| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Tuesday, May 8, 1:30 PM–3:30 PM
Abstracts 153-163
Moderator(s): Hani Abujudeh
1:30 PM
Keynote Address: Imaging Features of Lung Cancers on PET and CT after Treatment with Radiofrequency Ablation
Don Yoo, Rhode Island Hospital, East Greenwich, RI
1:40 PM
153. FDG PET/CT Findings of Muscular Involvement of Lymphoma
Krishnamoorthy S. K.*; Chauhan A.; Froelich J.; Dhurairaj T. University of Minnesota, Minneapolis, MN
Address correspondence to S. Krishnamoorthy (krish020{at}umn.edu)
Objective: Muscular involvement by lymphoma occurs in 1.4% of malignant lymphomas. We present the PET/CT findings of muscular lymphoma in a series of patients. To our knowledge such work has not been previously reported.
Materials and Methods: From April 2005 to August 2006, our institution used PET/CT to image 305 patients with lymphoma, of which 77 patients had extranodal involvement. Twelve of these patients (47.8 years old ± 13.8 years) had muscular involvement in which 10 were biopsy proven and 2 were proven by imaging followup, for a total of 37 separate FDG PET/CT studies. Several patients had MR correlation. Eighty-four percent and 86% of the CT portions had intravenous and oral contrasts, respectively. Total PET scan time was 60 to 90 minutes. Standard uptake values (SUV) were calculated.
Results: PET/CT was indicated for primary (pretreatment) staging in 7 of the 12 patients, with all 12 undergoing followup imaging. The histological subtypes of lymphoma were: 7 patients with diffuse B cell, 2 with post transplant lymphoproliferative disorder, 1 with anaplastic large cell, 1 with T cell, and 1 patient with mixed Hodgkin's. Six patients had predominantly muscular lymphoma with or without bone involvement; 3 patients had nodal lymphoma and developed muscular involvement on subsequent imaging; 2 patients had hepatic metastases; 1 patient had an extensive malignancy that involved the face and underlying structures. Twelve true-positive lesions, by biopsy or follow-up imaging, demonstrated PET abnormalities that were CT occult. In one case, a lymph node that was persistently enlarged on CT remained negative on all PET studies.
Conclusion: PET/CT is better for early detection of muscular lymphoma than CT alone. In addition response to treatment can be assessed by PET/CT. Although MR is sensitive for identifying tumor involvement in muscle, PET/CT allows assessment of the whole body. Additional work is needed to prospectively confirm these findings.
154. Significant Incidental Abnormalities on CT Portion of PET/CT Examinations
Schaaf W. E.*; Johnson L. S. Eastern Virginia Medical School, Norfolk, VA
Address correspondence to W. Schaaf (weschaaf{at}gmail.com)
Objective: Determine the frequency of significant incidental CT findings on PET/CT. Assess the impact of oral contrast on these CT findings.
Materials and Methods: PET/CT reports by a nuclear radiologist of 345 consecutive patients who underwent standard "eyes to thighs" PET/CT exams were retrospectively reviewed. CTs were acquired immediately before FDG PET images. Oral CT contrast was instituted during this period and was given to 162 patients. Incidental CT findings were tabulated and assigned a level of significance on a scale of 1 to 5. CT findings which corresponded to PET abnormalities, or which were already known from prior CT, were NOT counted. A score of 1 represented "doubtful significance," such as a renal cyst; 2 "questionable significance," such as IVC filter; 3 "possibly significant," such as vertebral compression deformity; 4 "probably significant," such as abdominal aortic aneurysm (AAA) <5 cm; while a score of 5 represented a "significant" finding, such as AAA >5 cm. A score of 3 or above constitutes a finding which a radiologist would be expected to report.
Results: 171 patients (50%) had at least 1 CT finding rated 3 or higher. 96 patients (28%) had at least 1 CT finding rated 4 or 5, and 25 patients (7%) had at least 1 CT finding rated 5. Studies with versus without oral CT contrast had the same number of incidental CT findings. 302 patients (88%) had no prior neck CT. Forty-six patients (13%) had no prior CT of the chest, while 162 (47%) and 175 (51%) patients had no prior CT of the abdomen or pelvis, respectively.
