POSITRON EMISSION TOMOGRAPHY
53
Andrew Newberg, MD
1. What instructions do patients need to prepare for positron emission tomography
(PET) scan with fluorodeoxyglucose (FDG)?
The most important issue regarding a PET scan is that FDG competes with nonradioactive glucose. If a patient
has recently eaten or has diabetes with a blood glucose level greater than 150 mg/dL, the sensitivity of the
PET scan would be diminished. Typically, blood glucose levels greater than 200 mg/dL should exclude a patient
from having a PET scan. Nondiabetic patients should have nothing by mouth (NPO) overnight or at least for
4 to 6 hours if their scan is later in the day. Diabetic patients should be managed carefully because some have
difficulty forgoing eating. Also, insulin drives glucose, including FDG, into the muscles, so care should be taken
regarding the administration of insulin too close to the scan time. It is preferred that the patient not inject insulin
or eat before the PET scan, but if this is impossible, the referring physician must balance glucose levels, eating,
and insulin in the diabetic patient. The scan itself takes place approximately 30 minutes after the intravenous
injection of FDG, and performance of a whole-body scan takes an additional 30 to 60 minutes, depending on the
type of scanner.
2. Is bowel activity normal on the FDG PET scan in Fig. 53-1?
Generally, bowel activity can be a normal finding on a PET scan, which sometimes makes the evaluation of
cancer in the bowel more difficult. Some centers recommend a bowel preparation to minimize constipation,
which might result in substantial smooth muscle uptake
in the bowel. Bowel uptake that is very focal and
intense should raise suspicion, but it is very important
to correlate such findings with either anatomic imaging
or endoscopy.
3. What structures in the head and neck
normally take up FDG? Is the neck uptake
normal in Fig. 53-2?
Normal uptake in the head and neck can be observed in the
facial muscles, tongue (especially when patients are talking
at the time of injection), neck muscles, brown fat, thyroid
tissue, and vocal cords. The uptake of FDG can be very
intense and can mimic or obscure cancer in these regions.
Minimizing talking and patient movement and keeping the
environment quiet and dimly lit during injection and for
approximately 15 to 20 minutes after injection may help to
diminish uptake in these areas.
4. Is the lung nodule in the right upper lobe
benign or malignant in Fig. 53-3?
A lung nodule that has a moderate amount of uptake
needs to be evaluated further for malignancy. Typically, a
standardized uptake value (SUV), a quantity that incorporates
the patient s size and the injected dose, that is more than
2.0 is considered to be suggestive of malignancy, whereas
lesions with SUVs less than this value are considered to be
benign. The SUV of this nodule in the right upper lobe is 4.0.
Benign lesions include inflammatory or infectious etiologies.
There are examples, however, in which active infectious or
inflammatory processes may result in SUVs comparable to
Figure 53-1. Normal FDG PET scan shows mild diffuse
SUVs observed in malignant disease, so clinical history and uptake throughout the bowel (normal variant), mild uptake in
the liver, significant uptake in the brain and heart, and marked
anatomic correlation are required for adequately assessing
uptake in the bladder.
any finding on FDG PET.
369
C
H A P T E R
370 POSITRON EMISSION TOMOGRAPHY
Figure 53-3. FDG PET scan of a patient with lung cancer
Figure 53-2. The neck uptake in this patient is related to
shows intense uptake in the right upper lobe with a SUV
muscle uptake or brown fat. Other structures normally seen
of 4.0. Additional areas of increased uptake in the left
in the neck include salivary glands, oropharynx, tongue, vocal
supraclavicular region and the right inguinal region are also
cords, and sometimes the thyroid. Uptake in the heart, brain,
concerning for malignancy, and would require additional
kidneys, and bladder is normal. There are several small foci of
follow-up. Uptake in the heart, kidneys, and bladder is
mildly increased uptake in the lungs, which suggests an active
normal.
metabolic process, such as neoplasm.
5. Is the patient in Fig. 53-3 a good surgical candidate?
One of the important uses of FDG PET is presurgical evaluation of patients with known lung cancer. If FDG PET scan
shows malignant lesions on the contralateral hilar regions or distant metastases, the patient no longer is a surgical
candidate and would proceed to systemic therapy. In this particular patient, there are additional areas of uptake that are
concerning for malignancy in the left supraclavicular region and the right inguinal region, which would require additional
follow-up. If these areas are confirmed to be malignant, the patient would no longer be a surgical candidate.
