Gengivite é a doença periodontal mais prevalente em todas as populações. No entanto, existe uma grande variedade entre as estimativas de prevalência (50-100%) entre diferentes estudos epidemiológicos. O objetivo do presente estudo foi avaliar as mudanças na gengivite após 4 anos e determinar os possíveis fatores de proteção em adultos brasileiros de Porto Alegre. Na avaliação inicial, uma amostra representativa de 1.023 adultos com 35 anos ou mais foi obtida por meio de uma estratégia de amostragem probabilística de múltiplos estágios. Após 4 anos, 402 indivíduos com pelo menos 1 dente foram reexaminados. Foi aplicado um questionário estruturado e foi realizado o índice de sangramento gengival (ISG) em quatro sítios de todos os dentes. Os indivíduos foram dicotomizados como aqueles que apresentaram reduções em ISG acima de 15% e aqueles que apresentaram reduções inferiores a 15% ou aumento do ISG. Os modelos de regressão de Poisson foram ajustados para estimar os riscos relativos (RR) e intervalos de confiança de 95% (IC 95%). A extensão geral dos sítios com ISG reduziu significativamente de 25.9% para 20.1%. Esta redução foi maior e significativa nas superfícies livres (34.1% para 24.0%), enquanto nas superfícies proximais a redução não foi significativa (17.9% para 16.1%). O percentual de indivíduos com redução de ISG ≥ 15% em todos os sítios e apenas em superfícies livres foi de 31.0% e 38.8%, respectivamente. Para todas as superfícies, a frequência de escovação auto-referida ≥ 2 vezes/dia aumentou a probabilidade de redução de ISG em 72% em comparação com ≤ 1 vez/dia (RR = 1.72; IC 95%: 1.01-3.16). Para superfícies livre, a frequência de escovação e o tabagismo permaneceram associados, e os indivíduos com peso normal apresentaram uma probabilidade 26% maior de redução de ISG do que indivíduos obesos com excesso de peso (RR = 1.26; IC 95%: 1.01-1.62). Em conclusão, foi observada uma redução na extensão da gengivite ao longo do tempo nesta população brasileira. A frequência de escovação e o peso normal foram fatores protetores para gengivite. Fumar reduziu gengivite ao longo do tempo, mas provavelmente devido ao seu efeito de vasoconstrição na margem gengival. / Gingivitis is the most prevalent periodontal disease in all populations. However, there is a great variety between estimates of prevalence of gingivitis (50-100%) among different epidemiological studies. The aim of the present study was to assess changes in gingivitis over 4 years and to determine possible protective factors in Brazilian adults. At baseline, a representative sample of 1,023 adults 35 years and older was obtained using a multistage probability sampling strategy. After 4 years, 402 individuals with at least 1 tooth were re-examined. A structured questionnaire was applied and gingival bleeding index (GB) was performed at four sites of all teeth. Individuals were dichotomized into those showing reductions in GB higher than 15% and those showing reductions under 15% or increase in GB. Poisson regression models were fitted to estimate relative risks (RR) and 95% confidence intervals (95%CI). The overall extent of sites with GB reduced significantly from 25.9% to 20.1%. This reduction was higher and significative in free surfaces (34.1% to 24.0%), whereas in proximal surfaces the reduction was not significant (17.9% to 16.1%). The percentage of individuals with GB reduction ≥15% in all sites and only free sites was 31.0% and 38.8%, respectively. For all surfaces, self-reported brushing frequency ≥2 times/day increased the probability of reducing GB by 72% compared to ≤1/day (RR=1.72; 95%CI 1.01-3.16). For each 10 packyears smoked, the probability of reduction in GB was 3% higher (RR=1.03; 95%CI 1.01-1.04). For free surfaces, brushing frequency and smoking remained associated, and normal weight individuals had 26% higher probability of reducing GB than overweight-obese individuals (RR=1.26; 95%CI 1.01-1.62). In conclusion, it was observed a reduction in the extension of gingivitis over time in this Brazilian population. Brushing frequency and normal weight were found to be protective factors for gingivitis. Smoking reduced gingivitis over time, but probably due to its vasoconstriction effect on the gingival margin.

