Stroke is characterized with loss of one or more functions of the body resulted by inadequate blood supply to the brain. Most of the cases result from a blood clot forms on an atherosclerotic plaque in the brain which is called as ischemic stroke. Structure of the arteries and vascular tone are listed in major determinants in the development of the disorder. Soluble epoxide hydrolase (sEH, EPHX2) catalyzes conversion of epoxyeicosatrienoic acids to inactive diol metabolites. EETs are potent vasodilators that participate in the regulation of vascular tone and cerebral blood flow. Microsomal epoxide hydrolase (mEH, EPHX1) is a critical phase I enzyme that catalyzes the conversion of various xenobiotic epoxide substrates and polycyclic aromatic hydrocarbons (PAHs). Animal studies show that tobacco smoke mutagens such as PAHs and heterocyclic amines directly increase the development of atherosclerotic lesions. The main purpose of this study is evaluation of effect of Arg287Gln single nucleotide polymorphism of EPHX2 gene and Tyr113His and His139Arg single nucleotide polymorphisms of EPHX1 gene as a risk factor for ischemic stroke in Turkish population. Blood samples of 237 ischemic stroke patients and 120 controls were collected and all polymorphisms were determined by PCR-RFLP method. Mutant allele frequencies in terms of Arg287Gln polymorphism of EPHX2 gene (A) were found as 0.08 for patient group and 0.09 for controls. Tyr113His polymorphism of EPHX1 gene (C) were found as 0.27 for patient group and 0.31 for controls when, His139Arg polymorphism of EPHX1 gene (G) were 0.820 and 0.814 for patient and control groups, respectively. The differences between mutant allele frequencies of patients and controls were not found to be statistically significant. Subgroup analysis was used to investigate the effects of conventional vascular factors according to the genotypes in the stroke susceptibility. Smoking, diabetes, obesity and hypertension were found to significantly increase the risk of having stroke. More detailed analysis on these factors with respect to genotypes showed that the risk of hypertensive individuals having ischemic stroke was higher in wild type homozygous genotype groups of Tyr113His (TT) and His139Arg (AA) polymorphisms and heterozygous and mutant homozygous genotypes of Arg287Gln (GA+AA) polymorphism than their counterparts (OR= 3.21, 3.15 and 4.69, respectively). Smoker people within the heterozygous and mutant homozygous genotypes group of Arg287His (GA+AA) polymorphism and wild type homozygous group of His139Arg (AA) polymorphism were found to be more susceptible to have stroke (OR= 11.81 and 4.78 respectively). Finally, diabetes mellitus was found to double the risk of having stroke regardless of the genetic background. Logistic regression analyses were used to ascertain the effects of vascular factors, lipid parameters and genotypes in the stroke susceptibility. LDL-cholesterol (OR=1.46 / 95%CI, 1.12-1.89, P=0.00), smoking (OR=3.46 / 95%CI, 1.66-7.21, P=0.00) and hypertension (OR=3.19 / 95%CI, 1.92-5.30, P=0.00) were found to be significant risk factors for ischemic stroke, whereas HDL (OR=0.27 / 95%CI, 0.12-0.65, P=0.02) was found to be a protective factor in general population. In this study, the relation of Tyr113His and His139Arg polymorphisms of EPHX1 gene and risk of ischemic stroke is investigated for the first time in literature while, Arg287Gln polymorphism and ischemic stroke risk in Turkish population was studied for the first time.

