The Association Between Ventilator Dyssynchrony, Delivered Tidal Volume, and Sedation Using a Novel Automated Ventilator Dyssynchrony Detection Algorithm

Application of a cloud platform that identifies patient-ventilator asynchrony and enables continuous monitoring of mechanical ventilation in intensive care unit

Background

Patient-ventilator asynchrony (PVA) frequently occurs in mechanically ventilated patients within the ICU and has the potential for harm. Depending solely on the health care team cannot accurately and promptly identify PVA. To address this issue, our team has developed a cloud-based platform for monitoring mechanical ventilation (MV), comprising the PVA-RemoteMonitor system and the 24-h MV analysis report. We conducted a survey to evaluate physicians' satisfaction and acceptance of the platform in 14 ICUs.

Methods

Data from medical records, clinical information systems, and ventilators were uploaded to the cloud platform and underwent data processing. The data were analyzed to monitor PVA and displayed in the front-end. The 24-h analysis report for MV was generated for clinical reference. Critical care physicians in 14 hospitals' ICUs that involved in the platform participated in a questionnaire survey, among whom 10 physicians were interviewed to investigate physicians' acceptance and opinions of this system.

Results

The PVA-RemoteMonitor system exhibited a high level of specificity in detecting flow insufficiency, premature cycle, delayed cycle, reverse trigger, auto trigger, and overshoot, with sensitivities of 90.31 %, 98.76 %, 99.75 %, 99.97 %, 100 %, and 99.69 %, respectively. The 24-h analysis report supplied essential data about PVA and respiratory mechanics. 86.2 % (75/87) of physicians supported the application of this platform.

Conclusions

The PVA-RemoteMonitor system accurately identified PVA, and the MV analysis report provided guidance in controlling PVA. Our platform can effectively assist ICU physicians in the management of ventilated patients.

Patient-ventilator asynchrony classification in mechanically ventilated patients: Model-based or machine learning method?

BACKGROUND AND OBJECTIVE

Patient-ventilator asynchrony (PVA) is associated with poor clinical outcomes and remains under-monitored. Automated PVA detection would enable complete monitoring standard observational methods do not allow. While model-based and machine learning PVA approaches exist, they have variable performance and can miss specific PVA events. This study compares a model and rule-based algorithm with a machine learning PVA method by retrospectively validating both methods using an independent patient cohort.

METHODS

Hysteresis loop analysis (HLA) which is a rule-based method (RBM) and a tri-input convolutional neural network (TCNN) machine learning model are used to classify 7 different types of PVA, including: 1) flow asynchrony; 2) reverse triggering; 3) premature cycling; 4) double triggering; 5) delayed cycling; 6) ineffective efforts; and 7) auto triggering. Class activation mapping (CAM) heatmaps visualise sections of respiratory waveforms the TCNN model uses for decision making, improving result interpretability. Both PVA classification methods were used to classify incidence in an independent retrospective clinical cohort of 11 mechanically ventilated patients for validation and performance comparison.

RESULTS

Self-validation with the training dataset shows overall better HLA performance (accuracy, sensitivity, specificity: 97.5 %, 96.6 %, 98.1 %) compared to the TCNN model (accuracy, sensitivity, specificity: 89.5 %, 98.3 %, 83.9 %). In this study, the TCNN model demonstrates higher sensitivity in detecting PVA, but HLA was better at identifying non-PVA breathing cycles due to its rule-based nature. While the overall AI identified by both classification methods are very similar, the intra-patient distribution of each PVA type varies between HLA and TCNN.

CONCLUSION

The collective findings underscore the efficacy of both HLA and TCNN in PVA detection, indicating the potential for real-time continuous monitoring of PVA. While ML methods such as TCNN demonstrate good PVA identification performance, it is essential to ensure optimal model architecture and diversity in training data before widespread uptake as standard care. Moving forward, further validation and adoption of RBM methods, such as HLA, offers an effective approach to PVA detection while providing clear distinction into the underlying patterns of PVA, better aligning with clinical needs for transparency, explicability, adaptability and reliability of these emerging tools for clinical care.

Neuromuscular Blocking Agent Use in Critical Care Transport Not Associated With Intubation

OBJECTIVE

Variable indications exist for neuromuscular blocking agents (NMBAs) in the critical care transport setting beyond facilitation of intubation.

METHODS

This retrospective cohort study included adult patients (≥ 18 years) who underwent critical care transport from July 1, 2020, to May 2, 2023, and received NMBAs during transport that was not associated with intubation. The primary outcome was the indication for NMBA administration. Secondary outcomes included the characterization of NMBA use, mean Richmond Agitation Sedation Scale score before NMBA administration, sedation strategy used, and continuation of NMBAs within 48 hours of hospital admission.

RESULTS

One hundred twenty-six patients met the inclusion criteria. The most common indication for NMBA administration was ventilator dyssynchrony (n = 71, 56.4%). The majority of patients received rocuronium during transport (n = 113, 89.7%). The mean pre-NMBA Richmond Agitation Sedation Scale score was -3.7 ± 2.4. The most common sedation strategy was a combination of continuous infusion and bolus sedatives (76.2%). One hundred (79.4%) patients had sedation changes in response to NMBA administration. Seventy-two (57.1%) received NMBAs during the first 48 hours of their intensive care unit admission.

CONCLUSION

NMBAs were frequently administered for ventilator dyssynchrony and continuation of prior therapy. Optimization opportunities exist to ensure adequate deep sedation and reassessment of NMBA indication.

Identifying low-dimensional trajectories of mechanically-ventilated patient systems: Empirical phenotypes of joint patient+care processes to enhance temporal analysis in ARDS research

Refined management of mechanically ventilation is an obvious target for improving patient outcomes, but is impeded by the nature of data for study and hypothesis generation. The connections between clinical outcomes and temporal development of iatrogenic injuries current lung-protective ventilator settings remain poorly understood. Analysis of lung-ventilator system (LVS) evolution at relevant timescales is frustrated by data volume and multiple sources of heterogeneity. This work motivates, presents, and validates a computational pipeline for resolving LVS systems into the joint evolution of data-conditioned model parameters and ventilator information. Applied to individuals, the workflow yields a concise low-dimensional representation of LVS behavior expressed in phenotypic breath waveforms suitable for analysis. The effectiveness of this approach is demonstrated through application to multi-day observational series of 35 patients. Individual patient analyses reveal multiple types of patient-oriented dynamics and breath behavior to expose the complexity of LVS evolution; less than 10% of phenotype changes related to ventilator settings changes. Dynamics are shown to including both stable and unstable phenotype transitions as well as both discrete and continuous changes unrelated to ventilator settings. At a cohort scale, 721 phenotypes constructed from individual data are condensed into a set of 16 groups that empirically organize around certain settings (positive end-expository pressure and ventilator mode) and structurally similar pressure-volume loop characterizations. Individual and cohort scale phenotypes, which may be refined by hypothesis-specific constructions, provide a common framework for ongoing temporal analysis and investigation of LVS dynamics.

Driving pressure, as opposed to tidal volume based on predicted body weight, is associated with mortality: results from a prospective cohort of COVID-19 acute respiratory distress syndrome patients

OBJECTIVE

To evaluate the association between driving pressure and tidal volume based on predicted body weight and mortality in a cohort of patients with acute respiratory distress syndrome caused by COVID-19.

