DOI:10.2214/AJR.06.0109
AJR 2007; 188:866-873
© American Roentgen Ray Society
Brain Abnormalities Detected on Whole-Body 18F-FDG PET in Cancer Patients: Spectrum of Findings
Erin Stubbs1,
Jonathan Kraas1,
Kathryn A. Morton2 and
Paige B. Clark1
1 Department of Radiology, Wake Forest University School of Medicine, Medical
Center Blvd., Winston-Salem, NC 27157.
2 Department of Radiology, University of Utah Health Sciences Center, Salt Lake
City, UT 84132.
Received January 20, 2006;
accepted after revision May 30, 2006.
Address correspondence to P. B. Clark.
Abstract
OBJECTIVE. The purpose of this article is to discuss and show
examples of the PET appearance of common brain abnormalities that radiologists
encounter when interpreting whole-body 18F-FDG PET examinations of
cancer patients.
CONCLUSION. Knowledge of the PET appearance of various brain
abnormalities can yield diagnostically relevant information in cancer
patients. Detection of brain abnormalities on whole-body PET often requires
adjusting window settings to reduce the intensity of normal brain FDG
activity. Often, close correlation of PET/CT and MRI with clinical history
offers the most complete radiologic diagnosis.
Keywords: arachnoid cyst • brain cancer • cerebrovascular accident • FDG PET • oncologic imaging • stroke • whole-body imaging
Introduction
Oncologic imaging with 18F-FDG PET is a valuable clinical
tool for staging malignancies, including lymphoma; melanoma; and head and
neck, lung, esophageal, colorectal, and breast cancers. In addition to
providing initial staging that is more accurate than anatomic imaging alone,
new data show that FDG PET can be used to effectively monitor therapeutic
response in patients with lymphoma, lung cancer, and esophageal cancer
[1-3].
The now common use of PET/CT in academic and private centers allows
radiologists to provide superior accuracy in staging and follow-up of
malignancy when compared with anatomic imaging alone, such as CT or MRI, or
with side-by-side correlation of PET and CT
[2].
Whole-body PET scans at our institution include images from the top of the
skull through the thighs that are obtained 45 minutes after IV injection of up
to 740 MBq (20 mCi) of FDG. Each bed position is imaged for 7 minutes in the
2D mode. The compilation of cases presented in this article includes a
spectrum of incidentally detected brain lesions in patients undergoing PET for
oncologic staging. Institutional review board approval for this study was
obtained.
Attenuation correction was performed using unenhanced CT or transmission
scans, depending on the PET unit used at the time. Because of improved
anatomic correlation with PET/CT, this technique has been in use in our
institution since 2005. During review, manually adjusting the window settings
to decrease normal FDG activity in the brain (as compared with the settings
used for whole-body images) was performed. This was followed by qualitative
analysis of PET abnormalities by two experienced reviewers.
Brain Metastases
In all patients with malignancy, 10-35% will develop brain metastases
during the disease course [4,
5]. Metastases usually reach
the brain hematogenously and often localize to the corticomedullary junction.
Most metastases are supratentorial (80%), with the remaining metastases
occurring in the cerebellum (18%) and brainstem (2%)
[6]. Solitary lesions account
for up to 30% of brain metastases.
Clinically, brain metastases can be asymptomatic or affected patients can
present with focal neurologic deficits, headache, nausea or vomiting, or
seizure. Cerebellar metastases can also cause ataxia
[6,
7]. Asymptomatic brain
metastases are most common in those with melanoma and in lung cancer patients
[8].
The efficacy of FDG PET in depicting cerebral metastases is controversial.
The sensitivity of FDG PET in revealing cerebral metastases in patients with
malignancy has been reported at 68-82% when compared with anatomic imaging
[9]. This large range of
sensitivity likely reflects two factors: first, the high FDG activity in the
brain; and, second, the inability of PET scanners to show lesions smaller than
6-10 mm.
One focus or several foci of increased FDG activity as compared with normal
brain glucose metabolism in a cancer patient is highly suspicious for
metastases [10]. Because FDG
accumulates in normal cerebral cortex, cerebellum, and basal ganglia, manually
adjusting the window settings used for the whole-body scan by decreasing the
intensity of the brain image is often necessary to detect brain metastases
(Figs. 1A,
1B,
2A,
2B,
3A,
3B,
3C,
4A,
4B,
5A,
5B,
5C). Subcentimeter lesions
that can be detected on anatomic imaging, such as contrast-enhanced CT, MRI,
or both, are often missed on PET because of its 6- to 8-mm limit of resolution
(Fig. 6A,
6B). A mass lesion with
enhancement and surrounding edema on contrast-enhanced CT, MRI, or both can
often confirm the radiologic diagnosis
[8]. Conversely, in patients
with known brain metastases, PET may be able to localize the primary
malignancy (Fig. 7A,
7B,
7C). Although FDG PET does
reveal unsuspected brain abnormalities, as shown in this pictorial essay,
anatomic imaging will likely remain the gold standard for ruling out cerebral
metastases in cancer patients.
