Background and objective: Model-based lung mechanics monitoring can provide clinically useful information for guiding mechanical ventilator treatment in intensive care. However, many methods of measuring lung mechanics are not appropriate for both fully and partially sedated patients, and are unable provide lung mechanics metrics in real-time. This study proposes a novel method of using lung mechanics identified during passive expiration to estimate inspiratory lung mechanics for spontaneously breathing patients. Methods: Relationships between inspiratory and expiratory modeled lung mechanics were identified from clinical data from 4 fully sedated patients. The validity of these relationships were assessed using data from a further 4 spontaneously breathing patients. Results: For the fully sedated patients, a linear relationship was identified between inspiratory and expiratory elastance, with slope 1.04 and intercept 1.66. The r value of this correlation was 0.94. No cohort-wide relationship was determined for airway resistance. Expiratory elastance measurements in spontaneously breathing patients were able to produce reasonable estimates of inspiratory elastance after adjusting for the identified difference between them. Conclusions: This study shows that when conventional methods fail, typically ignored expiratory data may be able to provide clinicians with the information needed about patient condition to guide MV therapy. (C) 2019 Elsevier B.V. All rights reserved.

Monitoring of respiratory mechanics is required for guiding patient-specific mechanical ventilation settings in critical care. Many models of respiratory mechanics perform poorly in the presence of variable patient effort. Typical modelling approaches either attempt to mitigate the effect of the patient effort on the airway pressure waveforms, or attempt to capture the size and shape of the patient effort. This work analyses a range of methods to identify respiratory mechanics in volume controlled ventilation modes when there is patient effort. The models are compared using 4 Datasets, each with a sample of 30 breaths before, and 2–3 minutes after sedation has been administered. The sedation will reduce patient efforts, but the underlying pulmonary mechanical properties are unlikely to change during this short time.Model identified parameters from breathing cycles with patient effort are compared to breathing cycles that do not have patient effort. All models have advantages and disadvantages, so model selection may be specific to the respiratory mechanics application. However, in general, the combined method of iterative interpolative pressure reconstruction, and stacking multiple consecutive breaths together has the best performance over the Dataset. The variability of identified elastance when there is patient effort is the lowest with this method, and there is little systematic offset in identified mechanics when sedation is administered.

Models of respiratory mechanics can be used to titrate patient-specific mechanical ventilation (MV) settings in critical care, but often perform poorly in the presence of patient breathing effort. Respiratory mechanics are conventionally calculated using only inspiratory data. Muscle activity is normally assumed relatively minimal or absent during passive expiration regardless of the presence of inspiratory spontaneous breathing (SB) efforts. Hence, this study assesses whether expiratory lung elastance can be used to estimate inspiratory lung elastance for spontaneously breathing, reverse triggered patients. Clinical data from recruitment manoeuvres in fully sedated patients were used to determine a relationship between inspiratory and expiratory modeled lung elastance. The validity of this relationship was assessed using data recorded pre- and post- sedation from different patients.

Respiratory mechanics of fully sedated patients can be easily estimated as the ventilator has full control of patient’s work of breathing. However, in spontaneously breathing patients or patients whose work of breathing is only partially assisted, respiratory mechanics estimation is much more difficult. This difficulty is caused by un-modelled and variable patient effort introduced into the system. The Time-varying elastance model is a model that can estimate respiratory mechanics of spontaneously breathing patients with use of dynamic elastance. The model has a negative elastance component where measured airway pressure is decreasing, yet the air is flowing into the lungs due to patient effort. In this study, quantification of the negative elastance component is studied. Airway flow, pressure and Electrical Activity of Diaphragm signals from 22 invasively ventilated patients using Pressure Support mode are used for this analysis. Two methods have been used to quantify negative elastance, the zero-crossing method and trapezoidal method. The estimated median values of negative elastance using the zero-crossing method is: -3.289 [Interquartile range (IQR: -4.803~-2.504] cmH2O/L and for trapezoidal method is: -1.899 [IQR: -2.362-1.664] cmH2O/L. The correlation between electrical activity of the diaphragm and negative elastance is very weak as the R values across patients are: -0.0697 [IQR: -0.4972~ -0.0255] for zero-crossing method and 0.0939 [IQR: -0.0293-0.3003] for trapezoidal method. Negative elastance is a conceptual component of the model and can be used to quantify patient demand. However, in this study, quantifying patient effort using negative elastance has little similarity to electrical activity of the diaphragm, as negative elastance appears to capture more than only patient effort. Further research is needed to use this metric to observe patient effort.

Randomised control trials have sought to seek to improve mechanical ventilation treatment. However, few trials to date have shown clinical significance. It is hypothesised that aside from effective treatment, the outcome metrics and sample sizes of the trial also affect the significance, and thus impact trial design. In this study, a Monte-Carlo simulation method was developed and used to investigate several outcome metrics of ventilation treatment, including 1) length of mechanical ventilation (LoMV); 2) Ventilator Free Days (VFD); and 3) LoMV-28, a combination of the other metrics. As these metrics have highly skewed distributions, it also investigated the impact of imposing clinically relevant exclusion criteria on study power to enable better design for significance. Data from invasively ventilated patients from a single intensive care unit were used in this analysis to demonstrate the method. Use of LoMV as an outcome metric required 160 patients/arm to reach 80% power with a clinically expected intervention difference of 25% LoMV if clinically relevant exclusion criteria were applied to the cohort, but 400 patients/arm if they were not. However, only 130 patients/arm would be required for the same statistical significance at the same intervention difference if VFD was used. A Monte-Carlo simulation approach using local cohort data combined with objective patient selection criteria can yield better design of ventilation studies to desired power and significance, with fewer patients per arm than traditional trial design methods, which in turn reduces patient risk. Outcome metrics, such as VFD, should be used when a difference in mortality is also expected between the two cohorts. Finally, the non-parametric approach taken is readily generalisable to a range of trial types where outcome data is similarly skewed.