METHODS

This was a prospective, observational study that included patients with acute respiratory distress syndrome due to COVID-19 admitted to two intensive care units. We performed multivariable analyses to determine whether driving pressure and tidal volume/kg predicted body weight on the first day of mechanical ventilation, as independent variables, are associated with hospital mortality.

RESULTS

We included 231 patients. The mean age was 64 (53 - 74) years, and the mean Simplified Acute and Physiology Score 3 score was 45 (39 - 54). The hospital mortality rate was 51.9%. Driving pressure was independently associated with hospital mortality (odds ratio 1.21, 95%CI 1.04 - 1.41 for each cm H2O increase in driving pressure, p = 0.01). Based on a double stratification analysis, we found that for the same level of tidal volume/kg predicted body weight, the risk of hospital death increased with increasing driving pressure. However, changes in tidal volume/kg predicted body weight were not associated with mortality when they did not lead to an increase in driving pressure.

CONCLUSION

In patients with acute respiratory distress syndrome caused by COVID-19, exposure to higher driving pressure, as opposed to higher tidal volume/kg predicted body weight, is associated with greater mortality. These results suggest that driving pressure might be a primary target for lung-protective mechanical ventilation in these patients.

Flow-Limited and Reverse-Triggered Ventilator Dyssynchrony Are Associated With Increased Tidal and Dynamic Transpulmonary Pressure

OBJECTIVES

Ventilator dyssynchrony may be associated with increased delivered tidal volumes (V t s) and dynamic transpulmonary pressure (ΔP L,dyn ), surrogate markers of lung stress and strain, despite low V t ventilation. However, it is unknown which types of ventilator dyssynchrony are most likely to increase these metrics or if specific ventilation or sedation strategies can mitigate this potential.

DESIGN

A prospective cohort analysis to delineate the association between ten types of breaths and delivered V t , ΔP L,dyn , and transpulmonary mechanical energy.

SETTING

Patients admitted to the medical ICU.

PATIENTS

Over 580,000 breaths from 35 patients with acute respiratory distress syndrome (ARDS) or ARDS risk factors.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Patients received continuous esophageal manometry. Ventilator dyssynchrony was identified using a machine learning algorithm. Mixed-effect models predicted V t , ΔP L,dyn , and transpulmonary mechanical energy for each type of ventilator dyssynchrony while controlling for repeated measures. Finally, we described how V t , positive end-expiratory pressure (PEEP), and sedation (Richmond Agitation-Sedation Scale) strategies modify ventilator dyssynchrony's association with these surrogate markers of lung stress and strain. Double-triggered breaths were associated with the most significant increase in V t , ΔP L,dyn , and transpulmonary mechanical energy. However, flow-limited, early reverse-triggered, and early ventilator-terminated breaths were also associated with significant increases in V t , ΔP L,dyn , and energy. The potential of a ventilator dyssynchrony type to increase V t , ΔP L,dyn , or energy clustered similarly. Increasing set V t may be associated with a disproportionate increase in high-volume and high-energy ventilation from double-triggered breaths, but PEEP and sedation do not clinically modify the interaction between ventilator dyssynchrony and surrogate markers of lung stress and strain.

CONCLUSIONS

Double-triggered, flow-limited, early reverse-triggered, and early ventilator-terminated breaths are associated with increases in V t , ΔP L,dyn , and energy. As flow-limited breaths are more than twice as common as double-triggered breaths, further work is needed to determine the interaction of ventilator dyssynchrony frequency to cause clinically meaningful changes in patient outcomes.

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.

The Development, Optimization, and Validation of Four Different Machine Learning Algorithms to Identify Ventilator Dyssynchrony

Invasive mechanical ventilation can worsen lung injury. Ventilator dyssynchrony (VD) may propagate ventilator-induced lung injury (VILI) and is challenging to detect and systematically monitor because each patient takes approximately 25,000 breaths a day yet some types of VD are rare, accounting for less than 1% of all breaths. Therefore, we sought to develop and validate accurate machine learning (ML) algorithms to detect multiple types of VD by leveraging esophageal pressure waveform data to quantify patient effort with airway pressure, flow, and volume data generated during mechanical ventilation, building a computational pipeline to facilitate the study of VD.

MATERIALS AND METHODS

We collected ventilator waveform and esophageal pressure data from 30 patients admitted to the ICU. Esophageal pressure allows the measurement of transpulmonary pressure and patient effort. Waveform data were cleaned, features considered essential to VD detection were calculated, and a set of 10,000 breaths were manually labeled. Four ML algorithms were trained to classify each type of VD: logistic regression, support vector classification, random forest, and XGBoost.

RESULTS

We trained ML models to detect different families and seven types of VD with high sensitivity (>90% and >80%, respectively). Three types of VD remained difficult for ML to classify because of their rarity and lack of sample size. XGBoost classified breaths with increased specificity compared to other ML algorithms.

DISCUSSION

We developed ML models to detect multiple types of VD accurately. The ability to accurately detect multiple VD types addresses one of the significant limitations in understanding the role of VD in affecting patient outcomes.

CONCLUSION

ML models identify multiple types of VD by utilizing esophageal pressure data and airway pressure, flow, and volume waveforms. The development of such computational pipelines will facilitate the identification of VD in a scalable fashion, allowing for the systematic study of VD and its impact on patient outcomes.

The oesophageal balloon for respiratory monitoring in ventilated patients: updated clinical review and practical aspects

There is a well-recognised importance for personalising mechanical ventilation settings to protect the lungs and the diaphragm for each individual patient. Measurement of oesophageal pressure (P oes) as an estimate of pleural pressure allows assessment of partitioned respiratory mechanics and quantification of lung stress, which helps our understanding of the patient's respiratory physiology and could guide individualisation of ventilator settings. Oesophageal manometry also allows breathing effort quantification, which could contribute to improving settings during assisted ventilation and mechanical ventilation weaning. In parallel with technological improvements, P oes monitoring is now available for daily clinical practice. This review provides a fundamental understanding of the relevant physiological concepts that can be assessed using P oes measurements, both during spontaneous breathing and mechanical ventilation. We also present a practical approach for implementing oesophageal manometry at the bedside. While more clinical data are awaited to confirm the benefits of P oes-guided mechanical ventilation and to determine optimal targets under different conditions, we discuss potential practical approaches, including positive end-expiratory pressure setting in controlled ventilation and assessment of inspiratory effort during assisted modes.

Reverse triggering neural network and rules-based automated detection in acute respiratory distress syndrome

PURPOSE

Dyssynchrony may cause lung injury and is associated with worse outcomes in mechanically ventilated patients. Reverse triggering (RT) is a common type of dyssynchrony presenting with several phenotypes which may directly cause lung injury and be difficult to identify. Due to these challenges, automated software to assist in identification is needed.

MATERIALS AND METHODS

This was a prospective observational study using a training set of 15 patients and a validation dataset of 13 patients. RT events were manually identified and compared with "rules-based" programs (with and without esophageal manometry and reverse triggering with breath stacking), and were used to train a neural network artificial intelligence (AI) program. RT phenotypes were identified using previously defined rules. Performance of the programs was compared via sensitivity, specificity, positive predictive value (PPV) and F1 score.

RESULTS

33,244 breaths were manually analyzed, with 8718 manually identified as reverse-triggers. The rules-based and AI programs yielded excellent specificity (>95% in all programs) and F1 score (>75% in all programs). RT with breath stacking (24.4%) and mid-cycle RT (37.8%) were the most common phenotypes.

