We calculated the in vitro delivery accuracy of a computer software (CCIP) linked to an Alaris® GH syringe pump to deliver propofol by target-controlled infusion in dogs. This was studied by simulating an infusion of propofol at two target concentrations (2 and 6 µg ml–1) in a 6 and 22 kg dog, each for 10, 60 and 240 minutes in a crossover design. Each infusion was replicated three times (36 infusions). The total volume of propofol delivered at the end of each infusion was measured using a high precision scales and compared with the volume predicted by the software and that displayed by the syringe driver. Median prediction error (MDPE), median absolute performance error (MDAPE), divergence and intraclass correlation coefficient (ICC) were calculated as indexes of bias, accuracy, time-related changes in accuracy and reliability, respectively. The ranges of MDPA and MDAPE for all the infusions were –2.08-4.28% and 1.85-4.28%, respectively and the median (95% CI) was –1.75% (–1.84-1.02%) and 2.18% (2.23-4.13%), respectively. The divergence was 0.010 ± 0.004% h⁻¹. The ICC was 1.00% (0.99-1.00%; P < 0.0001). The volume of propofol infused by the CCIP linked to an Alaris® GH syringe pump is accurate and has a small tendency to overestimate the real volume delivered. The prediction error fell in the ± 5% range of delivery performance considered clinically acceptable. The performance of the CCIP ameliorates with time and the error will decrease for long infusions, reducing the risk of administering an excessive dose of propofol and increasing the safety of CCIP. The CCIP had also an excellent index of reliability for all the targets, animal body weight and length of infusions tested. In conclusion, CCIP linked to an Alaris® GH syringe pump is precise, reliable and suitable for clinical practice.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.