Association Study Of Atopic Dermatitis,Eczema And Asthma With Ageing-Related Phenotypes

Although some studies have suggested that the occurrence and development of allergic diseases such as atopic dermatitis,eczema and asthma are closely related to a variety of aging related phenotypes such as diabetes,hypertension and cognitive decline,the existing evidence has certain defects in research design and sample size,and the conclusions are quite heterogeneous.This thesis selects a variety of aging related phenotypes that have not been reported or remain debated in previous studies.Through four independent studies,we explored the correlation between atopic dermatitis(AD),eczema,asthma with age-related phenotypes at the population level,as follows:Study 1:Association between atopic dermatitis and asthma with serum lipidsObjective:This work is to evaluate the association between asthma/AD and serum lipids(SLs)and investigate the potential causal relationship.Methods:A large-scale cross-sectional study based on the UK Biobank(UKB).The four main blood lipid indexes,including total cholesterol(TC),LDL-C,HDL-C,and triglycerides(TG)were logarithmically transformed,and the univariate and multivariate linear regression models were established with atopic dermatitis and asthma.We used the published summary statistical dataset of genome wide association studies(GWAS)an inverse variance weighted(IVW)two sample Mendelian randomization(MR)study to explore the causal association between allergic diseases and SLs.Results:A total of 502505 participants were included in analysis.After full adjustment,AD was associated with lower TG(β=-0.006,P=0.006),lower LDL(β=-0.004,P<0.001),and lower TC(β=-0.004,P<0.001)but insignificantly correlated to HDL(P=0.794).Asthma was also inversely correlated to TG(β=-0.005,P<0.001),LDL(β=-0.003,P<0.001),and TC(β=-0.002,P<0.001),but was positively correlated to HDL(β=0.004,P<0.001),respectively.In subsequent MR analysis,both allergic diseases(β=-0.066,P=0.017)and asthma(β=-0.057,P=0.001)showed a protective effect on TC.Allergic diseases(β=-0.051,P=0.028)and asthma(β=-0.037,P=0.008)showed a negative effect on LDL(β:-0.066-0.019).Conclusion:Collectively,we identify a protective causal effect of allergic diseases on serum lipids.Study 2:Association between atopic dermatitis and asthma with sarcopeniaObjective:To explore the association of allergic phenotypes including AD,asthma,and eosinophils count(EC),and sarcopenia.Methods:We conducted cross-sectional,cohort study and MR analysis based on UKB,the China Health and Retirement Longitudinal Study(CHARLS)and data from GWAS.A large-scale cross-sectional study was carried out using the population data of UK biobank(UKB)to explore the relationship between atopic dermatitis,asthma and the number of eosinophils in peripheral blood and sarcopenia and pre-sarcopenia through logistic regression.A cohort study was conducted using the China Health and Retirement Longitudinal Study(CHARLS)to explore the association between asthma and preclinical sarcopenia through log-binomial regression.The published GWAS data set and the population data of UKB were used to perform Mendelian randomized analysis of two samples and one sample respectively to explore the effects of atopic dermatitis,asthma and peripheral blood eosinophil data on muscle strength and appendicular lean mass.Results:Based on the UKB data,asthma and EC were positively correlated with sarcopenia(adjusted OR=1.533 aOR=1.533,P<0.001for asthma,adjusted OR=1.604,P=0.019 for EC)and presarcopenia(adjusted OR=1.404 for asthma,adjusted OR=1.366 for EC,P<0.001)both in crude and adjusted models(P<0.005).Based on the CHARLS cohort,asthma significantly added 162.2%risks for presarcopenia in the crude model(RR=1.73,P<0.001).After adjustment,asthma could still add 109.4%extra risk of pre-sarcopenia compared to the control group(RR=2.094,P=0.002).According to the one-sample MR analysis,both asthma(β=-2.011,P=0.004)and EC(β=-1.535,P=0.001)exerted casual effects on significantly decreased hand grip.Marginally significant casual effects of AD(β=-0.043,P=0.058)and asthma(β=0.026,P=0.061)on decreased appendicular lean mass(ALM)were observed in two-sample MR.When the outlier I Vs were removed,the casual relationship between AD/asthma and ALM became significant,although the effect sizes remained small.EC showed a significant effect on ALM.Conclusion:Atopic dermatitis,asthma and the increase in the number of eosinophils in peripheral blood have a significant causal impact on sarcopema.Study 