| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Massachusetts General Hospital, FND 202, 55 Fruit
St., Boston, MA 02114.
2 Present address: Department of Radiology, University of Cincinnati Medical
Center, Cincinnati, OH.
3 Institute for Technology Assessment, Massachusetts General Hospital, Boston,
MA.
Received August 15, 2005;
accepted after revision October 7, 2005.
Address correspondence to S. L. Aquino
(saquino{at}partners.org).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Eighty combined PET/CT scans were obtained in 64 patients (30 women, 34 men; mean age, 57 years; range, 19-86 years). The 80 PET/CT scans consisted of five group of patients (16 PET/CT scans per group) who underwent whole-body combined 18F-FDG PET/CT with different CT breathing protocols: expiration, mid suspended breath-hold, quiet breathing, small breath in, and regular breath in. The quality of alignment was analyzed at the diaphragm, aortic arch, heart, thoracic spine, and lung apices using a scale of ratings from 1 (very poor) to 5 (excellent). The Kruskal-Wallis test was used to compare alignment between breathing protocols for each anatomic reference point.
RESULTS. Alignment of the PET and CT data sets was excellent with three breathing protocols: expiration, mid suspended breath-hold, and quiet breathing, with no statistical differences. Significant misalignment occurred at the diaphragm (p < 0.0001) and heart (p < 0.0001) with the small breath-in and regular breath-in techniques.
CONCLUSION. Excellent image fusion of combined PET/CT data sets in the thorax, especially at the diaphragm and heart, can be achieved with expiration, mid suspended breath-hold, or quiet breathing. Quiet breathing is recommended for optimal patient comfort during acquisition of attenuation-correction CT data sets.
Keywords: computer-assisted • image analysis • CT • image reconstruction • positron emission • PET/CT
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
For optimal image fusion with PET/CT, the lung volumes should be well matched during emission PET and transmission CT acquisition. This alignment will prevent attenuation-correction errors in the lower lungs, peridiaphragmatic regions, and upper abdomen. The emission PET scan is acquired over approximately 30 minutes during quiet respiration. Transmission attenuation-correction CT data sets, however, are acquired in seconds, during which the patient is instructed to breathe quietly or to perform a breath-hold maneuver.
Goerres et al. [5, 6] evaluated the impact of different breathing maneuvers during attenuation-correction CT on anatomic alignment of combined PET/CT scans. They evaluated free breathing, maximum inspiration, normal expiration, and maximum expiration. They found that optimal image fusion occurred when CT data sets were acquired during normal expiration. Similar studies also have shown a reduction in the severity and frequency of respiratory motion artifacts when a limited expiratory breath-hold protocol was used [7, 8].
Sustained expiratory breath-hold during whole-body CT scan acquisition may be difficult in elderly patients or in those with underlying lung disease. In such instances, significant respiratory motion artifact may result if a patient resumes respiration (which can be heavy because of compensation) during the latter part of the CT scan. Misalignment may occur in the lower lungs, diaphragm, and as a result, the upper abdomen. Therefore, additional breathing maneuvers such as suspended breath-hold or quiet breathing may be more appropriate.
None of these maneuvers—expiration, suspended breath-hold, or quiet breathing—is optimal for diagnostic imaging of the lungs. Inspiration is necessary to overcome atelectasis and fully display the lung parenchyma. To our knowledge, there has been no prior study that has comprehensively evaluated this full spectrum of breathing maneuvers. Therefore, the purpose of this study was to determine which of the following breathing protocols provides the best anatomic fusion: expiration, mid suspended breath-hold, quiet breathing, small breath in, or regular breath in.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Eighty PET/CT data sets from these 64 patients were analyzed. The 80 PET/CT data sets consisted of five groups of 16 PET/CT scans. In each group (n = 16 patients), the CT scans were acquired using a different breathing protocol—normal expiration, mid suspended breath-hold, quiet breathing, small breath in, or regular breath in. The patient groups, breathing protocol, and numbers of patients are summarized in Table 1.
|
The study groups were derived from clinical whole-body PET/CT scans obtained from March 2004 through November 2004. During this period, our institution's combined PET/CT protocol consisted of a whole-body attenuation-correction CT scan and a subsequent contrast-enhanced diagnostic CT scan of the chest and upper abdomen during the same imaging session. From March through November 2004, breathing protocols were modified during attenuation-correction CT acquisition to optimize image fusion and patient comfort. Breathing instructions evolved from expiration (group 1), to suspended breath-hold (group 2), to our currently used quiet-breathing protocol (group 3). For the subsequent diagnostic contrast-enhanced CT scans obtained during the same imaging session, we instructed our patients to take a small breath in (group 4) not only to optimize lung imaging but also to avoid significant mismatch in the lower thorax and upper abdomen. The patients who did not receive IV contrast material (due to allergy or renal insufficiency) received a single whole-body CT scan that functioned both as the attenuation-correction CT and diagnostic CT scan. In this group, an additional CT scan of the thorax (from the lung apices through the upper abdomen) was acquired at full breath in (group 5) to comprehensively image the lung parenchyma.
