Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
AJR:159,
November
MR
1992
A
OF
CLIVAL
1073
ABNORMALITIES
B
C
Fig. 16.-Infection of petrous apex (Gradenigo’s syndrome) and clivus. A, Contrast-enhanced axial TI-weighted MR Image shows epidural enhancing mass (arrow) in cerebellopontine angle. Bone involvement cannot be determined with contrast enhancement. Gradenigo’s syndrome classically consists of triad of sixth-nerve palsy, pain in trigeminal nerve distribution, and chronic otitis media In patients with petrous aplcftis. B and C, CT scans show epidural mass (white arrow) and bone erosion of petrous and clival bones (arrowheads). Note inflammatory mass in middle ear (black arrows).
Fig. 17.-Mucocele
involving
sphenoidal
sinus
and cllvus In a 60-year-old man. Contrast-enhanced midline sagittal MR Image shows mass of Intermediate signal markedly expanding sphenoidal sinus and clivus. Only mucopenosteum enhances (arrows). Classically, a mucocele develops from obstruction of a sinaI ostium.
Fig. 18.-Rheumatoid arthritis in a 64-year-old woman. A, Midline sagfttal Ti-weighted MR Image reveals pannus formation (open arrow) about dens disrupting normal infraclival fat (white arrow). Note markedly widened atiantodens Interspace of 11 mm measuring from cortex (black line representing compact cortical bone Is shown by solid black arrows). Normal atlantodens interspace in adults is 2.5 mm and in children It is up to 5 mm. Anteroposterior diameter of spinal canal is narrowed. B, Lateral tomogram shows atlantoaxial subluxation.
and T2-weighted images. They found the normal adult clival signal usually iso- or hyperintense relative to the pons on Ti weighted images and approximately isointense with the pons on T2-weighted images. Abnormalities were hypointense in relation to the pons on the Ti-weighted images and varied
Administration of gadopentetate dimeglumine is useful in delineating vascular anatomy and disease in the paraclival
from diffusely
disease, [3].
-
on T2-weighted
hypointense images.
to hyperintense
relative to the pons
region;
however,
enhance unenhanced
many
and become images
lesions
in the marrow
isointense are
more
with sensitive
and are more sensitive
normal
of the clivus will marrow. Thus,
for detecting
than T2-weighted
marrow
images
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
i 074
CHAUUB
ET
AL.
AJR:159, November 1992
Fig. 19.-Thrombosed aneurysm in vertebrobasilar artery along dorsal clivus in a 57-year-old woman. A, Midline sagittal Ti-weighted MR Image shows posterior displacement of medulla (curved arrow) (straight
by a large intermediate-signal mass arrow). Round shape and predominant
lack of flow void suggest thrombosed aneurysm. B, Selected lateral vertebral arteriogram shows posterior displacement of vertebral artery (solid arrow) and ectasia of basilar artery (open arrow), but no aneurysmal
Fig. 20.-jf-Thalassemla
Midline Ti-weighted ment
of normal
diploetic sue.
space
bright
In a 22-year-old
man.
MR image shows replacein clivus
and
of calvaria by hematopoletic
fafty
signal
tis-
Fig. 21.-FIbrous
dysplasla
In a 54-year-old
man. Sagittal TI-weighted MR image shows low signal In sphenoid bone and clivus (arrow). A CT scan showed osseous expansion and hazy sole-
rosis characteristic
of fibrous dysplasla.
filling.
Fig. 22.-Tornwaldt’s cyst in a 3i-year-old man. Midline sagfttalTi-weighted MR image displays a high-signal mass along midline of superior nasopharynx and ventral clivus (arrow). Track from notochordal remnants may give rise to this cyst, which can be complicated by infec-
tion.
REFERENCES 1 . Freedy AM, Miller KD Jr. Granulocytic sarcoma (chloroma): sphenoidal sinus and paraspinal involvement as evaluated by CT and MR. AJNR i99i;12:259-262 2. Kimura F, Kim KS, Friedman H, et al. MR imaging of the normal and abnormal clivus. AJNR 1990;11 :1015-1021 3. Stimac GK, Porter BA, Olson DO, et al. Gadolinium-DTPA-enhanced MR
imaging of spinal neoplasms: preliminary investigation and comparison with enhanced spin-echo and STIR sequences. AJNR 1988:9:839-846 SUGGESTED
READING
1 . Lame FJ, Nadel L, Braun IF. CT and MR imaging of the central skull base. Part 1 : Techniques, embryologic development and anatomy. RadioGraphics 1990:10:591-602
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
i 075
Case Report .. ‘.
