Kidney Week

Abstract: FR-PO023

The Value of Plasma Inflammatory Biomarkers in Sepsis-Associated AKI

Session Information

• AKI: Epidemiology, Risk Factors, Prevention - II
  November 08, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 101 AKI: Epidemiology, Risk Factors, and Prevention

Authors

• Kwong, Yuenting Diana, University of California at San Francisco School of Medicine, San Francisco, California, United States

Yuenting Diana Kwong,

• Koyner, Jay L., University of Chicago, Chicago, Illinois, United States

Jay L. Koyner,

• Kangelaris, Kirsten N., University of California at San Francisco School of Medicine, San Francisco, California, United States

Kirsten N. Kangelaris,

• Calfee, Carolyn, University of California at San Francisco School of Medicine, San Francisco, California, United States

Carolyn Calfee,

• Abbott, Jason, University of California at San Francisco School of Medicine, San Francisco, California, United States

Jason Abbott,

• Liu, Kathleen D., University of California at San Francisco School of Medicine, San Francisco, California, United States

Kathleen D. Liu,

Background

Sepsis-associated acute kidney injury (SA-AKI) may be mediated by inflammation. We evaluated the association of plasma inflammatory biomarkers with SA-AKI beyond traditional AKI risk factors and markers of structural damage (uNGAL and UACR).

Methods

We included 453 adults with sepsis (SIRS and adjudicated infection) with collected samples within 24 hours of ICU admission. Admission AKI risk factors were adjudicated into the following categories: hypotension, hypovolemia, nephrotoxins, obstruction and other (e.g. rhabdomyolysis). AKI was defined by ≥ 50% serum creatinine (SCr) rise within 7 days from preadmission SCr or nadir SCr if the former was missing. Plasma and urine biomarkers were log₁₀-transformed. We evaluated the relationship between plasma biomarkers and AKI using univariate and multivariable logistic models with and without backwards stepwise selection (p<0.05). Analysis was repeated for severe AKI (KDIGO stage 2-3).

Results

275 subjects (60%) had AKI, 100 (22%) had severe AKI, 42 (9%) required dialysis and 140 (31%) died in-hospital. A model with demographics and clinical variables (Table 1) had AUC of 0.68 for AKI. The addition of structural damage markers improved the AUC to 0.71, and all plasma biomarkers further increased the AUC to 0.73 (p≤0.01). No plasma biomarkers remained after backwards selection for AKI. Findings were replicated for severe AKI. Inclusion of all plasma biomarkers improved the AUC from 0.75 to 0.79 (p=0.03). Only ICAM remained during backwards selection.

Conclusion

Inflammatory biomarkers only modestly improve SA-AKI prediction. Inflammation and structural damage are likely to have occurred before ICU admission. Efforts targeting systemic inflammation to prevent SA-AKI in the ICU may be limited.

Funding

• NIDDK Support