Goal-Directed Haemodynamic Therapy: Are Closed-Loop Systems the Future?

Goal-Directed Haemodynamic Therapy: Are Closed-Loop Systems the Future?

  • Issue 76

The Young Teaching Recognition Award 2018

Goal-Directed Haemodynamic Therapy: Are Closed-Loop Systems the Future?

Sean Coeckelenbergh, Philippe Van der Linden, Luc Barvais, Edgard Engelman,
Luc Van Obbergh, Joseph Rinehart, Alexandre JoostenErasme

University Hospital, Brussels, Belgium

Goal-directed haemodynamic therapy (GDHT) is an important topic in perioperative medicine. In the 230 million major inpatient surgical procedures performed worldwide each year, 20–40% of inpatients will develop postoperative complications and up to 4% will die before discharge.1-3Although morbidity and mortality vary among centres and countries, the majority of deaths occur in high risk patients.2GDHT may be useful in this surgical population since it has been shown to improve perioperative outcome in moderate to high risk surgery.4-6GDHT is the management of fluids, vasopressors, and inotropes to optimise tissue oxygen delivery.7This strategy can be guided using such monitors as the invasive Swan-Ganz pulmonary artery catheter or the less invasive transoesophageal Doppler. Some of the more recent and popular techniques are based on pulse contour analysis and can be semi-invasive or non-invasive.8Despite its potential to improve perioperative outcome, however, implementation of GDHT remains low, due in part to poor compliance and inter-provider variability in goals and therapy.9,10

Closed-loop systems may be a solution to poor compliance because an automated system that consistently applies GDHT would increase protocol adherence and decrease busywork. Closed-loops are systems that evaluate a variable and, if this variable reaches a predefined threshold, apply an intervention. In medicine, closed-loops require a monitor and a treatment. Several intermediate steps are necessary before creating a functional closed-loop. The first is to automate monitoring and therapy, for example by creating a monitor capable of making the diagnosis and a pump capable of administering the treatment. Once both have been automated, the next step is to create a practitioner-controlled hybrid monitoring/interventional open loop system. Finally, once the open-loop system has been validated, physicians can implement a fully automated closed-loop system.

Closed-loop goal-directed fluid therapy (GDFT) has been successfully implemented.11-13This closed-loop (Sironis, Irvine, CA, USA) interprets the patient’s fluid response status through either non-invasive or semi-invasive pulse contour technology using Flo-Track technology (Edwards, Irvine, CA, USA) and delivers a fluid bolus with a Sapphire infusion pump (Q Core Medical, Netanya, Israel). The system consists of two layers, a model layer that uses information from previous patients to apply a standard baseline intervention and an adaptive layer that evaluates the patient’s fluid response and adapts the infusion threshold to better optimize stroke volume. Investigators validated this system at several levels. In 2011, it was shown to be theoretically feasible during simulated haemorrhage scenarios.11It was then compared in 2015 to manual GDFT and was shown to increase protocol compliance and maintain patients more consistently in a preload independent state.12Then, in 2018, it was shown to be associated with improved patient outcome in a historical before-after study that compared closed-loop GDFT to standard fluid therapy in a GDFT naïve institution. Closed-loop GDFT was associated with improved fluid balance, decreased incidence of postoperative complications, and decreased hospital length of stay.13This system thus increases protocol compliance, consistently applies therapy better than humans, is safe, and adapts to the patient. Like any device, however, the closed-loop GDFT system also has limitations. It takes time to gain approval from regulatory institutions, it will not function if the monitor and/or pump fail, and it does not foresee perioperative events. In addition, it must be used by a physician who is competent in GDFT and who can regularly assess the patient’s fluid status as well as the system’s interventions to guarantee that it functions properly.

Closed-loops may very well be the future of GDHT and several key steps have already been taken towards its automation. In addition to GDFT, vasopressor administration via closed-loop technology has also been validated in silico and in animals (under review) and is currently being evaluated in patients. The research potential is huge and it is up to our generation to develop automated systems. Clearly, closed-loop GDHT has its advantages, but closed-loops should not be limited to haemodynamics. Other goal-directed strategies, such as goal-directed analgesic therapy or goal-directed hypnotic therapy, could be useful in the near future. To do so, it is essential that the medical industry invests in research and development of monitoring, therapies, and closed-loop systems. Ultimately, these systems could be integrated and allow physicians to consistently apply high level of care. The question is not “Will closed-loops replace us?”, but rather “When will closed-loops allow us to practice evidence-based medicine consistently?”.


  1. Weiser TG, Regenbogen SE, Thompson KD, et al. Lancet2008;372(9633):139-44.
  2. Pearse RM, Moreno RP, Bauer P, et al. Lancet2012;380(9847):1059-65.
  3. Ghaferi AA, Birkmeyer JD, Dimick JB. New Engl J Med2009;361(14):1368-75.
  4. Benes J, Giglio M, Brienza N, Michard F. Critical Care2014;18(5):584.
  5. Hamilton MA, Cecconi M, Rhodes A. Anesth Analg2011;112(6):1392-402.
  6. Pearse RM, Harrison DA, MacDonald N, et al. JAMA2014;311(21):2181-90.
  7. Cecconi M, Corredor C, Arulkumaran N, et al. Critical Care(London) 2013;17(2):209.
  8. Vincent JL, Pelosi P, Pearse R, et al. Crit Care2015;19(1):224.
  9. Cannesson M, Pestel G, Ricks C, et al. Crit Care2011;15(4):R197.
  10. Miller TE, Roche AM, Gan TJ. Anesth Analg 2011;112(6):1274-6.
  11. Rinehart J, Alexander B, Le Manach Y, et al.Crit Care2011;15(6):R278.
  12. Rinehart J, Lilot M, Lee C, et al. Crit Care2015;19:94.
  13. Joosten A, Coeckelenbergh S, Delaporte, A et al.Eur J Anaesthesiol2018;35(9):650-8.