Procedure Guideline for Tumor Imaging with 18F-FDG PET/CT 1.0* Dominique Delbeke1, R. Edward Coleman2, Milton J. Guiberteau3, Manuel L. Brown4, Henry D. Royal5, Barry A. Siegel5, David W. Townsend6, Lincoln L. Berland7, J. Anthony Parker8, Karl Hubner9, Michael G. Stabin10, George Zubal11, Marc Kachelriess12, Valerie Cronin13, and Scott Holbrook14 1Vanderbilt University Medical Center, Nashville, Tennessee; 2Duke University Medical Center, Durham, North Carolina; 3Christus St. Joseph Hospital, Houston, Texas; 4Henry Ford Hospital, Detroit, Michigan; 5Mallinckrodt Institute of Radiology, St. Louis, Missouri; 6University of Tennessee, Knoxville, Tennessee; 7University of Alabama Hospital, Birmingham, Alabama; 8Beth Israel Deaconess Hospital, Boston, Massachusetts; 9University of Tennessee Medical Center, Knoxville, Tennessee; 10Vanderbilt University, Nashville, Tennessee; 11Yale University, New Haven, Connecticut; 12Institute of Medical Physics, University of Erlangen-Nurnberg, Erlangen, Germany; 13Mercy Hospital, Buffalo, New York; and 14Precision Nuclear, Gray, Tennessee I. PURPOSE available for several years, the readily apparent and doc- umented advantages of having PET and CT in a single device The purpose of these guidelines is to assist physicians in have resulted in the rapid dissemination of this technology recommending, performing, interpreting, and reporting the in the United States. This Procedure Guideline pertains results of 18F-FDG PET/CT for oncologic imaging of adult only to combined PET/CT devices. and pediatric patients. II. BACKGROUND INFORMATION AND DEFINITIONS Deﬁnitions PET is a tomographic scintigraphic technique in which a A. A PET/CT scanner is an integrated device containing computer-generated image of local radioactive tracer dis- both a CT scanner and a PET scanner with a single tribution in tissues is produced through the detection of patient table and therefore capable of obtaining a CT annihilation photons that are emitted when radionuclides scan, a PET scan, or both. If a patient does not move introduced into the body decay and release positrons. 18F- between the scans, the reconstructed PET and CT FDG PET is a tomographic imaging technique that uses a images will be spatially registered. radiolabeled analog of glucose, 18F-FDG, to image relative B. PET/CT registration is the process of aligning PET glucose use rates in various tissues. Because glucose use is and CT images for the purposes of combined image increased in many malignancies, 18F-FDG PET is a sensi- display (fusion) and image analysis. tive method for detecting, staging, and monitoring the C. PET/CT fusion is the combined display of registered effects of therapy of many malignancies. CT is a tomo- PET and CT image sets. Superimposed data typically graphic imaging technique that uses an x-ray beam to are displayed with the PET data color coded to the produce anatomic images. This anatomic information is CT data in gray scale. used to detect and to help determine the location and extent D. PET/CT acquisitions can include the whole body, an of malignancies. Combined PET/CT devices provide both extended portion of the body, or a limited portion of the metabolic information from 18F-FDG PET and the the body. These acquisitions are deﬁned in Current anatomic information from CT in a single examination. As Procedural Terminology 2005 as follows: shown in some clinical scenarios, the information obtained 1. Whole-body tumor imaging: from the top of the by PET/CT appears to be more accurate in evaluating pa- head through the feet tients with known or suspected malignancies than does the 2. Skull base–to–midthigh tumor imaging information obtained from either PET or CT alone or the 3. Limited-area tumor imaging results obtained from PET and CT separately but inter- E. Methods of attenuation correction: preted side by side. 1. CT transmission imaging with PET/CT scanners 18F-FDG PET and CT are proven diagnostic procedures. 2. Transmission scanning with an isotopic source: Although techniques for registration and fusion of images not commonly used with PET/CT scanners obtained from separate PET and CT scanners have been Received Mar. 10, 2006; accepted Mar. 10, 2006. For correspondence or reprints contact: Dominique Delbeke, Vanderbilt III. EXAMPLES OF CLINICAL OR RESEARCH University Medical Center, 21st Ave. S. and Garland, Nashville, TN 37232- 2675. APPLICATIONS E-mail: firstname.lastname@example.org *YOU CAN ACCESS THIS ACTIVITY THROUGH THE SNM WEB SITE Indications for 18F-FDG PET/CT include but are not (http://www.snm.org/guidelines). limited to the following: A. Differentiating benign from malignant lesions a. For brain imaging, the patient should be in a quiet B. Searching for an unknown primary tumor when and dimly lit room for 18F-FDG administration metastatic disease is discovered as the ﬁrst manifes- and the subsequent uptake phase. tation of cancer or when the patient presents with a b. For body imaging, the patient should remain seated paraneoplastic syndrome or recumbent for 18F-FDG administration and the C. Staging known malignancies subsequent uptake phase to avoid muscular uptake. D. Monitoring the effect of therapy on known malig- c. The blood glucose level should be checked before nancies 18F-FDG administration. Tumor uptake of 18F-FDG E. Determining whether residual abnormalities detected is reduced in hyperglycemic states. Most institu- on physical examination or on other imaging studies tions reschedule the patient if the blood glucose after treatment represent tumor or posttreatment ﬁ- level is greater than 150–200 mg/dL. Reducing the brosis or necrosis serum glucose level by administering insulin can F. Detecting tumor recurrence, especially in the pres- be considered, but the administration of 18F-FDG ence of elevated levels of tumor markers should be delayed after insulin administration (with G. Selecting the region of a tumor most likely to yield the duration of the delay being dependent on the diagnostic information for biopsy type and route of administration of insulin). H. Guiding radiation therapy planning d. For either a CT scan done for attenuation cor- I. Nononcologic applications, such as evaluation of rection/anatomic localization (AC/AL) or a diag- infection and atherosclerosis