Association Between Metal Exposure And Type 2 Diabetes Mellitus In Chinese Han Population

Objectives1.To establish an efficient and practical pretreatment technology for blood serum samples detection following direct sampling dilution.2.To find out an accurate analysis method by ICP-MS to determine serum Co,Ni,Cu,Zn,Ga,V,Cr,Mn,Fe,As,Se,Rb,Sr,Ru,Rh,Pd,Ag,Cd,Cs,Ba and so on.3.Aimed to investigate the distribution levels and characteristics of serum metal levels,providing the scientific basis for police-making of serum reference ranges.4.Analyzing the influence of demography factors(gender,sex,et al)and environmental aspects(smoke,drink,physical work and diet habit)to type 2 diabetes.5.To find out the potential association between different concentrations of metals and risk catching type 2 diabetes.6.To get the most influential types of metal elements,whose reciprocal actions with demography and environmental factors will be studied.Methods1.Study subjects and participantsThe case-control study consisted of 3114 subjects(1551 T2DM patients and 1563 controls)from Suzhou in Jiangsu Province.The T2DM cases who have been diagnosed by professional clinicians were selected without any restriction in age or sex.The diagnostic criteria were,fasting blood-glucose(FBG)?7.0mml/L,or random blood-glucose?11.0mml/L,or 2hPG in OGTT?11.1mml/L,people who have been diagnosed as T2DM and have been taking medicines were also seen as cases in our study.The controls,who were frequency-matched to the cases by age and sex,consisted of individuals from the same community.2.Questionnaire and biological samplesPre-investigation has been taken before the formal ones.The purpose,contents and attentions were informed to volunteers before taking questionnaire that was administered through face-to-face interviews by trained interviewers.After fully understanding,each or the vol unteers was suggested to sign an informed consent.The questionnaire contained items concerning demographic data(name,gender,nationality,age,culture,et al.),pattern of life(smoking,drinking status,previous passive smoking exposure during life,physical activities and diet information),previous and present medical conditions,hereditary factors such as family history(diabetes mellitus,hypertension,coronary heart disease,stroke,hyperlipidemia,et al.).All workers and subjects should conform to the regulation during the questionnaire and write carefully,avoiding mistake or repetition.Each subject donated 5-ml venous blood sample drawn into EDTA microanticoagulant tube by technical doctors and mixed immediately,the number of blood vessels and questionnaire should be consistent,all the blood samples were sent back to laboratory through the cold chain transmission every day.3.Confirmation of instrument conditions and methodology validationUse the tuned liquid of Li,Ce,Co,In,U,Bi and Ba with 10.00ng/ml to adjust equipment,making the signal-to-noise ratios of all elements in tuned liquid reach the highest.,The optimal internal standard elements,isotope and response mode were determined by analyzing the interference in the process of ICP-MS determination.The method to detect serum levels of elements by ICP-MS with determined sample processing procedures and instrument conditions were tested.The detection limits,recovery rates,linear range,precisions concluding within-day and between-day were calculated and assessed respectively.4.Determination of serum samplesInternal standard method was taken to determine the metal elements in human serum in our study,the selection principles of internal standards were as follows:(1)The physical and chemical properties,nucleon number and ionization energy of internal standards should be similar with analytical metal elements.(2)The internal standards can reflect the characteristics of metals in ICP-MS.(3)The internal standards ought not to be disturbed by other factors and also should not disturb analyzed metal elements.(4)Be very infrequent in natural world.(5)The analytes do not contain the internal standards or have rare in them.We selected Bi,Y,In,Tb as internal standards according to the above standards.After dilution with 0.5%HNO3+0.005%Triton X-100,the serum metals were directly determined by ICP-MS using internal standard method to adjust the instrument.Extracting 200?L serum sample into a 10ml polypropylene centrifugal tube,and then 200?L configuration fluid of internal standards(100?g/L),demarcating volume to 4mL with dilution(diluted by 20 times).After above processing,serum samples were detected for metals conversation by ICP-MS.The serum level of metal elements could be computed by multiplying the dilution ratios as twenty.5.Quality controlQuality control was requested in the procedures of investigation,sampling,collection of reagents,preliminary treatments to data processing.Selection principles of subjects were ensured before the rgular survey.The work of questionnaire,physical examination and collecting samples should be well controlled during the whole process.We took the measures of repeated measurements and catering to the reference ranges of technical control standards(SeronormTM)to meet the requirements of quality control in detecting serum metals.During the determination process,standard recovery rates of reagent blank and sample blank should be measured every time before and after detecting,meet the demand of 70%-120%.Serum samples whose metal levels were less than LOQ ought to be formulate as LOQ/2.The laboratory indicators were determined within the acceptable range of laboratory variation.A new-standard-curve was operated every 40 samples as determination of serum metal levels.All reaction reagents