Patient-Ventilator Dyssynchrony

Quantifiable identification of flow-limited ventilator dyssynchrony with the deformed lung ventilator model

BACKGROUND

Ventilator dyssynchrony (VD) can worsen lung injury and is challenging to detect and quantify due to the complex variability in the dyssynchronous breaths. While machine learning (ML) approaches are useful for automating VD detection from the ventilator waveform data, scalable severity quantification and its association with pathogenesis and ventilator mechanics remain challenging.

OBJECTIVE

We develop a systematic framework to quantify pathophysiological features observed in ventilator waveform signals such that they can be used to create feature-based severity stratification of VD breaths.

METHODS

A mathematical model was developed to represent the pressure and volume waveforms of individual breaths in a feature-based parametric form. Model estimates of respiratory effort strength were used to assess the severity of flow-limited (FL)-VD breaths compared to normal breaths. A total of 93,007 breath waveforms from 13 patients were analyzed.

RESULTS

A novel model-defined continuous severity marker was developed and used to estimate breath phenotypes of FL-VD breaths. The phenotypes had a predictive accuracy of over 97% with respect to the previously developed ML-VD identification algorithm. To understand the incidence of FL-VD breaths and their association with the patient state, these phenotypes were further successfully correlated with ventilator-measured parameters and electronic health records.

CONCLUSION

This work provides a computational pipeline to identify and quantify the severity of FL-VD breaths and paves the way for a large-scale study of VD causes and effects. This approach has direct application to clinical practice and in meaningful knowledge extraction from the ventilator waveform data.

Acute Respiratory Failure in Severe Acute Brain Injury

Acute respiratory failure is commonly encountered in severe acute brain injury due to a multitude of factors related to the sequelae of the primary injury. The interaction between pulmonary and neurologic systems in this population is complex, often with competing priorities. Many treatment modalities for acute respiratory failure can result in deleterious effects on cerebral physiology, and secondary brain injury due to elevations in intracranial pressure or impaired cerebral perfusion. High-quality literature is lacking to guide clinical decision-making in this population, and deliberate considerations of individual patient factors must be considered to optimize each patient's care.

A Novel Application of Spectrograms with Machine Learning Can Detect Patient Ventilator Dyssynchrony

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.

DYSPNEA AND MECHANICAL VENTILATION: APPLYING PHYSIOLOGY TO GUIDE THERAPY

While advances in our understanding of mechanical ventilation have improved mortality from acute respiratory distress syndrome, recent studies indicate a rising incidence of post-ventilation mental health sequelae, including post-traumatic stress disorder (PTSD). Concurrent research on the physiology of dyspnea provides insights about the role of multiple sources of sensory information underlying respiratory discomfort along with the contribution of efferent-afferent dissociation to dyspnea, and the subsequent relationship of dyspnea to a range of affective responses, including fear and anxiety. An understanding of the mechanisms of dyspnea may provide holistic approaches to managing acute respiratory failure that can achieve the best physical and emotional outcomes for patients requiring mechanical ventilation.

Safety and Considerations of the Anaesthetic Management during Bronchoscopic Lung Volume Reduction Treatments

BACKGROUND

Different bronchoscopic lung volume reduction approaches are available for a select group of patients with advanced COPD. General anaesthesia is the recommended method of sedation during these procedures. However, this patient population is at an increased risk of anaesthetic complications, and the best approach to general anaesthesia and mechanical ventilation is unknown.

OBJECTIVES

The aims of this study were to describe the anaesthetic management techniques used during bronchoscopic lung volume reduction procedures and to investigate the number of anaesthesia-related events.

METHODS

Data were retrospectively collected from all endobronchial valve and lung volume reduction coil procedures performed between January 2018 and March 2020 in our hospital. Primary outcomes measures were anaesthetic technique including airway management; ventilation mode and settings; and the incidence of anaesthesia-related events, classified as catastrophic, severe, significant, or moderate.

RESULTS

202 procedures were included. One procedure was performed under procedural sedation, 198 (98%) under general anaesthesia with endotracheal intubation, and 3 (1.5%) under general anaesthesia with laryngeal mask airway. Volume-controlled ventilation was used in 64% of the procedures and pressure-controlled in 36%. Patients were ventilated with a median respiration rate of 9.9 (IQR: 9.6-10.6) breaths per minute, mean tidal volume of 5.8 ± 1.4 mL/kg, and median inspiratory to expiratory (I:E) ratio of 1:2.8 (IQR: 1:2.1-1:3.2). No catastrophic anaesthesia-related events were observed. Hypotension was the most observed anaesthesia-related event.

