| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Tuesday, May 8, 1:30 PM–3:30 PM
Abstracts 142-152
Moderator(s): Pamela Schaefer and Mark Mullins
1:30 PM
Keynote Address: Safety of MRI Contrast Agents in Patients with Renal Disease
Matthew Omojola, Creighton University Medical Center, Omaha, NE
1:40 PM
142. Overnight Preliminary Head CT Interpretations Provided by Residents: Locations of Misinterpreted Intracranial Hemorrhage
Strub W. M.; Leach J. L.*; Tomsick T.; Vagal A. University of Cincinnati, Cincinnati, OH
Address correspondence to J. Leach (James.Leach{at}uc.edu)
Objective: The purpose of this study was to assess the performance of radiology residents in the interpretation of head CT images and to determine patterns of error in the detection of intracranial hemorrhage. Follow-up studies were also reviewed to determine if there was any adverse effect on patient outcome as a result of these preliminary interpretations.
Materials and Methods: Radiology residents prospectively interpreted 22,590 head CT examinations while on-call from January 1, 2002 to July 31, 2007. The following morning the studies were interpreted by staff neuroradiologists. Discrepancies were then documented in the final dictation of the report and communicated to the Emergency Department or the ordering in house physician. All discrepancies were reviewed and either confirmed or adjudicated by a single neuroradiologist. Patient charts were reviewed for clinical outcomes and any imaging follow-up.
Results: There were a total of 1037 discrepancies identified of which 141 were due to intracranial hemorrhage. Thirteen of these cases were determined to be false positives. The most common types of intracranial hemorrhage that were missed were subdural and subarachnoid hemorrhage occurring in 39% and 33% of the cases, respectively. The most common location for missed subdural hemorrhage was either parafalcine or frontal in location. The most common location of missed subarachnoid hemorrhage was in the interpeduncular cistern. There was one case of non traumatic subarachnoid hemorrhage that was not described in a patient who was admitted despite the fact that these findings were not described on the preliminary report. Fourteen patients were brought back to the emergency department after being discharged for short term follow-up imaging. Thirteen of these patients were sent home the same day from the emergency department and one patient was admitted for increasing headache.
Conclusion: The proportion of discrepancies due to intracranial hemorrhage is not insignificant and is usually due to cases of subdural or subarachnoid hemorrhage. A more complete understanding of the locations of the missed hemorrhage can hopefully help decrease the discrepancy rate to help improve patient care.
143. Bone Subtraction CTA: Two Advanced Algorithms for Compensating Interscan Motion
Lell M.1,3*; Ditt H.2; Panknin C.3; Sayre J. W.3; Ruehm S.3; Villablanca J.3 1. Friedrich Alexander University, Erlangen, Germany; 2. Siemens Medical Solutions, Forchheim, Germany; 3. UCLA, Los Angeles, CA
Address correspondence to M. Lell (mlell{at}mednet.ucla.edu)
Objective: Bone subtraction techniques have been shown to enhance CT angiography interpretation, but motion can lead to incomplete bone removal. Aim of this study was to evaluate two novel registration techniques to compensate for patient motion.
Materials and Methods: Fifty-four patients underwent bone subtraction CTA (BSCTA) for the evaluation of the neck vessels with 64-slice CT. Three different registration procedures were tested, pure rigid registration (BSCTA), slab-based registration (SB-BSCTA) and a partially rigid registration (PR-BSCTA) approach. Subtraction quality for the assessment of different vascular segments was evaluated by two examiners in a blinded fashion. Cohen's Kappa test was applied to test interobserver variability, the Wilcoxon signed-rank test to test for differences between the procedures. Motion between the corresponding data sets was measured and plotted against image quality scores.
Results: Algorithms with motion compensation revealed higher image quality scores (SB-BSCTA, mean 4.31; PR-BSCTA, mean 4.43) than pure rigid registration (BSCTA, mean 3.88). PR-BSCTA was rated superior to SB-BSCTA for the evaluation of the cervical internal, and external carotid artery (p < 0.001), while there was no significant difference for the other vessels (p = 0.157–0.655). Both algorithms were clearly superior to pure rigid registration for all vessels except the basilar and ophthalmic artery; interobserver agreement was high (kappa = 0.46–0.98).
