Introdução: Glicemia de jejum alterada tem sido associada a maior risco de desenvolver diabetes, comparando a indivíduos normoglicêmicos. Apesar de diabetes ser relacionado a aterosclerose e a piores desfechos cardiovasculares, os dados de literatura relacionando glicemia de jejum alterada à doença aterosclerótica são conflitantes. Os objetivos deste trabalho foram determinar (a) a incidência de diabetes em indivíduos com glicemia de jejum normal ou alterada em 1998 após um seguimento de até 12 anos; (b) se a glicemia de jejum alterada em 1998 e/ou diabetes incidente estiveram associados com aterosclerose subclínica no Estudo Longitudinal da Saúde do Adulto (ELSA-Brasil) ou à variável combinada de eventos clínicos não-fatais e escore de cálcio coronariano maior ou igual a 400. Métodos: Avaliamos 1.536 trabalhadores da Universidade de São Paulo, que participaram de um programa de avaliação em 1998 (idade 23-63 anos) e da linha de base do ELSA-Brasil (2008-2010). Apresentamos as taxas de incidência de diabetes brutas e ajustadas para todos os indivíduos e também estratificados por gênero e por índice de massa corpórea (IMC) em 1998. Utilizamos modelos de regressão brutos e ajustados para estimar a associação entre glicemia de jejum alterada em 1998 ou diabetes incidente com a espessura de íntima-média de carótidas (EIMC), escore de cálcio coronariano (CACS, do inglês Coronary Artery Calcium Score) e a variável composta CACS >= 400 ou eventos cardiovasculares incidentes (infarto do miocárdio ou revascularização). Resultados: Encontramos diabetes incidente em 177 indivíduos. A incidência de diabetes em nossa amostra foi de 9,8/1.000 pessoas-ano (Intervalo de confiança de 95% [IC95%]: 7,7-13,6). A incidência foi mais elevada entre os homens (11,2/1.000 pessoas-ano, IC95%: 8,6-15,0) do que entre as mulheres (8,5/1.000 pessoas-ano, IC95%: 5,3-15,3). Glicemia de jejum alterada em 1998 mostrou associação com maior risco de progressão para diabetes ao longo do seguimento (hazard ratio [HR]: 3,17; IC95%: 2,14-4,68) e HR: 7,42; IC95%: 4,75-11,57 para glicemias de jejum entre 100 e 109mg/dl e entre 110 e 125mg/dl, respectivamente). Glicemias entre 110 e 125mg/dl em 1998 foram associadas a maiores valores de EIMC (beta=+0,028; IC95%: 0,003 a 0,053) na linha de base do ELSA-Brasil. Ao excluir da análise aqueles com diabetes incidente, houve associação limítrofe, não-significativa, entre maiores valores de EIMC e glicemia de jejum entre 110 e 125mg/dl em 1998 (?=0,030; IC95%: -0,005 a 0,065). Ambos os níveis de glicemia de jejum alterada em 1998 não se mostraram associados ao CACS ou à variável composta de CACS >= 400 ou eventos cardiovasculares incidentes nos modelos de ajuste completo. Diabetes incidente foi associado a maiores valores de EIMC (em milímetros) (?=0,034; IC95%: 0,015 a 0,053), a CACS >= 400 (Razão de chances=2,84; IC95%: 1,17-6,91) e ao desfecho combinado de CACS >= 400 ou eventos cardiovasculares incidentes (Razão de chances=3,50; IC95%: 1,60-7,65). Conclusões: Glicemia de jejum alterada em 1998, especialmente nos valores mais próximos dos limiares de corte para diabetes, foram associados a maior incidência de diabetes ao longo do seguimento e a maiores valores de EIMC quando da avaliação inicial do ELSA-Brasil. Diabetes incidente entre as avaliações foi associado a maior risco cardiovascular / Introduction: Impaired fasting glucose has been associated with higher risk of incident diabetes, compared to normoglycemic individuals. Although diabetes mellitus is related to atherosclerosis and higher long-term cardiovascular burden, there are conflicting data about the association between impaired fasting glucose and atherosclerotic disease. We aimed (a) to determine diabetes incidence rates in individuals with normal or impaired fasting glucose in 1998 after follow-up of up to 12 years, (b) whether impaired fasting glucose in 1998 and/or incident diabetes were associated with subclinical atherosclerosis in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) or the combined variable of non-fatal clinical events or a coronary calcium score >= 400. Methods: We evaluated 1,536 civil servants from