Normal variants and pitfalls in whole-body PET imaging with 18F FDG.
A glucose analogue, FDG is transported into the cells by glucose transporters and is phosphorylated intracellularly but not metabolized further. (2-4) The FDG distribution throughout the body mainly reflects the glucose metabolism of the individual tissues. One exception is the liver, which has a higher phosphatase activity than most other tissues and, therefore, has a higher washout of FDG. Glucose is the only nutrient that can be utilized by the brain, and brain uptake of FDG is consistently very high. Free fatty acids are the predominant metabolic substrate for the myocardium. However, after high carbohydrate intake, the insulin level is high and myocardial glucose utilization is predominant. Inflammatory cells also have increased glucose metabolism, and inflammation constitutes a major diagnostic challenge in oncologic PET imaging. (4,5) The main excretion pathway of FDG is through the kidneys, where FDG is processed differently from glucose without tubular reabsorption. The combination of an increased concentration of glucose transporters, increased glucose phosphorylation, and low phosphatase activity results in relatively high concentrations of FDG in most cancer cells. (2-4)
A pertinent patient history is needed for proper patient preparation. This is essential to optimize image quality and to reduce artifacts that may result in decreased diagnostic accuracy. The acquisition time should be adjusted based on patient size. Some patients will need sedation or relaxants. Information regarding concomitant relevant diseases is important. Patients with diabetes will need specific information about how to prepare for the study. Information about extent and location of inflammatory skin disease will be important if the clinical problem is malignant melanoma. Knowledge of prior biopsies or surgical procedures and a thorough knowledge of normal uptake, distribution, excretion, normal variations, and responses to different treatment regimes, artifacts, and potential pitfalls are essential for correct image interpretation. (6-12)
The uptake of FDG in skeletal muscle depends on the glucose insulin level. In the resting state, fatty acid oxidation is the major energy source. At a low exercise level, fat may provide 50% to 60% of the energy for muscle contraction, while glucose is the preferred fuel during exercise. The glucose is derived both from glycogen breakdown in the muscle and from dietary sources. Postprandial muscle uptake may be very extensive. During and after exercise, glycolysis becomes the major source of energy resulting in significant FDG uptake. To reduce FDG uptake in skeletal muscles, the patient should avoid any muscle effort before injection and during FDG distribution. Heavy muscle use before injection will result in increased uptake in the strained muscle groups. (6,7) Patients are therefore requested to abstain from heavy exercise during the 24 hours before a FDG-PET examination. As an example, walking with crutches will cause intense muscle uptake in the arms. (12) Spastic paresis will result in increased muscle uptake in the involved muscles. (9) Stress-induced muscle tension with the result of increased FDG uptake in the posterior cervical muscle groups, neck muscles, trapezius muscles, and paraspinal muscles is common. Asymmetric uptake in the sternocleidomastoid muscles, longus capitis, and longus colli is not uncommon. Attention to the position of the head during imaging and during the uptake phase is of prime importance. Use of a pillow to ensure slight head flexion and complete head relaxation is necessary to avoid unnecessary muscle uptake interference. Some have advocated the use of a benzodiazepine drug to suppress muscle uptake. Scalene muscle uptake is often seen after neck dissection, where the sternocleidomastoid muscle has been removed. Uptake in the intrinsic laryngeal musculature is a common finding and may be rather intense if the patient was talking during the uptake of FDG. (6,9,12,13) A typical finding is symmetric high uptake at the muscle origin and insertion of the arytenoid cartilage, posterior cricoarytenoid muscles, and some less intense uptake along the course of thyroarytenoid and vocalis muscle (11) (Figure 1). Paresis will result in asymmetric uptake. Especially after unilateral surgery or radiation therapy, the cervical muscle uptake may be remarkably asymmetric. The lower neck images sometimes show uptake only at the origin of the neck muscles, which can be confused with bilateral supraclavicular lymph node disease. The muscle uptake is, however, usually symmetric and linear, and additional confirmation of benign uptake can be found if palpation or CT correlation indicates no nodes of sufficient size to account for the amount of uptake seen on the scan. More rarely, temporalis, pterygoid, masseter, or other head and neck muscles will accumulate tracer, sometimes depending on specific use by the patient, so that obtaining a history of physical activity or intense chewing may be helpful. Taking steps to reduce or eliminate muscle uptake can be crucial to correct image interpretation. Ensuring that the patient is not chewing gum, chewing tobacco, reading, or talking during the uptake phase is important. Hyperventilation or forced respiration from bronchopulmonary disease may cause increased uptake in the diaphragm, crura, and intercostal muscles. (9,12) Orbital muscles are fast-twitch, fast-glycolytic muscles with high glycolytic enzyme content; they show consistently intense FDG uptake. With a high-resolution system, the individual eye muscles may be identified.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
In patients with poorly controlled diabetes, high muscle uptake can present diagnostic problems and the FDG accumulation in tumors can be decreased. (14) Not only will the tumor or muscle uptake be decreased, but the tumor/blood ratios may be reduced as well (Figure 2). Blood glucose levels should be checked prior to FDG injection and patients with whole blood values (plasma value = whole blood value x 1.12) >150 to 200 dL/mg (8.3 to 11.1 mmol/L) should be rescheduled for imaging when their blood glucose value is under control. Diabetic patients should take their diabetic control medication and have a morning meal no fewer than 4 hours prior to their appointment time. (7)
Intramuscular injection, eg, in the buttocks, may result in local inflammation and mild focal FDG uptake, but the pattern is normally easy to recognize.
