Sean Coeckelenbergh, Alexandre Joosten
Numerous studies have shown that haemodynamic monitoring coupled with treatment protocols, commonly known as goal-directed haemodynamic therapy (GDHT), can improve patient outcome.1-3 This strategy, which is recommended by several anaesthesia societies, can be loosely defined as the use of data from advanced haemodynamic monitors (e.g., flow-related and/or dynamic parameters of fluid responsiveness) to titrate therapeutic interventions (e.g., intravenous fluids, vasopressors, and inotropes). The ultimate aim of GDHT is to optimise peripheral tissue perfusion and oxygenation.
This concept has greatly improved during the past thirty years. Technologies have evolved from being very invasive to completely non-invasive, and the philosophy of maximising oxygen delivery has given way to optimisation of cardiac output and perfusion pressure. However, despite major progress and the consistent benefits of this strategy, GDHT adoption continues to be quite infrequent and a lack of standardised criteria for perioperative fluid administration has resulted in significant clinical variability among practitioners.4Automation is a possible solution as it allows physicians to plan haemodynamic goals and frees them from time-consuming tasks. A closed-loop consisting of a controller that monitors and intervenes to maintain a predefined target is one option to automate care. This type of system, which can be seen as extra pairs of eyes and hands, decreases workload associated with repetitive tasks. It has recently been shown to consistently apply goal-directed fluid therapy (GDFT), which is an important component of GDHT.
Closed-loop fluid infusion was introduced in the twentieth century. The first generation of closed-loop GDFT systems focused on diuresis to control fluid administration in burn patients.5 Once flow-related and dynamic parameters of fluid responsivenessbecame available, perioperative medicine entered a new era of automated GDFT. The closed-loop GDFT system developed by Sironis (Irvine, USA) uses stroke volume increase, a dynamic parameter of fluid responsiveness derived from pulse contour analysis, to guide fluid infusion.6 Studies have shown its feasibility, safety, and impact in simulation, animal, and patient studies.7-10 This system can be guided using semi-invasive and non-invasive monitors and the controller has both closed-loop and open-loop (i.e., decision support) options.9,11,12 The closed-loop option eliminates the need for an anaesthesiologist to intervene, while the open-loop interface adds an extra level of safety by requiring physician confirmation for fluid administration. This system was shown to increase protocol compliance and was associated with improved outcome when compared to fluid therapy guided by static parameters of fluid responsiveness.9
Closed-loop GDFT is a first step in automating GDHT and several groups are now developing new automated systems. Several vasopressor closed-loop systems, for example, are currently under investigation.13,14 There are many developments in both fluid and vasopressor automation and the coming years are sure to shed new light on the potential of GDHT automation.15 Only then will we be able to determine if fully automated closed-loop systems can improve outcome in high-risk surgical patients by consistently optimising cardiac output and perfusion pressure.
- Cannesson M, Ramsingh D, Rinehart J, et al. Crit Care2015;19:261
- Lima MF, Mondadori LA, Chibana AY, et al. J Clin Monit Comput2018:1-7.
- Habicher M, Balzer F, Mezger V, et al. Perioper Med (Lond)2016;5:31.
- Lilot M, Ehrenfeld JM, Lee C, et al. Br J Anaesth2015;114:767-76.
- Bowman RJ, Westenskow DR. IEEE Trans Biomed Eng1981;28:475-9.
- Joosten A, Alexander B, Delaporte A, et al. Anaesthesiol Intensive Ther2015;47:517-23.
- Rinehart J, Chung E, Canales C, et al. J Cardiothorac Vasc Anesth2012;26:933-9.
- Rinehart J, Lee C, Canales C, et al. Anesth Analg2013;117:1119-29.
- Joosten A, Coeckelenbergh S, Delaporte A, et al. Eur J Anaesthesiol2018;35:650-8.
- Joosten A, Delaporte A, Ickx B, et al. Anesthesiology2018;128:55-66.
- Joosten A, Raj Lawrence S, Colesnicenco A, et al. Anesth Analg2018; doi: 10.1213/ANE.0000000000003553
- Joosten A, Hafiane R, Pustetto M, et al. J Clin Monit Comput2019;33:15-24.
- Joosten A, Delaporte A, Alexander B, et al. Anesthesiology2019;130:394-403.
- Rinehart J, Joosten A, Ma M, et al. J Clin Monit Comput2018; doi: 10.1007/s10877-018-1234-0
- Libert N, Chenegros G, Harrois A, et al. Ann Intensive Care2018;8(1):89.