Apolipoprotein L1 Genotypes and the Association of Urinary Potassium Excretion with CKD Progression

Abstract

Background and objectives: Progressive CKD in Black individuals is strongly associated with polymorphisms in the APOL1 gene, but it is unknown whether dietary risk factors for CKD progression vary in high- versus low-risk APOL1 genotypes. We investigated if APOL1 genotypes modify associations of dietary potassium and sodium with CKD progression and death.

Design, setting, participants, & measurements: We analyzed 1399 self-identified Black participants enrolled in the Chronic Renal Insufficiency Cohort from April 2003 to September 2008. Exposures were calibrated 24-hour urine potassium and sodium excretion. The primary outcome was CKD progression defined as the time to 50% decline in eGFR or kidney failure. The secondary outcome was CKD progression or death. We tested for an interaction between urinary potassium and sodium excretion and APOL1 genotypes.

Results: Median 24-hour urinary sodium and potassium excretions in Black participants were 150 mmol (interquartile range, 118-188) and 43 mmol (interquartile range, 35-54), respectively. Individuals with high- and low-risk APOL1 genotypes numbered 276 (20%) and 1104 (79%), respectively. After a median follow-up of 5.23 years, CKD progression events equaled 605, and after 7.29 years, CKD progression and death events equaled 868. There was significant interaction between APOL1 genotypes and urinary potassium excretion with CKD progression and CKD progression or death (P=0.003 and P=0.03, respectively). In those with high-risk APOL1 genotypes, higher urinary potassium excretion was associated with a lower risk of CKD progression (quartiles 2-4 versus 1: hazard ratio, 0.83; 95% confidence interval, 0.50 to 1.39; hazard ratio, 0.54; 95% confidence interval, 0.31 to 0.93; and hazard ratio, 0.50; 95% confidence interval, 0.27 to 0.93, respectively). In the low-risk APOL1 genotypes, higher urinary potassium excretion was associated with a higher risk of CKD progression (quartiles 2-4 versus 1: hazard ratio, 1.01; 95% confidence interval, 0.75 to 1.36; hazard ratio, 1.23; 95% confidence interval, 0.91 to 1.66; and hazard ratio, 1.53; 95% confidence interval, 1.12 to 2.09, respectively). We found no interaction between APOL1 genotypes and urinary sodium excretion with CKD outcomes.

Conclusions: Higher urinary potassium excretion was associated with lower versus higher risk of CKD progression in APOL1 high-risk and low-risk genotypes, respectively.

Keywords: APOL1; chronic kidney disease; gene environment interaction; genotype; potassium.

Graphical abstract

Figure 1.

Higher urinary potassium excretion is associated with lower risk of CKD progression and CKD progression or death in high-risk APOL1 genotypes. (A) Time to event analyses for calibrated 24-hour urinary potassium excretion with CKD progression defined as 50% decline in eGFR or kidney failure are stratified by APOL1 genotypes in the Chronic Renal Insufficiency Cohort (CRIC). The fully adjusted model covariates are clinic site, age, sex, education, urinary creatinine excretion, waist circumference, body mass index, cigarette smoking, alcohol drinking, physical activity, history of hypercholesterolemia, history of diabetes, history of cardiovascular disease, use of diuretics, use of renin-angiotensin system blocking agents, use of other antihypertensive medications, baseline eGFR, and 24-hour calibrated urinary sodium excretion. For CKD progression, the P for the interaction urinary potassium excretion × APOL1 risk allele in the unadjusted model is <0.001; for the fully adjusted model, the P for interaction is 0.003. (B) Time to event analyses for calibrated 24-hour urinary potassium excretion with CKD progression or death are stratified by APOL1 genotypes. The fully adjusted model covariates are clinic site, age, sex, education, urinary creatinine excretion, waist circumference, body mass index, cigarette smoking, alcohol drinking, physical activity, history of hypercholesterolemia, history of diabetes, history of cardiovascular disease, use of diuretics, use of renin-angiotensin system blocking agents, use of other antihypertensive medications, baseline eGFR, and calibrated 24-hour urinary sodium excretion. For CKD progression or death, the P for the interaction urinary potassium excretion × APOL1 risk allele in the unadjusted model is 0.007; for the fully adjusted model, the P for interaction is =0.03. 95% CI, 95% confidence interval; HR, hazard ratio.

Figure 2.

Restricted cubic splines showing higher urinary potassium excretion is associated with a lower risk of CKD progression in high-risk APOL1 genotypes. The figure shows restricted cubic spline models of the associations of 24-hour urine–calibrated potassium with CKD outcomes adjusted for covariates. (Fully adjusted models are adjusted for clinic site, age, sex, education, urinary creatinine excretion, waist circumference, body mass index, cigarette smoking, alcohol drinking, physical activity, history of hypercholesterolemia, history of diabetes, history of cardiovascular disease, use of diuretics, use of renin-angiotensin system blocking agents, use of other antihypertensive medications, baseline eGFR, and urinary sodium excretion.) The x axis represents 24-hour urinary potassium in millimoles per day with the median 24-hour urinary potassium (43.5 mmol/d) in the APOL1 high-risk genotypes or APOL1 low-risk genotypes (42.9 mmol/d) as the reference group. The shaded blue lines and surrounding shadows represent HRs and 95% CIs, respectively, at different values of calibrated urinary potassium compared with the median value, which is set as the reference. Gray shadows at the bottom represent the frequency distributions of 24-hour urine–calibrated potassium. (A) Associations of urinary potassium excretion with CKD progression defined as 50% decline in eGFR or kidney failure using restricted cubic splines in the APOL1 high-risk genotype group. (B) Associations of urinary potassium excretion with CKD progression defined as 50% decline in eGFR or kidney failure in the APOL1 low-risk genotype group. (C) Associations of urinary potassium excretion with CKD progression or death in the APOL1 high-risk genotype group. (D) Associations of urinary potassium excretion with CKD progression or death in the APOL1 low-risk genotype group.