Pictorial Essay
Nuclear Medicine
March 2007

Brain Abnormalities Detected on Whole-Body 18F-FDG PET in Cancer Patients: Spectrum of Findings

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.

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.
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.

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].

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].

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].

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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
Fig. 5A —55-year-old woman with diffuse large B-cell lymphoma. Sagittal PET image shows focus of increased 18F-FDG activity in brainstem (arrow).
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).
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.
Fig. 6A 79-year-old man with melanoma. Coronal T1-weighted MR image shows subcentimeter focus of enhancement (arrow), which is consistent with metastasis.
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.
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).
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).
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.
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.
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.
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).
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.
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).
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).
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).
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.
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.
Fig. 11A 48-year-old man with metastatic melanoma. Coronal PET image shows enlarged photopenic ventricles (arrow) surrounded by normal 18F-FDG activity in compressed gray matter (arrowhead).
Fig. 11B 48-year-old man with metastatic melanoma. Coronal FLAIR MR image shows ventriculomegaly (arrow), which is consistent with patient's history of communicating hydrocephalus.
Fig. 12A —64-year-old man with non-small cell lung cancer. Axial PET image shows mildly asymmetric increased 18F-FDG activity in left parietooccipital region (arrow).
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.
Fig. 13A 67-year-old man with non-small cell lung cancer. Axial PET image shows large focus of decreased 18F-FDG activity in left frontal lobe (arrow).
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.
Fig. 14A 67-year-old woman with non-small cell lung cancer. Axial PET image shows 18F-FDG activity in cerebellum (arrow) that is within normal limits.
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.

Footnote

Address correspondence to P. B. Clark.

References

1.
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
2.
Kumar R, Nadig MR, Chauhan A. PET: clinical applications in oncology. Part 1. Expert Rev Anticancer Ther 2005; 5:1079-1094
3.
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]
4.
Norden AD, Wen PY, Kesari S. Brain metastases. Curr Opin Neurol 2005; 18:654-661
5.
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
6.
Weisberg LA. Solitary cerebellar metastases: clinical and CT correlations. Arch Neurol 1985; 42:336-341
7.
Fadul C, Misulis KE, Wiley RG. Cerebellar metastases: diagnostic and management considerations. J Clin Oncol 1987; 5:1107-1115
8.
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
9.
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
10.
Larcos G, Maisey MN. FDG PET screening for cerebral metastases in patients with suspected malignancy. Nuc Med Comm 1996; 17:197-198
11.
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
12.
Samii M, Carvalho GA, Schuhmann MU, Matthies C. Arachnoid cysts of the posterior fossa. Surg Neurol 1999; 51:376-382
13.
Altman NR, Naidich TP, Braffman BH. Posterior fossa malformations. AJNR 1992; 13:691-724
14.
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
15.
Bondurant CP, Jimenez DF. Epidemiology of cerebrospinal fluid shunting. Pediatr Neurosurg 1995; 23:254-259
16.
Rogers LR. Cerebrovascular complications in cancer patients. Neurol Clin 2003; 21:167-192
17.
Rowley HA. Cerebrovascular disease. In: Brant WE, Helms CA, eds.Fundamentals of diagnostic radiology, 2nd ed. Baltimore, MD: Lippincott Williams & Wilkins, 1998:79-112
18.
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
19.
Kelly PJ, Stein J, Shafqat S, et al. Functional recovery after rehabilitation for cerebellar stroke. Stroke 2001; 32:530-534
20.
Macdonell RA, Kalnins RM, Donnan GA. Cerebellar infarction: natural history, prognosis, and pathology. Stroke 1987; 18:849-855

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 866 - 873
PubMed: 17312080

History

Submitted: January 20, 2006
Accepted: May 30, 2006

Keywords

  1. arachnoid cyst
  2. brain cancer
  3. cerebrovascular accident
  4. FDG PET
  5. oncologic imaging
  6. stroke
  7. whole-body imaging

Authors

Affiliations

Erin Stubbs
Department of Radiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157.
Jonathan Kraas
Department of Radiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157.
Kathryn A. Morton
Department of Radiology, University of Utah Health Sciences Center, Salt Lake City, UT 84132.
Paige B. Clark
Department of Radiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

There are no citations for this item

View Options

View options

PDF

View PDF

PDF Download

Download PDF

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media