As a new class of insecticides, neonicotinoid insecticides are widely used around the world. Recent studies have found that neonicotinoid insecticides have endocrine disrupting effects which can affect the homeostasis of sex hormones, but there are few epidemiological studies on children and adolescents.

This study is conducted to explore the associations between neonicotinoid insecticides exposure and serum sex hormones among children and adolescents.

This study was based on the public online data of the National Health and Nutrition Examination Survey (NHANES) (2015-2016). A total of 599 children and adolescents at 6-20 years of age were finally enrolled, who had basic characteristic information and test data of six neonicotinoid insecticidesimidacloprid (IMI), acetamiprid (ACE), thiacloprid (THD), clothiandin (CLO), N-desmethylacetamiprid (N-DMA), and 5-hydroxy imidacloprid (5-OH-IMI) and three sex hormonesestradiol (E2), testosterone (T), and sex hormone binding globulin (SHBG). Urine neonicotinoid pesticides were measured using high performance liquid chromatography-tandem mass spectrometry. Serum T and E2 were detected using liquid chromatography-tandem mass spectrometry. Serum SHBG was tested using electrochemiluminescence immunoassay. Except a high detection rate of N-DMA, the detection rates of other neonicotinoid insecticides were low. Therefore, the study subjects were divided into three groups (Q1-Q3) according to the logarithmic value of N-DMA level. Generalized linear model was used to analyze the association between N-DMA exposure and sex hormones in all participants and different genders.

The detection rates of the 6 neonicotinoid insecticidesN-DMA (40.7%), 5-OH-IMI (17.9%), CLO (7.7%), IMI (4.3%), ACE (0.5%), and THD (0.3%) were all lower than 50%. Only the P₇₅ (0.39 μg·L^-1) and P₉₅ (1.15 μg·L^-1) of N-DMA concentration were higher than its detection limit (0.2μg·L^-1); only the P₉₅ (1.1 μg·L^-1) of 5-OH-IMI concentration was higher than its detection limit (0.4 μg·L^-1); only the P₉₅ (0.4 μg·L^-1) of CLO concentration was higher than its detection limit (0.2 μg·L^-1); the IMI, ACE, and THD quartiles were all lower than their detection limits (0.4 μg·L^-1, 0.3 μg·L^-1, and 0.03 μg·L^-1 respectively). There were seasonal differences in the concentrations of N-DMA, 5-OH-IMI, and CLO, namely higher in summer and autumn than in winter and spring (P < 0.05). The generalized linear model results showed that the N-DMA was negatively correlated with serum T (b=-0.12, 95% CI: -0.22﹣-0.02) in the Q3 group, while positively correlated with serum SHBG (b=0.05, 95% CI: 0-0.09; b=0.08, 95% CI: 0.04-0.13) in the Q2 and Q3 groups when taking the Q1 group as reference, and these two associations both had a dose-response trend (P_trend=0.023 and < 0.001). After gender stratification, the N-DMA in the Q3 group was negatively correlated with boys' serum T (b=-0.15, 95% CI: -0.29-0) and serum E2 (b=-0.07, 95% CI: -0.13-0), and both had a dose-response trend (P_trend=0.042 and 0.032). The N-DMA in the Q2 and Q3 groups was positively correlated with boys' serum SHBG (b=0.06, 95% CI: 0.01-0.11; b=0.07, 95%CI: 0.02-0.13), and there was a dose-response trend (P_trend=0.010). However, none of these associations were observed in girls.

The exposure level of neonicotinoid insecticides in American children and adolescents is relatively low, and the detection rates of metabolites are higher than those of the parents. Exposure to N-DMA, a metabolite of ACE, may be related to the decrease in T levels and the increase in SHBG levels in children and adolescents. Moreover, there are gender differences. These findings indicate that exposure to neonicotinoid insecticides may affect the levels of sex hormones in children and adolescents.