A carregar...
Publicação
Polimorfismos genéticos e suas interações no risco de hipertensão arterial, numa população da Ilha da Madeira. associação com as lesões dos órgãos alvo
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 3.27 MB | Adobe PDF | |||
| 4.01 MB | Adobe PDF |
Orientador(es)
Resumo(s)
RESUMO: Introdução: A Hipertensão essencial (EH) afeta um quarto dos adultos em todo o mundo, e estima se o seu aumento para um terço em 2025. A sua prevalência global na população adulta
portuguesa é cerca de (42%), sendo (44.4% nos homens e 40.2% nas mulheres).
A etiologia da HTA é suscetivel de compreender como sendo um distúrbio multifatorial
resultante de fatores ambientais e genéticos e das suas interações.
Estudos epidemiológicos revelaram que 20-40% da variação da pressão arterial é
determinada geneticamente.
Objetivos:
Avaliar quais as variantes genéticas que se associam com EH, numa população da Ilha
da Madeira. Estudar as suas interações gene-gene e com os fatores de risco tradicionais
(FRT), no aparecimento de HTA.
Investigar, na população de Hipertensos, quais as variantes genéticas que se associam
com as lesões de orgão alvo (LOA), assim como as suas interações com os Fatores
preditivos de LOA, na eclosão destas lesões.
Construção de um score genético de risco hipertensivo (SRG), calcular o seu poder
preditivo e compará- lo com o risco conferido pelos FRT. Finalmente estimar o valor
adicional conferido por esta associação.
Métodos:
Com uma amostra de 1712 indivíduos com idade média de 51±7,9 anos (51% género
masculino), fizemos um estudo caso-controlo. O grupo dos casos constituido por 860
hipertensos com idade média 51.4 ± 8 anos; 53.3% género masculino). Ambos os grupos
foram ajustados em termos de sexo e idade. Considerou-se ter HTA quando os
individuos à entrada do estudo já estavam diagnosticados e /ou faziam terapêutica antihipertensora
ou apresentavam TA ≥140/90 mmHg em mais de 3 leituras. Os controlos
normotensos, não faziam qualquer medicação ou apresentavam PAS/PAD <140/90
mmHg.
Comparámos entre os grupos as suas carateristicas clínicas, parâmetros laboratorias,
assim como as 14 variantes genéticas estudadas que se associavam ao aparecimento
de HTA.
Foi investigada através de duas análises de regressão logística a associação dos fatores
de risco ambientais e genéticos no aparecimento de HTA.
Por Data mining nomeadamente Multifactor Dimensionality Reduction (MDR), foram
analisadas as interações gene-gene e gene-ambiente no aparecimento de HTA.
Através de duas curvas ROC foi comparado o valor preditivo dos fatores de risco
tradicionais e genéticos e o valor adicional desta associação, no aparecimento de HTA.
Seguidamente num grupo de 600 hipertensos foram estimadas as Lesões de Orgão Alvo
(LOA) e estudados os fatores genéticos e não genéticos de risco para LOA.
vi
Por MDR, foram investigadas as interações das variantes genéticas e fatores preditores
não genéticos que na análise univariada se associaram ao aparecimento de LOA. Foram
construídos dois modelos de Score genético (SRG), tendo sido usado o modelo
multiplicativo e avaliado através de duas curvas ROC o seu poder preditivo de HTA assim
como o valor adicional da junção aos FRT. Este modelo foi aplicado aos dois géneros.
Resultados:
Das variantes genéticas investigadas as que, na nossa população e no melhor modelo
genético se associaram de forma significativa com a Hipertensão Arterial, foram a ACE
rs4340 (OR 1,233 p=0,032), ACE 2350 rs4343 (OR 1,244, p=0,036), ADRB1 rs1801253 (OR
1,408, p=0,033), ADD1 rs4961 (OR 2,763 p=0,003) e GNB3 rs5443 (OR 1,339, p=0,004).
