Knowledge of the relationship between in vivo contrast agent concentration and magnetic resonance (MR) signal response is an important requirement in contrast enhanced MR imaging in general and in MR based perfusion imaging in particular. This relationship is a complex function of the properties of the contrast agent as well as the structure of the target tissue. The aim of the present work was to quantify the effects of the iron oxide nanoparticle based intravascular contrast agent, NC100150 Injection, on proton relaxation rates in different tissue systems in vivo in a pig model and ex vivo in phantoms containing whole blood. Methods that enabled accurate relaxation rate measurements in these organs were developed, and validated. From these measurements, trans-compartmental water exchange rates and blood volume could be estimated and the MR signal response could be predicted as a function of contrast agent concentration under relevant imaging conditions. Using a 1.5 Tesla clinical MR system, the longitudinal (R1=1/T1) proton relaxation rates in blood, renal cortex, paraspinal muscle and myocardium were measured in vivo as a function of plasma concentration (Cp) of NC100150 Injection. The transverse (R2* = 1/T2*) relaxation rates were measured in vivo in blood, renal cortex and muscle as a function of Cp and ex vivo in blood as a function of Cp and blood oxygenation tension. The proton nuclear MR (NMR) linewidth and lineshape were analysed as a function of Cp and blood oxygen tension ex vivo at 7.05 T. In muscle and renal cortex, there was a linear correlation between R2* and Cp whereas R2* increased as a quadratic function of Cp in blood. The NMR linewidth increased linearly with Cp in fully oxygenated blood whereas in deoxygenated blood the linewidth initially decreased with increasing Cp, reaching a minimum and then increasing again with further increase in Cp. R1 increased linearly with Cp in blood and from the slope of R1 vs. Cp the T1-relaxivity (r1) of NC100150 Injection in blood at 1.5 T was estimated to be (mean ± SD) 13.9 ± 0.9 s-1mM-1. In tissue, the maximum increase in R1 was limited by the rate of water exchange between the intravascular and interstitial tissue compartments. Using a two-compartment exchange-limited relaxation model, the permeability surface area (PS) product was estimated to be 61.9 ± 2.9 mL/min/g in renal cortex and 10.1 ± 1.5 mL/min/g in muscle and the total myocardial water exchange rate, kt, was 13.5 ± 6.4 s-1. The estimated blood volumes obtained from the same model were 19.1 ± 1.4 mL/100 g, 2.4 ± 1.4 mL/100 g and 11.2 ± 2.1 mL/100 g, respectively in renal cortex, muscle and myocardium. Current T2* based first-pass MR perfusion methods assume a linear correlation between R2* and Cp both in blood and tissue and our results therefore suggest that quantitative perfusion values can not easily be obtained with existing tracer kinetic models. The correlation between MR signal response and Cp is further complicated in the kidney by a significant first-pass increase in R1 which may lead to an underestimation of Cp. In T1-based perfusion methods, low concentrations of NC100150 Injection must be used in order to maintain a linear dose-response relationship between R1 and Cp. The effect of blood oxygenation on the NMR linewidth in the presence of NC100150 Injection enabled accurate estimation of magnetic susceptibility of deoxyhemoglobin and the effect can potentially be used to determine blood oxygenation status. In conclusion, NC100150 Injection is well suited as a T2* perfusion agent due to the large magnetisation and intravascular biodistribution of this agent. T1-based perfusion imaging with this agent is limited by water exchange effects and large T2* effects at higher contrast agent concentrations. Quantitative perfusion assessment is unlikely to be feasible with any of these approaches due to the non-linear dose response.

