DOI:10.2214/AJR.04.1508
AJR 2005; 185:1131-1137
© American Roentgen Ray Society
Dynamic MRI for Distinguishing High-Flow from Low-Flow Peripheral Vascular Malformations
Yoshimitsu Ohgiya1,2,
Toshi Hashimoto1,
Takehiko Gokan1,
Shouji Watanabe3,
Masayoshi Kuroda3,
Masanori Hirose1,
Seishi Matsui1,
Hiroshi Nobusawa1,
Takashi Kitanosono2 and
Hirotsugu Munechika1
1 Department of Radiology, Showa University School of Medicine, Tokyo,
Japan.
2 Present address: Department of Diagnostic and Interventional Neuroradiology,
University of Rochester School of Medicine & Dentistry and University of
Rochester Medical Center, 601 Elmwood Ave., PO Box 648, Rochester, NY
14642.
3 Department of Plastic Surgery, Showa University School of Medicine, Tokyo,
Japan.
Received September 24, 2004;
revised December 6, 2004;
Address correspondence to Y. Ohgiya
(yogiya{at}qd5.so-net.ne.jp).
Abstract
OBJECTIVE. The purpose of our study was to assess the usefulness of
dynamic MRI in distinguishing high-flow vascular malformations from low-flow
vascular malformations, which do not need angiography for treatment.
SUBJECTS AND METHODS. Between September 2001 and January 2003, 16
patients who underwent conventional and dynamic MRI had peripheral vascular
malformations (six high- and 10 low-flow). The temporal resolution of dynamic
MRI was 5 sec. Time intervals between beginning of enhancement of an arterial
branch in the vicinity of a lesion in the same slice and the onset of
enhancement in the lesion were calculated. We defined these time intervals as
"artery–lesion enhancement time." Time intervals between the
onset of enhancement in the lesion and the time of the maximal percentage of
enhancement above baseline of the lesion within 120 sec were measured. We
defined these time intervals as "contrast rise time" of the
lesion. Diagnosis of the peripheral vascular malformations was based on
angiographic or venographic findings.
RESULTS. The mean artery–lesion enhancement time of the
high-flow vascular malformations (3.3 sec [range, 0–5 sec]) was
significantly shorter than that of the low-flow vascular malformations (8.8
sec [range, 0–20 sec]) (Mann-Whitney test, p < 0.05). The
mean maximal lesion enhancement time of the high-flow vascular malformations
(5.8 sec [range, 5–10 sec]) was significantly shorter than that of the
low-flow vascular malformations (88.4 sec [range, 50–100 sec])
(Mann-Whitney test, p < 0.01).
CONCLUSION. Dynamic MRI is useful for distinguishing high-flow from
low-flow vascular malformations, especially when the contrast rise time of the
lesion is measured.
Introduction
Vascular lesions of soft tissues are classified into hemangiomas and
vascular malformations on the basis of their natural history, cellular
turnover, and histology [1].
Hemangiomas are benign endothelial cell neoplasms that appear during infancy
and usually have a history of proliferation followed by spontaneous
involution. However, vascular malformations are present at birth and enlarge
in proportion to the growth of the child. Vascular malformations are
subcategorized as lymphatic, capillary, venous, arteriovenous, and mixed
malformations, according to the predominant type of vessel involved
[1–5].
Alternatively, vascular malformations can be classified into low-flow or
high-flow lesions on the basis of hemodynamic flow characteristics
[6–9].
Malformations with arterial components are considered to be high-flow lesions
(arterial malformations containing macrofistulas and arteriovenous
malformations containing microfistulas through a vascular nidus), and those
without arterial components are considered to be low-flow lesions (venous,
capillary, and lymphatic malformations).
Peripheral vascular malformations are treated by surgical resection,
arterial embolization, or direct percutaneous puncture with embolic materials
(sclerotherapy). Sclerotherapy is the least invasive and is considered to be
especially effective in treating low-flow vascular malformations
[1,
10,
11]. However, sclerotherapy is
not indicated for a high-flow lesion in which the infused agents are rapidly
washed out of the lesion. Arterial embolization appears to be the most
effective treatment in high-flow lesions, with occasional subsequent surgical
resection [9,
12,
13]. Therefore, evaluation of
the flow characteristics of vascular malformations is thought to play an
essential role in determining appropriate patient treatment
[13,
14]. Although some articles
[4,
5] reported the usefulness of
time-resolved MRI, we focus on hemodynamics in vascular lesions to
differentiate between high-flow and low-flow vascular malformations.
