Juergen Weber

Assessment of myocardial perfusion using multisection first-pass MRI and color-coded parameter maps: a comparison to 99mTc Sesta MIBI SPECT and systolic myocardial wall thickening analysis.

The most recently reported magnetic resonance first-pass myocardial perfusion studies were restricted to single slice imaging or a data analysis based on interactively placed regions of interest. This study was designed to investigate a new saturation recovery TurboFLASH sequence for multisection myocardial perfusion imaging and to develop a pixel-based software tool to calculate qualitative perfusion parameters. The findings of perfusion imaging were compared to percent systolic myocardial wall thickening analysis and 99mTc Sesta MIBI SPECT. Six healthy volunteers and twelve patients with proven coronary artery disease (CAD) or chronic myocardial infarction were examined. Diagnostic images were acquired for all volunteers and patients with the multisection saturation recovery TurboFLASH sequence. Perfusion defects could be visualized on parameter maps for signal intensity increase over baseline and signal intensity upslope. Sensitivity and specificity were 76.9% and 97.1% for first-pass perfusion MRI, and respectively 84.6% and 94.3% for CINE imaging. All perfusion defects determined with 99mTc Sesta MIBI SPECT were identified by the combined analysis of myocardial perfusion and wall thickening. The presented software demonstrated a pixel-based analysis of first-pass perfusion studies and simplified image interpretation in a clinical setting. The combination of perfusion and wall motion imaging provided complementary information for the treatment of patients suffering from CAD.

Regional relative blood volume MR maps of meningiomas before and after partial embolization

PURPOSE Our goal was to evaluate the role of relative blood volume (rBV) measurements in monitoring the embolization effect in meningiomas by using maps of susceptibility-weighted first pass MR data. METHOD Eighteen examinations of nine patients before and following partial embolization were performed on a 1.5 T scanner. Embolization was achieved by injection of particles (45-150 microns). During dynamic imaging a bolus (0.2 mmol/kg) of gadopentetate dimeglumine was injected. RESULTS The tumor rBV/gray matter rBV ratio was 3.11 +/- 1.40 for untreated tumors and 0.12 +/- 0.09 for successfully embolized tumors. The regional rBV in embolized meningiomas was significantly lower than that in untreated meningiomas (t test, p < 0.001). Vital tumor tissue showed positive enhancement in T1 and high rBV; nonvital tissue lacked T1 enhancement and bolus effect in the first pass, thus leading to low or missing rBV values. However, we also observed lack of bolus formation despite T1 enhancement (6/9 postembolization regions), possibly due to slow collateral flow. One of the patients treated with embolization bled during surgery. CONCLUSION Maps of relative regional cerebral BV provide hemodynamic information in meningiomas and monitor the treatment effect of embolization in meningiomas more precisely than T1-weighted contrast-enhanced imaging.

Multislice MR perfusion imaging and regional myocardial function analysis: complimentary findings in chronic myocardial ischemia

Purpose: The purpose of this study is to assess the reliability of multislice MR perfusion imaging in comparison to regional wall function and nuclear medicine and to test different qualitative and quantitative parameters for perfusion assessment. Material and methods: 15 patients with chronic myocardial ischemia underwent CINE and first-pass perfusion MR imaging. Functional myocardial imaging was performed using a segmented CINE FLASH sequence and systolic myocardial wall thickening was assessed after semiautomated segmentation. MR first-pass perfusion studies were performed using a multislice saturation recovery TurboFLASH sequence. Different parameters were calculated for assessment of hypoperfused segments and results of MR imaging compared to 99mTc-SestaMIBI SPECT. Results: MR perfusion imaging showed a sensitivity of 72% and a specificity of 98%. In combination with MR CINE imaging and wall thickening analysis we calculated a sensitivity of 100% and a specificity of 93%. Qualitative and quantitative perfusion parameter analysis showed significant differences between normal and hypoperfused segments for the signal intensity increase (p < 0.001), the signal intensity upslope (p < 0.001) as well as for the myocardial mean transit time (p < 0.001). Conclusion: The combination of systolic wall thickening analysis and myocardial perfusion can markedly improve the sensitivity of MRI in depiction of LV myocardial perfusion abnormalities. For assessment of hypoperfusion, different quantitative and qualitative parameters can be calculated showing significant differences between normal state and hypoperfusion.

MRI assessment of cerebral blood volume in patients with brain infarcts

Abstract MRI perfusion studies have focussed mainly on acute ischaemia and characterisation in ischaemia. Our purpose was to analyse regional brain haemodynamic information in acute, subacute, and chronic ischaemia. We performed 16 examinations of 11 patients on a 1.5 T MR images. Conventional and dynamic contrast-enhanced imaging were employed in all examinations. For the dynamic susceptibility sequences, a bolus (0.2 mmol/kg) of gadopentetate dimeglumine was injected. Reconstructed regional relative cerebral blood volume (rCBV) maps, bolus maps, and conventional images were analysed by consensus reading. In all examinations decreases in rCBV were observed in the lesions. The distribution of regional rCBV in lesions was heterogeneous. The rCBV of the periphery of the lesions was higher than that at their center. There was a correlation between the time since onset and abnormalities on the rCBV map and T2-weighted images (T2WI). In the early stage of acute stroke, the abnormalities tended to be larger on the rCBV than on T2WI. Many patterns of bolus passage were observed in ischaemic regions. rCBV maps provide additional haemodynamic information in patients with brain infarcts.

Comparison of the diagnostic information in relative cerebral blood volume, maximum concentration, and subtraction signal intensity maps based on magnetic resonance imaging of gliomas.