Introdução: A presença de cálcio nas artérias coronárias identifica a existência de aterosclerose coronariana, mesmo em fases precoces. Por outro lado, a decisão quanto à melhor forma de tratamento dessa entidade baseia-se no uso de exames funcionais, em especial, a cintilografia de perfusão do miocárdio (CPM). Existem dúvidas quanto à correlação desses dois exames, o que poderia ser, ao menos em parte, explicada pelo fato dos testes avaliarem fases distintas dessa entidade. Este estudo pretende avaliar o uso de uma abordagem anatômica e funcional combinada por meio da (CPM) e tomografia computadorizada coronária com multidetectores (TCMD) na determinação do escore de cálcio (CACS) e da presença de estenoses e isquemia e examinar a associação de seus resultados entre si e com outras variáveis demográficas, clínicas e funcionais em uma população brasileira com doença arterial coronária (DAC) suspeita ou conhecida. Métodos: foram analisados, retrospectivamente, 413 pacientes que se submeteram à CPM e TCMD por indicação clínica do médico assistente, durante o ano de 2009, com intervalo menor de 90 dias, entre os dois estudos. CACS foi definido automaticamente pelo software dedicado e o escore de Agatston foi calculado de forma semiquatitativa. Na CPM e na TCMD, os resultados foram obtidos por quantificação visual semiobjetiva. A correlação dos resultados dos exames foi analisada e foram obtidos dados da evolução tardia, baseados no contato com o médico assistente ou pela análise dos registros hospitalares. Procurou-se definir os preditores da ocorrência de eventos cardiovasculares adversos na evolução. Resultados: Foram selecionados 303 pacientes (73,3% homens, média de idade de 55,8 ± 10,6 anos, intervalo: 32-86 anos). Destes, 73,3% apresentaram estudos de perfusão normal, 71,6% tinham cálcio coronário e 45,2% tinha perfusão normal e TCDM com cálcio. No grupo sem suspeita de DAC (177), houve associação entre os resultados de TCMD e SPECT com: diabetes (DM) (p=0,045), hipertensão (HAS) (p=0,032), dislipidemia (p=0,030) testes funcionais, resultado do teste (p=0,022), percentil escore de cálcio (p<0,001) e CACS (p<0,001). Nos pacientes sem DAC, houve associação com defeitos de perfusão e TCMD com cálcio que mostraram em casos de DM (30,0%) e CACS igual ou acima do percentil 75 (60,0%) e com valores acima de 400 (40,0%). O grupo com TCMD e perfusão normal apresentou maior percentual de ausência de HAS (78,6%). De outra forma, o grupo com perfusão normal e TCMD com cálcio mostrou mais pacientes com dislipidemia (42,5%) e distribuição dos percentis de CACS e CACS, semelhantes ao grupo com defeitos de perfusão e TCMD com cálcio. No grupo dos pacientes com DAC conhecida, também houve associação entre CACS e a extensão da aterosclerose coronária. Não houve associações entre os resultados da cintilografia e TCMD e outras variáveis. No subgrupo de 128 pacientes, em que se conseguiu seguimento tardio (média de 824,5 dias; DP de 385,9), aconteceram dois óbitos (1,6%), nenhum de causa cardiovascular. O evento mais frequente foi cinecoronariografia (CAT) (21,1%), seguida da angioplastia (ATC) (9,4%) e revascularizações cirúrgicas (3,1%). Não houve infarto do miocárdio (IM). De forma similar, não foram observados determinantes significativos da evolução tardia desses pacientes. Conclusões: Embora o CAC tenha sido tão frequente, como perfusão normal no SPECT, menos da metade daqueles com SPECT normal podem apresentar cálcio coronário à TCMD. Esta combinação de resultados (SPECT normal e cálcio) tinha associação com dislipidemia. Além disso, DM, CACS > 400 e percentil igual ou superior a 75% foram associados com SPECT anormal e cálcio na MDCT. A MDTC combinada ao SPECT foi capaz de detectar isquemia em pacientes com estenose coronariana conhecida, mas não havia associações entre o grau de estenose ou a presença, ou ausência de isquemia e sua extensão, com os resultados combinados. Houve uma baixa taxa de eventos no subgrupo de acompanhamento. / Background: The calcium in coronary arteries identifies the presence of coronary atherosclerosis, even in early stages. On the other hand, the decision about the best form of treatment this entity is based on the use of functional tests, in particular the myocardial perfusion scintigraphy (SPECT). There are doubts about the correlation between the results of these two exams, what could be, at least in part explained by the fact of these