[Predictive capability of the TCI Diprifusor system in patients with terminal chronic renal insufficiency]

To evaluate the predictive capability of a target-controlled infusion (TCI) system in patients with terminal chronic renal failure by comparing real drug concentrations with predicted concentrations.

Forty ASA II-III patients undergoing kidney transplants were enrolled and grouped by age: group A 18-54 years old and group B 55-80 years old. Propofol was administered through the Diprifusor TCI system (Astra Zeneca) with the target concentration of propofol for induction set at 3 micrograms/mL for group A patients and 2.5 micrograms/mL for group B patients. Arterial blood samples were taken for analysis of drug concentrations at the following times: 2 and 5 minutes after starting the infusion; immediately after the incision; before and 5 minutes after increasing or decreasing the target concentration > 25%; before and 5 minutes after switching off the perfusion for surgery; upon eye opening; and 30 minutes after switching off the infusion pump. The predictive capability of the system was determined by performance error (PE). We calculated bias (%, median PE) and accuracy (%, median absolute PE).

Twenty group A and 20 group B patients were studied. The median PEs in groups A and B, respectively, were -3.45 (-20.3-28.4) and -1.1 (-19.7-15.4). Median absolute PEs were 21.2 (11.9-45.1) and 16.3 (11.5-27.4), respectively.

The results indicate that the predictive capability of the Diprifusor pharmacokinetic model is acceptable in patients with terminal kidney failure, given the minor bias of 10% to 20% and the degree of accuracy between 20% and 40%.