Os objetivos deste estudo foram estimar a prevalência da deficiência intelectual aos 7-8 anos de idade em uma coorte de nascimentos, através de investigação genética clínica e laboratorial e, também, investigar a etiologia da mesma e os fatores associados. Os participantes faziam parte de uma coorte acompanhada desde o nascimento e foram incluídos neste estudo por apresentar, em acompanhamentos anteriores, suspeita de atraso no desenvolvimento segundo o Teste de Rastreamento de Battelle, QI abaixo de 70 segundo a escala WPPSI e/ou problemas no comportamento observados durante entrevista. Das 4231 crianças da Coorte de 2004 de Pelotas, 214 foram selecionadas para a avaliação genética que constou de: anamnese, exame físico e dismorfológico e coleta de sangue e urina quando indicado. Criou-se um banco de dados incluindo variáves desta avaliação e dos acompanhamentos anteriores da Coorte, tais como: variáveis da gestação e do nascimento, sociodemográficas e relativas à saúde e estimulação da criança. Os dados foram processados no pacote estatístico Stata 13.0 e foi utilizada análise de variância (ANOVA). Foi considerada como tendo deficiência intelectual a criança que, além de apresentar um QI abaixo de 70, apresentava também problemas no comportamento adaptativo. Cento e setenta crianças das duzentas e quatorze selecionadas no início do estudo foram diagnosticadas com deficiência intelectual e classificadas em cinco grupos etiológicos. A maior parte das crianças (44,4%) foi classificada como tendo deficiência intelectual devida a causas não-biológicas, ou seja, ligada a fatores ambientais. O segundo maior grupo (16,6%) foi o grupo de crianças com deficiência intelectual genética que incluiu crianças com síndrome de Down, microdeleções e patologias autossômicas dominantes e patologias multifatoriais. A seguir, crianças com sequelas neonatais (13,3%) e deficiência intelectual associada a outras doenças (13,3%), como epilepsia e TDAH. O menor grupo foi o idiopático, constituído por crianças que, mesmo após investigação clínica e laboratorial, permaneceram sem diagnóstico definido. A prevalência de deficiência intelectual foi de 4,5% e a prevalência de deficiência intelectual genética de 0,66%. Apesar de algumas limitações como a identificação e seleção dos casos aos 4 anos para uma avaliação aos 7-8 anos, é importante considerar que, por ser um estudo de base populacional, com alta taxa de acompanhamento (92,0%), isto minimiza o viés de seleção. O fato dos dados serem colhidos no momento ou em um curto intervalo de tempo, considerando os diversos acompanhamentos, minimiza o viés de memória. Fora do mundo desenvolvido, são raros os estudos de coorte que avaliaram deficiência intelectual, seus fatores de risco e sua etiologia. Grande parte destes estudos, mesmo os conduzidos em países de renda alta, avaliaram a prevalência, mas não a etiologia. Os dados sugerem que boa parte destes casos poderia ser prevenida, principalmente considerando uma etiologia não-biológica, caso existissem, além do rastreamento de problemas no desenvolvimento, estratégias de intervenção educacional e de saúde. / The aims of this study were to estimate the prevalence and etiology of intellectual disability at 7-8 years of age in a birth cohort through clinical and laboratory investigation and associated factors. Participants were part of a cohort followed from birth and were included in this study due to suspected developmental delay according to the Battelle Screening Test, IQ below 70 according to WPPSI scale and / or behavior problems observed during the interview in previous follow-ups. Of the 4231 children in the 2004 Pelotas birth cohort, 214 were selected for genetic evaluation which included anamnesis, physical and dysmorphological examination and collection of blood and urine when indicated. A dataset including variables from this evaluation and the previous cohort of follow-ups such as variables of pregnancy and birth, social demographic and health-related and stimulation of the child. Data were analyzed using Stata version 13.0. Analysis of variance (ANOVA) was performed. To be considered as having intellectual disability the child that presenting an IQ below 70 and problems in adaptive behavior. One hundred and seventy children from two hundred fourteen selected at baseline were diagnosed with intellectual disability and they were classified into five etiologic groups. Most children (44.4 %) were classified as having intellectual disability due to no biological causes, i.e., linked to environmental factors. The second largest group (16.6%) was the group of children with genetic intellectual disability which included children with Down syndrome, microdeletions and autosomal dominant and multifactorial diseases. Children with neonatal sequelae accounted for 13.3% and intellectual disability associated with other diseases such as epilepsy and ADHD also accounted for 13.3%. The smallest group was idiopathic composed of children who even after clinical and laboratory investigation remained without a definite diagnosis. The prevalence of intellectual disability was 4.5 % and the prevalence of genetic intellectual disability 0.66 %. Despite some limitations such as the identification and selection of cases to four years for an assessment at 7-8 years it is important to consider that it is a population-based study with high follow-up rate (92.0 %) which minimizes selection and information bias. As data were collected in time or in a short period of time considering the several follow-ups minimize recall bias. Outside the developed world few cohort studies assessed intellectual disabilities, their risk factors and etiology. Most of these studies even those conducted in high-income countries assessed the prevalence but not the etiology. The data suggest that part of these cases could be prevented specially considering the non-biological etiology if there were screening of developmental delay and intervention strategies on health and educational bases.

O presente estudo teve por objetivo analisar fatores prognósticos de sobrevida inferior a três meses em pacientes com câncer de pulmão metastático. Foram incluídos pacientes com idade superior a 18 anos egressos de um hospital da região sul do Brasil, portadores de câncer de pulmão (estádio IV) com doença metastática, com qualquer tratamento anti-neoplásico prévio, efetuado até um mês após o diagnóstico. Análise consistiu na verificação da distribuição, análise das variáveis independentes controladas pelo percentual de perda de peso em 6 meses através da Regressão de Cox. Entre os 137 indivíduos, 31 (22,6%; IC95%: 15,9 – 30,5) foram a óbito dentro de 90 dias. A prevalência de percentual de perda de peso em 6 meses foi de 35,6% (IC95%: 27,4 – 44,4). Na análise multivariada foram testados modelos relacionados ao risco de morte em relação à perda de peso acima de 10%, e em todos os modelos testados foi evidenciado o impacto independente de morte para fatores como índice de massa corporal, circunferência muscular do braço, índice de Karnofsky, escala de Lasa (dor e dispneia), tabagismo, internação hospitalar e tipo de tratamento oncológico, sendo a variável prognóstica mais importante relacionada à sobrevida de pacientes com câncer de pulmão metastático. / The present study aimed to analyze prognostic factors of survival lesser than three months in patients with metastatic lung cancer. We included patients older than 18 years, discharged from a hospital in southern Brazil, with metastatic lung cancer (stage IV), receiving at least one month after the diagnosis. Analysis consisted in checking the distribution, analysis of the independent variables controlled by the percentage of weight loss in 6 months by Cox Regression. Among the 137 individuals, 31 (22.6% 95% CI 15.9 to 30.5) died within 90 days. The prevalence of percentage weight loss at 6 months was 35.6% (95% CI 27.4 to 44.4). In multivariate models were tested related to the risk of death compared to weight loss of over ten percent, and all models tested was evidenced the independent impact of death for factors such as body mass index, arm muscle circumference, index Karnofsky scale, Lasa (pain and dyspnea), smoking, hospitalization and type of cancer treatment, being the most important prognostic variable related to survival of patients with metastatic lung cancer.