CONCLUSIONS

Automated detection of RT demonstrated good performance, with the potential application of these programs for research and clinical care.

Hypothesis-driven modeling of the human lung-ventilator system: A characterization tool for Acute Respiratory Distress Syndrome research

Mechanical ventilation is an essential tool in the management of Acute Respiratory Distress Syndrome (ARDS), but it exposes patients to the risk of ventilator-induced lung injury (VILI). The human lung-ventilator system (LVS) involves the interaction of complex anatomy with a mechanical apparatus, which limits the ability of process-based models to provide individualized clinical support. This work proposes a hypothesis-driven strategy for LVS modeling in which robust personalization is achieved using a pre-defined parameter basis in a non-physiological model. Model inversion, here via windowed data assimilation, forges observed waveforms into interpretable parameter values that characterize the data rather than quantifying physiological processes. Accurate, model-based inference on human-ventilator data indicates model flexibility and utility over a variety of breath types, including those from dyssynchronous LVSs. Estimated parameters generate static characterizations of the data that are 50%-70% more accurate than breath-wise single-compartment model estimates. They also retain sufficient information to distinguish between the types of breath they represent. However, the fidelity and interpretability of model characterizations are tied to parameter definitions and model resolution. These additional factors must be considered in conjunction with the objectives of specific applications, such as identifying and tracking the development of human VILI.

Delay-induced uncertainty in the glucose-insulin system: Pathogenicity for obesity and type-2 diabetes mellitus

We have recently shown that physiological delay can induce a novel form of sustained temporal chaos we call delay-induced uncertainty (DIU) (Karamched et al. (Chaos, 2021, 31, 023142)). This paper assesses the impact of DIU on the ability of the glucose-insulin system to maintain homeostasis when responding to the ingestion of meals. We address two questions. First, what is the nature of the DIU phenotype? That is, what physiological macrostates (as encoded by physiological parameters) allow for DIU onset? Second, how does DIU impact health? We find that the DIU phenotype is abundant in the space of intrinsic parameters for the Ultradian glucose-insulin model—a model that has been successfully used to predict glucose-insulin dynamics in humans. Configurations of intrinsic parameters that correspond to high characteristic glucose levels facilitate DIU onset. We argue that DIU is pathogenic for obesity and type-2 diabetes mellitus by linking the statistical profile of DIU to the glucostatic theory of hunger.

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.

Artificial intelligence in critically ill diabetic patients: current status and future prospects

Recent years have witnessed increasing numbers of artificial intelligence (AI) based applications and devices being tested and approved for medical care. Diabetes is arguably the most common chronic disorder worldwide and AI is now being used for making an early diagnosis, to predict and diagnose early complications, increase adherence to therapy, and even motivate patients to manage diabetes and maintain glycemic control. However, these AI applications have largely been tested in non-critically ill patients and aid in managing chronic problems. Intensive care units (ICUs) have a dynamic environment generating huge data, which AI can extract and organize simultaneously, thus analysing many variables for diagnostic and/or therapeutic purposes in order to predict outcomes of interest. Even non-diabetic ICU patients are at risk of developing hypo or hyperglycemia, complicating their ICU course and affecting outcomes. In addition, to maintain glycemic control frequent blood sampling and insulin dose adjustments are required, increasing nursing workload and chances of error. AI has the potential to improve glycemic control while reducing the nursing workload and errors. Continuous glucose monitoring (CGM) devices, which are Food and Drug Administration (FDA) approved for use in non-critically ill patients, are now being recommended for use in specific ICU populations with increased accuracy. AI based devices including artificial pancreas and CGM regulated insulin infusion system have shown promise as comprehensive glycemic control solutions in critically ill patients. Even though many of these AI applications have shown potential, these devices need to be tested in larger number of ICU patients, have wider availability, show favorable cost-benefit ratio and be amenable for easy integration into the existing healthcare systems, before they become acceptable to ICU physicians for routine use.

Ignorance Isn't Bliss: We Must Close the Machine Learning Knowledge Gap in Pediatric Critical Care

Pediatric intensivists are bombarded with more patient data than ever before. Integration and interpretation of data from patient monitors and the electronic health record (EHR) can be cognitively expensive in a manner that results in delayed or suboptimal medical decision making and patient harm. Machine learning (ML) can be used to facilitate insights from healthcare data and has been successfully applied to pediatric critical care data with that intent. However, many pediatric critical care medicine (PCCM) trainees and clinicians lack an understanding of foundational ML principles. This presents a major problem for the field. We outline the reasons why in this perspective and provide a roadmap for competency-based ML education for PCCM trainees and other stakeholders.

Discordance Between Respiratory Drive and Sedation Depth in Critically Ill Patients Receiving Mechanical Ventilation

OBJECTIVES

In mechanically ventilated patients, deep sedation is often assumed to induce "respirolysis," that is, lyse spontaneous respiratory effort, whereas light sedation is often assumed to preserve spontaneous effort. This study was conducted to determine validity of these common assumptions, evaluating the association of respiratory drive with sedation depth and ventilator-free days in acute respiratory failure.

DESIGN

Prospective cohort study.

SETTING

Patients were enrolled during 2 month-long periods in 2016-2017 from five ICUs representing medical, surgical, and cardiac specialties at a U.S. academic hospital.

PATIENTS

Eligible patients were critically ill adults receiving invasive ventilation initiated no more than 36 hours before enrollment. Patients with neuromuscular disease compromising respiratory function or expiratory flow limitation were excluded.

INTERVENTIONS

Respiratory drive was measured via P0.1, the change in airway pressure during a 0.1-second airway occlusion at initiation of patient inspiratory effort, every 12 ± 3 hours for 3 days. Sedation depth was evaluated via the Richmond Agitation-Sedation Scale. Analyses evaluated the association of P0.1 with Richmond Agitation-Sedation Scale (primary outcome) and ventilator-free days.

MEASUREMENTS AND MAIN RESULTS

Fifty-six patients undergoing 197 bedside evaluations across five ICUs were included. P0.1 ranged between 0 and 13.3 cm H2O (median [interquartile range], 0.1 cm H2O [0.0-1.3 cm H2O]). P0.1 was not significantly correlated with the Richmond Agitation-Sedation Scale (RSpearman, 0.02; 95% CI, -0.12 to 0.16; p = 0.80). Considering P0.1 terciles (range less than 0.2, 0.2-1.0, and greater than 1.0 cm H2O), patients in the middle tercile had significantly more ventilator-free days than the lowest tercile (incidence rate ratio, 0.78; 95% CI, 0.65-0.93; p < 0.01) or highest tercile (incidence rate ratio, 0.58; 95% CI, 0.48-0.70; p < 0.01).

CONCLUSIONS

Sedation depth is not a reliable marker of respiratory drive during critical illness. Respiratory drive can be low, moderate, or high across the range of routinely targeted sedation depth.

