Cargando…

1107. Vancomycin AUC Dosing: Is One Concentration in the Hand Worth Two in the Bush?

BACKGROUND: Recent guidelines recommend a transition from trough-based to area-under the curve-based (AUC) monitoring for vancomycin for serious invasive methicillin-resistant Staphylococcus aureus infections. Due to the challenges of performing AUC monitoring in clinical practice, this study sought...

Descripción completa

Detalles Bibliográficos
Autores principales: Spivey, Justin, Shroba, Jenny, Deri, Connor, Nys, Cara, Wrenn, Rebekah, Yarrington, Michael E
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8644980/
http://dx.doi.org/10.1093/ofid/ofab466.1301
Descripción
Sumario:BACKGROUND: Recent guidelines recommend a transition from trough-based to area-under the curve-based (AUC) monitoring for vancomycin for serious invasive methicillin-resistant Staphylococcus aureus infections. Due to the challenges of performing AUC monitoring in clinical practice, this study sought to compare the accuracy of an AUC calculated from two points using trapezoidal calculations and from a single steady-state trough combined with population assumptions. METHODS: This prospective cohort analysis included hospitalized patients with stable renal function from 10.2020 to 12.2020 with two vancomycin concentrations obtained at steady-state during a single dosing interval. For each patient, AUC was calculated via trapezoidal equations utilizing peak and trough concentrations (P/T) and using the trough concentration (T) combined with population volume of distribution. Appropriate concentrations were defined as a peak at least 2 hours after the end of the infusion and a trough within one hour of the next dose. The percent and actual differences were calculated between the P/T and T AUC assessments for each patient. A patient level review was independently conducted by two clinical pharmacists to evaluate if a change in dosing would have been made according to AUC estimation methodology. RESULTS: Thirty-one patients had appropriate steady-state P/T obtained. Baseline demographics are shown in Table 1 with the majority of patients being overweight with normal renal function. The mean calculated AUC for both groups was similar, P/T 544.8 and T 549.8. The mean and median percent differences were 1.85% and 0.65%, with a standard deviation of 7.3% and an apparent normal distribution (Figure 1, p = 0.94 by Shapiro’s test). The median absolute difference in AUC was 25.82 mg*h/L between methodologies. Both methods would have resulted in the same modification to the vancomycin regimen based on patient level chart review. [Image: see text] [Image: see text] CONCLUSION: The single-trough method performed similarly to the more laborious P/T method. No patient would have received a dose adjustment based on the two different AUC estimation methods. The single-trough method may represent a resource and workflow conscious AUC estimation method for patients meeting population assumptions. DISCLOSURES: All Authors: No reported disclosures