Introdução: A influência da depressão pós-acidente vascular cerebral (AVC) como preditor de mortalidade é pouca investigada. Assim, nosso objetivo foi avaliar a influência da depressão na sobrevida em longo prazo nos participantes do Estudo de Mortalidade e Morbidade do Acidente Vascular Cerebral (Estudo EMMA), uma coorte em andamento no Hospital Universitário da Universidade de São Paulo, Brasil. Métodos: Foi avaliada prospectivamente uma subamostra de 191 indivíduos com 35 anos ou mais, acometidos por acidente vascular cerebral do tipo isquêmico e hemorrágico, participantes do Estudo EMMA. As características do AVC na admissão hospitalar, além dos dados sociodemográficos e fatores de risco cardiovasculares foram avaliados de acordo com o primeiro episódio de depressão após o evento agudo. O rastreamento de depressão foi realizado com cada participante através de entrevista telefônica com o questionário \"Patient Health Questionnaire-9\" [transtorno depressivo maior (escore ≥ 10 pontos) e depressão menor (escore ≥ cinco pontos)] 30 dias após o evento agudo e periodicamente durante um ano de seguimento. As análises de mortalidade foram realizadas através das curvas de sobrevivência de Kaplan-Meier e pelas razões de risco (\"hazard ratio\") proporcionais que foram calculadas por modelos de regressão logística de Cox ajustados para potenciais fatores de confusão. Resultados: Nesta subamostra de 191 participantes do estudo EMMA, 164 (85,9%) indivíduos foram diagnosticados com acidente vascular cerebral isquêmico e 27 (14,1%) com acidente vascular cerebral hemorrágico. A frequência geral de transtorno depressivo maior foi de 25,1% (48/191) e a frequência de depressão menor foi de 57,1% (109/191) durante um ano de seguimento, independentemente do subtipo de AVC. Foi observada uma taxa de sobrevida menor entre os indivíduos que desenvolveram transtorno depressivo maior pós-AVC em comparação com aqueles que não desenvolveram esta condição após um ano de seguimento (85,4% vs. 96,5%, log-rank p = 0,006). Após análise de regressão de Cox múltipla, observamos um risco maior de mortalidade por todas as causas dentre aqueles que desenvolveram transtorno depressivo maior em comparação aos participantes sem transtorno depressivo maior (razão de risco = 5,97; IC 95% =1,43-24,91, p = 0,01). Os participantes que desenvolveram depressão menor não apresentaram diferenças estatisticamente significativas nas taxas de sobrevida pós-AVC comparativamente aos que não apresentaram estes sintomas (93,2% vs. 94%, log-rank p = 0,94); assim como não apresentaram um aumento no risco de morte em até um ano de seguimento (razão de risco = 0,76; IC 95% = 0,18 3,19, p = 0,71). Conclusão: Nossos resultados sugerem que a ocorrência do primeiro episódio de transtorno depressivo maior foi um preditor potencial de mau prognóstico caracterizado por um risco aumentado de morte um ano após o AVC / Introduction: The influence of post-stroke depression as a predictor of mortality is poorly investigated. Thus, our aim was to evaluate the influence of depression on long-term survival in participants from the Study of Stroke Mortality and Morbidity (EMMA Study), an ongoing stroke cohort performed in the Hospital Universitário da Universidade de São Paulo, Brazil. Methods: We prospectively evaluated a subset of 191 EMMA participants with 35 years or more, affected by ischemic and hemorrhagic stroke. The stroke characteristics at hospital admission, as well as, sociodemographic and cardiovascular risk factors were evaluated according to the first episode of depression after acute event. Screening of depression was conducted by telephone interview with each participant using \"Patient Health Questionnaire-9\" [major depressive disorder (score ≥ 10 points) and minor depression (score ≥ five points)] 30 days after acute event and periodically during one year of follow-up. Mortality analyzes were performed using Kaplan-Meier survival curves and the hazard proportional ratios were calculated by Cox logistic regression models adjusted for potential confounders. Results: In this subsample of 191 EMMA participants, 164 (85.9%) patients were diagnosed with ischemic stroke and 27 (14.1%) with hemorrhagic stroke. The overall frequency of major depressive disorder was 25.1% (48/191) and lower depression rate was 57.1% (109/191) during one year of follow-up, regardless of stroke subtype. It was observed a lower survival rate among individuals who developed major depressive disorder after stroke compared to those who did not develop this condition after one year of follow-up (85.4% vs. 96.5% was observed, log-rank p = 0.006). After multiple Cox regression analysis, we observed an increased all-cause mortality among those who developed major depressive disorder when compared to participants without major depressive disorder (risk ratio = 5.97, 95% CI = 1.43 to 24.91, p = 0.01). Participants who developed minor depression showed no statistically significant differences in post-stroke survival rates compared to those without these symptoms (93.2% vs. 94%, log-rank p = 0.94); as well as, they did not show an increased risk of death within one year of follow-up (risk ratio = 0.76, 95% CI = 0.18 to 3.19, p = 0.71). Conclusion: Our results suggest that the occurrence of the first episode of major depressive disorder was a potential predictor of poor prognosis characterized by an increased risk of death one year after the stroke