Brown fat uptake, which is typically patchy and fairly symmetrically distributed, may be very intense and extensive (Figure 3). It is commonly located in the neck, suboccipital, supraclavicular, and paraspinal regions in the neck and chest. (9-11,15-17) Intense FDG uptake in brown fat may also be located on scattered perivascular areas in the mediastinum and in the retrocrural area. The brown fat uptake was previously believed to be of muscle origin, even if the pattern in many cases was hardly consistent with the anatomical configuration of muscles. On PET/CT, however, it has been recognized as uptake in fat. (15) It is important that the fat uptake is identified and understood and not mistaken as tumor uptake. However, brown fat, hypermetabolic tumor, and metastatic lymph nodes may coexist, and differentiation may be difficult without PET/CT. Uptake in brown fat may be so high and extensive that quantitative whole-body distribution may be influenced.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Brown fat is a highly specialized heat-producing tissue. (18) Repeat imaging, making sure that the patient is warm and comfortable during injection and uptake, may be needed to resolve the issue. To prevent as much brown fat uptake as possible, we typically use warm blankets and keep uptake rooms warm. Benzodiazepines can be used as an extra measure to reduce fat uptake for repeated scans when necessary. Beta-blockers are much more effective than benzodiazepines in rats for this purpose; however, no experience in humans is available. (19)
Consistent, mildly enhanced uptake corresponding to the skin on attenuation correction images is probably more a reconstruction artifact than actual high uptake in the skin. Inflammatory skin diseases, such as eczematous dermatitis and psoriasis, may cause focally increased FDG uptake. Intertrigo (eg, under heavy breasts and in the sacrococcygeal region between the buttocks) is common in obese persons and may result in linear increased FDG uptake that is normally easy to recognize.
Brain and spinal cord
As already mentioned, glucose is the only nutrient that can be utilized by the brain, and the FDG uptake in the cerebral and cerebellar cortex, basal ganglia, and thalamus is very intense. Tumor uptake may be obscured by normal high gray matter uptake. (10,20) There is normally symmetric distribution of FDG in both hemispheres. The ventricles will not be as well defined as they are on CT and MRI, and white matter is much less metabolically active than gray matter. The cortical distribution will, to some extent, be influenced by a variety of possible stimuli that occur just before or during injection and distribution of FDG, such as visual activity and muscle activity. Visual and frontal cortex will normally have some slightly higher uptake than temporal and parietal cortex and, as opposed to cerebral blood pool single-photon-emission CT (SPECT) with technetium (Tc)-99m-labeled radiopharmaceuticals, cerebral cortical uptake is commonly more intense than cerebellar cortical uptake. Moderate uptake is seen in the brain stem, and the cervical and upper thoracic spinal cord frequently show mild uptake, gradually decreasing in the caudal direction (Figure 4). Since glucose consumption of the brain is fairly constant, brain activity has often been used as a reference for semiquantification.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Nasopharynx and nasal sinuses
Interpretation of PET images of the nasopharynx region presents particular challenges. (13,21) The region shows substantial variations in normal uptake that can present difficulties in the identification of pathology. Intense uptake in the abundant lymphoid tissue in the pharynx, forming the Waldeyer ring (pharyngeal tonsil, palatine tonsils, lingual tonsillar tissue, and lymphoid tissue along salpingopharyngeal folds) is easily recognized (Figure 4). A negative correlation between age and intensity of uptake in the palantine tonsils is described. There also seems to be a positive correlation between intensity of 18F FDG uptake in the palantine tonsils and the lingual tonsils. (13)
Some asymmetry of the pharyngeal and palatine tonsils is not uncommon; however, especially when looking for a squamous cell carcinoma with unknown primary, any asymmetry should be noted as a possible site for the primary malignancy. A characteristic "V"-shaped high uptake area in the floor of mouth along the medial borders of the mandible is a consistent finding and is most likely the sublingual glands (Figure 5). The sublingual gland is a predominant mucous gland, which may explain the consistently much higher uptake compared to the parotid (serous gland) and submandibular glands (mixed serous and mucous). The uptake in the parotid and submandibular glands is variable but can be intense. As for the palantine tonsils, a positive correleation between age and uptake and sublingual glands has been described. (13)
Maxillary sinusitis is common. Increased uptake along the bony lining of the sinus is a typical finding consistent with inflammatory sinusitis (Figure 6).