the University of São Paulo, who participated in both 1998 (aged 23-63 years) and ELSA-Brasil baseline (2008-2010) assessments and had complete data. We presented crude and adjusted diabetes incident rates for all individuals and then stratified by sex and body mass index (BMI) in 1998. We used crude and adjusted regression models to estimate the association between impaired fasting glucose in 1998 or incident diabetes and coronary intima-media thickness (CIMT), coronary artery calcium score (CACS) and the composite variable of a CACS?400 or incident cardiovascular events (myocardial infarction or revascularization). Results: We found incident diabetes in 177 individuals. Diabetes incidence in our sample was 9.8/1,000 person-years (95% confidence interval [95%CI]:7.7-13.6). Diabetes incidence was higher in men (11.2/1,000 person-years, 95%CI: 8.6-15.0) than women (8.5/1,000 person-years, 95%CI: 5.3 to 15.3). Impaired fasting glucose in 1998 was associated with a higher risk of progression to diabetes during follow-up (hazard ratio [HR]: 3.17; 95%CI: 2.14-4.68 and HR: 7.42; 95%CI: 4.75-11.57 for a fasting plasma glucose between 100 to 109mg/dl and 110 to 125 mg/dl, respectively). Fasting plasma glucose levels between 110 to 125 mg/dl in 1998 were associated with higher CIMT (beta=+0.028; 95%CI: 0.003 to 0.053) in ELSA-Brasil baseline. Excluding those with incident diabetes, there was a non-significant borderline association between higher CIMT (in mm) and fasting plasma glucose 110 to 125mg/dl (beta=0.030; 95%CI: -0.005 to 0.065). Fasting plasma glucose levels in 1998 were not associated with CACS or the composite variable of a CACS ? 400 or incident cardiovascular events in full-adjusted models. Incident diabetes was associated with higher CIMT (in mm) (beta=0.034; 95%CI: 0.015 to 0.053), CACS >= 400 (OR=2.84; 95%CI: 1.17-6.91) and the combined outcome of a CACS >= 400 or incident cardiovascular event (OR=3.50; 95%CI: 1.60-7.65). Conclusions: Elevated fasting plasma glucose in 1998, especially those near diabetes diagnosis limits were associated with higher diabetes incidence during follow-up and higher CIMT in ELSA-Brasil baseline assessment. Incident diabetes between assessments was associated with higher cardiovascular burden

Introdução: Doenças tireoidianas subclínicas incluem hipotireoidismo e hipertireoidismo subclínicos. A associação entre doença tireoidiana subclínica e morbimortalidade cardiovascular é controversa e os dados sobre a relação entre essas condições clínicas e aterosclerose subclínica são escassos. Objetivos: Este estudo objetiva avaliar a associação entre doença tireoidiana subclínica, calcificação arterial coronariana (CAC), doença arterial coronariana (DAC), índice de espessura de médio-íntima carotídea média (IMT) e velocidade de onda de pulso carotídeo-femoral (cf-VOP) no Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil). Métodos: Incluímos sujeitos eutireóideos, definidos como tendo TSH entre 0,4 e 4,0 mUI/L e T4L entre 0,8 e 1,9ng/dL, indivíduos com hipotireoidismo subclínico, definido como TSH > 4,0 mUI/L e T4L normal, e hipertireoidismo subclínico, definido como TSH < 0,4 mUI/L e T4L normal. Excluímos os indivíduos com as demais disfunções tireoidianas, em uso de medicação que altera a função tireoidiana, e com doença cardiovascular prévia. Na análise de angiotomografia, excluímos também os sujeitos com hipertireoidismo subclínico pelo pequeno número que impedia a análise e, na análise de cf-VOP, doença renal crônica, indivíduos em uso de diuréticos e de anti-hipertensivos. As associações entre quintis de TSH, CAC > 100 e DAC foram avaliadas por regressão logística e as associações entre IMT, VOP (como variáveis contínuas ou categorizadas com ponto de corte no percentil 75 amostral) e níveis de TSH ou doenças tireoidianas subclínicas foram avaliadas por regressões logísticas e lineares multivariadas. Todos os modelos foram ajustados por variáveis demográficas e fatores de risco cardiovasculares. Resultados: A análise de CAC incluiu 3.836 sujeitos, mediana de idade de 49 anos (IQR=44-56), 1.999 (52,1%) mulheres. CAC > 100 associou-se independentemente