Elsevier

Molecular Imaging & Biology

Volume 6, Issue 4, July–August 2004, Pages 188-199
Molecular Imaging & Biology

Review Article
Positron emission tomography/computed tomography–imaging protocols, artifacts, and pitfalls

https://doi.org/10.1016/j.mibio.2004.04.006Get rights and content

Abstract

There has been a longstanding interest in fused images of anatomical information, such as that provided by computed tomography (CT) or magnetic resonance imaging (MRI) systems, with biological information obtainable by positron emission tomography (PET). The near-simultaneous data acquisition in a fixed combination of a PET and a CT scanner in a combined PET/CT imaging system minimizes spatial and temporal mismatches between the modalities by eliminating the need to move the patient in between exams. In addition, using the fast CT scan for PET attenuation correction, the duration of the examination is significantly reduced compared to standalone PET imaging with standard rod-transmission sources. The main source of artifacts arises from the use of the CT-data for scatter and attenuation correction of the PET images. Today, CT reconstruction algorithms cannot account for the presence of metal implants, such as dental fillings or prostheses, properly, thus resulting in streak artifacts, which are propagated into the PET image by the attenuation correction. The transformation of attenuation coefficients at X-ray energies to those at 511 keV works well for soft tissues, bone, and air, but again is insufficient for dense CT contrast agents, such as iodine or barium. Finally, mismatches, for example, due to uncoordinated respiration result in incorrect attenuation-corrected PET images. These artifacts, however, can be minimized or avoided prospectively by careful acquisition protocol considerations. In doubt, the uncorrected images almost always allow discrimination between true and artificial finding. PET/CT has to be integrated into the diagnostic workflow for harvesting the full potential of the new modality. In particular, the diagnostic power of both, the CT and the PET within the combination must not be underestimated. By combining multiple diagnostic studies within a single examination, significant logistic advantages can be expected if the combined PET/CT examination is to replace separate state-of-the-art PET and CT exams, thus resulting in significantly accelerated diagnostics.

Introduction

Both, positron emission tomography (PET) and computed tomography (CT) are well-established noninvasive diagnostic modalities. Their application is wide, but no doubt the most dominant application are oncologic questions. Both modalities have their independent justification. CT images morphology and the diagnosis rely on structural changes. Therefore, CT is optimized to provide high-resolution images with as much contrast as possible. CT contrast is based on differences in X-ray attenuation between different tissues and these differences in contrast aid the diagnosis of pathology. Furthermore, morphologic imaging procedures depend on specific size criteria for detection of malignancy. These size criteria, however, have been shown to be an unreliable indicator for malignancy, which is most profound when assessing lymph nodes for malignant spread, applying a threshold of one cm for the differentiation of benign (< one cm) from malignant (> one cm) disease. Thus, previous studies have found up to 21% of nodes smaller than one cm to be malignant, while 40% of those larger than one cm were demonstrated to be benign.1., 2. Furthermore, malignant lesions, which appear isodense compared to their surrounding tissue may be indistinguishable from normal anatomy and, thus, be missed on CT evaluation. In part, this problem can be overcome by the use of oral and intravenous (IV) contrast agent agents, which increase the contrast between normal and diseased tissue. Due to the lack of functional information, however, the sensitivity of CT for the detection of early disease, as well as early detection of therapeutic response of a tumor are limited. PET is in many aspects complementary. Implicitly, this notifies a large variety of radiopharmaceuticals, which are sensitive to many different cellular aspects (e.g., receptors, metabolic pass ways, and milieu-dependent uptakes) and may detect a large spectrum of tumors or tumor characteristics. Speaking in mathematical terms, PET tests the differential change (in time) of the tumor, which makes the modality more sensitive to early detection both, of the tumor itself or of therapy response. This, however, presupposes the tumor to express the characteristics tested by the applied radiopharmaceutical. If that is not the case, even large tumors may be missed—although the diagnosis “no pathologic uptake” (receptor density, turn over, and so on) is made correctly. One of the major drawbacks of PET is the inability to determine the precise location of the lesion with respect to anatomical structures, which may result in the uncertainty of the tumor-bearing organ.

The combination of PET and CT images has several attractive aspects. The combination of two complementary modalities significantly increases the diagnostic accuracy compared to each of the two modalities, as well as the two imaging modalities viewed side-by-side.3., 4., 5., 6., 7. The number of inconclusive PET findings will be reduced by accurate identification of the site of the activity accumulation. This finding may be due to pathologically increased (tumor) turnover, pathologically increased turnover in a nonmalignant process (e.g., inflammation, thyroid nodule), or increased but physiological uptake in an activated organ (e.g., fatty tissue, muscle, endocrine gland).