CONCLUSIONS

Despite the presence of advanced COPD, general anaesthesia and mechanical ventilation are well tolerated by patients undergoing endobronchial valve or lung volume reduction coil treatment. This is presumably strongly linked to the strict selection criteria. Other important considerations are using a low respiratory rate, low tidal volume, and high I:E ratio.

Tracheotomy in ventilator-dependent patients with COVID-19: a cross-sectional study of analgesia and sedative requirements

Objective

During March 2020 in the United States, demand for sedatives increased by 91%, that for analgesics rose by 79%, and demand for neuromuscular blockers increased by 105%, all owing to the number of COVID-19 cases requiring invasive mechanical ventilation (MV). We hypothesize that analgesic and sedative requirements decrease following tracheotomy in this patient population.

Methods

In this cross-sectional study, we conducted a retrospective chart review to identify patients with COVID-19 who underwent tracheotomy (T) at an academic medical center between March 2020 and January 2021. We used a paired Student t-test to compare total oral morphine equivalents (OMEs), total lorazepam equivalents, 24-hour average dexmedetomidine dosage in μg/kg/hour, and 24-hour average propofol dosage in μg/kg/minute on days T−1 and T+2 for each patient.

Results

Of 50 patients, 46 required opioids before and after tracheotomy (mean decrease of 49.4 mg OMEs). Eight patients required benzodiazepine infusion (mean decrease of 45.1 mg lorazepam equivalents. Fifteen patients required dexmedetomidine infusion (mean decrease 0.34 μg/kg/hour). Seventeen patients required propofol (mean decrease 20.5 μg/kg/minute).

Conclusions

When appropriate personal protective equipment is available, use of tracheotomy in patients with COVID-19 who require MV may help to conserve medication supplies in times of extreme shortages.

Reverse Triggering: An Introduction to Diagnosis, Management, and Pharmacologic Implications

Reverse triggering is an underdiagnosed form of patient-ventilator asynchrony in which a passive ventilator-delivered breath triggers a neural response resulting in involuntary patient effort and diaphragmatic contraction. Reverse triggering may significantly impact patient outcomes, and the unique physiology underscores critical potential implications for drug-device-patient interactions. The purpose of this review is to summarize what is known of reverse triggering and its pharmacotherapeutic consequences, with a particular focus on describing reported cases, physiology, historical context, epidemiology, and management. The PubMed database was searched for publications that reported patients presenting with reverse triggering. The current body of evidence suggests that deep sedation may predispose patients to episodes of reverse triggering; as such, providers may consider decreasing sedation or modifying ventilator settings in patients exhibiting ventilator asynchrony as an initial measure. Increased clinician awareness and research focus are necessary to understand appropriate management of reverse triggering and its association with patient outcomes.

Accuracy of Algorithms and Visual Inspection for Detection of Trigger Asynchrony in Critical Patients : A Systematic Review

Objective

This study aimed to summarize the accuracy of the different methods for detecting trigger asynchrony at the bedside in mechanically ventilated patients.

Method

A systematic review was conducted from 1990 to 2020 in PubMed, Lilacs, Scopus, and ScienceDirect databases. The reference list of the identified studies, reviews, and meta-analyses was also manually searched for relevant studies. The reference standards were esophageal pressure catheter and/or electrical activity of the diaphragm. Studies were assessed following the QUADAS-2 recommendations, while the review was prepared according to the PRISMA criteria.

Results

One thousand one hundred and eleven studies were selected, and four were eligible for analysis. Esophageal pressure was the predominant reference standard, while visual inspection and algorithms/software comprised index tests. The trigger asynchrony, ineffective expiratory effort, double triggering, and reverse triggering were analyzed. Sensitivity and specificity ranged from 65.2% to 99% and 80% to 100%, respectively. Positive predictive values reached 80.3 to 100%, while the negative predictive values reached 92 to 100%. Accuracy could not be calculated for most studies.

Conclusion

Algorithms/software validated directly or indirectly using reference standards present high sensitivity and specificity, with a diagnostic power similar to visual inspection of experts.