The thyroid gland
The thyroid gland normally shows low uptake. (13) Diffuse thyroid uptake is most commonly consistent with chronic thyroiditis and may be very intense in Hashimoto's thyroiditis (22,23) (Figure 7). Our experience is that Hashimoto's thyroiditis may have uptake as intense as typically seen in anaplastic thyroid carcinoma and thyroid lymphoma and cautious interpretation is imperative. Diffusely increased uptake is also seen in active Graves' disease. (24) In Graves' disease, there seems to be a Connection between the glucose metabolism and hormone syntheses in the thyroid cells. The increased uptake in chronic thyroiditis with normal or reduced thyroid function is probably caused by glucose utilization by the inflammatory cells and not by the thyroid cells. (24,25)
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
Inhomogeneous thyroid uptake with multifocal uptake is seen in multinodular goiter. A hypermetabolic nodule in the thyroid gland as an incidental finding occurs in 1% to 2% of oncologic patients and is associated with malignant primary or secondary thyroid cancer in as many as 47% of the cases. (26-30) There are some reports indicating that standard uptake value (SUV) can be used to differentiate between malignant and benign lesions. (27,29,31) However, in a yet unpublished study of 28 patients with thyroid nodules with confirmed etiology, we did not find any significant difference in the SUV max between benign (18 of 28) and malignant (10 of 28) nodules (Trond Velde Bogsrud, MD, unpublished data, January 2006). It is important to know that a benign nodule may be very hot. We usually report that the finding of a hypermetabolic focus in the thyroid gland may represent a primary or secondary thyroid cancer, and we recommend follow-up with ultrasound and/or fine-needle aspiration biopsy.
There are some case reports of hypermetabolic nodules being autonomous thyroid nodules with subclinical hyperthyroidism. (32)
The thymus consistently shows triangular-shaped uptake with homogeneous mild to moderate intensity. After chemotherapy the thymus often shows a significant rebound hyperplasia with enlargement and increased FDG uptake that may last from 3 months to 1 year after completed chemotherapy, even in young adults. (2,9) The finding must not be mistaken for malignant thymic involvement or pathologic lymph nodes in the anterior mediastinal space. (2,6,8-12)
Lungs and mediastinum
Normal lungs have very faint uptake. Atelectasis, scarring, pleural inflammation, and pleural effusion may have mildly increased uptake, but it is generally less than mediastinal activity. (31) Pneumonia, including postradiation pneumonia, will have mild to fairly intense uptake. (6,8-10,33) Aspiration pneumonia is common in patients with head and neck cancer and must be differentiated from a synchronous or metachronous cancer. Lobar patterns of uptake or CT evidence of inflammatory consolidation may be of help in this regard. In patients with progressive fibrosis caused by silicosis, FDG uptake may be relatively intense. Other benign granulomas can show relatively intense FDG activity.
[FIGURE 10 OMITTED]
Mediastinal and hilar enlarged lymph nodes with uptake equal to or lower than mediastinum blood background are most likely inflammatory. Mediastinal and hilar lymph node enlargement is very common with sarcoidosis, and mild to fairly intense FDG uptake can pose a differential diagnostic challenge. (6,9,10,34) A variety of infectious agents can cause hilar and mediastinal lymph node enlargement with mild to moderate increased FDG uptake (Figure 8). Bilateral, symmetric hilar and mediastinal uptake is more typical for sarcoidosis than for malignancy. Malignant lymph node enlargement tends to have higher FDG uptake on average, but there is a broad overlap.
Free fatty acids are the predominant metabolic substrate for the myocardium. However, after high carbohydrate intake, the insulin level is high and glucose utilization increases. Oncologic FDG-PET imaging is performed in the fasting state ([greater than or equal to]4 hours since last meal) in order to reduce blood sugar and insulin levels and minimize FDG uptake in skeletal and cardiac muscle. In the fasting state, however, the myocardial uptake is still very variable in intensity and distribution. (35) Patchy atrial uptake may be more confusing than irregular ventricular uptake. Inhomogeneous myocardial uptake in a fasting patient must not be misinterpreted as coronary artery disease, neither paracardiac or of the parasternal hypermetabolic lymph nodes. Analysis of repeated studies shows intra-individual variation in FDG uptake even if the patient preparation is similar. (35)
The intensity of FDG uptake in breast tissue is variable. (9,10) Increased nipple uptake is seen frequently and may be asymmetric. Mild, diffuse breast uptake in the later phase of the menstrual cycle is reported. Fibrocystic changes may result in inhomogeneous mild to moderate, typically bilateral FDG uptake, and malignant tumors may be obscured. (9) Focal necrosis of fat tissue is followed by inflammation that will concentrate FDG. Fat necrosis in the breast may be Indistinguishable from carcinoma on clinical and mammographic examinations, and it is important to know that the FDG uptake may be high. As with surgical biopsies elsewhere, breast biopsies will result in focally increased uptake. Breast implants will result in a photopenic area, sometimes surrounded by a rim of mildly increased uptake. Rather intense breast uptake either diffusely or multifocally is seen during lactation.
Normal bone marrow shows mild uptake with homogeneous distribution. Bone marrow recovering after chemotherapy, after treatment with colony-stimulating factor (GCsF), and in anemic patients typically shows moderate to intense uptake, homogeneously distributed in the central skeleton and proximal extremities. (6,9,10,13) Malignant bone marrow involvement should be more irregular and asymmetric even though it may be extensive and widespread. Bone marrow recovering after chemotherapy in patients treated successfully for malignancy with bone marrow involvement may show inhomogeneous uptake in this recovery phase. Irradiated areas may show strikingly decreased uptake. (9) Sites for bone marrow biopsy or graft harvest will show focal or linear uptake that may last for about a month. Fibrous dysplasia may exhibit intensively increased FDG uptake. (9)
Mild uptake along the extent of the esophagus with more intense uptake at the gastroesophageal junction is normal. The marked diffuse uptake in the distal third of the esophagus may be due to inflammation caused by gastric reflux. (2,9) The stomach wall has variable uptake that can be mild to fairly intense. On coronal slices, the stomach may look tubular with photopenic lumen.