Conclusion: This work demonstrates that a substantial percentage of PET/CT studies contain noteworthy incidental CT findings, abnormalities being identified on CT for the first time and not evident on PET. We find that a substantial percentage of PET/CT patients have had no prior CT of one or more body regions. Thus, these CT images must be reviewed carefully by a qualified CT reader. In 2006 Medicare adds $25.74 to the professional fee for reading a standard PET/CT (including CT neck/chest/abdomen/pelvis) compared to reading a standard PET with no CT ($99.07 for PET, vs. $124.81 for PET/CT). Ironically, the entire PET/CT professional fee is far less than the fee for reading noncontrast CT of the neck/chest/abdomen/pelvis with no PET ($238.71 from Medicare). It requires considerable time and skill to read multiple CT studies per PET/CT exam, including identifying and characterizing the incidental CT abnormalities documented here. Professional reimbursement for PET/CT is markedly inequitable.
155. Impact of Lymphotrophic Nanoparticle-enhanced MRI (LNMRI) on CT-guided Percutaneous Lymph Node Biopsy
Saksena M.*; Gervais D. A.; Dawson S.; Hahn P.; Katkar A. S.; Harisinghani M. G. Massachusetts General Hospital, Boston, MA
Address correspondence to M. Saksena (mansisaksena{at}gmail.com)
Objective: The purpose of the study was to assess the impact of prebiopsy LNMRI on percutaneous lymph node biopsy in patients with pelvic malignancies.
Materials and Methods: 38 patients; M:F = 34:4; with prostate (22) and bladder (16) cancer underwent contrast enhanced diagnostic CT and LNMRI prior to percutaneous node biopsy and were included in the study. Mean time duration between diagnostic CT and biopsy was 10 days (4–20 days). Two radiologists retrospectively reviewed prebiopsy diagnostic CT scan to determine the lymph node most suitable for percutaneous biopsy. Reader 1 was an interventional radiologist with greater than 25 years of experience while reader 2 was an interventional radiologist with 10 years of experience. Readers were unaware of primary malignancy and LNMRI results. Percutaneous biopsy and LNMRI were then reviewed by study staff to assess if LNMRI results changed choice of the biopsied node.
Results: A total of 218 nodes were assessed in 38 patients with mean of 5 nodes per patient (mean size 3–55 mm; mean size 24 mm). Based on LNMRI results the node biopsied was changed in case of reader 1 in 5 patients (13%). These patients had both benign and malignant nodes with malignant node identified on LNMRI. Reader 1 would not have biopsied any node in 1 patient with multiple abnormal nodes less then 5 mm in size. In 5 patients the node was changed due to reasons other than LNMRI results such as ease of patient position. In case of reader 2, the node biopsied was different based on LNMRI results in 4 patients (10.5%). In 8 patients the node was changed due to reasons other than LNMRI results. Overall 7 patients (18.4 %) had nodes less than 8 mm in size. These patients would not have undergone biopsy based on CT criteria alone.
Conclusion: Patients with malignant nodal disease may harbor both benign and malignant nodes. LNMRI has a significant role in identifying the nodes for accurate and effective percutaneous lymph node biopsy.
156. Lymphotrophic Nanoparticle-enhanced MRI (LNMRI) in Prostate Cancer: Incidence of Metastatic Perirectal Lymphadenopathy
Saksena M.*; Braschi M.; Katkar A. S.; Ventura E.; Hahn P.; Harisinghani M. G. Massachusetts General Hospital, Boston, MA
Address correspondence to M. Saksena (mansisaksena{at}gmail.com)
Objective: To determine the incidence of metastatic perirectal lymphadenopathy in patients with prostate cancer based on malignant nodes identified by lymphotrophic nanoparticle enhanced MRI (LNMRI).
Materials and Methods: 106 patients with prostate cancer underwent LNMRI for nodal staging. The images were prospectively reviewed and nodes were classified according to their size, location and signal characteristics. Findings on LNMRI were then correlated with histology.
