Abstract
Purpose:
The purpose of this study was to evaluate the safety and efficacy of preoperative percutaneous n-butyl cyanoacrylate (nBCA) embolization of venous malformations in children.
Material and Methods:
Clinical data were retrospectively reviewed in children who underwent embolization using nBCA followed by resection of venous malformations.
Results:
A total of 17 embolizations were performed in 14 patients (9 females, mean age: 5.5 years; median age: 3 years; range 0.1-16 years). The venous malformations involved the lower extremity and the knee joint (n = 7), the trunk (n = 4), head and neck (n = 2), and hand (n = 1). n-Butyl cyanoacrylate was diluted with iodized oil at a ratio of 1:3 to 1:5. The mean and median volume of nBCA per procedure were 2.1 and 2 mL, respectively (range: 0.5-8 mL). There were no complications associated with the procedures. The mean and median time between final embolization and resection were 3.6 and 2 days, respectively. All children underwent successful resection of the symptomatic lesions. The estimated mean and median blood loss were 75 and 50 mL, respectively (range: 5-350 mL). The postprocedure course was uneventful, the days to discharge ranged between 1 and 6 days (mean 3 days).
Conclusion:
Initial results suggest that preoperative percutaneous n-butyl cyanoacrylate embolization of venous malformations is safe and effective in children, with the potential for minimizing blood loss and inpatient stay.
Introduction
Venous malformations (VMs) are the most common vascular malformations and result from developmental errors in vascular morphogenesis. 1 These slow-flow malformations are present at birth, tend to enlarge commensurate with the somatic growth of the child, and comprise a spectrum of dilated, distorted, valveless venous channels with thin walls and abnormal smooth muscle. 1,2 Although VMs may occur anywhere in the body, they are most frequently located in the head and neck (40%), extremities (40%), and trunk (20%). 3 Lesions may range from small, localized subcutaneous or intramuscular malformations to diffuse infiltrative malformations involving multiple deeper tissue planes and vital structures. 4,5
Sclerotherapy is considered as the first-line treatment of symptomatic VMs. 6 It is generally safer and more effective than resection but often requires multiple treatments. 6 Small, focal VMs can be resected as primary treatment, but excision of large and diffuse VMs is best performed after sclerotherapy. 1 Resection may be indicated in VMs that are unresponsive to sclerotherapy, have well-defined margins, are adherent to major extremity nerves where concern exists about nerve injury or to avoid joint impairment in intra-articular malformations. 7,8
Although preoperative sclerotherapy of facial VM results in less operative time per lesion volume and less operative blood loss, 9 preoperative embolization with n-butyl cyanoacrylate (nBCA, glue) has been reported to be a safe alternative method in head and neck VM. 10,11 Limitations of prior studies include the lack of documentation of blood loss, distribution, or the size of the resected lesion. 10,11
The purpose of this study was to evaluate the safety and effectiveness of preoperative embolization using nBCA in children with VMs.
Materials and Methods
The study was compliant with the Health Insurance Portability and Accountability Act and was approved by the institutional review board of Boston Children’s Hospital (Boston, Massachusetts) with informed consent waived. A retrospective review of the Vascular Anomalies Center database and the radiology information system was performed to identify patients who underwent preoperative embolization using nBCA of VMs as a planned preresection procedure. Only patients who had resection of the lesion within 1 month of the final embolization session were included. Patients with a longer period between the embolization and resection were excluded as the cyanoacrylate was primarily used as an adjunct to sclerotherapy, and the resection was only performed if there was limited response to endovascular treatment. Patient demographics (age and gender), lesion size and localization, clinical indication for the treatment, and follow-up were recorded.
Embolizations
Prior sclerotherapy, the number of embolization sessions, dilution ratio and total volume of nBCA, and additional sclerosing agents were recorded. Any procedure-related complications were noted.
Technique
Embolization procedures were performed under general anesthesia by pediatric interventional radiologists. The VMs were percutaneously punctured under real-time sonographic guidance. Venography was performed to outline the malformation and identify venous drainage. If there was no evidence of early drainage into systemic veins, embolization of the lesion was performed using nBCA (Trufill: Codman Neurovascular, Raynham, Massachusetts or Histoacryl; B. Braun Medical Ltd, Melsungen, Germany) diluted in iodized oil (Lipiodol; Laboratoire Guerbet, Roissy, France) under direct fluoroscopic guidance (Figures 1, 2). This is an off-label use of both cyanoacrylate compounds. Immediately before injection of nBCA, the catheter was primed with 5% nonionic dextrose solution to completely remove any ionic solutions from its lumen. The nBCA was then dispersed using 5% dextrose solution under continuous fluoroscopy and digital “road-mapping” technique. Multiple percutaneous punctures were performed in the presence of a lesion that was divided into several compartments. Following nBCA embolization, sclerotherapy was performed in some cases to treat portions of the lesion not planned for resection. All of the needles and catheters were then removed, and compression was applied until hemostasis was achieved.
