Urate epidemiology

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Marc Weisskopf, Alberto Ascherio and colleagues at Harvard (2007); Michael Schwarzschild, Ascherio and colleagues at Massachusetts General Hospital and Harvard, with the Parkinson Study Group PRECEPT investigators (2008).

What happened

• 2007 (risk). Weisskopf et al., American Journal of Epidemiology. A nested case-control study inside the Health Professionals Follow-up Study, using blood drawn in 1993–95 from over 18,000 men. 84 men developed Parkinson's by 2000; each was matched to two controls. Average urate was 5.7 mg/dl in cases and 6.1 mg/dl in controls. The top urate quarter had a rate ratio of 0.43 against the bottom (95% CI 0.18–1.02; trend p = 0.017), and 0.17 when cases diagnosed within four years of the blood draw were excluded.
• 2008 (progression). Schwarzschild et al., Archives of Neurology. Blood from the PRECEPT trial (804 people with early Parkinson's, tested with a drug called CEP-1347; average follow-up 21 months) was analysed. The top fifth for urate reached the need for dopamine drugs at about half the rate of the bottom fifth (hazard ratio 0.51, 95% CI 0.37–0.72). Brain scans (striatal β-CIT uptake, in 399 people) also showed less loss with higher urate. The link was stronger in men (HR 0.39) than in women (HR 0.77, not significant). The authors said it needed further testing before clinical use.

Why it moved us toward a cure

Urate is an antioxidant, and the link gave a cheap, testable idea: raise urate, protect neurons. It is the missing parent of the inosine trial (landmark 66), which raised urate in 298 people and did not slow decline, and of the genetic test in landmark 89, which found no causal protection. Both feed oxidative-stress's "pruned" status.

Outcome

Dead end as a treatment lead (as with smoking and coffee, landmark 82). The association is real in the data, but trials and genetics suggest it is not cause and effect. People with early, more severe disease may simply have lower urate for other reasons (for example diet, body size or illness). Dead end for the treatment idea; see dead ends.

Sources

• Weisskopf et al., "Plasma urate and risk of Parkinson's disease," Am J Epidemiol 2007;166(5):561–567, PMID 17584757 · ScienceDaily summary
• Schwarzschild et al., Arch Neurol 2008, PubMed 18413464, NCT00040404 · Harvard Gazette · Alzforum
• Correction: an earlier version cited DOI 10.1093/aje/kwm385 for Weisskopf 2007; that DOI is a different paper (Gao et al., diet and urate, 2008). The Weisskopf DOI was not found, so the PMID is given.
• Not yet verified: the Schwarzschild volume and pages. Note the 2007 CI (0.18–1.02) just crosses 1.0 for the main comparison; the 0.17 figure is the cases-excluded analysis.