were carefully to control against pollution.6.Statistical AnalysisA database was set up for the experiment data.The research data was scientifically processed.Analyses were operated using the statistical analysis software SAS 9.1.3 and SPSS22.0,the average level was presented as mean and standard eviation(x±s)or median and inter-quartile range.Univariate analysis of variance and Student's t-tests were used for the analysis of continuous data.Qualitative data were evaluated using Pearson ?2 contingency tables.Crude and adjusted odds ratios(ORs)were determined with 95%confidence intervals(95%CIs)by multivariate logistic regression analysis.P<0.05 was centified as statistically significant.Results1.The methods for determining serum metals by ICP-MS were set up.The results indicated that this method could conform to the demand of serum metals determination.The correlation coefficient r of each metal element were better than 0.999.The low,medium and high within-day precisions were below 10%and the between-day precisions were better than 20%.Standard addition recovery rates of each element were all between 70%-120%.2.There were 1551(49.81%)of cases and 1563(50.19%)of controls in our subject,and there were more female than male.The average age of patients and control one were 65.25±9.88 years and 65.54±10.17 years,separately.We found no significant difference between cases and controls in gender or age.Whereas BMI and waist circumference in patients were higher than that in control ones.Besides that,the cases have more persons with hypertension,hyperlipidemia,family history with diabetes mellitus and low education level,when comparing to non-diabetes ones.Unfortunately,the blood glucose level of cases were nearly triple of that in controls(8.15 mmol/L VS 5.38 mmol/L),whose difference was statistically significant(P<0.001).3.The percentages of smoking and drinking ones in diabetes group were higher than control ones,and the labour-intensity of cases who would like to choose fried and salted foods daily was seemly ligher than non-diabetes who inversely choose milk and vegetables as usual dinner.4.When referring to metal conversations between cases and controls,the levels of V,Cr,Mn,Fe,Cu,Zn,As,Se,Rb,Sr,Ru,Pd,Ag,Cd,Cs,Ba,La,Ce,Nd,Ir and Tl were stronger than that in controls,whereas the concentrations of Ni,Os and Th were lower than control ones.Unfortunately,no significant difference had been found in Co,Ga,Rh,Pr,Sm,Eu,Gd,Dy,Ho,Er,Tm,Yb,Lu,Re,Pt,Au,Pb and U.5.PCA(principal component analysis)showed that there were 8 types of principal components generated with the value of characteristic root greater than 1,the accumulate of variance devoting rates reached 69.065%.We could find out thirty metal elements(V,Cr,Mn,Fe,Co,Cu,Zn,As,Se,Rb,Sr,Ag,Cd,Cs,Ba,La,Ce,Pr,Nd,Sm,Eu,Dy,Ho,Er,Tm,Yb,Re,Os,Ir,Tl)as the candidate ones for following logistic analysis.6.The middle or high concentrations of metals V,Cr,Mn,Cu,Zn,As,Se,Rb,Sr,Cs,Ba,Nd,Tl,Fe,Cd and Ag were likely to be more susceptible to catch diabetes,when comparing to the owest group,while the levels of V,Cr,Mn,Cu,Zn,Se,Rb,Sr,Cs,Ba,Nd and Tl had obvious dose-response relationships with the risk of type 2 diabetes.For some metal elements,it is worth to mention that the risk to get diabetes of high concentration was 2-4 times as the lowest group,the OR values and 95%CIs of Mn,Cu,Cs,Ba,Tl were 1.84(1.45-2.33)vs 5.04(3.87-6.57)?2.67(2.10-3.40)vs 4.98(3.76-6.59)?1.52(1.19-1.95)vs 3.81(2.96-4.91)?1.79(1.40-2.29)vs 4.77(3.66-6.21)?2.42(1.87-3.13)vs 8.86(6.71-11.71),separately.The high concentrations of elements Co,Sm,Eu,Ho,Tm had less risk to suspect diabetes.However,no significance had been showed when concerning La,Ce,Pr,Gd,Dy,Er,Yb,Re,Os and Ir.7.Through the stratified analysis,there were nineteen metal elements whose highest levels had stronger risk getting diabetes than lowest group in men,but not in women.The highest concentrations of Eu and Tm had lower risk than reference both in male and female,while there had no significant difference in metal element Ag.Conclusions1.We have found a set of methods which was practical and efficient aimed at blood samples,serum can be directly analyzed after dilution,simplifying the treatment steps,shortening the time of the whole sample analysis and effectively avoiding pollution brought by complicated sample preparation.The methods detecting serum 42 kinds of elements by ICP-MS were tested in the area of accuracy,relative standard deviation(RSD)and limit of detection(LOD).The results showed that this method could meet the demands of elements determination for biological serum.The method is accurate,simple and high sensitivity,it requires less pre-treatment test doses and sample requirements,reducing pollution and preventing volatile component.A variety of elements could be analyzed simultaneously,so the analysis efficiency was high.The method could be well applied in research detection.2.We found that risk of diabetes mellitus were associated with obese,large waistline,poor education level,hypertension,hyperlipidemia,family history of type 2 diabetes mellitus,smoke,drink,light physical work,interesting in fried and salted foods.3.As the concentrations of V,Cr,Mn,Cu,Zn,As,Se,Rb,Sr,Cs,Ba,Nd,Tl,Fe,Cd and Ag grew,the risk of suspecting diabetes increased,suggesting that these metals exposure may be related to type 2 diabetes mellitus in the general population.However,the metals La,Ce,Pr,.Gd,Dy,Er,Yb,Re,Os and Ir may be not associated with type 2 diabetes mellitus.4.In the highest level group of metals V,Cr,Mn,Fe,Co,Cu,Zn,As,Se,Rb,Sr,Cd,Cs,Ba,Nd,Sm,Ho and Tl,male would be more susceptible to type 2 diabetes than female.