformations in oral and maxillofacial regions: an evaluation of 297 consecutive patients. Oral Oncol 2009; 45:872–876; http://dx.doi.org/10.1016/j.oraloncology.2009. 02.011, PubMed PMID: 19628423. 10. Yang Y, Sun M, Ma Q, et al. Bleomycin A5 sclerotherapy for cervicofacial lymphatic malformations. J Vasc Surgery 2011; 53:150 155; http://dx.doi.org/10.1016/j.jvs.2010.07.019, PubMed PMID: 20843632. 11. Muir T, Kirsten M, Fourie P, Dippenaar N, Ionescu GO. Intralesional bleomycin injection (IBI) treatment for haemangiomas and congenital vascular malformations. Pediatr Surg Int 2004; 19:766–773; http://dx.doi. org/10.1007/s00383-003-1058-6, PubMed PMID: 14740248. 12. Zhao JH, Zhang WF, Zhao YF. Sclerotherapy of oral and facial venous malformations with use of pingyangmycin and/or sodium morrhuate. Int J Oral Maxillofac Surg 2004; 33:463–466; http://dx.doi.org/10.1016/j.ijom. 2003.10.003, PubMed PMID: 15183410. 13. Zhi K, Wen Y, Li L, Ren W. The role of intralesional Pingyangmycin in the treatment of venous malformation of facial and maxillary region. Int J Pediatr Otorhinolaryngol 2008; 72:593–597; http://dx.doi.org/10.1016/ j.ijporl.2008.01.007, PubMed PMID: 18294701. 14. Hou R, Guo J, Hu K, et al. A clinical study of ultrasound-guided intralesional injection of bleomycin A5 on venous malformation in cervical-facial region in China. J Vasc Surg 2010; 51:940–945; http://dx. doi.org/10.1016/j.jvs.2009.11.038, PubMed PMID: 20347690. 15. Dompmartin A, Acher A, Thibon P, et al. Association of localized intravascular coagulopathy with venous malformations. Arch Dermatol 2008; 144:873–877; http://dx.doi.org/10.1001/archderm.144.7.873 PubMed PMID: 18645138. 16. Bajpai H, Bajpai S. Comparative analysis of intralesional sclerotherapy with sodium tetradecyl sulfate versus bleomycin in the management of low flow craniofacial soft tissue vascular lesions. J Maxillofac Oral Surg 2012; 11:13– 20; http://dx.doi.org/10.1007/s12663-011-0325-7, PubMed PMID: 23449774; PubMed Central PMCID: PMC3319813.
Ul Haq et al
’
JVIR
17. Spence J, Krings T, TerBrugge KG, Agid R. Percutaneous treatment of facial venous malformations: a matched comparison of alcohol and bleomycin sclerotherapy. Head Neck 2011; 33:125–130; http://dx.doi.org/ 10.1002/hed.21410, PubMed PMID: 20848431. 18. Zhang J, Li HB, Zhou SY, et al. Comparison between absolute ethanol and bleomycin for the treatment of venous malformation in children. Experimental and therapeutic medicine 2013; 6:305–309; http://dx.doi.org/ 10.3892/etm.2013.1144, PubMed PMID: 24137179; PubMed Central PMCID: PMC3786900. 19. Berenguer B, Burrows PE, Zurakowski D, Mulliken JB. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg 1999; 104:1–11, discussion 2-5. PubMed PMID: 10597669. 20. Khandpur S, Sharma VK. Utility of intralesional sclerotherapy with 3% sodium tetradecyl sulphate in cutaneous vascular malformations. Dermatol Surg 2010; 36:340–346; http://dx.doi.org/10.1111/j.1524-4725. 2009.01440.x, PubMed PMID: 20100267. 21. 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; http://dx.doi.org/10.1016/j.jvir.2006.12. 735, PubMed PMID: 17377179. 22. Siniluoto TM, Svendsen PA, Wikholm GM, Fogdestam I, Edstrom S. Percutaneous sclerotherapy of venous malformations of the head and neck using sodium tetradecyl sulphate (sotradecol). Scand J Plast Reconstr Surg Hand Surg 1997; 31:145–150, PubMed PMID: 9232699. 23. O’Donovan JC, Donaldson JS, Morello FP, Pensler JM, Vogelzang RL, Bauer B. Symptomatic hemangiomas and venous malformations in infants, children, and young adults: treatment with percutaneous injection of sodium tetradecyl sulfate. AJR Am J Roentgenol 1997; 169:723–729; http://dx.doi.org/10.2214/ajr.169.3.9275886, PubMed PMID: 9275886. 24. Umezawa H. Recent study on biochemistry and action of Bleomycin. In: Carter SK, Crook ST, Umezawa H, editors. Bleomycin: Current Status and New Developments. New York, NY: Academic Press; 1978. p. 15–20. 25. Yura J, Hashimoto T, Tsuruga N, Shibata K. Bleomycin treatment for cystic hygroma in children. Nippon Geka Hokan 1977; 46:607–614, PubMed PMID: 73364. 26. Tanigawa N, Shimomatsuya T, Takahashi K, et al. Treatment of cystic hygroma and lymphangioma with the use of bleomycin fat emulsion. Cancer 1987; 60:741–749, PubMed PMID: 2439189. 27. Mowad CM, Nguyen TV, Elenitsas R, Leyden JJ. Bleomycin-induced flagellate dermatitis: a clinical and histopathological review. Br J Dermatol 1994; 131:700–702, PubMed PMID: 7528043. 28. Vogelzang RL, Atassi R, Vouche M, Resnick S, Salem R. Ethanol embolotherapy of vascular malformations: clinical outcomes at a single center. J Vasc Interv Radiol 2014; 25:206–213; http://dx.doi.org/10.1016/ j.jvir.2013.10.055, PubMed PMID: 24461130.