tests assess distinct phases of this entity. This study aims to evaluate the use of a combined functional and anatomical approach through SPECT and coronary computed tomography with multi-detectors (MDCT) in determining the calcium score (CACS),presence of stenosis or ischemia and examine the combined results with other demographic, clinical and functional variables, in a Brazilian population with suspected or known coronary artery disease (CAD). Methods: we retrospectively analyzed 413 patients who underwent to SPECT and MDCT by their physician indications during the year 2009, with less than 90 days interval between the two studies. CACS was automatically defined by dedicated software and the Agatston score was semi-automatically calculated. SPECT and MDCT results were evaluated by semi-objective visual quantification. The correlations of both tests results were analyzed and follow-up data were obtained through contact with assistant physician or analysis of hospital records. Results: We included 303 patients with suspect or known DAC who underwent to both, SPECT and MDCT (75.9 % men, mean age 55.8 ± 10.6 years; range:32-86 years); it was observed that 73,3% had normal perfusion studies, CAC was present 71,6% patients and 48.2% had normal perfusion and MDCT with calcium. In 177 patients without DAC there were association between MDCT and SPECT results and following variables: diabetes (DM) (p=0.045), hypertension (HAS) (p=0.032), dyslipidemia (p=0.030), functional test results (p=0.022), percentile of calcium score (p<0.001) and CACS (p<0.001). Patients without DAC, with perfusion defects and calcium showed more occurrences of DM (30.0%) and CACS equal to or above the 75 percentile (60.0%) and with values above 400 (40.0%). The group with normal MDCT and perfusion had higher percentage of absence of HAS (78.6%). The group with normal perfusion and MDCT with calcium had the highest percentage of patients with dyslipidemia (42.5%) and distribution of the calcium score percentiles and CACS were similar to the group with perfusion defects and MDCT with calcium. In the group with known CAD there was association between DM, extent of coronary disease and CACS. There were no associations between the results of SPECT and MDCT and other variables. In the subgroup of 128 patients with follow-up, the average time was 824.5 days (SD 385.9), there were 2 deaths (1.6%) confirmed, however none of cardiovascular cause. Coronary angiography (CATH) (21.1%) was the most frequent event, followed by percutaneous angioplasty (PTCA) (9.4%). Surgical revascularizations were much less frequent (3.1%). There was no myocardial infarction (MI). Similarly, there were no markers of long term prognosis in this sample. Conclusions: Although CAC was as frequent as normal perfusion in SPECT, less than half with normal SPECT may have calcium on MDCT. This results combination (normal SPECT and calcium) had association with dyslipidemia. Also, DM, CACS >400 and percentile equal or above 75% were associate with abnormal SPECT and calcium on MDCT. MDCT combined to SPECT was able to detect ischemia in patients with known coronary stenosis, but there were no associations between the degree of stenosis, or the presence or absence of ischemia and its extent with the combined results. There was a low event rate in the follow-up subgroup.

Introdução e Objetivos: A craniectomia descompressiva com duroplastia (CDD) reduz a mortalidade e melhora o prognóstico funcional em doentes com acidente vascular encefálico isquêmico (AVEi) hemisférico e proporciona a redução da pressão intracraniana. Entretanto, pouco se sabe sobre sua repercussão na hemodinâmica cerebral. O objetivo do presente trabalho é o de avaliar com a tomografia computadorizada com estudo de perfusão (TCP) as alterações hemodinâmicas nos doentes com AVEi após a CDD e identificar possíveis marcadores prognósticos substitutos. Métodos: Foram avaliados 27 doentes com AVEi com indicação de CDD. Os parâmetros hemodinâmicos da TCP estudados no período pré-operatório e em até 24h após a cirurgia foram: duração média de trânsito (DMT), volume sanguíneo encefálico (VSE) e fluxo sanguíneo encefálico (FSE). O desfecho primário utilizado foi a melhora ou a ausência de melhora hemodinâmica. Os desfechos secundários foram a escala de Rankin modificada em seis meses, dicotomizada como favorável (0-3) ou desfavorável (4-6); casos fatais em um mês e em