O presente estudo teve por objetivo analisar fatores prognósticos de sobrevida inferior a três meses em pacientes com câncer de pulmão metastático. Foram incluídos pacientes com idade superior a 18 anos egressos de um hospital da região sul do Brasil, portadores de câncer de pulmão (estádio IV) com doença metastática, com qualquer tratamento anti-neoplásico prévio, efetuado até um mês após o diagnóstico. Análise consistiu na verificação da distribuição, análise das variáveis independentes controladas pelo percentual de perda de peso em 6 meses através da Regressão de Cox. Entre os 137 indivíduos, 31 (22,6%; IC95%: 15,9 – 30,5) foram a óbito dentro de 90 dias. A prevalência de percentual de perda de peso em 6 meses foi de 35,6% (IC95%: 27,4 – 44,4). Na análise multivariada foram testados modelos relacionados ao risco de morte em relação à perda de peso acima de 10%, e em todos os modelos testados foi evidenciado o impacto independente de morte para fatores como índice de massa corporal, circunferência muscular do braço, índice de Karnofsky, escala de Lasa (dor e dispneia), tabagismo, internação hospitalar e tipo de tratamento oncológico, sendo a variável prognóstica mais importante relacionada à sobrevida de pacientes com câncer de pulmão metastático. / The present study aimed to analyze prognostic factors of survival lesser than three months in patients with metastatic lung cancer. We included patients older than 18 years, discharged from a hospital in southern Brazil, with metastatic lung cancer (stage IV), receiving at least one month after the diagnosis. Analysis consisted in checking the distribution, analysis of the independent variables controlled by the percentage of weight loss in 6 months by Cox Regression. Among the 137 individuals, 31 (22.6% 95% CI 15.9 to 30.5) died within 90 days. The prevalence of percentage weight loss at 6 months was 35.6% (95% CI 27.4 to 44.4). In multivariate models were tested related to the risk of death compared to weight loss of over ten percent, and all models tested was evidenced the independent impact of death for factors such as body mass index, arm muscle circumference, index Karnofsky scale, Lasa (pain and dyspnea), smoking, hospitalization and type of cancer treatment, being the most important prognostic variable related to survival of patients with metastatic lung cancer.

Abstract Oligomenorrhea (menstrual disturbances) and hirsutism (excessive growth of body hair) are typical symptoms of polycystic ovary syndrome, a common endocrine disorder with long-term health risks among fertile-age women. Associations between body size development and polycystic ovary syndrome symptoms in a cohort design (528 symptomatic and 1479 asymptomatic women) and endocrine, metabolic and clinical characteristics of women with self-reported symptoms of oligomenorrhea or hirsutism in a nested case-control design (518 cases and 1036 controls) were investigated in this general population-based study. Gynecologic ultrasonographic examinations were performed in 196 cases and 67 controls to assess the morphology of the ovaries and its relationship to biochemical and clinical parameters. The study population was derived from the Northern Finland Birth Cohort 1966, which included all births with expected birth dates in 1966 in Northern Finland and is well representative of the general female population. Polycystic ovary syndrome symptoms in adulthood were associated with obesity, particularly abdominal obesity, in adolescence and in adulthood, but not with birth weight or being small for gestational age. Hormonal changes typical of polycystic ovary syndrome, i.e. higher circulating concentrations of testosterone, luteinizing hormone (LH) and insulin and lower levels of sex hormone-binding globulin (SHBG), were detected in women with self-reported symptoms of oligomenorrhea and/or hirsutism compared with the controls. Less favorable metabolic cardiovascular disease risk factor profiles, higher body mass index (BMI), waist-hip ratio (WHR), and triglyceride and C-reactive protein (CRP) concentrations and lower high density lipoprotein cholesterol (HDL-C) levels, were detected in women with symptoms, being the most severe among women who reported both hirsutism and oligomenorrhea. Unfavorable characteristics were pronounced in the presence of overweight or obesity. Women with symptoms more often had features characteristic of polycystic ovarian morphology associated with an endocrine and clinical profile reflecting polycystic ovary syndrome. This study shows that questioning in regard to symptoms of oligomenorrhea and hirsutism is useful in detecting women at risk of polycystic ovary syndrome and associated health risks. Avoidance of being overweight is important among young women to prevent the development of insulin resistance. Systematic follow-up of women with symptoms of oligomenorrhea and hirsutism is justified for prevention and early detection of long-term health risks.

