DOI: 10.1148/radiol.2262012157
(Radiology 2003;226:567-572.)
© RSNA, 2003
Hemangiomas in the Face and Extremities: MR–guided Sclerotherapy—Optimization with Monitoring of Signal Intensity Changes in Vivo1
Naoto Hayashi, MD,
Tomohiko Masumoto, MD,
Toshiyuki Okubo, MD,
Osamu Abe, MD,
Nobuyuki Kaji, MD,
Kazuyuki Tokioka, MD,
Shigeki Aoki, MD and
Kuni Ohtomo, MD
1 From the Departments of Radiology (N.H., T.M., T.O, O.A., S.A., K.O.) and Plastic Surgery (N.K., K.T.), Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. Received January 14, 2002; revision requested March 5; revision received June 3; accepted July 1. Address correspondence to N.H. (e-mail: naoto-tky@umin.ac.jp).
 |
ABSTRACT
|
|---|
The feasibility of a method of magnetic resonance (MR)-guided sclerotherapy for hemangioma was evaluated. The distribution of a test injection and a contiguous sclerosant injection was monitored clearly with MR fluoroscopy. Postsclerotherapy T2-weighted MR images depicted areas with and those without effective sclerosant concentration, which enabled performance of selective supplemental sclerotherapy. This method of MR-guided sclerotherapy seems feasible for clinical application.
© RSNA, 2003
Index terms: Angioma, soft tissues, 20.362, 40.362 • Gadolinium • Magnetic resonance (MR), guidance, 20.362, 40.362 • Sclerotherapy, 20.1269, 20.362, 40.1269, 40.362
|
INTRODUCTION
|
|---|
Since the development of fast imaging sequences for magnetic resonance (MR) fluoroscopy, various MR-guided interventional radiologic techniques have been reported (1–3). Compared with conventional fluoroscopy, MR fluoroscopy offers unique advantages, such as radiation-free imaging, excellent tissue contrast, and the ability to acquire images in arbitrary section orientation (4).
In this article, we use the term hemangioma as "a benign but nonreactive process in which there is an increased number of normal or abnormal-appearing vessels" (5). We also acknowledge that some investigators reserve the term hemangioma to describe "lesions that show increased endothelial mitotic activity," and they use vascular malformation to describe "vascular lesions that grow pari passu with the child, and show normal endothelial mitotic activity" (6). The semantics of hemangioma, however, will not be discussed in this article.
Hemangioma often needs to be treated because it may cause functional disability, local pain, and aesthetic problems (7). Sclerotherapy of hemangioma has an indispensable role when surgery is not recommended because of the risk of aesthetic and functional complications. Sclerotherapy acts on hemangioma by inducing endothelial injury and thrombosis of the vessels in the lesion. In our institution, we have treated hemangiomas with conventional fluoroscopy–guided percutaneous sclerotherapy with the sclerosant ethanolamine oleate (EO). Results of animal experiments show that EO induces complete endothelial injury at a concentration of greater than approximately 1% per volume with contact to the endothelium for 30 seconds (8). To our knowledge, however, it has not been possible to determine the concentration of sclerosant in the lesion in vivo because it is influenced by many factors, such as lesion volume, blood flow, and presence of connecting channels. The appropriate amount of sclerosant to inject at a certain site must be determined because excessive amounts may flow into the systemic circulation and increase the risk of hemolytic nephropathy, while injection of insufficient amounts will not induce endothelial injury (9).
The purpose of this study was to evaluate the feasibility of a method of sclerotherapy we have developed to treat soft-tissue hemangioma.