African Americans experience a disproportionate mortality, morbidity, and disability associated with ischemic stroke. Traditional risk factors offer some explanation for this finding, but novel risk factors have not been explored. APOE4 genotype, which is more prevalent in African Americans demonstrate a pro-inflammatory phenotype that may result in an exaggerated inflammatory response associated with ischemic stroke, resulting in worse outcomes. The purpose of this study was to examine relationships among APOE genotypes, inflammatory markers (CD11β, platelet leukocyte aggregates, IL-1β, IL-6, IL-8, and tissue necrosis factor alpha), the anti-inflammatory marker, IL-10, and risk factors (hypertension, diabetes type II, obesity, hyperlipidemia, and smoking) in African Americans at 3 days post stroke. Twenty five patients were enrolled with 12 patients in the APOE4 group and 13 patients in the non-APOE4 group. In the APOE4 group, 75% were male compared to 54% in the non-APOE4 group. The average age in the APOE4 group was 56.5 ± 9.0 compared to 66 ± 16.0 years in the non-APOE4 group. Females in the APOE4 group were younger with ages comparable to men. All participants had hypertension. Forty two percent of patients in the APOE4 group had two risk factors and 46% of patients in the non-APOE4 group had three risk factors. Major findings included 1) there were no statistical difference between inflammatory markers and APOE genotypes, and 2) the APOE4 carrier was not a predictor for overall inflammatory load among African Americans with ischemic stroke. The study was underpowered and small effect sizes were not sufficient to create statistical findings. This was the first study to examined APOE genotypes, inflammatory markers, and risk factors among African Americans with ischemic stroke. More studies are needed to not only investigate novel risk factors, but to also characterize inflammatory and genetic mechanisms with ischemic stroke and their associated outcomes among African Americans. Such studies may lead to primary and secondary prevention of ischemic stroke and reduce the health disparities associated with ischemic stroke among African Americans.

Disertacijoje nagrinėjamos medicininių vaizdų analizės problemos, taikant vaizdų spektrus. Pagrindiniai tyrimo objektai yra galvos kompiuterinės tomogramos bei magnetinio rezonanso angiogramos. Šie objektai yra svarb��s medicininės diagnostikos rezultatai, kurių detali analizė bei interpretacija leidžia parinkti tinkamiausią paciento gydymo būdą. Pagrindinis disertacijos tikslas yra sukurti metodikas ir algoritmus, skirtus automatiniam min÷tų vaizdų apdorojimui bei analizei. Sukurtų algoritmų taikymo sritis yra medicininė programinė įranga. Taip pat disertacijoje siekiama ištirti dažninio ir erdvinio dažninio spektrų taikymo galimybes medicininių vaizdų apdorojime ir analizėje, siekiant aptikti ir pažymėti vaizduose ligų sritis arba kitas specifines sritis. Darbe sprendžiami keli pagrindiniai uždaviniai: smegenų ir išeminio insulto sričių aptikimas tomogramose bei kraujagyslių pėdsakų sričių aptikimas angiogramose. Pirmasis uždavinys suformuluotas, atsižvelgiant į palyginti didelį išeminio insulto paplitimą ir jo diagnostikos aktualumą. Antrasis siejasi su aneurizmos srities aptikimo angiogramose problema. Disertaciją sudaro šeši skyriai, iš kurių paskutinis – rezultatų apibendrinimas. / The main idea of current PhD thesis is an analysis of frequency and joint space-frequency representations of medical images. The objects of investigation are digital images of computed tomography and magnetic resonance angiography of human brain. These objects are essential diagnostic results which require a detail visual analysis and interpretation for the successive selection of treatment. The aim of work covers the development of methods and algorithms suitable for automatic analysis of medical images which may be implemented in medical software. The other purpose encompasses an investigation of Fourier and Haar wavelet transforms in the analysis and processing of medical images. The result of analysis is assumed as an enhancement and segmentation of suspected regions of diseases or other specific regions. There are two main tasks under consideration. First one is an automatic detection of region of brain and ischemic stroke in the images of computed tomography. Second one is a detection of traces of blood vessels in magnetic resonance angiography images. The first task is formulated in the view of wide distribution and high relevance of diagnosis of ischemic stroke. Second task is related with the detection of diseases of blood vessels. The thesis is divided into six chapters including the conclusion’s chapter.