nostic CT scan of the abdomen or pelvis, an intraluminal gastrointestinal contrast agent may 18F-FDG PET/CT is not equally effective for all malig- be administered to provide adequate visualization nancies, but other tracers are available; however, many of of the gastrointestinal tract unless it is medically these are not yet approved by the U.S. Food and Drug contraindicated or unnecessary for the clinical Administration or reimbursable by the Medicare program. indication (see Section E.2.b.). The scientiﬁc literature concerning its utility continues to evolve rapidly. B. Information Pertinent to Performing Procedure See also the Society of Nuclear Medicine Procedure Guidelines for General Imaging. IV. PROCEDURE 1. Focused history, including the type and site of malig- A. Patient Preparation nancy, dates of diagnosis and treatment (biopsy results, The optimum preparation for patients about to undergo surgery, radiation, chemotherapy, and administration PET/CT is evolving. The major goals of preparation are to of bone marrow stimulants and steroids), and current minimize tracer uptake in normal tissues, such as the medications. myocardium and skeletal muscle, while maintaining uptake 2. History of diabetes, fasting state, and recent infection in target tissues (neoplastic disease). The following is a 3. Patient’s ability to lie still for the duration of the commonly used protocol. acquisition (15–45 min) 4. History of claustrophobia 1. Pregnancy and breast-feeding: see the Society of Nuclear 5. Patient’s ability to put his or her arms overhead Medicine Procedure Guidelines for General Imaging 2. Before arrival C. Precautions Patients should be instructed to fast and not con- sume beverages, except for water, for at least 4–6 h See the Society of Nuclear Medicine Procedure Guide- before the administration of 18F-FDG to decrease lines for General Imaging. physiologic glucose levels and to reduce serum insu- lin levels to near basal levels. Oral hydration with D. Radiopharmaceutical water is encouraged. Intravenous ﬂuids containing With PET/CT, the radiation dose to the patient (Table 1) dextrose or parenteral feedings also should be with- is the combination of the radiation dose from the PET held for 4–6 h. radiopharmaceutical and the radiation dose from the CT When intravenous contrast material is to be used, portion of the study. Radiation dose in diagnostic CT has patients should be screened for a history of iodinated attracted considerable attention in recent years, in particular contrast material allergy, use of metformin for the for pediatric examinations. It can be very misleading to treatment of diabetes mellitus, and renal disease. state a ‘‘representative’’ dose for a CT scan because of the Intravenous contrast material should not be adminis- wide diversity of applications, protocols, and CT systems. tered when the serum creatinine level is above 2.0 This caveat also applies to the CT component of a PET/CT mg/dL. study. For example, a body scan may include various portions 3. Before injection of the body and may use protocols aimed to reduce the TABLE 1 18F-FDG Radiation Dosimetry for Adults and Children Organ receiving the largest radiation dose, Effective dose, Patient Intravenously administered activity mGy/MBq (rads/mCi) mSv/MBq (rems/mCi) Adult 370–740 MBq (10–20 mCi) Bladder, 0.16* (0.59) 0.019 (0.070) Child (5 y old) 5.18–7.4 MBq/kg (0.14–0.20 mCi/kg) Bladder, 0.32y (1.2) 0.050 (0.18) *Voiding interval, 3.5 h. Changes in bladder wall dose are approximately linear with changes in voiding interval; therefore, for a voiding interval of 2.0 h, dose to bladder wall would change by a factor of 2/3.5. y Voiding interval, 2.0 h. Data are from International Commission on Radiological Protection. Radiation Dose to Patients from Radiopharmaceuticals. St. Louis, MO: Elsevier; 2000:49. ICRP publication 80. radiation dose to the patient or aimed to optimize the CT scan performing whole-body tumor imaging offers the for diagnostic purposes. The effective dose could range from advantage of staging the entire body. approximately 5 to 80 mSv (0.5–8.0 rems) for these options. c. For optimal imaging of the body, the arms should It is therefore advisable to estimate the CT dose speciﬁc to the be elevated over the head if that position can be CT system and protocol being used. tolerated by the patient. Arms along the side may Pediatric and adolescent patients should have their CT produce beam-hardening artifacts over the torso. examinations performed at milliampere-seconds settings However, for optimal imaging of the head and appropriate for patient size, regardless of the CT protocol neck, the arms should be positioned along the used, because radiation dose to the patient increases sig- side. niﬁcantly as the diameter of the patient decreases. d. The patient should void the bladder before the acquisition of the images to limit the radiation dose to the renal collecting system and bladder. E. Image Acquisition e. Metallic objects should be removed from the See also the Society of Nuclear Medicine PET Tumor patient whenever possible. Imaging Guidelines and the ‘‘Speciﬁcations of the Exam- 2. Protocol for CT imaging ination’’ and ‘‘Documentation’’ sections of the American The CT component of a PET/CT examination can College of Radiology Practice Guideline for the Perfor- be performed either for AC/AL or as an optimized mance of Computed Tomography of the Extracranial Head diagnostic CT scan. An AC/AL CT scan has not and Neck in Adults and Children, the American College necessarily been optimized as a diagnostic CT of Radiology Practice Guideline for the Performance of examination, whereas for a diagnostic CT scan, such Pediatric and Adult Thoracic Computed Tomography optimization has been attempted. In some circum- (CT), and the American College of Radiology Practice stances, both an initial CT acquisition for AC/AL Guideline for the Performance of Computed Tomography (before the PET data acquisition) and diagnostic (CT) of the Abdomen and Computed Tomography (CT) of CT (after the PET data acquisition) are performed. the Pelvis. Optimization of the CT technique used in PET/CT 1. Field of view, positioning, and preacquisition preparation continues to evolve. a. Skull base–to–proximal thigh imaging