Cardiovascular and Kidney Outcomes of Non-Diabetic CKD by Albuminuria Severity: Findings From the CRIC Study

Abstract

Rationale & objective: The clinical trajectory of normoalbuminuric chronic kidney disease (CKD), particularly in the absence of diabetes, has not yet been well-studied. This study evaluated the association of kidney and cardiovascular outcomes with levels of albuminuria in a cohort of patients with nondiabetic CKD.

Study design: Prospective cohort study.

Setting & participants: 1,463 adults with nondiabetic CKD without known glomerulonephritis and diagnosed with hypertensive nephrosclerosis or unknown cause of CKD participating in the Chronic Renal Insufficiency Cohort (CRIC) Study.

Exposure: Albuminuria stage at study entry.

Outcome: Primary outcome: Composite kidney (halving of estimated glomerular filtration rate [eGFR], kidney transplantation, or dialysis), Secondary outcomes: (1) eGFR slope, (2) composite cardiovascular disease events (hospitalization for heart failure, myocardial infarction, stroke, or all-cause death), (3) all-cause death.

Analytical approach: Linear mixed effects and Cox proportional hazards regression analyses.

Results: Lower levels of albuminuria were associated with female sex and older age. For the primary outcome, compared with normoalbuminuria, those with moderate and severe albuminuria had higher rates of kidney outcomes (adjusted hazard ratio [AHR], 3.3 [95% CI, 2.4-4.6], and AHR, 8.6 [95% CI, 6.0-12.0], respectively) and cardiovascular outcomes (AHR, 1.5 [95% CI, 1.2-1.9], and AHR, 1.5 [95% CI, 1.1-2.0], respectively). Those with normoalbuminuria (<30μg/mg; n=863) had a slower decline in eGFR (-0.46mL/min/1.73m² per year) compared with those with moderate (30-300μg/mg, n=372; 1.41mL/min/1.73m² per year) or severe albuminuria (>300μg/mg, n=274; 2.63mL/min/1.73m² per year). In adjusted analyses, kidney outcomes occurred, on average, sooner among those with moderate (8.6 years) and severe (7.3 years) albuminuria compared with those with normoalbuminuria (9.3 years) whereas the average times to cardiovascular outcomes were similar across albuminuria groups (8.2, 8.1, and 8.6 years, respectively).

Limitations: Self-report of CKD etiology without confirmatory kidney biopsies; residual confounding.

Conclusions: Participants with normoalbuminuric nondiabetic CKD experienced substantially slower CKD progression but only modestly lower cardiovascular risk than those with high levels of albuminuria. These findings inform the design of future studies investigating interventions among individuals with lower levels of albuminuria.

Plain-language summary: Diabetes and hypertension are the leading causes of chronic kidney disease (CKD). Urine albumin levels are associated with cardiovascular and kidney disease outcomes among individuals with CKD. However, previous studies of long-term clinical outcomes in CKD largely included patients with diabetes. As well, few studies have evaluated long-term outcomes across different levels of urine albumin among people without diabetes. In this study, we found individuals with nondiabetic CKD and low urine albumin had much slower decline of kidney function but only a modestly lower risk of a cardiovascular events compared with those with high levels of urine albumin. Individuals with low urine albumin were much more likely to have a cardiovascular event than progression of their kidney disease. These findings inform the design of future studies investigating treatments among individuals with lower levels of albuminuria.

Keywords: Albuminuria; CKD; cardiovascular disease; dialysis; eGFR; proteinuria; renal insufficiency.