Associations of Biomarkers of Kidney Tubule Health with Retinal Microvascular Signs

Key Points Kidney injury molecule-1 and soluble urokinase plasminogen activator receptor are associated with retinal microvascular changes in individuals without CKD, diabetes, or cardiovascular disease. Tubule biomarkers may reveal microvascular pathways linking kidney dysfunction to cardiovascular risk beyond eGFR or albumin-to-creatinine ratio. Background CKD is strongly associated with cardiovascular disease (CVD), yet the etiology responsible for this link remains elusive. Novel blood and urine biomarkers reflecting kidney tubule dysfunction and injury may provide novel insights to mechanisms linking the kidney to CVD. Methods In 470 participants of the Multi-Ethnic Study of Atherosclerosis without type 2 diabetes, CVD, or CKD, we measured six plasma (kidney injury molecule-1 [KIM-1], monocyte chemoattractant protein-1, soluble urokinase plasminogen activator receptor, tumor necrosis factor receptor 1 and 2, and anti–chitinase-3-like protein 1) and six urinary ( α 1 microglobulin, EGF, KIM-1, monocyte chemoattractant protein-1, anti–chitinase-3-like protein 1, and uromodulin) kidney tubule health biomarkers. To assess microvascular health, we used retinal microvascular measurements assessed from fundus photography: central retinal arteriolar and venular equivalents (central retinal artery equivalent [CRAE] and central retinal venular equivalent [CRVE], respectively). Multivariable linear regression evaluated associations of tubule biomarkers and kidney function with CRAE and CRVE. Results The mean participant age was 60±10 years with 52% female. The racial and ethnic distribution was 46% White, 24% Black, 18% Hispanic, and 11% Chinese. The mean eGFR was 92.1±13.3 ml/min per 1.73 m 2 , and the median urine albumin-to-creatinine ratio was 4.7 mg/g (interquartile range, 3.0–9.4). Higher plasma KIM-1 ( β , −5.14; 95% confidence interval [CI], −9.84 to −0.45) and urine KIM-1 ( β, −5.68; 95% CI, −10.15 to −1.22) concentrations were individually associated with narrower CRAE, while plasma soluble urokinase plasminogen activator receptor concentrations were individually associated with wider CRAE ( β , 9.15; 95% CI, 0.89 to 17.4) and CRVE ( β , 21.49; 95% CI, 9.39 to 33.59). There were no significant associations between the remaining tubule health biomarkers and CRAE or CRVE nor were there associations between eGFR or urine albumin-to-creatinine ratio with CRAE and CRVE. Conclusions In this study of community-living individuals without CKD, diabetes, or CVD, selected kidney tubule health markers are associated with retinal microvascular changes. These findings suggest that kidney tubule biomarkers may reflect or contribute to systemic microvascular dysfunction, above and beyond glomerular damage. Tubular biomarkers may help elucidate the shared microvascular mechanisms linking CKD and CVD.