generally is a. If the CT scan is obtained for AC/AL, a low recommended to survey the body in the search for milliampere-seconds setting is recommended to areas of abnormal 18F-FDG accumulation for most decrease the radiation dose to the patient. tumor types. Such PET/CT scans typically are b. For an optimized diagnostic CT scan, standard acquired from the external auditory meatus to the CT milliampere-seconds settings are recom- midthigh region. For tumors with a high likelihood mended to optimize the spatial resolution of the of scalp, skull, or brain involvement or lower- CT scan. Tube current modulation may be used extremity involvement, whole-body tumor imag- to minimize radiation dose to the patient. In ing is performed. some cases, intravenous or oral contrast material b. Limited-area tumor imaging can be considered may be used. A separate CT acquisition may be when critical abnormalities are likely to be local- necessary to produce an optimized diagnostic ized in a known region of the body (e.g., solitary CT scan that is requested for a particular region pulmonary nodule, probable lung cancer, evalua- of the body. For many indications, the exami- tion of hilar lymph node involvement, diagnosis of nation is performed with intravenous contrast head and neck cancer, and monitoring of therapy material and appropriate injection techniques. of locally advanced breast cancer). However, High intravascular concentrations of intravenous contrast material may cause an attenuation cor- imaging and is based on the administered activity, rection artifact on the PET image, but the impact patient body weight, and sensitivity of the PET usually is modest. This artifact is minimized scanner (as determined largely by detector com- on scanners by use of appropriate correction position and acquisition method). Typically, for factors. imaging skull to midthigh, the total acquisition c. For either a CT scan done for AC/AL or an time ranges from 15 to 45 min. The imaging time optimized diagnostic CT scan of the abdomen or typically is prolonged for the acquisition of brain pelvis, an intraluminal gastrointestinal noncaloric images or for images of a limited region of interest. contrast agent may be administered to provide c. Semiquantitative estimation of tumor glucose me- adequate visualization of the gastrointestinal tract tabolism by use of the SUV is based on relative unless it is medically contraindicated or unneces- lesion radioactivity measured on images corrected sary for the clinical indication. This agent may be for attenuation and normalized for the injected dose a positive contrast agent (such as dilute barium), and body weight, lean body mass, or body surface an oral iodinated contrast agent, or a negative area. This measurement is obtained on a static contrast agent (such as water). Collections of highly emission image typically acquired more than 45 concentrated barium or iodinated contrast agents min after injection. The accuracy of SUV measure- can result in an attenuation correction artifact that ments depends on the accuracy of the calibration of leads to a signiﬁcant overestimation of the regional the PET scanner, among other factors. The repro- 18F-FDG concentration; other, dilute positive and ducibility of SUV measurements depends on the negative oral contrast agents cause less overestima- reproducibility of clinical protocols, for example, tion and do not affect PET image quality. dose inﬁltration, time of imaging after 18F-FDG d. With regard to the breathing protocol for CT administration, type of reconstruction algorithms, transmission scanning, in PET/CT, the position type of attenuation maps, size of the region of of the diaphragm on the PET emission images interest, changes in uptake by organs other than the should match as closely as possible that on the CT tumor, and methods of analysis (e.g., maximum and transmission images. Although a diagnostic CT mean). scan of the chest typically is acquired during end- d. Semiquantitative estimation of tumor metabolism inspiration breath holding, this technique is not can be based on the ratio of 18F-FDG uptake in a optimal for PET/CT because it may result in lesion to 18F-FDG uptake in internal reference substantial respiratory motion misregistration on regions, such as the blood pool, mediastinum, PET and CT images. Some facilities perform CT liver, and cerebellum. transmission scans during breath holding at mid- inspiration volume, and others prefer that the F. Interventions patient continue shallow breathing during the CT 1. Intense urinary bladder tracer activity degrades image acquisition. Respiratory motion results in inaccu- quality and can confound the interpretation of ﬁndings rate localization of lesions at the base and periph- in the pelvis. Hydration and a loop diuretic, without or ery of the lungs, in the dome of the liver, or near with bladder catheterization, may be used to reduce any lung–soft tissue interface and may result in accumulated urinary tracer activity in the bladder. spurious standardized uptake value (SUV) deter- Bladder catheterization with a 3-way ﬂushing catheter minations. Motion correction or respiratory gating running a continuous bladder ﬂush after injection until is recommended when available. imaging has been used successfully to clear bladder 3. Protocol for PET emission imaging activity. a. The radiopharmaceutical should be injected at a 2. Keeping the patient in a warm room for 30–60 min site contralateral to the site of concern. Emission before the injection of 18F-FDG, particularly in cold images should be obtained at least 45 min after climates and air-conditioned environments, will help radiopharmaceutical injection. The optimal 18F- to minimize brown fat uptake. Lorazepam or diazepam FDG distribution phase is controversial. Many fa- given before the injection of 18F-FDG may reduce cilities start the acquisition of the images at 60 or 90 uptake by brown adipose tissue or skeletal muscle. min after 18F-FDG administration. Some facilities b-Blockers also may reduce uptake by brown fat. obtain a second set of images to assess the change in uptake over time. The 18F-FDG uptake time should G. Processing be constant whenever possible and certainly when 2 1. PET reconstruction: PET emission data consist of