Associations Between Changes In Anthropometric Indices And Urinary 8-hydroxydeoxyguanosine Levels

Part One Development of gradient elution high-performance liquid chromatography method for urinary 8-hydroxydeoxyguanosine levelsObjectives:The purposes of this study were to develop a gradient-elution high-performance liquid chromatography with electrochemical detection(HPLC-ECD)method and to measure urinary 8-hydroxydeoxyguanosine(8-OHdG)levels among 3-years follow-up Wuhan participants of the Wuhan-Zhuhai Cohort Study.Methods:Urine samples were pre-treated by using Bond Elut LRC C18-OH solid phase extraction columns.Urinary 8-OHdG levels in the extracted samples were analyzed by gradient-elution HPLC-ECD method.X SELECT*T(3.5?m,4.6×150mm)was selected as the analytical column.The mobile phase consisted of1%methanol solution(A)and 20%methanol solution(B).The flow rate was1.0mL/min.Results:The calibration curves for 8-OHdG had good linearity in the concentration range of 28-1400nmol/L(R~2=0.9996-0.9999).The detection limit was 2.57nmol/L and the average recovery was 85.7%.The relative standard deviation ranged from0.07%to 7.97%.Conclusions:The established gradient-elution HPLC-ECD method is available tomeasure urinary 8-OHdG levels,particularly in large populations.The method has the advances in saving time and higher repeatability and provides a technical support for assessing oxidative DNA damage.Part Two Associations between changes in anthropometric indices and urinary8-hydroxydeoxyguanosine levelsObjectives:The purpose of this study aimed to explore the associations between changes in anthropometric indices and urinary 8-OHdG levels.Methods:Participants were from the Wuhan residents at baseline(n=3053)of the Wuhan-Zhuhai Cohort Study(established in 2011).Among them,2283(74.78%)participants finished a 3-year follow-up.We excluded individuals with missing data on anthropometric indices(n=34)and urine samples(n=206)at baseline.Then,we further excluded individuals with missing data on anthropometric indices(n=743)and urinary 8-OHdG levels(n=149)at follow-up.Finally,1151 participants were included in the longitudinal analysis.Urinary levels of 8-OHdG were measured by gradient-elution HPLC-ECD method and calibrated by the creatinine values.Differences between means in both groups(such as weight,waist circumference,hipcircumference(HC)and body mass index(BMI))were compared by Student's t-test.Differences between frequencies in both groups(such as education levels,smoking status and drinking status)by using Chi-square test.Non-parametric variables(such as waist-to hip ratio(WHR)and urinary 8-OHdG levels)were compared by Mann-Whitney U test.Restricted cubic spline function(RCS)was used to analyze theassociations between changes in anthropometric indices and urinary 8-OHdG levels after adjusting for confounding factors.Furthermore,urinary 8-OHdG levels were adjusted for urinary creatinine levels and then log-transformed prior to analysis due to the skewed distribution.Participants were classified into four groups of?-2.5%,-2.5%-2.5%(reference group),2.5%-5%and>5%,according to proportional changes in values of hip circumference(HC)during 3-year follow-up survey period.Statistical differences in normal distribution variables(such as age and urinary 8-OHdG levels)were determined by one-way analysis of variance(ANOVA),followed by post-hocmultiple comparison analysis with Fisher's least significant difference test(LSD-t test).The generalized linear models were used to analyze the relationships between percentage changes in selected variables and change in urinary 8-OHdG levels during3-year follow-up survey period.Results:The results are summarized as follows:(1)After adjusted for age,gender,educational levels,active smoking,passive smoking,alcohol consumption,physical activity,dietary intake frequency,napping time,sleep duration and sleep quality,RCS models showed a does-response relationship between change in HC values andchange in urinary 8-OHdG levels among all participants(P for overall association<0.05).(2)Only individuals in the?-2.5%(-0.08±1.05?mol/mol Cr)and>5%(0.21±1.06?mol/mol Cr)in HC groups had greater change in urinary 8-OHdG levels compared with those in the-2.5%to 2.5%change in HC group(-0.37±1.10?mol/mol Cr)(both P<0.05).(3)The fitted models revealed that among individuals with normal weight(BMI<24kg/m~2),overweight(BMI?24kg/m~2)or central obesity(males:WHR?0.90,females:WHR?0.85)at baseline,percentage change in HC wasassociated with change in urinary 8-OHdG levels in a dose-response manner(P fortrend<0.05).The generalized linear models showed that after adjusted for age,gender,educational levels,active and passive smoking,alcohol consumption,physicalactivity,dietary intake frequency,napping time,sleep duration and sleep quality,as compared with the reference group(HC gain-2.5%-2.5%),baseline individuals with central obesity plus?-2.5%or>5%change in HC had a 0.290?mol/mol Cr(95%CI:0.108,0.472)or 0.553?mol/mol Cr(95%CI:0.273,0.833)increase in urinary 8-OHdG levels(both P<0.05),however,only baseline individuals with non-central obesity plus?-2.5%change in HC had a 0.361?mol/mol Cr(95%CI:0.096,0.625)increase in urinary 8-OHdG levels.(4)Compared with non-central obesity males at both baseline and 3-year follow-up visit,we observed a 0.30?mol/mol Cr(95%CI:0.06,0.55)increase in urinary 8-OHdG levels for central obesity males at 3-year follow-up visit but non-central obesity at baseline.Males with central obesity at both baseline and 3-year follow-up visit had a 0.46?mol/mol Cr(95%CI:0.16,0.75)increase in urinary 8-OHdG levels rather than those with non-central obesity males at both baseline and 3-year follow-up visit.Conclusions:The associations between HC gain and increase in urinary 8-OHdG levels a dose-dependent manner were found among all individuals.Central obesity males at both baseline and 3-year follow-up visit were at higher risk of increased urinary 8-OHdG levels.