Miliary Tuberculosis Gloria
T. Gee,1
Canlos
Bazan
Ill, and
Involving J. Randy
Jinkins
We report a case of miliary cerebral tuberculosis in a 14month-old child who began having seizures while on quadrupIe drug antituberculous therapy. MR imaging showed numenous round, very small enhancing lesions scattered throughout the brain, compatible with this diagnosis. Case
Report
A 14-month-old girl was referred for evaluation and control of seizures of several months’ duration. Tuberculous meningitis had been diagnosed 3 months earlier on the basis of clinical history and family exposure to tuberculosis. While the child was being treated with antituberculous oral medications, she began having seizures, and left-sided weakness developed suddenly. Physical examination revealed a temperature of 38.8#{176}C, slight weakness of the left leg, and clonus at the right ankle. The chest radiograph obtained when she was admitted showed a bilateral pattern of diffuse reticulonodular opacities suggestive of miliary pneumonia. Examination of CSF revealed an inflammatory pleocytosis and a low glucose content. A tuberculin skin test (PPD) was initially nonreactive, but a second PPD test 2 months later showed a 1 3 x 1 2 mm area of erythema after 24 hr. MR examination of the brain showed focal areas of hypointensity in the right temporopanietal and left occipital lobes on Ti-weighted images and increased signal on T2-weighted images (Figs. iA and 1 B). After IV administration of gadopentetate dimeglumine (0.1 mmol/ kg), we noted numerous round, very small tered throughout the brain (Fig. 1 C). Most proximately 1 mm in diameter.
enhancing of these
lesions scatmeasured ap-
Discussion tuberculosis
can be described
geal or parenchymal)
Cranial
and type (diffuse
Received 1
AJR
(menin-
tubencu-
loma, tubenculous abscess, or focal cerebnitis) [i ]. Associated changes also are commonly encountered, such as artenitis, infarction, hemorrhage, and hydrocephalus [1 2]. Tubencubus meningitis represents less than 5% of the cases of bacterial meningitis reported in children. Infection in children usually results from inhalation of organisms shed by adults with tuberculosis [1]. Entry of Mycobacterium tuberculosis into the CNS is believed to be by hematogenous spread from the lung to the meninges, where small tuberculomas are formed. These may rupture into the subanachnoid space and ,
lead to meningitis, the severity of which is partly related each patient’s local tissue reaction to the organism. Sixty-eight are reported
percent to have
nadiognaphs,
and
to
of children with tubenculous meningitis concomitant miliany tuberculosis of the brain. Conversely, in only 1 % of children with miliany cerebral tuberculosis does tubenculous meningitis develop subsequently [1]. Few patients with miliary tuberculosis have fever at presentation. Up to 50% have abnormal findings on chest only
25% have a positive
tuberculin
skin
test [3]. Fever, an elevated erythrocyte sedimentation rate, and evidence of extracranial tubenculous infection have been reported in 32-50% of patients with cerebral tubenculomas [2,4]. In cases of tuberculous meningitis, the basal cisterns are isodense to hyperdense compared with normal CSF on unenhanced CT scans. After IV administration of iodinated contrast medium, intense enhancement of the affected cisterns was noted. Communicating hydnocephalus is common and has been reported to be a sensitive indicator of tubencubus meningitis in children [5]. lschemic infancts are most common in the distribution of
18, 1 992; accepted after revision May 15, 1992. Neuroradiology Section, The University of Texas Health Science Center, 7703 Floyd Curl Dr. , San Antonio, TX 78284-7800.
March
All authors:
requests
by location
meningitis,
the Brain: MR Findings
to J. A. Jinkins.
159:1075-1076,
November
1992 0361-803X/92/1
595-1075
© American Roentgen Ray Society
Address reprint
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1076
GEE
A
ET AL.
AJR:159,
B
November
C
Fig. 1.-A, Ti-weighted (600/20) axial spin-echo MR Image shows hypointense areas in right temporoparietal lobe and left occipital lobe. B, T2-welghted (2500/20) MR image shows hyperintense areas in right temporoparietal lobe, left occipital lobe, and left side of thalamus. C, TI-weighted (600/20) axIal spin-echo MR image after IV contrast administration shows numerous round, very small enhancing lesions brain, with confluent areas appearing in right temporoparietal and left occipital areas.
the middle
cerebral
artery,
including
the lenticulostniate
arten-
ies [6, 7], and are associated with tubenculous arteritis caused by the basal meningitis [8]. MR imaging has been reported to be more sensitive than CT in detecting disease affecting the basal cisterns [8]. Contrast-enhanced MR is particularly useful for detecting small foci of meningeal inflammation [8]. MR is reported to be more sensitive than CT in detecting tuberculomas of the cerebral parenchyma. Tuberculomas are reported to be isointense with gray matter on Ti -weighted MR images. On T2-weighted images, lesions show central hypenmntensity. In some cases, a hypointense ring is apparent within the wall of the tuberculoma on T2-weighted images.
Most
tuberculomas
are further
by a collar
of high
signal resulting from edema on T2-weighted images. culomas typically enhance after the IV administration
Tuberof ga-
dopentetate dimeglumine The MR examination in areas of enhancement in left occipital lobe, which tubenculomas,
weighted
concomitant
images,
multiple
outlined
in a solid or ring pattern. the current case showed confluent the right temporoparietal lobe and may have represented coalescent ischemic disease, or both. On T2small
hyperintense
1992
foci were
scat-
tered throughout the brain parenchyma, all ofwhich enhanced with IV administration of contrast medium, consistent with
throughout
the diagnosis of miliary cerebral tuberculosis. The homogeneous enhancement pattern suggests that the lesions are granulomas rather than abscesses, which would be expected to have a central
fluid collection.