the studies are being compared by use of semiquanti- number of events along lines of response between tative parameters, especially the SUV. detector pairs. The emission data must be corrected b. The emission image acquisition time varies from for detector efﬁciency (normalization), system dead 2 to 5 min or longer per bed position for body time, random coincidences, scatter, attenuation, and sampling nonuniformity. Some of these corrections H. Interpretation Criteria (e.g., attenuation) can be incorporated directly into the 1. Normal physiologic uptake of 18F-FDG can be seen to reconstruction process. Scanners with retractable some extent in every viable tissue, including the brain, septa can acquire data in both 2-dimensional (2D) myocardium (where the uptake is signiﬁcant in some and 3-dimensional (3D) modes, whereas scanners patients despite prolonged fasting), breast, liver, without septa acquire data in the 3D mode only. spleen, stomach, intestines, kidneys and urine, mus- Datasets acquired in the 3D mode either can be cle, lymphoid tissue (e.g., tonsils), bone marrow, rebinned into 2D data and reconstructed with a 2D salivary glands, thymus, uterus, ovaries, testes, and algorithm or can be reconstructed with a fully 3D brown adipose tissue (see Section K). algorithm. Iterative reconstruction approaches are 2. For whole-body surveys, studies have shown that 18F- now widely available for clinical applications in both FDG PET of the brain is relatively insensitive for 2D and 3D modes, largely replacing the direct, ﬁltered the detection of cerebral metastases, because of high backprojection methods used previously. For a given physiologic 18F-FDG uptake in the gray matter. algorithm, the appropriate reconstruction parameters 3. Increased uptake of 18F-FDG can be seen in neo- will depend on the acquisition mode, the type of plasms, granulation tissue (e.g., healing wounds), scanner, and the imaging task. It is considered good infections, and other inﬂammatory processes. practice to archive reconstructions both with and 4. Although the pattern of 18F-FDG uptake and speciﬁc without attenuation correction to resolve issues arising CT ﬁndings as well as the correlation with history, from potential artifacts generated by the CT-based physical examination, and other imaging modalities attenuation correction procedure. The reconstructed im- usually are the most helpful features in differentiating age volume can be displayed in transaxial, coronal, and benign from malignant lesions, semiquantitative esti- sagittal planes and as a rotating maximum-intensity- mates (e.g., SUV) also may be of value, especially for projection image. evaluating changes over time or with therapy. 2. CT reconstruction: CT sinograms are reconstructed by ﬁltered backprojection at full ﬁeld of view for the data I. Reporting used for attenuation correction of the PET emission See also the Society of Nuclear Medicine Procedure data and separately for CT interpretation with appro- Guidelines for General Imaging. priate zoom, slice thickness and overlap, and recon- 1. Study identiﬁcation struction algorithms for the particular region of the 2. Clinical information body scanned. The ﬁltered backprojection can be a. Indication for the study either 2D after appropriate portions of the spiral CT b. Relevant history data are collected into axial or tilted planes or fully c. Information needed for billing 3D. In addition to the reconstruction kernel that 3. Procedure description and imaging protocol adjusts in-plane features, such as spatial resolution a. Radiopharmaceutical, including administered ac- and noise texture, longitudinal ﬁltration (along the tivity, route of administration, and 18F-FDG up- z-axis) is used to modify the z-resolution and the slice take time sensitivity proﬁles. In addition, there are techniques b. Other drugs administered and procedures performed, for emphasizing certain image features, for example, such as placement of intravenous line; hydration; bone, lung, or head algorithms. For attenuation cor- insertion of Foley catheter (size of catheter); admin- rection, only the standard kernels are used. Because istration of furosemide (amount and time), muscular CT volumes today are nearly isotropic, reslicing in relaxants, or pain medications; and sedation proce- coronal, sagittal, or even curved displays often is dures (brieﬂy describe the procedure, state the type of preferred. Advanced display techniques, such as vol- medication and time of sedation in relation to the ume rendering and maximum- or minimum-intensity radiotracer injection, and state the patient condition projections applied to the complete volume or to at the conclusion of the PET study) thick, arbitrarily oriented sections, often are used. c. Field of view and patient positioning: whole body, Organ- and task-speciﬁc automatic or semiautomatic skull base to midthigh, or limited area and position segmentation algorithms and special evaluation algo- of the arms rithms also are in routine use. d. Baseline glucose level 3. Display: With an integrated PET/CT system, typically e. CT transmission protocol (for AC/AL or diagnostic the software packages provide registered and aligned CT protocol with or without oral or intravenous CT images, 18F-FDG PET images, and fusion images contrast material and with the appropriate protocol in the axial, coronal, and sagittal planes as well as for the clinical scenario and body region of interest) maximum-intensity-projection images for review in the f. PET emission protocol: see the Society of Nuclear 3D cine mode. 18F-FDG PET images with and without Medicine Procedure Guidelines for General attenuation correction should be available for review. Imaging 4. Description of ﬁndings a. Whenever possible, a precise diagnosis should be a. Quality of the study: for example, limited because given of motion, muscular uptake, or hyperglycemia b. When appropriate, a differential diagnosis should b. Describe the location, extent, and intensity of be given abnormal 18F-FDG uptake in relation to uptake c. When appropriate, follow-up and additional diag- in normal comparable tissues and describe the nostic studies needed to clarify or conﬁrm the relevant morphologic ﬁndings related to PET ab- impression should be recommended