seis meses. Resultados: 18 (70,3%) doentes eram do sexo feminino e 12 (44,4%) tinham idade superior a 55 anos. Houve melhora da DMT (queda de 8,74 para 8,24, p=0,01) e tendência a melhora do FSE (aumento de 22,37 para 25,26, p=0,06) após a CDD. Não houve diferença estatística em relação ao VSC (aumento de 2,14 para 2,26, p=0,33). A idade superior a 55 anos foi o preditor independente de prognóstico desfavorável (p=0,03) e a DMT pré-operatória, foi preditora hemodinâmica para mortalidade em seis meses (8,20 vs 9,23, p=0,04). Conclusões: A craniectomia descompressiva com expansão dural determinou melhora hemodinâmica na maioria dos doentes com AVEi hemisférico. A DTM préoperatória é um bom marcador substituto para a possibilidade de óbito em seis meses / Background and Objectives: Decompressive craniectomy (DC) reduces the mortality and improves the functional outcome in patients with malignant cerebral infarction (MCI). This procedure causes a decrease of the intracranial pressure, however, little is known about its impact in brain hemodynamics. Therefore, our goal is to study through CT perfusion the hemodynamics changes that may occur in patients with MCI after the DC. Methods: 27 patients with MCI treated with DC were studied. The CT perfusion hemodynamic parameters - the mean transit time (MTT), the cerebral blood volume (CBV) and cerebral blood flow (CBF) - were evaluated preoperatively and within the first 24 hours after the DC. The primary outcome measure was improvement or lack of improvement in cerebral hemodynamics. Secondary outcomes were the modified Rankin scale in 6 months, classified as favorable (0-3) and unfavorable (4-6); and, fatal cases at 1 month and 6 months. Results: 18 (70.3%) patients were female and 12 (44.4%) were older than 55 years. There was improvement of MTT (decrease from 8.74 to 8.24, p = 0.01) and a trend towards improvement of the CBF (increase from 22.37 to 25.26, p = 0.06) after DC. There was no statistical difference in the CBV before and after DC (increase from 2.14 to 2.26, p = 0.33). Patients over 55 years had poorer prognosis (p=0.03) and preoperative MTT was an independent hemodynamic predictor of mortality at 6 months (8.20 vs 9.23, p=0.04). Conclusions: DC improved cerebral hemodynamics in most patients with malignant ischemic stroke. Preoperative MTT seems to be a good marker for case fatality in 6 months

L'IRM de perfusion et de spectroscopie restent encore peu utilisées en raison de leur mise en oeuvre difficile et de leur manque de quantification. L'objectif de ces travaux a été d'optimiser et de valider des techniques IRM totalement non invasives chez l'Homme en vue d'applications cliniques permettant une exploration sur un large volume cérébral et une quantification absolue des paramètres de perfusion et du métabolisme cérébraux. Concernant la perfusion, 3 séquences de type marquage de spins,PASL PICORE, PASL FAIR et pCASL, ont été comparées en termes de sensibilité et de reproductibilité. pCASL a ensuite été intégrée dans un protocole de recherche sur des patients atteints de sclérose en plaques ou SEP. Quant au métabolisme cérébral, un protocole a été mis en place afin d'accéder à une quantification absolue et pseudo absolue des métabolites par la normalisation du signal de l'eau issue de la CSI par la densité de protons acquise en IRM. Cette technique a été validée en CSI 2D puis transposée en 3D avec la séquence EPSI sur deux orientations différentes : CACP et CACP+15°afin de constituer des valeurs normatives fiables des métabolites principaux sur tout le cerveau. L'élaboration de ces techniques en spectroscopie a abouti à une étude sur des patients souffrant de SEP démontrant la faisabilité de l'utilisation de ces techniques en clinique. Ces travaux démontrent que la quantification absolue en IRM de perfusion et en IRM de spectroscopie peut être obtenue sur un large volume cérébral de manière fiable sur un système IRM disponible en environnement clinique dans un temps d'acquisition acceptable à travers les corrections diverses spécifiques à chaque imagerie. / Conventional MRI's lack of specificity in clinical routine limits our ability to perform correct diagnoses or follow-ups of pathological diseases. Two forms of NMR imaging, perfusion weighed and spectroscopic imaging provide information about two closely related characteristics :cerebral perfusion and metabolism. However, these techniques are not