Results: Overall 4 (3.7%) patients demonstrated perirectal lymphadenopathy (7 nodes) of which 3 (2.8%) had metastatic nodes as seen on LNMRI. All nodes were less than 10 mm in size (range 3–9 mm). Malignant nodes ranged in size from 4–8 mm. All patients that had metastatic perirectal nodes also had nodal metastases in other pelvic nodes. Pathological confirmation was obtained in all patients demonstrating malignant lymphadenopathy.
Conclusion: We identified a subset of prostate cancer patients (2.8%) who have metastatic disease within perirectal lymph nodes. This region is not evaluated in routine pelvic lymphadenectomy and preoperative imaging studies must be performed to evaluate these nodes.
157. Variation in the Size of Lymph Nodes on Pre and Postcontrast Nanoparticle-enhanced MRI
Katkar A. S.1*; Braschi M.1; Hahn P. F.1; Seethamraju R. T.2; Harisinghani M. G.1 1. Massachusetts General Hospital, Boston, MA; 2. Siemens Medical Solutions, Charlestown, MA
Address correspondence to A. Katkar (akatkar{at}partners.org)
Objective: Lymphotrophic nanoparticle-enhanced MRI has been shown to be an accurate technique for differentiating benign from malignant lymph nodes. As the nanoparticle-enhanced imaging relies on susceptibility, the aim of our study was to compare the size of lymph nodes on pre and postcontrast images to assess for any change in nodal size that can be due to the susceptibility artifact.
Materials and Methods: Forty-one lymph nodes in 15 patients with primary prostate cancer were evaluated on pre and post nanoparticle-enhanced MRI. The MR images were retrospectively reviewed and the short-axis diameter of the nodes was recorded on the pre and postcontrast FSE T2-weighted and T2*-weighted (TE: 14 msec) images. Student's t test was used to compare the mean sizes on the pre and postcontrast images.
Results: All 41 lymph nodes were benign on histopathology and showed homogenous uptake of nanoparticles. The distribution of mean short axis node size ± SD was as follows; precontrast FSE T2W (11.56 ± 4.62); precontrast T2* (12.36 ± 4.9); postcontrast FSE T2W (11.24 ± 5.02); postcontrast T2* (12.09 ± 4.93). There was no statistically significant difference between the pre- and postcontrast mean values for FSE T2W (p = 0.76) and T2* GRE (p = 0.80) images.
Conclusion: There is no significant size variation due to susceptibility artifact in benign lymph nodes following the administration of magnetic nanoparticles. Size of lymph nodes can be reliably measured on nanoparticle-enhanced MRI since susceptibility artefacts produced by nanoparticles does not affect size parameter of lymph nodes.
158. FDG PET/CT Findings in Primary Nasal-paranasal Lymphomas
Chauhan A.*; Krishnamoorthy S. K.; Froelich J.; Dhurairaj T. University of Minnesota, Minneapolis, MN
Address correspondence to A. Chauhan (chauh009{at}umn.edu)
Objective: Primary nasal-paranasal lymphomas (PNPLs) constitute less than 1% of all head and neck malignancies. The diagnosis is based on clinical presentation and imaging. MRI is the imaging modality of choice, owing to its excellence in delineating the soft tissue lesion and its extent. PET/CT imaging has never been studied in PNPLs. This case series evaluates the role of PET/CT in patients with PNPLs for initial staging and restaging.
Materials and Methods: From April 2005 to August 2006, our institution used PET/CT to image 305 patients with lymphomas, of which 77 had extranodal involvement. Five of these patients had PNPLs (3 men, 2 women, age: 27–62 years). PET/CT was performed on a Siemens Biograph 16 PET/CT scanner, 60–90 minutes after IV administration of 11.8–14.3 mCi of F-18-FDG. A total of 9 PET/CT examinations were performed in 5 patients. MRI was considered for comparison when performed within one month of PET/CT. The diagnosis of lymphoma was made on histopathology and/or clinical follow-up.
