Battery Life of Portable Home Ventilators: Effects of Ventilator Settings

Effect of Breathing Frequency

The increase in breathing frequency from 10 to 20 breaths/min (+100%) was associated with a decrease in ventilator BL of 18 ± 11%. Therefore, the decrease in ventilator BL is not inversely proportional to the increase in breathing frequency. The ventilators tested in this study use turbines that are also powered during expiration although at a lower speed. Accordingly, the battery drain is more important during inspiration but not limited to it, meaning that ventilator BL is not solely dependent on the number of inspiratory cycles.

Effects of PEEP and Mode of Mechanical Ventilation

Surprisingly, PEEP did not affect ventilator BL. This may be explained by the low level of PEEP tested in our study considering that, 12 y ago, a study demonstrated that PEEP 20 cm H₂O had important effects on BL in electrically driven portable ventilators.⁷ Nevertheless, the use of PEEP at 20 cm H₂O is now less common than at 5 cm H₂O. It is also probable that the new generation of home ventilators requires less power for operating the solenoid dedicated to driving the exhalation valve compared with the older ones. In addition, when analyzing the individual data, we observed that 2 of the 5 ventilators had a longer BL with the addition of PEEP. We checked these results by duplicating the PEEP condition with these 2 ventilators and observed similar results. One possible explanation could be a lower decrease in turbine speed during expiration in order to meet the PEEP level, which may reduce the energy cost of turbine acceleration during inspiration.

VC-CMV Versus PSV

Changing the VC-CMV mode for a PSV mode did not affect ventilator BL. This contrasts with the data reported by Campbell et al,⁷ who observed a decrease in ventilator BL with PSV because of the need for a higher flow delivery during early inspiration with this mode. In fact, 3 ventilators had a constant flow during VC-CMV (Elisée 150, PB560, and Trilogy 100), whereas the other two (Vivo 50 and Monnal T50) presented pressure and flow wave contours during VC-CMV that were very similar to those during PSV. When comparing differences in BL and in pressure and flow wave contours during VC-CMV and PSV for each ventilator, there seemed to be no effect between various pressure and flow wave contours on ventilator BL when the same volume was delivered during a similar inspiratory time.

Effects of Leaks During VC-CMV and PSV

The increase in ventilator BL during leaks with VC-CMV may be explained by the decrease in impedance imposed on the ventilator when the circuit is partially opened to the atmosphere. On the other hand, the decrease in ventilator BL during leaks with PSV may be explained by the increase in turbine speed to reach the target pressure despite the opening of the circuit. In addition, leaks induce an increase in the inspiratory time because, during PSV, the flow cycle occurs when the inspiratory flow falls below a set percentage of peak inspiratory flow. It has been demonstrated that flow cycle criteria are less easily attained during leaks. Indeed, leaks may prolong inspiratory time.⁵

Limitations and Clinical Implications

Because a short BL could increase the risk of power failure and because we have recently lost a home-ventilated patient for this unique reason, we believe that our bench evaluation is clinically relevant. Our experimental condition was ideal considering that it represents a subject with a normal respiratory system impedance. Unfortunately, we did not test the effects of decreasing compliance and increasing resistance of the simulated respiratory system. However, because leak occurrence (which reduces the impedance imposed on the ventilator) increased BL during VC-CMV, we can speculate that, when impedance increases, BL decreases when delivering the same V_T at the same frequency. Similarly, because the increase in respiratory system impedance during PSV should reduce the delivered V_T, it would be necessary to increase the level of PSV to obtain the desired V_T, which reduces BL.

It is difficult to state that the manufacturers underestimate BL (Table 1) considering that this duration can depend on the patient's characteristics (mechanical properties of the respiratory system, patient's respiratory muscle activity, and patient's minute ventilation need) and on the ventilator settings, which were different considering that manufacturers prefer to use the pressure-targeted mode, whereas our baseline condition was VC-CMV.⁸ Nevertheless, even if we can understand the difference between our results and the information on BL delivered by the manufacturers, the BL delivered by the manufacturers is not representative of the differences that we observed between ventilators considering that, for example, the ventilator that had the highest BL in our bench study, for each condition, was the penultimate one when comparing the company's reported duration (Table 1). This suggests that all manufacturers should delineate the conditions of mechanical ventilation used when measuring BL.

Finally, our study cannot give to prescribers an idea of BL considering that this can change with the patient's characteristics, but it does provide information that could be used at the bedside. Our results suggest that for the same V_T adjustment and in the absence of leaks, there is no difference in BL between VC-CMV and PSV. However, in the presence of leaks, BL increases systematically during VC-CMV and decreases during PSV. This result reinforces the preference of VC-CMV over PSV when using mouthpiece ventilation during the daytime.⁹ Our results also suggest that for all ventilators, the BL (for all models) varies significantly from that reported by the manufacturers. Therefore, clinicians should not rely on the BL delivered by the manufacturer. Moreover, if a low level of PEEP is required, preference should be given to the ventilators that improve BL. Interestingly, if PEEP is used to improve the speech of patients with tracheotomies, the activation of a key-pinch switch under the patient's control may be an option for the future¹⁰ so that PEEP is used only when needed (ie, for speech).