This study assessed the usefulness of dynamic contrast-enhanced MRI in
distinguishing vascular malformations with arterial components (high-flow
type) from venous malformations (low-flow type) that do not need angiography
for treatment.
Subjects and Methods
Patients
For this study, we prospectively enrolled 16 consecutive patients (six male
and 10 female; age range 1–43 years; median age, 11 years) who were
suspected of having peripheral vascular malformations. Patients were examined
between September 2001 and January 2003. They had no therapy before undergoing
this examination. This study group consisted of six vascular malformations
showing arterial components (high-flow type) and 10 venous malformations
(low-flow type). Lesions were located in the face (n = 11), lower
extremity (n = 2), neck (n =1), upper extremity (n
= 1), and buttock (n =1). The diameters of the vascular malformations
ranged from 10 to 151 mm (mean, 61.6 mm). This study protocol was approved by
the institutional review board. Informed consent was obtained from all adult
patients and all parents of child patients.
MRI
All examinations were performed with a 1.5-T whole-body imager (Magnetom
Vision, Siemens Medical Solutions) equipped with surface coils (head coil,
neck array coil, phased-array body coil, knee coil, and extremity coil). In
all patients, MRI preceded angiography and venography. Sedation was used for
infants. The MRI protocol consisted of fat-suppressed T1-weighted spin-echo
imaging, fat-suppressed T2-weighted fast spin-echo imaging, dynamic
contrast-enhanced MRI, and contrast-enhanced fat-suppressed T1-weighted
spin-echo imaging. Detailed imaging parameters are listed in
Table 1. Contrast-enhanced
fat-suppressed T1-weighted spin-echo images were obtained using the same
sequences as the fat-suppressed T1-weighted fast spin-echo sequences. The
starting time for the acquisition of dynamic contrast-enhanced MR images was
synchronized with the start of an IV injection of gadopentetate dimeglumine
(Magnevist, Schering) at a dose of 0.2 mL/kg of body weight. The injection
rate using a power injector was 2 mL/sec; the injection was immediately
followed by a saline flush at a dose of 0.2 mL/kg of body weight at the same
injection rate. In infants we injected the contrast medium and saline by hand.
Dynamic contrast-enhanced images were continuously obtained every 5 sec for
120 sec. The first unenhanced dynamic image was subtracted from the
contrast-enhanced dynamic images.
Angiography and Venography
Angiography or venography was performed in all patients using an
angiographic unit (C-vision, Shimazu Corporation). Closed-system venography
was performed by direct percutaneous contrast injection into the lesion with a
fine needle. The diagnosis of peripheral vascular malformation was based on
angiographic or venographic findings.
Data Analysis
Two radiologists without knowledge of the clinical and angiographic
findings independently reviewed the presence or absence of flow voids on
conventional MR images (fat-suppressed T1-weighted spin-echo images,
fat-suppressed T2-weighted fast spin-echo images, dynamic contrast-enhanced MR
images, and contrast-enhanced fat-suppressed T1-weighted spin-echo images).
They judged whether flow voids were present on all conventional MR sequences.
If their opinions differed, a conclusion was reached by consensus. A round
region of interest at least 10 mm in diameter was placed on the 16 peripheral
vascular malformations, and the signal intensities of the malformations were
measured on dynamic contrast-enhanced images. Regions of interest were drawn
over an area of the lesion on the slice on which it was largest.
The percentage of enhancement above baseline was calculated using the
following formula:
We analyzed time intervals between beginning of enhancement of an arterial
branch in the vicinity of the lesion in the same slice and the onset of
enhancement in the lesion by visual inspection of the dynamic
contrast-enhanced subtraction images. We defined these time intervals as
"artery–lesion enhancement time." Time intervals between the
onset of the enhancement in the lesion and the time of the maximal percentage
of enhancement above baseline of the lesion within 120 sec were measured. We
defined these time intervals as "contrast rise time" of the
lesion. Statistically significant differences (p < 0.05) were
determined using the Mann-Whitney test for comparison between high-flow and
low-flow vascular malformations in terms of artery–lesion enhancement
time and contrast rise time.
Results
Results of the artery–lesion enhancement times and the contrast rise
times are summarized in Table
2. The time–signal intensity curves with each type of
vascular malformation on dynamic contrast-enhanced MRI are illustrated in
Figure 1. Three of the
high-flow vascular malformations showed flow voids on conventional MR images,
but none of the venous malformations showed flow voids. The sensitivity of the
flow voids for differentiating the high-flow from the low-flow malformations
was 50% (3/6), with specificity of 100% (10/10).
