AJR 2005; 185:92-94
© American Roentgen Ray Society
Intrathecal Gadolinium-Enhanced MR Myelography Showing Multiple Dural Leakages in a Patient with Marfan Syndrome
Nils Kraemer,
Ansgar Berlis and
Martin Schumacher
Department of Neuroradiology, University Hospital Freiburg, Breisacher
Strasse 64, 79106 Freiburg, Germany.
Received January 5, 2004;
accepted after revision September 7, 2004.
Address correspondence to A. Berlis
(berlis{at}nz.ukl.uni-freiburg.de).
Introduction
Loss of CSF or imbalance of its turnover can cause symptoms such as
orthostatic headaches, nausea, and neck pain. These symptoms are frequently
the result of CSF leakage from the dura and can occur spontaneously or after
trauma, lumbar puncture, or neurosurgery. It has been suggested that disorders
of the connective tissues predispose patients to spontaneous CSF leaks
[1]. Hence, localization and
identification of dural leakages are crucial for satisfactory treatment.
We present the case of a patient with Marfan syndrome and multiple dural
leaks. The exact site of CSF leakage was not apparent on conventional
myelography or high-resolution CT myelography, but was shown using intrathecal
contrast-enhanced MR myelography.
Case Report
A 19-year-old man with Marfan syndrome presented with severe headaches,
neck pain, and nausea. These symptoms were triggered by orthostatic conditions
involving an erect posture (e.g., sitting, standing, or walking). However, the
symptoms resolved after lying down for several minutes. Blood analysis and
neurologic examinations presented normal parameters. Intrathecal pressure was
not measurable after lumbar puncture. Spontaneous intracranial hypotension
(SIH) and chronic loss of CSF were assumed to be responsible for the symptoms.
MR images using T1- and T2-weighted and constructive interference in
steady-state 3D imaging showed dural ectasia at different levels of the
lumbosacral spine and extradural fluid with CSF-like signal performance
dorsally to L5 and S1, but no connection to the subarachnoid space was proven.
A CT scan of the brain showed normal cisterns, no caudal displacement of the
brainstem, and no evidence of low intracranial pressure. Myelography, a
postmyelography high-resolution CT scan, and T2-weighted MR myelography (Figs.
1A and
1B) showed dysplasia of the
lumbosacral subarachnoid space and a CSF-filled compartment with
contrast-medium uptake dorsal to the dura. However, the examinations failed to
localize the dural leakages.
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. Dural ectasia (arrows) and extradural CSF were noted
dorsally to L5 and S1 on conventional myelography and CT myelography image
(not shown) obtained 1 day before C-F.
|
|
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. T2-weighted MR myelography image obtained 1 day before C-F
shows extradural CSF (arrows) but fails to show exact localization of
leakages.
|
|
In an attempt to localize the leakages and determine the best form of
therapy, a contrast-enhanced MR examination was performed after intrathecal
injection of a gadolinium contrast agent. As this agent was not approved for
intrathecal application, the patient was informed of the experimental nature
of this method and written consent was obtained. For this examination, 0.3 mL
of 0.5 mol/L gadobenate dimeglumine (MultiHance, Bracco) was diluted in 10 mL
of 0.9% saline and injected directly into the lumbar subarachnoid space. With
an estimated CSF volume of 150 mL, the final concentration of gadobenate
dimeglumine was about 1 µmol/mL of CSF. This concentration has been shown
as ideal for enhancement of CSF
[2].
MRI was performed on a 1.5-T magnet (Magnetom Vision, Siemens Medical
Solutions) at 45 min, 3 hr, and 6 hr after injection of gadobenate
dimeglumine. Axial and coronal T1- and T2-weighted images were acquired.
At 45 min after contrast injection, conspicuous enhancement of the CSF was
noted on T1-weighted images. The examination revealed contrast filling of the
dorsal extraspinal compartment (arrow,
Fig. 1C) with a small channel
(arrowheads, Fig. 1C)
communicating with the subarachnoid space. On T2-weighted imaging at the same
time, an inverse loss of signal intensity was observed.