|
Materials and Methods
|
|---|
All studies were performed with a 0.2-T open-configuration MR imaging system (Magnetom Open; Siemens Medical Systems, Erlangen, Germany). A ring-shaped multipurpose coil functioned as a radio-frequency transmitter and receiver. Fast spin-echo T2-weighted MR imaging (5,000/102 [repetition time msec/echo time msec]; echo train length, seven; matrix size, 256 x 224; field of view, 200–300 mm; section thickness, 7–10 mm) was performed before and immediately after completion of sclerotherapy to evaluate the therapeutic effects. MR fluoroscopy was performed with two types of steady-state free precession sequences: (a) fast imaging with steady-state precession (FISP) (21/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200–300 mm; section thickness, 7–10 mm), and (b) reversed FISP (PSIF) (22/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200–300 mm; section thickness, 7–10 mm). The MR fluoroscopic images were obtained in three contiguous parallel planes or three orthogonal planes with an imaging time of 2 seconds for each image. The FISP sequence, with short repetition time and large flip angle, produced images weighted by T2 divided by T1 (T2/T1), whereas the PSIF sequence produced images that were similar to T2-weighted MR images. FISP MR fluoroscopy was used to image the test injection (gadopentetate dimeglumine [0.5 mmol/mL], Magnevist; Nihon Schering, Osaka, Japan), and PSIF MR fluoroscopy was used to image injection of the sclerosant (10% EO, Oldamin; Grelan Pharmaceutical, Tokyo, Japan). Data from the manufacturer of gadopentetate dimeglumine confirmed that the chelate in the contrast agent is stable even when it is mixed with EO (unpublished data, Magnevist-Oldamin interaction laboratory test report, Nihon Schering, 1998).
In Vitro Study for Determining Gadopentetate Dimeglumine Concentration for MR Fluoroscopy
Samples of various concentrations of contrast material were imaged with FISP and PSIF MR fluoroscopy to determine the optimum concentration for test injection. Test tube samples with 10.0%, 1.0%, 0.1%, 0.01%, 0.001%, and 0.0001% saline dilutions of 0.5 mmol/mL gadopentetate dimeglumine were imaged with a large nickel-sulfate–doped phantom (which consisted of 1.25 g of nickel sulfur and 5 g of sodium chloride in 1,000 g of pure water).
Theory and in Vitro Study for Estimating Sclerosant Concentration in Vivo
The signal intensity (SI) of one spin-echo pulse sequence with paramagnetic agent enhancement is shown by the following equation (10):
where N(H) is proportional to hydrogen density; F(V) is a function of flow; TR is repetition time and TE is echo time; T1d and T2d are the relaxation times of tissue in the absence of paramagnetic agent; R1 and R2 are the relaxation rate changes per unit concentration of paramagnetic agent with respect to T1 and T2, respectively; and C is the concentration of the paramagnetic agent. From the Equation, the theoretic SI of blood in a T2-weighted spin-echo MR image can be shown as a function of concentration of the paramagnetic agent (Fig 1).
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Schematic shows relationship between concentration of gadopentetate dimeglumine and SI of blood doped with the contrast agent. In the Equation, N(H) = 1, F(V) = 1, TR = 5.0 sec, TE = 0.1 sec, T1d = 1.5 sec, T2d = 1.5 sec, R1 = 5 mmol-1 · sec-1, and R2 = 5 mmol-1 · sec-1). The SI of blood doped with gadopentetate dimeglumine changes within a relatively narrow concentration range and is virtually 0 above a certain concentration.
|
|
The SI of blood doped with gadopentetate dimeglumine changes within a relatively narrow concentration range and is virtually 0 above a certain concentration. This characteristic of SI changes was applied to predict the adequate concentration of sclerosant in vivo. A mixture of sclerosant and gadopentetate dimeglumine was prepared to show the disappearance of SI when the mixture is above the therapeutic concentration. In short, this mixture will have sclerosing capacity if it does not show any SI, but it will have inadequate sclerosing capacity if it shows some SI. This should theoretically allow monitoring of the sclerosing capacity of the sclerosant mixture in vivo. Concerning the sclerosant EO, it has been reported that a concentration of 1% EO is the minimum therapeutic concentration to induce endothelial injury necessary for vascular occlusion (8). Ideally, therefore, the EO mixture should show SI at the concentration level of approximately 1% EO.
The relationship between SI and concentration changes with dilution of the gadopentetate dimeglumine–EO mixture in blood was examined. Sixteen samples of human blood from one volunteer were mixed in vitro with sclerosant mixtures that included various concentrations of gadopentetate dimeglumine, and then the samples were imaged with the fast spin-echo T2-weighted MR sequence. Test tube samples of 16%, 8%, 4%, and 2% saline dilution of 0.5 mmol/mL gadopentetate dimeglumine were mixed with the same volume of 10% EO, which made compositions of 8% gadopentetate dimeglumine and 5% EO, 4% gadopentetate dimeglumine and 5% EO, 2% gadopentetate dimeglumine and 5% EO, and 1% gadopentetate dimeglumine and 5% EO, respectively. These samples were then diluted with human blood to concentrations of 50.0%, 25.0%, 12.5%, and 6.3%, respectively. Then all dilutions were imaged together with the fast spin-echo T2-weighted MR sequence.