Conclusion: The algorithms provided high image quality scores in the neck and brain region, algorithms with motion compensation revealed better results in cases of complex motion. The visualization of the ICA and ECA was rated superior with PR-BSCTA.
144. Bone Subtraction CTA: Results from Non-enhanced Versus Late Venous Phase Scans to Create a Bone Mask
Panknin C.3*; Ditt H.2; Sayre J. W.3; Ruehm S. G.3; Villablanca J.3; Lell M.1,3 1. Friedrich Alexander University, Erlangen, Germany; 2. Siemens Medical Solutions, Forchheim, Germany; 3. UCLA, Los Angeles, CA
Address correspondence to C. Panknin (christoph.panknin{at}gmx.net)
Objective: In CT angiography, removal of bone can enhance visualization and detection of vascular disease. Bone subtraction algorithms have been shown to be superior to segmentation approaches, but require an additional data set to create a bone mask. The aim of this study was to evaluate whether a second scan, performed after the CTA scan can replace a non-enhanced scan for creation of the bone mask.
Materials and Methods: Thirty-eight patients underwent bone subtraction CTA (BSCTA) for the evaluation of the neck vessels with a 64-slice CT system. In 18 patients a low-dose non-enhanced CT scan (NECT) was performed prior to CTA, 18 patients received a late venous scan (LVCT) after CTA. Subtraction quality for the assessment of different vascular segments was evaluated by two examiners in a blinded fashion with a time span of approximately two weeks between the readings. Cohen's Kappa test was applied to test interobserver variability, the Wilcoxon signed-rank test to test for differences between the two procedures. Motion between the two scans was measured and correlated to image quality.
Results: Bone subtraction could be successfully performed with both NECT and LVCT. Arterial enhancement in the subtracted data sets was comparable in both groups. High interobserver agreement was found (kappa values 0.5–1). Image quality did not differ significantly (borderline significance for the common carotid artery (reader 2: p = 0.05), p-values for the other vascular regions ranging from 0.10–1) although bone remnants from misregistration due to interscan motion were slightly more frequently encountered in subtractions with LVCT.
Conclusion: LVCT is a promising alternative to a NECT scan if a late venous scan is clinically indicated or CTA reveals a situation in which BSCTA would be desirable.
145. Grading of Gliomas. Assessment with Diffusion-weighted MR Imaging and Proton MR Spectroscopy
Abdel Razek A.*; Samir S.; Nada N. Mansoura Faculty of Medicine, Mansoura, Egypt
Address correspondence to A. Abdel Razek (arazek{at}mans.eun.eg)
Objective: To evaluate the usefulness of apparent diffusion coefficient map and multivoxel proton MR spectroscopy in grading of gliomas.
Materials and Methods: This study included 48 patients (36M, 12F aged 57–69 years: mean 61 years) with gliomas underwent routine MR, diffusion weighted MR imaging and MR spectroscopy. Diffusion MR imaging was done using echoplanar imaging. Proton MR spectroscopy was performed using multivoxel chemical shift imaging (CSI) with TE = 135. Apparent diffusion coefficient (ADC) values were determined in the solid part of the tumor and the peritumoral hyperintense area on T2WI. We analyzed the spectral peaks and calculated ratio for lactate/creatine, choline/Cr, NAA/Cr. Neuropathologic grading of patients was performed according to established WHO criteria: low grade (WHO I,II) and high grade (WHO III and IV).
Results: Histologic grading were grade II astrocytoma (n = 7), grade III anaplastic astrocytoma (n = 13) and grade IV glioblastoma multiforme (n = 28). The mean ADC value of solid part of high-grade glioma was statistically lower (p < 0.001) than low-grade glioma. Peritumoral ADCs were statistically higher (p < 0.01) in low-grade than high-grade glioma. Proton MR spectra of high-grade gliomas showed significantly lower level of myoinositol and presence of lipid compared with that of low-grade gliomas. There was statistically difference in lactate/Cr ratio (p < 0.002) and insignificant difference of choline/Cr and NAA/Cr ratio (p < 0.6) between high- and low-grade gliomas.