View larger version (75K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —70-year-old woman with non-small cell lung cancer and
extensive thoracic and abdominal metastases. Anterior
maximal-intensity-projection (MIP) PET image with window settings optimized to
show abnormalities in chest, abdomen, and pelvis shows numerous foci of
increased 18F-FDG activity in thorax and abdomen, which is
consistent with diffuse metastatic lung cancer. FDG activity in brain appears
within normal limits.
|
|
View larger version (62K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —70-year-old woman with non-small cell lung cancer and
extensive thoracic and abdominal metastases. Anterior MIP PET image after
window settings were adjusted to optimize visualization of abnormalities in
brain reveals small focus of increased FDG activity in right temporal lobe
(arrow), which is consistent with brain metastasis.
|
|
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —55-year-old man with non-small cell lung cancer. Sagittal PET
image shows foci of increased 18F-FDG activity in mediastinal lymph
nodes (arrow). Heterogeneous FDG activity in brain
(arrowhead) is suspicious for metastatic disease.
|
|
View larger version (67K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —55-year-old man with non-small cell lung cancer. Sagittal PET
image after window settings were adjusted shows several foci of abnormal
18F-FDG activity in brain. Two lesions show central photopenia
(arrows), which is consistent with necrosis. Biopsy confirmed
non-small cell lung cancer and necrosis.
|
|
View larger version (55K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —48-year-old man with non-small cell lung cancer. Coronal PET
image shows small focus of increased 18F-FDG activity in left upper
lobe (arrow), which is consistent with patient's known lung cancer.
FDG activity in brain appears within normal limits.
|
|
View larger version (52K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —48-year-old man with non-small cell lung cancer. Coronal PET
image after window settings were adjusted shows small focus of increased FDG
activity in left cerebellum (arrow) that is suspicious for new
metastatic disease.
|
|
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —48-year-old man with non-small cell lung cancer. Axial
T1-weighted MR image obtained with contrast material shows focus of
enhancement (arrow) corresponding to that shown on PET (B) and
confirms diagnosis of cerebellar metastasis.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —59-year-old woman with metastatic breast carcinoma who presented for
follow-up examination after right mastectomy, chemotherapy, and radiation
therapy. Posterior maximal-intensity-projection (MIP) PET image shows liver
metastasis (arrow) and multiple bone metastatic lesions
(arrowheads). Note 18F-FDG activity in brain appears
within normal limits.
|
|
View larger version (70K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —59-year-old woman with metastatic breast carcinoma who presented for
follow-up examination after right mastectomy, chemotherapy, and radiation
therapy. Posterior MIP PET image after window settings were adjusted to
optimize visualization of brain abnormalities shows small focus of increased
FDG activity in right cerebellum (arrow), which is consistent with
new metastatic disease.
|
|
View larger version (63K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —55-year-old woman with diffuse large B-cell lymphoma. Axial
PET image shows brainstem focus (arrow) as seen in A in
addition to focus of increased activity in left temporal lobe
(arrowhead).
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C —55-year-old woman with diffuse large B-cell lymphoma. Axial
spin-echo MR image obtained with contrast material shows enhancement in
brainstem (arrow) and left temporal lobe (arrowhead). Biopsy
confirmed large B-cell lymphoma.
|
|
View larger version (71K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B —79-year-old man with melanoma. Coronal PET image depicts
normal 18F-FDG activity likely because this lesion (arrow)
is at limits of PET resolution. This lesion was treated with gamma knife
radiation therapy.
|
|
View larger version (163K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A —70-year-old man with poorly differentiated metastatic brain
neoplasm of unknown primary cancer. Coronal T1-weighted MR image shows
enhancing lesion in left inferior parietal lobe (arrow).