Section snippets

PET/CT application concepts

We define a PET/CT as the hardware-based combination of a PET and a CT scanner in which PET attenuation and scatter corrections are performed using the CT data. The development of PET/CT with rod sources is expected to be abundant in the future. PET/CT may be used in a variety of applications, which are based on different philosophies concerning the use of the combined modalities and which may require different imaging protocols. PET/CT may be used as a PET scanner (faster-PET) with built-in

Clinical protocols

The full diagnostic power of combined PET/CT is gained only, by avoiding separate CT investigations prior to the PET examination, which, in turn, requires additional education of the referring physicians. Figure 1 displays a decision tree for PET/CT indications at our hospital. By following this scheme we aim at avoiding unnecessary radiation exposure of the patient and costs from repeated CT investigations. The nuclear medicine physician makes the final indication for the PET, whereas the

General patient preparation before initiation of PET

The patient preparation for PET/CT studies is the same as for stand-alone PET. There is general agreement that patients preferably should be fasted overnight or at least for six hours prior to the injection of the 2-deoxy-2-[18F]fluoro-D-glucose (FDG). Diabetic patients should be in a stable state of disease management (dietary, oral antidiabetic medication, or insulin), nevertheless, a reduced sensitivity for the detection of malignant disease in diabetics has been reported; insulin injections

Patient preparation prior to the CT scan

No long-term preparation is necessary for a routine oncology CT examination, however, contraindications for the injections of IV contrast agents should be ruled out beforehand. Depending on the severity, thyroid disorder, such as hyperthyroidism or autonomous functioning thyroid tissue, may require premedication, that is, blocking of the thyroid gland with perchlorate, which competes with the iodine uptake into the thyroid cells, and/or thioamide derivatives, which block the intracellular

Patient positioning

Prior to the exam patients remove all metal (e.g., bracelets, dental braces, belt buckles, pants with zippers, etc.), which may lead to artifacts on the CT-transmission scan. On the scanner table patients should be supported with adequate positioning aids, that is, a roll to support the knees, an arm rest for all studies with the arms positioned over the head, and a head rest (Figure 3). Nevertheless, especially PET/CT studies involving the head and neck frequently suffer from local

CT beam hardening

In standalone PET imaging, the scan is usually performed with the patient's arms lying along the side of the body. This ensures whole-body imaging without increasing the axial scan length. Having the arms in the field-of-view (FOV) during a CT scan, however, causes serious beam hardening artifacts on the CT images due to the preferential absorption of the lower energetic X-ray components of the polychromatic X-ray beam, as well as scatter build-up (Figure 4). The resulting image quality is not

Truncation

Obese patients, as well as patients with their arms down may extend outside the transverse FOV of the CT scanner (50-cm diameter in commercial PET/CT systems), resulting in inconsistent CT projection data, which cause truncation artifacts in the CT images. Furthermore, the transverse FOV of the PET scanner is larger (about 60 cm) than that of the CT scanner, resulting in missing data for the CT-based attenuation correction. The resulting discrepancy in imaging FOV results in artifacts of the

External radiation therapy planning

PET/CT imaging provides molecular information about a tumor in the spatial coordinate system utilized by the radiation treatment planning system. Recently, great hope has been expressed in the radiation oncology community that the resulting biological conformality may improve tumor control by radiation therapy.12 Prior to using PET/CT images in the radiation therapy planning process, the data transfer between the diagnostic imaging device and the treatment planning system must be validated and

Absolute quantification of positron emitter concentrations in the PET study

PET imaging, which originated as a research tool, always had the potential for accurate in vivo quantitation of the activity distribution of a positron emitting radiopharmaceutical. Absolute quantitation requires accurate scatter and attenuation correction, and, for objects which are small compared to the system resolution, a size dependent recovery correction.13., 14.

Scatter and attenuation correction require an accurate attenuation map at a photon energy of 511 keV. Therefore, the use of the

IV Contrast

The acquisition of a long CT spiral in oncology PET/CT, which typically covers the body from the head to the symphysis, requires a contrast injection protocol, which is modified from the one used for the more familiar shorter spirals in single organ CT studies. For an optimization of the flow rate and the delay between injection and initiation of the CT scan the duration of the CT acquisition over the entire scan range, the direction (cranio-caudal versus caudo-cranial), and the desired peak

CT acquisition and reconstruction parameters

The acquisition and reconstruction protocols depend in detail on the specific hardware and software used. We acquire all whole-body studies or partial studies of the torso with 5-mm slice thickness, 8-mm table feed per rotation, and a slice spacing of 2.4 mm with tube settings of 130 kVp and an effective tube current of 140 mAs, which is limited by the tube heat capacity in case of extended scan ranges. The CT images are reconstructed with a medium sharp filter and displayed in a soft tissue

PET-acquisition protocol and image reconstruction

In adult patients of nominal body weight (75 kg), we inject 350 MBq FDG one hour prior to the PET emission scan. During the uptake period the patients are kept relaxed in a comfortable chair in a softly lit room. Before being placed on the scanner bed, all patients must void. The typical acquisition time per bed position in three-dimensional mode on our system—based on the ECAT EXACT HR+ PET scanner—is three to five minutes with the shorter acquisition time being reserved for patients under 65