REFERENCES 1 . Jinkins JR. Computed
tomography of intracranial tuberculoma. Neurora126-1 35 2. Draouat S, Abdenabi B. Ghanem M, Bourjat P. Computed tomography of cerebral tuberculoma. J Comput Assist Tomogr 1987;i 1 : 594-597 3. Witham RR, Johnson RH, Roberts DL. Diagnosis of miliary tuberculosis by cerebral computerized tomography. Arch Intern Med 1979;139: 479-480 4. de Castro CC, Hesselink JR. Tuberculosis. Neuroimaging C/in North Am diology 1991:33:
1991;1
:119-135
5. Waecker NJ, Connor JD. Central nervous system tuberculosis in children: a review of 30 cases. Pediatr Infect Dis J i990;9:539-543 6. Menkes JH. Infections of the nervous system: Chronic and granulomatous infections of the meninges. In: Textbook of child neurology. Philadelphia: Lea & Febiger, 1985:335-414 7. Leisuarda A, Berthier M, Starkstein S. Ischemic infarction in 25 children with tuberculous meningitis. Stroke i988;19:200-204 8. Kee H-C, Moon HH, Jae KR, et al. Gd-DTPA-enhanced MR imaging of the brain in patients with meningitis. AJR 1990:154:809-816
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1077
Artery
Femoral
Pseudoaneurysms:
Value of Color Doppler Predicting Which Ones Without Treatment
H Enik K. Paulson1 Barbara S. Hertzberg Susan S. Paine Barbara A. Carroll
.
OBJECTIVE. The purpose of this study teristics of femoral artery pseudoaneurysms
determine would
if color Doppler sonography
ultimately
MATERIALS
studied spectral
thrombose
can
be used
thrombosis
of the
which pseudoaneurysms
to predict
spontaneously.
METHODS.
AND
was to evaluate the Doppler Imaging characoccurring after cardiac catheterization to
Over a 30-month-period,
with serial color Doppler sonography. images and clinical records of patients
spontaneous repair
Sonography in Will Thrombose
were compared
with
24
pseudoaneurysms
were
The color flow, B-mode, and Doppler whose pseudoaneurysms demonstrated those of patients who required surgical
pseudoaneurysm.
RESULTS.
We
found
no statistically
significant
differences
between
the
patients
whose pseudoaneurysms demonstrated spontaneous thrombosis and those treated surgically in regard to clinical parameters or the volume of the pseudoaneurysm, of flow within the pseudoaneurysm, ratio of forward to reversed flow velocity in the pseudoaneurysm neck, duration of diastolic flow in the pseudoaneurysm neck, or length of pseudoaneurysm neck. However, on color Doppler images, the volume of flow in the lumens of pseudoaneurysms that thrombosed spontaneously (1.8 ± 3.3 ml) was significantly smaller than the volume of flow in the lumens of those treated surgically (4A ± 3.2, ml, p = .02). CONCLUSION. We conclude that pseudoaneurysms with small volumes of flow in the lumen are more likely to thrombose than are those with large volumes of flow in the Iumen However, color Doppler sonographic characteristics cannot be used to predict percentage
subsequent AJR
thrombosis.
159:i077-i081,
Femoral
November
1992
are a known complication of cardiac catheterization for vascular access [i -3]. With increased use of largediameter arterial sheaths in interventional cardiac procedures, these pseudoaneurysms are being encountered more frequently [i 4-6]. Traditional teaching suggests that pseudoaneurysms should be repaired surgically to avoid rupture [4, 79]. Although surgery is effective and safe, it is expensive and is associated with prolonged hospitalization. Recent reports raise questions about the standard surgical approach. First, many pseudoaneurysms thrombose spontaneously, suggesting that close observation and serial imaging may be indicated in some patients [5, 6, 1 0-i 3]. Second, nonsurgical sonographically guided compression-occlusion therapy has been suggested as a viable alternative to surgical management [14]. The purpose of this study was to evaluate the imaging characteristics of in which
pseudoaneurysms
the groin
is used
,
Received February 1 1 , 1992; accepted after revision May 22, 1992.