1. Ištyrus homocisteino koncentraciją kraujo serume ir trombocitų agregaciją kraujo plazmoje su pagrindiniais natūriniais agregantais (adenozino difosfatas, adrenalinas, kolagenas) sergančiųjų ūminiais galvos smegenų kraujotakos sutrikimais, nustatyta koreliacija tarp:  padidėjusios homocisteino koncentracijos ir insulto su dideliu neurologiniu deficitu, išeminės širdies ligos, kreatinino ir C-reaktyviojo baltymo koncentracijų;  vidinės miego arterijos stenozės laipsnio ir homocisteino koncentracijos kraujo serume;  homocisteino koncentracijos kraujo serume ir amžiaus;  padidintos homocisteino koncentracijos ir pasikartojančio insulto, todėl homocisteino koncentracijos lygis gali būti panaudotas kaip prognostinis rodiklis naujam ar kartotiniam ūminiam galvos smegenų kraujotakos sutrikimui išsivystyti. 2. Trombocitų agregacija turtingoje trombocitais plazmoje ūminio išeminio galvos smegenų kraujotakos sutrikimo periode statistiškai patikimai padidėja insultu sergančiųjų grupėje, nepriklausomai nuo susirgimo sunkumo. 3. Profilaktinės aspirino dozės hiperhomocisteinemijos fone dalį ligonių neapsaugo nuo išeminio insulto išsivystymo. Aspirino vartojimas turi veiksmingesnį antiagregacinį poveikį moterims, nei vyrams: moterų plazmoje trombocitų agregacijos intensyvumas buvo žemesnis su visais agonistais, o vyrų trombocitai, atvirkščiai, į juos reagavo viršnorminiu atsaku. 4. Išeminiu insultu sirgusių ligonių kraujyje dažniau randami policitemija, padidėjęs leukocitų... [toliau žr. visą tekstą] / 1. When examining homocysteine concentration in blood serum and platelet aggregation in plasma using nature aggregants (adenosine diphosphate, adrenalin, collagen) in patients with acute cerebrovascular disorders, the positive corelation was found between the following parameters:  the elevated amount of homocysteine and stroke with the high neurological deficit, ischemic heart disease, creatinin and C-reactive protein concentrations;  the degree of internal carotid artery stenosis and homocysteine concentration in blood serum;  homocysteine concentration in blood serum and age;  elevation of homocysteine concentration and recurrent stroke, thus, homocysteine concentration may be a predisposing indicator to the development of new or recurrent acute cerebrovascular disorder. 2. Platelet aggreagation in platelet-rich plasma during the period of acute cerebrovascular disorder statistically significantly increases in stroke group independent upon severity of illness. 3. Preventive aspirin doses in the light of hyperhomocysteinemia do not protect a certain part of patients from ischemic stroke development (among the debatable causes of this outcome may be aspirin resistance). Aspirin intake has a more effective antiaggregate influence on women than men: the intensity of platelet aggregation in women plasma was lower using all agonists, while in men, on the contrary, platelets responded by reaching levels above normal. 4. The patients, who experienced ischemic stroke... [to full text]

Images of CT and MRI are grayscale images and usually they are analyzed by radiologists on display either on transparency screen. Then objects in images are recognized by their properties like average luminosity of the region, localization, shape, dimensions, etc. The process of visual recognition is affected by many factors. The totality of factors determines uncertainties of diagnostics. More and more often there are used software utilities in the analysis of images which help to lower the uncertainties and speed up the analysis time. The aim of the work is investigation of medical image processing and analysis using frequency and joint space-frequency domains and developing of algorithms suitable for segmentation of regions of diseases. / Žmogaus galvos KT bei MR vaizdus analizuoja apmokyti diagnostikos specialistai – radiologai. Vaizdai, kuriuos turi analizuoti radiologas, yra pilkieji vaizdai, pateikiami monitoriuje arba skaidrėse. Juose tiriami objektai yra atpažįstami pagal tam tikrų vaizdo sričių pilkumo lygį, išsidėstymą, formą, dydį ir kitus požymius. Atpažinimo procesui turi įtakos eilė veiksnių, kurių visuma lemia diagnostines neapibrėžtis. Jų sumažinimui bei diagnostikos paspartinimui vis plačiau taikoma pagalbinė kompiuterinė programinė įranga, kuri, be to, suteikia daug papildomų trimačio vizualizavimo ir skaičiavimo galimybių. Šio darbo tikslas yra ištirti medicininių vaizdų analizės ir apdorojimo galimybes, taikant dažninius ir erdvinius dažninius spektrus, bei sudaryti algoritmus, skirtus ligų sričių ar kitų specifinių vaizdo sričių segmentavimui.

Thermal imaging is a non-invasive and marker-free approach for intraoperative measurements of small temperature variations. In this work, we demonstrate the abilities of active dynamic thermal imaging for analysis of tissue perfusion state in case of cerebral ischemia. For this purpose, a NaCl irrigation is applied to the exposed cortex during hemicraniectomy. The cortical temperature changes are measured by a thermal imaging system and the thermal signal is recognized by a novel machine learning framework. Subsequent tissue heating is then approximated by a double exponential function to estimate tissue temperature decay constants. These constants allow us to characterize tissue with respect to its dynamic thermal properties. Using a Gaussian mixture model we show the correlation of these estimated parameters with infarct demarcations of post-operative CT. This novel scheme yields a standardized representation of cortical thermodynamic properties and might guide further research regarding specific intraoperative diagnostics.

