Association Between Urinary Protein-to-Creatinine Ratio and Chronic Kidney Disease Progression: A Secondary Analysis of a Prospective Cohort Study

Abstract

Objective: Studies on the association between urinary protein-to-creatinine ratio (UPCR) and chronic kidney disease (CKD) progression are limited. This study aimed to investigate the relationship between UPCR and CKD progression in a Japanese population.

Methods: The present research was a secondary analysis of a prospective cohort study. Eight hundred and ninety-six subjects from the research of CKD-ROUTE in Japan were included. All the patients were new visitors or first referred to the participating centers of nephrology between October 2010 and December 2011. The target-independent variable was UPCR measured at baseline. The dependent variable was CKD progression and the estimated glomerular filtration rate (eGFR) changes during follow-up. We used Cox proportional hazards regression to investigate the association between UPCR and CKD progression risk. To address UPCR and CKD progression's non-linearity, a multivariate Cox proportional hazards regression analysis with cubic spline functions model and smooth curve fitting (penalized spline method) were conducted. We further used a generalized linear mixed model to explore the relationship between UPCR and the changes of eGFR.

Result: The mean age of the included patients was 67.2 ± 13.4 years old. Two hundred and thirty-four people occurred CKD progression during follow-up. The present study showed that UPCR was independently associated with CKD progression in the multivariate analysis [HR = 1.164, 95% CI (1.116, 1.215)]. The non-linear relationship between UPCR and CKD progression was explored in a dose-dependent manner, with an obvious inflection point of 1.699. Furthermore, our findings indicated that the tendency of the effect sizes on both the left and right sides of the inflection point was not consistent [left HR: 4.377, 95% CI (2.956, 6.483); right HR: 1.100, 95% CI (1.049-1.153)]. Using the linear mixed-effects regression model, we found that UPCR was an independent predictor of the longitudinal changes in eGFR (p < 0.001 for the interaction term with time).

Conclusion: This study demonstrates a nonlinear positive relationship between UPCR and CKD progression in the Japanese population. UPCR is also an independent predictor of the longitudinal changes in eGFR.

Keywords: Cox proportional hazards regression; chronic kidney disease progression; linear mixed-effects regression model; non-linearity; urinary protein-to-creatinine ratio.