Test-bench experimental validation of pressure accuracy in a mechanical threshold device for inspiratory muscle training: The threshold IMT® case.

Inspiratory muscle training (IMT) is widely used in rehabilitation, yet the accuracy of mechanical threshold devices has been only partially characterised. The aim of this study was to evaluate the validity and stability of the pressure delivered by the Threshold IMT® device (Philips Respironics) using independent calibrated instrumentation, with the measurement chain itself validated against a weighted-plunger reference method. Three units were tested on a bench equipped with a calibrated differential pressure transducer (YOKOGAWA EJA110E; full-scale (FS) error ≤0.055%), an ISO 5167-3-referenced differential flow meter (uncertainty <5%), and an industrial fan. Five measurements per nominal pressure level (9, 25, and 41 cmH2O) were collected for each device across a range of inspiratory flow rates. The pressure reported by the bench corresponds to the total differential developed at the patient side, i.e. the load the inspiratory muscles must overcome at the mouth. A two-way ANOVA was used to assess precision and inter-device reproducibility. The measurement chain reproduced the weighted-plunger reference pressures to within 2.5%, confirming its accuracy. The patient-side pressures were systematically lower than the nominal scale values at all three levels, corresponding to approximately 40-50% of the set value (linear regression slope = 0.49; R2 ≈ 0.99). Reproducibility was good (no significant device main effect: F [2,36] = 0.24; p = 0.79), indicating a design-inherent characteristic rather than a manufacturing artefact. The pressure-flow relationship followed a logarithmic curve for each device and threshold level. For the device tested, the nominal scale value did not correspond to the pressure measured at the patient side. These findings have implications for how Threshold IMT® settings are interpreted when designing and reporting IMT protocols.