| Title: | Kidney Function Measures and Mortality: A Mendelian Randomization Study |
| Journal: | American Journal of Kidney Diseases |
| Published: | 25 Dec 2023 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/38151225/ |
| DOI: | https://doi.org/10.1053/j.ajkd.2023.10.014 |
| URL: | http://www.ajkd.org/article/S0272638623009940/pdf |
Publication 9074
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
RATIONALE & OBJECTIVE: Individuals with a low estimated glomerular filtration rate (eGFR) are at a high risk of death. However, the causes underpinning this association are largely uncertain. This study aimed to assess the causal relationship of low eGFR with all-cause and cause-specific mortality.</p>
STUDY DESIGN: Retrospective cohort study incorporating Mendelian randomization (MR).</p>
SETTING & PARTICIPANTS: Individual-level data from 436,214 White participants (54.3% female; aged 56.8±8.0 years) included in the UK Biobank.</p>
EXPOSURES: eGFR estimated using cystatin C (eGFRcyst).</p>
OUTCOMES: The outcomes of interest included all-cause mortality, cardiovascular mortality, cancer mortality, infection mortality, and other-cause mortality.</p>
ANALYTICAL APPROACH: Cox proportional hazards analysis for the conventional observational analyses; linear and nonlinear MR analyses implemented using genetic allele scores as instrumental variables representing kidney function to estimate the effect of kidney function on the survival outcomes.</p>
RESULTS: During a median follow-up of 12.1 years, there were 30,489 deaths, 6,098 of which were attributed to cardiovascular events, 15,538 to cancer, 1,516 to infection, and 7,227 to other events. In the conventional observational analysis, eGFRcyst exhibited a nonlinear association with all the outcomes. MR analysis suggested that a genetically predicted lower eGFRcyst was linearly associated with a higher rate of cardiovascular mortality (HR, 1.43; 95% CI, 1.18-1.75) across the entire measurement range (every 10-mL/min/1.73m2 decrement). Nonetheless, no causal associations between eGFRcyst and all-cause mortality (HR, 1.07; 95% CI, 0.98-1.17) or any types of noncardiovascular mortality were detected.</p>
LIMITATIONS: Potential misclassification of the actual cause of death, a nonrepresentative sample, and potential error in the interpretation of the magnitude of associations generated in MR analyses.</p>
CONCLUSIONS: These findings suggest a potential causal association between low eGFR and cardiovascular mortality in the general population, but no causal relationship with all-cause mortality or noncardiovascular mortality was observed. Further studies in other populations are warranted to confirm these findings.</p>
PLAIN-LANGUAGE SUMMARY: This study investigated the existence of a causal relationship between lower kidney function and death of different causes. Using data from 436,214 people in the United Kingdom, we applied conventional statistical analyses and those incorporating genetic data to implement Mendelian randomization, an approach that estimates causal associations. The observational analysis showed a nonlinear association between kidney function and various types of mortality outcomes. However, Mendelian randomization analysis suggested a linear increase in the risk of cardiovascular mortality with lower kidney function, but no causal link between the level of kidney function and all-cause or noncardiovascular mortality was identified. Managing kidney health may help reduce cardiovascular mortality, but caution is needed in interpreting the magnitudes of these results. Further validation in other populations and in those with advanced kidney failure is needed.</p>
14 Authors
- Ying Shan
- Jingwen Zhang
- Yueqi Lu
- Jinlan Liao
- Yuyang Liu
- Liang Dai
- Jing Li
- Congying Song
- Guobin Su
- Sara Hägg
- Zuying Xiong
- Dorothea Nitsch
- Juan Jesus Carrero
- Xiaoyan Huang
1 Application
| Application ID | Title |
|---|---|
| 65814 | Leveraging functional genomics resources to improve the power of phenotype prediction -- a new machine learning approach |
Enabling scientific discoveries that improve human health