Driven by industry demands and the need for anesthesia that consumes fewer resources ("green anesthesia"), manufacturers of anesthesia ventilators have developed closed-circuit ventilators. In a closed-circuit system, during the exhalation phase, gases are not released outside the ventilator into the atmosphere but are injected into the inspiratory limb of the circuit after being warmed and having CO2 removed. It is a closed loop. However, the loop is not completely closed because a very small stream of air is additionally injected into the inspiratory limb to limit the risk of hypoxia. This small stream of air is called "Fresh Gas Flow" or FGF.

To optimize this closed-loop recirculation, manufacturers have designed machines capable of minimizing the fresh gas flow. In the 2000s, machines operated with FGFs of 2 L/min, dropping to flows as low as 0.5 L/min by 2010. This low-flow regime allows for an 80% reduction in halogenated gas (sevoflurane) consumption while maintaining the same efficiency. Recently, even more sophisticated machines have been designed with FGFs of 0.3 L/min. Some of these new machines will be commercially available on a large scale in 2026. As part of an industrial partnership, the Nîmes University Hospital has validated the post-CE marking of these machines for certain manufacturers (including General Electric); the post-CE marking allows the manufacturer to have access to usage data Given that this product has only recently been introduced to the market, very little data has been published on the use of sevoflurane in daily practice at very low FGF flow rates (0.3 to 0.5 L/min).

For the past four months, the Nîmes University Hospital has had two CE-marked machines (GE, CS 850) capable of ventilating at flow rates of 0.3 L/min. These machines are connected to software (Carestation Insight) that allows all anonymized ventilator data to be recorded independently of the users. This software has been in continuous use in 12 operating rooms since 2024 within our institution and has been the subject of numerous publications (see Ref. 19). All ventilation and sevoflurane data are continuously recorded and stored in a secure, anonymous cloud (Ref. 19).

The objective of this retrospective study is to analyze the database of sevoflurane consumption derived from the Carestation Insight software for patients who underwent general anesthesia with ventilation at FGF flow rates of 0.3 L/min and to compare these data with our database of patients under anesthesia at flow rates of 0.5 to 2 L/min, also derived from this same cloud.

Inclusion Criteria:

• Adult patients undergoing surgery under general anesthesia with mechanical ventilation

  • Connected to a ventilator linked to the Carestaion software
  • No refusal as indicated by a letter of consent

Exclusion Criteria:

• Refusal to participate
• Minor patients

Patients who underwent general anesthesia with ventilation at a minute volume of 0.3 L/min at the Nîmes University Hospital