normalities on the CT images. An estimate of the intensity of 18F-FDG uptake can be provided by the SUV; however, the intensity of uptake may be J. Quality Control described as mild, moderate, or intense or in rela- 1. Radiopharmaceuticals tion to the background uptake in normal hepatic See the Society of Nuclear Medicine Procedure parenchyma (average SUV: 2.0–3.0; maximum Guidelines for Use of Radiopharmaceuticals. SUV: 3.0–4.0). The integrated PET/CT report 2. Instrumentation speciﬁcations should include any detected incidental ﬁndings See also the ‘‘Equipment Speciﬁcations’’ and on the CT scan that are relevant to patient care. If ‘‘Quality Control’’ sections of the American College the CT scan was requested and performed as a of Radiology Practice Guideline for the Performance diagnostic examination, then the CT component of Computed Tomography of the Extracranial Head of the study may be reported separately, if neces- and Neck in Adults and Children, the American sary, to satisfy regulatory, administrative, or re- College of Radiology Practice Guideline for the imbursement requirements. In that case, the Performance of Pediatric and Adult Thoracic Com- PET/CT report can refer to the diagnostic CT scan puted Tomography (CT), and the American College of report for ﬁndings not related to the PET/CT Radiology Practice Guideline for the Performance of examination. Computed Tomography (CT) of the Abdomen and c. Limitations: when appropriate, identify factors Computed Tomography (CT) of the Pelvis. that can limit the sensitivity and speciﬁcity of a. Equipment performance guidelines the examination (e.g., small lesions or an inﬂam- For patient imaging, state-of-the-art scanners matory process) meet or exceed the following speciﬁcations: d. Clinical issues: address or answer any pertinent For the CT scanner: see the American College clinical questions raised in the request for the of Radiology guidelines mentioned above. imaging examination For the PET scanner: These speciﬁcations are e. Comparative data based on manufacturers’ data obtained according Comparisons with previous examinations and to National Electrical Manufacturers Association reports, whenever possible, should be part of the (NEMA) 2001 protocols; the uniformity estimate radiologic consultation and report. PET/CT is based on the NEMA 1994 protocol. For lutetium studies are more valuable when correlated with oxyorthosilicate–based scanners, a modiﬁed previous diagnostic CT, previous PET, previous NEMA protocol is deﬁned to account for the PET/CT, previous MRI, and all appropriate natural radioactive background from lutetium imaging studies and clinical data that are relevant. oxyorthosilicate. When PET/CT is performed for monitoring d In-plane spatial resolution: ,6.5 mm therapy, a comparison of the extent and intensity d Axial resolution: ,6.5 mm of uptake may be summarized as metabolic pro- d Sensitivity (3D): .4.0 cps/kBq gressive disease, metabolic stable disease, meta- d Sensitivity (2D): .1.0 cps/kBq bolic partial response, or metabolic complete d Uniformity: ,5% response. The European Organization for Re- For the PET/CT scanner: search and Treatment of Cancer has published d Maximum co-scan range (CT and PET): .160 criteria for these categories, although these criteria cm have not yet been validated in outcome studies. A d Maximum patient weight: at least 157.50 kg change in the intensity of uptake with semiquan- (350 lb) titative measurements, expressed in absolute d Patient port diameter: at least 59 cm values and percent change, may be appropriate All currently available commercial PET/CT in some clinical scenarios. However, the technical scanners meet or exceed these speciﬁcations. protocol and analysis of images need to be con- The CT ﬁeld of view should be at least 50 cm sistent in the 2 sets of images, as described in in diameter to minimize artifacts from CT-based Section E.3. attenuation correction because of mismatch be- 5. Impression (conclusion or diagnosis) tween the CT and the PET ﬁelds. A workstation with the capability to display d Brown adipose tissue CT, PET, and fused images with different per- d Thymus, especially in children centages of CT and PET blending also should be d Lactating breast available. The workstation should allow multi- d Areola planar display with linked CT and PET cursors. d Skeletal and smooth muscles (e.g., neck or Postcollection registration of the PET and CT paravertebral; hyperinsulinemia) datasets and registration with other imaging stud- d Gastrointestinal (e.g., esophagus, stomach, or ies, including nonrigid registration, are desirable. bowel) b. Equipment quality control d Urinary tract structures (containing excreted 18F-FDG) PET performance monitoring should be in ac- cordance with the American College of Radiology d Female genital tract (e.g., uterus during menses Technical Standard for Medical Nuclear Physics or corpus luteum cyst) Performance Monitoring of Nuclear Medicine b. Inﬂammatory processes Imaging Equipment. d Postsurgical inﬂammation, infection, or hema- CT performance monitoring should be in ac- toma; biopsy site; or amputation site cordance with the American College of Radiology d Postradiation (e.g., radiation pneumonitis) Technical Standard for Medical Physics Perfor- d Postchemotherapy mance Monitoring of Computed Tomography d Local inﬂammatory disease, especially granu- (CT) Equipment. lomatous processes (e.g., sarcoidosis, fungal The quality control procedures for PET/CT disease, or mycobacterial disease) should incorporate both CT procedures and PET d Ostomy site (e.g., trachea or colon) and drainage procedures according to the American College of tubes Radiology technical standards mentioned above. d Injection site The quality control procedures for CT should d Thyroiditis include air and water calibrations in Hounsﬁeld d Esophagitis, gastritis, or inﬂammatory bowel units for a range of kilovolts. The quality control disease procedures for PET should include a calibration d Acute and occasionally chronic pancreatitis measurement of activity in a phantom containing a d Acute cholangitis and cholecystitis known concentration, generally as a function of d Osteomyelitis, recent fracture sites, or joint axial position within the scanner ﬁeld of view. A prostheses