Presented at the annual meeting of the American Roentgen Ray Society, Orlando, FL, May 1992. ‘ All authors: Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 2771 0. Address reprint requests to E. K. Paulson. 0361-803X/92/1 © American
595-1077
Roentgen
Ray Society
pseudoaneurysms managed conservatively to determine phy can be used to predict which pseudoaneurysms
if color Doppler will thrombose
sonograwithout
treatment. Materials
and
Methods
A computer-based search of sonographic studies conducted over a 30-month period 1 , 1989, to July 1 , 1991) indicated that 430 patients had had color Doppler
(January
PAULSON
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1078
sonography of the groin. Most of these examinations were performed after cardiac catheterization in order to rule out a pseudoaneurysm or arteriovenous fistula. Seventy-nine pseudoaneurysms were identified in 78 patients; their color Doppler sonographic reports were reviewed. Fifty-five of the pseudoaneurysms were imaged only once by using color Doppler sonography and were excluded from the study group. The remaining 24 were examined at least twice with color Doppler sonography; they form the basis of this report. Twenty-three patients were included in
this study. One patient had two cardiac catheterizations
21 months apart; each procedure was complicated aneurysm
that
was
examined
sequentially.
rysms were imaged twice (original
Fourteen
examination
performed by a pseudopseudoaneu-
and one follow-up
examination) and 1 0 were imaged three times (original examination and two follow-up examinations). Thus, a total of 58 color Doppler sonograms were reviewed. Patients with pseudoaneurysms were divided into two groups on the basis of clinical outcome. One group included patients with pseudoaneurysms in whom spontaneous thrombosis occurred, without surgical or nonsurgical intervention. The other group included patients who had color Doppler sonography but had surgery before thrombosis was detected. All were followed up as inpatients except for four patients who had spontaneous thrombosis who were dis-
charged with a known pseudoaneurysm with a scheduled follow-up study. All patients had cardiac catheterization except for one in the spontaneous
All color
thrombosis group who had iliac angioplasty. Doppler images were obtained by using commercially
available real-time equipment (Acuson 128, Mountain View, CA) with a 5-MHz linear-array transducer. Experienced physicians or sonographers performed each examination. The sensitivity setting for color Doppler imaging was routinely adjusted for each patient such that color flow was shown throughout the femoral artery but without bleeding’ of color into the adjacent soft tissues. The following criteria
were used to diagnose
a pseudoaneurysm:
demonstration
of “swirls”
of color in a hypoechoic collection outside the arterial lumen; presence of a track of color from the artery to the pseudoaneurysm, representing the “neck” of the pseudoaneurysm; and demonstration of a “to-and-fro” Doppler waveform in either the pseudoaneurysm or the
ET AL.
AJR:159, November 1992
neck. Transverse and longitudinal color flow images were obtained to show the full extent of the volume of flow in the lumen of the pseudoaneurysm.
The color flow, B-mode, and image-directed Doppler waveforms and the clinical records of all patients were reviewed retrospectively. In order
to determine
the relative
amounts
of thrombus
and pulsatile
flow within each pseudoaneurysm, the entire hypoechoic collection volume and the volume of flow in the lumen were calculated. blood
Measurements were obtained by using either electronic calipers placed on the static images at the time of the examination or handheld calipers (Fig. 1). The volume ofthe pseudoaneurysm (in milliliters)
was calculated length
by using the following
x 0.52
[1 5]. The percentage
rysm was determined in the lumen doaneurysm
formula:
of flow within
by using the following
height
x width
x
each pseudoaneu-
formula:
volume of flow
#{247} volume
of the pseudoaneurysm x 1 00%. If a pseuwas bibbed or trilobed, each lobe was measured sep-
arately and the values were added. Hand-held
calipers
were used to determine
the forward
and reverse
flow velocities of the Doppler spectra obtained from the neck of the pseudoaneurysm. The ratio of forward (systolic) to reverse (diastolic) velocity
and the percentage
of the diastolic
cycle that showed flow were estimated. pseudoaneurysm
Patients’
was measured
records
portion
of the cardiac
The length of the neck of the
by using
and catheterization
hand-held
reports
calipers.
were reviewed
to
determine the type of procedure used (diagnostic catheterization vs interventional catheterization), use of arterial sheaths, heparinization,
use of tissue plasminogen the group
who
activator,
and presence
of hematoma.
In
had spontaneous
thrombosis, the presence or absence of a bruit was described in only eight cases. The clinical and demographic parameters of each group were compared by using an independent groups t-test and exact version of the x2-test. The initial and final color Doppler studies that showed flow were analyzed by using the exact Wilcoxon test. Analyses were done by using Stat Xact (Cytel Softwear Corp., Cambridge, MA) and SAS (Version 6.04, SAS Institute, Cary, NC) software. One patient in the surgery group was excluded from all analyses because in this case we could not distinguish the margin of the pseudoaneurysm from a large hematoma.
Fig. i.-43-year-old man with congestive cardiomyopathy, 3 days after femoral artery puncture for cardiac catheterization. A, Color Doppler sonogram shows partially thrombosed pseudoaneurysm with rim of hypoechoic thrombus. Overall volume of pseudoaneurysm was determined by measuring outer margin of hypoechoic rim (arrows). volume of flow in lumen was determined by measuring outer margin of color Doppler signal (arrowheads). Femoral artery (A) lies deep to pseudoaneurysm. B, Color Doppler sonogram obtained 2 days after A shows spontaneous thrombosis (T) of pseudoaneurysm. Femoral artery (A) lies deep to pseudoaneurysm.
