RESUMO
We describe a real time, noninvasive method of estimating work of breathing (esophageal balloon not required) during noninvasive pressure support (PS) that uses an artificial neural network (ANN) combined with a leak correction (LC) algorithm, programmed to ignore asynchronous breaths, that corrects for differences in inhaled and exhaled tidal volume (VT) from facemask leaks (WOBANN,LC/min). Validation studies of WOBANN,LC/min were performed. Using a dedicated and popular noninvasive ventilation ventilator (V60, Philips), in vitro studies using PS (5 and 10 cm H2O) at various inspiratory flow rate demands were simulated with a lung model. WOBANN,LC/min was compared with the actual work of breathing, determined under conditions of no facemask leaks and estimated using an ANN (WOBANN/min). Using the same ventilator, an in vivo study of healthy adults (n = 8) receiving combinations of PS (3-10 cm H2O) and expiratory positive airway pressure was done. WOBANN,LC/min was compared with physiologic work of breathing/min (WOBPHYS/min), determined from changes in esophageal pressure and VT applied to a Campbell diagram. For the in vitro studies, WOBANN,LC/min and WOBANN/min ranged from 2.4 to 11.9 J/min and there was an excellent relationship between WOBANN,LC/breath and WOBANN/breath, r = 0.99, r(2) = 0.98 (p < 0.01). There were essentially no differences between WOBANN,LC/min and WOBANN/min. For the in vivo study, WOBANN,LC/min and WOBPHYS/min ranged from 3 to 12 J/min and there was an excellent relationship between WOBANN,LC/breath and WOBPHYS/breath, r = 0.93, r(2) = 0.86 (p < 0.01). An ANN combined with a facemask LC algorithm provides noninvasive and valid estimates of work of breathing during noninvasive PS. WOBANN,LC/min, automatically and continuously estimated, may be useful for assessing inspiratory muscle loads and guiding noninvasive PS settings as in a decision support system to appropriately unload inspiratory muscles.