View this table:
[in this window]
[in a new window]
|
TABLE 2: Mean Artery–Lesion Enhancement Time and Contrast Rise Time for
High- and Low-Flow Vascular Malformations on Dynamic MRI
|
|
The range of the artery–lesion enhancement time in the high-flow
vascular malformations (Fig.
2A,
2B,
2C,
2D) was 0–5 sec. The
range of the artery–lesion enhancement time in the low-flow vascular
malformations (Fig. 3A,
3B,
3C,
3D,
3E) was 0–20 sec. The
mean artery–lesion enhancement time of the high-flow vascular
malformations was significantly shorter than that of the low-flow vascular
malformations (Mann-Whitney test, p < 0.05).
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —2-year-old boy with peripheral high-flow vascular
malformation. Transverse T2-weighted fast spin-echo MR image (TR/TE, 4,000/96)
shows vascular malformation (arrow) in right temporalis region.
|
|
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —2-year-old boy with peripheral high-flow vascular
malformation. Dynamic contrast-enhanced subtraction MR image shows start of
arterial enhancement (short arrow) and no lesion enhancement
(long arrow) 15 sec after start of IV bolus of gadopentetate
dimeglumine.
|
|
View larger version (106K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —2-year-old boy with peripheral high-flow vascular
malformation. Dynamic contrast-enhanced subtraction MR image, obtained at same
level as B but 5 sec later, shows immediate and intense lesion
enhancement (arrow).
|
|
View larger version (120K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —2-year-old boy with peripheral high-flow vascular
malformation. Selective angiogram of right superficial temporal artery shows
characteristics of high-flow vascular malformation. Note dilatation of
afferent arteries (long arrow) followed by early enhancement of
enlarged efferent veins (short arrow).
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —36-year-old man with peripheral low-flow vascular
malformation. Dynamic contrast-enhanced subtraction MR image shows start of
arterial enhancement (arrow) 15 sec after start of IV bolus of
gadopentetate dimeglumine.
|
|
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —36-year-old man with peripheral low-flow vascular
malformation. Dynamic contrast-enhanced subtraction MR image, obtained at same
level as B but 5 sec later, shows slight lesion enhancement
(arrow).
|
|
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —36-year-old man with peripheral low-flow vascular
malformation. Dynamic contrast-enhanced subtraction MR image, obtained at same
level as B but 75 sec later, shows more intense lesion enhancement
(arrow).
|
|
A scatterplot of the artery–lesion enhancement time for each vascular
malformation is shown in Figure
4. The range of the contrast rise time of the high-flow vascular
malformations (Fig. 2A,
2B,
2C,
2D) was 5–10 sec. The
range of the contrast rise time of the low-flow vascular malformations (Fig.
3A,
3B,
3C,
3D,
3E) was 50–100 sec. The
mean contrast rise time of the high-flow vascular malformations was
significantly shorter than that of the low-flow vascular malformations
(Mann-Whitney test, p < 0.01). A scatterplot of the contrast rise
time of each vascular malformation is shown in
Figure 5.
View larger version (3K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —Scatterplot of artery–lesion enhancement time for each
vascular malformation. All high-flow vascular malformations show an
artery–lesion enhancement time of less than 10 sec. High-flow vascular
malformations show shorter artery–lesion enhancement time than low-flow
malformations, with overlap between the two. Use of a threshold
artery–lesion enhancement time of 5 sec would result in 100% (6/6)
sensitivity and 60% (6/10) specificity for differentiation of high-flow from
low-flow malformations. Note: artery–lesion enhancement time = interval
from beginning of enhancement of an arterial branch in vicinity of lesion in
same slice to onset of enhancement in lesion. Start of arterial enhancement is
defined as 0 sec. Number next to  is number of patients at that
enhancement time.
|
|
View larger version (3K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —Scatterplot shows contrast rise time for each vascular
malformation. All high-flow vascular malformations show contrast rise time of
less than 20 sec. High-flow vascular malformations show shorter contrast rise
time than low-flow vascular malformations, with no overlap between the two.
Use of threshold contrast rise time of 30 sec would result in 100% (6/6)
sensitivity and 100% (10/10) specificity for differentiation of high-flow from
low-flow malformations. Note: contrast rise time = time between onset of
lesion enhancement and time of maximal percentage of enhancement above
baseline within 120 sec. Onset of lesion enhancement is defined as 0 sec.
