2025/11/14 by Andrea Gentile, Michele Introna, Michel Struys +1 · 1 voice · 1 citation
Medicine · Engineering · Health Professions · #Anesthesia and Sedative Agents #Intravenous Infusion Technology and Safety #Patient Safety and Medication Errors
paper · pdf · doi:10.1111/anae.70078
Target-controlled infusions have been criticised as too slow for a rapid sequence induction (RSI), due primarily to the maximum infusion rate of commercially available pumps (approximately 1200 ml.h-1). Despite the widespread adoption of total intravenous anaesthesia, there is limited consensus on what truly defines a ‘rapid’ induction of general anaesthesia, and no standardised description of RSI components exists [1]. A new feature in commercial target-controlled infusion pumps enables administration of an additional manual bolus to accelerate induction of anaesthesia [2]. The rationale relies on comparing a 5 s manual bolus with a capped 1200 ml.h-1 infusion rate, supported by proof-of-concept simulation in a single virtual subject. This study extends that concept by providing a simulation-based pharmacokinetic proof of feasibility for target-controlled infusions use in RSI. Simulations were performed using the Python Anaesthesia Simulator [3], which incorporates the Eleveld population model for propofol. Virtual patients were defined as 35-year-old males, 175 cm in height, with bodyweights ranging from 40 kg to 160 kg and assuming concomitant opioid administration. To isolate the effect of bodyweight on model performance and infusion limitations, we kept age, sex, height and opioid co-administration fixed across simulations, varying weight only. This approach ensured that differences in onset dynamics were attributable to weight-dependent changes in pharmacokinetics and to the infusion-rate cap of the pump, rather than to covariate interactions. Using propofol 1%, three induction doses were evaluated (1.5, 2.0 and 2.5 mg.kg-1), administered by two modalities: target-controlled infusion at a high infusion rate of 1200 ml.h-1; and a manual bolus delivered at 3360 ml.h-1. Without standardised guidelines for propofol administration during RSI, a manual bolus of 2 mg.kg-1 over 15 s in a 70 kg patient was assumed. Simulation infusions continued until the full dose was delivered. Time to peak effect-site concentration was used as a marker of induction speed, while peak plasma concentration of propofol served as a surrogate for risk of overshoot and haemodynamic instability. Although these indices do not capture loss of consciousness directly or clinical responses, they provide mechanistically relevant markers of onset dynamics, including the trajectory of the effect-site of propofol and systemic exposure. Primary outputs included: plasma and effect-site propofol concentration trajectories; time to peak effect-site concentration; maximum plasma site concentration; duration of infusion; and area under the curve at time-to-peak effect-site concentration. Results were summarised as median (IQR [range]) across bodyweights, with relative differences in proportions reported for time to peak effect-site and plasma concentration. Across all weight ranges (40–160 kg), effect-site concentrations at time-to-peak effect were nearly identical between target-controlled and manual infusions, differing by < 1% at all doses (Table 1 and Fig. 1a). This indicates that target-controlled infusions preserve effect-site kinetics relevant to induction despite differences in infusion modality. In contrast, time-to-peak effect was consistently longer with target-controlled infusions, showing a dose-dependent increase in delay: 8% at 1.5 mg.kg-1; 12% at 2.0 mg.kg-1; and 14% at 2.5 mg.kg-1 (Fig. 1b). These differences remained modest across weights, reflecting only minimal practical impact on effect-site equilibration time. Peak plasma propofol concentration showed the most pronounced divergence, being attenuated with target-controlled infusions. Compared with a manual bolus, target-controlled infusions reduced maximum plasma site concentration by approximately 14%, 19%, and 22% at 1.5, 2.0, and 2.5 mg.kg-1, respectively, with greater relative reductions observed at higher doses and lower bodyweights (Fig. 1c). The use of target-controlled infusions for RSI has gained renewed attention recently due to the lack of updated guidelines [4]. Our simulations show that target-controlled infusions, even at the maximum pump rate, achieve effect-site kinetics compatible with RSI across a wide bodyweight range. Compared with a manual bolus, target-controlled infusions caused only a modest delay in time-to-peak effect-site concentration (approximately 10 s) while reducing peak plasma levels consistently, suggesting a potential advantage in limiting adverse effects. Importantly, effect-site dynamics were preserved, indicating that pharmacodynamic onset was largely maintained. The time-to-peak effect-site concentration reflects the equilibration delay between plasma and effect-site compartments, governed primarily by the effect-site rate constant, not the plasma profile; therefore explaining the minimal differences observed between infusion modalities. Although limited to pharmacokinetic surrogates, these results provide a technical proof-of-concept. Time-to-peak effect-site concentration and peak plasma concentration do not predict clinical endpoints directly, making clinical validation essential. However, hypnotic onset is known to depend on effect-site equilibration rather than plasma peaks [5]. In this context, manual bolus may be overly aggressive, especially in patients who are haemodynamically unstable or living with obesity, where plasma overshoot increases risk [6-8]. Our simulations show that a target-controlled infusion smooths plasma kinetics, preserves effect-site exposure and achieves RSI concentrations without delaying onset. These findings challenge the common perception that target-controlled infusions are too slow for RSI and instead support its non-inferiority to manual bolus administration, offering a more controlled and potentially safer pharmacokinetic profile. This work was supported/partially supported by the Italian Ministry of Health. MI's research group has received research grants from Becton Dickinson (Eysins, Switzerland) and consultancy fees from Aspen Healthcare FZ LLC (Dubai, UAE). MS's research group/department received research grants and consultancy fees from Masimo (Irvine, CA, USA); Becton Dickinson (Eysins, Switzerland); Fresenius-Kabi (Bad Homburg, Germany); Paion (Aachen, Germany); Medcaptain Europe (Andelst, The Netherlands); Baxter (Chicago, IL, USA); and HanaPharm (Seoul, Republic of Korea). MS receives royalties on intellectual property from Demed Medical (Sinaai, Belgium) and Ghent University (Ghent, Belgium). Open access funding provided by BIBLIOSAN.