Conclusion: Apparent diffusion coefficient map and MR spectroscopy are noninvasive imaging modalities that provide valuable information for preoperative grading of gliomas.
146. Values from Dynamic Contrast-enhanced Imaging and Cerebral Blood Volume Imaging Correlate in Patients with High-grade Glial Neoplasms
Provenzale J. M.3*; Golding L. D.4; Choma M.1; York G.2 1. Boston Children's Hospital, Boston, MA; 2. Brooke Army Medical Center, Ft Sam Houston, TX; 3. Duke University Medical Center, Durham, NC; 4. Moses Cone Hospital, Greensboro, NC
Address correspondence to J. Provenzale (prove001{at}mc.duke.edu)
Objective: To correlate degree of contrast enhancement on dynamic contrast-enhanced (DCE) T1-weighted images and relative cerebral blood volume (rCBV) values on T2*-weighted images in patients with high-grade brain neoplasms.
Materials and Methods: Ten patients with biopsy-proven high-grade gliomas underwent DCE imaging (T1-weighted FSPGR TR 8.3 ms, TE 1.5 ms during infusion of 0.1 mmol/kg MR contrast medium). This sequence was followed within 5 minutes by DSC imaging (TR 1500 ms, TE 80 ms, infusion of 0.2 mmol/kg MR contrast medium). DCE analysis was performed using the maximum signal intensity algorithm and DSC analysis was performed using the negative enhancement integral program, which were both operating in Functool on an Advantage Windows workstation. For DCE analysis, we used a software program developed by our group that measures relative permeability by depicting signal intensity as number of standard deviations (SD) above the mean for normal tissue. For DSC analysis, rCBV was measured in the same tumor-containing ROI for each patient as for DCE analysis and expressed as percent of rCBV measured within the lentiform nucleus. For each tumor, we performed two comparisons: (1) average DCE and rCBV values within an ROI around entire contrast-enhancing tumor on a single image through center of lesion, and (2) highest DCE and highest rCBV values within each tumor. Statistical analysis of the first comparison was performed using Pearson correlation coefficient, R2 correlation coefficient and Spearman rank correlation and for the second comparison using Kendall's tau correlation.
Results: Mean signal intensity values ranged between 3.48 and 7.16 SD above baseline values (mean: 4.89 SD). Mean rCBV values ranged from 57.9% to 122.7% of normal lentiform nucleus (mean: 76.6%). Pearson correlation coefficient was 0.867, R2 correlation coefficient 0.752 and Spearman rank correlation 0.794 (p = 0.001). DCE values from the region of highest signal intensity ranged from 7.7 to 48.6 SD above baseline values (mean: 17.3 SD). Highest rCBV values ranged from 105% to 400% of normal lentiform nucleus (mean: 292%). The correlation was estimated at 0.7778 and was statistically significant (p = 0.0035).
Conclusion: We found a high correlation between degree of contrast enhancement on DCE images and rCBV values in whole tumor as well as in regions having the highest degree of contrast enhancement in this small study. Our findings suggest relative permeability and rCBV may be correlated in high-grade glial neoplasms.
147. Diffusion Tensor Imaging-derived Tractography Visualizes the Altered Corticospinal Tracts Around Dilated Ventricles
Fan M.1*; Poulanc J.2; Wu R.1; Mikulis D.2 1. Department of Medical Imaging, The 2nd Hospital, Shantou University Medical College, Shantou, China; 2. Department of Neuroradiology, Toronto Western Hospital, University of Toronto, Toronto, Canada
Address correspondence to M. Fan (fanok05{at}hotmail.com)
Objective: It's known that the corticospinal tracts (CST) is one of the two main white matter fiber bundles (another one is corpus callosum) close to the ventricles on the DTI-derived tractography. The purpose of this study is to report that DTI-derived fiber-tracking technique can show water diffusion changes of CST which is close to the enlarged ventricles compared with normal controls.
