O15	Perfusion imaging of the brain using ultrasound Privatdozent Dr. Günter Seidel, MD Department of Neurology, Medical University Lübeck In this review, methodological aspects of cerebral perfusion imaging with ultrasound signal enhancing agents are being described. The various experimental bases, contributing to the understanding of the phenomena are summarised and the resulting human investigation techniques are being illustrated. By means of harmonic imaging technology, human cerebral perfusion can be depicted as a two-dimensional scan. The two major principles of contrast measurement are analysis of the bolus kinetics and analysis of the refill kinetics. Using the bolus method, hypoperfused areas in stroke patients can be visualised and parameter images of wash-in and wash-out curves can be generated off-line. The recently developed theory on the refill kinetics of UCA enables us to calculate quantitative parameters for the description of the cerebral microcirculation, being less affected by the depth dependence of the contrast effect. These parameters, too, can be visualised as parameter images. The ultrasound methods described in this review represent new minimal-invasive bedside techniques for analysing brain perfusion. Although their development is still in an early state, the potential of these ultrasound technologies to compete with perfusion-CT, perfusion-MRI or SPECT in the diagnostic arsenal of brain imaging techniques is becoming evident.
O16	UMEDS and Beyond -- Future Microbubble Applications in Stroke Stephen Meairs, Neurologische Klinik, Universitätsklinikum Mannheim, Germany "Ultrasonographic Monitoring and Early Diagnosis of Stroke" (UMEDS) is a new project funded by the European Commission for development of methods for ultrasonographic brain imaging and novel monitoring techniques to assess morphologic and functional parameters of evolving stroke. The project encompasses the design and manufacture of optimized experimental echographic agents for stroke diagnosis and brain perfusion monitoring, which in turn correspond both to clinical demands and foreseen optimizations in harmonic imaging technologies for transcranial applications. UMEDS addresses the emerging field of molecular imaging, which encompasses the noninvasive in vivo diagnosis of complex pathologic processes by detection of unique molecular signatures. Localization of specific biochemical epitopes with targeted contrast agents affords the opportunity for imaging of thrombus material in acute vessel occlusions as well as a possible means for enhanced detection of microembolic signals. The promise of gene therapy in acute stroke lies in the potential to increase the brain’s resistance to ischemic damage by upregulating genes known to improve cell survival. For successful development of clinical gene therapy, however, effective gene delivery is needed. Recent work suggests that ultrasound may play a central role in the development of new approaches for gene delivery in stroke patients. Microbubbles can be used to enhance the effects of ultrasound on gene expression. They may also be employed as carriers of gene therapeutic agents. The ability to focus ultrasound and cause local cavitation with these new gene carriers may provide a powerful new tool for gene delivery in stroke patients.
O17	Semi-Quantitative Ultrasonic Brain Perfusion Imaging: Technical and Theoretical Basis of Contrast Burst Depletion Imaging (CODIM) ^1W. Wilkening, ^2J. Eyding, ^2S. Meves, ^2T. Postert, ^1H. Ermert ^1Dept. of Electrical Eng., ^2Dept. of Neurology, Ruhr-University, Bochum, Germany Background: Functional sonography greatly benefits from ultrasound contrast agents (UCAs). Although the mean flow velocity in the microcirculation, i. e. in sub-resolution vessels, still cannot be measured directly, UCA-specific imaging techniques allow to determine the UCA concentration over time as a time-intensity curve (TIC). The microbubble used for UCAs undergo gradual destruction when insonified. This effect, observed on a short time scale, is used for highly sensitive UCA imaging, e. g. contrast burst imaging (CBI). The same effect also has an impact on TICs and has to be considered for the semi-quantitative analysis of TICs. TICs that can be observed after a bolus injection of an UCA are not well suited for a semi-quantitative analysis because of many unknown and unpredictable influences of physiological effect, vessel topology, shadowing artefacts etc. Moreover, only one imaging plane can be assessed per injection. Methods: We propose a new semi-quantitative perfusion imaging approach called CODIM (contrast burst depletion imaging). Insonation at 1 – 5 frames per second starts after a fairly constant microbubble concentration has been reached in the imaging plane (>40 s after a bolus injection or UCA infusion). The acquisition of each of the 20 – 40 ultrasound images then reduces the UCA concentration while perfusion re-increases it between acquisitions. A theoretical model was develop to describe the TIC as a function of time as well as local blood volume, destruction coefficient (DC), and perfusion coefficient (PC). This approach was investigated in vitro using a perfusion phantom made from agar and a sponge with open pores. To account for the size of the phantom, a 7.5 MHz linear array transducer was used. Mean flow velocities ranged from 0 –1 cm/s. Results: In repeated measurements, it could be shown that the PC calculated from the image series is a monotonic function of mean flow velocity and is only weakly sensitive to initial UCA concentration an acoustic power. Conclusions: Contrast Burst Depletion Imaging was shown to provide semi-quantitative information on tissue perfusion in an in vitro setup, where mean flow velocities in the range of 0 –1 cm/s were considered. Contrast Burst Imaging has previously been show to be a sensitive contrast agent imaging technique for transcranial applications in vivo.