We sought a composition that changed its SI from some SI to signal void at a concentration between 12.5% and 25.0% dilution, which corresponded to EO concentrations of 0.625% and 1.25%, respectively. This composition could signal a concentration of 1% EO by changing its SI to signal void when its concentration was greater than 1% EO.
Clinical Study
This study was approved by the institutional ethical committee. From July to November 1997, written informed consent was obtained from patients for each procedure. Thirteen consecutive patients (nine female and four male patients; age range, 5–28 years; mean and median age, 16 years) with low-flow hemangiomas in the face or extremities were selected for this study. Capillary hemangiomas of infancy were excluded because they are likely to involute in natural course. Patients with a lesion in the vicinity of the orbit were also excluded because of the risk that sclerosant might drain into the superior ophthalmic vein and cause occlusion. Fourteen MR-guided sclerotherapy procedures were performed in 13 patients (two procedures for one patient).
The target lesions for sclerotherapy were 13 low-flow hemangiomas. Nine of the lesions were located in the face, and four were located in the extremities. There were four localized hemangiomas (mean size, 7.3 mL; size range, 2.9–9.9 mL), seven diffuse hemangiomas (mean size, 134 mL; size range, 23–531 mL), and two Klippel-Trenaunay-Weber syndrome lesions (size not assessed). Lesion sizes were calculated on presclerotherapy and follow-up T2-weighted MR images by means of a stereologic method (11).
The purpose of treatment was aesthetic and functional improvement in three patients, aesthetic improvement only in seven patients, and local pain control in three patients. We did not necessarily aim to eliminate the lesions since they are benign in nature. The procedure was performed with intravenous anesthesia and sedation (thiopental sodium, Lavonal, Tanabe Seiyaku, Osaka, Japan, and pentazocine, Pentagin injection, Sankyo, Tokyo, Japan). Vital functions of the patients were monitored with MR-compatible electrocardiography and pulse oximetry. A Foley catheter was placed in the bladder to monitor the volume of urination and the presence of any hematuria. The lesion was monitored during the procedure by means of direct inspection to evaluate the presence of cutaneous devascularization and local soft-tissue swelling. During the procedure, the MR images were evaluated on the in-room console monitor. These evaluations were made with consensus of at least two of the authors. The patients were hospitalized for at least 3 days after sclerotherapy and for longer than 1 week if there were any complications such as skin necrosis. All procedures were completed successfully.
Baseline imaging study.—A presclerotherapy study was performed with fast spin-echo T2-weighted MR imaging to cover the whole lesion with contiguous sections. For MR fluoroscopy, the plane that included the target area was selected.
Test injection with FISP MR fluoroscopy.—After puncture, 1% gadopentetate dimeglumine was injected into the lesion, guided with FISP MR fluoroscopy, to predict the distribution of the sclerosant during the following sclerotherapy (Fig 2a). The 1% gadopentetate dimeglumine causes an increase in SI as a result of the T1-shortening effect. When the 1% gadopentetate dimeglumine did not spread in the target area, another puncture was tried until an appropriate needle position was attained. The skin overlying the lesion was monitored for discoloration caused by the injection, which would imply devascularization and necrosis of the skin when the sclerosant was injected.
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Schematic shows test and sclerosant injections guided with MR fluoroscopy. (a) FISP MR fluoroscopic imaging. Left: Lesion is isointense. Middle: One percent gadopentetate dimeglumine is hyperintense against isointense lesion. Right: Fully injected lesion is hyperintense against surrounding soft-tissue structures. (b) PSIF MR fluoroscopic imaging. Left: Lesion is hyperintense. Middle: EO-gadopentetate dimeglumine mixture is hypointense against hyperintense lesion. Right: When effective concentration is achieved, lesion is signal void.
|
|
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Schematic shows test and sclerosant injections guided with MR fluoroscopy. (a) FISP MR fluoroscopic imaging. Left: Lesion is isointense. Middle: One percent gadopentetate dimeglumine is hyperintense against isointense lesion. Right: Fully injected lesion is hyperintense against surrounding soft-tissue structures. (b) PSIF MR fluoroscopic imaging. Left: Lesion is hyperintense. Middle: EO-gadopentetate dimeglumine mixture is hypointense against hyperintense lesion. Right: When effective concentration is achieved, lesion is signal void.