com o primeiro quintil (OR ajustado=1,57, IC 95%=1,05-2,35, P=0,027), usando o terceiro como referência. Na análise de angiotomografia, foram incluídos 796 sujeitos, mediana de idade de 55 anos (IQR=48-60 anos), 406 (51%) mulheres. O primeiro quintil associou-se independentemente com CAC (OR ajustado=1,76, IC 95%=1,09-2,82, P= 0,02), DAC (OR ajustado=1,73, IC 95%=1,08-2,79, P=0,023), mas não com extensão de doença. Na análise de IMT, foram incluídos 8.623 sujeitos, mediana de idade de 50 anos (IQR=45-57 anos), 4.624 (53,6%) mulheres, na subanálise de hipotireoidismo subclínico, e 8.193, com mediana de idade de 50 anos (IQR=44-57 anos), 4.382 (53,5%) mulheres, na subanálise de hipertireoidismo subclínico. Hipotireoidismo subclínico, mas não hipertireoidismo subclínico, associou-se ao IMT como variável contínua (beta=0,010, IC 95%=0,0004-0,019, P=0,041) e categorizado no percentil 75 ajustado para sexo, idade e raça (OR ajustado=1,30, IC95%=1,07-1,61, P=0,010). Na análise de cf-VOP, foram incluídos 8.341 sujeitos, mediana de idade de 50 anos (IQR=44-56 anos), 4.383 (52,5%) mulheres, na subanálise de hipotireoidismo subclínico, e 7.790, mediana de idade de 50 anos (IQR=44-57 anos), 4.191 (53,8%) mulheres, na subanálise de hipertireoidismo subclínico. Cf-VOP não se associou com doença tireoidiana subclínica. Conclusões: Em análises diferentes, CAC e DAC associaram-se com primeiro quintil de TSH usando-se o terceiro como referência. O IMT associou-se com hipotireoidismo subclínico e a cf-VOP não se associou com disfunção tireoidiana subclínica / Introduction: Subclinical thyroid disease includes subclinical hypothyroidism and subclinical hyperthyroidism. Association between subclinical thyroid disease and cardiovascular morbidity and mortality is controversial and data about the relationship between those clinical conditions and subclinical atherosclerosis is scarce. Objectives: This study aims to evaluate the association between subclinical thyroid disease, coronary artery calcification (CAC), coronary artery disease (CAD), mean common carotid intima-media thickness (IMT) and carotid-femoral pulse wave velocity (cf-PWV) in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Methods: We included euthyroid subjects, defined as TSH between 0.4 and 4.0 mIU/l and FT4 between 0.8 and 1.9 ng/dL, and individuals with subclinical hypothyroidism, defined as TSH > 4.0 mIU/l and normal FT4, and subclinical hyperthyroidism, defined as TSH < 0.4 mIU/L and normal FT4. We excluded individuals with other thyroid disorders, subjects who used medication that altered thyroid function, subjects with past of cardiovascular disease. In computed angiotomography analysis, we have excluded subjects with subclinical hyperthyroidism because of the small sample, and in cf-PWV analysis, we have excluded individuals with chronic kidney disease, use of anti-hypertensive and diuretics. The association between TSH quintiles was evaluated in logistic regression models for CAC and CAD, and the association between IMT, cf-PWV (as continuous variables or as factor, categorized at 75th sample\'s percentile) and TSH levels or subclinical thyroid diseases was evaluated by multivariate logistic and linear regression models. All models were adjusted for demographic variables and cardiovascular risk factors. Results: CAC analysis included 3,836 subjects, median of age 49 years (IQR=44-56), 1,999 (52.1%) women. CAC > 100 was independently associated with first quintile of TSH, using the third quintile as the reference (adjusted OR=1.57, 95% CI=1.05-2.35, P=0.027). Computed angiotomography analysis included 796 subjects, median of age 55 years (IQR=48-60), 406 (51%) women. CAD and CAC > 0 was independently associated with first quintile in comparison with third quintile (adjusted OR=1.73, 95% CI=1.08-2.79, P=0.023 and adjusted OR=1.76, 95% CI=1.09-2.82, P= 0.02, respectively), but not with burden of disease. In IMT analysis, 8,623 subjects were included, median of age 50 years (IQR=45-57 years), 4,624 (53.6%) women in the subclinical hypothyroidism subanalysis, and 8,193, median age 50 years (IQR = 44-57 years), 4,382 (53.5%) women, in the