Expiratory Pause Maneuver to Assess Muscle Pressure During Simulated Assisted Mechanical Ventilation Part 2: Intra- and Inter-Observer Reproducibility

Background:

In Part 1 of our study we demonstrated that simulated breathing effort (ie, inspiratory muscle pressure or ΔPmus) during AMV could be measured accurately by introducing an airway occlusion at end expiration to measure ΔPaw during the subsequent patient-triggered inspiration. In Part 2 of this bench study we examined the variability of Pmus measurements by EPM made by individual clinicians and between clinicians.

Methods:

Spontaneous breathing was simulated using an Ingmar ASL-5000 (Ingmar, Pittsburgh, PA) with a compliance of 80 mL/cm H₂O and resistance of 5 cm H₂O per L/s. The breathing pattern was f of 20 at 3 ΔPmus levels reflecting mild, moderate and high effort (5, 10, 15 cm H₂O) using a 20% pressure rise, 10% hold and 10% decay (inspiratory time or Ti: 1.2s). The test series consisted of 15 breaths/Pmus. A PB-980 ventilator (Medtronics, Minneapolis, MN) was used with 3 modes: continuous positive airway pressure (CPAP: 5 cm H₂O), pressure support ventilation (PSV: Δ15/5 cm H₂O), and volume control ventilation (VCV: tidal volume: 500mL, f: 15, peak flow: 60L/m, Ti: 0.9s, positive end-expiratory pressure: 5 cm H₂O). Trigger sensitivity was set at 3 L/m. EPM measured Pmus by using the ventilator's Negative Inspiratory Force function and releasing the pause-hold as soon as a negative Paw deflection was observed. There were 45 observation periods (ie, 9 Pmus and mode conditions and 5 clinicians). For each test condition intra-individual measurement variability was assessed by the magnitude of the standard deviation (SD) whereas measurement variability between clinicians was assessed by one-way ANOVA and Tukey Kraemer post-test. Alpha was set at 0.05.

Results:

In 37 of 45 (82%) observation periods the individual observer SD ranged from 0 to 0.5 cm H₂O (ie. upper 95% of measurement variation of 1 cm H₂O). In only 2 observation periods (4%) the SD was 1 and 1.3 cm H₂O (upper 95% of measurement variation of 2 and 2.6 cm H₂O) that occurred during VCV. Measurement comparisons between observers although statistically significant were in general clinically insignificant (Table). In only 3 instances were these differences > 1 cm H₂O.

Conclusions:

With a minimum practice period (approximately 5m) clinicians are able to produce highly consistent measurements of Pmus during simulated AMV using the EPM technique.

Mean Pmus Measurements by EPM Technique