PET body oncology scans have always been interpreted, to varying degrees, in conjunction with CT scans or other anatomic imaging. Now with the advent of PET-CT scanners, the merging of metabolic and anatomic diagnosis has become explicit and routine. Adding PET-CT to a practice involves merging several aspects of body FDG PET and body CT practice.


Using metabolic imaging and anatomic imaging in combination is neither a new nor a difficult concept. Accordingly, performing high-quality FDG PET and diagnostic body CT scans together on a PET-CT scanner should not be a difficult concept. Many centers still perform PET-CT using a reduced beam current noncontrast CT scan for attenuation correction and anatomic localization. Unless the CT tube beam current is very low, however, CT scans depict important anatomic diagnostic features, which must be recognized and incorporated into the final imaging diagnosis. Even scans performed without contrast material and with beam currents of 60 to 100mA, as typically used for anatomic localization, will on contemporary CT scanners contain significant diagnostic information beyond anatomic localization, including findings unrelated to abnormalities on the PET portion of the examination. Hence, the CT portion of a PET-CT examination is not performed for anatomic localization only; the CT provides both anatomic localization and anatomic diagnostic information. Further, due to improvements in PET image quality, cancer lesions well under 1 cm are now routinely detected. This has raised the bar for the CT image quality suitable for the highly detailed anatomic localization requirements increasingly encountered in PET-CT oncology imaging. PET-CT, then, involves the performance and interpretation of a FDG PET imaging examination and a diagnostic CT imaging examination, both optimized for oncology diagnosis.


Surprisingly, the CT protocols, which have been well developed over the years specifically for oncology diagnosis in the chest, abdomen, and pelvis, and in the head and neck, as of yet have not been widely adapted in the setting of PET-CT. In some institutions the use of an optimized CT protocol with contrast material is resisted as if to do so would breach some forbidden threshold. In many centers the mistaken notion that a contrast-enhanced CT cannot be used for attenuation correction of the PET emission scan persists, resulting in subjecting patients to two CT scans during a PET-CT examination; a low beam current noncontrast “non-diagnostic CT” for attenuation correction and anatomic localization followed, after the PET acquisition, by a “full diagnostic CT” including the use of contrast material. Both of these CT scans provide anatomic localization and anatomic diagnostic information; they are in fact both diagnostic CT scans. The later scan, however, is optimized for both anatomic diagnosis and anatomic localization. Hyperconcentrated positive oral contrast in stomach and bowel and undiluted intravenous contrast material in central veins can cause spurious apparent increased tracer activity on the attenuation corrected PET images when the contrast-enhanced CT scan is used for attenuation correction; however, this is not common or typically confounding.1 In our practice, using contemporary multi-detector CT protocols involving oral and intravenous contrast material, no perceptible artifacts have been encountered on the attenuation corrected PET images when the contrast-enhanced CT is used for attenuation correction. Recent published findings similarly found no artifacts or significant changes in standardized uptake value measurements when the contrast-enhanced CT was used for attenuation correction.2 Established CT protocols are thus relatively easily adapted to PET-CT with a single CT acquisition. Positive oral contrast can be given during the FDG uptake period of 1-2 hours before the scan acquisition with good opacification of the entire small bowel and no colon hyperconcentration. Negative (water density) oral contrast material provides similar complete small bowel opacification with no potential for attenuation correction artifacts from colon hyperconcentration. In our practice intravenous contrast infusion rates of 3 mL/sec resulted in no perceptible artifacts in both head and neck and whole body protocols when routine saline chase contrast infusion technique was used. As with any scan performed on contemporary multi-detector CT scanners, special attention must be given to contrast bolus timing and length, particularly for the neck to hips whole body scan protocols used for most cancer indications (see Figure below). Full optimization of chest and abdomen CT scan involves breath holding to avoid respiratory movement artifact, and this can lead to misregistration of PET and CT images if the position of the diaphragm and chest expansion on the CT do not match the averaged position of these structures on the PET images. Best results have been obtained with partial expiratory breath holds at end tidal volume, although in practice consistent results can be difficult to obtain. With a very fast CT scanner (16 or more detectors with gantry rotation periods of 0.5 second), respiratory artifact on quiet breathing during CT scan acquisition is relatively modest in most patients.