subclinical hyperthyroidism subanalysis. Subclinical hypothyroidism, but not subclinical hyperthyroidism, was independently associated with IMT as continuous variable (beta=0.010, IC 95%=0.0004-0.019, P=0.041) or as factor categorized at 75th percentile adjusted for age, sex and race (adjusted OR=1.30, 95% CI=1.07-1.61, P=0.010). In cf-PWV analysis, 8,341 subjects were included, median of age 50 years (IQR=44-56 years), 4,383 (52.5%) women in the subclinical hypothyroidism subanalysis, and 7,790, median age 50 years (IQR = 44-57 years), 4,191 (53.8%) women in subclinical hyperthyroidism subanalysis. Cf- PWV was not associated with subclinical thyroid disease. Conclusion: In separated analysis, CAC and CAD was independently associated with first quintile of TSH using the third as the reference; IMT was independently associated with subclinical hypothyroidism, and cf-PWV was not associated with subclinical thyroid diseases

Cardiovascular disease varies by ethnicity in the UK. South Asians (SA) have higher coronary heart disease (CHD) and diabetes prevalence, while African-Caribbeans (AfC) have greater stroke, but intriguingly lower CHD rates despite higher blood pressures and diabetes risk than Europeans. Conventional risk factors do not fully explain such differences. This cross-sectional study tested the hypothesis that the hormones, vitamin D measured as 25(OH)D and aldosterone, would be independently associated with intermediate cardiovascular outcome markers in these ethnic groups. Community-dwelling men 40-80 years old (AfC: n=67, 55±10yr; SA: n=68, 55±10yr; European: n=63, 57±8yr) were sampled from Greater Manchester’s multi-ethnic population. The intermediate markers examined were aortic pulse wave velocity (aPWV), left ventricular (LV) mass and function, and carotid intima media thickness (CIMT), measured non-invasively by ultrasound, and hemodynamic profiling methods (the Arteriograph) in the total sample and by magnetic resonance imaging (MRI) in a subsample of 50. Adjusted for age, systolic blood pressure and diabetes, mean(SE) aPWV by the Arteriograph, was 0.5(0.2) m/s higher in SA than AfC and Europeans (p=0.01), which paralleled known cross-ethnic CHD risk differences in the UK. By MRI, aPWV along the descending aorta in SA was 0.7(0.3) and 0.8(0.3) m/s higher than that in AfC and Europeans, but aPWV along the aortic arch was not significantly different. Unlike aldosterone, 25(OH)D was independently and inversely correlated with aPWV (unstandardised B(SE)=-0.013[0.004] m/s, p<0.001), and partly explained the ethnic variation in aPWV. Similar inverse correlations were found between 25(OH)D and LV concentricity measured by echocardiography and MRI. Compared to Europeans, SA and AfC, had 21(3) and 14(3) nmol/L lower mean(SE) 25(OH)D, respectively (p<0.01). Mean(SE) of relative wall thickness, an index of LV concentricity by echocardiography, was 0.05(0.01) higher in SA and AfC than Europeans. Lower 25(OH)D levels were also associated with higher myocardial deformation rates measured by MRI myocardial tagging (n=50), supporting previous animal experimental evidence. A one standard deviation (SD) decrease in 25(OH)D was associated with a 0.38 SD increase in absolute systolic strain rate (p=0.003) and 0.22 SD rise in diastolic strain rate (p=0.04). Right and left CIMT showed different relations with 25(OH)D and aldosterone. Left-right CIMT differences varied by ethnicity and were related to SA ethnicity and aldosterone levels. Two related technical studies investigated the relatively new method of hemodynamic profiling, the Arteriograph, used here. The results suggested a standardisation method of aortic length estimation for purely central aPWV, which significantly improved aPWV agreement between the Arteriograph and MRI (reference method here), and was used for calibrating the Arteriograph aPWV in the above-mentioned results for the total sample. Future well-designed trials are necessary to investigate any cause-effect relationship between vitamin D deficiency and the unfavourable cardiovascular intermediate outcomes found here in a cross-sectional design and multi-ethnic background.