Introdução: Diversas condições clínicas podem modificar a resposta ao tratamento trombolítico no acidente vascular isquêmico agudo. O grau de comorbidade dos pacientes medido pelo Índice de Charlson, um índice que mede o grau de comorbidades clínicas em AVC, tem valor prognóstico na incapacidade pós-AVC tanto em populações com acidente vascular do tipo hemorrágico como isquêmico. Objetivo: Avaliar o efeito do grau de comorbidade aferido pelo índice de Charlson na resposta ao tratamento trombolítico no acidente vascular isquêmico e a incapacidade na alta hospitalar. Métodos: Estudo de coorte prospectivo de 96 pacientes tratados com trombólise para o acidente vascular isquêmico, avaliando o impacto das comorbidades clínicas na resposta ao tratamento trombolítico no AVC isquêmico. Os pacientes foram divididos em dois grupos, aqueles com alto ou baixo grau de comorbidades clínicas, conforme o índice de Charlson. A evolução após o tratamento foi aferida pelo escore de gravidade dos sintomas de acordo com a escala do NIHSS medido antes da infusão, imediatamente após o tratamento, 24horas e 7 dias após a trombólise. A incapacidade na alta foi avaliada pela escala modificada de Rankin sendo, considerada boa resposta a pontuação 0-1 e sua frequência comparada entre os dois grupos de pacientes. Resultados: A comparação dos escores médios do NIHSS mostraram diferenças significativas nos diferentes momentos entre os grupos de alta e baixa comorbidade (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Pacientes com índice de comorbidade baixo apresentaram redução do escore do NIHSS de 10.13 para 2.9, enquanto que no grupo com alta comorbidade, o tratamento trombolítico demostrou pouco efeito. Uma boa evolução, definida como incapacidade 0 e 1 na escala modificada de Rankin, foi observada em (73%) dos pacientes com baixo índice de comorbidade, enquanto somente (15%) dos pacientes com alto índice de comorbidade apresentaram essa evolução favorável, uma diferença clinicamente muito significativa (RR 5.62; 95% CI = 2.97 a 10.65; p< 0.001). Conclusão: A presença de comorbidades clínicas medida peloíndice de Charlson foi associada a uma menor resposta neurológica no tratamento trombolítico do AVC isquêmico e a um maior grau de incapacidade funcional na alta. / Background and purpose: Clinical comorbidities modify prognosis in haemorrhagic and ischaemic stroke. Charlson Comorbidity index is a validated and useful tool for evaluating comorbidity in stroke. In this study we evaluated the effect of clinical comorbidities as measured by Charlson Comorbidity Index in the in ischaemic stroke thrombolysis. Methods: Prospective cohort study of 96 thrombolysis treated ischaemic stroke patients. The cohort population was divided in two groups according with severity of Charlson Comorbidity Index. During study, NIHSS score was evaluated four times (pre, post, 24 hours and 7 days after thrombolysis) and lower or higher comorbidities groups were compared using repeated measures ANOVA. Response to thrombolysis in both groups was also analysed with disability modified Rankin scale. Results: We observed differences in evolution of mean NIHSS scores between higher and lower clinical comorbidity groups. Patients with low clinical comorbidities experiencing a significant reduction of NIHSS score that ranged from 10.13 to 2.9 points, while patients in the HIC group had initial NIHSS score of 14.75 and final NIHSS score of 13.78 (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Lack of response to thrombolysis had direct relation with disability at hospital discharge. Better clinical outcome, as evaluated by modified Rankin scale of 0 and 1, was markedly different between groups, with 23 (73%) versus 9 (15%) in low and high clinical comorbidities patients respectively (RR=5.62; 95%CI=2.97 to 10.65; p< 0.001). Conclusion: High level of clinical comorbidities negatively influences response to thrombolysis, attenuating treatment related reduction of stroke symptoms severity and increasing the frequency of disabled patients at discharge.

