2.6. Statistical analysisDistributi

2.6. Statistical analysisDistribution of continuous data was analyzed using ShapiroWilks, parametric and nonparametric tests were also appliedconsequently. The difference of positive wells between test groupsand positive controls was calculated with Gilbert Z2 test. Differences in the biomarkers between groups were calculated withWilcoxon-Mann-Whitney test for continuous variable and the Chisquare test for nominal data. The correlation between biomarkersand exposure time was evaluated by Pearson correlation. Statisticalsignificant level was set at P < 0.05. Data were analyzed using SPSSversion 19.0 (SPSS Inc.).3. Result3.1. Demographic characteristics of study subjectsThe demographic characteristics of hospital workers exposed toADs and controls are listed in Table 1. No significant differences inthe age, gender, BMI, percentages of smokers and consumption ofsmoked or fried foods were observed between the two groups.3.2. Evaluation of biomarkers in exposed workersThe urinary 8-OHdG/Cr concentrations in 158 workers occupationally exposed to ADs was 22.05 ± 17.89 ng/mg Cr, which wassignificant higher than the levels observed in a control population(17.36 ± 13.50 ng/mg Cr (P = 0.014)) (Fig. 1).The frequencies of urinary positive mutagenic in exposed andcontrol group at different dose of urine concentrates is shown inTable 2. The mutagenic activity was significantly higher in exposedsubjects as compared to the control group (P < 0.05). As for TA 100plus or minus S9mix, no statistically significant difference in themutagenic activity was observed in exposed subjects as comparedto controls. Notably, the mutagenic activity positively correlatedwith the dose of urine concentrates. No statistically significantdifferences in early lymphocyte apoptosis were found betweenexposed and control group (Fig. 2).