Number next to is number of patients at that enhancement time.
|
|
Discussion
Goals of imaging peripheral vascular malformations are twofold: to define
the anatomic extent of the lesion and to distinguish low-flow vascular
malformations from high-flow vascular malformations. The most important
characterizing feature of vascular malformations is whether the lesion is a
high-flow or low-flow vascular malformation, because direct percutaneous
sclerotherapy is described as the treatment of choice for low-flow vascular
malformations [1,
10,
11]. Direct puncture of
vascular channels is performed using a combination of sonographic and
fluoroscopic guidance. Therefore, correct diagnosis of low-flow vascular
malformations with MRI could eliminate angiography for differentiating the two
types of vascular malformations.
A main feature on conventional MRI is reported to be the presence or
absence of flow voids in categorizing vascular malformations
[4,
8]. High-flow vessels of
high-flow vascular malformations are shown as linear signal voids on spin-echo
imaging and bright signal on gradient echo sequences. Meyer et al.
[15] reported that all three
high-flow vascular malformations had flow voids on spin-echo images, but only
two had vascular enhancement on gradient echo images. One small high-flow
vascular malformation was dark, perhaps from turbulent flow. Rak et al.
[8] reported the presence of
flow voids in all untreated arterial and arteriovenous malformations, whereas
van Rijswijk et al. [4]
reported that all arterial malformations and only two of four arteriovenous
malformations showed flow voids in their study. Van Rijswijk et al. performed
dynamic contrast-enhanced MRI in an attempt to better differentiate the
various categories of vascular malformations. They reported that dynamic
enhancement could not be used as a feature to differentiate high- and low-flow
malformations using artery–lesion enhancement time because all high-flow
malformations, and some of the low-flow malformations, displayed early
enhancement. Their findings are supported by the results of our study using
artery–lesion enhancement time.
Actually, artery–lesion enhancement time does not reflect the
hemodynamics of vascular lesions directly. We attempted to measure contrast
rise time with dynamic contrast-enhanced MRI to distinguish high-flow from
low-flow malformations because it can reflect hemodynamics of vascular
lesions. The contrast rise time of the lesion was therefore shorter in
high-flow malformations than in low-flow malformations, and no overlap was
observed between the two groups because it can reflect hemodynamics of
vascular lesions. Herborn et al.
[5] reported that dynamic
time-resolved contrast-enhanced 3D MR angiography allowed reviewers to
correctly classify the lesions in all patients. This technique could depict
feeding and draining vessels with high spatial resolution. However, the
technique provided limited temporal resolution to estimate the hemodynamics of
vascular lesions.
The two noninvasive imaging techniques that are most useful for the
examination of vascular malformations are MRI and sonography. Sonography has
been advocated as useful in examining vascular malformations
[16,
17]. Certainly, Doppler
sonography has been used in differentiating low-from high-flow vascular
malformations [18]. However,
sonography has limitations, including the small field of view and restricted
depth of penetration, especially with high-frequency transducers. Thus,
sonography cannot always be substituted for MRI when determination of the full
extent of vascular malformations is necessary.
In addition to differentiating low-from high-flow vascular malformations,
it is important to know in which tissues the vascular malformation is involved
and whether adjacent vital structures, such as neurovascular bundles, are
involved by the lesion. Such information is vital to planning surgery or
imaging-guided procedures. MRI is excellent for defining the extension of
vascular malformations and their relationship to adjacent structures such as
neurovascular bundles. In fact, Donnelly et al.
[9] reported that MRI is the
primary imaging technique for the evaluation of suspected vascular
malformations. Most information needed to examine the lesion is available from
conventional MR images. In particular, the high accuracy of heavily
T2-weighted images in defining the extent of vascular malformations has been
described before and is already widely used in clinical practice
[4,
8,
9]. In addition to defining the
extent of vascular malformations on conventional MR images, dynamic
contrast-enhanced MR images can provide information about the hemodynamics of
vascular lesions. MRI may become a one-stop examination for the evaluation of
vascular malformations.
Our study has several potential limitations. First, the patient population
was small. Larger-scale studies are needed to validate our results. Second, we
did not include lymphatic malformations or capillary malformations (port-wine
stains) in our study. We cannot evaluate the differentiation of venous
malformations from lymphatic malformations. Generally, low-flow vascular
malformations include venous, lymphatic, and mixed malformations.