O melhor modelo de interação gene/ambiente, no risco de HTA, foi o que era constituído
pela junção da variante ADD1, pela existência de obesidade e de diabetes na população,
com OR nos dados de teste = 3,79 (IC a 95% 1,92-7,50) e com consistência de validação
cruzada 10/10 (p<0,0001).
Com duas curvas AUC calculámos e comparámos a capacidade preditiva dos FRT e das 5
variantes genéticas + FRT. Existia diferença significativa (p=0,014), entre as duas curvas.
No grupo de doentes hipertensos com lesão/Insuficiência Renal o polimorfismo ATP2B1
associou-se, de forma significativa, com a existência de Insuficiência/lesão renal (OR
3,726; p=0,028).
As variantes genéticas que se associaram na análise univariada com o aparecimento de
Lesão de Órgãos Alvo, nos hipertensos, foram o ADRB1 (p=0,005), ATP2B1 (p=0,041) e
SCNN1G (p=0,014) (de forma protetora).
O melhor modelo de interação entre fatores genéticos e não genéticos preditivos de
Lesão de Órgãos Alvo foi o constituído pela idade, tempo de diagnóstico e pelos genes
ADRB1 e ATP2B1 com OR no treino de 2,58 (IC a 95% 1,82 – 3,66; p<0,0001) e com
consistência de validação cruzada de 10/10.
Dos dois scores elaborados, o multiplicativo mostrou melhor poder preditivo de HTA (OR
1,592; p <0,0001).
Quando aos Fatores de Risco Tradicionais se adicionava o SRGm, o incremento adicional
do risco de HTA foi significativo (p=0,0004). No género masculino este acréscimo foi
também significativo (p=0,004) e a sua significância manteve-se no género feminino
(p=0,006), embora com um valor ligeiramente inferior.
Conclusões:
Existem na nossa população da Ilha da Madeira, variantes genéticas que se associaram
com HTA (ACE, ACE 2350, ADRB1, ADD1, GNB3).
Quando estudada a interação gene ambiente, o melhor modelo foi o que incluía a
variante ADD1, obesidade e diabetes.
Nos hipertensos encontrámos 3 variantes genéticas que influenciaram o aparecimento
de LOA: ADRB1, ATP2B1 e SCNN1G.
O SRGm associou-se de forma significativa e independente com o aparecimento de HTA
Quando aos FRT se adicionou o SRGm o modelo apresentou incremento no seu valor
preditivo de HTA.
ABSTRACT: Essential Hypertension (EH) affects one-quarter of adults worldwide, and is estimated to increase to one-third by 2025. Its overall prevalence in the Portuguese adult population is about 42%, of which 44.4% is in men and 40.2% in women. EH is a complex and multifactorial disorder resulting from environmental and genetic factors and their interactions. Epidemiological studies have shown that 20-40% of blood pressure variation is genetically determined. Objectives Our goal was to evaluate the genetic variants associated with EH in a population of Madeira Island, as well as to study gene-gene and gene-traditional risk factors (TRF) interactions in EH onset. We also intend to investigate, in the hypertensive population, which genetic variants are associated with the target organ damage (TOD), as well as their interactions with the predictive factors of TOD, in the occurrence of this injuries. Our intention was to construct a hypertensive genetic risk score (GRS), calculate its predictive power and compare it with the risk conferred by TRF in order to finally estimate the additional value conferred by this association. Methods We performed a case-control study with a cohort of 1712 individuals with a mean age of 51± 7.9 years (51% male), namely 860 patients diagnosed with EH (mean age 51.4 ± 8; 53.3% male) and 852 normotensive controls (mean age 50.7 ± 7.7; 48.7% male). Cases and controls were adjusted in terms of sex and age. EH was considered when patients, at the entry to this study, were already diagnosed and/or had been on antihypertensive medication. The diagnosed hypertensives had