BackgroundFew studies investigated the influence of the circuit applied during non-invasive ventilation (NIV) with a total face mask. The aim of this bench study was to evaluate the effects of separated inflow and outflow ports in a total face mask on patient ventilator interaction and performance during NIV through a total face mask. MethodsA mannequin was connected to an active lung simulator. NIV was applied both via a standard total face mask (STFM) with a Y-piece connector for inflow/outflow gases and a modified total face mask (MTFM) with 2 different connectors for inflow and outflow gases. ResultsThe MTFM showed both a significantly better patient-ventilator interaction and a significantly higher performance.The MTFM showed a significantly lower Δtrigger compared to STFM (p<0.01) and shorter value of PTPtrigger during all ventilator setting tested (p < 0.01).Significant differences in PTP 200, PTP 300, and PTP 500 were observed between the MTFM and STFM (p < 0.01) in all conditions tested. Concerning PTP 500 ideal index, in all the conditions tested, the MTFM presented higher values compared to STFM, although those differences were not statistically significant.At both RRsim and ventilator settings tested, the MTFM showed a significantly shorter Delaytrinsp and Delaytrexp compared to STFM (p < 0.01).At both RRsim tested and both ventilator settings, the MTFM showed a significantly longer Timesync compared to STFM (p < 0.01). ConclusionsThe MTFM showed a significantly better patient ventilator interaction and a better ventilator performance, suggesting that this kind of total face mask design should be preferred in clinical practice.