Diffusely distributed patchy uptake in the bowel is a consistent finding. (6,8-12,35) The uptake is most likely located in the bowel wall and may be related to smooth muscle, lymphatic system or inflammation; however, the exact mechanism is uncertain. Enemas do not reduce the bowel uptake. The bowel activity varies in the same patient on subsequent studies and is independent of blood glucose levels and the length of the fasting period. (9,35) Extensive intense uptake is sometimes seen in the colon and particularly in the cecum, the ascendant colon, and sigmoid colon (Figure 9). Moderate to intense uptake in the rectum is a common normal finding. Physiological bowel uptake or uptake related to inflammation will typically follow the contour and course of the bowel. However, patchy bowel uptake could be misinterpreted as mesenteric lymph nodes, bowel wall tumor, or pelvic tumor. Hiatal and abdominal wall hernias containing bowel may show moderately increased uptake. (9) Laparoscopic portals and colostomies will regularly exhibit increased uptake. (2) Inflammatory bowel disease, such as Crohn's disease or diverticulitis, shows focal, segmental, or diffuse increased FDG uptake. (2,36,37) On the contrary, segmental or diffuse increased FDG uptake is not equivalent with inflammatory bowel disease. Irritable bowel syndrome does not show increased uptake. Focal uptake that is more intense than the bowel background activity is worrisome for malignancy, and local wall thickening should be looked for on CT. Colorectal adenomas may show focal intense FDG uptake. (2) Further diagnostic workup is warranted for unexplained focal bowel uptake.
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
Hernias and enterostomies will show mild to moderately increased uptake, but normally do not cause any differential diagnostic problem.
Liver, spleen, pancreas, and adrenal glands
Low-level, diffusely mottled liver activity is common and may obscure small malignant lesions. Hepatocytes have a higher concentration of phosphatase enzymes compared with most other tissues, which results in dephosphorylation of FDG-6-phosphate and a faster FDG washout. This is the rationale behind delayed imaging of the liver that has been suggested by some authors for indeterminate liver lesions. (38) Side-by-side comparison with nonattenuated images may also be of help. Normal gallbladder contents will be photopenic. Splenic uptake is normally less intense and more homogeneous than the liver but may be very intense after treatment with granulocyte colony-stimulating factor. (9,10)
The pancreas normally shows only mild uptake. Fat necrosis may show very intense uptake and may be difficult to differentiate from malignant tumor.
Normal adrenal glands do not have increased uptake. Benign adrenal adenomas may have slightly increased uptake, but the intensity will be lower than it is in the liver. (10)
Degenerative joint disease and bursitis
Mild to relatively intense FDG uptake in degenerative joint disease is a common finding and must not be mistaken for malignancy (Figure 10). Low to moderate uptake in trochanteric bursitis is a common finding. (9) Increased uptake can be present surrounding prosthetic devices. (9,10) Moderately increased FDG uptake in the caput and collum area after hip joint replacement is seen regularly, while increased uptake on the bone-prosthesis interface along the femur shaft may indicate loosening. (9)
[FIGURE 13 OMITTED]
Bones and healing fractures
Paget's disease and fibrous dysplasia may show fairly intense uptake and must not be mistaken for bone metastases or primary bone tumors. (9,10) Traumatic and surgical fractures show increased uptake with highest intensity 2 to 3 weeks after the incident and is usually normalized within 3 months (39,40) (Figure 11). Persistent uptake may indicate delayed healing, nonunion or complicating osteomyelitis. Acute osteoporotic compression fracture of the vertebral bodies is reported not to have increased or only mildly increased FDG uptake. (41) Focally increased uptake at the sites of bone marrow biopsy and bone harvesting will already show increased FDG uptake the following day.
Injection site and dose infiltration
The site of injection should always be indicated. Taking a focal uptake on a wrist or elbow for granted as dose infiltration may lead to risk of overlooking metastasis. Dose infiltration may result in clot formation and clots may be trapped in lymph nodes or in the lungs and may mimic lymph node metastases and hypermetabolic lung nodules. The focal lung uptake caused by clots is typically small, intense, peripherally located, and sharply marginated and should be suspected when there is no corresponding CT finding (Figure 12). High activity in a dose infiltrate may create reconstruction artifacts, which may mask, eg, metastatic lesions in the liver. (9)
The marked vessel uptake in the major arteries that is frequently seen in elderly patients is located in the vessel wall and is likely related to the inflammatory component of the atherosclerotic process and not to calcification. (42) There is substantial evidence of FDG uptake at the sites of atheromatous plaques. In some modern PET systems using CT for attenuation correction, focal calcification may be artificially hot on PET caused by a reconstruction artifact. Nonattenuated images showing a photopenic focus will confirm the artifactual nature.