PET/CT mismatches despite simultaneous acquisition

Mismatches of breathing patterns in combined PET/CT examinations have been described as a source of artifacts in PET images following CT-based attenuation correction.20., 21. These artifacts are most severe, for example, when the CT scan is acquired during breath hold at maximum inspiration, that is, with the typical protocol for a standalone CT scan of the thorax. They are caused by the mismatch of the anatomy of the thoracic and abdominal organs at maximum inspiration versus the anatomy when

Metal artifacts

Oncology patients frequently present with metal implants, such as chemotherapy ports, metal braces in the spine, artificial joints, or dental fillings. Unlike in standard PET transmission scanning, where metal implants cause little or no artifacts, they are often severe at CT energies (Figure 9). Just as for iodine-based contrast agents, this is due to the significantly higher photon absorption from high-Z materials (e.g., metals) compared to low-Z materials (e.g., soft tissues) at X-ray CT

PET/CT reporting

The primary idea of PET/CT seems to be the transformation of the PET—functional information into the three-dimensional spatial coordinate system of CT. The second and attractive step is obviously to identify the PET finding with morphologic structures—normal or pathologic. Finally a joint report is necessary, which is an easy task in case of concordant findings. However, there is no accepted rule for dealing with contradictory PET and CT findings. We come to our conclusion depending on

Conclusion

PET/CT combines the diagnostic power of PET and CT. The appropriate use of this new modality creates synergistic effects. Consequently, PET/CT is widely applied already shortly after its clinical introduction. The sources for artifacts and pitfalls have been identified. They will be reduced or avoided by ongoing technical developments as well as by the skillful use of the combined tomographs. The more integrated the use of the two modalities, the more dependent become their protocols from each

Acknowledgements

The clinical success and the generation of optimized clinical protocols at the University Hospital Essen is largely based on the contributions by our technologists Sandra Heistrüvers, Gianina Marchese, Slavko Maric, Sandra Pabst, Dorothea Porsch-Plotek, Lydia Schostock, and Bärbel Terschüren, as well as by the involved physicians Drs Thomas Egelhof, Roman Pink, and Sandra Rosenbaum.

References (24)

  • C.C. Ling et al.

    Towards multidimensional radiotherapy (MD-CRT): biological imaging and biological conformality

    Int. J. Radiat. Oncol. Biol. Phys.

    (2000)
  • J. Deslauriers et al.

    Clinical and surgical staging of non-small cell lung cancer

    Chest

    (2000)
  • C. Staples et al.

    Mediastinal nodes in bronchogenic carcinoma: comparison between CT and mediastinoscopy

    Radiology

    (1998)
  • P.G. Kluetz et al.

    Combined PET/CT imaging in oncology: Impact on patient management

    Clin. Posit. Imag.

    (2001)
  • T.F. Hany et al.

    PET diagnostic accuracy: improvement with in-line PET-CT System

    Initial Results. Radiology

    (2002)
  • R. Bar-Shalom et al.

    Clinical performance of PET/CT in evaluation of cancer: additional value for diagnostic imaging and patient management

    J. Nucl. Med.

    (2003)
  • D. Lardinois et al.

    Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography

    N. Engl. J. Med.

    (2003)
  • G. Antoch et al.

    Non-small cell lung cancer: dual-modality PET/CT imaging in preoperative staging

    Radiology

    (2003)
  • G. Antoch et al.

    Effect of oral contrast agents on computed tomography-based positron emission tomography attenuation correction in dual-modality positron emission tomography/computed tomography imaging

    Invest. Radiol.

    (2003)
  • G. Antoch et al.

    Introduction and evaluation of a negative oral contrast agent to avoid contrast-induced artifacts in dual-modality PET/CT imaging

    Radiology

    (2004)
  • T. Beyer et al.

    Acquisition protocol considerations for combined PET/CT imaging

    J. Nucl. Med.

    (2004)
  • T. Beyer

    Design; construction; and validation of a combined PET/CT tomograph for clinical oncology, in Department of Physics

    (2000)
  • Cited by (128)

    • Pitfalls on PET/CT Due to Artifacts and Instrumentation

      2021, Seminars in Nuclear Medicine
      Citation Excerpt :

      Moreover, the performance of the networks could depend on the noise level of the images used for training and vary with applications.61 Depending on the role of CT in PET data processing, different artifacts could emerge62-64 Most artifacts observed in PET/CT imaging can be attributed to the misalignment between PET and CT data, errors in the CT-derived attenuation coefficients or CT image truncation. As the PET and CT data are usually obtained sequentially in practice, misalignment between these two datasets could happen due to motion of the subject between the acquisitions, and/or within either/both of the acquisitions.

    • SPECT and SPECT/CT

      2021, Molecular Imaging: Principles and Practice
    View all citing articles on Scopus
    View full text