RESUMO
Patients in intensive care units are frequently supported by mechanical ventilation. There is increasing awareness of patient-ventilator dyssynchrony (PVD), a mismatch between patient respiratory effort and assistance provided by the ventilator, as a risk factor for infection, narcotic exposure, lung injury, and adverse neurocognitive effects. One of the most injurious consequences of PVD are double cycled (DC) breaths when two breaths are delivered by the ventilator instead of one. Prior efforts to identify PVD have limited efficacy. An automated method to identify PVD, independent of clinician expertise, acumen, or time, would potentially permit early, targeted treatment to avoid further harm. We performed secondary analyses of data from a clinical trial of children with acute respiratory distress syndrome. Waveforms of ventilator flow, airway pressure and esophageal manometry were annotated to identify DC breaths and underlying PVD subtypes. Spectrograms were generated from those waveforms to train Convolutional Neural Network (CNN) models in detecting DC and underlying PVD subtypes: Reverse Trigger (RT) and Inadequate Support (IS). The DC breath detection model yielded AUROC of 0.980, while the multi-target detection model for underlying dyssynchrony yielded AUROC of 0.980 (RT) and 0.976 (IS). When operating at 75% sensitivity, DC breath detection had a number needed to alert (NNA) 1.3 (99% specificity), while underlying PVD had a NNA 1.6 (98.5% specificity) for RT and NNA 4.0 (98.2% specificity) for IS. CNNs using spectrograms of ventilator waveforms can identify DC breaths and detect the underlying PVD for targeted clinical interventions.
RESUMO
Rationale: Reverse triggering (RT) occurs when respiratory effort begins after a mandatory breath is initiated by the ventilator. RT may exacerbate ventilator-induced lung injury and lead to breath stacking.Objectives: We sought to describe the frequency and risk factors for RT among patients with acute respiratory distress syndrome (ARDS) and identify risk factors for breath stacking.Methods: We performed a secondary analysis of physiologic data from children on synchronized intermittent mandatory pressure-controlled ventilation enrolled in a single-center randomized controlled trial for ARDS. When children had a spontaneous effort on esophageal manometry, waveforms were recorded and independently analyzed by two investigators to identify RT.Results: We included 81,990 breaths from 100 patient-days and 36 patients. Overall, 2.46% of breaths were RTs, occurring in 15/36 patients (41.6%). A higher tidal volume and a minimal difference between neural respiratory rate and set ventilator rate were independently associated with RT (P = 0.001) in multivariable modeling. Breath stacking occurred in 534 (26.5%) of 2,017 RT breaths and in 14 (93.3%) of 15 patients with RT. In multivariable modeling, breath stacking was more likely to occur when total airway Δpressure (peak inspiratory pressure - positive end-expiratory pressure [PEEP]) at the time patient effort began, peak inspiratory pressure, PEEP, and Δpressure were lower and when patient effort started well after the ventilator-initiated breath (higher phase angle) (all P < 0.05). Together, these parameters were highly predictive of breath stacking (area under the curve, 0.979).Conclusions: Patients with higher tidal volume who have a set ventilator rate close to their spontaneous respiratory rate are more likely to have RT, which results in breath stacking >25% of the time.Clinical trial registered with ClinicalTrials.gov (NCT03266016).