systolic blood pressure (SBP)/diastolic blood pressure (DBP) ≥140/90 mmHg measured on at least 3 occasions. The normotensive controls had never been treated with antihypertensive medication and presented a SBP/DBP <140/90 mmHg. We compared, between the groups, the clinical characteristics, laboratorial parameters as well as the 14 genetic variants studied that were associated with the advent of EH. Receiver Operating Characteristic Through two analysis of logistic regression we investigated the association of environmental and genetic risk factors in the emergence of EH. By Data mining, namely Multifactor Dimensionality Reduction (MDR), we analysed the gene-gene and gene-environment interactions in the onset of EH. The predictive value of traditional and genetic risk factors and the additional value of this association at the onset of EH were compared through two receiver operating characteristic (ROC) curve. Next, in a group of 600 hypertensive patients, the TOD and genetic and non-genetic risk factors for TOD were estimated. Using MDR, the interactions of the genetic variants and non-genetic predictive factors, which were found to be associated with TOD in the univariate analysis, were investigated. Two genetic score models (GRS) were built, using the multiplicative model and evaluated through two ROC curves, to access EH predictive power as well as the additional value of the TRF junction. This model was applied to both genders. Results Amongst the genetic variants investigated the ones that, in our population, were significantly associated in the best genetic model with EH were: ACE rs4340 (OR 1.233 p=0.032), ACE 2350 rs4343 (OR 1.244, p=0.036), ADRB1 rs1801253 (OR 1.408, p=0.033), ADD1 rs4961 (OR 2.763 p=0.003) and GNB3 rs5443 (OR 1.339, p=0.004). The best model of gene/environment interaction in the risk of EH was the one including the combination of the ADD1 variant, the presence of obesity and diabetes in the population, with an OR in test data of 3.79 (95% CI 1.92-7.5) and a cross-validation consistency of 10/10 (p<0.0001). With two AUC curves we calculated and compared the predictive capacity of the TRF and the 5 genetic variants + TRF. There was a significant difference (p=0.014) between the two curves. In the group of hypertensive patients with lesion/Renal Insufficiency, ATP2B1 polymorphism was significantly associated with the existence of renal insufficiency/injury (OR 3.726, p=0.028). The genetic variants that were associated in the univariate analysis with the appearance of TOD in hypertensive patients were ADRB1 (p=0.005), ATP2B1 (p=0.041) and SCNN1G (p=0.014) (in a protective way). The best interaction model between genetic and non-genetic predictors of TOD included age, time of diagnosis, and ADRB1 and ATP2B1 genes with training OR of 2.58 (95% CI 1.82-3.66, p <0.0001) and a cross-validation consistency of 10/10. Of the two elaborated scores, the multiplicative showed a better predictive power of EH (OR 1,592; p<0.0001). When SRGm was added to the TRF, the additional increase in EH risk was significant (p=0.0004). In the male gender this increase was also significant (p=0.004) and its significance remained in the female gender (p=0.006), albeit with a slightly lower value. Conclusions There are genetic variants in our population in Madeira Island, that were associated with EH (ACE, ACE 2350, ADRB1, ADD1, GNB3). When studying the gene-environment interaction, the best model was the one that included the variant ADD1, obesity and diabetes. In the hypertensive group we found 3 genetic variants that influenced the appearance of TOD: ADRB1, ATP2B1 and SCNN1G. When the SRGm was added to the TRF, the model showed an increase in its predictive value of HD.