|
|
Sclerotherapy with PSIF MR fluoroscopy.— A mixture of EO and gadopentetate dimeglumine was prepared to show SI changes with 1% EO, which is the minimum therapeutic concentration (Fig 2b). The mixture was prepared by mixing 0.8 mL of gadopentetate dimeglumine, 4.2 mL of distilled water, and 5 mL of 10% EO; this resulted in 10 mL of the 8%/5% (gadopentetate dimeglumine/EO) solution, which is clear and highly viscous. The mixture was then injected, with PSIF MR fluoroscopic guidance, into the lesion that showed high SI before injection. Distribution of the sclerosant was confirmed by a remarkable decrease in SI as a result of the strong T2-shortening effect of the sclerosant mixture. Since the solution of 1% gadopentetate dimeglumine that was injected during the test injection had little T2-shortening effect, it barely influenced the SI of the lesion at PSIF MR fluoroscopy. The volume of sclerosant that was injected was recorded. The presence of immediate complications, if any, was recorded. If evident hematuria occurred during sclerotherapy, 2,000–4,000 U of haptoglobin (Haptoglobin injection; Mitsubishi Pharma, Osaka, Japan) was administered to prevent renal tubular dysfunction (12).
Postsclerotherapy MR imaging study.—Postsclerotherapy fast spin-echo T2-weighted MR images were obtained immediately after injection of the sclerosant to evaluate the distribution of the sclerosant at therapeutic concentration. The images were evaluated to determine whether the high SI of the target area changed to signal void. If there was substantial residual lesion (ie, high-SI area) with no sclerosant distribution, additional sclerotherapy was performed with the same technique described earlier.
Follow-up.—All patients were followed up for at least 3 months with clinical examinations by the plastic surgeon authors. Follow-up MR imaging was performed in 12 of the 13 patients at 3–6 months after sclerotherapy. One patient was lost to follow-up before follow-up MR imaging. The patients were evaluated for clinical improvement or deterioration, changes in lesion volume, and any complications. They were evaluated especially for the maintenance of their initial therapeutic response.
|
Results
|
|---|
Gadopentetate Dimeglumine Concentration for MR Fluoroscopy
At FISP MR fluoroscopy, the 1% gadopentetate dimeglumine solution had the highest SI (Fig 3). At PSIF MR fluoroscopy, the 10% gadopentetate dimeglumine solution was completely signal void, while other concentrations showed some SI. The 1% and 10% gadopentetate dimeglumine solutions were selected for use in the test injection at FISP and PSIF MR fluoroscopy, respectively.
View larger version (81K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. MR fluoroscopic images of test samples (eight test tubes) of various gadopentetate dimeglumine concentrations were obtained with (a) FISP (18/8; flip angle, 90°) and (b) PSIF (22/10; flip angle, 90°) MR fluoroscopic sequences. Top row, left to right: Test samples with 10.0%, 1.0%, 0.1%, and 0.01% gadopentetate dimeglumine. Bottom row, left to right: Test samples with 75% ethanol, water, 0.001% gadopentetate dimeglumine, and 0.0001% gadopentetate dimeglumine. In a, 1% gadopentetate dimeglumine solution has highest SI. In b, 10% gadopentetate dimeglumine solution is signal void, while other concentrations show some SI.
|
|
View larger version (102K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. MR fluoroscopic images of test samples (eight test tubes) of various gadopentetate dimeglumine concentrations were obtained with (a) FISP (18/8; flip angle, 90°) and (b) PSIF (22/10; flip angle, 90°) MR fluoroscopic sequences. Top row, left to right: Test samples with 10.0%, 1.0%, 0.1%, and 0.01% gadopentetate dimeglumine. Bottom row, left to right: Test samples with 75% ethanol, water, 0.001% gadopentetate dimeglumine, and 0.0001% gadopentetate dimeglumine. In a, 1% gadopentetate dimeglumine solution has highest SI. In b, 10% gadopentetate dimeglumine solution is signal void, while other concentrations show some SI.
|
|
Composition of Sclerosant Mixture
Among the compositions that were tested, the 8%/5% (gadopentetate dimeglumine/EO) solution changed its SI at the designated 1% concentration level from signal void to some SI at a concentration between 25.0% and 12.5% of the original mixture (Fig 4).