Introdução: Diversas condições clínicas podem modificar a resposta ao tratamento trombolítico no acidente vascular isquêmico agudo. O grau de comorbidade dos pacientes medido pelo Índice de Charlson, um índice que mede o grau de comorbidades clínicas em AVC, tem valor prognóstico na incapacidade pós-AVC tanto em populações com acidente vascular do tipo hemorrágico como isquêmico. Objetivo: Avaliar o efeito do grau de comorbidade aferido pelo índice de Charlson na resposta ao tratamento trombolítico no acidente vascular isquêmico e a incapacidade na alta hospitalar. Métodos: Estudo de coorte prospectivo de 96 pacientes tratados com trombólise para o acidente vascular isquêmico, avaliando o impacto das comorbidades clínicas na resposta ao tratamento trombolítico no AVC isquêmico. Os pacientes foram divididos em dois grupos, aqueles com alto ou baixo grau de comorbidades clínicas, conforme o índice de Charlson. A evolução após o tratamento foi aferida pelo escore de gravidade dos sintomas de acordo com a escala do NIHSS medido antes da infusão, imediatamente após o tratamento, 24horas e 7 dias após a trombólise. A incapacidade na alta foi avaliada pela escala modificada de Rankin sendo, considerada boa resposta a pontuação 0-1 e sua frequência comparada entre os dois grupos de pacientes. Resultados: A comparação dos escores médios do NIHSS mostraram diferenças significativas nos diferentes momentos entre os grupos de alta e baixa comorbidade (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Pacientes com índice de comorbidade baixo apresentaram redução do escore do NIHSS de 10.13 para 2.9, enquanto que no grupo com alta comorbidade, o tratamento trombolítico demostrou pouco efeito. Uma boa evolução, definida como incapacidade 0 e 1 na escala modificada de Rankin, foi observada em (73%) dos pacientes com baixo índice de comorbidade, enquanto somente (15%) dos pacientes com alto índice de comorbidade apresentaram essa evolução favorável, uma diferença clinicamente muito significativa (RR 5.62; 95% CI = 2.97 a 10.65; p< 0.001). Conclusão: A presença de comorbidades clínicas medida peloíndice de Charlson foi associada a uma menor resposta neurológica no tratamento trombolítico do AVC isquêmico e a um maior grau de incapacidade funcional na alta. / Background and purpose: Clinical comorbidities modify prognosis in haemorrhagic and ischaemic stroke. Charlson Comorbidity index is a validated and useful tool for evaluating comorbidity in stroke. In this study we evaluated the effect of clinical comorbidities as measured by Charlson Comorbidity Index in the in ischaemic stroke thrombolysis. Methods: Prospective cohort study of 96 thrombolysis treated ischaemic stroke patients. The cohort population was divided in two groups according with severity of Charlson Comorbidity Index. During study, NIHSS score was evaluated four times (pre, post, 24 hours and 7 days after thrombolysis) and lower or higher comorbidities groups were compared using repeated measures ANOVA. Response to thrombolysis in both groups was also analysed with disability modified Rankin scale. Results: We observed differences in evolution of mean NIHSS scores between higher and lower clinical comorbidity groups. Patients with low clinical comorbidities experiencing a significant reduction of NIHSS score that ranged from 10.13 to 2.9 points, while patients in the HIC group had initial NIHSS score of 14.75 and final NIHSS score of 13.78 (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Lack of response to thrombolysis had direct relation with disability at hospital discharge. Better clinical outcome, as evaluated by modified Rankin scale of 0 and 1, was markedly different between groups, with 23 (73%) versus 9 (15%) in low and high clinical comorbidities patients respectively (RR=5.62; 95%CI=2.97 to 10.65; p< 0.001). Conclusion: High level of clinical comorbidities negatively influences response to thrombolysis, attenuating treatment related reduction of stroke symptoms severity and increasing the frequency of disabled patients at discharge.