Patient-Ventilator Dyssynchrony in Critically Ill Patients

Patient–ventilator dyssynchrony is a mismatch between the patient’s respiratory efforts and mechanical ventilator delivery. Dyssynchrony can occur at any phase throughout the respiratory cycle. There are different types of dyssynchrony with different mechanisms and different potential management: trigger dyssynchrony (ineffective efforts, autotriggering, and double triggering); flow dyssynchrony, which happens during the inspiratory phase; and cycling dyssynchrony (premature cycling and delayed cycling). Dyssynchrony has been associated with patient outcomes. Thus, it is important to recognize and address these dyssynchronies at the bedside. Patient–ventilator dyssynchrony can be detected by carefully scrutinizing the airway pressure–time and flow–time waveforms displayed on the ventilator screens along with assessing the patient’s comfort. Clinicians need to know how to depict these dyssynchronies at the bedside. This review aims to define the different types of dyssynchrony and then discuss the evidence for their relationship with patient outcomes and address their potential management.

An interpretable 1D convolutional neural network for detecting patient-ventilator asynchrony in mechanical ventilation

BACKGROUND AND OBJECTIVE

Patient-ventilator asynchrony (PVA) is the result of a mismatch between the need of patients and the assistance provided by the ventilator during mechanical ventilation. Because the poor interaction between the patient and the ventilator is associated with inferior clinical outcomes, effort should be made to identify and correct their occurrence. Deep learning has shown promising ability in PVA detection; however, lack of network interpretability hampers its application in clinic.

METHODS

We proposed an interpretable one-dimensional convolutional neural network (1DCNN) to detect four most manifestation types of PVA (double triggering, ineffective efforts during expiration, premature cycling and delayed cycling) under pressure control ventilation mode and pressure support ventilation mode. A global average pooling (GAP) layer was incorporated with the 1DCNN model to highlight the sections of the respiratory waveform the model focused on when making a classification. Dilation convolution and batch normalization were introduced to the 1DCNN model for compensating the reduction of performance caused by the GAP layer.

RESULTS

The proposed interpretable 1DCNN exhibited comparable performance with the state-of-the-art deep learning model in PVA detection. The F1 scores for the detection of four types of PVA under pressure control ventilation and pressure support ventilation modes were greater than 0.96. The critical sections of the waveform used to detect PVA were highlighted, and found to be well consistent with the understanding of the respective type of PVA by experts.

CONCLUSIONS

The findings suggest that the proposed 1DCNN can help detect PVA, and enhance the interpretability of the classification process to help clinicians better understand the results obtained from deep learning technology.

Patient-ventilator asynchrony in acute brain-injured patients: a prospective observational study

Background

Patient–ventilator asynchrony is common in mechanically ventilated patients and may be related to adverse outcomes. Few studies have reported the occurrence of asynchrony in brain-injured patients. We aimed to investigate the prevalence, type and severity of patient–ventilator asynchrony in mechanically ventilated patients with brain injury.

Methods

This prospective observational study enrolled acute brain-injured patients undergoing mechanical ventilation. Esophageal pressure monitoring was established after enrollment. Flow, airway pressure, and esophageal pressure–time waveforms were recorded for a 15-min interval, four times daily for 3 days, for visually detecting asynchrony by offline analysis. At the end of each dataset recording, the respiratory drive was determined by the airway occlusion maneuver. The asynchrony index was calculated to represent the severity. The relationship between the prevalence and the severity of asynchrony with ventilatory modes and settings, respiratory drive, and analgesia and sedation were determined. Association of severe patient–ventilator asynchrony, which was defined as an asynchrony index  ≥ 10%, with clinical outcomes was analyzed.

Results

In 100 enrolled patients, a total of 1076 15-min waveform datasets covering 330,292 breaths were collected, in which 70,156 (38%) asynchronous breaths were detected. Asynchrony occurred in 96% of patients with the median (interquartile range) asynchrony index of 12.4% (4.3%–26.4%). The most prevalent type was ineffective triggering. No significant difference was found in either prevalence or asynchrony index among different classifications of brain injury (p > 0.05). The prevalence of asynchrony was significantly lower during pressure control/assist ventilation than during other ventilatory modes (p < 0.05). Compared to the datasets without asynchrony, the airway occlusion pressure was significantly lower in datasets with ineffective triggering (p < 0.001). The asynchrony index was significantly higher during the combined use of opioids and sedatives (p < 0.001). Significantly longer duration of ventilation and hospital length of stay after the inclusion were found in patients with severe ineffective triggering (p < 0.05).

Conclusions

Patient–ventilator asynchrony is common in brain-injured patients. The most prevalent type is ineffective triggering and its severity is likely related to a long duration of ventilation and hospital stay. Prevalence and severity of asynchrony are associated with ventilatory modes, respiratory drive and analgesia/sedation strategy, suggesting treatment adjustment in this particular population.

Trial registration The study has been registered on 4 July 2017 in ClinicalTrials.gov (NCT03212482) (https://clinicaltrials.gov/ct2/show/NCT03212482).