daily check on the stability of the individual d Lymphadenitis detectors also should be performed to identify c. Benign neoplasms detector failures and drifts. d Pituitary adenoma In addition, for PET/CT, a check on the align- d Adrenal adenoma ment between the CT and the PET scanners should d Thyroid follicular adenoma be performed periodically. Such a gantry align- d Salivary gland tumors (e.g., Warthin’s tumor or ment check should determine any offset between pleomorphic adenoma) the CT and the PET scanners to be incorporated d Colonic adenomatous polyps and villous into the fused image display to ensure accurate adenoma image alignment. d Ovarian thecoma and cystadenoma 3. Emergency procedures d Giant cell tumor An emergency cart containing appropriate medi- d Aneurysmal bone cyst cations and resuscitation equipment must be readily d Leiomyoma available to treat adverse contrast material reactions. d. Hyperplasia or dysplasia d Graves’ disease K. Sources of Error d Cushing’s disease Processes other than malignancies may cause false- d Bone marrow hyperplasia (e.g., anemia or cy- positive and false-negative results. The following list, tokine therapy) although not all-inclusive, includes the most commonly d Thymic rebound hyperplasia (postchemotherapy) encountered causes: d Fibrous dysplasia 1. False-positive ﬁndings d Paget’s disease a. Physiologic uptake that may lead to false-positive e. Ischemia interpretations d Hibernating myocardium d Salivary glands and lymphoid tissues in the head f. Artifacts and neck d Misalignment between PET and CT data can d Thyroid cause attenuation correction artifacts. PET images without attenuation correction and fusion images their observations subsequently integrated, there is a need can be used to help identify these artifacts. to deﬁne the training for people who can interpret and d Inaccuracies in converting from polychromatic integrate both components of PET/CT scans. Regardless of CT energies to the 511-keV energy of annihila- previous training, imaging experts interpreting PET/CT tion radiation can cause artifacts around metal scans should have appropriate training in both PET and or dense barium, although these artifacts are less CT. Ideally, a diagnostic radiologist who has not received common with newer conversion algorithms. training and experience in PET should have training and 2. False-negative ﬁndings experience in PET similar to that of a nuclear medicine d Small size (,2 times the resolution of the system) physician, and a nuclear medicine physician should have d Tumor necrosis training and experience in CT similar to that of a diagnostic d Recent chemotherapy or radiotherapy radiologist, including treatment of contrast material reac- d Recent high-dose steroid therapy tions. In most instances, it is not feasible for a practicing d Hyperglycemia and hyperinsulinemia diagnostic radiologist to duplicate exactly the PET training d Some low-grade tumors (e.g., sarcoma, lymphoma, that a nuclear medicine physician receives during a nuclear or brain tumor) medicine residency, nor is it possible for a practicing d Tumors with large mucinous components nuclear medicine physician to duplicate the CT training d Some hepatocellular carcinomas, especially well- obtained in a diagnostic radiology residency. differentiated tumors An article summarizing discussions regarding issues d Some genitourinary carcinomas, especially well- relating to imaging with PET, CT, and PET/CT recently differentiated tumors was published by a collaborative working group with d Prostate carcinoma, especially well-differentiated representatives from the American College of Radiology, tumors the Society of Nuclear Medicine (SNM), and the Society of d Some neuroendocrine tumors, especially well- Computed Body Tomography and Magnetic Resonance (J differentiated tumors Nucl Med. 2005;46:1225–1239). These organizations agree d Some thyroid carcinomas, especially well- that only appropriately trained, qualiﬁed physicians should differentiated tumors interpret PET/CT images. Traditionally, appropriate train- d Some bronchioloalveolar carcinomas ing has been quantiﬁed by the number of continuing d Some lobular carcinomas of the breast medical education credits earned and the number of cases d Some skeletal metastases, especially osteoblastic or interpreted. The training recommendations from the col- sclerotic tumors laborative working group are summarized in Table 2. d Some osteosarcomas Alternative approaches, such as determining the accuracy of each physician’s interpretation of images compared with V. QUALIFICATION OF PERSONNEL that of his or her peers by use of a workstation simulator A. Physicians and a report generation and scoring system, may have equal The use of PET/CT technology is becoming common or greater validity. practice. Because it is inefﬁcient to have PET/CT images In the future, the requirements of radiology and nuclear interpreted by 2 different imaging experts and then to have medicine residency training programs will include training TABLE 2 Summary of PET/CT On-The-Job Training Requirements No. of PET/CT No. of CT interpretations interpretations PET/CT CME CT CME Training Board certiﬁcation (supervised)* (supervised)y credits credits Nuclear medicine ABNM 150 500 8 100 Diagnostic radiology (recent CT)z ABR 150 35 Nuclear radiology (recent CT)z ABR 150 8 Radiology (recent CT)z ABR and ABNM 150 8 Diagnostic radiology (no recent CT) ABR 150 500 35 100 *Supervision should be performed by qualiﬁed nuclear medicine physicians or diagnostic radiologists who have interpreted more than 500 PET/CT studies. y Supervision should be performed by qualiﬁed diagnostic radiologists as deﬁned in American College of Radiology Practice Guidelines for Performing and Interpreting Diagnostic Computed Tomography. CT cases should include reasonable distribution of head and neck, chest, abdomen, and pelvis. z Includes radiologists or nuclear radiologists with recent experience in body CT (100 body CT cases/y for preceding 5 y). ABNM 5 American Board of Nuclear Medicine. in the interpretation and supervision of integrated PET/CT The Nuclear Medicine Technology Certiﬁcation Board studies. Certifying and recertifying examinations will in- (NMTCB) has developed a PET specialty examination