More localized diffuse or focally increased uptake in the walls of large vessels is seen in giant-cell arteritis and Takayasu arteritis; however, differentiation between an atheromatous process and large vessel vasculitis should not be based on PET alone. The atheromatous process mainly involves the lower extremities. When only the aorta and upper extremities are affected without involvement of the arteries of the lower extremities, arteritis and not an atheromatous process should be suspected.
Increased uptake corresponding to vascular grafts is a normal finding and must not be misinterpreted as infection or rejection (9) (Figure 13). The high uptake appears to persist for years.
Kidneys and urine collecting system
High activity of FDG is seen in the calyces, renal pelvis, ureters, bladder, and uretra, and may interfere with the interpretation of findings close to the course of the urinary system. (6,7,9,10) Special consideration must be taken to avoid contamination with radioactive urine. If urinary contamination occurs, it can cause artifacts and difficulty in scan interpretation. Urinary spillage may even contaminate the imaging table and have a consequence for subsequent patients.
Activity in the renal calyces may be misinterpreted as adrenal uptake or retroperitoneal hypermetabolic lymph nodes; and increased uptake in renal cell carcinoma may be misinterpreted as activity in the calyces. Combined PET/CT is of great help. Urinary activity in the ureters is typically linear but may be focal and asymmetric and mimic hypermetabolic lymph nodes. An alternation between transversal, coronal, and sagittal images and referencing the ureter is often needed to make sure that the activity of concern is really excreted urine. To facilitate clearance of activity from the renal collecting system, 20 to 40 mg of furosemide may be administered. (7) A duplicating collecting system and a horseshoe kidney may be confusing without PET/ CT or a diagnostic CT for comparison.
The bladder activity may obscure pelvic structures and may also create artifacts even with iterative reconstructive techniques. Primary bladder cancer is difficult to diagnose with FDG because of the normal high uptake in the bladder. Bladder catherization is often used to reduce bladder activity and artifacts in the pelvic region. (7) Different approaches are used. The simplest procedure is a regular bladder catheter kept open during the examination. A 3-way urinary catheter with continuous saline flushing during uptake and aquisition of the pelvic region allows the bladder to collapse and results in better clearing of the urinary activity but frequently leaves small foci of high intensity that may be confusing. When the bladder is collapsed, there may be confusion with adjacent small intestine activity. Urinary catheter clamping after emptying and saline infusion (200 mL) into the bladder just before pelvic imaging accentuates bladder definition without creating artifacts from urinary activity. The full bladder pushes the small bowel loops away from the lesser pelvis, making the evaluation of the ovaries, uterus, prostate gland, rectum, sigmoid colon, and pelvic lymph nodes optimal. After completed imaging of the pelvic region, the irrigation is stopped and the urinary catherer is unclamped. This latter approach also corresponds to the preferred CT technique of imaging with a full bladder.
[FIGURE 14 OMITTED]
[FIGURE 15 OMITTED]
Focal urinary activity in the urethra and residual urine in the bladder neck that are seen after transurethral prostate resection should not be misinterpreted as tumor. Experience from a limited number of patients at the Mayo Clinic with focally increased uptake in the prostate gland as an incidental finding indicates that prostate cancer is a common pathologic finding that needs to be considered if more lateral uptake in the prostate is seen (Mark Nathan, MD, unpublished data, June 2006; to be presented as an abstract at the Society of Nuclear Medicine meeting) (Figure 14).
The reproductive system
Uterine fibroids, ovaries, and menstruating endometrium may show relatively intense FDG uptake and must not be misinterpeted as ovarian cancer, endometrial cancer, or metastatic pelvic lymph nodes (2,9,10) (Figure 15). Increased ovarian uptake seems to be common after ovulation. Uterine fibroma may be relatively intense and can be difficult to differentiate from malignant tumor based on PET alone. Testicular uptake is variable and usually symmetric and may be very intense. (2,9)
Inflammatory changes secondary to radiation and surgery
The accumulation of FDG is not specific for malignant tumors. Infectious and inflammatory processes will also accumulate FDG. (5,6,8,10,43) It is important to recognize changes that were caused by surgery and radiation therapy and to differentiate them from malignancy. A short time after completed radiation therapy, radiation-induced inflammation may result in mildly to moderately increased uptake that gradually decreases over weeks and months until the tissue uptake is normalized or slightly reduced. It is not uncommon to see the radiation volume with some reduced uptake. Reduced uptake in irradiated bone marrow may be striking. Radiation-induced inflammation may sometimes last for years (eg, pleural inflammation after irradiation of lung cancer or mediastinal lymphoma).
Surgical wounds may be intensely positive for a short time, with a reduction in intensity over a couple of months. Surgical scars that are complicated with infection may be very intense.