AJR:159,
November
MR
1992
A
OF
CLIVAL
1073
ABNORMALITIES
B
C
Fig. 16.-Infection of petrous apex (Gradenigo’s syndrome) and clivus. A, Contrast-enhanced axial TI-weighted MR Image shows epidural enhancing mass (arrow) in cerebellopontine angle. Bone involvement cannot be determined with contrast enhancement. Gradenigo’s syndrome classically consists of triad of sixth-nerve palsy, pain in trigeminal nerve distribution, and chronic otitis media In patients with petrous aplcftis. B and C, CT scans show epidural mass (white arrow) and bone erosion of petrous and clival bones (arrowheads). Note inflammatory mass in middle ear (black arrows).
Fig. 17.-Mucocele
involving
sphenoidal
sinus
and cllvus In a 60-year-old man. Contrast-enhanced midline sagittal MR Image shows mass of Intermediate signal markedly expanding sphenoidal sinus and clivus. Only mucopenosteum enhances (arrows). Classically, a mucocele develops from obstruction of a sinaI ostium.
Fig. 18.-Rheumatoid arthritis in a 64-year-old woman. A, Midline sagfttal Ti-weighted MR Image reveals pannus formation (open arrow) about dens disrupting normal infraclival fat (white arrow). Note markedly widened atiantodens Interspace of 11 mm measuring from cortex (black line representing compact cortical bone Is shown by solid black arrows). Normal atlantodens interspace in adults is 2.5 mm and in children It is up to 5 mm. Anteroposterior diameter of spinal canal is narrowed. B, Lateral tomogram shows atlantoaxial subluxation.
and T2-weighted images. They found the normal adult clival signal usually iso- or hyperintense relative to the pons on Ti weighted images and approximately isointense with the pons on T2-weighted images. Abnormalities were hypointense in relation to the pons on the Ti-weighted images and varied
Administration of gadopentetate dimeglumine is useful in delineating vascular anatomy and disease in the paraclival
from diffusely
disease, [3].
-
on T2-weighted
hypointense images.
to hyperintense
relative to the pons
region;
however,
enhance unenhanced
many
and become images
lesions
in the marrow
isointense are
more
with sensitive
and are more sensitive
normal
of the clivus will marrow. Thus,
for detecting
than T2-weighted
marrow
images
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
i 074
CHAUUB
ET
AL.
AJR:159, November 1992
Fig. 19.-Thrombosed aneurysm in vertebrobasilar artery along dorsal clivus in a 57-year-old woman. A, Midline sagittal Ti-weighted MR Image shows posterior displacement of medulla (curved arrow) (straight
by a large intermediate-signal mass arrow). Round shape and predominant
lack of flow void suggest thrombosed aneurysm. B, Selected lateral vertebral arteriogram shows posterior displacement of vertebral artery (solid arrow) and ectasia of basilar artery (open arrow), but no aneurysmal
Fig. 20.-jf-Thalassemla
Midline Ti-weighted ment
of normal
diploetic sue.
space
bright
In a 22-year-old
man.
MR image shows replacein clivus
and
of calvaria by hematopoletic
fafty
signal
tis-
Fig. 21.-FIbrous
dysplasla
In a 54-year-old
man. Sagittal TI-weighted MR image shows low signal In sphenoid bone and clivus (arrow). A CT scan showed osseous expansion and hazy sole-
rosis characteristic
of fibrous dysplasla.
filling.
Fig. 22.-Tornwaldt’s cyst in a 3i-year-old man. Midline sagfttalTi-weighted MR image displays a high-signal mass along midline of superior nasopharynx and ventral clivus (arrow). Track from notochordal remnants may give rise to this cyst, which can be complicated by infec-
tion.
REFERENCES 1 . Freedy AM, Miller KD Jr. Granulocytic sarcoma (chloroma): sphenoidal sinus and paraspinal involvement as evaluated by CT and MR. AJNR i99i;12:259-262 2. Kimura F, Kim KS, Friedman H, et al. MR imaging of the normal and abnormal clivus. AJNR 1990;11 :1015-1021 3. Stimac GK, Porter BA, Olson DO, et al. Gadolinium-DTPA-enhanced MR
imaging of spinal neoplasms: preliminary investigation and comparison with enhanced spin-echo and STIR sequences. AJNR 1988:9:839-846 SUGGESTED
READING
1 . Lame FJ, Nadel L, Braun IF. CT and MR imaging of the central skull base. Part 1 : Techniques, embryologic development and anatomy. RadioGraphics 1990:10:591-602
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
i 075
Case Report .. ‘.