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Fast spin-echo T2-weighted MR images (5,000/102; echo train length, seven) depict test samples (17 test tubes) with various gadopentetate dimeglumine concentrations diluted with blood. Top row: 8%/5% (gadopentetate dimeglumine/EO) solution; left to right, 100.0%, 50.0%, 25.0%, 12.5%, 6.3% dilutions. Second row: 4%/5% solution; left to right, 100.0%, 50.0%, 25.0%, 12.5%, 6.3% dilutions. Third row: 2%/5% solution; left to right, 100.0%, 50%.0, 25.0%, 12.5%, 6.3% dilutions. Bottom row: 1%/5% solution; left to right, 100.0%, 50.0%, 25.0%, 12.5%, 6.3% dilutions; far right, one-sample control (water). With the 8%/5% solution, changes in SI from signal void to some SI occur at a concentration between 25.0% and 12.5% (arrows); the six samples on the left do not show any SI.
|
|
Clinical Study
All hemangiomas were depicted as isointense on FISP MR fluoroscopic images and as hyperintense on PSIF MR fluoroscopic images and fast spin-echo T2-weighted MR images. The contrast material in the test injections was depicted as hyperintense on FISP MR fluoroscopic images in all cases (Fig 5a). The sclerosant mixture was depicted as hypointense on PSIF MR fluoroscopic images in all cases (Fig 5b). On the PSIF MR fluoroscopic images, signal voids were depicted in the surrounding hyperintense area. This provided more conspicuous contrast of the sclerosant mixture than that seen on the FISP MR fluoroscopic images (a hyperintense focus in the surrounding isointense area). The postsclerotherapy fast spin-echo T2-weighted MR images helped distinguish between areas with and those without the effective sclerosant concentration. This enabled us to perform selective additional sclerotherapy in areas without adequate sclerosant concentration (Fig 5c, 5d). The mean volume of injection at each puncture site was 4.8 mL.
View larger version (57K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. MR images depict MR-guided sclerotherapy of a hemangioma in the upper lip of a 24-year-old woman. (a) Midsagittal FISP MR fluoroscopic image (21/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). (b) Midsagittal PSIF MR fluoroscopic image (22/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). In a, lesion (arrow) shows increasing SI after test injection of gadopentetate dimeglumine. In b, lesion (arrow) shows expansion of hypointense area after injection of sclerosant mixture; hypointensity of sclerosant mixture contrasts well against hyperintensity of surrounding lesion. (c, d) Transverse fast spin-echo T2-weighted MR images (5,000/102; echo train length, seven; matrix size, 196 x 256; field of view, 150 x 200; section thickness, 5 mm; two signals acquired) were obtained (c) before sclerotherapy and (d) immediately after sclerotherapy. Hyperintense lesion (arrow) in upper lip becomes signal void after sclerotherapy, which indicates adequate sclerosant concentration in vivo.
|
|
View larger version (72K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. MR images depict MR-guided sclerotherapy of a hemangioma in the upper lip of a 24-year-old woman. (a) Midsagittal FISP MR fluoroscopic image (21/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). (b) Midsagittal PSIF MR fluoroscopic image (22/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). In a, lesion (arrow) shows increasing SI after test injection of gadopentetate dimeglumine. In b, lesion (arrow) shows expansion of hypointense area after injection of sclerosant mixture; hypointensity of sclerosant mixture contrasts well against hyperintensity of surrounding lesion. (c, d) Transverse fast spin-echo T2-weighted MR images (5,000/102; echo train length, seven; matrix size, 196 x 256; field of view, 150 x 200; section thickness, 5 mm; two signals acquired) were obtained (c) before sclerotherapy and (d) immediately after sclerotherapy. Hyperintense lesion (arrow) in upper lip becomes signal void after sclerotherapy, which indicates adequate sclerosant concentration in vivo.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. MR images depict MR-guided sclerotherapy of a hemangioma in the upper lip of a 24-year-old woman. (a) Midsagittal FISP MR fluoroscopic image (21/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). (b) Midsagittal PSIF MR fluoroscopic image (22/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). In a, lesion (arrow) shows increasing SI after test injection of gadopentetate dimeglumine. In b, lesion (arrow) shows expansion of hypointense area after injection of sclerosant mixture; hypointensity of sclerosant mixture contrasts well against hyperintensity of surrounding lesion. (c, d) Transverse fast spin-echo T2-weighted MR images (5,000/102; echo train length, seven; matrix size, 196 x 256; field of view, 150 x 200; section thickness, 5 mm; two signals acquired) were obtained (c) before sclerotherapy and (d) immediately after sclerotherapy. Hyperintense lesion (arrow) in upper lip becomes signal void after sclerotherapy, which indicates adequate sclerosant concentration in vivo.