PET artifacts caused by CT attenuation
State-of-the-art PET systems use CT for attenuation correction and a fusion of PET and CT images for anatomic localization of PET findings or metabolic characterization of CT findings as an integrated part of the examination. (2) A typical PET aquisition takes 20 to 40 minutes, and a CT of the chest or upper abdomen imaged within a few seconds with breath-holding will not match the PET in the fused dataset. To best correspond to the PET images, the CT is acquired during gentle shallow breathing or during a breath-hold at midexpiratory position. Respiratory motion may create CT artifacts and lead to locoregional Misregistration of the fused PET/CT data. (2) The most critical artifacts are those in the lower chest and upper abdomen. Curvilinear cold artifacts that parallel the dome of the diaphragm at the lung bases are frequently present on PET images with misregistration. The artifact is mild in severity in most patients and is more noticable in the right lung than in the left. The artifacts may cause interpretation problems with lesions located in the upper liver or lower lung fields.
Because of the standard 20 to 40 minutes of imaging time, a number of patients will not be able to position their arms over their heads, even if this were desired. Thus, a CT scan from a combined PET/CT study will often be deteriorated not only with respiratory artifacts, but also with beam-hardening artifacts from the patient's arms. (44,45)
Some misalignment between PET and CT caused by patient motion is also inevitable. Slightly changed head, jaw, or shoulder position, and a change in breathing depth as the study progresses and the patient gets more relaxed are all factors that will cause some misalignment between PET and CT. It is important that the physician reading the PET/ CT study identifies the misalignment and is able to interpolate the information from the fused images.
The attenuation correction using CT (40 to 140 keV) is based on conversion of the Hounsfield units to linear attenuation coefficients ([micro]-values) to be used at 511 keV. In this conversion, metallic objects tend to give too large [micro]-values, which result in an overcorrection that causes artificially high intensity in the reconstructed attenuation-corrected PET image. Examples of metallic objects that can cause artifacts are pacemakers, joint prostheses, dental implants, tracheal tubes, surgical clips, and venous ports. (2,9) Artifacts caused by metallic objects, by densely calcified lymph nodes, and by intravenous (IV) and oral contrast are reported to interfere with image interpretation. A mild inflammatory process with mildly to moderately increased FDG uptake around joint prostheses, pacemakers, tracheal tubes, and venous ports are common, but when increased FDG uptake is located corresponding to the center of the dense object, artifactually increased uptake as a consequence of overcorrection should be considered. Attenuation-corrected and noncorrected images will need to be reviewed side-by-side. The degree of overcorrection and artificially high uptake corresponding to dense objects or contrast seems to vary among different PET/CT systems on the market. In a newly published study, Yau et al (46) did not find any statistically or clinically significant spuriously elevated SUV level that might potentially interfere with the diagnostic value of PET/CT as a result of the application of IV iodinated contrast. In some institutions, a diagnostic contrast-enhanced CT is performed after completion of the PET/CT study.
If the CT transverse field-of-view is smaller than the PET field-of-view, truncation artifacts will be created if the patient is imaged with arms down or if the patient is large. (2,7)
The predominantly used radiopharmaceutical for diagnostic oncologic whole-body PET is FDG, a glucose analogue that is transported into the cells by glucose transporters and is phosphorylated intracellularly but not further metabolized. Approximately 50% of the injected activity is excreted unmetabolized in the urine. The distribution throughout the body mainly reflects the glucose metabolism of the individual tissues and the excretory pathway. Thus, the highest uptake will be in the urinary tract and in the brain, and there will be variable high uptake in the heart. Diffusely distributed, patchy, rather intense uptake in the bowel may be difficult to differentiate from malignancy. Uptake in skeletal muscle is highly variable, depending on muscle use before and during FDG uptake and the glucose/insulin status. Intense uptake in brown fat must not be misinterpreted as malignant lymph nodes. Oncologic PET is based on the trapping of FDG in most malignant tumors. However, the mechanism is not unique for malignancy, and inflammatory cells also concentrate FDG. Discriminating inflammatory processes from malignancy is one of the major challenges in PET interpretation.
State-of-the art PET systems use CT for attenuation correction and a fusion of PET and CT images for anatomic localization of PET findings or metabolic characterization of CT findings. CT attenuation has introduced new artifacts (eg, those caused by breathing and overcorrection of dense material) that are important to recognize.
Combining CT with PET has increased specificity as well as physicians' confidence in reading the PET. As with other imaging modalities, repeat or follow-up studies may be needed to increase specificity. Either PET/CT or PET fused with a diagnostic CT is of great help in differentiating between physiologic muscle uptake, brown fat, and pathologic uptake in tumor or metastatic lymph nodes. If no combined PET/CT or program for fusion of diagnostic CT and PET is available, CT and/or MRI images should be present for side-by-side comparison.
(1.) Townsend D. A future for PET/CT. In: Czernin J, Dahlbom M, Ratib O, Schiepers C, eds. Atlas of PET/CT Imaging in Oncology. New York, NY: Springer-Verlag; 2004:12-20.
(2.) Czernin J, Yap C. From FDG-PET to FDG-PET/CT. In: Czernin J, Dahlbom M, Ratib O, Schiepers C, eds. Atlas of PET/CT Imaging in Oncology. New York, NY: Springer-Verlag; 2004:46-53.
(3.) Oehr P. Metabolism and transport of glucose and FDG. In: Ruhlmann J, Oehr P, Biersack HJ, eds. PET in Oncology. Basics and clinical applications. Berlin, Germany: Springer-Verlag; 1999:43-57.