6. How can FDG PET change the management of patients with lung cancer?
Studies have revealed that FDG PET changes the surgical management of patients in 40% of cases. In some cases,
distant metastases or restaging indicates that the cancer is inoperable, preventing surgery that would not have been
useful. In 20% of patients, PET shows that enlarged nodes that may have prevented surgery from being considered are
benign, so that surgery can be performed. This ability to change management has also contributed to the discovery that
FDG PET is cost-effective in the management of patients with lung cancer because it can greatly alter the management
and guide appropriate treatment.
Key Points: Indications for PET
1. Oncologic staging: St aging or restaging of non small cell lung cancer, breast cancer,
colorectal cancer, melanoma, lymphoma, head and neck cancer, and esophageal cancer
(this indication is approved by Medicare)
2. Brain tumors : Differentiation of residual or recurrent brain tumor from radiation necrosis
3. Seizures: Presurgical, interictal identification of refractory brain seizure foci
4. Cardiac imaging: Metabolic assessment of myocardial viability
NUCLEAR RADIOLOGY 371
7. What are potential confounding
diseases that can mimic lung
cancer?
Several nonmalignant diseases can mimic lung
cancer by appearing as focally intense areas
of increased uptake on FDG PET scan. Most
infectious processes, such as pneumonia, active
tuberculosis, or other abscesses, can appear as
focal areas of increased activity. Sarcoidosis and
other granulomatous diseases can also manifest
as focally increased uptake (Fig. 53-4). Iatrogenic
causes, such as postsurgical or postradiation
therapy, can have increased activity, although
radiation changes are usually more diffuse,
involving most of the radiation field.
8. How is dual time point imaging
used?
The physiology of glucose uptake is such that
malignant cells continue to take up more and
more glucose over time, whereas normal tissues
establish an equilibrium. By taking FDG PET images
at several time points, the later time point images
should reveal malignant regions as increasing in
intensity, especially compared with the background
levels. Dual time point imaging can contribute to
making more accurate diagnoses, especially when
Figure 53-4. FDG PET scan of a patient with sarcoidosis shows
FDG uptake is borderline high on earlier images multiple areas of intensely increased activity in the hilar regions; these
findings appear similar to findings in lung cancer. This scan also shows
and increases on later images.
increased activity in other regions, such as the left axilla, abdominal, and
pelvic lymph nodes.
9. How important prognostically is FDG
PET scan with negative results after treatment for lymphoma?
Studies evaluating the prognostic value of PET imaging for lymphoma after therapy have been quite dramatic. Patients
with a PET scan with negative results have generally been shown to have greater than a 90% cure rate with excellent
long-term survival. Patients with PET scans with positive results after therapy have less than a 10% chance of cure,
and most have a survival of less than 2 to 3 years. FDG PET after chemotherapy for patients with lymphoma has very
important prognostic implications and may suggest additional therapy for patients with scans with positive results (Fig. 53-5).
10. What are the current indications for FDG PET in patients with lymphoma?
Currently, FDG PET scans are recommended for initial staging, especially to evaluate whether there is disease above and
below the diaphragm. PET imaging can also be valuable in the early stages of therapy to assess whether the particular
intervention is working. PET can be used to obtain prognostic information after therapy. PET can be used to evaluate
the possibility of recurrence of disease and can be a primary means of long-term follow-up of patients. PET can also be
used for restaging if tumor recurrence is already observed.
11. How useful is PET in the evaluation of colon cancer?
A meta-analysis of studies evaluating almost 600 patients with colorectal cancer showed a sensitivity and specificity
of whole-body FDG PET of 97% and 75%. The specificity was found to be greater for local recurrence and hepatic
metastases (>95%). FDG PET imaging was responsible for a change in patient management in almost 30% of cases.
12. How does correlation with anatomic imaging help in detecting cancer accurately?
Fusing computed tomography (CT) and PET images has significant advantages over PET or CT alone. Studies have
suggested that combining CT with PET may increase the diagnostic accuracy in 40% of patients compared with PET
or CT alone. Specific advantages of fusing CT and PET images include discriminating metastases from physiologic foci
of activity, improving lesion detection on PET and CT, precisely localizing metastatic foci, differentiating bone from soft
tissue, differentiating liver from adjacent bowel, and identifying specific structures of the neck. These advantages have
led to a change in management in 10% to 20% of patients over PET alone.