Materials and Methods: Diffusion tensor imaging examinations at 3.0 T were performed on 4 patients with enlarged ventricles and 10 healthy control subjects. Based on DTI orientation-color maps, Fractional anisotropy (FA), average DC (ADC) of CST portions of right-left sides in each case were symmetrically measured, starting from their lower positions at the caudal pons-medulla level to the higher positions at the superior internal capsule level; and accordingly, the fibers of CST were shown with 3D tractographic techniques.
Results: In the healthy subjects, the mean values of FA and ADC of CST portions at different positions had no significant differences (p > 0.05) in the symmetrical right-left sides, respectively. The FA values of CST portions surrounding the dilated lateral ventricles in all patients were higher than the mean FA of healthy control subjects; the corresponding ADC values in three of the four patients were mostly higher, except one was lower than the healthy controls. Tractography demonstrated CST fibers were brighter and expanded along the dilated ventricles, compared with the controls.
Conclusion: DTI suggested that the expanded CST with unusual higher FA in hydrocephalous be mainly caused by the compress of the enlarged ventricles. Higher FA in white matter tracts is speculated to be an early response against outside force, and its significance is needed to be further investigated.
148. Assessment of Regional Cerebrovascular Reactivity Using BOLD fMRI and Carbon Dioxide Challenge: Reproducibility and Regional Heterogeneity
Krishan S.2*; Goode S.1; Auer D.1 1. Department of Academic Radiology, Nottingham, United Kingdom; 2. Radiology Academy, Leeds General Infi rmary, United Kingdom
Address correspondence to S. Krishan (sonalkrishan11{at}yahoo.co.uk)
Objective: Cerebral blood flow (CBF) and volume (CBV) increase proportionally, over a reasonable range of increases in carbon dioxide (CO2); because of unchanged oxygen extraction this causes decreased deoxyhemoglobin, and consequently increased BOLD MR signal. BOLD based cerebrovascular reactivity (CVR) maps can be generated using CO2 challenge. Knowledge of the accuracy of the technique, a precondition for clinical use, is however very limited. The aim of our study was to develop a clinically applicable study design and to assess the test-retest performance in healthy volunteers.
Materials and Methods: Sixteen healthy volunteers underwent MR scanning. All the subjects had at least two periods of high (breathing 8% CO2) and normal (breathing air) end-tidal CO2. BOLD imaging was performed with a standard gradient echo EPI sequence (TR/TE 3500/60 ms). Expired CO2, arterial oxygen saturation, and blood pressure were continuously monitored. Mean signal intensity time-courses were derived from the raw data after standard pre-processing using SPM. The data was normalized to the change in end tidal CO2. Percent signal change was calculated for whole brain and various regions of interest. The reproducibility was tested by measuring coefficient of variation (CV) of percent signal change per mmHg change in end tidal CO2 for the whole brain, and the specific regions. Color-coded parametric maps were generated using FSL.
Results: The mean percent signal change for whole brain was 2.2 ± 0.79%. The intrasubject coefficient of variation for whole brain signal change was 7.16%.The color-coded parametric maps showed symmetrical and uniform increase in blood flow in both hemispheres.
Conclusion: We have established a standardized clinically applicable CVR mapping technique that can be added to standard MRI. The normalized whole brain CVR is in good agreement with published data and furthermore a good intrasubject reproducibility of peak signal change was shown. This paves the way for wider clinical applications in patients with suspected CVR impairment.
149. Spectrocopic Imaging of the Human Brain at 7 T
Sammet S.2*; Koch R. M.2; Murdoch J.1; Knopp M. V.2 1. Philips Medical Systems, Cleveland, OH; 2. The Ohio State University, Department of Radiology, Columbus, OH
Address correspondence to S. Sammet (sammet.5{at}osu.edu)
Objective: Spectroscopic imaging (SI) is used in magnetic resonance spectroscopy to determine local variations in the concentrations of metabolic compounds in vivo. The spatial location is phase encoded and a spectrum is recorded at each phase encoding step to allow the spectra acquisition in a number of volumes covering the whole sample. This study demonstrates that SI of the human brain is feasible at ultra-high magnetic field strengths but has its limitations due to field inhomogeneities and susceptibility artifacts.
