Fig 1.
Patient selection and classification.
Eligible patients with urinary stones and control subjects selected from the database. We excluded 629 patients without stone analysis, 42 patients without complete clinical data, and 46 struvite stones. The remaining 602 urinary stone patients included in the study. The subjects were divided into two groups according to UA levels: the UA-high group with hyperuricemia (UA ≥ 7.0 mg/dL) or the UA-low group with normal UA levels (UA < 7.0 mg/dL). The control subjects and stone patients were pair-matched according to age, sex, body mass index, comorbidities (HTN, DM and CVD), hyperlipidemia hemoglobin, serum Alb, and serum UA levels using propensity score matching.
Fig 2.
The relationship between serum UA level and eGFR.
The correlation between eGFR and serum UA levels were significant in Ctrl, CaOx/CaP, and UA stone patients. R2 values showed higher correlation in UA stone patients (R2 = 0.211, P < 0.001) than that in the Ctrl (R2 = 0.044, P < 0.001) or in the CaOx/CaP or (R2 = 0.070, P < 0.001) (A). The number of CaOx/CaP patients with stage 3 CKD was not significantly different between the UA-low and UA-high (B). The number of UA stone patients with stage 3 CKD was significantly higher in the patients with UA-high than with UA-low group (C).
Table 1.
Clinical characteristic of the population in the present study.
Fig 3.
The prevalence of renal impairment in pair-matched subjects stratified serum UA level.
In the UA-low group (serum UA < 7.0 mg/dL), eGFR was significantly lower in patients with UA stones compared with control subjects (A). The prevalence of stage 3 CKD was significantly greater in UA stone patients (48%) compared with control subjects (24%) (P = 0.005). In the UA-high group (serum UA ≥ 7.0 mg/dL), eGFR was significantly lower in patients with UA stones compared with control subjects (B). The prevalence of stage 3 CKD was significantly greater in UA stone patients (76%) compared with control subjects (45%) (P = 0.020). In the UA-low group, eGFR was significantly lower in patients with UA stones compared with CaOx/CaP (P = 0.034) (C). The prevalence of stage 3 CKD was not significantly different between in patients with UA stone (53%) and CaOx/CaP (40%) (P = 0.414). In the UA-high group, eGFR was significantly lower in patients with UA stones compared with CaOx/CaP (P = 0.032) (D). The prevalence of stage 3 CKD was significantly greater in UA stone patients (76%) compared with CaOx/CaP (38%) (P = 0.028).
Table 2.
Clinical characteristic of pair-matched subjects (analysis 1).
Table 3.
Clinical characteristic of pair-matched subjects (analysis 2).
Fig 4.
Multivariate logistic regression analysis for stage 3 CKD.
Independent risk factors for development of stage 3 CKD in the 3082 control subjects and 602 stone patients were evaluated using multivariate logistic regression analysis including 12 variables. UA stone, CaOx/CaP, hyperuricemia, presence of CVD, body mass index, age and hemoglobin were selected as independent risk factors for stage 3 CKD.
Fig 5.
Multivariate logistic regression analysis for stage 3B CKD.
Age, body mass index, presence of CVD, DM, hemoglobin, hyperuricemia, CaOx/CaP, and UA stone were selected as independent risk factors for stage 3B CKD.
Table 4.
Independent risk factors for stage 3 CKD by uni- and multivariate logistic regression analysis (analysis 3).
Table 5.
Independent risk factors for stage 3B CKD by uni- and multivariate logistic regression analysis (analysis 3).