13. How does PET help with patient planning for radiation therapy?
FDG PET can play several roles with respect to planning for radiation therapy. Because PET can detect distant metastatic
disease, it may change the need for performing radiation therapy in the first place. PET can also be beneficial in
helping to evaluate the area that requires radiation. Numerous studies have shown that PET may show that the area of
372 POSITRON EMISSION TOMOGRAPHY
Figure 53-5. FDG PET scans in a patient with lymphoma before (left ) and after (right ) chemotherapy. The scan before chemotherapy
shows multiple foci of increased activity in the neck and abdominal lymph node chains. The scan after chemotherapy reveals complete
resolution of these findings, suggesting a good response to therapy and good overall prognosis.
increased metabolism extends beyond the abnormal region on anatomic imaging, which would result in expanding the
target volume for radiation treatment. Alternatively, PET may show that some areas believed to be involved with cancer
on anatomic imaging are hypometabolic and are not actively neoplastic. In this setting, the radiation volume can be
reduced. Such changes in the plan for radiation therapy can occur in 25% of patients.
14. Is FDG PET useful for evaluating peritoneal seeding?
FDG PET can detect peritoneal seeding, but the sensitivity and specificity of PET for making such a diagnosis are much
less than for other, more focal lesions. Peritoneal seeding can be difficult to detect on other imaging modalities as
well; PET can still be useful when CT or magnetic resonance imaging (MRI) findings are equivocal because diffusely
increased FDG uptake in various abdominal areas can suggest peritoneal seeding. PET can often show negative results,
however, in patients with peritoneal seeding.
15. How useful is FDG PET in the evaluation of osteomyelitis?
FDG PET has been shown to be useful for evaluating infections in bones and soft tissues. Studies have suggested
that FDG PET has a high sensitivity (90% to 100%) and specificity (81% to 89%) for detecting osteomyelitis in lower
limb prostheses. Similar sensitivities and specificities have been reported in the detection of chronic osteomyelitis.
Areas of osteomyelitis typically have intense uptake that is localized in the bone itself. More recently, PET has been
used to evaluate soft tissue infectious or inflammatory disease, such as vascular graft infection, inflammatory bowel
disease, and fever of unknown origin (Fig. 53-6).
16. How is PET used in the interictal evaluation of patients with seizure disorders?
FDG PET scans are commonly used to evaluate seizure patients who are refractory to medications before surgery is
performed to remove specific seizure foci. FDG PET scan findings are also correlated with other clinical findings, such
as electroencephalography, Wada test, MRI findings, and neuropsychologic tests. FDG PET has a sensitivity of more than
70% in the detection of seizure foci. The most common approach is to perform an interictal scan in which a seizure
focus is expected to have reduced glucose metabolism. A clearly observed seizure focus on FDG PET can confirm the
location of the seizure focus so that surgery can be performed. In Fig. 53-7, there is hypometabolism in the left temporal
lobe, which is consistent with a seizure focus.
17. What are the most common areas for seizure foci, and what are the implications for
finding multiple abnormal areas?
The temporal lobe is the most common focus of partial epilepsy and is the region that can be most accurately evaluated
with FDG PET. The frontal lobes are the next most common area where seizure foci arise. This area is also relatively easy
NUCLEAR RADIOLOGY 373
to observe on FDG PET, and the sensitivity and
specificity for the detection of seizure foci in both
regions are similar. On scans in which more than one
area appears to be involved, or when the regions
of hypometabolism extend to other structures, the
likelihood of resolution of seizures after surgery is
typically diminished, sometimes substantially.
18. How useful is FDG PET for the
evaluation of primary brain tumors
or new metastatic disease?
FDG PET has generally not been useful for detecting
primary brain or new metastatic tumors because
the normal brain has such a high metabolism, and
it is often difficult to find a tumor embedded in
areas that already have high metabolism. Some
studies have suggested, however, that the degree of
hypermetabolism in tumors that have been detected
may have prognostic implications because highly
metabolic tumors tend to be the most aggressive and
portend the worse prognosis. Low-grade tumors have
lower metabolism and have an overall better prognosis.
19. What are the sensitivity and
specificity for differentiating tumor
recurrence from radiation necrosis
in patients with brain cancer?
