Recently, the FLOW-i anaesthesia ventilator was developed based on the SERVO-i intensive care ventilator. The aim of this study was to test the FLOW-i's tidal volume delivery in the presence of a leak in the breathing circuit. We ventilated a test lung model in volume-, pressure-, and pressure-regulated volume-controlled modes (VC, PC, and PRVC, respectively) with a FLOW-i. First, the circuit remained airtight and the ventilator was tested with fresh gas flows of 6, 1, and 0.3 L/min in VC, PC, and PRVC modes and facing 4 combinations of different resistive and elastic loads. Second, a fixed leak in the breathing circuit was introduced and the measurements repeated. In the airtight system, FLOW-i maintained tidal volume (VT) and circuit pressure at approximately the set values, independently of respiratory mode, load, or fresh gas flow. In the leaking circuit, set VT = 500 mL, FLOW-i delivered higher VTs in PC (about 460 mL) than in VC and PRVC, where VTs were substantially less than 500 mL. Interestingly, VT did not differ appreciably from 6 to 0.3 L/min of fresh air flow among the 3 ventilatory modes. In the absence of leakage, peak inspiratory pressures were similar, while they were 35-45 % smaller in PRVC and VC than in PC mode in the presence of leaks. In conclusion, FLOW-i maintained VT (down to fresh gas flows of 0.3 L/min) to 90 % of its preset value in PC mode, which was 4-5 times greater than in VC or PRVC modes.
Keywords: Air leaks; Low flow anaesthesia; Mechanical loads; Ventilatory modality; Volume reflector.