A systematic review and meta-analysis and mendelian randomization analysis of serum phosphorus, albumin, and CRP as risk factors for death in hemodialysis patients.

BACKGROUND

We aimed to identify the main mortality risk factors in hemodialysis patients using data from relevant cross-sectional literature. We used Mendelian randomization (MR) to assess the causality of those identified risk factors using pertinent Genome-Wide Association Study (GWAS) Single Nucleotide Polymorphism (SNP) data.

METHODS

Sixteen publications detailing hemodialysis-related mortality implicated 32 death-related risk factors. Based on heterogeneity testing, we utilized randomand stratified-effects models for meta-analysis. Sensitivity analysis and bias testing were used to evaluate data reliability. Mr analysis identified type-2 diabetes (T2D), serum phosphorus, serum albumin, and age as risk factors, with hematology as the outcome. Inverse-variance weighting (IVW) analysis was used in the main study. The consistency of the IVW analysis results was evaluated simultaneously using four different methods: Mr Egger regression, weighted median estimator (WME), weighted mode, and simple mode. Horizontal pleiotropy was assessed using the Mr Egger regression intercept term; heterogeneity was evaluated using Cochran's Q.

RESULTS

Using randomand stratified-effects models, a meta-analysis of 16 published articles revealed that the following factors were associated with a greater mortality risk in hemodialysis patients: T2D; serum phosphorus, albumin, and CRP; and the Charlson comorbidity index (CCI). The results were deemed reliable based on bias (P=0.1186, I2=99.53%) and sensitivity (T=0.39, df=116, P=0.6953) analyses. IVW indicated a genetic-level positive causal relationship between T2D and hematology (OR=1.2572, 95% CI=1.0375-1.5235; P=0.0195). Genetic-level serum pathology and hematology were positively correlated (OR=2.0269, 95% CI=1.0614-3.8708; P=0.0323). However, age (OR=11.1112, 95% CI=0.83) was a factor. No discernible genetic causal relationship occurred between hematology and serum ferritin (OR=0.6707, 95% CI=0.4612-0.9707; P=0.4612) or albumin (OR=1.2933, 95% CI=0.8931-1.8729; P=0.1733).

CONCLUSIONS

Meta-analysis identified the number of dialysis sessions, serum CRP, CCI, T2D, serum phosphorus, and serum albumin as mortality risk factors in hemodialysis patients. The Mr results showed positive causal relationships between T2D incidence and serum phosphorus with hemodialysis risk.