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B —70-year-old man with poorly differentiated metastatic brain
neoplasm of unknown primary cancer. Coronal PET image shows focus of increased
18F-FDG activity in left inferior parietal lobe
(arrow).
|
|
View larger version (86K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7C —70-year-old man with poorly differentiated metastatic brain
neoplasm of unknown primary cancer. Coronal PET image of thorax shows small
focus of increased FDG activity in right upper lobe (arrow). Biopsy
confirmed bronchogenic carcinoma.
|
|
The specificity of FDG PET in evaluating cerebral abnormalities is less
clear than its sensitivity. In one study published in 1996 of 402 lung cancer
patients, researchers reported a 38% specificity of FDG PET alone when
compared with anatomic imaging
[11]. However, today, FDG PET
is rarely evaluated without CT. Likely the low specificity reported in that
study [10] would be higher in
the era of FDG PET/CT, which can often depict cerebral infarction, anatomic
variants, and atrophy, all of which can alter the appearance of normal brain
on FDG PET [9].
Postoperative and Postradiation Changes
In patients with known brain metastases, treatment options include surgery,
whole-brain irradiation, stereotactic radiosurgery, chemotherapy, and hormonal
therapy. Patients with single metastatic lesions are most commonly treated
with surgery followed by whole-brain radiation, which improves prognosis. If
these lesions are not accessible by conventional surgical means, stereotactic
radiosurgery is used. In patients with multiple metastases, single lesions can
be resected for a tissue diagnosis or to palliate neurologic symptoms
[4].
Photopenia (i.e., decreased FDG activity) in the areas of successful
surgical and stereotactic radiation therapy of metastatic lesions in the brain
can be evident on PET (Figs.
8A,
8B and
9A,
9B,
9C). Areas where surgery was
performed for benign brain abnormalities, such as temporal lobectomy for
epilepsy, would also have this appearance. Clinical history, CT, and MRI often
confirm PET findings.
View larger version (79K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A —49-year-old woman with non-small cell lung cancer and brain
metastases who presented for follow-up examination after craniotomy and
resection of left parietal lobe lesion. Axial PET image shows photopenia
(arrow) in left parietal lobe with normal 18F-FDG activity
in surrounding gray matter.
|
|
View larger version (165K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B —49-year-old woman with non-small cell lung cancer and brain
metastases who presented for follow-up examination after craniotomy and
resection of left parietal lobe lesion. Axial spin-echo MR image shows
postsurgical changes (arrow) after left parietal craniotomy.
|
|
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A —65-year-old man with melanoma who presented for follow-up
examination after craniotomy and resection of left frontoparietal metastatic
lesion. Sagittal (A) and coronal (B) PET images show photopenia
(arrows) in left frontoparietal region with normal 18F-FDG
activity in surrounding gray matter. C, Coronal T1-weighted MR image
obtained with contrast material shows postcraniotomy changes
(arrow).
|
|
View larger version (89K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B —65-year-old man with melanoma who presented for follow-up
examination after craniotomy and resection of left frontoparietal metastatic
lesion. Sagittal (A) and coronal (B) PET images show photopenia
(arrows) in left frontoparietal region with normal 18F-FDG
activity in surrounding gray matter.
|
|
View larger version (166K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9C —65-year-old man with melanoma who presented for follow-up
examination after craniotomy and resection of left frontoparietal metastatic
lesion. Coronal T1-weighted MR image obtained with contrast material shows
postcraniotomy changes (arrow).
|
|
Arachnoid Cyst
Arachnoid cysts are congenital abnormalities that encompass 1% of all
intracranial lesions. They are most commonly found in the middle cranial fossa
(50%) and the posterior cranial fossa (25-30%), followed by the anterior
cranial fossa, suprasellar subarachnoid cistern, and quadrigeminal plate
cistern [8,
12]. Many are clinically
silent. However, arachnoid cysts in the posterior fossa are more likely to be
symptomatic, causing headache, dizziness, ataxia, tinnitus, or nausea
[13]. Treatment, when
necessary, includes resection, shunting of CSF, marsupialization of the cyst,
or a combination of these approaches
[12].
FDG PET can reveal photopenia corresponding to an arachnoid cyst, with the
adjacent compressed brain parenchyma showing relatively normal FDG activity
(Fig. 10A,
10B,
10C,
10D). CT and MRI findings show
an arachnoid cyst as a CSF-attenuation fluid-filled cyst without peripheral
contrast enhancement [13].
View larger version (53K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A —77-year-old man with metastatic non-small cell lung cancer
and primary renal cell carcinoma. Axial (A) and sagittal (B) PET
images show displacement of cerebellar hemispheres (arrowheads) by
midline photopenic defect (arrows).
|
|
View larger version (80K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B —77-year-old man with metastatic non-small cell lung cancer
and primary renal cell carcinoma. Axial (A) and sagittal (B) PET
images show displacement of cerebellar hemispheres (arrowheads) by
midline photopenic defect (arrows).
|
|
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10C —77-year-old man with metastatic non-small cell lung cancer
and primary renal cell carcinoma. Unenhanced axial (C) and sagittal
(D) CT images show collection of CSF-attenuation fluid with
nonenhancing borders in posterior fossa (arrows) that is consistent
with retrocerebellar arachnoid cyst.
