RENAL10: Urinary Vitamin D Binding Protein and Kidney Injury Molecule-1 May Predict the Risk of Acute Kidney Injury in Heart Failure Patients Undergoing CF-LVAD Implantation

Background: Renal dysfunction in advanced heart failure (HF) and durable mechanical circulatory support (MCS) is common and harmful, yet current tests are inaccurate and confounded by body composition, fluid status, and other factors. Newer biomarkers more directly reflecting specific aspects of renal health/function have been tested in other fields but have yet to be tested in this complex setting. Urine vitamin-D-binding protein (uVDBP) and urine kidney injury molecule-1 (uKIM-1) have been shown to identify and stage severity of parenchymal kidney injury in several settings. The purpose of our study is to explore the potential role of uVDBP and uKIM-1 as biomarkers of acute renal damage among HF patients receiving CF-LVAD support. Methods: Our cohort included 29 HF patients undergoing CF-LVAD implantation. Early morning voided urine was collected immediately pre-implantation. Measurements of uVDBP and uKIM-1 were performed with ELISA kits and data were normalized to urine creatinine (ng/mg Cr.). Relationships of uVDBP and uKIM-1 and important baseline clinical characteristics (presence of kidney dysfunction, defined as eGFR<60 ml/min/1.73m2, and hemodynamic instability, defined as INTERMACS 1/2) were assessed using the Wilcoxon test. Relationships of baseline uVDBP and uKIM-1 (log2 transformed) with risk of acute kidney Injury (AKI) developing within 90 days after CF-LVAD implant were assessed, using logistic regression with adjustment for baseline kidney function and for baseline hemodynamic stability. AKI was assessed using RIFLE (Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease) criteria. Results: The cohort (n=29), had mean (SD) age 58 (14) years, 17.2% were female, and destination therapy goal for 96.6%. Baseline kidney dysfunction and hemodynamic instability were present in 37.9% and 44.8% of patients respectively, and 44.8% had AKI within 90 days. Baseline uVDBP [median (IQR)] was 0.88 (0.20-2.3) ng/mg Cr., and baseline uKIM-1 [median (IQR)] was 0.013 (0.0032-0.042) ng/mg Cr. Baseline uVDBP was found to be significantly associated with baseline kidney risk category (p <0.001), with higher uVDBP associated with worse baseline kidney health/function risk (assessed by combination of baseline low kidney function and hemodynamic instability, Fig.1A). Baseline uKIM-1 was significantly associated with baseline kidney function but not with baseline kidney risk category (Fig.1B). Two-fold higher baseline uKIM-1 was associated with nearly two-fold increased odds of 90-day AKI (Table.1). Similar results were obtained for baseline uVDBP. Conclusions: We have demonstrated that both uVDBP and uKIM-1 may provide additional information about baseline kidney health beyond that provided by serum creatinine-based and hemodynamic instability-based measures, assessed by risk for adverse 90-day kidney outcomes. Further study in larger cohorts is needed. Table 1. - Association of baseline urine markers with post-implant AKI Odds ratios (95% confidence intervals) Unadjusted Model 1 Model 2 uKIM-1 Per two-fold greater uKIM-1 1.98 (1.28-3.71) 4.39 (1.79-22.2) 8.76 (2.03-311) uVDBP Per two-fold greater uVDBP 1.99 (1.27-3.68) 2.25 (1.23-4.90) 7.74 (2.07-296) Model 1 was adjusted for baseline kidney dysfunction;Model 2 was adjusted for baseline kidney dysfunction and baseline hemodynamic instability