Combined PET/CT Imaging
Combined FDG PET/CT was performed on a 16-MDCT scanner (Biograph Sensation
16, Siemens Molecular Imaging). This system consists of an in-line combined
PET/CT system with a lutetium oxyorthosilicate (LSO) detector and a 16-MDCT
unit (Somatom Sensation, Siemens Medical Solutions). Approximately 555 MBq (15
mCi) of 18F-FDG was administered IV as a bolus, and imaging was
performed 60 minutes later.
Unenhanced attenuation-correction CT was performed from the skull base through the upper thighs with the arms positioned overhead. Each attenuation-correction CT scan was obtained with the following parameters: variable milliampere settings, 0.5 seconds per tube rotation, 120-kV tube voltage, helical pitch of 1.5, reconstructed slice thickness of 5 mm, and 16 simultaneous slice acquisitions. The average scan duration was 15 seconds. The attenuation-correction CT data sets were obtained with one of three breathing protocols: expiration (group 1, n = 16), mid suspended breath-hold (group 2, n = 16), or quiet breathing (group 3, n = 16).
The PET emission scan data sets were acquired from the mid thigh level to the skull base. The spatial resolution of the PET scanner was 4.6 mm full width at half maximum (FWHM) with a slice thickness of 3.5 mm.
For all patients, a subsequent diagnostic CT scan was obtained during the same imaging session (without moving the patient) with the following CT parameters: variable milliampere settings, 0.5 seconds per tube rotation, 140-kV tube voltage, helical pitch of 1.5, reconstructed slice thickness of 2 mm, and 16 simultaneous slice acquisitions. The average scan duration was 20 seconds. IV contrast material (ioxilan 300 mg I/mL) was administered at an injection rate 2 mL/s to all patients except the 16 patients in group 5 who had contraindications to contrast material (allergy or renal insufficiency). The diagnostic CT scan was obtained with a small breath in (group 4, n = 16) or regular breath in (group 5, n = 16).
For all CT scans, breathing instructions were given and rehearsed with the patient before image acquisition. For the expiratory breath-hold technique, each patient was asked to breath normally, release a small breath out, and then stop breathing. For the mid suspended breath-hold technique, each patient was asked to stop in the middle of a normal breath whether during inspiration or expiration. For the quiet-breathing technique, each patient was asked to continuously breathe quietly in small breaths with no large breaths in or out. For the small breath-in technique, each patient was asked to take a small breath in and hold it. For the regular breath-in protocol, each patient was instructed to take a deep breath in and hold it.
Image Reconstruction and Analysis
Attenuation correction was system-generated. PET image reconstruction was
performed with Fourier rebinning (FORE) and attenuation-weighted ordered
subsets expectation maximization (AW-OS-EM). All PET/CT data sets were
displayed on a workstation (Reveal-MVS, Mirada Solutions Ltd.) as stand-alone
or superimposed data sets of axial, coronal, and sagittal reformatted images.
Image registration using software was not performed.
Combined PET/CT data sets were analyzed for quality of alignment at five anatomic locations: the diaphragm, aortic arch, left ventricle contour, thoracic spine, and lung apices. The quality of alignment was rated on a scale of 1 (very poor or complete lack of superimposition of the anatomic structure on axial, coronal, and sagittal views) to 5 (excellent or complete superimposition of the anatomic structure on axial, coronal, and sagittal views). Image analysis and scoring were performed by two radiologists experienced in PET, CT, and combined PET/CT image interpretation. All disputes were handled by consensus. The data sets were evaluated in a random order, and the reviewers were blinded to the breathing maneuver performed.