O18	COMPARISON OF TRANSCRANIAL BRAIN TISSUE PERFUSION IMAGES BETWEEN ULTRAHARMONIC, SECOND HARMONIC, AND POWER HARMONIC IMAGING T. Shiogai^1), C. Uebo^1), M. Makino^2), T. Mizuno^2), K. Nakajima^2), H. Furuhata^3), Department of Clinical Neurosciences, Kyoto Takeda Hospital^1), Department of Neurology, Kyoto Prefectural University of Medicine^2); ME Lab, Tokyo Jikei University School of Medicine^3), Japan Objective: To clarify optimal brain tissue perfusion images visualized by transcranial contrast-enhanced harmonic imaging, we compared the gray-scale images of ultraharmonic imaging (UHI) and second harmonic imaging (SHI), and power harmonic imaging (PHI) in patients both with and without an intact temporal skull. Methods: The subjects were 14 various neurological patients (aged 40-84 years, mean 70), 10 with an intact temporal skull and 4 who were undergoing temporal craniectomy. Utilizing an ultrasound system with an S3 transducer (SONOS 5500), transient response images taken (at every 2 heart beats) after a 7ml bolus injection of Levovist^® (2.5g) were evaluated at the horizontal plain involving the temporal lobe (TL), basal ganglia (BG) and Thalamus (Th) via temporal windows. The transmitting/receiving frequencies (MHz) of each imaging process were as follows: gray-scale integrated backscatter images of UHI (1.3/2.6) and SHI (1.3/3.6); power Doppler (PD) and B-mode images of PHI 2.6 (PD 1.3/2.6, B-mode 1.6/3.2) and PHI 3.2 (PD 1.6/3.2, B-mode 1.6/3.2). Contrast images were compared in terms of 1) size and location (TL, BG, and Th), 2) peak intensity (PI), and 3) visualization of background image and sharpness of contrast area demarcation. Results: a) Contrast area: 1) A large contrast area was most frequently observed in SHI images followed by UHI, PHI 2.6, and PHI 3.2. In two craniectomized patients, the contrast area was larger in UHI images than in other procedures. 2) No contrast area was observable through SHI or UHI in 2 intact temporal skull cases. 3) The contrast effects of PHI were more closely located to the transducer. b) Intensity: There were more cases of high PI in UHI images. c) Background and demarcation: 1) The contrast area was defined in all cases of UHI and PHI images, but the area was not always defined by SHI.2) It was difficult to identify the location of the contrast area by UHI due to weak background signals. Conclusions: In transcranial contrast-enhanced harmonic imaging, images utilizing a low receiving frequency of 2.6 MHz are superior to those of 3.2 or 3.6 MHz in cases of intact temporal skull. In comparison with gray-scale imaging, the contrast area of PHI can be more easily identified from background anatomical localization.
O19	IS THE CONTRAST TRANSCRANIAL DOPPLER TEST FOR VENOUS TO ARTERIAL CIRCULATION SHUNTS REPRODUCIBLE? S Sastry, KJ Daly, T Chengodu, CN McCollum South Manchester University Hospital, United Kingdom. Background: Paradoxical embolism through venous-to-arterial circulation shunts (v-aCS) as a cause of stroke may be investigated with transcranial Doppler ultrasound (TCD) using agitated saline contrast. We investigated the reproducibility of this simple non-invasive TCD test. Methods: 40 patients aged 16-39 years, suffering ischaemic stroke or myocardial infarction, had a v-aCS test performed on two separate occasions by the same investigator and a third test performed by a second blinded investigator. Up to two contrast injections at rest, two with cough provocation and two with a Valsalva manoeuvre were performed. Inter and intra-investigator test results were analysed using the kappa test for categorical variables, and the mean bias test and Spearman’s rank correlation (rs) for continuous variables. Results: The repeated test results (positive/negative) were identical for investigator 1 in 39/40 cases (kappa 0.935). There was agreement on shunt size (small/moderate/large) for investigator 1 in 36/40 cases (kappa 0.842). Good agreement was also found for the highest number of bubbles detected after any injection (mean bias –2.67 [95% CI -23.95, 14.59]; rs=0.95 [0.91, 0.98]). Positive/negative test results were the same for investigators 1 and 2 in 35/38 cases (kappa 0.813), and there was agreement for shunt size in 31/38 cases (kappa 0.738). There was good agreement for the highest number of bubbles detected (mean bias 3.24 [-11.08, 21.82]; rs=0.92 [0.84, 0.96]). Conclusions: This study shows that the contrast TCD test for investigation of a v-aCS has good inter- and intra-investigator reproducibility.