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10D —77-year-old man with metastatic non-small cell lung cancer
and primary renal cell carcinoma. Unenhanced axial (C) and sagittal
(D) CT images show collection of CSF-attenuation fluid with
nonenhancing borders in posterior fossa (arrows) that is consistent
with retrocerebellar arachnoid cyst.
|
|
Hydrocephalus and Ex Vacuo Ventriculomegaly
Hydrocephalus results from an imbalance of production and resorption of CSF
in the ventricular system
[14]. This imbalance can be
congenital or acquired secondary to bacterial or viral CNS infection,
hemorrhage, or tumor. The prevalence of CSF shunts in the United States, which
are often used to treat hydrocephalus, is estimated at more than 125,000
[15]. Ex vacuo
ventriculomegaly due to parenchymal brain atrophy—commonly associated
with aging—can also be seen.
On PET, the prominence of CSF spaces is indicated by large areas of
photopenia surrounded by normal FDG activity in the surrounding gray matter
(Fig. 11A,
11B). On CT and MRI,
hydrocephalus is characterized by ventricular enlargement out of proportion to
the cerebral sulci. However, ex vacuo ventriculomegaly is characterized by
enlargement of both the sulci and the ventricles
[14].
Hemorrhagic Tumor
Metastatic brain tumors are more likely to hemorrhage than primary brain
tumors. The mechanisms of hemorrhage into a metastatic lesion include tumor
necrosis, rupture of nascent blood vessels, and tumoral invasion of
parenchymal vessels. Solid tumor metastases that cause brain hemorrhage
include melanoma, papillary thyroid cancer, and lung cancer. The most common
hemorrhagic primary brain tumor is malignant glioma
[16].
Often present in multiple metastases, hemorrhage leads to acute symptoms
such as headache, seizure, and obtundation. Focal neurologic deficits may also
be present. If intratumoral hemorrhage is clinically life threatening,
resection may be necessary. In cases in which there is no known primary
cancer, resection of a hematoma may lead to a diagnosis.
PET of an acute hemorrhagic mass shows focally increased FDG activity (Fig.
12A,
12B). This increased activity
may be due to radiopharmaceutical extravasation or to subacute inflammation.
CT and MR images may show contrast extravasation, heterogeneity of an
enhancing mass, and edema
[16].
View larger version (156K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12B —64-year-old man with non-small cell lung cancer. Axial T1-weighted
MR image obtained with contrast material shows corresponding hemorrhagic mass
(arrow), which is consistent with metastasis.
|
|
Cerebrovascular Accidents
Cerebrovascular accidents (CVAs) are the third leading cause of death in
the United States and cause significant disability. Eighty percent of strokes
are ischemic and 20% are hemorrhagic in origin
[17,
18]. In one study, 14% of
patients with cancer were found to have cerebrovascular infarction or
hemorrhage at autopsy. Common causes of CVA in cancer patients include
coagulopathy, metastases, therapeutic complications, and infection
[16]. Clinical symptoms
include headache, hemiparesis, dysarthria, seizure, and obtundation.
Encompassing 2-3% of all strokes, cerebellar strokes may be asymptomatic or
may present with ataxia, dizziness, vertigo, and nausea or vomiting
[19,
20].
A remote CVA on PET shows photopenia in the gliotic scar and
encephalomalacia. An acute CVA may show increased FDG activity due to
radiopharmaceutical extravasation or inflammation in a cerebrovascular
territory, although we have not seen a case of this on FDG PET. Mass effect
could also be evident (Figs.
13A,
13B and
14A,
14B). Depending on the acuity
of the CVA, CT and MRI may show vessel changes, edema, mass effect,
encephalomalacia, or gliosis
[17].
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13B —67-year-old man with non-small cell lung cancer. Axial
T1-weighted MR image shows corresponding area of encephalomalacia
(arrow) that is consistent with prior cerebrovascular infarction.
|
|
View larger version (97K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14B —67-year-old woman with non-small cell lung cancer. Obtained 1
year after A, axial PET image shows focus of photopenia
(arrow) in mid right cerebellar hemisphere that is consistent with
previously undiagnosed cerebrovascular infarction. Previously diagnosed remote
cerebrovascular infarction (arrowhead) is also evident in left
parietal lobe.
|
|
Conclusion
Knowledge of the PET appearance of various brain abnormalities can yield
diagnostically relevant information in cancer patients. Detection of brain
abnormalities on whole-body PET often requires adjusting the window settings
to reduce the intensity of normal brain 18F-FDG activity. Often,
close correlation of PET/CT and MRI with clinical history offers the most
complete radiologic diagnosis.