that clude testing on CT, PET, and PET/CT. Eligibility for is open to certiﬁed or registered nuclear medicine technol- taking the recertiﬁcation examinations will mandate par- ogists, registered radiologic technologists, CT technolo- ticipation in a maintenance of certiﬁcation (MOC) program gists, and registered radiation therapists, as long as they and will include training in the interpretation of PET, CT, have fulﬁlled the required prerequisites, as deﬁned on and PET/CT images. Some components of the MOC will the NMTCB Web site (www.nmtcb.org). The American include evaluation of the accuracy of each physician’s Registry of Radiologic Technologists (ARRT) has adapted interpretation of images compared with that of his or her its CT certiﬁcation examination and has allowed certiﬁed peers by use of a workstation simulator and a report gen- or registered nuclear medicine technologists who have eration and scoring system. Performing and interpreting met the required prerequisites to take this examination. physicians should participate in and be able to show Eligibility criteria are located on the ARRT Web site evidence of participation in continuing medical education (www.arrt.org). in the techniques and interpretation related to the proce- Licensure and regulation deﬁnitely are affecting the dures discussed in these guidelines. Where MOC programs opportunities that nuclear medicine technologists have for exist, physicians should be able to show evidence of obtaining the CT experience needed to take the ARRT CT participation. examination as well as the opportunities that radiologic If no physicians with training in the interpretation of technologists have for gaining the PET experience needed integrated PET/CT studies are available, PET scans should to take the NMTCB PET examination. The SNMTS is be interpreted by nuclear medicine physicians with exper- approaching these issues through both legislative and reg- tise in PET, and CT scans should be interpreted by diag- ulatory pathways. The SNMTS has been promoting the nostic radiologists with expertise in CT. Strong opinions Consumer Assurance of Radiologic Excellence bills pend- have been expressed that a single report should be issued to ing before the U.S. Congress. The Consumer Assurance of avoid inconsistencies, confusion, and redundancy, although Radiologic Excellence bills would establish minimum ed- reimbursement issues still are being debated. ucation and credentialing standards for those who perform medical imaging and therapeutic procedures. The second B. Technologists pathway recognizes the regulatory route in addressing these PET/CT technology presents similar practice issues re- practice issues through a collaborative liaison relationship garding the education, training, and certiﬁcation of tech- that has been established with the Conference of Radiation nologists to become appropriately qualiﬁed and competent Control Program Directors (www.crcpd.org), the profes- to perform PET/CT. Additional issues arise with regard to sional organization of state radiation regulators. ensuring competency, standardizing the educational expe- rience of technologists, and barriers placed by licensure and regulation at the state level. The Society of Nuclear Med- C. Qualiﬁed Medical Physicists icine Technologist Section (SNMTS) and the American A qualiﬁed medical physicist is an individual who is Society of Radiologic Technologists (ASRT) have jointly competent to practice independently one or more of the developed a master plan and set into motion mechanisms to subﬁelds of medical physics. The SNM considers certiﬁ- sort out the practice issues regarding PET/CT. This master cation and continuing education in the appropriate sub- plan was crafted in July 2002 during a stakeholders’ ﬁeld(s) to demonstrate that an individual is competent to meeting known as the PET/CT Consensus Conference. The practice one or more of the subﬁeld(s) of medical physics recommendations from this meeting can be found in a and to be a qualiﬁed medical physicist. The SNM recom- report of the PET/CT Consensus Conference (J Nucl Med mends that the individual be certiﬁed in the appropriate Technol. 2002;30:201–204) and are also accessible on the subﬁeld(s) by the American Board of Radiology (ABR) or SNM Web site (www.snm.org). the American Board of Science in Nuclear Medicine It is the responsibility of professional associations to (ABSNM). establish standards, delineate mechanisms for obtaining the The appropriate subﬁelds of medical physics are as training necessary to promote a qualiﬁed and competent follows: workforce to perform these procedures, and partner with d ABR: ‘‘Medical Nuclear Physics’’ with initially at organizations that can assist in sorting out practice issues. least 15 h of continuing education credits in CT To address educational needs, the ASRT and the SNMTS imaging physics, or spearheaded the development of a PET/CT curriculum, d ‘‘Diagnostic Radiological Physics’’ with initially at which was endorsed by numerous professional organiza- least 15 h of continuing education credit in PET tions and distributed to each state radiation control board imaging physics, or and every program director in the United States; it is also d ABSNM: ‘‘Nuclear Medicine Physics and Instrumen- posted on the SNM Web site (www.snm.org) and the ASRT tation’’ with initially at least 15 h of continuing Web site (www.asrt.org). education credits in CT imaging physics. A qualiﬁed medical physicist must have at least 40 h of E. Cohade C, Osman M, Nakamoto Y, et al. Initial experi- practical experience providing physics support for both the ence with oral contrast in PET/CT: phantom and PETand the CT components in an established PET/CT facility. clinical studies. J Nucl Med. 2003;44:412–416. A qualiﬁed medical physicist’s continuing education F. Coleman RE, Delbeke D, Guiberteau MJ, et al. should be in accordance with the American College of Concurrent PET/CT with an integrated imaging system: Radiology Practice Guideline for Continuing Education intersociety dialogue from the Joint Working Group of and should include at least 15 h in PET and CT physics the American College of Radiology, the Society of combined in a 3-y period. Nuclear Medicine, and the Society of Computed Body A qualiﬁed medical physicist or other qualiﬁed scientist Tomography and Magnetic Resonance. J Nucl Med. performing physics services in support of a PET/CT facility 2005;46:1225–1239. should meet all of the following criteria: G. Czernin J. PET/CT: imaging structure and function. 