(4.) Kowalsky RJ, Falen SW, eds. Radiopharmaceuticals in Nuclear Pharmacy and Nuclear Medicine. 2nd ed. Washington DC: American Pharmacists Association. 2004.
(5.) Zhuang H, Alavi A. 18-fluorodeoxyglucose positron emission tomographic imaging in the detection and monitoring of infection and inflammation. Semin Nucl Med. 2002;32:47-59.
(6.) Shreve PD, Bui CDH. Normal variants in FDG PET imaging. In: Wahl RL, ed. Principles and Practice of Positron Emission Tomography. Philadelphia, PA: Lippincott Williams & Wilkins; 2002:111-136.
(7.) Hamblen SM, Lowe VJ. Clinical 18F-FDG oncology patient preparation techniques. J Nucl Med Technol. 2003;31:3-7.
(8.) Cook GJR. Artifacts and normal variants in PET imaging. In: Valk PE, Bailey DL, Townsend DW, Maisey MN, eds. Positron Emission Tomography: Basic Science and Clinical Practice. London, UK: Springer-Verlag; 2003:495-505.
(9.) Kipper MS, Tartar M. Clinical Atlas of PET: With Imaging Correlation. Philadelphia, PA: W. B. Saunders; 2004.
(10.) Seltzer M, Schiepers C. Normal pattern and common pitfalls of FDG-PET image interpretation. In: Czernin J, Dahlbom M, Ratib O, Schiepers C, eds. Atlas of PET/CT Imaging in Oncology. New York, NY: Springer-Verlag; 2004:54-59.
(11.) Czernin J, Dahlbom M, Ratib O, Schiepers C, eds. Atlas of PET/CT Imaging in Oncology. New York, NY: Springer-Verlag; 2004.
(12.) Cook GJ, Fogelman I, Maisey MN. Normal physiological and benign pathological variants of 18-fluoro-2-deoxyglucose positron-emission tomography scanning: Potential for error in interpretation. Semin Nucl Med. 1996;26:308-314.
(13.) Nakamoto Y, Tatsumi M, Hammoud D, et al. Normal FDG distribution patterns in the head and neck: PET/CT evaluation. Radiology. 2005; 234:879-885.
(14.) Torizuka T, Zasadny KR, Wahl RL. Diabetes Decreases FDG accumulation in primary lung cancer. Clin Positron Imaging. 1999;2:281-287.
(15.) Hany TF, Gharehpapagh E, Kamel EM, et al. Brown adipose tissue: A factor to consider in symmetric tracer uptake in the neck and upper chest region. Eur J Nucl Med Mol Imaging. 2002;29:1393-1398.
(16.) Cohade C, Osman M, Pannu HK, Wahl RL. Uptake in the supraclavicular area fat ("USA-FAT"): Description on 18F-FDG PET/CT. J Nucl Med. 2003;44:170-176.
(17.) Truong MT, Erasmus JJ, Munden RF, et al. Focal FDG uptake in mediastinal brown fat mimicking malignancy: A potential pitfall resolved on PET/CT. AJR Am J Roentgenol. 2004;183:1127-1132.
(18.) Weber WA. Brown adipose tissue and nuclear medicine imaging. J Nucl Med. 2004;45:1101-1103.
(19.) Tatsumi M, Engles JM, Ishimori T, et al. Intense (18)F-FDG uptake in brown fat can be reduced pharmacologically. J Nucl Med. 2004;45:1189-1193.
(20.) Van Heertum RL, Tikofsky RS. Brain tumors, other diseases, and activation. In: Van Heertum RL, Tikofsky RS. Functional Cerebral SPECT and PET Imaging. Philadelphia, PA: Lippincott Williams & Wilkins; 2000:275-312.
(21.) Lowe VJ, Stack BC. PET imaging in head and neck cancer. In: Valk PE, Bailey DL, Townsend DW, Maisey MN, eds. Positron Emission Tomography: Basic Science and Clinical Practice. London, UK: Springer-Verlag; 2003:535-546.
(22.) Yasuda S, Shohtsu A, Ide M, et al. Chronic thyroiditis: Diffuse uptake of FDG at PET. Radiology. 1998;207:775-778.
(23.) Gianoukakis AG, Karam M, Cheema A, Cooper JA. Autonomous thyroid nodules visualized by positron emission tomography with 18F-fluorodeoxyglucose: A case report and review of the literature. Thyroid. 2003;13:395-399.
(24.) Boerner AR, Voth E, Theissen P, et al. Glucose metabolism of the thyroid in Graves' disease measured by F-18-fluoro-deoxyglucose positron emission tomography. Thyroid. 1998;8:765-772.
(25.) Boerner AR, Voth E, Theissen P, et al. Glucose metabolism of the thyroid in autonomous goiter measured by F-18-FDG-PE. Exp Clin Endocrinol Diabetes. 2000;108:191-196.
(26.) Bogsrud TV, Karantanis D, Nathan MA, et al. Focal high uptake in the thyroid gland as an incidental finding on 18F-FDG PET. Thyroid. 2004;14:760, abstract 240.
(27.) Cohen MS, Arslan N, Dehdashti F, et al. Risk of malignancy in thyroid incidentalomas identified by fluorodeoxyglucose-positron emission tomography. Surgery. 2001;130:941-946.