ABSTRACT: Essential Hypertension (EH) affects one-quarter of adults worldwide, and is estimated to increase to one-third by 2025. Its overall prevalence in the Portuguese adult population is about 42%, of which 44.4% is in men and 40.2% in women. EH is a complex and multifactorial disorder resulting from environmental and genetic factors and their interactions. Epidemiological studies have shown that 20-40% of blood pressure variation is genetically determined. Objectives Our goal was to evaluate the genetic variants associated with EH in a population of Madeira Island, as well as to study gene-gene and gene-traditional risk factors (TRF) interactions in EH onset. We also intend to investigate, in the hypertensive population, which genetic variants are associated with the target organ damage (TOD), as well as their interactions with the predictive factors of TOD, in the occurrence of this injuries. Our intention was to construct a hypertensive genetic risk score (GRS), calculate its predictive power and compare it with the risk conferred by TRF in order to finally estimate the additional value conferred by this association. Methods We performed a case-control study with a cohort of 1712 individuals with a mean age of 51± 7.9 years (51% male), namely 860 patients diagnosed with EH (mean age 51.4 ± 8; 53.3% male) and 852 normotensive controls (mean age 50.7 ± 7.7; 48.7% male). Cases and controls were adjusted in terms of sex and age. EH was considered when patients, at the entry to this study, were already diagnosed and/or had been on antihypertensive medication. The diagnosed hypertensives had systolic blood pressure (SBP)/diastolic blood pressure (DBP) ≥140/90 mmHg measured on at least 3 occasions. The normotensive controls had never been treated with antihypertensive medication and presented a SBP/DBP <140/90 mmHg. We compared, between the groups, the clinical characteristics, laboratorial parameters as well as the 14 genetic variants studied that were associated with the advent of EH. Receiver Operating Characteristic Through two analysis of logistic regression we investigated the association of environmental and genetic risk factors in the emergence of EH. By Data mining, namely Multifactor Dimensionality Reduction (MDR), we analysed the gene-gene and gene-environment interactions in the onset of EH. The predictive value of traditional and genetic risk factors and the additional value of this association at the onset of EH were compared through two receiver operating characteristic (ROC) curve. Next, in a group of 600 hypertensive patients, the TOD and genetic and non-genetic risk factors for TOD were estimated. Using MDR, the interactions of the genetic variants and non-genetic predictive factors, which were found to be associated with TOD in the univariate analysis, were investigated. Two genetic score models (GRS) were built, using the multiplicative model and evaluated through two ROC curves, to access EH predictive power as well as the additional value of the TRF junction. This model was applied to both genders. Results Amongst the genetic variants investigated the ones that, in our population, were significantly associated in the best genetic model with EH were: ACE rs4340 (OR 1.233 p=0.032), ACE 2350 rs4343 (OR 1.244, p=0.036), ADRB1 rs1801253 (OR 1.408, p=0.033), ADD1 rs4961 (OR 2.763 p=0.003) and GNB3 rs5443 (OR 1.339, p=0.004). The best model of gene/environment interaction in the risk of EH was the one including the combination of the ADD1 variant, the presence of obesity and diabetes in the population, with an OR in test data of 3.79 (95% CI 1.92-7.5) and a cross-validation consistency of 10/10 (p<0.0001). With two AUC curves we calculated and compared the predictive capacity of the TRF and the 5 genetic variants + TRF. There was a significant difference (p=0.014) between the two curves. In the group of hypertensive patients with lesion/Renal Insufficiency, ATP2B1 polymorphism was significantly associated with the existence of renal insufficiency/injury (OR 3.726, p=0.028). The genetic variants that were associated in the univariate analysis with the appearance of TOD in hypertensive patients were ADRB1 (p=0.005), ATP2B1 (p=0.041) and SCNN1G (p=0.014) (in a protective way). The best interaction model between genetic and non-genetic predictors of TOD included age, time of diagnosis, and ADRB1 and ATP2B1 genes with training OR of 2.58 (95% CI 1.82-3.66, p <0.0001) and a cross-validation consistency of 10/10. Of the two elaborated scores, the multiplicative showed a better predictive power of EH (OR 1,592; p<0.0001). When SRGm was added to the TRF, the additional increase in EH risk was significant (p=0.0004). In the male gender this increase was also significant (p=0.004) and its significance remained in the female gender (p=0.006), albeit with a slightly lower value. Conclusions There are genetic variants in our population in Madeira Island, that were associated with EH (ACE, ACE 2350, ADRB1, ADD1, GNB3). When studying the gene-environment interaction, the best model was the one that included the variant ADD1, obesity and diabetes. In the hypertensive group we found 3 genetic variants that influenced the appearance of TOD: ADRB1, ATP2B1 and SCNN1G. When the SRGm was added to the TRF, the model showed an increase in its predictive value of HD.
Descrição
Palavras-chave
Hypertension Gene Polymorphism Genotype Gene-gene interation Gene - environment interation Population Madeira Island