Successful radical resection of a venous malformation after n-butyl cyanoacrylate (nBCA) embolization in a 2-year-old boy (Patient #1). A, Extensive venous malformation in the right lumbar region (arrows). B Sagittal T2 image shows the subcutaneous and intramuscular hyperintense venous malformation (arrow). C, Lateral fluoroscopy image demonstrates the distribution of radio-opaque nBCA mixture within the lesion. D, Intraoperative photograph of radical resection of the venous malformation 5 days after embolization. E, Photograph obtained 3 months following resection shows marked decrease in bulk.
An 18-month-old boy with venous malformation of the right hand (patient #8). A, A large mass is seen in the the thenar region, most prominent in the first webspace (arrow). B, T2-weighted axial magnetic resonance imaging (MRI) image. A hyperintense lesion with fluid–fluid levels is seen in insinuating between the first and fourth metacarpals. C, Intraprocedural fluoroscopy image demonstrates nBCA within the VM. D, T2-weighted axial image. Magnetic resonance imaging performed 2.5 years postresection shows marked decrease in size of the malformation.
Resections
Parameters noted included procedures performed, estimated blood loss, blood transfusions, immediate postprocedure course, and the days to discharge.
Results
Participants
In a 5-year period, a total of 17 embolizations of VMs were performed in 14 patients (Table 1). Patients with syndromic VMs or fibroadipose vascular anomaly (FAVA) were excluded. The VMs involved the lower extremity and the knee joint (n = 7), the trunk (n = 4), head and neck (n = 2) and hand (n = 1). The lesions varied in size from 1 to 541 cm3, mean 90 cm3; in patients with diffuse extremity involvement, the size of the malformation was difficult to accurately determine. The mean and median age at first embolization were 5.5 years and 3 years, respectively (range: 0.1-16 years); 9 of the patients were female. The most common presenting symptoms were pain (with or without swelling; n = 14) and painless mass (n = 5). In 2 patients with intra-articular involvement, there was evidence of recurrent hemarthroses.
Patient and Procedure Details.
Abbreviations: nBCA, n-butyl cyanoacrylate; VM, venous malformation.
Embolizations
Eleven patients underwent 1 nBCA embolization sessions. Three patients underwent 2 sessions of embolizations. In 1 patient, the procedure was staged due to the size of the lesion and performed 1 month apart. In the other 2 patients, the initial embolization was combined with sclerotherapy to target the nonarticular portion of the malformation (n = 2). In these 2 patients, the second session was performed to embolize the residual lesion immediately prior to resection of the intra-articular malformation and was performed 3 months after the primary embolization. Embolizations were performed using nBCA, diluted at a ratio of 1:3 to 1:5 with lipiodol. The volume of nBCA applied per procedure ranged from 0.9 to 4.5 mL (median: 2.1 mL, mean 2 mL).
In 8 procedures, additional sclerosants (STS 3% and absolute ethanol) were used to address components of the VM that were not intended for resection.
Resections
All of the lesions were resected within 1 month of embolization. One patient who had sclerotherapy of a large nonarticular component of the malformation was discharged, and the resection of the knee malformation performed 26 days later following reduction in inflammation. In the remainder of patients, resection was performed at the same admission as the final session of embolization, with the period between the embolization and resection determined primarily by scheduling rather than clinical reasons.The mean period between embolization and resection was 3.5 days (median 2 days; range: 0-26 days). The mean and median estimated blood loss (EBL) were 75 and 40 mL, respectively (range: 5- 350 mL). Two patients were transfused, one with packed red blood cells (50 mL) and other with 75 mL of cell salvage. For 7 patients undergoing resection of intra-articular knee malformation with synovectomy, the EBL was 90 mL (range: 5-350 mL). The immediate postprocedure course was uneventful in all patients and the days to discharge following resection ranged between 1 and 6 days (mean: 3.1 days, median: 2.5 days).
The mean follow-up time period was 2.2 years (median: 3 years, range: 0.2-4 years). Follow-up imaging was available in 8 patients (sonography in 5, magnetic resonance imaging in 3). All of the patients reported an improvement in symptoms following resection. One patient developed accessory (cranial nerve XI) nerve injury, with resolution of symptoms of over time. There were no other long-term complications related to resection. Residual malformation was seen in 6 patients, all of whom had disease extending beyond the area of resection. Two patients had recurrent symptoms in the area of the treated malformation. In 1 patient with recurrent knee swelling and pain, repeat resection and synovectomy were performed 2 years after initial treatment. The other patient presented 3 years after resection with painless swelling due to small residual posterior auricular malformation; this was then excised. No embolization was performed prior to the repeat procedures due to the small residual component of the malformation.
