Contact-free assessments of respiratory rate and volume with load cells under the bed legs in ventilated patients: a prospective exploratory observational study

Development of reliable noncontact unrestrained respiratory monitoring is capable of augmenting the safety of hospitalized patients in the recovery phase. We previously discovered respiratory-related centroid shifts along the long axis of the bed with load cells under the bed legs [bed sensor system (BSS)]. This prospective exploratory observational study examined whether noncontact measurements of respiratory-related tidal centroid shift amplitude (TA-BSS; primary variable) and respiratory rate (RR-BSS; secondary variable) were correlated with tidal volume (TV-PN) and respiratory rate (RR-PN), respectively, measured by pneumotachograph in 14 ICU patients under mechanical ventilation. Among the 10-min average data automatically obtained for a 48-h period, 14 data samples were randomly selected from each patient. Successfully and evenly selected 196 data points for each variable were used for the purpose of this study. A good agreement between TA-BSS and TV-PN (Pearson’s r = 0.669) and an excellent agreement between RR-BSS and RR-PN (r = 0.982) were observed. Estimated minute ventilatory volume [3.86 · TA-BSS · RR-BSS (MV-BSS)] was found to be in very good agreement with true minute volume (MV-PN) (r = 0.836). Although Bland–Altman analysis evidenced accuracy of MV-BSS by a small insignificant fixed bias (−0.02 L/min), a significant proportional bias of MV-BSS (r = −0.664) appeared to produce larger precision (1.9 L/min) of MV-BSS. We conclude that contact-free unconstrained respiratory monitoring with load cells under the bed legs may serve as a new clinical monitoring system, when improved. NEW & NOTEWORTHY We previously discovered that four load cells placed under the bed legs capture a centroid shift during respiration in bedridden human subjects. In 14 ICU patients under mechanical ventilation, this study evidenced that contact-free measurements of respiratory rate, tidal volume, and minute ventilation with the load cells correlated well with those measured by pneumotachograph. Possible clinical usefulness of this approach as a new clinical respiratory monitor is indicated.