A Damaged-Informed Lung Ventilator Model for Ventilator Waveforms

Motivated by a desire to understand pulmonary physiology, scientists have developed physiological lung models of varying complexity. However, pathophysiology and interactions between human lungs and ventilators, e.g., ventilator-induced lung injury (VILI), present challenges for modeling efforts. This is because the real-world pressure and volume signals may be too complex for simple models to capture, and while complex models tend not to be estimable with clinical data, limiting clinical utility. To address this gap, in this manuscript we developed a new damaged-informed lung ventilator (DILV) model. This approach relies on mathematizing ventilator pressure and volume waveforms, including lung physiology, mechanical ventilation, and their interaction. The model begins with nominal waveforms and adds limited, clinically relevant, hypothesis-driven features to the waveform corresponding to pulmonary pathophysiology, patient-ventilator interaction, and ventilator settings. The DILV model parameters uniquely and reliably recapitulate these features while having enough flexibility to reproduce commonly observed variability in clinical (human) and laboratory (mouse) waveform data. We evaluate the proof-in-principle capabilities of our modeling approach by estimating 399 breaths collected for differently damaged lungs for tightly controlled measurements in mice and uncontrolled human intensive care unit data in the absence and presence of ventilator dyssynchrony. The cumulative value of mean squares error for the DILV model is, on average, ≈12 times less than the single compartment lung model for all the waveforms considered. Moreover, changes in the estimated parameters correctly correlate with known measures of lung physiology, including lung compliance as a baseline evaluation. Our long-term goal is to use the DILV model for clinical monitoring and research studies by providing high fidelity estimates of lung state and sources of VILI with an end goal of improving management of VILI and acute respiratory distress syndrome.

Patient-ventilator asynchrony, impact on clinical outcomes and effectiveness of interventions: a systematic review and meta-analysis

Background

Patient–ventilator asynchrony (PVA) is a common problem in patients undergoing invasive mechanical ventilation (MV) in the intensive care unit (ICU), and may accelerate lung injury and diaphragm mis-contraction. The impact of PVA on clinical outcomes has not been systematically evaluated. Effective interventions (except for closed-loop ventilation) for reducing PVA are not well established.

Methods

We performed a systematic review and meta-analysis to investigate the impact of PVA on clinical outcomes in patients undergoing MV (Part A) and the effectiveness of interventions for patients undergoing MV except for closed-loop ventilation (Part B). We searched the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, ClinicalTrials.gov, and WHO-ICTRP until August 2020. In Part A, we defined asynchrony index (AI) ≥ 10 or ineffective triggering index (ITI) ≥ 10 as high PVA. We compared patients having high PVA with those having low PVA.

Results

Eight studies in Part A and eight trials in Part B fulfilled the eligibility criteria. In Part A, five studies were related to the AI and three studies were related to the ITI. High PVA may be associated with longer duration of mechanical ventilation (mean difference, 5.16 days; 95% confidence interval [CI], 2.38 to 7.94; n = 8; certainty of evidence [CoE], low), higher ICU mortality (odds ratio [OR], 2.73; 95% CI 1.76 to 4.24; n = 6; CoE, low), and higher hospital mortality (OR, 1.94; 95% CI 1.14 to 3.30; n = 5; CoE, low). In Part B, interventions involving MV mode, tidal volume, and pressure-support level were associated with reduced PVA. Sedation protocol, sedation depth, and sedation with dexmedetomidine rather than propofol were also associated with reduced PVA.

Conclusions

PVA may be associated with longer MV duration, higher ICU mortality, and higher hospital mortality. Physicians may consider monitoring PVA and adjusting ventilator settings and sedatives to reduce PVA. Further studies with adjustment for confounding factors are warranted to determine the impact of PVA on clinical outcomes.

Trial registration protocols.io (URL: https://www.protocols.io/view/the-impact-of-patient-ventilator-asynchrony-in-adu-bsqtndwn, 08/27/2020).

Supplementary Information

The online version contains supplementary material available at 10.1186/s40560-021-00565-5.

Effect of Lowering Vt on Mortality in Acute Respiratory Distress Syndrome Varies with Respiratory System Elastance

Rationale: If the risk of ventilator-induced lung injury in acute respiratory distress syndrome (ARDS) is causally determined by driving pressure rather than by Vt, then the effect of ventilation with lower Vt on mortality would be predicted to vary according to respiratory system elastance (Ers). Objectives: To determine whether the mortality benefit of ventilation with lower Vt varies according to Ers. Methods: In a secondary analysis of patients from five randomized trials of lower- versus higher-Vt ventilation strategies in ARDS and acute hypoxemic respiratory failure, the posterior probability of an interaction between the randomized Vt strategy and Ers on 60-day mortality was computed using Bayesian multivariable logistic regression. Measurements and Main Results: Of 1,096 patients available for analysis, 416 (38%) died by Day 60. The posterior probability that the mortality benefit from lower-Vt ventilation strategies varied with Ers was 93% (posterior median interaction odds ratio, 0.80 per cm H₂O/[ml/kg]; 90% credible interval, 0.63-1.02). Ers was classified as low (<2 cm H₂O/[ml/kg], n = 321, 32%), intermediate (2-3 cm H₂O/[ml/kg], n = 475, 46%), and high (>3 cm H₂O/[ml/kg], n = 224, 22%). In these groups, the posterior probabilities of an absolute risk reduction in mortality ≥ 1% were 55%, 82%, and 92%, respectively. The posterior probabilities of an absolute risk reduction ≥ 5% were 29%, 58%, and 82%, respectively. Conclusions: The mortality benefit of ventilation with lower Vt in ARDS varies according to elastance, suggesting that lung-protective ventilation strategies should primarily target driving pressure rather than Vt.

Identifying Patient-Ventilator Asynchrony on a Small Dataset Using Image-Based Transfer Learning

Mechanical ventilation is an essential life-support treatment for patients who cannot breathe independently. Patient-ventilator asynchrony (PVA) occurs when ventilatory support does not match the needs of the patient and is associated with a series of adverse clinical outcomes. Deep learning methods have shown a strong discriminative ability for PVA detection, but they require a large number of annotated data for model training, which hampers their application to this task. We developed a transfer learning architecture based on pretrained convolutional neural networks (CNN) and used it for PVA recognition based on small datasets. The one-dimensional signal was converted to a two-dimensional image, and features were extracted by the CNN using pretrained weights for classification. A partial dropping cross-validation technique was developed to evaluate model performance on small datasets. When using large datasets, the performance of the proposed method was similar to that of non-transfer learning methods. However, when the amount of data was reduced to 1%, the accuracy of transfer learning was approximately 90%, whereas the accuracy of the non-transfer learning was less than 80%. The findings suggest that the proposed transfer learning method can obtain satisfactory accuracies for PVA detection when using small datasets. Such a method can promote the application of deep learning to detect more types of PVA under various ventilation modes.

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.

Reverse Triggering Dyssynchrony 24 h after Initiation of Mechanical Ventilation

BACKGROUND

Reverse triggering is a delayed asynchronous contraction of the diaphragm triggered by passive insufflation by the ventilator in sedated mechanically ventilated patients. The incidence of reverse triggering is unknown. This study aimed at determining the incidence of reverse triggering in critically ill patients under controlled ventilation.

METHODS

In this ancillary study, patients were continuously monitored with a catheter measuring the electrical activity of the diaphragm. A method for automatic detection of reverse triggering using electrical activity of the diaphragm was developed in a derivation sample and validated in a subsequent sample. The authors assessed the predictive value of the software. In 39 recently intubated patients under assist-control ventilation, a 1-h recording obtained 24 h after intubation was used to determine the primary outcome of the study. The authors also compared patients' demographics, sedation depth, ventilation settings, and time to transition to assisted ventilation or extubation according to the median rate of reverse triggering.