Introdução: Diversas condições clínicas podem modificar a resposta ao tratamento trombolítico no acidente vascular isquêmico agudo. O grau de comorbidade dos pacientes medido pelo Índice de Charlson, um índice que mede o grau de comorbidades clínicas em AVC, tem valor prognóstico na incapacidade pós-AVC tanto em populações com acidente vascular do tipo hemorrágico como isquêmico. Objetivo: Avaliar o efeito do grau de comorbidade aferido pelo índice de Charlson na resposta ao tratamento trombolítico no acidente vascular isquêmico e a incapacidade na alta hospitalar. Métodos: Estudo de coorte prospectivo de 96 pacientes tratados com trombólise para o acidente vascular isquêmico, avaliando o impacto das comorbidades clínicas na resposta ao tratamento trombolítico no AVC isquêmico. Os pacientes foram divididos em dois grupos, aqueles com alto ou baixo grau de comorbidades clínicas, conforme o índice de Charlson. A evolução após o tratamento foi aferida pelo escore de gravidade dos sintomas de acordo com a escala do NIHSS medido antes da infusão, imediatamente após o tratamento, 24horas e 7 dias após a trombólise. A incapacidade na alta foi avaliada pela escala modificada de Rankin sendo, considerada boa resposta a pontuação 0-1 e sua frequência comparada entre os dois grupos de pacientes. Resultados: A comparação dos escores médios do NIHSS mostraram diferenças significativas nos diferentes momentos entre os grupos de alta e baixa comorbidade (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Pacientes com índice de comorbidade baixo apresentaram redução do escore do NIHSS de 10.13 para 2.9, enquanto que no grupo com alta comorbidade, o tratamento trombolítico demostrou pouco efeito. Uma boa evolução, definida como incapacidade 0 e 1 na escala modificada de Rankin, foi observada em (73%) dos pacientes com baixo índice de comorbidade, enquanto somente (15%) dos pacientes com alto índice de comorbidade apresentaram essa evolução favorável, uma diferença clinicamente muito significativa (RR 5.62; 95% CI = 2.97 a 10.65; p< 0.001). Conclusão: A presença de comorbidades clínicas medida peloíndice de Charlson foi associada a uma menor resposta neurológica no tratamento trombolítico do AVC isquêmico e a um maior grau de incapacidade funcional na alta. / Background and purpose: Clinical comorbidities modify prognosis in haemorrhagic and ischaemic stroke. Charlson Comorbidity index is a validated and useful tool for evaluating comorbidity in stroke. In this study we evaluated the effect of clinical comorbidities as measured by Charlson Comorbidity Index in the in ischaemic stroke thrombolysis. Methods: Prospective cohort study of 96 thrombolysis treated ischaemic stroke patients. The cohort population was divided in two groups according with severity of Charlson Comorbidity Index. During study, NIHSS score was evaluated four times (pre, post, 24 hours and 7 days after thrombolysis) and lower or higher comorbidities groups were compared using repeated measures ANOVA. Response to thrombolysis in both groups was also analysed with disability modified Rankin scale. Results: We observed differences in evolution of mean NIHSS scores between higher and lower clinical comorbidity groups. Patients with low clinical comorbidities experiencing a significant reduction of NIHSS score that ranged from 10.13 to 2.9 points, while patients in the HIC group had initial NIHSS score of 14.75 and final NIHSS score of 13.78 (Wilk's Lambda test F (1,92) = 24.293; p< 0.001). Lack of response to thrombolysis had direct relation with disability at hospital discharge. Better clinical outcome, as evaluated by modified Rankin scale of 0 and 1, was markedly different between groups, with 23 (73%) versus 9 (15%) in low and high clinical comorbidities patients respectively (RR=5.62; 95%CI=2.97 to 10.65; p< 0.001). Conclusion: High level of clinical comorbidities negatively influences response to thrombolysis, attenuating treatment related reduction of stroke symptoms severity and increasing the frequency of disabled patients at discharge.