widely used due to the complexity of implementation and a lack of quantification.The general aim was to optimize and validate completely non-invasive NMR techniques for further human clinical applications in the context of exploring large cerebral volumes and determining absolute or pseudo-absolute quantification of cerebral perfusion and metabolism. Concerning perfusion, three arterial spin labeling sequences, PASL PICORE, PASL FAIR and pCASL, were compared in terms of sensitivity and reproducibility. The pCASL sequence was then integrated to a protocol applied to patients suffering from multiple sclerosis. In relation to metabolism, a protocol was applied in order to access absolute and pseudo-absolute metabolite quantification by water SI normalization from MRI proton density. This technique was validated on 2D CSI and then on 3D with EPSI sequence with two orientations, AC-PC and AC-PC+15 in order to generate reliable normative values of metabolites for the whole brain. The use of those spectroscopic techniques on patients suffering from multiple sclerosis allowed demonstrating the feasibility in clinic.This work demonstrates that reliable absolute quantification in perfusion weighted and spectroscopic imaging can be obtained with extensive coverage and with an acquisition time compatible with the reality of clinical exams.

The purpose of this study is to introduce the methods of the deconvolution and to programme some of them. For the simulation, the tissue homogeneity model and the model of arterial input fiction were used. These models were engaged as the test procedures with the aim of verify the functionality and utility of the Wiener filter, the Lucy-Richardson algorithm and the Singular value decomposition.

The result of this thesis, focused on medical image registration, is an automatic image registration algorithm. It is constructed to be used on real ultrasonography images, created by perfusion imaging. In its introductory part, the thesis deals with registrations methods, next it describes types of optimization principles and single criteria functions served to determine correct image transformation. Based on the theoretical part, there are realized three optimization algorithms using three criterial functions, which served to registration of provided ultrasonography sequences. These algorithms are tested and results are passed on analysis, on its ground are judged its advantages and disadvantages.

Single photon emission tomography (SPECT) data from myocardial perfusion imaging (MPI) are normally displayed as a set of three slices orthogonal to the left ventricular (LV) long axis for both ECG-gated (GSPECT) and non-gated SPECT studies. The total number of slices presented for assessment depends on the size of the heart, but is typically in excess of 30.  A requirement for data presentation is that images should be orientated about the LV axis; therefore, a set of radial slice would fulfill this need. Radial slices are parallel to the LV long axis and arranged diametrically. They could provide a suitable alternative to standard orthogonal slices, with the advantage of requiring far fewer slices to adequately represent the data. In this study a semi-automatic method was developed for displaying MPI SPECT data as a set of radial slices orientated about the LV axis, with the aim of reducing the number of slices viewed, without loss of information and independent on the size of the heart. Input volume data consisted of standard short axis slices orientated perpendicular to the LV axis chosen at the time of reconstruction.  The true LV axis was determined by first determining the boundary on a central long axis slice, the axis being in the direction of the y-axis in the matrix. The skeleton of the myocardium were found and the true LV axis determined for that slice. The angle of this axis with respect to the y-axis was calculated. The process was repeated for an orthogonal long axis slice. The input volume was then rotated by the angles calculated. Radial slices generated for presentation were integrated over a sector equivalent to the imaging resolution (1.2 cm); assuming the diameter of the heart is about 8cm then non-gated data could be represented by 20 radial slices integrated over an 18 degree section. Gated information could be represented with four slices spaced at 45 intervals, integrated over a 30 degree sector.

No description available.

