Abstract: TH-PO108
Risk Factors and Outcomes of AKI Subphenotypes Based on Serum Creatinine Trajectory After Vancomycin Exposure
Session Information
- AKI: Biomarkers, Drugs, Onco-Nephrology
November 07, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- Ramirez-Sandoval, Juan Carlos, Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubiran, Mexico City, Mexico
- Paz-Cortes, Jose A., Universidad Nacional Autónoma de México, Mexico City, Mexico
- García-Sánchez, Lourdes G., Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubiran, Mexico City, Mexico
- Vergara-Juarez, Josue J., Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubiran, Mexico City, Mexico
- Correa-Rotter, Ricardo, Institutor Nacional de la Nutricion, Mexico City, Mexico
Background
Vancomycin-associated (VA) acute kidney injury (AKI) is poorly characterized. We hypothesized that the phenotyping of VA-AKI according to the time course of changes in serum creatinine (sCr ‘trajectory’) could identify different risk factors and prognosis.
Methods
Cohort study. We included all subjects (2017 to 2019) admitted to a tertiary referral hospital exposed to vancomycin IV ≥4 days without CKD G5 or dialysis treatment. We collected daily sCr and vancomycin serum concentrations, calculated the area under the concentration-time curve (AUC24h), and eGFR 30-days after hospital discharge.
Results
We included 361 subjects. In survivors (332/361,94%), we identified 4 phenotypes based on sCr trajectory: 1 No AKI-VA (n=229, 68%): subjects without AKI who did not have sCr changes during exposure; 2 Severe AKI-VA (n=19,6%): subjects with an accelerated rise in sCr not related to other clinical factors, with a median time of vancomycin exposure 10 days (IQR: 7-18) with tubular injury (biopsy in 2 cases); 3 non-severe AKI-VA (n=55, 17%): subjects who at the beginning of vancomycin prescription had AKI, which was improving but had slight and slow sCr increases during treatment, usually in context of other AKI risk factors (sepsis relapsed, nephrotoxic drugs, bleeding); 4 recovery of AKI (n=29, 9%): subjects with sepsis induced AKI who had improvement without relapses during treatment. In a multivariate analysis, risk factors for group 2 were the slope of the initial day 2-4 vancomycin drug levels (OR:2.0 95%IC 1.22-2.7) and baseline sCr (OR:1.7 95%IC 1.1-2.8). Risk factors for group 3 non-severe AKI-VA were vancomycin drug levels >15 ng/mL (OR 1.6 per each 10 ng/mL, 95%IC 1.1-3.5) and AUC24h target of ≥ 600 mg*h/L (OR 2.9 95%IC 1.6-5.9). After 30-d of discharge, severe AKI-VA was usually reversible (12% had CKD G3A1). Group 3 had a higher frequency of CKD (78% CKD G3) compared to other groups. Total daily vancomycin dose, accumulated dose, and duration of therapy were not risk factor for any group.
Conclusion
VA-AKI occurs in different clinical presentations. Abrupt and severe AKI-VA can be predicted according to slope change in first doses of vancomycin levels. Non-severe AKI-VA was multifactorial, had a higher risk of CKD, and had pre-AKI high vancomycin trough levels.