1. Advanced training directed at the speciﬁc area of J Nucl Med. 2004;45(suppl 1):1S–103S. responsibility (e.g., medical physics, health physics, H. Dizendorf E, Hany TF, Buck A, et al. Cause and mag- or instrumentation) nitude of the error induced by oral CT contrast agent in 2. Licensure, if required by state regulations CT-based attenuation correction of PET emission 3. Documented regular participation in continuing edu- studies. Eur J Nucl Med. 2003;44:732–738. cation in the area of speciﬁc involvement to maintain I. Donnelly LF. Lessons from history. Pediatr Radiol. competency 2002;32:287–292. 4. Knowledge of radiation safety and protection and of J. Fearon T, Vucich J. Pediatric patient exposures from CT all rules and regulations applying to the area of examinations: GE CT/T 9800 scanner. AJR. 1985; practice 144:805–809. K. Gambhir SS, Czernin J, Schimmer J, et al. A tabulated VI. ISSUES REQUIRING FURTHER CLARIFICATION summary of the 18F-FDG PET literature. J Nucl Med. 2001;42(suppl):1S–93S. A. Use of AC/AL CT, use of optimized diagnostic CT, L. Kinahan PE, Hasegawa BH, Beyer T. X-Ray based or both may depend on the indication; the selection attenuation correction for PET/CT scanners. Semin of the CT protocol to be used with PET/CT is under Nucl Med. 2003;33:166–179. evolution M. Nakamoto Y, Chin BB, Kraitchman DL, et al. Effects of B. Role of reregistration to correct for respiratory and nonionic intravenous contrast agents at PET/CT imag- other motion artifacts ing: phantom and canine studies. Radiology. 2003; C. Role of axial collimation 227:817–824. D. Role of respiratory gating for PET, CT, or both N. Osman MM, Cohade C, Nakamoto Y, Wahl RH. E. Optimal distribution time for 18F-FDG, scan time per Clinically signiﬁcant inaccurate localization of lesions bed position, image regularization (smoothing), re- with PET/CT: frequency in 300 patients. J Nucl Med. construction algorithm, and best method for CT-based 2003;44:240–243. attenuation correction O. Paquet N, Albert A, Foidart J, Hustinx R. Within- F. Optimal methods for semiquantitative measurements patient variability of 18F-FDG: standardized uptake (e.g., SUV) values in normal tissues. J Nucl Med. 2004;45:784–788. P. PET-CT Consensus Conference, SNMTS, American VII. CONCISE BIBLIOGRAPHY Society of Radiologic Technologists (ASRT). Fusion A. Antoch G, Freudenberg LS, Egelhof T, et al. Focal imaging: a new type of technologist for a new type of tracer uptake: a potential artifact in contrast-enhanced technology. J Nucl Med Technol. 2002;30:201–204. dual-modality PET/CT scans. J Nucl Med. 2002; Q. Thie J. Understanding the standardized uptake value, its 43:1339–1342. methods, and implications for usage. J Nucl Med. 2004; B. Antoch G, Freudenberg LS, Stattaus J, et al. Whole- 45:1431–1434. body positron emission tomography-CT: optimized CT R. Townsend DW, Beyer T, Blodgett TM. PET/CT scan- using oral and intravenous contrast materials. AJR. ners: a hardware approach to image fusion. Semin Nucl 2002;179:1555–1560. Med. 2003;33:193–204. C. Antoch G, Jentzen W, Freudenberg LS, et al. Effect of S. Weber WA. Use of PET for monitoring therapy and oral contrast agents on computed tomography-based predicting outcome. J Nucl Med. 2005;46:983–995. positron emission tomography attenuation correction in T. Yau YY, Chan WS, Tam YM, et al. Application of dual-modality positron emission tomography/computed intravenous contrast in PET/CT: does it really introduce tomography imaging. Invest Radiol. 2003;38:784–789. signiﬁcant attenuation correction error? J Nucl Med. D. Beyer T, Townsend DW, Brun T, et al. A combined 2005;46:283–291. PET/CT scanner for clinical oncology. J Nucl Med. U. Young H, Baum R, Cremerius U, et al. Measurement of 2000;41:1369–1379. clinical and subclinical tumour response using [18F]- ﬂuorodeoxyglucose and positron emission tomography: physician when considering the circumstances presented. review and 1999 EORTC recommendations. European Therefore, an approach that differs from the guidelines is Organization for Research and Treatment of Cancer not necessarily below the standard of care. A conscientious (EORTC) PET Study Group. Eur J Cancer. 1999; practitioner may responsibly adopt a course of action 35:1773–1782. different from that set forth in the guidelines when, in his or her reasonable judgment, that course of action is indi- VIII. DISCLAIMER cated by the condition of the patient, limitations on avail- The Society of Nuclear Medicine has written and ap- able resources, or advances in knowledge or technology proved these guidelines as an educational tool designed to subsequent to publication of the guidelines. promote the cost-effective use of high-quality nuclear All that should be expected is that the practitioner will medicine procedures in medical practice or in the conduct follow a reasonable course of action based on current of research and to assist practitioners in providing appro- knowledge, available resources, and the needs of the patient priate care for patients. These guidelines should not be to deliver effective and safe medical care. The sole purpose deemed inclusive of all proper procedures or exclusive of of these guidelines is to assist practitioners in achieving this other procedures reasonably directed to obtaining the same objective. results. They are neither inﬂexible rules nor requirements of Advances in medicine occur at a rapid rate. The date of a practice and are not intended nor should they be used to guideline always should be considered in determining its establish a legal standard of care. For these reasons, the current applicability. Society of Nuclear Medicine cautions against the use of these guidelines in litigation in which the clinical decisions IX. APPROVAL of a practitioner are called into question. These guidelines were approved by the Board of Direc- The ultimate judgment about the propriety of any spe- tors of the Society of Nuclear Medicine on February 11, ciﬁc procedure or course of action must be made by the 2006.