Nous proposons dans le cadre de cette thèse une nouvelle approche permettant de s'affranchir de l'atténuation mammaire en tomographie par émission monophotonique (TEMP) de perfusion myocardique. Elle est constituée de deux parties : - la première consiste à rendre les projections acquises consistantes. - la deuxième consiste à pondérer ces même les projections corrigées durant la reconstruction. Nous avons effectué l'étude de validité de nos méthodes sur quelques exemples de simulation TEMP de perfusion myocardique simulant l'atténuation mammaire et sur quelques exemples d'études patients réelles notamment : des cas d'atténuation mammaire, d'infarctus inférieure, d'infarctus apical, d'infarctus antérieur, d'ischémie antérieure et inférieure. Les résultats semblent encourageants. Il s'agit dans le proche avenir de mener une étude de validation chez les patients versus un gold standard (coronarographie, coroscanner) / We propose in this thesis a new approach to correct the breast attenuation in SPECT myocardial perfusion imaging. It consists of two parts: -The first is to make the acquired projections consistent with each other. - The second is to weight the corrected attenuated projection during the reconstruction. We conducted a validation of our methods on some examples of myocardial perfusion SPECT imaging simulating the breast attenuation and some examples of real patient studies including: breast attenuation, anterior myocardial infarction, inferior myocardial infarction, anterior myocardial ischemia and inferior myocardial ischemia. The obtained results are encouraging. At this step, it is interesting in the near future to conduct a validation study in patients versus a gold standard (angiography, coroscan).Key words: SPECT, tomographic reconstruction, breast attenuation, Iterative reconstruction, attenuation correction, myocardial perfusion imaging, nuclear medicine

Apesar dos avanços no entendimento da fisiopatologia e dos recursos terapêuticos atualmente disponíveis, a cardiomiopatia dilatada (CMD) permanece como uma condição com alta mortalidade, sendo que a disfunção microvascular é um dos mecanismos relacionados à piora da função cardíaca. Nos últimos anos, novas técnicas de ecocardiografia vêm sendo utilizadas para avaliação da disfunção microvascular, incluindo a medida de velocidade de fluxo coronário pelo Doppler da artéria coronária descendente anterior (ADA), e a análise quantitativa do fluxo miocárdico pela ecocardiografia com perfusão miocárdica em tempo-real (EPMTR). No presente estudo, avaliamos o valor prognóstico da reserva de velocidade de fluxo coronário (RVFC), obtida pelo Doppler da ADA, e da reserva de fluxo miocárdico (RFM), obtida pela EPMTR, para predizer morte e transplante cardíaco em pacientes com CMD de origem não isquêmica. Adicionalmente, avaliamos se as medidas de reserva de fluxo acrescentam valor prognóstico sobre variáveis clínicas e ecocardiográficas que já são conhecidas como preditores de eventos nesta população. Estudamos 195 pacientes com CMD (130 homens, média etária 54 ± 12 anos) que apresentavam fração de ejeção do ventrículo esquerdo inferior a 35% pelo ecocardiograma e ausência de sinais de doença arterial coronária obstrutiva por angiografia coronária invasiva ou por método não invasivo (cintilografia de perfusão miocárdica ou angiotomografia de coronárias). Foram analisados parâmetros ecocardiográficos convencionais de função sistólica e diastólica do ventrículo esquerdo em repouso. A velocidade de fluxo coronário foi determinada pelo Doppler pulsado na ADA e a dinâmica das microbolhas no miocárdio foi quantificada pela EPMTR utilizando programas computacionais específicos, tanto em repouso como durante o estresse pelo dipiridamol (0,84 mg/Kg). As RVFC, RFM e reserva de velocidade de repreenchimento de microbolhas no miocárdio (reserva ) foram obtidas pela relação entre os parâmetros de fluxo durante a hiperemia e em repouso, sendo consideradas diminuídas quando os valores estavam abaixo de 2,0. O tempo médio de acompanhamento foi de 29 meses (variando de 6 a 69 meses). Neste período, 45 pacientes (24%) apresentaram eventos, sendo 43 mortes de causa cardíaca e 2 transplantes cardíacos. Na análise univariada foram preditores de eventos: etiologia chagásica, classe funcional de insuficiência cardíaca, uso de inibidores da enzima conversora de angiotensina e/ou bloqueadores dos receptores de angiotensina II (fator protetor), diâmetros