We selected the 2017-2018NHANES database to participate in this study, of which 1810 participants were excluded because they did not have HDL results, then 1034 participants were excluded because they did have basic biochemical test results. Finally, 4499 participants were excluded because they did not have insulin test results or BMI information. A total of 1,902 participants were enrolled in the study.(Figure 1)
Of the 1,902 eligible subjects, we divided them into two groups based on whether they developed insulin resistance. Four hundred fifty subjects with an IR value greater than 4.4 were assigned to the exposed group, and the remaining 1,452 subjects were assigned to the unexposed group. The age of the exposed group and the unexposed group were 53.19±16.80 and 50.42±17.79, respectively, and the difference in age distribution between the two groups was statistically significant (P < 0.001).BMI of the exposed group was higher than that of the unexposed group, and there was statistical significance (P<0.001). There was statistical significance in the race, smoking, waist circumference, hypertension, diabetes, and biochemical indexes such as ALT, ALP, BUN and other conventional biochemical indexes between the two groups (P<0.001). There was no statistical significance in annual income, marriage or gender (P<0.005).(Table 1)
It showed the relationship between insulin resistance and sex, smoking, education and diabetes mellitus and hypertension. Using men as the baseline, the risk of insulin resistance in women was reduced by 15%, P= 0.1246, because P>0.05 was not statistically significant. Based on recent non-smoking, the risk of recent smoking and recent non-smoking increased by 112% and 43%, respectively, P<0.0001, which was statistically significant. Participants with higher education had a 47% lower risk of developing insulin resistance than those without higher education (P<0.0001), which was statistically significant. The risk of insulin resistance without hypertension and diabetes was reduced by 52% and 79%, respectively, with statistical significance (P<0.0001).(table 2)
In the study, a total of 450 participants (40%) developed insulin resistance. A multivariate logistic regression model was used to analyze the association between TG/HDL-C and insulin resistance risk. After adjusting for age, gender, BMI, waist circumference, smoking and other factors, the probability of insulin resistance increased with the increase of TG/HDL ratio (OR: 1.91, 95%CI: 1.59-2.30, P<0.0001). At the same time, we conducted a smoothing curve fitting. We found a non-linear relationship between TG/HDL and IR Even when the adjusted covariable was deleted from the model, and the correlation remained unchanged. The inflexion point of TG/HDL-C smoothing curve was 0.95. When TG/HDL-C < 0.95, the effect value was very significant (OR: 27.34, 95%CI, 10.61-70.47, P<0.0001), and when TG/HDL-C≥0.95, the effect value was relatively lower (OR: 1.29, 95%CI, 1.03-1.61, P<0.0001). We grouped them by inflexion points, and multiple regression equations showed a two-fold increase in the risk of developing insulin resistance after adjustment for groups greater than 0.95 compared with groups less than 0.95
At the same time, we conducted a smoothing curve fitting. We found that there was a non-linear relationship between TG/HDL and IR. Even when the adjusted covariable was deleted from the model, the correlation remained unchanged. The inflexion point of TG/HDL-C smoothing curve was 0.95. When TG/HDL-C < 0.95, the effect value was very significant (OR: 27.34, 95%CI, 10.61-70.47, P<0.0001), and when TG/HDL-C≥0.95, the effect value was relatively lower (OR: 1.29, 95%CI, 1.03-1.61, P<0.0001). We grouped them by inflexion points. Multiple regression equations showed a two-fold increase in the risk of developing insulin resistance after adjustment for groups greater than 0.95 compared with groups less than 0.95.(figure 2,table 4)
Figure 3 showed associations between triglyceride/HDL cholesterol ratio and race, waist circumference, high blood pressure, BMI, diabetes, and ALT. In the non-Hispanic Black population, the increase of TG/HDL ratio had the most significant effect on insulin resistance. In the middle and old age of 42-60, the probability of insulin resistance increased by 177% for each increase of TG/HDL ratio by one unit. When waist circumference ranged from 63.2 to 92.3, the effect value was 2.83, suggesting that the risk of developing insulin resistance increased by 183% with each increase of TG/HDL ratio by one unit. Regardless of which factor and which group, TG/HDL-C was significantly associated with the occurrence of insulin resistance.