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. Axial T1-weighted image obtained after intrathecal injection of
gadobenate dimeglumine reveals existence of extradural CSF (arrow)
and small structure communicating with this compartment
(arrowheads).
|
|
The CSF was still enhanced on T1-weighted images 3 hr after gadobenate
dimeglumine injection while a normal hyperintense signal was noted on
T2-weighted images. In addition, on T1-weighted images, multiple sites of
dural leakage (arrowheads, Fig.
1D) were noted in the thoracolumbosacral spine and the enhanced
CSF (arrows, Figs. 1D,
1E,
1F) was visualized outside the
subarachnoid space in the retroperitoneum alongside the psoas muscle on both
sides.
View larger version (153K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. Obtained 3 hr after C, axial T1-weighted image shows multiple
sites of contrast extravasation (arrowheads) at different levels.
Arrows show gadobenate dimeglumine-enhanced CSF in retroperitoneum.
|
|
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1E —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. Obtained 3 hr after C, coronal T1-weighted images show
gadobenate dimeglumine-enhanced CSF is also seen in retroperitoneum
(arrows). Radiographic and T2-weighted MR images (both not shown)
failed to show these structures.
|
|
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1F —19-year-old man with Marfan syndrome suffering from severe
headaches. Spontaneous intracranial hypotension was assumed to cause these
symptoms. Obtained 3 hr after C, coronal T1-weighted images show
gadobenate dimeglumine-enhanced CSF is also seen in retroperitoneum
(arrows). Radiographic and T2-weighted MR images (both not shown)
failed to show these structures.
|
|
The subarachnoid space and cystic dorsal structure were still enhanced 6 hr
after the injection of gadobenate dimeglumine, although enhancement in the
retroperitoneum could no longer be shown.
No side effects or adverse reactions appeared during or after the
examination.
The multiple fistulas were too small and numerous to occlude by
neurosurgery. Treatment was limited to bed rest for 6 weeks in expectation of
self-occlusion of the fistulas.
On follow-up at 4 months, the patient could sit for about 3 hr without
symptoms. Currently, walking and standing cause marginal problems.
Discussion
Orthostatic headache, nausea, nuchal pain, stiffness, and vomiting are the
main symptoms of CSF loss or imbalance of intracranial pressure
[3]. The idiopathic appearance
of these symptoms is mostly described as SIH, which is generally considered a
rare dysfunction. CSF leakages are responsible for the majority of SIH and
subsequent headaches [3]. The
cause of SIH appearance and dural leakages often is unknown, although
connective tissue disorders may predispose to the appearance of spontaneous
dural leakages [1].
The most common findings of Marfan syndrome are changes in the skeletal,
cardiovascular, and ocular system
[4], but as the meningeal
fibers are also of poor quality, a specific symptom in many Marfan syndrome
patients is the existence of dural ectasia
[5], where fistulas can
occur.
In order to confirm the diagnosis and decide on further treatment (e.g.,
neurosurgery, blood patches, or conventional therapy), it is important to
detect these fistulas.
The localization and identification of small dural fistulas is often
challenging. Myelography and CT myelography are the most common diagnostic
techniques. In our case, these procedures failed to show the fistulas. The use
of T2-weighted MR myelography may be useful for the visualization of
hyperintense CSF without application of a contrast agent. However, in certain
circumstances, depiction of CSF pathologies may be difficult on T2-weighted MR
myelography [6] alone. In such
cases, T1-weighted MRI after intrathecal application of MR contrast media may
give additional information.
Unlike iodinated contrast media, increasing concentrations of gadolinium
contrast agents do not necessarily give a linear increase in signal intensity.
As with other gadolinium agents, gadobenate dimeglumine enhances the MR signal
intensity on T1-weighted imaging but has no effect on the signal on
T2-weighted sequences. However, at high concentrations, signal loss may occur
because of the advent of T2* effects. In the present case, we
obtained excellent enhancement of the CSF on T1-weighted images after
approximately 150 µmol of diluted gadobenate dimeglumine was directly
injected into the subarachnoid space, to give a final concentration of about 1
µmol of gadobenate dimeglumine per milliliter of CSF
[2]. However, a loss of signal
intensity on T2-weighted imaging was noted on early scans because the contrast
agent was still mostly located in the lumbar subarachnoid space.