Miliary Tuberculosis Gloria
T. Gee,1
Canlos
Bazan
Ill, and
Involving J. Randy
Jinkins
We report a case of miliary cerebral tuberculosis in a 14month-old child who began having seizures while on quadrupIe drug antituberculous therapy. MR imaging showed numenous round, very small enhancing lesions scattered throughout the brain, compatible with this diagnosis. Case
Report
A 14-month-old girl was referred for evaluation and control of seizures of several months’ duration. Tuberculous meningitis had been diagnosed 3 months earlier on the basis of clinical history and family exposure to tuberculosis. While the child was being treated with antituberculous oral medications, she began having seizures, and left-sided weakness developed suddenly. Physical examination revealed a temperature of 38.8#{176}C, slight weakness of the left leg, and clonus at the right ankle. The chest radiograph obtained when she was admitted showed a bilateral pattern of diffuse reticulonodular opacities suggestive of miliary pneumonia. Examination of CSF revealed an inflammatory pleocytosis and a low glucose content. A tuberculin skin test (PPD) was initially nonreactive, but a second PPD test 2 months later showed a 1 3 x 1 2 mm area of erythema after 24 hr. MR examination of the brain showed focal areas of hypointensity in the right temporopanietal and left occipital lobes on Ti-weighted images and increased signal on T2-weighted images (Figs. iA and 1 B). After IV administration of gadopentetate dimeglumine (0.1 mmol/ kg), we noted numerous round, very small tered throughout the brain (Fig. 1 C). Most proximately 1 mm in diameter.
enhancing of these
lesions scatmeasured ap-
Discussion tuberculosis
can be described
geal or parenchymal)
Cranial
and type (diffuse
Received 1
AJR
(menin-
tubencu-
loma, tubenculous abscess, or focal cerebnitis) [i ]. Associated changes also are commonly encountered, such as artenitis, infarction, hemorrhage, and hydrocephalus [1 2]. Tubencubus meningitis represents less than 5% of the cases of bacterial meningitis reported in children. Infection in children usually results from inhalation of organisms shed by adults with tuberculosis [1]. Entry of Mycobacterium tuberculosis into the CNS is believed to be by hematogenous spread from the lung to the meninges, where small tuberculomas are formed. These may rupture into the subanachnoid space and ,
lead to meningitis, the severity of which is partly related each patient’s local tissue reaction to the organism. Sixty-eight are reported
percent to have
nadiognaphs,
and
to
of children with tubenculous meningitis concomitant miliany tuberculosis of the brain. Conversely, in only 1 % of children with miliany cerebral tuberculosis does tubenculous meningitis develop subsequently [1]. Few patients with miliary tuberculosis have fever at presentation. Up to 50% have abnormal findings on chest only
25% have a positive
tuberculin
skin
test [3]. Fever, an elevated erythrocyte sedimentation rate, and evidence of extracranial tubenculous infection have been reported in 32-50% of patients with cerebral tubenculomas [2,4]. In cases of tuberculous meningitis, the basal cisterns are isodense to hyperdense compared with normal CSF on unenhanced CT scans. After IV administration of iodinated contrast medium, intense enhancement of the affected cisterns was noted. Communicating hydnocephalus is common and has been reported to be a sensitive indicator of tubencubus meningitis in children [5]. lschemic infancts are most common in the distribution of
18, 1 992; accepted after revision May 15, 1992. Neuroradiology Section, The University of Texas Health Science Center, 7703 Floyd Curl Dr. , San Antonio, TX 78284-7800.
March
All authors:
requests
by location
meningitis,
the Brain: MR Findings
to J. A. Jinkins.
159:1075-1076,
November
1992 0361-803X/92/1
595-1075
© American Roentgen Ray Society
Address reprint
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1076
GEE
A
ET AL.
AJR:159,
B
November
C
Fig. 1.-A, Ti-weighted (600/20) axial spin-echo MR Image shows hypointense areas in right temporoparietal lobe and left occipital lobe. B, T2-welghted (2500/20) MR image shows hyperintense areas in right temporoparietal lobe, left occipital lobe, and left side of thalamus. C, TI-weighted (600/20) axIal spin-echo MR image after IV contrast administration shows numerous round, very small enhancing lesions brain, with confluent areas appearing in right temporoparietal and left occipital areas.
the middle
cerebral
artery,
including
the lenticulostniate
arten-
ies [6, 7], and are associated with tubenculous arteritis caused by the basal meningitis [8]. MR imaging has been reported to be more sensitive than CT in detecting disease affecting the basal cisterns [8]. Contrast-enhanced MR is particularly useful for detecting small foci of meningeal inflammation [8]. MR is reported to be more sensitive than CT in detecting tuberculomas of the cerebral parenchyma. Tuberculomas are reported to be isointense with gray matter on Ti -weighted MR images. On T2-weighted images, lesions show central hypenmntensity. In some cases, a hypointense ring is apparent within the wall of the tuberculoma on T2-weighted images.
Most
tuberculomas
are further
by a collar
of high
signal resulting from edema on T2-weighted images. culomas typically enhance after the IV administration
Tuberof ga-
dopentetate dimeglumine The MR examination in areas of enhancement in left occipital lobe, which tubenculomas,
weighted
concomitant
images,
multiple
outlined
in a solid or ring pattern. the current case showed confluent the right temporoparietal lobe and may have represented coalescent ischemic disease, or both. On T2small
hyperintense
1992
foci were
scat-
tered throughout the brain parenchyma, all ofwhich enhanced with IV administration of contrast medium, consistent with
throughout
the diagnosis of miliary cerebral tuberculosis. The homogeneous enhancement pattern suggests that the lesions are granulomas rather than abscesses, which would be expected to have a central
fluid collection.