RESULTS

The positive and negative predictive value of the software for detecting reverse triggering were 0.74 (95% CI, 0.67 to 0.81) and 0.97 (95% CI, 0.96 to 0.98). Using a threshold of 1 μV of electrical activity to define diaphragm activation, median reverse triggering rate was 8% (range, 0.1 to 75), with 44% (17 of 39) of patients having greater than or equal to 10% of breaths with reverse triggering. Using a threshold of 3 μV, 26% (10 of 39) of patients had greater than or equal to 10% reverse triggering. Patients with more reverse triggering were more likely to progress to an assisted mode or extubation within the following 24 h (12 of 39 [68%]) vs. 7 of 20 [35%]; P = 0.039).

CONCLUSIONS

Reverse triggering detection based on electrical activity of the diaphragm suggests that this asynchrony is highly prevalent at 24 h after intubation under assist-control ventilation. Reverse triggering seems to occur during the transition phase between deep sedation and the onset of patient triggering.

EDITOR’S PERSPECTIVE

Intelligent decision support with machine learning for efficient management of mechanical ventilation in the intensive care unit - A critical overview

BACKGROUND

Effective management of Mechanical Ventilation (MV) is vital for reducing morbidity, mortality, and cost of healthcare.

OBJECTIVE

This study aims to synthesize evidence for effective MV management through Intelligent decision support (IDS) with Machine Learning (ML).

METHOD

Databases that include EBSCO, IEEEXplore, Google Scholar, SCOPUS, and the Web of Science were systematically searched to identify studies on IDS for effective MV management regarding Tidal Volume (TV), asynchrony, weaning, and other outcomes such as the risk of Prolonged Mechanical ventilation (PMV). The quality of the articles identified was assessed with a modified Joanna Briggs Institute (JBI) critical appraisal checklist for cross-sessional research.

RESULTS

A total of 26 articles were identified for the study that has IDS for TV (n = 2, 7.8 %), asynchrony (n = 9, 34.6 %), weaning (n = 12, 46.2 %), and others (n = 3, 11.5 %). It was affirmed that implementing IDS in MV management will enhance seamless ICU patient management following the utilization of various Machine Learning (ML) algorithms in decision support. The studies relied on (n = 14) ML algorithms to predict the TV, asynchrony, weaning, risk of PMV and Positive End-Expiratory Pressure (PEEP) changes of 11-20262 ICU patients records with model inputs ranging from (n = 1) for timeseries analysis of TV to (n = 47) for weaning prediction.

CONCLUSIONS

The small data size, poor study design, and result reporting, with the heterogeneity of techniques used in the various studies, hampered the development of a unified approach for managing MV efficiency in TV monitoring, asynchrony, and weaning predictions. Notwithstanding, the ensemble model was able to predict TV, asynchrony, and weaning to a higher accuracy than the other algorithms.

Future Medical Artificial Intelligence Application Requirements and Expectations of Physicians in German University Hospitals: Web-Based Survey

BACKGROUND

The increasing development of artificial intelligence (AI) systems in medicine driven by researchers and entrepreneurs goes along with enormous expectations for medical care advancement. AI might change the clinical practice of physicians from almost all medical disciplines and in most areas of health care. While expectations for AI in medicine are high, practical implementations of AI for clinical practice are still scarce in Germany. Moreover, physicians' requirements and expectations of AI in medicine and their opinion on the usage of anonymized patient data for clinical and biomedical research have not been investigated widely in German university hospitals.

OBJECTIVE

This study aimed to evaluate physicians' requirements and expectations of AI in medicine and their opinion on the secondary usage of patient data for (bio)medical research (eg, for the development of machine learning algorithms) in university hospitals in Germany.

METHODS

A web-based survey was conducted addressing physicians of all medical disciplines in 8 German university hospitals. Answers were given using Likert scales and general demographic responses. Physicians were asked to participate locally via email in the respective hospitals.

RESULTS

The online survey was completed by 303 physicians (female: 121/303, 39.9%; male: 173/303, 57.1%; no response: 9/303, 3.0%) from a wide range of medical disciplines and work experience levels. Most respondents either had a positive (130/303, 42.9%) or a very positive attitude (82/303, 27.1%) towards AI in medicine. There was a significant association between the personal rating of AI in medicine and the self-reported technical affinity level (H4=48.3, P<.001). A vast majority of physicians expected the future of medicine to be a mix of human and artificial intelligence (273/303, 90.1%) but also requested a scientific evaluation before the routine implementation of AI-based systems (276/303, 91.1%). Physicians were most optimistic that AI applications would identify drug interactions (280/303, 92.4%) to improve patient care substantially but were quite reserved regarding AI-supported diagnosis of psychiatric diseases (62/303, 20.5%). Of the respondents, 82.5% (250/303) agreed that there should be open access to anonymized patient databases for medical and biomedical research.

CONCLUSIONS

Physicians in stationary patient care in German university hospitals show a generally positive attitude towards using most AI applications in medicine. Along with this optimism comes several expectations and hopes that AI will assist physicians in clinical decision making. Especially in fields of medicine where huge amounts of data are processed (eg, imaging procedures in radiology and pathology) or data are collected continuously (eg, cardiology and intensive care medicine), physicians' expectations of AI to substantially improve future patient care are high. In the study, the greatest potential was seen in the application of AI for the identification of drug interactions, assumedly due to the rising complexity of drug administration to polymorbid, polypharmacy patients. However, for the practical usage of AI in health care, regulatory and organizational challenges still have to be mastered.

Can Big Data and Machine Learning Improve Our Understanding of Acute Respiratory Distress Syndrome?

Acute respiratory distress syndrome (ARDS) accounts for 10% of all diagnoses in the Intensive Care Unit, and about 40% of the patients succumb to the disease. Clinical methods alone can result in the under-recognition of this heterogeneous syndrome. The purpose of this study is to evaluate the role that big data and machine learning (ML) have played in understanding the heterogeneity of the disease and the development of various prediction algorithms. Most of the work in the field of ML in ARDS has been in the development of prediction models that have comparable efficacies to that of traditional models. Prediction algorithms have been useful in identifying new variables that may be important to consider in the future, supplementing the unknown information with the help of available noninvasive parameters, as well as predicting mortality. Phenotype identification using an unsupervised ML algorithm has been pivotal in classifying the heterogeneous population into more homogenous classes. Big data generated from ventilators in the form of ventilator waveform analysis and images in the form of radiomics have also been leveraged for the identification of the syndrome and can be incorporated into a clinical decision support system. Although the results are promising, lack of generalizability, "black box" nature of algorithms and concerns about "alarm fatigue" should be addressed for more mainstream adoption of these models.

Automated detection and quantification of reverse triggering effort under mechanical ventilation

BACKGROUND

Reverse triggering (RT) is a dyssynchrony defined by a respiratory muscle contraction following a passive mechanical insufflation. It is potentially harmful for the lung and the diaphragm, but its detection is challenging. Magnitude of effort generated by RT is currently unknown. Our objective was to validate supervised methods for automatic detection of RT using only airway pressure (Paw) and flow. A secondary objective was to describe the magnitude of the efforts generated during RT.

METHODS

We developed algorithms for detection of RT using Paw and flow waveforms. Experts having Paw, flow and esophageal pressure (Pes) assessed automatic detection accuracy by comparison against visual assessment. Muscular pressure (Pmus) was measured from Pes during RT, triggered breaths and ineffective efforts.

