补充硒对成人血糖指标和氧化应激生物标志物的影响：随机对照试验的系统综述、荟萃分析和剂量-反应分析。
完整摘要 · 中文译文
背景：糖尿病和代谢综合征是与氧化应激、炎症和胰岛素信号受损密切相关的重大公共卫生挑战。硒是一种具有抗氧化和免疫调节作用的必需微量元素，已被提出作为代谢调节的辅助策略；然而，随机试验的证据仍不一致。我们评估了补充硒对成人血糖指标、氧化应激相关生物标志物、脂质代谢及相关代谢指标的影响，并探讨了潜在的效果修饰因素。 方法：我们检索了PubMed、Embase、Cochrane CENTRAL、Web of Science、CINAHL、CNKI、万方、维普和中国生物医学文献数据库截至2025年8月12日的文献，纳入在成人（包括妊娠期糖尿病（GDM）孕妇）中口服补充硒的随机对照试验。结局包括空腹血糖、糖化血红蛋白、空腹胰岛素、胰岛素抵抗和β细胞功能指标、氧化应激相关和炎症生物标志物、脂质参数以及人体测量学和血流动力学指标。随机效应荟萃分析估计了加权均数差及95%置信区间。进行了亚组、剂量-反应和敏感性分析，并使用GRADE评估证据确定性。 结果：共纳入52个治疗组，涉及10，916名参与者。补充硒降低了空腹胰岛素、HOMA-IR和HOMA-β值并提高了QUICKI，但对空腹血糖或糖化血红蛋白未见明确的总体效应。有利 在选定的抗氧化和炎症生物标志物中观察到可测量的变化，包括谷胱甘肽、总抗氧化能力、超氧化物歧化酶、谷胱甘肽过氧化物酶和血清硒升高，以及丙二醛和高敏C反应蛋白降低，对一氧化氮无明显影响。血脂参数也观察到适度变化，包括高密度脂蛋白胆固醇升高和低密度脂蛋白胆固醇、甘油三酯及总胆固醇降低。在人体测量学和血流动力学结局中，仅舒张压降低。效应因剂量、持续时间、参与者特征和地理硒暴露背景而异。 结论：补硒与胰岛素敏感性标志物、选定抗氧化和炎症生物标志物及血脂参数的适度改善相关，而对空腹血糖或糖化血红蛋白未见明确的总体效应。这些发现可能最适用于基线硒状态边缘性或偏低的人群，向硒充足环境外推时应谨慎。需要长期、高质量随机对照试验（RCT）来阐明基线状态特异性效应、安全剂量范围和长期安全性。 试验注册：PROSPERO CRD420251124852。
完整摘要 · 英文原文
BACKGROUND: Diabetes and metabolic syndrome are major public health challenges closely linked to oxidative stress, inflammation, and impaired insulin signaling. Selenium, an essential trace element with antioxidant and immunomodulatory roles, has been proposed as an adjunctive strategy for metabolic regulation; however, evidence from randomized trials remains inconsistent. We evaluated the effects of selenium supplementation on glycemic indices, oxidative-stress-related biomarkers, lipid metabolism, and related metabolic indicators in adults, and explored potential effect modifiers. METHODS: We searched PubMed, Embase, Cochrane CENTRAL, Web of Science, CINAHL, CNKI, WanFang, VIP, and SinoMed through 12 August 2025 for randomized controlled trials of oral selenium supplementation in adults, including pregnant women with gestational diabetes mellitus (GDM). Outcomes included fasting plasma glucose, glycated hemoglobin, fasting insulin, indices of insulin resistance and β-cell function, oxidative-stress-related and inflammatory biomarkers, lipid parameters, and anthropometric and hemodynamic indicators. Random-effects meta-analyses estimated weighted mean differences with 95% confidence intervals. Subgroup, dose-response, and sensitivity analyses were performed, and certainty of evidence was assessed using GRADE. RESULTS: Fifty-two treatment arms involving 10,916 participants were included. Selenium supplementation reduced fasting insulin, HOMA-IR, and HOMA-β values and increased QUICKI, but showed no clear overall effect on fasting plasma glucose or glycated hemoglobin. Favorable changes were observed in selected antioxidant and inflammatory biomarkers, including increases in glutathione, total antioxidant capacity, superoxide dismutase, glutathione peroxidase, and serum selenium, and reductions in malondialdehyde and high-sensitivity C-reactive protein, with no clear effect on nitric oxide. Modest changes in lipid parameters were also observed, including higher high-density lipoprotein cholesterol and lower low-density lipoprotein cholesterol, triglycerides, and total cholesterol. Among anthropometric and hemodynamic outcomes, only diastolic blood pressure was reduced. Effects varied by dose, duration, participant characteristics, and geographical selenium exposure context. CONCLUSIONS: Selenium supplementation was associated with modest improvements in insulin-sensitivity markers, selected antioxidantand inflammatory biomarkers, and lipid parameters, while showing no clear overall effect on fasting plasma glucose or glycated hemoglobin. These findings may be most applicable to populations with marginal or low baseline selenium status and should be extrapolated cautiously to selenium-replete settings. Long-term, high-quality RCTs are needed to clarify baseline-status-specific effects, safe dose ranges, and long-term safety. TRIAL REGISTRATION: PROSPERO CRD420251124852.