Dipstick proteinuria predicts all-cause mortality in general population: A study of 17 million Korean adults

Abstract

Objective A quantitative basis for the use of dipstick urinalysis for risk assessment of all-cause mortality is scarce. Therefore, we investigated the association between dipstick proteinuria and all-cause mortality in a general population and evaluated the effect of confounders on this association. Methods The study population included 17,342,956 adults who underwent health examinations between 2005 and 2008 under the National Health Insurance System. Proteinuria was determined using a single dipstick urinalysis, and the primary outcome of this study was all-cause mortality. The prognostic impact of proteinuria was assessed by constructing a multi-variable Cox model. Results The mean age of the study population (53.24% male) was 46.06 years; 724,681 deaths from all causes occurred over a median follow-up period of 9.34 years (interquartile range 8.17-10.16), and the maximum follow-up was 12.12 years. After full adjustment for covariates, a higher level of dipstick proteinuria indicated a higher risk of all-cause death [Hazard ratios (95% confidence intervals); 1.22 (1.20-1.24), 1.47 (1.45-1.49), 1.81 (1.77-1.84), 2.32 (2.24-2.41), 2.74 (2.54-2.96); trace to 4+, respectively], and various subgroup analyses did not affect the main outcome for the total population. >= 1+ proteinuria in the group without metabolic diseases (hypertension, diabetes, dyslipidemia, or obesity) resulted in higher hazard ratios than those in the group with metabolic diseases and negative or trace proteinuria. Conclusions Our study showed a strong association between dipstick proteinuria and all-cause mortality in this nationwide population-based cohort in South Korea.