Materials and Methods: Spectroscopic Imaging of the human brain was performed on 7 volunteers on a 7-T whole body MR-scanner (Philips Medical Systems) with a transmit-receive head-coil. Transverse slice spin-echo localization was performed with the following parameters TR = 1500 ms, TE = 144 ms; field of view = 250 mm; 28 x 28 matrix; slice thickness = 20 mm; 1 average; 8 REST slabs. SI was performed in one of the volunteers additionally with a short echo time of TE = 26 ms.
Results: Highly-resolved spectra were acquired in the brain of all volunteers. Resonances of choline, creatine, N-acetylaspartate and myo-inositol could be easily identified in all CSI voxels of the brain. SI with short echo time additionally resolved the resonances of glutamate, glutamine and scyllo-inositol.
Conclusion: SI at ultra-high magnetic fields is possible and increases the sensitivity and spectral resolution in magnetic resonance spectroscopy. PRESS localization is not applicable for CSI at ultra-high fields due to chemical shift displacement. It is preferable to use slice spin-echo localization instead. Good shimming avoids broadening of spectral peaks induced by stronger susceptibility effects at higher field strengths in vivo. SI at ultra-high fields requires distortion corrections of the acquired data. Ultra-high-field SI improves the noninvasive characterization and quantification of molecular markers with clinical utility for improving detection and treatment for a variety of neurological diseases.
150. Hemodynamic Evaluation of Basilar Tip Aneurysms with Computational Fluids Dynamics
Karmonik C.2*; Benndorf G.2; Nasseri F.2; Klucznik R.1; Haykal H. A.1; Strother C. M.1 1. The Methodist Neurological Institute, Houston, TX; 2.The Methodist Research Institute, Houston, TX
Address correspondence to C. Karmonik (ckarmonik{at}tmh.tmc.edu)
Objective: Wall shear stress (WSS) is considered one of the main hemodynamic factors in the development of intracranial aneurysms. Magnitude and distribution of the WSS in and around three basilar tip aneurysms were determined using computational fluid dynamics (CFD). Flow patterns at the level of aneurysm ostium were analyzed to determine inflow and outflow zones.
Materials and Methods: Three-dimensional digital subtraction angiography (3D DSA) image data of three basilar tip aneurysms were retrospectively obtained from diagnostic angiogram studies. High-resolution 3D reconstructions of the acquired 3D DSA data and computational meshes were created and then imported into the CFD solver Fluent (Fluent Inc.) Unsteady simulations were performed with the assumption of Newtonian fluid property for blood and a rigid wall approximation. Three areas for assessment of WSS were defined: the aneurysm area (AA) as the wall area of the basilar tip (including the aneurysm), the basilar artery area (BAA) as the area of a healthy section of the basilar artery and the aneurysm dome area (DA) located at the top of the aneurysm dome. For assessment of inflow and outflow zone, the aneurysm ostium area was defined using a cut plane through the aneurysm neck.
Results: Inflow and outflow areas relative to the total ostium area did not significantly change between the times of maximum flow and minimum flow in the parent artery (inflow: case 1: 45 %, case 2: 32 %, case 3: 52 %; outflow: case 1: 55 %, case 2: 68 %, case 3: 48 %). Variations in the inflow conditions (100 % - 10 %) result in changes of the average WSS magnitude relative to the average WSS magnitude of the BA area for the wall areas AA (case 1: 95%–84%; case 2: 86%–68%; case 3: 86%–76%) and DA (case 1: 50%–34%; case 2: 71%–48%; case 3: 30%–15%).
Conclusion: Not only the absolute values but also the relative values of the WSS magnitude obtained at different wall segments in the basilar system using CFD techniques strongly depend on the inflow conditions. Accurate determination of these inflow conditions is therefore essential to be able to obtain realistic WSS values. The results of our study also demonstrate that CFD can be utilized to determine areas of inflow and outflow at the aneurysm ostium.