REFERENCES 1 . Jinkins JR. Computed
tomography of intracranial tuberculoma. Neurora126-1 35 2. Draouat S, Abdenabi B. Ghanem M, Bourjat P. Computed tomography of cerebral tuberculoma. J Comput Assist Tomogr 1987;i 1 : 594-597 3. Witham RR, Johnson RH, Roberts DL. Diagnosis of miliary tuberculosis by cerebral computerized tomography. Arch Intern Med 1979;139: 479-480 4. de Castro CC, Hesselink JR. Tuberculosis. Neuroimaging C/in North Am diology 1991:33:
1991;1
:119-135
5. Waecker NJ, Connor JD. Central nervous system tuberculosis in children: a review of 30 cases. Pediatr Infect Dis J i990;9:539-543 6. Menkes JH. Infections of the nervous system: Chronic and granulomatous infections of the meninges. In: Textbook of child neurology. Philadelphia: Lea & Febiger, 1985:335-414 7. Leisuarda A, Berthier M, Starkstein S. Ischemic infarction in 25 children with tuberculous meningitis. Stroke i988;19:200-204 8. Kee H-C, Moon HH, Jae KR, et al. Gd-DTPA-enhanced MR imaging of the brain in patients with meningitis. AJR 1990:154:809-816
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1077
Artery
Femoral
Pseudoaneurysms:
Value of Color Doppler Predicting Which Ones Without Treatment
H Enik K. Paulson1 Barbara S. Hertzberg Susan S. Paine Barbara A. Carroll
.
OBJECTIVE. The purpose of this study teristics of femoral artery pseudoaneurysms
determine would
if color Doppler sonography
ultimately
MATERIALS
studied spectral
thrombose
can
be used
thrombosis
of the
which pseudoaneurysms
to predict
spontaneously.
METHODS.
AND
was to evaluate the Doppler Imaging characoccurring after cardiac catheterization to
Over a 30-month-period,
with serial color Doppler sonography. images and clinical records of patients
spontaneous repair
Sonography in Will Thrombose
were compared
with
24
pseudoaneurysms
were
The color flow, B-mode, and Doppler whose pseudoaneurysms demonstrated those of patients who required surgical
pseudoaneurysm.
RESULTS.
We
found
no statistically
significant
differences
between
the
patients
whose pseudoaneurysms demonstrated spontaneous thrombosis and those treated surgically in regard to clinical parameters or the volume of the pseudoaneurysm, of flow within the pseudoaneurysm, ratio of forward to reversed flow velocity in the pseudoaneurysm neck, duration of diastolic flow in the pseudoaneurysm neck, or length of pseudoaneurysm neck. However, on color Doppler images, the volume of flow in the lumens of pseudoaneurysms that thrombosed spontaneously (1.8 ± 3.3 ml) was significantly smaller than the volume of flow in the lumens of those treated surgically (4A ± 3.2, ml, p = .02). CONCLUSION. We conclude that pseudoaneurysms with small volumes of flow in the lumen are more likely to thrombose than are those with large volumes of flow in the Iumen However, color Doppler sonographic characteristics cannot be used to predict percentage
subsequent AJR
thrombosis.
159:i077-i081,
Femoral
November
1992
are a known complication of cardiac catheterization for vascular access [i -3]. With increased use of largediameter arterial sheaths in interventional cardiac procedures, these pseudoaneurysms are being encountered more frequently [i 4-6]. Traditional teaching suggests that pseudoaneurysms should be repaired surgically to avoid rupture [4, 79]. Although surgery is effective and safe, it is expensive and is associated with prolonged hospitalization. Recent reports raise questions about the standard surgical approach. First, many pseudoaneurysms thrombose spontaneously, suggesting that close observation and serial imaging may be indicated in some patients [5, 6, 1 0-i 3]. Second, nonsurgical sonographically guided compression-occlusion therapy has been suggested as a viable alternative to surgical management [14]. The purpose of this study was to evaluate the imaging characteristics of in which
pseudoaneurysms
the groin
is used
,
Received February 1 1 , 1992; accepted after revision May 22, 1992.