Abstract Schizophrenia is associated with a shortened life expectancy and increased somatic comorbidity with e.g. cardiovascular disorders. The purpose of this study was to evaluate hyperlipidemia and metabolic syndrome in schizophrenia and thus find specific risk factors for excess mortality and morbidity. The study population was a subsample of the Northern Finland 1966 Birth Cohort, a general population-based birth cohort. In 1997, 8,463 members of the cohort were invited to a clinical examination, where e.g. blood samples were taken after an overnight fast. Total cholesterol (TC), high-density lipoprotein (HDL), low-density lipoprotein (LDL) and triglycerides (TG) were determined. The following psychiatric diagnostic categories were used: 1) DSM-III-R schizophrenia (n = 31), 2) other psychoses (n = 21), 3) non-psychotic disorders (n = 104), 4) comparison group (n = 5,498), having no psychiatric hospital treatment. Mean TC (5.5 mmol/l) and TG (1.5 mmol/l) were significantly higher in the schizophrenia group than in the comparison group (5.1 mmol/l and 1.2 mmol/l, respectively). To evaluate serum lipid levels in subjects with and without antipsychotic medication the sample was analyzed according to used medication. The prevalence of hypercholesterolemia, high LDL cholesterol and hypertriglyceridemia was high in persons using antipsychotic medication (31%, 20% and 22%, respectively) compared to persons without such medication (12%, 10% and 7%, respectively). We found higher triglyceride levels in patients who were ≤ 20 years old at the onset of schizophrenia (mean 1.7 mmol/l; N = 17) as compared with patients with later onset (mean 1.4 mmol/l; N = 14) or non-hospitalized controls (mean 1.2 mmol/l; N = 5,453). The difference between the first and third group was significant (p < 0.01), and there was a negative correlation between the age at onset and the level of serum triglycerides (r = -0.35, p = 0.05). To evaluate the prevalence of metabolic syndrome, the subjects were assessed for the presence of metabolic syndrome according to the criteria of the National Cholesterol Education Program. The prevalence of metabolic syndrome was high in subjects with schizophrenia compared with the comparison group (19% vs. 6%, p = 0.010). The results indicate an elevated risk for hyperlipidemia and metabolic syndrome in persons with schizophrenia or on antipsychotic medication. Regular monitoring of weight, serum lipid and glucose levels and blood pressure is important. Comprehensive efforts directed at controlling weight and improving physical activity are needed.

L’augmentation de la prévalence des hémorragies cérébrales (HC) sous traitement antithrombotique constitue un problème de Santé Publique à cause du mauvais pronostic vital et fonctionnel et du risque compétitif de récidive hémorragique ou ischémique chez les survivants, qui pose le problème d’une prévention secondaire adaptée._x000D_L’objectif de la thèse était d’étudier l’impact des traitements antithrombotiques sur le pronostic des HC. Le terme d’HC, entendu au sens large, regroupe les HC spontanées et les microhémorragies cérébrales.La première partie avait comme objectif de comparer les caractéristiques des HC survenant avec ou sans traitement par antagonistes de la vitamine K (AVK), stratifiant l’analyse selon la localisation de l’HC. L’analyse a porté sur 545 patients de la cohorte Lilloise (PITCH), parmi lesquels 83 (15%) étaient sous AVK. La prise d’un traitement par AVK n’influençait pas la localisation de l’HC, mais était prédictive d’HC plus volumineuses (25 ml vs 12 ; p=0.002) chez les victimes d’HC non lobaire, alors que aucune différence existait en cas d’HC lobaire (26 ml vs 30; p=0.507). Ces résultats suggèrent que l’impact des AVK diffère en fonction de la localisation de l’HC et de la vasculopathie sous-jacente.La deuxième partie avait comme objectif d’analyser l’impact du traitement antithrombotique sur l’apparition de nouvelles microhémorragies cérébrales stratifiant l’analyse selon la localisation de l’HC. L’étude a porté sur 168 survivants à 6 mois d’une HC (cohorte PITCH) ayant bénéficié d’une IRM cérébrale 1.5T lors de l’HC et durant le suivi. Lors de l’HC, 89 (53%) patients présentaient des microhémorragies, et 80 (48%) ont développé des microhémorragies durant le suivi. La présence sur la première IRM de microhémorragies (aOR 2,3; IC 95%1,2-4,3), leur position mixte, lobaire et profonde (aOR 3.7; IC95% 1,7-8,3), et la présence de séquelles de macrohémorragies (aOR 6,8; IC95% 2,8-16,7), étaient associées à l’apparition de microhémorragies. Chez les patients avec HC non lobaire, l’apparition de microhémorragies était associée à l’utilisation d’un traitement antithrombotique durant le suivi (aOR 2,9 ; IC95% 1,1-7,3) et avec l’apparition de lacunes (aOR: 2,9; 95%CI 1,0-7,8). Chez les patients avec HC lobaire, l’apparition de microhémorragies était associée à l’apparition de macrohémorragie (aOR 9.8 ; IC95% 1,1-88,8). Ces résultats suggèrent que l’impact des traitements antithrombotiques diffère