Selenium & Kashin-Beck disease

Almost everything in this section is a hypothesis about whether a food might protect cartilage. This is the one place where a cartilage disease was caused by what people ate, and then largely removed by changing it. It is also the entry where the limits of that fact need stating most carefully.

What Kashin-Beck disease is

An endemic osteoarthropathy with necrosis of growth-plate and articular cartilage, striking in childhood and leaving permanent deformity and short stature. It occupies a long, narrow geographic band running through China, with historic pockets in Siberia and North Korea, and it tracks farming regions rather than cities. Multivariate analysis of the Chinese surveillance data puts its cause in the food supply: fungal contamination of stored grain, selenium deficiency, and imbalanced protein intake.

The population result

From the 1990s, preventive programmes reached more than 300 million residents of the endemic areas. National incidence fell in steps: 22.1% in 1990, 16.0% in 1995, 12.3% in 2000, 5.5% in 2005, 0.38% in 2010, and 0.18% in 2015. The disease has not gone — it still affected 574,925 people across a population of 22.6 million in 2016 — but the epidemiology of a cartilage disease was bent by a food intervention at national scale, which has happened exactly once.

The trials

Fifteen randomized controlled trials in 2,931 children were pooled in a network meta-analysis comparing five forms of selenium: selenium salt, sodium selenite, selenium-enriched yeast, and sodium selenite combined with vitamin E or vitamin C. All five produced more metaphyseal improvement on X-ray than placebo, with selenium salt ranked highest.

The authors then do something worth noticing: they downgrade their own primary outcomes from high to low or very low on GRADE, and state that the ranking may be misleading and the evidence insufficient to say which form is best. The direction is consistent; the confidence is not.

An earlier meta-analysis of eleven community trials reported larger numbers — primary prevention odds ratio 0.16 (95% CI 0.08–0.33), clinical improvement 6.57 (3.33–12.93), metaphyseal repair 5.53 (2.92–10.47). Those figures circulate widely. An Expression of Concern was published on that paper in November 2025, which is why they appear here with that attached rather than as support.

Why selenium would matter to cartilage

Selenium's biology runs through selenoproteins — principally the glutathione peroxidases and thioredoxin reductases that handle oxidative stress — and several of them are expressed in chondrocytes. That makes a deficiency disease of cartilage biologically coherent rather than a geographic coincidence, and it is the reason the association was pursued in the first place.

The other explanation

T-2 toxin, a trichothecene mycotoxin produced by fungi on damp stored grain, tracks with Kashin-Beck prevalence and progression in its own meta-analysis. Selenium-poor soils and mould-prone grain storage occur in the same isolated, poor, damp farming regions, and the preventive programmes changed both — along with drinking water, and in some places where people lived.

So the honest statement of the population result is that a package of food-related changes nearly eliminated a cartilage disease, and that the contribution of selenium within that package is supported by trials whose own authors rate the evidence low.

Who this applies to

Kashin-Beck disease is a deficiency disease of childhood, and the endpoint in every trial is repair of a metaphyseal lesion in a child who was short of selenium. Osteoarthritis in an adult with adequate selenium intake is a different disease with a different mechanism, and nothing in this literature suggests that raising selenium intake above sufficiency does anything for it. No cohort of selenium status against osteoarthritis in a replete population turned up in the search behind this entry, and no trial of selenium supplementation with an osteoarthritis endpoint exists outside the Kashin-Beck work.

Selenium also has an unusually narrow window between the recommended intake and the tolerable upper limit, and selenosis is a real toxicity. This entry is not an argument for taking any.

What would change this entry

Somebody looking. The Kashin-Beck literature has established that cartilage can be destroyed by a dietary deficiency and partly repaired by correcting it — which is a stronger proof of principle than anything else in this section offers. Whether selenium status varies enough within an ordinary population to matter for ordinary osteoarthritis is a question with no data attached to it at all, and it is answerable with a cohort that already has toenail or serum selenium in its freezer.