Progression of Chronic Kidney Disease Risk Categories and Risk of Cardiovascular Disease and Total Mortality: Coronary Artery Risk Development in Young Adults Cohort

Abstract

Background Previous studies of worsening chronic kidney disease (CKD) based on declining estimated glomerular filtration rate (eGFR) or increasing urine albumin-creatinine ratio (UACR) are limited to later middle-age and older adults. We examined associations of CKD progression and incident cardiovascular disease (CVD) and mortality in younger adults. Methods and Results We studied 4382 adults in CARDIA (Coronary Artery Risk Development in Young Adults) initially aged 27 to 41 years and prospectively over 20 years. Five-year transition probabilities across CKD risk categories were based on eGFR and UACR measured at each exam. Proportional hazards models predicted incident CVD and all-cause mortality by time-varying CKD risk category, adjusting for demographics and CVD risk factors. Progression of CKD risk categories over 20 years occurred in 28.7% (1256/4382) of participants, driven by increases in UACR, but including 5.8% (n=255) with eGFR<60 mL/min per 1.73 m² or UACR ≥300 mg/g. Compared with eGFR ≥60 and UACR <10, demographic and smoking-adjusted hazard ratios for CVD were 1.62 (95% CI, 1.21-2.18) for low CKD risk (eGFR ≥60 with UACR 10-29) and 13.65 (95% CI, 7.52-24.79) for very high CKD risk (eGFR <30 or eGFR 30-44 with UACR 30-299; or eGFR 30-59 with UACR ≥300). Corresponding hazard ratios for all-cause mortality were 1.42 (95% CI, 1.08-1.88) and 14.75 (95% CI, 9.97-21.82). Although CVD associations were attenuated after adjustment for mediating CVD risk factors, all-cause mortality associations remained statistically significant. Conclusions Among young to middle-aged adults, progression to higher CKD risk category was common. Routine monitoring eGFR and UACR holds promise for prevention of CVD and total mortality.

Keywords: CKD risk categories; KDIGO; all‐cause mortality; cardiovascular disease; progression; transition; young adults.

Figure 1. Definition of 5 CKD modified KDIGO risk categories employed in this study*.

*KDIGO 2012 risk matrix (reference [4]), modified to capture mild severity. Given relatively few people in G3aA1, those people were included with others who had eGFR <60 (orange category), whereas in the KDIGO 2012 risk matrix they were in the yellow category. Each block is formed based on combination categories of eGFR and UACR. CKD risk category numbering is such that a higher number reflects presumed greater risk. CKD indicates chronic kidney disease; eGFR, estimated glomerular filtration rate; KDIGO, Kidney Disease: Improving Global Outcomes; and UACR, urine albumin‐creatinine ratio.

Figure 2. Shift in multinomial CKD risk category prevalence between exam years 10 and 30.

Sample size with kidney function markers obtained or found with severe kidney disease on annual followup varied across exams, namely, 3461, 3311, 3369, 3403, and 3010 at exam years 10, 15, 20, 25, and 30, respectively. Cumulative worst CKD risk category for each exam, carrying forward the most recent nonmissing value in the case of missing information, was therefore obtained in 3461, 4032, 4261, 4376, and 4382 participants from exam years 10 to 30 exam. Number of new entries at exam years 15, 20, 25, and 30 were 571, 229, 115, and 6, respectively. Numbers in the very low CKD risk category at exam years 15, 20, 25, and 30 were 2966 (85.7%), 3240 (80.4%), 3113 (73.1%), 2907 (66.4%), and 2651(60.5%), respectively. CKD risk category classification was cumulative over time. Participants were classified in a CKD category at their first attended examination and that classification was updated at the next examination or carried forward if the next examination was missed. CARDIA indicates Coronary Artery Risk Development in Young Adults; and CKD, chronic kidney disease.