en fonction de la vasculopathie sous-jacente.La troisième partie avait comme objectif de : (i) comparer les proportions de survivants d’une HC ayant une indication formelle au traitement antithrombotique chez lesquels ce traitement était repris à la sorite ; (ii) identifier les caractéristiques associées à la reprise du traitement antithrombotique, au sein de 5 cohortes européennes (Lille, n=542; Utrecht, n=389 ; Amsterdam, n= 403 ; Londres, n=667 ; Lothian, n=137). Un traitement antithrombotique était recommencé chez 96 (20%) des 469 survivants ayant une indication formelle, mais en proportions différentes dans les 5 cohortes (Lille 18%, Utrecht 45%, Amsterdam 30% ; Londres 11% ; Lothian 15%; p<0.001). Nous n’avons retrouvé aucun autre facteur prédictif de reprise du traitement antithrombotique en dehors de la cohorte d’origine. Ces résultats montrent que la décision de reprise du traitement antithrombotique ne repose pas sur des critères reproductibles, et soulignent la nécessité de réaliser des essais randomisés.La quatrième partie, actuellement en cours, prévoit d’évaluer chez ces mêmes patients (survivants d’une HC ayant une indication formelle au traitement antithrombotique) le risque de récidive hémorragique ou ischémique en fonction de la reprise ou de l’arrêt du traitement antithrombotique. Il s’agit d’une étude observationnelle portant sur 274 patients de 4 cohortes Européennes, avec un suivi d’environ 2.5 ans. L’analyse statistique est en cours. / The proportion of patients who are taking antithrombotic drugs at the time they suffer a spontaneous intracerebral hemorrhage (ICH) is increasing over time, and this is a major issue in public health because of the poor prognosis of patients. Also, the decision to restart or not antithrombotic drugs in survivors is still a clinical dilemma. _x000D_The aim of this work was to evaluate the impact of antithrombotic drugs on ICH prognosis. The term ICH regroups (i) spontaneous ICH and (ii) cerebral microbleeds (CMBs).As a first step, we compared baseline characteristics of 545 consecutive patients with ICH included in the PITCH cohort (Lille, France), receiving or not Vitamin K Antagonists (VKAs) at the time of ICH, stratifying the analysis according to the ICH location (lobar vs non lobar). VKAs-ICH accounted for 83 patients (15%). The use of VKAs did not influence ICH location, but influenced ICH volume: in non lobar ICH, VKAs use was associated with larger ICH volumes (25 ml versus 12 ml, p=0.002). In lobar ICH, no difference was observed (26ml versus 30ml; p=0.507). The different impact of VKAs on ICH volumes according to location suggests a different susceptibility of the underlying vasculopathies to VKAs.As a second step, we aimed to evaluate the impact of antithrombotic drugs on the incidence of CMBs in 168 ICH survivors from the PITCH cohort who underwent 1.5T Magnetic resonance imaging (MRI) at ICH onset and during follow-up, stratifying the analysis according to the index ICH location. At the time of ICH, 89 (53%) patients had CMBs at ICH onset, and 80 (48%) showed new CMBs during follow-up. Predictors of incident CMBs were the presence on the first MRI of: at least 1 CMB (aOR 2.3; 95% CI: 1.2-4.3), old macro-hemorrhage (aOR 6.8; 95%CI 2.8-16.7), and CMBs in mixed location (aOR 3.7; 95%CI 1.7-8.3). In non lobar ICH, incident CMBs were associated with incident lacunes (aOR: 2.9; 95%CI 1.0-7.8) and with the use of antiplatelet agents (aOR 2.9; 95%CI 1.1-7.3). In lobar ICH, incident CMBs were associated with incident macro-hemorrhage (aOR 9.8; 95%CI 1.1-88.8). These results suggest that the impact of antithrombotic drugs differs according to the index ICH location, and therefore according to the underlying vasculpathy.The third step consisted in comparing the characteristics and proportions of patients taking antithrombotic drugs at ICH discharge in five European cohorts (Lille, France, n=542; Utrecht, The Netherlands, n=389; Amsterdam, The Netherlands, n=403; multicenter UK cohort, n=667; Lothian, Scotland, n=137). We then identified characteristics associated with restarting. Antithrombotic drugs were restarted in 96 (20%) of the 469 survivors who had an indication for antithrombotic drugs (secondary prevention or atrial fibrillation), but in different proportions according to the cohort of origin (Lille 18%, Utrecht 45%, Amsterdam 30%, CROMIS-2 ICH 11%, Lothian 15%; p<0.001). We did not find other consistent associations with restarting antithrombotic drugs. The variation in clinical practice in restarting antithrombotic drugs after ICH supports the need for randomized controlled trials.In the fourth step, we aim to analyse the risk of ICH recurrences or ischemic events in the same population of patients (ICH survivors who suffered from ICH while on antithrombotic drugs because of secondary prevention or atrial fibrillation) according to the antithrombotic drug status during follow-up. We included 274 patients from 4 European cohorts, with a median follow-up of 2.5 years. Statistical analysis is ongoing.

