The Bidirectional Relationship Between Chronic Kidney Disease And Hyperuricemia:A Prospective Cohort Study

Objective This study aims to investigate the association between chronic kidney disease（CKD）,estimated glomerular filtration rate（eGFR）at the baseline,and the dynamic changes of eGFR with the risk of hyperuricemia（HUA）,and the association between HUA,serum uric acid（sUA）at the baseline,and the dynamic changes of sUA with the risk of CKD in the Jinchang cohort.In addition,this study also aims to demonstrate the causal association between eGFR and sUA and to determine the direction of the causal association,thus providing a scientific basis for the prevention and control of CKD and HUA as well as for controlling sUA and improving eGFR levels.Methods Based on data from the Jinchang cohort at the baseline and the first-round follow-up,this study was conducted in three parts,with the first and second parts of the study followed to the end of HUA（N=25433）and CKD（N=28422）,respectively.The Cox regression model was used to clarify the relationship between the CKD and eGFR at the baseline with the incidence risk of HUA,as well as the relationship between the HUA and sUA at the baseline with the incidence risk of CKD,and the HRs value and 95%CIs were calculated.The multiplicative and additive interactions between CKD and other associated factors on the risk of HUA and between HUA and other associated factors on the risk of CKD were explored using SAS macros,respectively.The dose-response relationships between eGFR at the baseline with the risk of HUA and between sUA at the baseline with the risk of CKD were then analyzed by restricted cubic splines.The third part of the study（N=31028）applied a cross-lagged model to demonstrate the causal association between sUA and eGFR and to determine the direction of the causal association.Results 1.Among 25433 participants free of HUA at the baseline,1597new-onset HUA were observed during 56698.43 person-years of follow-up,with a incidence rate of 6.28%,and an incidence density of 28.17/1000 person-years.2.There was an association between CKD and the risk of HUA,the risk of HUA in CKD population was 1.58 times higher than that in non-CKD population（HR=1.58,95%CI:1.28-1.95）.This association was more evident among participants who were between 45 and 64 years old,65 years old and above,female,ex-smokers,ex-drinkers,hypertension,workers,and had a BMI<24.0 kg·m^-2.The interaction results showed a positive multiplicative interaction between CKD and age on the risk of HUA.Besides,There was a positive multiplicative and additive interaction between CKD and female,hypertension on the risk of HUA.3.There were U-shaped relationships between eGFR at the baseline with incident HUA(P_overall<0.001,P_nonlinear<0.001).The adjusted HRs（95%CIs）of HUA were 3.56（2.50-5.05）for the participants in the group of eGFR less than 60m L·min^-1·1.73 m^-2,1.61（1.42-1.83）for those in the group of eGFR between 60 and90 m L·min^-1·1.73 m^-2,and 1.74（1.42-2.14）for those in the group of eGFR more than 120 m L·min^-1·1.73 m^-2,compared with the group of eGFR between 90 and 120m L·min^-1·1.73 m^-2.Importantly,a remarkably higher risk of HUA was observed in participants with a sustained reduction in eGFR.Compared with the group of N-N,the adjusted HR（95%CI）of HUA was 3.12,（2.36-4.13）for the participants in the group of N-R,3.90（2.63-5.77）for the participants in the group of R-R,respectively.4.Among 28422 participants free of CKD at the baseline,1212 new-onset CKD were observed during 63574.32 person-years of follow-up,with a incidence rate of4.26%,and an incidence density of 19.06/1000 person-years.5.There was an association between HUA and the risk of CKD,the risk of CKD in HUA population was 1.28 times higher than that in non-HUA population（HR=1.28,95%CI:1.12-1.47）.This association was more robust among participants who were over 65 years old,female,with BMI<24.0 kg·m^-2,ex-smokers,workers,non-diabetes and hypertension.The interaction results showed that there was a positive multiplicative and additive interaction between HUA and hypertension on the risk of CKD.6.There was a positive linear dose-response relationship between sUA at the baseline and the risk of CKD(P_overall<0.05,P_nonlinear>0.05).Compared with the first quintile of sUA,the adjusted HR（95%CI）of CKD was 1.24（1.01-1.51）for the individuals in the fourth quantile In addition,a notably higher risk of CKD was observed in participants with persistently elevated sUA.Compared with the group of N-N,the adjusted HR（95%CI）of CKD was 1.80,（1.50-2.16）for the participants in the group of N-H,1.83（1.54-2.16）for the participants in the group of H-H,respectively.7.The results of the cross-lagged model showed that eGFR at the baseline predicted sUA at the follow-up（ρ₁=-0.024,P<0.001）,while sUA at the baseline also predicted eGFR at the follow-up（ρ₂=-0.015,P<0.01）in the total population.The results of subgroup analysis showed that there was a bidirectional relationship in both male and female populations,in those aged 45-64 years,and in workers and other occupations.Conclusions 1.CKD was a risk factor for the risk of HUA,and there was a synergistic effect with age,sex,and hypertension on the development of HUA.2.There was a U-shaped dose-response relationship between eGFR and the risk of HUA,and the dynamic decrease of eGFR was related to the increase of the incidence risk of HUA.3.HUA was a risk factor for the risk of CKD,and had a synergistic effect with hypertension on the development of CKD.4.There was a positive linear dose-response relationship between sUA with the incidence of CKD,and an association between dynamically elevated sUA with the increased risk of CKD.5.There was a bidirectional association between eGFR and sUA,with eGFR at the baseline negatively predicting sUA at the follow-up and sUA at the baseline negatively predicting eGFR at the follow-up.