References
- Levine EA, Farmer MR, Clark P, et al. Predictive value of
18-fluoro-deoxy-glucose-positron emission tomography (18F-FDG PET)
in the identification of responders to chemoradiation therapy for the
treatment of locally advanced esophageal cancer. Ann
Surg 2006; 243:472
-478[CrossRef][Medline]
- Kumar R, Nadig MR, Chauhan A. PET: clinical applications in
oncology. Part 1. Expert Rev Anticancer Ther2005; 5:1079
-1094[CrossRef][Medline]
- Meany HJ, Gidvani VK, Minniti CP. Utility of PET scans to predict
disease relapse in pediatric patients with Hodgkin lymphoma.
Pediatr Blood Cancer 2006;Mar
2 [Epub ahead of print]
- Norden AD, Wen PY, Kesari S. Brain metastases. Curr Opin
Neurol 2005; 18:654
-661[Medline]
- Takeshima H, Kuratsu J, Nishi T, Ushio Y. Prognostic factors in
patients who survived more than 10 years after undergoing surgery for
metastatic brain tumors: report of five cases and review of the literature.
Surg Neurol 2002;58
: 118-123[CrossRef][Medline]
- Weisberg LA. Solitary cerebellar metastases: clinical and CT
correlations. Arch Neurol 1985;42
: 336-341[Abstract/Free Full Text]
- Fadul C, Misulis KE, Wiley RG. Cerebellar metastases: diagnostic
and management considerations. J Clin Oncol1987; 5:1107
-1115[Abstract/Free Full Text]
- Koeller KK. Central nervous system neoplasms. In: Brant WE, Helms
CA, eds. Fundamentals of diagnostic radiology, 2nd ed.
Baltimore, MD: Lippincott Williams & Wilkins, 1998:113
-142
- Ludwig V, Komori T, Kolb D, Martin WH, Sandler MP, Delbeke D.
Cerebral lesions incidentally detected on
2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography images
of patients evaluated for body malignancies. Mol Imaging
Biol 2002; 4:359
-362[CrossRef][Medline]
- Larcos G, Maisey MN. FDG PET screening for cerebral metastases in
patients with suspected malignancy. Nuc Med Comm1996; 17:197
-198
- Palm I, Hellwig D, Leutz M, et al. Brain metastases of lung cancer:
diagnostic accuracy of positron emission tomography with fluorodeoxyglucose
(FDG-PET) [in German]. Med Klin (Munich)1999; 94:224
-227[Medline]
- Samii M, Carvalho GA, Schuhmann MU, Matthies C. Arachnoid cysts of
the posterior fossa. Surg Neurol 1999;51
: 376-382[CrossRef][Medline]
- Altman NR, Naidich TP, Braffman BH. Posterior fossa malformations.
AJNR 1992; 13:691
-724[Medline]
- Barakos JA. White matter and neurodegenerative diseases. In: Brant
WE, Helms CA, eds. Fundamentals of diagnostic radiology, 2nd
ed. Baltimore, MD: Lippincott Williams & Wilkins,1998
: 169-186
- Bondurant CP, Jimenez DF. Epidemiology of cerebrospinal fluid
shunting. Pediatr Neurosurg 1995;23
: 254-259[Medline]
- Rogers LR. Cerebrovascular complications in cancer patients.
Neurol Clin 2003;21
: 167-192[CrossRef][Medline]
- Rowley HA. Cerebrovascular disease. In: Brant WE, Helms CA, eds.Fundamentals of diagnostic radiology, 2nd ed.
Baltimore, MD: Lippincott Williams & Wilkins, 1998:79
-112
- Mountz JM, Liu HG, Deutsch G. Neuroimaging in cerebrovascular
disorders: measurement of cerebral physiology after stroke and assessment of
stroke recovery. Semin Nucl Med 2003;33
: 56-76[CrossRef][Medline]
- Kelly PJ, Stein J, Shafqat S, et al. Functional recovery after
rehabilitation for cerebellar stroke. Stroke2001; 32:530
-534[Abstract/Free Full Text]
- Macdonell RA, Kalnins RM, Donnan GA. Cerebellar infarction: natural
history, prognosis, and pathology. Stroke1987; 18:849
-855[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