Os objetivos deste estudo foram estimar a prevalência da deficiência intelectual aos 7-8 anos de idade em uma coorte de nascimentos, através de investigação genética clínica e laboratorial e, também, investigar a etiologia da mesma e os fatores associados. Os participantes faziam parte de uma coorte acompanhada desde o nascimento e foram incluídos neste estudo por apresentar, em acompanhamentos anteriores, suspeita de atraso no desenvolvimento segundo o Teste de Rastreamento de Battelle, QI abaixo de 70 segundo a escala WPPSI e/ou problemas no comportamento observados durante entrevista. Das 4231 crianças da Coorte de 2004 de Pelotas, 214 foram selecionadas para a avaliação genética que constou de: anamnese, exame físico e dismorfológico e coleta de sangue e urina quando indicado. Criou-se um banco de dados incluindo variáves desta avaliação e dos acompanhamentos anteriores da Coorte, tais como: variáveis da gestação e do nascimento, sociodemográficas e relativas à saúde e estimulação da criança. Os dados foram processados no pacote estatístico Stata 13.0 e foi utilizada análise de variância (ANOVA). Foi considerada como tendo deficiência intelectual a criança que, além de apresentar um QI abaixo de 70, apresentava também problemas no comportamento adaptativo. Cento e setenta crianças das duzentas e quatorze selecionadas no início do estudo foram diagnosticadas com deficiência intelectual e classificadas em cinco grupos etiológicos. A maior parte das crianças (44,4%) foi classificada como tendo deficiência intelectual devida a causas não-biológicas, ou seja, ligada a fatores ambientais. O segundo maior grupo (16,6%) foi o grupo de crianças com deficiência intelectual genética que incluiu crianças com síndrome de Down, microdeleções e patologias autossômicas dominantes e patologias multifatoriais. A seguir, crianças com sequelas neonatais (13,3%) e deficiência intelectual associada a outras doenças (13,3%), como epilepsia e TDAH. O menor grupo foi o idiopático, constituído por crianças que, mesmo após investigação clínica e laboratorial, permaneceram sem diagnóstico definido. A prevalência de deficiência intelectual foi de 4,5% e a prevalência de deficiência intelectual genética de 0,66%. Apesar de algumas limitações como a identificação e seleção dos casos aos 4 anos para uma avaliação aos 7-8 anos, é importante considerar que, por ser um estudo de base populacional, com alta taxa de acompanhamento (92,0%), isto minimiza o viés de seleção. O fato dos dados serem colhidos no momento ou em um curto intervalo de tempo, considerando os diversos acompanhamentos, minimiza o viés de memória. Fora do mundo desenvolvido, são raros os estudos de coorte que avaliaram deficiência intelectual, seus fatores de risco e sua etiologia. Grande parte destes estudos, mesmo os conduzidos em países de renda alta, avaliaram a prevalência, mas não a etiologia. Os dados sugerem que boa parte destes casos poderia ser prevenida, principalmente considerando uma etiologia não-biológica, caso existissem, além do rastreamento de problemas no desenvolvimento, estratégias de intervenção educacional e de saúde. / The aims of this study were to estimate the prevalence and etiology of intellectual disability at 7-8 years of age in a birth cohort through clinical and laboratory investigation and associated factors. Participants were part of a cohort followed from birth and were included in this study due to suspected developmental delay according to the Battelle Screening Test, IQ below 70 according to WPPSI scale and / or behavior problems observed during the interview in previous follow-ups. Of the 4231 children in the 2004 Pelotas birth cohort, 214 were selected for genetic evaluation which included anamnesis, physical and dysmorphological examination and collection of blood and urine when indicated. A dataset including variables from this evaluation and the previous cohort of follow-ups such as variables of pregnancy and birth, social demographic and health-related and stimulation of the child. Data were analyzed using Stata version 13.0. Analysis of variance (ANOVA) was performed. To be considered as having intellectual disability the child that presenting an IQ below 70 and problems in adaptive behavior. One hundred and seventy children from two hundred fourteen selected at baseline were diagnosed with intellectual disability and they were classified into five etiologic groups. Most children (44.4 %) were classified as having intellectual disability due to no biological causes, i.e., linked to environmental factors. The second largest group (16.6%) was the group of children with genetic intellectual disability which included children with Down syndrome, microdeletions and autosomal dominant and multifactorial diseases. Children with neonatal sequelae accounted for 13.3% and intellectual disability associated with other diseases such as epilepsy and ADHD also accounted for 13.3%. The smallest group was idiopathic composed of children who even after clinical and laboratory investigation remained without a definite diagnosis. The prevalence of intellectual disability was 4.5 % and the prevalence of genetic intellectual disability 0.66 %. Despite some limitations such as the identification and selection of cases to four years for an assessment at 7-8 years it is important to consider that it is a population-based study with high follow-up rate (92.0 %) which minimizes selection and information bias. As data were collected in time or in a short period of time considering the several follow-ups minimize recall bias. Outside the developed world few cohort studies assessed intellectual disabilities, their risk factors and etiology. Most of these studies even those conducted in high-income countries assessed the prevalence but not the etiology. The data suggest that part of these cases could be prevented specially considering the non-biological etiology if there were screening of developmental delay and intervention strategies on health and educational bases.