Presented at the annual meeting of the American Roentgen Ray Society, Orlando, FL, May 1992. ‘ All authors: Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 2771 0. Address reprint requests to E. K. Paulson. 0361-803X/92/1 © American
595-1077
Roentgen
Ray Society
pseudoaneurysms managed conservatively to determine phy can be used to predict which pseudoaneurysms
if color Doppler will thrombose
sonograwithout
treatment. Materials
and
Methods
A computer-based search of sonographic studies conducted over a 30-month period 1 , 1989, to July 1 , 1991) indicated that 430 patients had had color Doppler
(January
PAULSON
Downloaded from www.ajronline.org by NYU Langone Med Ctr-Sch of Med on 07/21/15 from IP address 128.122.253.212. Copyright ARRS. For personal use only; all rights reserved
1078
sonography of the groin. Most of these examinations were performed after cardiac catheterization in order to rule out a pseudoaneurysm or arteriovenous fistula. Seventy-nine pseudoaneurysms were identified in 78 patients; their color Doppler sonographic reports were reviewed. Fifty-five of the pseudoaneurysms were imaged only once by using color Doppler sonography and were excluded from the study group. The remaining 24 were examined at least twice with color Doppler sonography; they form the basis of this report. Twenty-three patients were included in
this study. One patient had two cardiac catheterizations
21 months apart; each procedure was complicated aneurysm
that
was
examined
sequentially.
rysms were imaged twice (original
Fourteen
examination
performed by a pseudopseudoaneu-
and one follow-up
examination) and 1 0 were imaged three times (original examination and two follow-up examinations). Thus, a total of 58 color Doppler sonograms were reviewed. Patients with pseudoaneurysms were divided into two groups on the basis of clinical outcome. One group included patients with pseudoaneurysms in whom spontaneous thrombosis occurred, without surgical or nonsurgical intervention. The other group included patients who had color Doppler sonography but had surgery before thrombosis was detected. All were followed up as inpatients except for four patients who had spontaneous thrombosis who were dis-
charged with a known pseudoaneurysm with a scheduled follow-up study. All patients had cardiac catheterization except for one in the spontaneous
All color
thrombosis group who had iliac angioplasty. Doppler images were obtained by using commercially
available real-time equipment (Acuson 128, Mountain View, CA) with a 5-MHz linear-array transducer. Experienced physicians or sonographers performed each examination. The sensitivity setting for color Doppler imaging was routinely adjusted for each patient such that color flow was shown throughout the femoral artery but without bleeding’ of color into the adjacent soft tissues. The following criteria
were used to diagnose
a pseudoaneurysm:
demonstration
of “swirls”
of color in a hypoechoic collection outside the arterial lumen; presence of a track of color from the artery to the pseudoaneurysm, representing the “neck” of the pseudoaneurysm; and demonstration of a “to-and-fro” Doppler waveform in either the pseudoaneurysm or the
ET AL.
AJR:159, November 1992
neck. Transverse and longitudinal color flow images were obtained to show the full extent of the volume of flow in the lumen of the pseudoaneurysm.
The color flow, B-mode, and image-directed Doppler waveforms and the clinical records of all patients were reviewed retrospectively. In order
to determine
the relative
amounts
of thrombus
and pulsatile
flow within each pseudoaneurysm, the entire hypoechoic collection volume and the volume of flow in the lumen were calculated. blood
Measurements were obtained by using either electronic calipers placed on the static images at the time of the examination or handheld calipers (Fig. 1). The volume ofthe pseudoaneurysm (in milliliters)
was calculated length
by using the following
x 0.52
[1 5]. The percentage
rysm was determined in the lumen doaneurysm
formula:
of flow within
by using the following
height
x width
x
each pseudoaneu-
formula:
volume of flow
#{247} volume
of the pseudoaneurysm x 1 00%. If a pseuwas bibbed or trilobed, each lobe was measured sep-
arately and the values were added. Hand-held
calipers
were used to determine
the forward
and reverse
flow velocities of the Doppler spectra obtained from the neck of the pseudoaneurysm. The ratio of forward (systolic) to reverse (diastolic) velocity
and the percentage
of the diastolic
cycle that showed flow were estimated. pseudoaneurysm
Patients’
was measured
records
portion
of the cardiac
The length of the neck of the
by using
and catheterization
hand-held
reports
calipers.
were reviewed
to
determine the type of procedure used (diagnostic catheterization vs interventional catheterization), use of arterial sheaths, heparinization,
use of tissue plasminogen the group
who
activator,
and presence
of hematoma.
In
had spontaneous
thrombosis, the presence or absence of a bruit was described in only eight cases. The clinical and demographic parameters of each group were compared by using an independent groups t-test and exact version of the x2-test. The initial and final color Doppler studies that showed flow were analyzed by using the exact Wilcoxon test. Analyses were done by using Stat Xact (Cytel Softwear Corp., Cambridge, MA) and SAS (Version 6.04, SAS Institute, Cary, NC) software. One patient in the surgery group was excluded from all analyses because in this case we could not distinguish the margin of the pseudoaneurysm from a large hematoma.
Fig. i.-43-year-old man with congestive cardiomyopathy, 3 days after femoral artery puncture for cardiac catheterization. A, Color Doppler sonogram shows partially thrombosed pseudoaneurysm with rim of hypoechoic thrombus. Overall volume of pseudoaneurysm was determined by measuring outer margin of hypoechoic rim (arrows). volume of flow in lumen was determined by measuring outer margin of color Doppler signal (arrowheads). Femoral artery (A) lies deep to pseudoaneurysm. B, Color Doppler sonogram obtained 2 days after A shows spontaneous thrombosis (T) of pseudoaneurysm. Femoral artery (A) lies deep to pseudoaneurysm.