Although FDG PET seems to be useful in grading
brain tumors and determining their prognosis, the
Figure 53-6. FDG PET scan of a patient with suspected major use of PET imaging is its ability to distinguish
osteomyelitis of the spine reveals intense activity in two adjacent
radiation necrosis from tumor recurrence. The
mid-thoracic vertebral bodies. This finding is consistent with
ability to differentiate these two entities has critical
osteomyelitis.
clinical implications and is often difficult with MRI
or CT. Generally, radiation necrosis should have
virtually no metabolism, whereas recurrent tumor
has increased metabolism. The ability to detect
tumor recurrence is enhanced by the contrast of
the tumor over the background hypometabolism
induced by the radiation therapy (or surgery).
Reports have generally scored the sensitivity and
specificity of PET for detecting tumor recurrence as
approximately 85% and 60% (Fig. 53-8).
20. What are the characteristic features
of Alzheimer disease on FDG PET in
Fig. 53-9?
The classic pattern of Alzheimer disease on FDG
PET is hypometabolism in the temporoparietal
regions; this may also involve the posterior
Figure 53-7. Interictal FDG PET brain scan in a patient with seizures
reveals decreased metabolism in the left temporal lobe, consistent with cingulate gyrus. The subcortical areas,
a seizure focus.
sensorimotor area, visual cortex, and cerebellum
are generally less affected. More recent studies
have shown, however, that other areas may be hypometabolic, particularly when the patient has specific neurocognitive
deficits. Temporoparietal hypometabolism can also be observed in other conditions, including Parkinson disease,
bilateral parietal subdural hematomas, bilateral parietal stroke, and bilateral parietal radiation therapy ports.
21. What is the typical metabolic pattern in the PET study of a patient with depression
(Fig. 53-10)?
FDG PET studies of depressed patients usually show decreased metabolism, which can be global or affect more specific
regions, such as the frontal lobes. More recent studies have suggested that certain areas may have increased activity,
such as the limbic regions. This more global pattern can usually be distinguished from specific neurodegenerative
diseases, such as Alzheimer or Pick disease, which typically affect the temporoparietal (Alzheimer disease) and frontal
(Pick disease) lobes.
374 POSITRON EMISSION TOMOGRAPHY
Figure 53-8. FDG PET brain scan in a patient after radiation therapy of a tumor in the left parietal region reveals decreased metabolism
in the left parietal lobe with an intense region of activity centrally. These findings are consistent with recurrent brain tumor in the setting
of radiation necrosis.
Figure 53-9. FDG PET brain scan in a patient with Alzheimer
disease shows bilateral temporoparietal hypometabolism. This
is the most typical feature in Alzheimer disease, although other
structures, including the frontal lobes and visual cortex, can also
be involved.
22. How is cardiac PET used clinically?
Cardiac PET usually uses two different tracers during a stress test. One tracer (ammonia N 13 or rubidium-82) typically
evaluates perfusion, and FDG evaluates metabolism. In ischemia, perfusion is decreased in the affected area. In the
heart, ischemic areas switch from use of fatty acids to glucose for energy, however, so the FDG scan shows preserved
or increased uptake. PET scans also can help show viability because preserved FDG uptake suggests ischemic but
viable tissue, whereas decreased perfusion and metabolism suggests infarcted tissue.
NUCLEAR RADIOLOGY 375
Figure 53-10. FDG PET brain scan in a patient with moderate depression reveals globally decreased cerebral glucose metabolism
compared with the subcortical structures of the basal ganglia and thalami.
BIBLIOGRAPHY
[1] R. Hustinx, F. Benard, A. Alavi, Whole-body FDG-PET imaging in the management of patients with cancer, Semin. Nucl. Med. 32 (2002)
35 46.
[2] L. Kostakoglu, S.J. Goldsmith, 18F-FDG PET evaluation of the response to therapy for lymphoma and for breast, lung, and colorectal
carcinoma, J. Nucl. Med. 44 (2003) 224 239.
[3] A.B. Newberg, A. Alavi, Role of positron emission tomography in the investigation of neuropsychiatric disorders, in: M.P. Sandler, R.E.
Coleman, J.A. Patton, et al. (Eds.), Diagnostic Nuclear Medicine, fourth ed., Lippincott Williams & Wilkins, Philadelphia, 2003, pp. 783 819.
[4] P.E. Valk, D.L. Bailey, D.W. Townsend, M.N. Maisey (Eds.), Positron Emission Tomography: Basic Science and Clinical Practice,
London, Springer, 2003.
[5] H. Zhuang, A. Alavi, 18-Fluorodeoxyglucose positron emission tomographic imaging in the detection and monitoring of infection and
inflammation, Semin. Nucl. Med. 32 (2002) 47 59.