Introdução: O Acidente Vascular Cerebral (AVC) é a maior causa de morte no Brasil e um dos maiores responsáveis por incapacitação e invalidez. Existem informações insuficientes quanto aos principais fatores associados à ocorrência de óbito nos pacientes vítimas desta enfermidade. Alguns escores preditores foram desenvolvidos porém não foram validados em população brasileira até o momento. Uma das ações mais importantes na redução do ônus do AVC é o atendimento sistematizado destes pacientes de forma mulltidisciplinar em Unidades de AVC (UAVC) com potencial aumento do uso da terapia trombolítica, além da estratificação dos pacientes, possibilitando decisões terapêuticas mais precoces. Este estudo traz informações sobre o perfil epidemiológico dos pacientes admitidos na UAVC do Hospital Geral de Fortaleza (HGF), bem como identifica fatores preditores de mortalidade e incapacidade até a alta hospitalar e busca validar o Escore de Risco do Registro da Rede Canadense de AVC (IScore), possibilitando a utilização desta ferramenta na estratificação de risco de morte e incapacidade em uma população distinta daquela originalmente realizada. Objetivos: avaliar perfil clínico-epidemiológico dos pacientes e identificar fatores preditores independentes de mortalidade e incapacidade (primários); validar o iScore para morte ou incapacidade e desenvolver um escore na amostra para morte e incapacidade (secundários). Métodos: Foram selecionados pacientes consecutivos admitidos na Unidade de AVC do HGF entre novembro de 2009 até maio de 2012 com diagnóstico clínico de AVC isquêmico. Os dados foram coletados por equipe treinada e através de um formulário específico. Foi realizada análise univariada (método do quiquadrado) e análise multivariada (com regressão logística, stepwise forwardbackward) para descrição das características e identificação dos fatores associados ao desfecho. Teste de correlação de Pearson e curva ROC foram utilizados para medidas de correlação e desempenho dos escores prognósticos. Resultados: no período entre novembro de 2009 e maio de 2012 foram elegíveis 1433 pacientes, sendo 780 analisados. Houve predomíno do sexo masculino e a média de idade (± desvio padrão) foi de 66,1 anos (± 15,44). A forma de apresentação mais comum foi a fraqueza muscular (653 pacientes, 83,6%). O desfecho combinado ocorreu em 423 pacientes (45,8%) e 40 pacientes (5,1%) morreram. Foram identificados 8 fatores preditores independentes para o desfecho. O iScore apresentou bom desempenho, com AUC de 0,797 e Correlação de Pearson de 0,989. Conclusão: Pacientes com AVCi tem altas taxas de incapacidade ou morte até a alta de uma unidade de AVC. Medidas populacionais de informação tem potencial para reduzir a ocorrência dos desfechos. Foram identificados oito fatores preditores de mortalidade ou incapacidade. O iScore apresentou bom desempenho na amostra e pode ser utilizado com acurácia na população brasileira como ferramenta prognóstica. / Intoduction: Stroke is the leading cause of death and one of the most important disease associated with disability in Brazil. There is insufficient information about factors associated with death in stroke patients. Some death risk score has been developed, but none of them were applied in the Brazilian population yet. One of the most important actions to be done to reduce the burden of the stroke is the multidisciplinary assessment of the patients in stroke units (UAVC), with the potential to improve the thrombolytic therapy utilization and the early stratification of patients, allowing earlier treatment decisions. The present study, provides information on the epidemiological profile of patients admitted to the stroke unit in the Hospital Geral de Fortaleza (HGF), identifies predictors of in-hospital mortality and disability and seeks to validate the IScore, allowing the use of this tool to stratify the risk of death and disability in a population different from that which was originally derived. Objectives: to evaluate patient epidemiologic and clinical patterns and factors independently associated with death and disability at hospital discharge (primary objectives); to validate the iScore fitness to predict mortality and/or disability and to develop a new risk score to predict mortality and disability at discharge (secondary objectives). Methods: all consecutive patients admitted to the Hospital Geral de Fortaleza Stroke Unit since November 2009 until May 2012 were elegible. Data were collected by a trained team and by using a specific clinical research form. Univariable analysis (by chi-square test) followed by multivariable analysis (with logistic regression) were performed to identify and establish the variables associated with the outcome (death or disability at hospital discharge). Additionally, Pearson correlation test and ROC curve to measure the iScore correlation and discrimination ability were conducted. Results: a total of 1433 patients were selected and 781 considered eligible were included for the analysis. Male gender were more frequent; mean age was 66,1 (± 15,44). The most common clinical pattern at hospital arrival was \"weakness\" (653 pacientes, 83,6%). Outcome occurred in 423 patients (58,6%) and 40 patients (5,1%) had died. Eight factors were independently associated with outcome. The iScore had good performance, with AUC of 0,797 and Pearson Correlation Test of 0,985. Conclusion: Stroke patients have substantial rate of death or disability at hospital discharge. Populationbased strategies to inform about the signs and symptoms of stroke have potential to decrease this rate. Eight factors were identified as predictors of death or disability and might be used to support patient risk stratification. The iScore had a good performance in the sample and can be used with accuracy as a prognostic tool in Brazil.