3:Association between atopic dermatitis and asthma with dementia and Parkinson’s diseasesObjective:To evaluate the association and potential causal relationship between AD,asthma and dementia and Parkinson’s disease.Methods:A large-scale cohort study was conducted using the population data of the UK biobank(UKB)to explore the relationship between AD,asthma,dementia and Parkinson’s disease through Cox regression.A subgroup analysis was conducted based on education level.Two sample MR analysis was conducted using the published GWAS data set to explore the causal relationship between AD and asthma and Alzheimer’s disease,cognitive function and Parkinson’s disease.Results:In the cohort study,10049 and 483043 participants entered the AD exposure group and its control group respectively,and 58059 and 425714 participants entered the asthma exposure group and its control group respectively.In univariate analysis,AD significantly increased the additional risk of Parkinson’s disease by 31.7%(HR=1.31,P=0.025),asthma significantly increased the extra risk of dementia by 22.0%(HR=1.31,P<0.001),while the effect of AD on dementia or asthma on Parkinson’s disease was not statistically significant(P>0.05).After adjusting for age,gender,alcohol consumption,smoking,body mass index,family income,and physical activity,the HR for AD on Parkinson’s disease was 1.37(P=0.010)and asthma on dementia was 1.27(P<0.001).Subgroup analysis based on education level showed that the effect of asthma on dementia was greatly affected by education level,and the population association was still significant at O/GESE level and NVQ/HND/HNC level.The other groups were not significantly different from incidence rate of dementia after UKB cohort education.MR analysis did not support the causal relationship between asthma and Alzheimer’s disease(P=0.679).However,asthma has a negative causal effect on cognitive function(β=-0.052,P=0.011),and this effect did not have pleiotropy(P>0.679).The effect of AD on Parkinson’s disease was not significant(P=0.871).Conclusion:AD can significantly increase the risk of Parkinson’s disease,while asthma can significantly increase the risk of dementia.Subgroup analysis shows that the effect of asthma on dementia is closely related to the level of education.The effect of asthma on dementia may be related to its causal effect on cognitive level.Study 4:Association between eczema and asthma with allostatic loadObjective:To explore the relationship between the prevalence of different allergic diseases and allostatic load.Methods:A cross-sectional study was conducted using data from the National health and Nutrition Examination Survey(NHANES)in the United States from 1999 to 2010.The included variables related to allergic diseases mainly include current eczema and eczema in the past year,current asthma,asthma in the past year and asthma ever.10 indexes including systolic blood pressure,diastolic blood pressure,body mass index,glycosylated hemoglobin,total cholesterol,high-density cholesterol,total cholesterol/high-density lipoprotein,serum albumin,creatinine clearance and C-reactive protein were included for allostatic load score(ALS)calculation,and the clinical recommended value was taken as the cut-off value for scoring.Logistic regression was used for correlation analysis,and the weight in the survey was considered.Results:Except for the previous asthma group and the group with eczema in the past one year,there were significant differences in adaptive load scores between the allergic phenotype groups and their respective control groups(P<0.05).The results of regression analysis showed that the correlation was not significant neither between ALS and current eczema nor ALS and eczema in past one year after adjusting covariates(P>0.05).For asthma,however,there was a significant positive correlation between ALS and current asthma and asthma in the past one year.For previous asthma,although the association between ALS and previous asthma in univariate regression was not statistically significant(P=0.842),the association became significant after adjustment(OR=1.026;P=0.014).Sensitivity analysis showed that the regression performance of each group was more consistent with the results of analysis based on clinical cut-off value,indicating that the results of correlation analysis were valid.Conclusion:Several phenotypes of asthma were significantly correlated with ALS.From the perspective of aging,our results suggest that the onset of asthma may have an impact on allostatic load.