Results: In assessing initial staging, PET/CT correctly identified all five primary PNPLs, SUVs ranging from 4.1 to 40.1. Two patients had three other extranodal lesions found on PET/CT in mandibular (one lesion as true positive, SUV 9.2) and tonsillar (bilateral and symmetrical in one patient as false positive, SUV 10.9) regions. PET/CT could not detect two lesions in insular area in one patient, showing enhancement on brain MRI. In assessing nodal staging, PET/CT was correct in all five patients. Follow-up PET/CT was available in three patients. In one patient, PET/CT and anatomical imaging findings were concordant. In remaining two patients, PET/CT showed complete resolution of PNPLs, where CT/MRI still showed mass and enhancement.
Conclusion: PNPLs differ from other lymphomas in terms of aggressiveness. PET/CT is a potentially effective imaging modality in PNPLs, especially in assessing early response to therapy. Due to mucosal changes in this region and associated bony erosion, it may be difficult at times to characterize these lesions on follow-up on anatomical imaging and this drawback may be overcome with PET/CT as a complementary imaging.
159. Sensitivity of PET-positive Adrenal Nodules in a Cancer Population to Determine Malignancy
Vikram R.*; Iyer R. B.; Yeung H.; Macapinlac H. A. M D Anderson Cancer Centre, Houston, TX
Address correspondence to R. Vikram (rvikram{at}mdanderson.org)
Objective: To determine sensitivity and specificity of PET/CT in distinguishing adrenal metastasis from benign nodules in a cancer population
Materials and Methods: Retrospective study of patients who presented for PET/CT with known malignancy and adrenal nodules equal to or greater than 1cm was performed. Various malignancies including lung, melanoma, lymphoma and breast carcinoma were studied. The commonest was lung cancer. Seventy-three adrenals lesions on 62 patients who had PET/CT for cancer staging or surveillance were identified. 2D PET was performed one hour after administration of 15–20 mCi of 18F-FDG. Noncontrast CT was performed for attenuation correction and diagnosis. Patients were scanned using a Discovery ST 16 (GE Healthcare-Milwaukee, WI) system and images were reviewed on a GE Advantage-Windows workstation. Review of PET/CT studies was performed by three radiologists/nuclear medicine physicians with experience in PET/CT. Nodules were considered PET positive if average SUV was greater than average SUV of the liver. The follow-up data and biopsy reports were used to determine the pathology of the adrenal nodule.
Results: A total 17 adrenal lesions were malignant of which one was an adrenocortical tumor. Fourteen of the 17 malignant nodules were PET positive. Two of 56 benign nodules were PET positive. The sensitivity is 0.82 (95% CI 0.56–0.95) and specificity is 0.96 (95% CI 0.87–0.99). The positive predictive value for detecting malignant lesions is 0.88 and negative predictive value is 0.95. Of the 3 malignant adrenal nodules that were PET negative, one patient was on vaccine therapy for melanoma, one patient was having radiotherapy and one patient was status 5 months post chemotherapy on a lesion which measured 1 cm.
Conclusion: PET/CT is very valuable in differentiating metastatic from benign lesions. FDG-avid adrenal masses are likely to be metastatic with a high negative predictive value. However, in patients receiving therapy, there is a small but significant false-negative rate that makes PET/CT less sensitive and should be an important factor to be considered while interpreting the study.
160. An Algorithm to Fuse Coronary Arteries from CTCA with Epicardial Surfaces from Perfusion SPECT-Quantitative Validation
Sirineni G.1*; Faber T.1; Santana C.1; Verdes L.1; Riera J.2; Garcia E. V.1 1. Emory University School of Medicine, Atlanta, GA; 2. Hospital Universitari Vall d' Hebron, Barcelona, Spain
Address correspondence to G. Sirineni (sgkrad{at}gmail.com)
Objective: An algorithm to align and fuse coronary arteries detected from computed tomographic coronary angiography (CTCA) with the left ventricle detected from SPECT images was previously described by us. We intend to evaluate the algorithm's accuracy in actual patient data.