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d. MR images depict MR-guided sclerotherapy of a hemangioma in the upper lip of a 24-year-old woman. (a) Midsagittal FISP MR fluoroscopic image (21/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). (b) Midsagittal PSIF MR fluoroscopic image (22/10; flip angle, 90°; matrix size, 64 x 128; field of view, 200 x 200 mm; section thickness, 8 mm; acquisition time, 2 seconds per image [from right to left]). In a, lesion (arrow) shows increasing SI after test injection of gadopentetate dimeglumine. In b, lesion (arrow) shows expansion of hypointense area after injection of sclerosant mixture; hypointensity of sclerosant mixture contrasts well against hyperintensity of surrounding lesion. (c, d) Transverse fast spin-echo T2-weighted MR images (5,000/102; echo train length, seven; matrix size, 196 x 256; field of view, 150 x 200; section thickness, 5 mm; two signals acquired) were obtained (c) before sclerotherapy and (d) immediately after sclerotherapy. Hyperintense lesion (arrow) in upper lip becomes signal void after sclerotherapy, which indicates adequate sclerosant concentration in vivo.
|
|
The initial therapeutic response was maintained in all patients. Local lesion volume was reduced in 11 patients, pain control was maintained in three patients, and functional improvement in jaw movement was maintained in three patients.
No major complication (eg, massive hemolysis, nerve injury, severe scars) occurred after sclerotherapy guided with MR fluoroscopy. Hematuria was noted in seven patients during the procedure, and haptoglobin was immediately administered intravenously; none of these patients developed renal dysfunction. Minor skin or mucosal breakdowns that occurred in four patients healed without conspicuous scars within the 3-month follow-up. Minor complications such as local swelling and pain were well controlled.
|
Discussion
|
|---|
Findings in our study show that MR fluoroscopy–guided sclerotherapy can depict the distribution of test and sclerosant injections and can monitor the effective sclerosant concentration in vivo. The safety and feasibility of MR-guided sclerotherapy has been reported by Lewin et al (13). Our results also show that MR-guided sclerotherapy has the potential to be a more refined treatment method than conventional sclerotherapy for low-flow hemangiomas.
In our experience with sclerotherapy guided with conventional fluoroscopy, contrast material in the lesion had to be washed out between the test and sclerosant injections to attain adequate visualization. The low flow of the lesion makes it difficult to expect rapid washout with natural flow; therefore, saline is frequently injected to promote rapid washout. However, repeated injections may cause dislocation of the injection needle and injection of the sclerosant into inappropriate spaces. With our method, the distribution of the test and sclerosant injections can be depicted with two MR sequences; therefore, the sclerosant can be injected immediately after the test injection without waiting for washout of contrast material. This should decrease the risk of needle dislocation and save time during sclerotherapy at each injection site.
Although the sclerosant used in this study is a relatively safe material, it would be harmful if it were used in an excessive amount. An appropriate amount of sclerosant should be selected for each lesion to achieve maximum therapeutic effect and avoid possible side effects. Though the therapeutic effect of the sclerosant is concentration dependent, to our knowledge no method has been reported previously with which to monitor the concentration of the sclerosant in vivo. Our method enabled monitoring of the effective sclerosant concentration in vivo; this made it possible to determine the minimal amount of sclerosant for injection. With our method, it is possible to distinguish areas with adequate concentrations of sclerosant from those with inadequate concentrations; therefore, it is also possible to perform multiple injections during one procedure without the risk of overdose in the area targeted for sclerotherapy. The mean volume injected at each site was small in our procedure, most likely because we avoided injection of excessive amounts.
One disadvantage of this method of sclerotherapy guided with MR fluoroscopy in its current implementation, however, is that it may be more time-consuming for detailed evaluation than that guided with conventional fluoroscopy. One injection may last as long as 5 minutes to prepare and complete with MR fluoroscopic guidance, while the procedure is usually shorter than 1 minute with conventional fluoroscopic guidance. Further advancements in the software and hardware for interventional MR imaging are desirable if it is to be more expeditious than ultrasonography or conventional fluoroscopy, which have been the major modalities in current interventional radiology.