Discussion
Symptoms of VMs include swelling, pain sensation, bleeding, deformation of anatomical structures, and functional disturbances or dysfunction. Symptoms are dependent on the size and location of the malformation. The excision of symptomatic VM alone poses the risk of excessive bleeding. 6 The risk can be lowered in cases where a tourniquet can be used (extremity cases); however, in those cases, once the malformation is entered, it is decompressed and removing the entire lesion can be difficult due to the loss of wall integrity and tension at the periphery, which allows better visualization and dissection. Injection of cyanoacrylate into the lesion can form a cast of the VM, allowing a more complete resection while minimizing blood loss. 10-11 Hence, another reason to use the glue injection into the VM is that it forms a cast around the VM and makes removal easier.
n-Butyl cyanoacrylate is a liquid embolic material that polymerizes upon exposure to blood or water. The embolization mechanism does not depend on body’s own coagulation cascade. Takasawa et al have shown that polymerization time is prolonged, diffusing capacity increased, and embolization of a larger portion of the malformation accomplished with decreasing nBCA concentration. 12 Furthermore, with higher dilution, there is more distal spread of nBCA/lipiodol mixture. 12 Additionally, it has been shown that the use of low concentrations of nBCA (20:80) induces a tissue response similar to that of high concentrations (50:50). The use of both the low and high concentration form showed no evidence of recanalization up to 3 months after embolization. 13 Based on these findings, we used concentrations between 1:3 and 1:5 to achieve successful percutaneous embolization of the lesions. Since nBCA is slowly absorbed, it leaves a firm mass in the tissue that improves localization of the embolized lesion during resection.
There are a few previous reported cases of preoperative glue embolization of VMs of the head and neck VMs. Tieu et al performed 11 preoperative single-staged embolizations of localized VM in the head and neck region using glue. 10 The embolizations were followed by immediate surgical excision. 10 In this study, the authors noted subjectively improved intraoperative blood loss and no complications related to embolization or resection. 10
Cil et al preoperatively embolized 13 patients with direct puncture of the VM in the craniofacial region using nBCA. 11 The mean volume of nBCA used per session (5.8 mL) and multisession embolization in 4 patients suggest considerable size of the lesions. 11 However, the indication for successful resection was the improvement of cosmetic disfigurement. Since further symptoms including the definitive distribution and the size of the lesions were not mentioned, it remains unclear how extensive the VM and the caused symptoms were in fact. Furthermore, the EBL was not specified. 11
The indication for surgery in our study population was based on severe pain, recurrent joint effusion, and bleeding. We did not include patients with signs of FAVA, a newly recognized entity that requires a different management approach. 14 Fibroadipose vascular anomaly is defined by ectatic veins/phlebecatsia and fibrofatty replacement of muscles, it is distinct from common VMs and shows poor response to sclerotherapy. Although sclerotherapy and embolization can be performed, most patients require resection of their lesions to address the dominant solid fibrofatty component. 14
The EBL in this study was favorable compared to the estimated operative blood loss after preoperative sclerotherapy of facial VM reported by James et al 9 We favored nBCA embolization over sclerotherapy since sclerotherapy may result in inflammation of the surgical bed and increased risk of extravasation of the neuropathic sclerosing agents. Furthermore sclerotherapy is not the mainstay for intra-articular VMs as the sclerosing agents can damage the intra-articular cartilage. 7,15 Preoperative embolization with nBCA resulted in an average blood loss of 90 mL in patients with knee VMs compared to 470 mL in a study with 6 children with VM and Klippel-Trenaunay syndrome undergoing knee synovectomy without prior embolization or sclerotherapy. 8 In addition, the development of a small postoperative hematoma in the suprapatellar pouch in 1 of the 6 patients further emphasizes the benefits of preoperative embolization. 8 In our study, we did not observe any complications related to the embolization. This is in accordance with preoperative glue embolization followed by surgical excision of localized VMs of the head and neck 10 and less than the complication rate reported by Cil et al: 1 of 13 patients who underwent glue embolization of VMs of the craniofacial region experienced skin necrosis 3 days after embolization. 11 There is a theoretical risk of developing compartment syndrome if injected into a large lesion within a small compartment, such as the hand and foot.
Pireau et al performed surgical resections of the intra-articular knee VMs in 8 children and observed persistence of the VM in 6 patients after a mean follow-up of 5.1 years, but 4 patients were free of symptoms. 7 Two patients (one of them with coagulopathy) developed postoperative hemarthroses resulting in drainage of the hemarthroses, rehospitalization, and extension of the hospital stay to 13 days. The median hospital stay of the remaining 6 patients was 5 days. 7
The mean inpatient stay following resection in our series was 3 days (median: 2.5; range: 1-6), which is comparable to a median hospital stay of 2.5 days after preoperative sclerotherapy of facial VMs reported by James et al 9
Limitations include the retrospective nature of the study and the relatively small number of patients. In addition, the fact that lesions of varying sizes and from multiple locations were included made it difficult to obtain an accurate control group.
In conclusion, percutaneous preoperative glue embolization of VMs is a safe and effective procedure in infants and children. It has a very low complication rate and has the potential for reducing intraoperative blood loss allowing resection of large symptomatic VMs and a relatively short inpatient period.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