RESULTS

Tracings from 20 hypoxemic patients were used (mean age 65 ± 12 years, 65% male, ICU survival 75%). RT was present in 24% of the breaths ranging from 0 (patients paralyzed or in pressure support ventilation) to 93.3%. Automatic detection accuracy was 95.5%: sensitivity 83.1%, specificity 99.4%, positive predictive value 97.6%, negative predictive value 95.0% and kappa index of 0.87. Pmus of RT ranged from 1.3 to 36.8 cmH₂0, with a median of 8.7 cmH₂0. RT with breath stacking had the highest levels of Pmus, and RTs with no breath stacking were of similar magnitude than pressure support breaths.

CONCLUSION

An automated detection tool using airway pressure and flow can diagnose reverse triggering with excellent accuracy. RT generates a median Pmus of 9 cmH₂O with important variability between and within patients.

TRIAL REGISTRATION

BEARDS, NCT03447288.

Delay-induced uncertainty for a paradigmatic glucose-insulin model

Medical practice in the intensive care unit is based on the assumption that physiological systems such as the human glucose-insulin system are predictable. We demonstrate that delay within the glucose-insulin system can induce sustained temporal chaos, rendering the system unpredictable. Specifically, we exhibit such chaos for the ultradian glucose-insulin model. This well-validated, finite-dimensional model represents feedback delay as a three-stage filter. Using the theory of rank one maps from smooth dynamical systems, we precisely explain the nature of the resulting delay-induced uncertainty (DIU). We develop a framework one may use to diagnose DIU in a general oscillatory dynamical system. For infinite-dimensional delay systems, no analog of the theory of rank one maps exists. Nevertheless, we show that the geometric principles encoded in our DIU framework apply to such systems by exhibiting sustained temporal chaos for a linear shear flow. Our results are potentially broadly applicable because delay is ubiquitous throughout mathematical physiology.

Use of Machine Learning to Screen for Acute Respiratory Distress Syndrome Using Raw Ventilator Waveform Data

To develop and characterize a machine learning algorithm to discriminate acute respiratory distress syndrome from other causes of respiratory failure using only ventilator waveform data.

Design

Retrospective, observational cohort study.

Setting

Academic medical center ICU.

Patients

Adults admitted to the ICU requiring invasive mechanical ventilation, including 50 patients with acute respiratory distress syndrome and 50 patients with primary indications for mechanical ventilation other than hypoxemic respiratory failure.

Interventions

None.

Measurements and Main Results

Pressure and flow time series data from mechanical ventilation during the first 24-hours after meeting acute respiratory distress syndrome criteria (or first 24-hr of mechanical ventilation for non-acute respiratory distress syndrome patients) were processed to extract nine physiologic features. A random forest machine learning algorithm was trained to discriminate between the patients with and without acute respiratory distress syndrome. Model performance was assessed using the area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value. Analyses examined performance when the model was trained using data from the first 24 hours and tested using withheld data from either the first 24 hours (24/24 model) or 6 hours (24/6 model). Area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value were 0.88, 0.90, 0.71, 0.77, and 0.90 (24/24); and 0.89, 0.90, 0.75, 0.83, and 0.83 (24/6).

Conclusions

Use of machine learning and physiologic information derived from raw ventilator waveform data may enable acute respiratory distress syndrome screening at early time points after intubation. This approach, combined with traditional diagnostic criteria, could improve timely acute respiratory distress syndrome recognition and enable automated clinical decision support, especially in settings with limited availability of conventional diagnostic tests and electronic health records.

Propofol-Associated Hypertriglyceridemia in Coronavirus Disease 2019 Versus Noncoronavirus Disease 2019 Acute Respiratory Distress Syndrome

Supplemental Digital Content is available in the text.

Objectives:

To characterize the incidence and characteristics of propofol-associated hypertriglyceridemia in coronavirus disease 2019 versus noncoronavirus disease 2019 acute respiratory distress syndrome.

Design:

Single-center prospective, observational cohort study.

Setting:

Medical ICU and regional infectious containment unit.

Patients:

Patients with acute respiratory distress syndrome admitted from April 7, 2020, to May 15, 2020, requiring continuous propofol administration.

Interventions:

None.

Measurements and Main Results:

Of 50 patients enrolled, 54% had coronavirus disease 2019 acute respiratory distress syndrome. Median Acute Physiology and Chronic Health Evaluation II and Sequential Organ Failure Assessment scores were 35.5 (interquartile range, 30.2–41) and 8 (interquartile range, 6–9). Pao₂/Fio₂ ratio was 130.5 (interquartile range, 94.5–193.8). Patients with coronavirus disease 2019-associated acute respiratory distress syndrome experienced a higher rate of hypertriglyceridemia (triglyceride ≥ 500 mg/dL) than noncoronavirus disease 2019-associated acute respiratory distress syndrome (9 [33.3%] vs 1 [4.3%]; p = 0.014). Those with coronavirus disease 2019, compared with those without, received more propofol prior to becoming hypertriglyceridemic (median, 5,436.0 mg [interquartile range, 3,405.5–6,845.5 mg] vs 4,229.0 mg [interquartile range, 2,083.4–4,972.1 mg]; p = 0.027). After adjustment for propofol dose with logistic regression (odds ratio, 5.97; 95% CI, 1.16–59.57; p = 0.031) and propensity score matching (odds ratio, 8.64; 95% CI, 1.27–149.12; p = 0.025), there remained a significant difference in the development of hypertriglyceridemia between coronavirus disease 2019-associated acute respiratory distress syndrome and noncoronavirus disease 2019-associated acute respiratory distress syndrome. There was no difference between groups in time to hypertriglyceridemia (p = 0.063). Serum lipase was not different between those who did or did not develop hypertriglyceridemia (p = 0.545). No patients experienced signs or symptoms of pancreatitis.

Conclusions:

Patients with coronavirus disease 2019 acute respiratory distress syndrome experienced a higher rate of propofol-associated hypertriglyceridemia than noncoronavirus disease 2019 acute respiratory distress syndrome patients, even after accounting for differences in propofol administration.

Myorelaxants in ARDS patients

Neuromuscular blocking agents (NMBAs) inhibit patient-initiated active breath and the risk of high tidal volumes and consequent high transpulmonary pressure swings, and minimize patient/ ventilator asynchrony in acute respiratory distress syndrome (ARDS). Minimization of volutrauma and ventilator-induced lung injury (VILI) results in a lower incidence of barotrauma, improved oxygenation and a decrease in circulating proinflammatory markers. Recent randomized clinical trials did not reveal harmful muscular effects during a short course of NMBAs. The use of NMBAs should be considered during the early phase of severe ARDS for patients to facilitate lung protective ventilation or prone positioning only after optimising mechanical ventilation and sedation. The use of NMBAs should be integrated in a global strategy including the reduction of tidal volume, the rational use of PEEP, prone positioning and the use of a ventilatory mode allowing spontaneous ventilation as soon as possible. Partial neuromuscular blockade should be evaluated in future trials.