Materials and Methods: In our study 27 patients who underwent CCTA on either a 16-slice or 64-slice scanner were recruited. These patients also had a perfusion SPECT on a separate scanner. Furthermore, catheter coronary angiography was performed on all these patients. From this pool images from 11 patients who had coronary artery stenoses with corresponding perfusion defects were selected. The above-described algorithm was used to align the arteries segmented from the CTCA with the epicardial surface obtained from SPECT. The physiological area-at-risk (PA), as defined by the abnormally perfused (blackout) region of the SPECT, was compared with the anatomical area-at-risk (AA), as defined by the region of the epicardium perfused by the capillary bed distal to the location of arterial stenosis seen on the CTCA. The overlap of the physiological and anatomical areas-at-risk were also computed. Note that AA depends on the location of the stenoses in relation to the epicardial surface, and that relation depends upon the alignment of the arteries and myocardium. If the arteries are properly aligned with the SPECT, the anatomical and physiological areas-at-risk should overlap, and similarly, normal areas should overlap.
Results: AA was on average 8.9 g larger than PA. AA correlated with PA as y = 1.1+3.0 g, r = 0.89. This is consistent with the current viewpoint maintaining that SPECT underestimates the size and/or extent of perfusion defects. The average overlap of AP with AA was 76%, while the overlap of the normal regions was 61%. Overlap was highest in patients with lesions in the LAD territories, and lowest with patients with lesions in the LCX territories.
Conclusion: Coronary arteries obtained from CTCA can be accurately fused with perfusion SPECT studies even when the images are obtained from separate scanners.
161. Anterior Displacement of Excreted F18-fluorodeoxyglucose by Intravenous Iodinated Contrast Material in the Bladder on PET/CT: Frequency and Etiology
Purcell D. D.*; Coakley F. V.; Hawkins R.; Franc B.; Boddington S.; Yeh B. M. UC San Francisco, San Francisco, CA
Address correspondence to D. Purcell (derk.purcell{at}radiology.ucsf.edu)
Objective: To investigate the frequency and etiology of anterior layering of excreted F18-fluorodeoxyglucose (FDG) in the bladder on PET/CT performed with intravenous iodinated contrast.
Materials and Methods: We retrospectively reviewed 100 consecutive PET/CT studies performed in the supine position with (n = 87) or without (n = 13) intravenous iodinated contrast to determine the relative frequency of anterior displacement and layering of excreted F18-FDG in the bladder. In vitro prospective PET/CT imaging of a specially constructed phantom containing agitated and nonagitated solutions of diluted F18-FDG and iodinated contrast was undertaken to investigate the potential etiology of anterior layering of excreted F18-FDG.
Results: Anterior layering of excreted F18-FDG was more commonly seen in PET/CT studies performed with (61 of 87, or 70 %) than without intravenous iodinated contrast material (0 or 13 studies, p < 0.001). In one patient, F18-FDG uptake in a malignant posterior bladder mass was unmasked by the anterior displacement of F18-FDG. In vitro imaging of an agitated phantom containing diluted F18-FDG and iodinated contrast demonstrated the development a gradual separation gradient over time with the F18-FDG activity becoming higher in the nondependent portion of the phantom. A slow dependent infusion of iodinated contrast resulted in a distinct fluid-fluid level with complete separation of the two solutions.
Conclusion: Anterior layering of excreted F18-FDG in the bladder is commonly (and only) seen on PET/CT scans performed with intravenous iodinated contrast and may occasionally unmask PET avid posterior bladder pathology; the phenomenon is likely due to anterior displacement of excreted FDG by high physical density excreted iodinated contrast.
162. Intense FDG Activity on PET/CT in a Pulmonary Embolus Mimicking Metastatic Disease
Ryan A. T.*; Amesur N.; Blodgett T.; McCook B.; Marsh W. University of Pittsburgh Medical Center, Pittsburgh, PA
Address correspondence to A. Ryan (ryanat{at}upmc.edu)
Objective: FDG-PET is a sensitive modality for the detection of metastatic disease for various malignancies. In general, malignant lesions demonstrate intense FDG activity whereas benign processes generally demonstrate less uptake. FDG is not specific for malignancy, and several other infectious and inflammatory processes have been reported to take up the radiotracer. Pulmonary emboli (PE) have been reported to accumulate mild to moderate FDG activity often allowing differentiation from potential malignancy. We report a case of intense FDG activity within a PE originally interpreted as malignant due to the intensity of the FDG uptake. We are not aware of any other reports of intense FDG uptake in a PE. As the treatments for benign PE and tumor thrombus secondary to metastatic disease differ greatly, awareness of this potential pitfall has great clinical relevance.