The strong enhancing power of gadobenate dimeglumine on T1-weighted imaging
allowed visualization of very small volumes of CSF. This was noted
particularly for CSF located outside the subarachnoid space and permitted the
presentation of extremely small fistulas that were not seen on CT myelography.
Distribution of gadobenate dimeglumine within the subarachnoid space and
subsequent rapid elimination within 3 hr of injection resulted in reduced
concentrations of gadobenate dimeglumine in the subarachnoid space and a
return to normal hyperintense signal on T2-weighted images, while the CSF
retained an enhanced signal on T1-weighted images.
Numerous studies have shown gadolinium-based MR contrast agents to be safe
for IV use [7].
Even if the lack of adverse events in the present case and the absence of
gadobenate dimeglumine diffusion into the parenchyma of the spinal cord are
evidence of the good tolerance of intrathecally injected gadobenate
dimeglumine, this procedure still must be validated with further studies.
However, previous animal studies on intrathecal gadobenate dimeglumine
[2] and clinical trials on
other intrathecal MR contrast media
[8] are encouraging.
In conclusion, the present case suggests that the intrathecal application
of gadobenate dimeglumine may be an effective alternative for the depiction of
dural leakages. Conspicuous enhancement in small volumes may permit the
visualization of tiny structures that are too small for efficient contrasted
visualization during X-ray examinations.
References
- Mokri B, Maher CO, Sencakova D. Spontaneous CSF leaks: underlying
disorder of connective tissue. Neurology2002; 58:814
-816[Abstract/Free Full Text]
- Kramer N, Berlis A, Klisch J, Kubalek R, Miosczka H, Schumacher M.
Intrathecal gadolinium-enhanced MR-cisternography: depiction of the
subarachnoidal space and evaluation of gadobenat-dimeglumin-(Gd-BOPTA,
"Multihance") toxicity in an animal model and a clinical case.
Acad Radiol2002; 9[suppl 2]:S447
-S451
- Lay CM. Low cerebrospinal fluid pressure headache. Curr
Treat Options Neurol 2002;4:357
-363[Medline]
- Pyeritz RE. The Marfan syndrome. Annu Rev
Med 2000;51:481
-510[CrossRef][Medline]
- Fattori R, Nienaber CA, Descovich B, et al. Importance of dural
ectasia in phenotypic assessment of Marfan's syndrome.
Lancet 1999;354:910
-913[CrossRef][Medline]
- Hergan K, Amann T, Vonbank H, Hefel C. MR-myelography: a comparison
with conventional myelography. Eur J Radiol1996; 21:196
-200[CrossRef][Medline]
- Kirchin MA, Runge VM. Contrast agents for magnetic resonance
imaging: safety update. Top Magn Reson Imaging2003; 14:426
-435[CrossRef][Medline]
- Tali ET, Ercan N, Krumina G, et al. Intrathecal gadolinium
(gadopentetate dimeglumine) enhanced magnetic resonance myelography and
cisternography: results of a multicenter study. Invest
Radiol 2002;37:152
-159[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
P-H Tsai, J-L Fuh, J-F Lirng, and S-J Wang
Comparisons Between Heavily T2-Weighted MR and CT Myelography Studies in Two Patients with Spontaneous Intracranial Hypotension
Cephalalgia,
June 1, 2008;
28(6):
653 - 657.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Albayram, A. Gunduz, and H. Ozer
Contrast Agent for Intrathecal Gadolinium Administration
Am. J. Roentgenol.,
October 1, 2006;
187(4):
W446 - W446.
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. Kramer and A. Berlis
Reply
Am. J. Roentgenol.,
October 1, 2006;
187(4):
W447 - W447.
[Full Text]
[PDF]
|
|
|
Top
Introduction
Case Report
