Kidney Week

Abstract: TH-PO851

Serum Amyloid A and Clinical Outcomes in Early Diabetic Kidney Disease

Session Information

• Diabetic Kidney Disease: Basic - I
  October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Diabetic Kidney Disease

• 601 Diabetic Kidney Disease: Basic

Authors

• Dieter, Brad, Providence Health Care, Spokane, Washington, United States

Brad Dieter,

• McPherson, Sterling, Washington State University College of Medicine, Spokane, Washington, United States

Sterling McPherson,

• Alicic, Radica Z., Providence Medical Research Center, Spokane, Washington, United States

Radica Z. Alicic,

• Saulnier, Pierre Jean, CHU poitiers, Poitiers, France

Pierre Jean Saulnier,

• Nelson, Robert G., National Institutes of Health, Phoenix, Arizona, United States

Robert G. Nelson,

• Kretzler, Matthias, U.Michigan, Ann Arbor, Michigan, United States

Matthias Kretzler,

• Tuttle, Katherine R., University of Washington School of Medicine, Spokane, Washington, United States

Katherine R. Tuttle,

Background

Risk of end-stage renal disease (ESRD) and death is increased by higher systemic levels of serum amyloid A (SAA) in patients with advanced diabetic kidney disease (DKD), but not among those with early or no DKD. The study aim was to determine relationships between indicators of locally-produced SAA and clinical outcomes in these patients.

Methods

In Pima Indians with type 2 diabetes, SAA protein was measured in urine (enzyme-linked immunosorbent assay, n=159) and SAA mRNA was quantified (microarray, n=46) in kidney tissue from research biopsies. Cox-proportional hazard models tested whether urinary SAA forecasted future ESRD and all-cause death. Linear regression models tested for cross-sectional associations of SAA mRNA in kidney tissue with urine albumin-to-creatinine ratio (UACR) or estimated glomerular filtration rate (eGFR; CKD-EPI-creatinine). Urine and blood samples for these analyses were collected within 6 months of the biopsies.

Results

Baseline characteristics included: age 47±11 (mean±SD) years, women (117/159; 74%), hemoglobin A1c 9.1±2.3%, diabetes duration 17±7 years, body mass index 36±8 kg/m², renin angiotensin system inhibitor use (64/159; 40%), systolic blood pressure 124±16 mm Hg, UACR 39,14-209 mg/g (median, interquartile range), eGFR 103, 91-114 mL/min/1.73m². During a median of 12.5 years, 25 participants developed ESRD and 53 died. SAA-positive urine forecasted ESRD (hazards ratio [HR] 3.73, 95% confidence interval [CI] 1.11–12.53, p=0.022), but not death (HR 1.71, 95% CI 0.53– 5.50). In kidney tissue, tubulointersitial SAA mRNA was associated with a 6.3 mL/min/1.73m² lower eGFR (p<0.001) and a 0.80 log-fold increase in UACR per log-fold change in SAA mRNA (p<0.001). Glomerular SAA mRNA was associated with a 4.5 mL/min/1.73m² lower eGFR (p<0.001) and a 0.58 log-fold increase in UACR per log-fold change in SAA mRNA (p<0.001).

Conclusion

Urinary SAA forecasted ESRD and higher levels of SAA mRNA in kidney tissue correlated with lower eGFR and higher UACR. Associations between indicators of locally-produced SAA and clinical outcomes point to a candidate mechanism for DKD onset and progression in diabetic patients with early or no DKD.