Age differences in the relationships between risk factors and loss of kidney function: a general population cohort study

Abstract

Background: Annual decline in kidney function is a widely applied surrogate outcome of renal failure. It is important to understand the relationships between known risk factors and the annual decline in estimated glomerular filtration rate (eGFR) according to baseline age; however, these remain unclear.

Methods: A community-based retrospective cohort study of adults who underwent annual medical examinations between 1999 and 2013 was conducted. The participants were stratified into different age groups (40-49, 50-59, 60-69, 70-79, and ≥ 80 years) to assess the risk for loss of kidney function. A mixed-effects model was used to estimate the association between risk factors and annual changes in eGFR.

Results: In total, 51,938 participants were included in the analysis. The age group of ≥80 years included 8127 individuals. The mean annual change in eGFR was - 0.39 (95% confidence interval: - 0.41 to - 0.37) mL/min/1.73 m² per year. Older age was related to faster loss of kidney function. In the older age group, higher systolic blood pressure, proteinuria, and current smoking were related to faster loss of kidney function (p trend < 0.01, 0.03, and < 0.01, respectively). Conversely, each age group showed similar annual loss of kidney function related to lower hemoglobin levels and diabetes mellitus (p trend 0.47 and 0.17, respectively).

Conclusions: Higher systolic blood pressure, proteinuria, and smoking were related to faster loss of kidney function, and a greater effect size was observed in the older participants. More risk assessments for older people are required for personalized care.

Keywords: Aged; Chronic renal insufficiency; Glomerular filtration rate.

Fig. 1

Risk factors for loss of kidney function and differences from the mean estimated glomerular filtration rate slopes according to age. Number of participants in each group: n = 1775 (age 40–49 years); n = 4396 (age 50–59 years); n = 19,015 (age 60–69 years); n = 18,635 (age 70–79 years); and n = 8127 (age ≥ 80 years). Values represent differences from the mean slopes of each age group. Each variable was adjusted for all other variables. The error bars represent 95% confidence intervals. The p trend value was obtained to test the consistency of the age relationships among the age groups

Fig. 2

Estimated glomerular filtration rate (eGFR) slopes according to age groups with respect to representative risk factors. Values are mean (95% confidence intervals) changes in eGFR slopes adjusted for sex, systolic blood pressure, diastolic blood pressure, body mass index, proteinuria, hemoglobin, total cholesterol, smoking status, history of coronary disease, history of stroke, and diabetes mellitus. For variables not mentioned in the axis labels, the average values were applied for each age group. The values of the representative risk factors (blood pressure, urinary protein, current smoking status) were chosen arbitrarily