Does high blood uric acid actually cause kidney disease?
People whose genes give them lifelong higher uric acid do not have worse kidney function, suggesting the familiar link between uric acid and kidney disease is not causal.
Source
No causal effects of serum urate levels on the risk of chronic kidney disease: A Mendelian randomization study
Study at a glance
- Design
- Mendelian randomisation — Two-sample MR using 26 urate-raising genetic variants as instruments against published GWAS summary statistics for eGFR and CKD, analysed with 7 pleiotropy-robust methods; replicated in UK Biobank and by a genetic risk score in 4 US population cohorts, with gout as a positive control.
- N
- No single N: the urate GWAS had 110,347 people; CKD summary data had 12,385 cases and 104,780 controls; UK Biobank added 335,212 people (5,615 CKD cases); the individual-level cohorts totalled 13,425.
- Population
- Adults of European ancestry in genetic consortia, UK Biobank, and the ARIC, CARDIA, CHS and Framingham cohorts.
- Outcome
- Estimated glomerular filtration rate (eGFR) and chronic kidney disease (eGFR below 60 or clinical diagnosis codes); gout as a positive-control outcome.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers used Mendelian randomization, treating 26 genetic variants that raise serum urate as a natural experiment assigned at conception. They tested whether these variants predicted kidney function (eGFR) or chronic kidney disease in large genome-wide datasets, using 7 different methods designed to cope with variants that might affect the kidney through other pathways. They repeated the test in UK Biobank using hospital diagnosis codes, and in 4 US population cohorts using each person's genetic risk score, and checked that the same variants did predict gout, which urate is known to cause.
What they found
None of the 7 methods found a causal effect of urate on eGFR or on CKD, and the UK Biobank and cohort analyses were also null (genetic risk score and CKD: odds ratio 1.05, not significant). The same variants clearly raised gout risk, showing the instruments worked. In contrast, ordinary observational analysis of the same cohorts found that each 1 mg/dl of urate went with 1.48 times the odds of CKD. The study had over 99% power to detect an effect as large as the observational one.
The limits
What it doesn't show
Mendelian randomization estimates the effect of lifelong, genetically raised urate, which may not match the effect of lowering urate with a drug later in life, and it cannot rule out benefits of xanthine oxidase inhibitors that act through reduced oxidative stress rather than lower urate. All participants were of European ancestry, so the result may not generalise to other populations. The genetic instruments showed substantial pleiotropy, and some GWAS participants may have been on urate-lowering drugs; the authors addressed both with robust methods and sensitivity analyses, but these remain assumptions rather than proofs.
Key terms
- Mendelian randomization
- Using genetic variants that influence an exposure as a natural experiment, because alleles are randomly allocated at conception and are therefore largely free of confounding and reverse causation.
- Instrumental variable
- A variable (here a genetic variant) that affects the outcome only through the exposure of interest, allowing a causal estimate.
- Horizontal pleiotropy
- When a genetic variant affects the outcome through a pathway other than the exposure, which can bias Mendelian randomization.
- Positive control
- An outcome known to be caused by the exposure (here gout), used to show the method can detect a real effect.
- eGFR
- Estimated glomerular filtration rate, a measure of kidney filtering capacity calculated from blood creatinine; CKD was defined as eGFR below 60.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What natural feature does Mendelian randomization rely on to reduce confounding?
Common questions
If observational studies show high urate goes with kidney disease, why isn't that enough?
Kidney disease itself reduces urate excretion, and shared factors like obesity or hypertension raise both, so the association could reflect reverse causation or confounding rather than urate damaging the kidney.
How do we know the study wasn't simply too small to find an effect?
The authors calculated more than 99% power to detect an effect as large as the observational association, and the same variants did detect urate's known effect on gout.
Does this mean allopurinol is useless for kidneys?
It suggests that lowering urate by itself is unlikely to prevent CKD, but it does not rule out other effects of these drugs, and randomized trials remain the definitive test.
More on Kidney disease