Introduction: With the current COVID-19 pandemic, concerns have raised regarding the risk for NIV to promote airborne transmission. In case of hospital admission, continuation of therapy in patients undergoing chronic NIV is necessary and several protective circuit configurations have been recommended to reduce the risk of aerosol dissemination. However, all these configurations increase instrumental dead space. We therefore designed this study to evaluate their effects on the tidal volume (VT E) required to preserve stable end-tidal CO 2 partial pressure (P ET CO 2) with constant respiratory rate. Methods: A bench consisting of a test lung connected to an adult-sized mannequin head was set up. The model was ventilated through usual domiciliary configuration (single limb circuit with facial vented mask) which was used as reference. Then, five different circuit configurations including non-vented facial mask with viral/bacterial filter, modification of leak position, and change from single to double-limb circuit were evaluated. For each configuration, pressure support (PS) was gradually increased to reach reference P ET CO 2. Resulting VT E was recorded as primary outcome. Results: Reference P ET CO 2 was 38(0) mmHg, with a PS set at 10 cmH 2 O, resulting in a VT E of 432(2) mL. Compared to reference, all the configurations evaluated required substantial increase in VT E to preserve alveolar ventilation, ranging from +79(2) to +216(1) mL. Conclusions: Modifications of NIV configurations in the context of COVID-19 pandemic result in substantial increase of instrumental dead space. Re-evaluation of treatment efficiency and settings is crucial whenever protective measures influencing NIV equipment are considered.