The proper posology of antibiotics in the critically ill in CRRT is difficult to assess. We therefore performed a prospective observational cohort study to make clear hints in this topic. Our results reveal a high Sieving Coefficient for all antibiotics, equal to or higher than those described in previous papers. CVVH clearance in relation to total body clearance was significant, (i.e., >than 25% for all classes). A strong correlation between the antibiotic concentrations obtained in plasma and ultrafiltrate was found both at the peak and in the valley, with the determination of two equations that allow a new method for calculating the amount of antibiotic lost in CVVH both for trough levels and peak. Based on the results of our study and considering the limitations we believe that we can extrapolate the following final considerations: (1) it is likely to carry out a loading dose for the main antibiotics (2) subsequent administrations must take into account the daily loss identified by the linear regression equation. This angular coefficient gives the idea that the average daily loss of given antibiotic is about 25%; this implies that on the basis of the linear regression equation that correlates ultrafiltered/plasma antibiotic concentration, the dosage should be increased by 25% every day, while still ensuring a daily plasma TDM of the drug.
Keywords: CVVH; acute kidney failure; antibiotic; critically ill patients; parmacokinentics; pharmacodinamic.