Does contemporary vancomycin dosing achieve therapeutic targets in a heterogeneous clinical cohort of critically ill patients? Data from the multinational DALI study

Abstract

Introduction: The objective of this study was to describe the pharmacokinetics of vancomycin in ICU patients and to examine whether contemporary antibiotic dosing results in concentrations that have been associated with favourable response.

Methods: The Defining Antibiotic Levels in Intensive Care (DALI) study was a prospective, multicentre pharmacokinetic point-prevalence study. Antibiotic dosing was as per the treating clinician either by intermittent bolus or continuous infusion. Target trough concentration was defined as ≥15 mg/L and target pharmacodynamic index was defined as an area under the concentration-time curve over a 24-hour period divided by the minimum inhibitory concentration of the suspected bacteria (AUC0-24/MIC ratio) >400 (assuming MIC ≤1 mg/L).

Results: Data of 42 patients from 26 ICUs were eligible for analysis. A total of 24 patients received vancomycin by continuous infusion (57%). Daily dosage of vancomycin was 27 mg/kg (interquartile range (IQR) 18 to 32), and not different between patients receiving intermittent or continuous infusion. Trough concentrations were highly variable (median 27, IQR 8 to 23 mg/L). Target trough concentrations were achieved in 57% of patients, but more frequently in patients receiving continuous infusion (71% versus 39%; P = 0.038). Also the target AUC0-24/MIC ratio was reached more frequently in patients receiving continuous infusion (88% versus 50%; P = 0.008). Multivariable logistic regression analysis with adjustment by the propensity score could not confirm continuous infusion as an independent predictor of an AUC0-24/MIC >400 (odds ratio (OR) 1.65, 95% confidence interval (CI) 0.2 to 12.0) or a Cmin ≥15 mg/L (OR 1.8, 95% CI 0.4 to 8.5).

Conclusions: This study demonstrated large interindividual variability in vancomycin pharmacokinetic and pharmacodynamic target attainment in ICU patients. These data suggests that a re-evaluation of current vancomycin dosing recommendations in critically ill patients is needed to more rapidly and consistently achieve sufficient vancomycin exposure.

Figure 1

Boxplots of vancomycin dosages administered stratified for renal replacement therapy and various glomerular filtration rates. ^*Data on glomerular filtration were only available in 30 patients. No difference between groups could be demonstrated (P = 0.359).

Figure 2

Boxplot of vancomycin trough concentrations stratified for renal replacement therapy and various glomerular filtration rates. ^*Data on glomerular filtration were only available in 30 patients. Red line indicates vancomycin trough concentration target (15 mg/L). No difference between groups could be demonstrated (P = 0.233).

Figure 3

Vancomycin trough concentration according to daily dose administered. ^*Red line indicates target trough concentration of 15 mg/L; circles represent patients with continuous infusion; triangles represent patients with intermittent dosing. Spearman rank correlation for (i) all patients: R = 0.540 (P <0.001), (ii) patients with continuous infusion: R = 0.587 (P = 0.062), and patients with intermittent dosing: R = 0.127 (P = 0.615).

Figure 4

Boxplot of vancomycin AUC_0–24/MIC ratios stratified for renal replacement therapy and various glomerular filtration rates. ^*Data on glomerular filtration were only available in 30 patients. Red line indicates the AUC_0–24/MIC target ratio (400). No difference between the groups could be demonstrated (P = 0.224). AUC_0–24/MIC, ratio of the area under the concentration-time curve over a 24-hour period divided by the minimum inhibitory concentration of the suspected bacteria.

Figure 5

AUC_0–24/MIC ratios of vancomycin for intermittent dosing versus continuous infusion. ^*Boxes indicate lower quartile (bottom of box), median (band near the middle of the box), and upper quartile (top of the box); whiskers indicate the lowest value still within 1.5 of the interquartile range from the lower quartile, and the highest value still within 1.5 of the interquartile range from the upper quartile; circles indicate outliers. The red line indicates the minimum the target threshold for optimizing outcomes (AUC_0–24/MIC >400). Difference between the groups: P = 0.029. AUC_0–24/MIC, ratio of the area under the concentration-time curve over a 24-hour period divided by the minimum inhibitory concentration of the suspected bacteria.