ventriculares e do átrio esquerdo, volumes ventriculares, fração de ejeção, disfunção diastólica, grau de insuficiência mitral, RVFC, RFM e reserva . Na análise multivariada, foram preditores independentes de eventos o diâmetro do átrio esquerdo (razão de chances = 1,16 por unidade de aumento; intervalo de confiança 95% = 1,078 1,264; p<0,001) e a reserva diminuída (razão de chances = 3,219; intervalo de confiança 95% = 1,178 8,795; p<0,001). No modelo sequencial de predição de eventos, fração de ejeção e diâmetro do átrio esquerdo adicionaram valor prognóstico às variáveis clínicas (aumento do qui-quadrado de 15,2 para 58,5; p < 0,001). A reserva adicionou poder ao modelo (qui-quadrado de 70,2; p<0,001). Concluímos que tanto a RVFC, obtida pelo Doppler da ADA, como a RFM e reserva , obtidas pela EPMTR sob estresse pelo dipiridamol, são preditoras de morte e transplante cardíaco em pacientes com CMD não isquêmica. Entretanto, somente o diâmetro do átrio esquerdo e a reserva mostraram valor prognóstico independente e incremental sobre variáveis clínicas e ecocardiográficas que já são conhecidas como preditores de eventos nesta população / Despite advances in understanding of the pathophysiology and therapeutic approaches, dilated cardiomyopathy (DCM) remains as a condition with high mortality and one of the mechanisms involved in this process seems to be microvascular dysfunction. Recently, new echocardiographic techniques have been incorporated in the clinical practice and used for the assessment of microvascular dysfunction, including evaluation of left anterior descending coronary artery (LAD) by Doppler and quantitative analysis of myocardial flow by real-time myocardial perfusion echocardiography (RTMPE). In the present study, we evaluated the prognostic value of coronary flow velocity reserve (CFVR), obtained by transthoracic echocardiography, and parameters of myocardial flow reserve (MFR), obtained by RTMPE for predicting cardiac death and heart transplantation in patients with nonischemic DCM. In addition, we sought to detect the incremental value of flow reserve over clinical and echocardiographic parameters already known as predictors of events in this population. We studied 195 patients with DCM (130 men, mean age 54 ± 12 years) who had left ventricular ejection fraction (LVEF) less than 35% by echocardiography and no signs of obstructive coronary artery disease by invasive coronary angiography or noninvasive methods (nuclear medicine or angiography by computed tomography). We assessed conventional echocardiographic parameters of systolic and diastolic left ventricular function at rest. Coronary flow velocity was determined by pulsed Doppler in LAD and dynamics of microbubbles in the myocardium was measured by RTMPE using specific computer programs, both at rest and during dipyridamole stress (0.84 mg/kg). CFVR, MFR and the reserve of velocity of microbubbles in the myocardium ( reserve) were obtained as the ratio between parameters of flow during hyperemia and at rest, being considered abnormal when these were below 2.0. The mean follow-up was 29 months (ranging from 6 to 69 months). During this period, 45 patients (24%) had events, 43 cardiac deaths and 2 heart transplantations. In the univariate analysis, predictors of events were: Chagas disease, New York Heart Association functional class, use of angiotensin-converting enzyme inhibitors/angiotensin II receptor blocker (protective effect), left ventricular and atrial diameters, left ventricular volumes, LVEF, diastolic dysfunction, degree of mitral regurgitation, CVFR, MFR and reserve. By multivariate analysis, the only independent predictors of events were the left atrial diameter (OR = 1.16, 95% confidence interval = 1.078 to 1.264, p <0.001) and reserve 2.0 (OR = 3.219, confidence interval 95 % = 1.178 to 8.795, p <0.001). In the sequential model of events prediction, LVEF and left atrial diameter added prognostic value over clinical factors (chi-square from 15.2 to 58.5; p<0.001). reserve added power to the model (chi-square = 70.2, p<0.001). We concluded that CVFR, obtained by Doppler in LAD, MFR and reserve, obtained by dipyridamole stress RTMPE, are predictors of cardiac death and heart transplantation in patients with nonischemic DCM. However, only left atrial diameter and depressed reserve showed independent and incremental predictive value beyond that provided by current known prognostic clinical and echocardiographic factors