Statistical Analysis
The Kruskal-Wallis test was used to compare alignment between breathing
protocols for each anatomic reference point. If the test was not significant,
no further analysis was performed. If the test was significant, the most
discrepant group was removed and the Kruskal-Wallis test was applied to the
remaining groups until the test was not significant. A p value of
less than 0.05 was considered to be statistically significant. The mean scores
for the quality of the alignment were calculated at the five anatomic
locations to compare alignment among breathing protocols.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
Excellent thoracic anatomic alignment was also found at the aortic arch, spine, and lung apices using the small breath-in and regular breath-in breathing methods. The mean scores ranged from 4.81 to 5. However, significant misalignment of anatomic structures occurred at the diaphragm and heart (Fig. 2). For the small breath-in technique, the mean score at the diaphragm was 4.06 and at the heart was 4.19. For the regular breath-in technique, the mean score at the diaphragm was 3 and at the heart was 3.44.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A challenge that we encountered in our practice in using expiratory transmission CT was patient compliance and comfort. Although we found that expiratory data sets provided excellent image alignment, many patients were incapable of maintaining the expiratory breath-hold for the duration of the whole-body CT examination. This resulted in greater compensatory breathing artifacts at the mid scan level (the lower thorax to the upper abdomen) when the patient resumed breathing. We therefore evaluated alternative methods, including mid suspended breath-hold, small breath in, and quiet breathing, to determine whether other methods could provide adequate fusion of PET and CT data sets and maintain patient compliance. We also evaluated fusion of PET with the full inspiratory CT images (group 5) that were obtained during the same scanning session.
Our results show that anatomic alignment was similar whether patients had attenuation-correction CT acquired during expiration, mid suspended breath-hold, or quiet breathing. All three methods produced excellent data set superimposition, with no significant difference in scoring. Inspiratory maneuvers, such as small breath in and regular breath in, however, led to significant misalignment of PET and CT data sets at the heart and diaphragm. Although previous studies reported that shallow expiration provides optimal image coregistration over quiet breathing, our data indicate that expiration, quiet breathing, and mid suspended breath-hold protocols provide equally accurate fusion.
Quiet breathing, suspended breath-hold, and expiration during CT acquisition, however, preclude the acquisition of a truly diagnostic-quality CT scan. Inspiratory CT is important for adequate imaging of most of the thorax and upper abdomen, particularly the lung parenchyma. Significant lung disease, including nodules, interstitial lung disease, and pneumonia, may be masked by atelectasis and motion artifacts. However, the degree of misalignment of the diaphragm and heart between PET and CT during both small breath-in and regular breath-in protocols proves that neither method is appropriate for attenuation-correction CT.
In conclusion, for combined PET/CT of the thorax, excellent image fusion was achieved during expiration, mid suspended breath-hold, and quiet breathing with no significant variation in scoring of anatomic alignment. Inspiratory maneuvers, however, resulted in significant thoracic misalignment, primarily at the heart and diaphragm.
We recommend quiet breathing during attenuation-correction CT acquisition to achieve optimal image fusion and consistent patient compliance. However, for optimal image quality and diagnostic evaluation of the lung parenchyma, an additional diagnostic CT at full inspiration during the same session is recommended.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Kligerman and S. Digumarthy Staging of Non-Small Cell Lung Cancer Using Integrated PET/CT Am. J. Roentgenol., November 1, 2009; 193(5): 1203 - 1211. [Full Text] [PDF]
|
|
|
|
 |
|
 W. De Wever, S. Stroobants, J. Coolen, and J. A. Verschakelen Integrated PET/CT in the staging of nonsmall cell lung cancer: technical aspects and clinical integration Eur. Respir. J., January 1, 2009; 33(1): 201 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Slomka, L. Le Meunier, S. W. Hayes, W. Acampa, M. Oba, G. G. Haemer, D. S. Berman, and G. Germano Comparison of Myocardial Perfusion 82Rb PET Performed with CT- and Transmission CT-Based Attenuation Correction J. Nucl. Med., December 1, 2008; 49(12): 1992 - 1998. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kawano, E. Ohtake, and T. Inoue Deep-Inspiration Breath-Hold PET/CT of Lung Cancer: Maximum Standardized Uptake Value Analysis of 108 Patients J. Nucl. Med., August 1, 2008; 49(8): 1223 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Gould, T. Pan, C. Loghin, N. P. Johnson, and S. Sdringola Reducing Radiation Dose in Rest-Stress Cardiac PET/CT by Single Poststress Cine CT for Attenuation Correction: Quantitative Validation J. Nucl. Med., May 1, 2008; 49(5): 738 - 745. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Gould, T. Pan, C. Loghin, N. P. Johnson, A. Guha, and S. Sdringola Frequent Diagnostic Errors in Cardiac PET/CT Due to Misregistration of CT Attenuation and Emission PET Images: A Definitive Analysis of Causes, Consequences, and Corrections J. Nucl. Med., July 1, 2007; 48(7): 1112 - 1121. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