RATIONALE AND OBJECTIVES The authors investigate the validity of regional relative cerebral blood volume (rCBV) versus maximum concentration and subtraction signal intensity (SI) maps using simple reconstruction modes in patients with gliomas. METHODS Twenty-five patients were studied using a 1.5 T magnetic resonance imaging scanner. To calculate the rCBV map, the magnetic resonance susceptibility effect SI/time curves were first transformed into concentration/time curves; then a gamma-variate function was fitted and the area under the curve was integrated. From the concentration/time data, the maximum concentration (MAX) maps were calculated pixel per pixel as the maximum peak amplitude of the concentration/time curve. Subtraction (SUB) maps are a result of simple image subtraction of pixelwise baseline SI minus the highest peak of susceptibility change pixel per pixel. Region of interest means SI measurement of the different maps was compared using statistical t test correlation. RESULTS Normal gray to white matter contrast did not show a significant difference among the rCBV, MAX, and SUB maps. Based on statistical evaluation, the low-grade lesions did not differ significantly in the rCBV, MAX, and SUB maps. The group with high-grade lesions (12 patients) showed no significant difference in standardized rCBV, MAX, and SUB maps. CONCLUSION Compared to rCBV maps, the simple MAX and SUB maps demonstrated good correlation in both high-grade and low-grade gliomas. This simpler approach could establish first-pass reconstruction in clinical settings because it reduces the need for time-consuming postprocessing.

Comparison study of oxygen-induced MRI-signal changes and pO2 changes in murine tumors.

The purpose of this study was to compare the results from oxygen-induced MR-signal intensity changes with polarographic pO2 measurements in tumors. Balb-c mice with an intramuscular transplanted osteosarcoma were examined. To study the response of tumors to changes in oxygen supply, hyperoxia was induced by breathing pure oxygen for a short period. The examination of each animal started with T2* weighted MRI followed by the pO2 measurements (Eppendorf Histograph). During oxygen inhalation in all tumors, when the hypoxic tumor fraction drops, both areas of significant MR-signal intensity increase and decrease were observed in each animal.

Qualitative und quantitative assessment of myocardial perfusion using magnetic resonance imaging

SummaryPurpose: Purpose of this study was to test a triple slice saturation recovery turbo FLASH sequence for myocardial perfusion imaging. In addition data-evaluation-tools for qualitative and quantitative perfusion parameters are presented and preliminary tested. Material and Methods: We examined 8 healthy volunteers and 4 patients with myocardial infarction. Parameters of the saturation recovery turbo FLASH sequence were as follows: TR = 2.5 msec, TE = 1.2 msec, α = 8 °, 3 slices, thickness 10 mm. For data analysis signal-intensity time curves were calculated pixel by pixel and evaluated for signal-intensity-increase over baseline and signal-intensity-upslope. Images were displayed color-coded. For quantitative data analysis we used the indicator dilution theory and developed a deconvolution algorithm which takes the arterial input function into account to calculate the myocardial mean transit time (MTT). Results: The color-coded parametermaps showed uniform conditions in normal myocardium of volunteers, but reduced signal-intensity-increase over baseline and signal-intensity-upslope for infarcted areas in patients. The MTTs calculated using our algorithm were significantly shorter than those assessed with previous methods and matched better with values derived from literature. Infarcted areas show prolonged MTTs in comparison to normal myocardium. Conclusion: A triple slice saturation recovery turbo FLASH sequence is suitable for myocardial perfusion imaging. Color-coded parametermaps can visualize hypoperfused areas. For calculating myocardial MTTs using indicator dilution therapy a deconvolution algorithm is necessary.ZusammenfassungZielsetzung: Ziel der Studie war die Erprobung einer Dreischicht Saturation Recovery Turbo FLASH Sequenz bei der Untersuchung der regionalen Myokardperfusion. Für die Datenauswertung stellen wir qualitative und quantitative Nachverarbeitungsmethoden vor, die wir an Probanden und Patienten testeten. Material und Methode: Wir untersuchten 8 gesunde Probanden und 4 Patienten mit Myokardinfarkten. Parameter der verwendeten Saturation Recovery Turbo FLASH Sequenz waren: TR = 2,5 ms, TE = 1,2 ms und α = 8°. Es wurden drei 10 mm dicke Schichten akquiriert. Zur Datenanalyse berechneten wir pixelweise Signalintensitäts-Zeit Kurven und werteten diese nach der Signalintensitätszunahme über den Ausgangswert und Signalintensitätsanstieg aus. Die Ergebnisse stellten wir farbkodiert dar. Zur quantitativen Datenauswertung nach Indikatorverdünnungstheorie entwickelten wir einen Rückfaltungsalgorithmus, der die arterielle Inputfunktion bei der Berechnung der myokardialen Mean Transit Time (MTT) berücksichtigt. Ergebnisse: Mit Hilfe der farbkodierten Parameterbilder konnten normal perfundierte von hypoperfundierte Myokardarealen gut unterschieden werden. Die mit unseres Algorithmus berechneten MTTs waren kürzer als die mit bisherigen Methoden abgeschätzten. Mit aus der Literatur ermittelten MTT Werten stimmten unsere MTTs besser überein. Infarktareale zeigten verlängerte MTTs. Schlußfolgerung: Eine Saturation Recovery Turbo FLASH Sequenz ist für die Untersuchung der regionalen Myokardperfusion in bis zu drei Schichten geeignet. Farbkodierte Parameterbilder können hypoperfundierte Myokardareale qualitativ darstellen. Für eine Berechnung der myokardialen MTT liefert ein Rückfaltungsalgorithmus verbesserte Abschätzungen der myokardialen MTT.