Nontraditional Risk Factors for Progression Through Chronic Kidney Disease Risk Categories: The Coronary Artery Risk Development in Young Adults Study

Abstract

Background: There may be nontraditional pathways of chronic kidney disease (CKD) progression that are complementary to classical pathways. Therefore, we aimed to examine nontraditional risk factors for incident CKD and its progression.

Methods: We used the generally healthy population (n = 4382) starting at age 27-41 years in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort, which is an observational longitudinal study. Nontraditional risk factors included forced vital capacity, inflammation, serum urate, and serum carotenoids. CKD risk category was classified using the estimated glomerular filtration rate (eGFR) and urinary albumin-to-creatinine ratio (UACR) measured in 1995-1996 and repeated every 5 years for 20 years: No CKD, low risk, moderate risk, high risk, and very high risk.

Results: At baseline, 84.8% had no CKD (eGFR ≥60 mL/min/1.73 m² and UACR <10 mg/g), 10.3% were in the low risk (eGFR ≥60 and UACR 10-29), and 4.9% had CKD (eGFR <60 and/or UACR ≥ 30). Nontraditional risk factors were significantly associated with the progression of CKD to higher categories. Hazard ratios per standard deviation of the predictor for incident CKD and its progression from the No CKD and low and moderate risk into CKD were inverse for forced vital capacity and serum carotenoids and positive for serum urate, GlycA, and C-reactive protein, the first 3 even after adjustment for conventional risk factors.

Conclusion: Several nontraditional markers were significantly associated with an increased risk of progression to higher CKD categories in generally healthy young to middle-aged adults.

Keywords: Antioxidant marker; Chronic kidney disease progression; Inflammatory marker; Longitudinal study; Lung function marker; Serum urate.

Figure 1.

Multivariable HRs for progression of chronic kidney disease (CKD) risk categories in relation to time-varying novel risk factors (per 1 SD) CKD risk category progression outcomes are (1) 1a vs 1b+ (from No CKD into Low Risk or higher); (2) 1a or 1b vs 2+ (from No CKD or Low Risk into Moderate Risk or higher); (3)1a, 1b, or 2 vs 3+ (from No CKD or Low or Moderate Risk into High or Very High Risk). Predictors are time-varying. See sample size and no. of cases in Supplementary Table 2. 1 SD for each predictor was as follows: FVC, 1.0 L; Plasma GlycA, 68.0 μmol/L; ln(serum hsCRP), 0.7 ln(μg/mL); serum urate, 1.4 mg/dL; sum of 4 serum carotenoids, 27.5 mg/dL. Blue markers represent hazard ratio <1, red markers represent hazard ratio ≥1. A. Models were adjusted for age, sex, race (Black or White), and maximal educational attainment. FVC was further adjusted for height. B. Models were further adjusted for the six conventional time-varying risk factors: pack-year of smoking ≥ 10, hypertension, diabetes, high triglycerides, low HDL-C, and obesity.

Figure 2.

Multivariable HRs for progression of chronic kidney disease (CKD) risk categories in relation to time-varying conventional predictors CKD risk category progression outcomes are (1) 1a vs 1b+ (from No CKD into Low Risk or higher); (2) 1a or 1b vs 2+ (from No CKD or Low Risk into Moderate Risk or higher); (3)1a, 1b, or 2 vs 3+ (from No CKD or Low or Moderate Risk into High or Very High Risk). Time-varying predictors indicate whether the condition had ever occurred at each examination. See sample size and no. of cases in Supplementary Table 2. Dichotomous time-varying predictors were defined as follows: hypertension (≥140/90 mm Hg or use of antihypertensive medications); diabetes (plasma glucose ≥126 mg/dL, 2-hr glucose ≥200 mg/dL, Hb1Ac ≥6.5%, or use of antidiabetic medications); high triglycerides (plasma triglycerides ≥150 mg/dL); low HDL-C (plasma HDL-C <50 mg/dL for female and <40 mg/dL for male); obesity (≥30 kg/m²). Blue markers represent hazard ratio <1, red markers represent hazard ratio ≥1. Abbreviations: CI, confidence interval; HbA1c, hemoglobin A1c; HDL-C, high-density lipoprotein; HR, hazard ratio. A. Models were adjusted for age, sex, race (Black or White), and maximal educational attainment. B. Models were further adjusted for all conventional predictors.