(28.) Van den Bruel A, Maes A, De Potter T, et al. Clinical relevance of thyroid fluorodeoxyglucose-whole body positron emission tomography incidentaloma. J Clin Endocrinol Metab. 2002;87:1517-1520.
(29.) Kang KW, Kim SK, Kang HS, et al. Prevalence and risk of cancer of focal thyroid incidentaloma identified by 18F-fluorodeoxyglucose positron emission tomography for metastasis evaluation and cancer screening in healthy subjects. J Clin Endocrinol Metab. 2003;88:4100-4104.
(30.) Van den Bruel A, Maes A, De Potter T, et al. Clinical relevance of thyroid fluorodeoxyglucose-whole body positron emission tomography incidentaloma. J Clin Endocrinol Metab. 2002;87:1517-1520.
(31.) Bloom AD, Adler LP, Shuck JM. Determination of malignancy of thyroid nodules with positron emission tomography. Surgery. 1993;114:728-735.
(32.) Gianoukakis AG, Karam M, Cheema A, Cooper JA. Autonomous thyroid nodules visualized by positron emission tomography with 18F-fluorodeoxyglucose: A case report and review of the literature. Thyroid. 2003;13:395-399.
(33.) McAdams HP, Erasums JJ, Patz EF, Goodman PC, Coleman RE. Evaluation of patients with round atelectasis using 2-[18F]-fluoro-2-deoxy-D-glucose PET.J Comput Assist Tomogr. 1998;22:601-604.
(34.) Kavanagh PV, Stevenson AW, Chen MY, Clark PB. Nonneoplastic diseases in the chest showing increased activity on FDG PET. AJR Am J Roentgenol. 2004;183:1133-1141.
(35.) de Groot M, Meeuwis AP, Kok PJ, et al. Influence of blood glucose level, age and fasting period on nonpathological FDG uptake in heart and gut. Eur J Nucl Med Mol Imaging. 2005;32:98-101.
(36.) Pandit-Taskar N, Schoder H, Gonen M, et al. Clinical significance of unexplained abnormal focal FDG uptake in the abdomen during whole-body PET. AJR Am J Roentgenol. 2004;183:1143-1147.
(37.) Neurath MF, Vehling D, Schunk K, et al. Noninvasive assessment of Crohn's disease activity: A comparison of 18F-fluorodeoxyglucose positron emission tomography, hydromagnetic resonance imaging, and granulocyte scintigraphy with labeled antibodies. Am J Gastroenterol. 2002;97:1978-1985.
(38.) Kubota K, Itoh M, Ozaki K, et al. Advantage of delayed whole-body FDG-PET imaging for tumour detection. Eur J Nucl Med. 2001;28:696-703.
(39.) Zhuang H, Sam JW, Chacko TK, et al. Rapid normalization of osseous FDG uptake following traumatic or surgical fractures. Eur J Nucl Med Mol Imaging. 2003;30:1096-1103.
(40.) Shon IH, Fogelman I. F-18 FDG positron emission tomography and benign fractures. Clin Nucl Med. 2003;28:171-175.
(41.) Schmitz A, Risse JH, Textor J, et al. FDG-PET findings of vertebral compression fractures in osteoporosis: Preliminary results. Osteoporos Int. 2002; 13:755-761.
(42.) Belhocine T, Blockmans D, Hustinx R, et al. Imaging of large vessel vasculitis with (18)FDG PET: Illusion or reality? A critical review of the literature data. Eur J Nucl Med Mol Imaging. 2003;30: 1305-1313.
(43.) Stumpe KD, Dazzi H, Schaffner A, von Schulthess GK. Infection imaging using whole-body FDG-PET. Eur J Nucl Med. 2000;27:822-832.
(44.) Osman MM, Cohade C, Nakamoto Y, Wahl RL. Respiratory motion artifacts on PET emission images obtained using CT attenuation correction on PET-CT. Eur J Nucl Med Mol Imaging. 2003; 30:603-606.
(45.) Nehmeh SA, Erdi YE, Rosenzweig KE, et al. Reduction of respiratory motion artifacts in PET imaging of lung cancer by respiratory correlated dynamic PET: Methodology and comparison with respiratory gated PET. J Nucl Med. 2003;44:1644-1648.
(46.) Yau YY, Chan WS, Tam YM, et al. Application of intravenous contrast in PET/CT: Does it really introduce significant attenuation correction error? J Nucl Med. 2005;46:283-291.
Trond Velde Bogsrud, MD and Val J. Lowe, MD
Dr. Bogsrud is the Head of Nuclear Medicine Oncology, Department of Nuclear Medicine, Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway. Dr. Lowe is an Associate Professor, Department of Radiology, Division of Nuclear Medicine, Mayo Clinic, Rochester, MN.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||positron emission tomography; fluoro-2- deoxy-D-glucose|
|Author:||Bogsrud, Trond Velde; Lowe, Val J.|
|Article Type:||Clinical report|
|Date:||Jun 1, 2006|
|Previous Article:||Upcoming meetings.|
|Next Article:||The future of ultrasound: a conversation with Arnd Kaldowski.|