Our method of MR-guided sclerotherapy can depict the distribution of both the test and contiguous sclerosant injections. It can also be used to monitor the effective sclerosant concentration in vivo. Findings in this study show the potential for MR-guided sclerotherapy to be a more refined method than conventional fluoroscopy–guided sclerotherapy for treatment of hemangioma.
|
FOOTNOTES
|
|---|
Abbreviations: EO = ethanolamine oleate,
FISP = fast imaging with steady-state precession,
PSIF = reversed FISP,
SI = signal intensity
Author contributions: Guarantor of integrity of entire study, N.H.; study concepts, N.H., T.O., O.A.; study design, N.H., T.M.; literature research, T.M., N.K.; clinical studies, T.M., N.K., K.T.; experimental studies, N.H., T.M.; data acquisition, N.H., T.M., T.O.; data analysis/interpretation, N.H., T.M.; manuscript preparation, N.H., T.M.; manuscript definition of intellectual content and editing, N.H., S.A.; manuscript revision/review, K.T., S.A., K.O.; manuscript final version approval, all authors.
|
REFERENCES
|
|---|
- Lewin JS, Petersilge CA, Hatem SF, et al. Interactive MR imaging-guided biopsy and aspiration with a modified clinical C-arm system. AJR Am J Roentgenol 1998; 170:1593-1601.[Abstract/Free Full Text]
- Schoenenberger AW, Steiner P, Debatin JF, et al. Real-time monitoring of laser diskectomies with a superconducting, open-configuration MR system. AJR Am J Roentgenol 1997; 169:863-867.[Abstract/Free Full Text]
- Lufkin RB. Interventional MR imaging. Radiology 1995; 197:16-18.[Free Full Text]
- Lenz G, Drobnitzky M. Interventional MRI with an open low-field system. In: Debatin JF, Adams G, eds. Interventional magnetic resonance imaging. Berlin, Germany: Springer-Verlag, 1998; 3-10.
- Enzinger FM, Weiss SW. Benign tumors and tumorlike lesions of blood vessels In: Soft tissue tumors. 3rd ed. St Louis, Mo: Mosby–Year Book, 1995; 579-626.
- Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982; 69:412-422.[Medline]
- Jackson IT, Carreno R, Potparic Z, Hussain K. Hemangiomas, vascular malformations, and lymphovenous malformations: classification and methods of treatment. Plast Reconstr Surg 1993; 91:1216-1230.[Medline]
- Masaki M, Obara K, Suzuki S, et al. The destructive effects of sclerosant ethanolamine oleate on mammalian vessel endothelium. Gastroenterol Jpn 1990; 25:230-235.[Medline]
- Ohta M, Hashizume M, Ueno K, Tanoue K, Sugimachi K. Albumin inhibits hemolysis of erythrocytes induced by ethanolamine oleate during endoscopic injection sclerotherapy. Hepatogastroenterology 1993; 40:65-68.[Medline]
- Davis PL, Parker DL, Nelson JA, Gillen JS, Runge VM. Interactions of paramagnetic contrast agents and the spin echo pulse sequence. Invest Radiol 1988; 23:381-388.[Medline]
- Vogels OJ, Zijlmans JC, van’t Hof MA, Thijssen HO, Horstink MW. MR volume estimation of subcortical brain lesions and ventricular cerebrospinal fluid: a simple and accurate stereologic method. AJNR Am J Neuroradiol 1995; 16:1441-1445.[Abstract]
- Miyoshi H, Ohshiba S, Matsumoto A, Takada K, Umegaki E, Hirata I. Haptoglobin prevents renal dysfunction associated with intravariceal infusion of ethanolamine oleate. Am J Gastroenterol 1991; 86:1638-1641.[Medline]
- Lewin JS, Merkle EM, Duerk JL, Tarr RW. Low-flow vascular malformations in the head and neck: safety and feasibility of MR imaging-guided percutaneous sclerotherapy—preliminary experience with 14 procedures in three patients. Radiology 1999; 211:566-570.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
K. N. Smolinski and A. C. Yan
Hemangiomas of Infancy: Clinical and Biological Characteristics
Clinical Pediatrics,
November 1, 2005;
44(9):
747 - 766.
[Abstract]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
