硒 Selenium × 免疫功能（一般免疫支持／NK／T 細胞活性／病毒感染抵抗）

結論：Authoritative Lancet review by Rayman concluding low selenium status is associated with poor immune function, increased mortality risk, and cognitive decline; higher status or supplementation has antiviral effects in deficient hosts. Critically describes a U-shaped relationship: those already replete may be harmed by additional supplementation, with type-2 diabetes risk signal at high intake. Selenoproteins (notably GPx, TrxR) underpin antioxidant and anti-inflammatory functions central to immune cell signaling.

結論：Avery & Hoffmann (Nutrients, 2018) review of 25 human selenoproteins and their immune functions. Adequate selenium intake supports T-cell proliferation, antibody response, and macrophage/NK cell activity; deficiency impairs these and increases susceptibility to viral and bacterial infection. Authors emphasize that 'too much' may also dysregulate immunity, and that supplementation benefits are most evident in deficient or marginal-status individuals; in selenium-replete adults the marginal benefit is unclear.

結論：Beck/Levander/Handy summary of the foundational Keshan-region mechanistic work: selenium-deficient or GPx1-knockout mice infected with normally benign coxsackievirus B3 develop myocarditis, and the virus acquires multiple genome mutations converting it to a virulent phenotype. Same effect demonstrated for influenza A. Establishes that host micronutrient deficiency (oxidative stress) can drive viral evolution toward greater virulence — one of the strongest mechanistic links between selenium status and population-level viral immunity. Does NOT address whether selenium-replete humans gain further benefit from supplementation.

結論：Cochrane review (Visser, Durao, Sinclair, Irlam, Siegfried, 2017) of 33 RCTs in 10,325 HIV-positive adults (excluding pregnancy). Single or dual micronutrient supplements (including selenium-only arms) were not adequately powered to assess mortality or morbidity outcomes. No clinically significant changes in CD4 count or viral load reported with selenium alone. One trial (Burbano 2002) suggested possible reduction in hospital admissions. Authors conclude routine micronutrient supplementation has 'little or no effect on mortality' but supplementation should not be denied where specific deficiencies exist.

結論：Kamwesiga et al. multicenter double-blind placebo-controlled RCT in 300 ART-naïve HIV-positive Rwandan adults (67% women, median age 33.5). Rate of CD4 depletion reduced 43.8% in selenium arm (3.97 → 2.23 cells/μL/month). HOWEVER: no treatment effect for viral suppression (OR 1.18, 95% CI 0.71-1.94) or for the composite ART-initiation outcome (HR 1.00, 95% CI 0.66-1.54). Trial underpowered for composite due to lower-than-expected event rate. Demonstrates a measurable surrogate-marker effect on CD4 trajectory without translation to clinically meaningful viral or progression endpoints.

結論：Kupka et al. (Fawzi group, AJCN 2008) RCT in 913 HIV-positive Tanzanian pregnant women. Selenium had NO significant effect on maternal CD4 count or viral load. Marginal reduction in low birth weight risk but signal of INCREASED fetal-death risk (harm signal). Reduced child mortality after 6 weeks (RR 0.43) but overall neonatal mortality unaffected; maternal mortality unaffected. Mixed outcome profile in this high-deficiency-risk population.

結論：Hawkes/Kelley/Taylor controlled-feeding study in 11 healthy North American men. Cytotoxic T-lymphocytes and activated T-cells trended higher in the high-selenium group (only approaching significance). Earlier lymphocyte proliferation response to pokeweed mitogen on day 45 (high-Se) vs day 100 (low-Se). Diphtheria booster antibody titer 2.5-fold higher in high-Se group. NO difference in primary influenza-vaccine response; serum immunoglobulins and complement unchanged. Small n=11 and surrogate immune markers; cannot infer clinical infection-rate benefit in healthy, selenium-replete adults.