151. Recurrent Vertebral Fractures after Kyphoplasty: Our Experience with 711 Fractures
Nasseri F.2*; Naeini R.3; Karmonik C.2; Haykal H. A.1 1. The Methodist Neurological Institute, Houston, TX; 2. The Methodist Research Institute, Houston, TX; 3. Baylor College of Medicine, Houston, TX
Address correspondence to F. Nasseri (fnasseri{at}tmh.tmc.edu)
Objective: Patients with osteoporotic vertebral fracture often return with additional vertebral fractures after polymethylmethacrylate (PMMA) kyphoplasty. There is conflicting data regarding the incidence of subsequent adjacent fractures due to intra-discal PMMA leakage or increased vertebral rigidity after vertebral augmentation. The purpose of this study is to evaluate the frequency of adjacent and remote fractures after kyphoplasty, and to determine any possible patterns of recurrent fractures and intra-discal PMMA leakage.
Materials and Methods: We retrospectively reviewed 711 osteoporotic vertebral fractures treated with kyphoplasty in 427 patients between October 2001 and September 2006. 60 patients, with 110 new fractures returned for vertebral augmentation by kyphoplasty. The incidences of new adjacent and remote vertebral fractures were determined. The incidence of adjacent intradiscal PMMA leakage was evaluated since that was felt to be a significant factor in recurrent fractures.
Results: 110 new vertebral fractures occurred during our follow-up period from October 2001 to September 2006, of which 68 (62%) vertebral fractures occurred immediately adjacent to previously treated vertebra and 42 (38%) vertebral fractures occurred in remote locations, separated by at least 1 vertebral body away from previously treated vertebra. Of the 68 adjacent fractures 18 (15%) noted to be adjacent to an intra-discal leak of the PMMA and 6 (9%) were located opposite to the end plate adjacent to the intra-discal leakage. These 6 fractures all occurred below a previously treated vertebra where the intra-discal leakage was to the disc above the first treated level (p < 0.01). No significant association between the presence of minimal intra-discal leakage and occurrence of the adjacent fracture was found. Subjective evaluation of intra-discal leak revealed that 30 leaks were noted to be minimal leaks to the annulus, 2 medium size leaks in to the annulus and 1 large leak penetrated to the nucleus pulposus.
Conclusion: Small intra-discal annular leaks of PMMA after kyphoplasty do not appear to be a significant factor inducing adjacent fractures. Our results indicate that the risk of recurrent fracture after kyphoplasty is less than that previously reported for vertebroplasty.
152. Preliminary Experiences with Functional Anesthetic Discography
Luchs J. S.*; Cho M.; Ortiz O. Winthrop-University Hospital, Mineola, NY
Address correspondence to J. Luchs (jluchs{at}winthrop.org)
Objective: To describe our preliminary experience with functional anesthetic discography (FAD) in the evaluation of patients with suspected discogenic low back pain.
Materials and Methods: FAD was performed in 19 consecutive patients (13 men, 6 women; mean age 47.2 years, range 34 to 69 years) who underwent lumbar discography for suspected discogenic low back pain. A total of 53 potentially pain-producing discs were studied. A 21-gauge FAD catheter was placed into one or more potentially pain producing discs as determined by a provocative discograms (32 discs) or abnormal discograms (26 discs) study using an over-the-wire exchange technique. Patients were kept blind to the level of disc injection and to the level of FAD placement. At the completion of the discogram, a multidetector CT scan was obtained in order to analyze the injected discs for the presence of radial tears and other pathology using the modified Dallas discogram scale. Patients were subsequently examined before and after (0, 5, 10, and 15 minutes post disc anesthetic injection) the injection of the disc via the FAD catheter using 1–2 mL of 0.5 to 2% local anesthetic agent. Patients were asked to perform maneuvers that would typically elicit their pain symptoms and their pain response was recorded.
Results: Nineteen out of 29 (65.5%) of the injected discs showed a favorable response (pain relief greater than 3 visual analog pain scale units) to anesthetic injection. Ten out of 29 (34.5 %) did not show a favorable response. In those patients with a favorable FAD response, 19 discs showed a provocative response during discography and 18 discs showed the presence of disc pathology on CT examination. In patients with no pain relief after anesthetic injection (an unfavorable FAD response), 8 discs showed a provocative discographic response, and 6 discs showed the presence of disc pathology on CT examination.
Conclusion: FAD is a useful procedure and can be used to confirm the presence of pathologic discs as the source of a patient's low back pain symptoms.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