Ventilator dyssynchrony - Detection, pathophysiology, and clinical relevance: A Narrative review

Mortality associated with the acute respiratory distress syndrome remains unacceptably high due in part to ventilator-induced lung injury (VILI). Ventilator dyssynchrony is defined as the inappropriate timing and delivery of a mechanical breath in response to patient effort and may cause VILI. Such deleterious patient–ventilator interactions have recently been termed patient self-inflicted lung injury. This narrative review outlines the detection and frequency of several different types of ventilator dyssynchrony, delineates the different mechanisms by which ventilator dyssynchrony may propagate VILI, and reviews the potential clinical impact of ventilator dyssynchrony. Until recently, identifying ventilator dyssynchrony required the manual interpretation of ventilator pressure and flow waveforms. However, computerized interpretation of ventilator waive forms can detect ventilator dyssynchrony with an area under the receiver operating curve of >0.80. Using such algorithms, ventilator dyssynchrony occurs in 3%–34% of all breaths, depending on the patient population. Moreover, two types of ventilator dyssynchrony, double-triggered and flow-limited breaths, are associated with the more frequent delivery of large tidal volumes >10 mL/kg when compared with synchronous breaths (54% [95% confidence interval (CI), 47%–61%] and 11% [95% CI, 7%–15%]) compared with 0.9% (95% CI, 0.0%–1.9%), suggesting a role in propagating VILI. Finally, a recent study associated frequent dyssynchrony-defined as >10% of all breaths-with an increase in hospital mortality (67 vs. 23%, P = 0.04). However, the clinical significance of ventilator dyssynchrony remains an area of active investigation and more research is needed to guide optimal ventilator dyssynchrony management.

Development and validation of a sample entropy-based method to identify complex patient-ventilator interactions during mechanical ventilation

Patient-ventilator asynchronies can be detected by close monitoring of ventilator screens by clinicians or through automated algorithms. However, detecting complex patient-ventilator interactions (CP-VI), consisting of changes in the respiratory rate and/or clusters of asynchronies, is a challenge. Sample Entropy (SE) of airway flow (SE-Flow) and airway pressure (SE-Paw) waveforms obtained from 27 critically ill patients was used to develop and validate an automated algorithm for detecting CP-VI. The algorithm's performance was compared versus the gold standard (the ventilator's waveform recordings for CP-VI were scored visually by three experts; Fleiss' kappa = 0.90 (0.87-0.93)). A repeated holdout cross-validation procedure using the Matthews correlation coefficient (MCC) as a measure of effectiveness was used for optimization of different combinations of SE settings (embedding dimension, m, and tolerance value, r), derived SE features (mean and maximum values), and the thresholds of change (Th) from patient's own baseline SE value. The most accurate results were obtained using the maximum values of SE-Flow (m = 2, r = 0.2, Th = 25%) and SE-Paw (m = 4, r = 0.2, Th = 30%) which report MCCs of 0.85 (0.78-0.86) and 0.78 (0.78-0.85), and accuracies of 0.93 (0.89-0.93) and 0.89 (0.89-0.93), respectively. This approach promises an improvement in the accurate detection of CP-VI, and future study of their clinical implications.

Leveraging IoTs and Machine Learning for Patient Diagnosis and Ventilation Management in the Intensive Care Unit

Future healthcare systems will rely heavily on clinical decision support systems (CDSS) to improve the decision-making processes of clinicians. To explore the design of future CDSS, we developed a research-focused CDSS for the management of patients in the intensive care unit that leverages Internet of Things (IoT) devices capable of collecting streaming physiologic data from ventilators and other medical devices. We then created machine learning (ML) models that could analyze the collected physiologic data to determine if the ventilator was delivering potentially harmful therapy and if a deadly respiratory condition, acute respiratory distress syndrome (ARDS), was present. We also present work to aggregate these models into a mobile application that can provide responsive, real-time alerts of changes in ventilation to providers. As illustrated in the recent COVID-19 pandemic, being able to accurately predict ARDS in newly infected patients can assist in prioritizing care. We show that CDSS may be used to analyze physiologic data for clinical event recognition and automated diagnosis, and we also highlight future research avenues for hospital CDSS.

Artificial Intelligence in the Intensive Care Unit

This article is one of ten reviews selected from the Annual Update in Intensive Care and Emergency Medicine 2020. Other selected articles can be found online at https://www.biomedcentral.com/collections/annualupdate2020. Further information about the Annual Update in Intensive Care and Emergency Medicine is available from http://www.springer.com/series/8901.

Neurally adjusted ventilatory assist vs. pressure support to deliver protective mechanical ventilation in patients with acute respiratory distress syndrome: a randomized crossover trial

Background

Protective mechanical ventilation is recommended for patients with acute respiratory distress syndrome (ARDS), but it usually requires controlled ventilation and sedation. Using neurally adjusted ventilatory assist (NAVA) or pressure support ventilation (PSV) could have additional benefits, including the use of lower sedative doses, improved patient–ventilator interaction and shortened duration of mechanical ventilation. We designed a pilot study to assess the feasibility of keeping tidal volume (V_T) at protective levels with NAVA and PSV in patients with ARDS.

Methods

We conducted a prospective randomized crossover trial in five ICUs from a university hospital in Brazil and included patients with ARDS transitioning from controlled ventilation to partial ventilatory support. NAVA and PSV were applied in random order, for 15 min each, followed by 3 h in NAVA. Flow, peak airway pressure (Paw) and electrical activity of the diaphragm (EAdi) were captured from the ventilator, and a software (Matlab, Mathworks, USA), automatically detected inspiratory efforts and calculated respiratory rate (RR) and V_T. Asynchrony events detection was based on waveform analysis.

Results

We randomized 20 patients, but the protocol was interrupted for five (25%) patients for whom we were unable to maintain V_T below 6.5 mL/kg in PSV due to strong inspiratory efforts and for one patient for whom we could not detect EAdi signal. For the 14 patients who completed the protocol, V_T was 5.8 ± 1.1 mL/kg for NAVA and 5.6 ± 1.0 mL/kg for PSV (p = 0.455) and there were no differences in RR (24 ± 7 for NAVA and 23 ± 7 for PSV, p = 0.661). Paw was greater in NAVA (21 ± 3 cmH₂O) than in PSV (19 ± 3 cmH₂O, p = 0.001). Most patients were under continuous sedation during the study. NAVA reduced triggering delay compared to PSV (p = 0.020) and the median asynchrony Index was 0.7% (0–2.7) in PSV and 0% (0–2.2) in NAVA (p = 0.6835).

Conclusions

It was feasible to keep V_T in protective levels with NAVA and PSV for 75% of the patients. NAVA resulted in similar V_T, RR and Paw compared to PSV. Our findings suggest that partial ventilatory assistance with NAVA and PSV is feasible as a protective ventilation strategy in selected ARDS patients under continuous sedation.

Trial registration ClinicalTrials.gov (NCT01519258). Registered 26 January 2012, https://clinicaltrials.gov/ct2/show/NCT01519258

Double Cycling During Mechanical Ventilation: Frequency, Mechanisms, and Physiologic Implications

OBJECTIVES

Double cycling generates larger than expected tidal volumes that contribute to lung injury. We analyzed the incidence, mechanisms, and physiologic implications of double cycling during volume- and pressure-targeted mechanical ventilation in critically ill patients.

DESIGN

Prospective, observational study.

SETTING

Three general ICUs in Spain.