Figure. PET-CT performed for staging lymphoma using single optimized diagnostic CT scan followed by the PET acquisition. Arterial and venous contrast opacification is relatively uniform over entire length of torso.


Requirements for technologists staffing a PET-CT facility vary depending on state regulations, institutional standards, or third-party payor requirements. Ideally, technologists operating PET-CT scanners should have background in both nuclear medicine and CT, and to this end a consensus regarding a curriculum for PET-CT technologists has been established by a joint working group of the American Registry of Radiologic Technologists and the Society of Nuclear Medicine Technologists Section, and pathways for dual registry are being established. In our practice nuclear medicine technologists underwent special training in CT prior to initiation of services and are to additionally achieve CT registry through the newly established pathways. Similarly, the training and qualifications for physicians supervising and interpreting PET-CT scans ideally would include competencies in both nuclear medicine/PET and body CT imaging. To this end, a Joint Working Group of the American College of Radiology, Society of Nuclear Medicine, and Society of Body Computed Tomography-MR have developed a collaborative white paper (soon to be published) on PET-CT outlining a proposed curriculum and training requirements for physicians supervising and interpreting PET-CT examinations. In our practice PET-CT examinations are interpreted only by radiologists with training and experience in both CT and nuclear medicine.

Storage (archiving) of PET-CT scans is dominated by the much larger CT image files, which can be up to 10 times larger than the PET image files. Due to the typical extensive axial coverage of a whole torso scan, uncompressed CT files from a whole body PET-CT scan can exceed 100 MB. Conventional nuclear medicine archives are promptly overwhelmed by the size of PET-CT files. When a PACS archive is available, such image files are relatively easy to accommodate. PET-CT is currently best displayed for interpretation on dedicated PET-CT workstations. These currently have their limitations stemming primarily from inadequate capability for rapid CT image display, although this limitation is being solved by incorporation of increasingly more powerful workstation hardware.

Current PACS displays have limited versatility in the display of the registered and aligned PET and CT images and the maximum intensity projection PET images, but the need for improved PACS display capability of PET-CT is now recognized among PACS vendors.


Referring physicians and their office staff need to understand the combined nature of PET-CT and how to properly order the scans. The third-party payors also need to understand that a PET-CT is a PET scan and a diagnostic CT scan (not a “fusion PET scan”), each performed under specific circumstances and indications, albeit in rapid succession using an integrated imaging system. It is important that the ordering physician separately specify the PET examination and the needed CT examinations with appropriate history to document medical necessity for all studies. If a CT scan has been performed shortly before the PET-CT examination (within 4-6 weeks for most cancers), it may not be appropriate to bill for a full CT examination. Billing under a limited service/follow-up code may be appropriate in some such circumstances as there is still (or at least should be) a CT interpretation and in effect reinterpretation of the recent prior CT under these circumstances. It should be noted that new PET/CT CPT codes describe the use of the CT for anatomic localization only, which, as noted above, would be a substandard performance and interpretation of a PET-CT examination in most instances. When billing under specific CT codes is anticipated, a separate report or self-contained section in a combined report is required for each CT examination (chest, abdomen, etc). PET-CT interpretation and reporting should always include pertinent anatomic diagnostic findings from the CT portion of the examination, with a fully integrated impression. The goal is to bring the full capabilities of both the metabolic and the anatomic imaging modalities together.

Paul Shreve, MD, is director, PET Medical Imaging Center, and a staff radiologist with Advanced Radiology Services, PC, Grand Rapids, Mich. This article has been excerpted from Improving Your PET Practice: Lessons Learned in Clinical Practice: PET-CT, which he presented at the 90th scientific assembly and annual meeting of the Radiological Society of North America on December 1, in Chicago.


  1. Antoch G, Freudenberg LS, Beyer T, et al. To enhance or not to enhance? F-18 FDG and CT contrast agents in dual-modality F-18 FDG PET-CT. J Nucl Med. 2004; 45:56S-65S.
  2. Yau YY, Chan W-S, Tam Y-M, et al. Application of intravenous contrast in PET/CT: Does it really introduce significant attenuation correction error? J Nucl Med. 2005; 46:283-291.