Contexte. La population agricole française, dans son ensemble, présente un plus faible risque de Cancer BronchoPulmonaire (CBP) par rapport à la population générale. Ce constat, retrouvé au niveau international, s’explique au moins en partie par un tabagisme plus faible notamment chez les plus âgés et chez les femmes ; un effet « protecteur » des endotoxines est également suspecté. Néanmoins le milieu professionnel agricole peut exposer à de nombreux cancérogènes pulmonaires avérés ou suspectés ; malgré cela, à l’exception de certains secteurs d’élevage, il n’existe que très peu de données sur les associations entre expositions ou situations de travail précises en milieu agricole et le risque de CBP. Objectifs. Le premier objectif de ce travail de thèse était d’étudier les associations, au sein de la cohorte AGRICAN (AGRIculture et CANcer), entre (1) CBP et exposition aux activités et tâches de cultures ; (2) CBP et utilisation de pesticides arsenicaux et herbicides dinitroanilines, et de mettre à profit le nombre important de sujets inclus et de cas de CBP pour réaliser ces analyses par sous-type histologique. Le deuxième objectif était d’initier un projet de métrologie de divers aérocontaminants en milieu professionnel agricole. Méthodes. Les analyses ont été menées en utilisant les données de l’étudeAGRICAN et de la matrice culture-exposition aux pesticides PESTIMAT. Nous avons utilisé un modèle de Cox, ajusté sur le sexe, le tabagisme, les élevages bovin et équin. Le projet métrologique AirExpA s’est limité à la Normandie, avec mesures personnelles des poussières totales (en temps réel), endotoxines et 38 mycotoxines (en cumulé), en élevage bovin, lors des foins et des moissons de blé/orge et pois fourrager/féverole. Résultats et discussion. Nous avons retrouvé une association inverse entre l’activité de culture de maïs et le risque de CBP, une association positive entre adénocarcinomes et viticulture, ainsi qu’entre carcinomes à petites cellules et culture de pois fourragers/féveroles. Les éléments de plausibilité d’un lien entre ces secteurs d’activité et le risque de CBP sont encore des hypothèses : la culture de maïs, par le stockage du grain, pourrait exposer aux endotoxines ; à l’inverse, la récolte de pois fourragers a été décrite comme empoussiérante, mais cet élément doit être confirmé par une caractérisation des poussières. Nous n’avons pas retrouvé d’association entre utilisation de pesticides arsenicaux, ni de dinitroanilines, et risque de CBP ; cependant les analyses stratifiées sur le sexe et le statut tabagique, montraient des associations positives chez les non-fumeurs et chez les femmes (en particulier pour les adénocarcinomes). Enfin, le projet AirExpA a permis de documenter un gradient d’exposition aux poussières et endotoxines en fonction de différentes tâches en élevage bovin, et une exposition plus modérée mais réelle, aux endotoxines, lors des activités de culture. Les analyses de mycotoxines sont toutes revenues inférieures aux limites de quantification. L’analyse des pics d’exposition, l’étude d’autres aérocontaminants et d’autres secteurs agricoles, sont des perspectives de ce projet. Conclusion. Notre travail soulève de nouvelles questions de recherche quant à des associations possibles entre certaines expositions professionnelles en milieu agricole (cultures et tâches, utilisation de certains pesticides) et risque de CBP, et contribue à une meilleure connaissance des expositions professionnelles des agriculteurs aux aérocontaminants. / Background. Farmers are at lower risk of lung cancer, compared to the general population, at least partly because of a reduced rate of smoking among the elderly and women. A “protective” effect of endotoxins has also been hypothesized. Even though some of the occupational exposures occurring during farming-related activities are well-known or suspected lung carcinogens, there is scarce data in the literature, about the associations between specific occupational exposures or tasks in farming, and the risk of lung cancer. Objectives.The first aim was to analyse, in the AGRICAN cohort, the associations between lung cancer and (1) occupational exposure to various crops and related tasks; (2) arsenical pesticides and dinitroaniline herbicides use in farming. We also aimed at stratifying our analysis by histological subtypes of lung cancer. The second aim was to initiate a metrological project, to assess farmers’ occupational exposure to various airborne pollutants, in relation with the risk of lung cancer. Methods. We used the data from the French AGRIculture and CANcer cohort (AGRICAN) and from the PESTIcide MATrix project (PESTIMAT), to assess the associations between occupational farming exposures and lung cancer. A Cox model was performed, with adjustment on gender, smoking intensity, cattle and horses breeding. Our metrological project AirExpA was first launched in Normandy and focused on a few exposures – total dusts in real-time ; endotoxins and 38 mycotoxins, cumulative – in cattle breeding, wheat/barley and peas harvesting, and haymaking. Results and discussion. We reported a negative association between lung cancer and corn growing, and positive associations between (1) vineyard growing and adenocarcinoma; (2) peas growing and small cell lung cancer. Corn growing, especially grain handling, might expose to significant levels of endotoxins, thus explaining the inverse association. Peas growing, especially harvesting, is reportedly a dusty task; the AirExpA project will enable us to better characterize the dusts. We found no association between arsenicals use and lung cancer, nor with dinitroaniline herbicides use. However, stratifying on gender and smoking status showed positive associations among women and never-smokers, especially for adenocarcinomas. The AirExpA project showed dust and endotoxins exposure gradients in cattle breeding, depending on the task, and a mild but real exposure to endotoxins in various tasks in crop growing. All samples for mycotoxins were below the limit of quantification. Analysing exposure peaks, extending the project to other breedings/crops and air pollutants are among the perspectives. Conclusion. Our results raise new research questions about possible associations between occupational agricultural exposures and lung cancer, and contribute to a better knowledge and understanding of farmers’ occupational exposure to air pollutants.