PATIENTS

Sixty-seven continuously monitored adult patients undergoing volume control-continuous mandatory ventilation with constant flow, volume control-continuous mandatory ventilation with decelerated flow, or pressure control-continuous mandatory mechanical ventilation for longer than 24 hours.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We analyzed 9,251 hours of mechanical ventilation corresponding to 9,694,573 breaths. Double cycling occurred in 0.6%. All patients had double cycling; however, the distribution of double cycling varied over time. The mean percentage (95% CI) of double cycling was higher in pressure control-continuous mandatory ventilation 0.54 (0.34-0.87) than in volume control-continuous mandatory ventilation with constant flow 0.27 (0.19-0.38) or volume control-continuous mandatory ventilation with decelerated flow 0.11 (0.06-0.20). Tidal volume in double-cycled breaths was higher in volume control-continuous mandatory ventilation with constant flow and volume control-continuous mandatory ventilation with decelerated flow than in pressure control-continuous mandatory ventilation. Double-cycled breaths were patient triggered in 65.4% and reverse triggered (diaphragmatic contraction stimulated by a previous passive ventilator breath) in 34.6% of cases; the difference was largest in volume control-continuous mandatory ventilation with decelerated flow (80.7% patient triggered and 19.3% reverse triggered). Peak pressure of the second stacked breath was highest in volume control-continuous mandatory ventilation with constant flow regardless of trigger type. Various physiologic factors, none mutually exclusive, were associated with double cycling.

CONCLUSIONS

Double cycling is uncommon but occurs in all patients. Periods without double cycling alternate with periods with clusters of double cycling. The volume of the stacked breaths can double the set tidal volume in volume control-continuous mandatory ventilation with constant flow. Gas delivery must be tailored to neuroventilatory demand because interdependent ventilator setting-related physiologic factors can contribute to double cycling. One third of double-cycled breaths were reverse triggered, suggesting that repeated respiratory muscle activation after time-initiated ventilator breaths occurs more often than expected.

Effects of sedatives and opioids on trigger and cycling asynchronies throughout mechanical ventilation: an observational study in a large dataset from critically ill patients

Background

In critically ill patients, poor patient-ventilator interaction may worsen outcomes. Although sedatives are often administered to improve comfort and facilitate ventilation, they can be deleterious. Whether opioids improve asynchronies with fewer negative effects is unknown. We hypothesized that opioids alone would improve asynchronies and result in more wakeful patients than sedatives alone or sedatives-plus-opioids.

Methods

This prospective multicenter observational trial enrolled critically ill adults mechanically ventilated (MV) > 24 h. We compared asynchronies and sedation depth in patients receiving sedatives, opioids, or both. We recorded sedation level and doses of sedatives and opioids. BetterCare™ software continuously registered ineffective inspiratory efforts during expiration (IEE), double cycling (DC), and asynchrony index (AI) as well as MV modes. All variables were averaged per day. We used linear mixed-effects models to analyze the relationships between asynchronies, sedation level, and sedative and opioid doses.

Results

In 79 patients, 14,166,469 breaths were recorded during 579 days of MV. Overall asynchronies were not significantly different in days classified as sedatives-only, opioids-only, and sedatives-plus-opioids and were more prevalent in days classified as no-drugs than in those classified as sedatives-plus-opioids, irrespective of the ventilatory mode. Sedative doses were associated with sedation level and with reduced DC (p < 0.0001) in sedatives-only days. However, on days classified as sedatives-plus-opioids, higher sedative doses and deeper sedation had more IEE (p < 0.0001) and higher AI (p = 0.0004). Opioid dosing was inversely associated with overall asynchronies (p < 0.001) without worsening sedation levels into morbid ranges.

Conclusions

Sedatives, whether alone or combined with opioids, do not result in better patient-ventilator interaction than opioids alone, in any ventilatory mode. Higher opioid dose (alone or with sedatives) was associated with lower AI without depressing consciousness. Higher sedative doses administered alone were associated only with less DC.

Trial registration

ClinicalTrial.gov, NCT03451461

Electronic supplementary material

The online version of this article (10.1186/s13054-019-2531-5) contains supplementary material, which is available to authorized users.

Early Neuromuscular Blockade in the Acute Respiratory Distress Syndrome

BACKGROUND

The benefits of early continuous neuromuscular blockade in patients with acute respiratory distress syndrome (ARDS) who are receiving mechanical ventilation remain unclear.

METHODS

We randomly assigned patients with moderate-to-severe ARDS (defined by a ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen of <150 mm Hg with a positive end-expiratory pressure [PEEP] of ≥8 cm of water) to a 48-hour continuous infusion of cisatracurium with concomitant deep sedation (intervention group) or to a usual-care approach without routine neuromuscular blockade and with lighter sedation targets (control group). The same mechanical-ventilation strategies were used in both groups, including a strategy involving a high PEEP. The primary end point was in-hospital death from any cause at 90 days.

RESULTS

The trial was stopped at the second interim analysis for futility. We enrolled 1006 patients early after the onset of moderate-to-severe ARDS (median, 7.6 hours after onset). During the first 48 hours after randomization, 488 of the 501 patients (97.4%) in the intervention group started a continuous infusion of cisatracurium (median duration of infusion, 47.8 hours; median dose, 1807 mg), and 86 of the 505 patients (17.0%) in the control group received a neuromuscular blocking agent (median dose, 38 mg). At 90 days, 213 patients (42.5%) in the intervention group and 216 (42.8%) in the control group had died before hospital discharge (between-group difference, -0.3 percentage points; 95% confidence interval, -6.4 to 5.9; P = 0.93). While in the hospital, patients in the intervention group were less physically active and had more adverse cardiovascular events than patients in the control group. There were no consistent between-group differences in end points assessed at 3, 6, and 12 months.

CONCLUSIONS

Among patients with moderate-to-severe ARDS who were treated with a strategy involving a high PEEP, there was no significant difference in mortality at 90 days between patients who received an early and continuous cisatracurium infusion and those who were treated with a usual-care approach with lighter sedation targets. (Funded by the National Heart, Lung, and Blood Institute; ROSE ClinicalTrials.gov number, NCT02509078.).

Intensive Care Unit Telemedicine in the Era of Big Data, Artificial Intelligence, and Computer Clinical Decision Support Systems

This article examines the history of the telemedicine intensive care unit (tele-ICU), the current state of clinical decision support systems (CDSS) in the tele-ICU, applications of machine learning (ML) algorithms to critical care, and opportunities to integrate ML with tele-ICU CDSS. The enormous quantities of data generated by tele-ICU systems is a major driver in the development of the large, comprehensive, heterogeneous, and granular data sets necessary to develop generalizable ML CDSS algorithms, and deidentification of these data sets expands opportunities for ML CDSS research.

Applying machine learning to continuously monitored physiological data

The use of machine learning (ML) in healthcare has enormous potential for improving disease detection, clinical decision support, and workflow efficiencies. In this commentary, we review published and potential applications for the use of ML for monitoring within the hospital environment. We present use cases as well as several questions regarding the application of ML to the analysis of the vast amount of complex data that clinicians must interpret in the realm of continuous physiological monitoring. ML, especially employed in bidirectional conjunction with electronic health record data, has the potential to extract much more useful information out of this currently under-analyzed data source from a population level. As a data driven entity, ML is dependent on copious, high quality input data so that error can be introduced by low quality data sources. At present, while ML is being studied in hybrid formulations along with static expert systems for monitoring applications, it is not yet actively incorporated in the formal artificial learning sense of an algorithm constantly learning and updating its rules without external intervention. Finally, innovations in monitoring, including those supported by ML, will pose regulatory and medico-legal challenges, as well as questions regarding precisely how to incorporate these features into clinical care and medical education. Rigorous evaluation of ML techniques compared to traditional methods or other AI methods will be required to validate the algorithms developed with consideration of database limitations and potential learning errors. Demonstration of value on processes and outcomes will be necessary to support the use of ML as a feature in monitoring system development: Future research is needed to evaluate all AI based programs before clinical implementation in non-research settings.