O20	OPTIMIZING THE TECHNIQUE OF CONTRAST TRANSCRANIAL DOPPLER ULTRASOUND IN THE DETECTION OF RIGHT-TO-LEFT SHUNTS ^1Ralf Dittrich, MD, ^1Dirk W. Droste, MD, ^1Stefan Lakemeier, ^2Thomas Wichter, MD, ^2Jörg Stypmann, MD, ^1Martin Ritter, ^3Martin Moeller, MD, ^3Michael Freund, MD, ^1E. Bernd Ringelstein, MD ^1Department of Neurology, ^2Department of Cardiology and Angiology, and ^3Department of Clinical Radiology, University of Münster, Germany Background: Cardiac right-to-left shunts can be identified by transesophageal echocardiography (TEE) and by transcranial Doppler ultrasound (TCD) using contrast agents and a Valsalva maneuver (VM) as a provocation procedure. This paper applies the modalities of these latter tests elaborated in previous studies to a large patient cohort and compares two contrast agents (saline and EchovistÒ-300) Methods: 81 patients were investigated by both TEE and bilateral TCD of the middle cerebral arteries. The following protocol with injections of 10 mL of agitated saline was applied in a randomized way: (1) no VM, (2) VM for 5 sec starting 5 sec after the beginning of contrast injection, (3) repetition of the test with VM, if the first test with VM was negative. The VM was performed for 5 sec starting exactly 5 sec after the begin of saline injection. Thereafter, the same protocol was repeated using 10mL of EchovistÒ-300 instead of saline. Results: 31 patients had a cardiac RLS. The EchovistÒ-300 investigation disclosed all these 31 shunts, but only 29 of them were disclosed by saline. 22 had a RLS only in at least one of the above TCD-tests, some of them even with a considerable shunt volume. Conclusions: Contrast-TCD performed with EchovistÒ-300, but not with saline, yields a 100% sensitivity to identify TEE-proven cardiac right-to-left shunts. The TCD- test should be repeated if negative at the first time. This manuscript gives detailed information for the optimization of the contrast-TCD technique. Extracardiac shunts only detected during contrast-TCD can have a considerable shunt volume and may also allow for paradoxical embolism.
O21	Intracranial pulseless PATTERN in PATIENTS WITH Takayasu arteritis: A TRANSCRANIAL DOPPLER STUDY. C. Cantú, F. Barinagarrementería, C. Pineda, M. Martínez-Lavín, Instituto Nacional Ciencias Médicas e Instituto Nacional Cardiología, Mexico Background: TCD may give important information about intracranial hemodynamic effects of arterial damage in the extracranial cerebral circulation in pts with Takayasu arteritis (TA). Objective: To describe TCD findings in TA pts and to investigate the clinical significance of an altered cerebral hemodynamics. Patients and Methods: Twenty-five consecutive patients who fulfilled ACR criteria for diagnosis of TA, independently of the presence or absence of neurologic complaints. TA was corroborated by conventional aortography in all cases. Patients age ranged from 12 to 40 years with a mean of 31.6 years. Cerebral hemodynamics was assessed by TCD, including the pulsatility index (PI) (normal = 0.87± 0.16). Results: 12 of 25 patients exhibited diverse degrees of impairement in cerebral hemodynamics, characterized by a damping of the waveform with a delayed upstroke, diminished flow velocity, slow acceleration and decreased PI, with a mean value of 0.42 ± 0.20 (range 0.60 to 0.13). Seven patients displayed a critically reduced pulsatility with a PI of 0.27± 0.14 Normal TCD was recorded in seven patients. In contrast, a high pulsatility spectrum was found in six patients with a mean of 1.52± 0.29, suggesting artery wall stiffness. The clinical relevance of this intracranial pulseless pattern was reflected in a higher frequency of stroke (10 vs. 3), syncope (5 vs. 0) and ischemic visual disturbances (7 vs. 2) when compared to those with a normal or highly pulsatile spectra. Conclusion: A critically reduced pulsatility intracranially in some TA patients was associated with a higher frequency of cerebral and retinal ischemic events.