Gadolinium-enhanced T1-weighted images typically show enhancement of the
slow-flowing venous channels and no central enhancement of the lymphatic
components [9]. Third,
histopathologic correlation of each lesion was impossible because all low-flow
malformations were treated with sclerotherapy. Fourth, we selected 5-sec
temporal resolution because of scanner limitations. In our study, the longest
contrast rise time of a high-flow vascular malformation was 10 sec. On the
other hand, the shortest contrast rise time of a low-flow vascular
malformation was 50 sec. The 5-sec temporal resolution was enough to
distinguish the high-flow from the low-flow vascular malformations. More
studies may be needed to consider whether this temporal resolution is optimal.
Fifth, we selected gadopentetate dimeglumine at a dose of 0.2 mL/kg of body
weight and a 2 mL/sec injection rate. In patients weighing more than 50 kg,
the bolus duration is longer than 5 sec. A faster injection rate would be
better because there is no point in having the bolus duration longer than the
5-sec acquisition time.
In conclusion, dynamic contrast-enhanced MRI is useful for distinguishing
high-flow from low-flow vascular malformations, especially when the contrast
rise time of the lesion is measured.
References
- 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]
- Burrows PE, Mulliken JB, Fellows JB, Strand RD. Childhood
hemangiomas and vascular malformations: angiographic differentiation.
AJR 1983; 141:483
–488[Abstract/Free Full Text]
- Fishman SJ, Mulliken JB. Hemangiomas and vascular malformations of
infancy and childhood. Pediatr Clin North Am1993; 40:1177
–1200[Medline]
- van Rijswijk CS, van der Linden E, van der Woude HJ, van Baalen JM,
Bloem JL. Value of dynamic contrast-enhanced MR imaging in diagnosing and
classifying peripheral vascular malformations. AJR2002; 178:1181
–1187[Abstract/Free Full Text]
- Herborn CU, Goyen M, Lauenstein TC, Debatin JF, Ruehm SG, Kroger K.
Comprehensive time-resolved MRI of peripheral vascular malformations.
AJR 2003; 181:729
–735[Abstract/Free Full Text]
- Inoue Y, Ohtake T, Wakita S, et al. Flow characteristics of
soft-tissue vascular anomalies evaluated by direct puncture scintigraphy.
Eur J Nucl Med 1997;24
: 505–510[Medline]
- Baker LL, Dillon WP, Hieshima GB, Dowd CF, Friden IJ. Hemangiomas
and vascular malformations of the head and neck: MR characterization.
AJNR 1993; 14:307
–314[Abstract]
- Rak KM, Yakes WF, Ray RL, et al. MR imaging of symptomatic
peripheral vascular malformations. AJR1992; 159:107
–112[Abstract/Free Full Text]
- Donnelly LF, Adams DM, Bisset GS. Vascular malformations and
hemangiomas: a practical approach in a multidisciplinary clinic.
AJR 2000; 174:597
–608[Free Full Text]
- Yakes WF, Haas DK, Parker SH, et al. Symptomatic vascular
malformations: ethanol embolotherapy. Radiology1989; 170:1059
–1066[Abstract/Free Full Text]
- Lee BB, Kim DI, Huh S, et al. New experiences with absolute ethanol
sclerotherapy in the management of a complex form of congenital venous
malformation. J Vasc Surg 2001;33
: 764–772[CrossRef][Medline]
- Yakes WF, Rossi P, Odink H. How I do it: arteriovenous malformation
management. Cardiovasc Intervent Radiol1996; 19:65
–71[CrossRef][Medline]
- Enjorlas O, Mulliken JB. The current management of vascular
birthmarks. Pediatr Dermatol 1993;10
: 311–333[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]
- Meyer JS, Hoffer FA, Barnes PD, Mulliken JB. Biological
classification of soft-tissue vascular anomalies: MR correlation.
AJR 1991; 157:559
–564[Free Full Text]
- Dubois J, Soulez G, Oliva VL, Berthiaume MJ, Lapierre C, Therasse
E. Soft-tissue venous malformations in adult patients: imaging and therapeutic
issues. Radiographics 2001;21
:1519
–1531[Abstract/Free Full Text]
- Trop I, Dubois J, Guibaud L, et al. Soft-tissue venous
malformations in pediatric and young adult patients: diagnosis with Doppler
US. Radiology. 1999;212
: 841–845[Abstract/Free Full Text]
- Paltiel HJ, Burrows PE, Kozakewich HP, Zurakowski D, Mulliken JB.
Soft-tissue vascular anomalies: utility of US for diagnosis.
Radiology 2000;214
: 747–754[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
) increases rapidly and decreases gradually, and that of
low-flow vascular malformations (
