Background

The associations among dietary selenium intake, serum selenium concentration, plasma glucose and glycosylated hemoglobin levels, and diabetes risk remain controversial. This study aimed to evaluate these associations in adults from the United States.

Methods

We conducted a cross-sectional study of participants aged 18 years and older who participated in the National Health and Nutrition Examination Survey. Between 1999 and 2006, a total of 41,474 participants were initially included in this study. Multivariable linear or logistic regression analysis was used to investigate the association between dietary selenium intake and serum selenium concentrations, glucose level, and diabetes risk.

Results

The average age of the participants was 30.32 ± 23.95 years, and 48.72 % were men. Their mean dietary selenium intake and mean serum selenium concentration were 98 ± 55 μg per day and 129 ± 22 ng/mL, respectively. Compared with t he lowest quartile of dietary selenium intake, the highest quartile was associated with elevated plasma glucose levels (β = 2.412, 95 % confidence interval [CI]: 0.420, 4.403, P = 0.018), glycosylated hemoglobin levels (β = 0.080, 95 % CI: 0.041, 0.119, P < 0.001), and diabetes risk (odds ratio [OR] = 2.139, 95 % CI: 1.763, 2.596, P < 0.001). Higher serum selenium was also associated with increased plasma glucose levels (β = 12.454, 95 % CI: 4.122, 20.786, P = 0.003) and glycosylated hemoglobin levels (β = 0.326, 95 % CI: 0.187, 0.465, P < 0.001). A generalized additive model with a spline curve suggested a nonlinear relationship between dietary selenium intake, serum selenium and glucose levels, and diabetes risk.

Conclusions

Dietary selenium intake and serum selenium were positively associated with elevated levels of plasma glucose and glycosylated hemoglobin, and the relationships were nonlinear. Additional selenium supplementation for patients with diabetes may not be recommended.

中文翻译：

---

膳食和血清硒浓度与血糖水平和糖尿病风险的关联：一项横断面研究的全国健康和营养检查调查，1999-2006

背景

膳食硒摄入量、血清硒浓度、血浆葡萄糖和糖化血红蛋白水平与糖尿病风险之间的关联仍存在争议。本研究旨在评估美国成年人的这些关联。

方法

我们对参加全国健康和营养检查调查的 18 岁及以上的参与者进行了横断面研究。1999 年至 2006 年间，最初共有 41,474 名参与者参与了这项研究。多变量线性或逻辑回归分析用于研究膳食硒摄入量与血清硒浓度、血糖水平和糖尿病风险之间的关联。

结果

参与者的平均年龄为 30.32 ± 23.95 岁，48.72% 为男性。他们的平均膳食硒摄入量和平均血清硒浓度分别为每天 98 ± 55 μg 和 129 ± 22 ng/mL。与膳食硒摄入量的最低四分位数相比，最高四分位数与血浆葡萄糖水平升高（β = 2.412，95% 置信区间 [CI]：0.420、4.403、P = 0.018）、糖化血红蛋白水平（β = 0.080 , 95 % CI: 0.041, 0.119, P < 0.001) 和糖尿病风险（比值比 [OR] = 2.139, 95% CI: 1.763, 2.596, P < 0.001）。较高的血清硒还与血浆葡萄糖水平升高有关（β = 12.454，​​95 % CI：4.122、20.786，P= 0.003) 和糖化血红蛋白水平 (β = 0.326, 95 % CI: 0.187, 0.465, P < 0.001)。具有样条曲线的广义加性模型表明膳食硒摄入量、血清硒和葡萄糖水平以及糖尿病风险之间存在非线性关系。

结论

膳食硒摄入量和血清硒与血浆葡萄糖和糖化血红蛋白水平升高呈正相关，并且这种关系是非线性的。可能不建议为糖尿病患者补充硒。