Materials and Methods: A 73-year-old man underwent colonic resection for colon cancer 8 years ago. A year later he underwent right lobe liver resection for tumor metastasis, subsequently followed by segment 2 resection. Recently he was noted to have recurrent tumor at the right lobectomy margin. He then presented with a pulmonary embolus. Images were obtained from a PET/CT scanner after 16.1 mCi of F-18 FDG was injected intravenously and following a 60-minute uptake period. PET images were reconstructed with and without attenuation correction using the CT attenuation coefficients. The blood glucose at the time of FDG injection was 114 mg/dL. CT was performed after 125 cc of Optiray-350 contrast was injected intravenously.
Results: A large left main pulmonary artery embolism was diagnosed on the CT portion of the exam extending into the left lower lobe segmental arteries. There was intense FDG activity present within the embolus with a maximum standardized uptake value of 11.5. This raised the possibility of it being tumor or tumor thrombus, rather than a bland embolus. Subsequent biopsy of the embolus demonstrated it to be negative for malignancy. Follow-up scans showed decreased size and slight recanalization, compatible with bland thrombus.
Conclusion: We report a case of intense FDG uptake in a pulmonary embolus that was later proven to be a benign PE by biopsy. In PET/CT imaging, PE must be considered in the differential for FDG-avid lesions in the vascular tree of the lung. Misinterpretation as tumor thrombus can lead to inappropriate and potentially harmful therapy.
163. Determination of Effective Half-Life (Teff) of Radioactive 131I in Thyroid Cancer Patients
Vu H. T.2*; Hrejsa A.1; Markovic A.1 1. Advocate Lutheran General Hospital, Park Ridge, IL; 2. Rosalind Franklin University of Medicine and Sciences, North Chicago, IL
Address correspondence to H. Vu (hoang.vu{at}rocketmail.com)
Objective: The purpose of this study was to determine the mean effective half-life (Teff) of 131I based on the measurements of the external exposure rates (mR/h). The null hypothesis stated that there was no statistical distinction between Teff and patients' gender and selected age groups.
Materials and Methods: A database of 492 cases (345 F/147M) of thyroid cancer patients have been surveyed and collected during the period of 1994–2006. The 131I activities ranged from 60 to 400 mCi (1.62–10.81 GBq) with a mean value of 173 mCi (4.68 GBq). All the patients were required to be hospitalized for at least one day. The ages ranged from 19 to 86 years old. The mean exposure rates were compared to those provided in the NCRP Report No. 124. The mean Teff was calculated and compared to other investigators. Using Sigma Stat and Sigma Plot 3.1 software, descriptive and inferential statistical analysis were used to determine whether or not there was a statistical difference or a correlation.
Results: The study evaluated the exposure rate as a function of initial source activity, distance, and the elapsed time after the administration of 131I. Using the simple exponential plot of exposure rate versus elapsed time post therapy, Teff for a particular patient can be mathematically derived. The mean Teff was found to be 14.16 ± SD 8.06 h (all patients); 14.14 ± 6.18 (F); and 15.68 ± 9.86 (M). These values were significantly higher compared to other authors (Williegaignon et al. 2006; North et al. 2001; Keating et al. 1950; 11.41, 14 and 11.4 h, respectively). Based on the results of independent T-Tests and Mann-Whitney Rank Sum Test, there was no statistical difference between the Teff and the patient's gender (p = 0.56 and p = 0.107, respectively). The results of One-Way ANOVA with Bonferroni correction factor revealed that there was a statistical difference between the Teff and the female patients' selected age groups (p = 0.007). However, such difference was not observed in male atients.
Conclusion: The results showed that the mean Teff varied in female patients due to the rate of 131I uptake by the remnant tissues or metastases and by levels of administered activities. Such information can be useful as instructions for releasing of patients receiving thyroid cancer therapy and for the general radiation protection protocols. Using the Teff, a more realistic and patient-specific distance and time of radioactive exposure after the administration of 131I can be determined for both the patients and the family members.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