Current methods to optimise mechanical ventilation involve increasing positive end expiratory pressure (PEEP) in steps to maximize recruitment. If PEEP is too high, overdistension and damage occur. There is thus an inherent risk involved when increasing PEEP. This study predicts dynamic elastance and lung mechanics for higher PEEP using clinically relevant elastance basis functions, capturing distension, recruitment and constant stiffness, in a first order model of lung mechanics. The clinically relevant basis functions were used to fit elastance using a single compartment lung model for 10 patients undergoing recruitment maneuvers, where 2-4 PEEP levels were analysed, and then used to predict the elastance and pressure waveforms for PEEP level increases of 5 and 10 cmH2O. The mean error for the pressure fits from the clinically relevant basis functions was 2.06%. Mean error for pressure predictions with a PEEP level increase of 5 cmH2O was 3.8-5.5%. Mean error for PEEP level increases of 10 cmH2O was slightly higher, between 5.0 and 6.6%. Good pressure fits and predictions show these basis functions accurately fit and predict elastance and thus lung behavior at increased PEEP levels. Each clinically relevant basis function behaved as expected, however improvements to the identifiability of distension would further improve the overall accuracy.

Mechanical ventilation (MV) is widely used in the Neonatal Intensive Care Unit (NICU) for patients suffering from respiratory distress syndrome (RDS). MV treatment is difficult due to intra-patient and inter-patient differences in lung mechanics over time, highlighting the need for patient-specific methods. Model-based methods allow identification of patient-specific lung mechanics which can be used to guide care. The aim of this study is to determine if the single compartment lung model can be used with neonatal MV data to provide more insight into their lung mechanics. Neonatal patient data was collected from published literature, and results were compared to data obtained from previously conducted clinical trials in the adult ICU. The single compartment lung model was found to fit the data very well (model fit error range: 2.2- 6.6%) giving patient-specific elastance and resistance values for each breath. Lung elastance was compared for adults and infants and it was found that infants have similar to 30x stiffer lungs than adults (elastance: 1 - 1.75 cmH(2)O /mL vs. 0.017 - 0.059 cmH(2)O/mL) for similar driving pressures. The ventilated neonatal lung has different lung mechanics to an adult's, suggesting that the lung of a neonate should not be treated as a small adult lung. Further work will validate these results using patient data collected from the NICU. Ultimately, this research will provide more knowledge into neonatal pulmonary mechanics and can be used as the first step towards optimised patient-specific care in the NICU. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

In respiratory mechanics models, resistance is often given less focus than elastance or compliance, and the impact of resistance is often simplified. Respiratory system resistance is likely to have some components that change during a breath. A variable resistance model of respiratory mechanics is presented as an extension of the single compartment model which allows resistance to change linearly with pressure during inspiration. The performance of this variable resistance model is tested against a single compartment, and a two compartment model using two clinical datasets using pressure support ventilation, and volume control ventilation with a combined 29346 breaths. The variable resistance model fits clinical data slightly better than a two compartment model, and much better than a single compartment model. The identified variable part of resistance is mostly positive, with median [IQR] of 0.665 s/L [0.484 0.903] in volume control data and 0.558 s/L [0.438 0.856] in pressure support data. This indicates that resistance increases during inspiration as pressure increases. This result seems counterintuitive, as there is no obvious physiological explanation, and may be due to ventilator artefacts.

The aim of mechanical ventilation (MV) is to provide sufficient breathing support for patients with respiratory failure in the intensive care unit (ICU). However, applying inappropriate ventilation parameters can result in ventilator induced lung injury. To prevent this, respiratory mechanics such as elastance and resistance can be estimated at the bedside to help guide MV parameters using respiratory mechanics models. Different models or methods provide different information and each have their own advantages and disadvantages. In this study, respiratory mechanics of 9 respiratory failure patients were estimated using the simple first order model (FOM) and viscoelastic model (VEM). These patients undergo different respiratory manoeuvres and their estimated respiratory mechanics using these models are studied and compared with a standard clinical method in estimating respiratory mechanics. The results showed that both models were able to capture patient-specific mechanics and responses. The VEM was able to provide higher correlation to the standard clinical method compared to FOM.

Automated therapy of intensive care patients is a development that comprises enourmos economical potential. Algorithms used for those automations become increasingly complex and need to be evaluated and validated during development and certification. Patient sim-ulators are a suitable tool in mimicking the physiological reaction of patients and thus allowing systematical evaluation. This paper is intended to present such a simulator that is software based and allows to simulate an artificially ventilated patient. The simulation results exhibit a physiologically plausible behaviour and are computed in real time or faster.

BACKGROUND Rugby is a highly popular team contact sport associated with high injury rates. Specifically, there is a chance of inducing internal lung injuries as a result of the physical nature of the game. Such injuries are only identified with the use of specific invasive protocols or equipment. This study presents a model-based method to assess respiratory mechanics of N=11 rugby players that underwent a low intensity experimental Mechanical Ventilation (MV) Test before and after a rugby game. METHODS Participants were connected to a ventilator via a facemask and their respiratory mechanics estimated using a time-varying elastance model. RESULTS All participants had a respiratory elastance <10 cmH2O/L with no significant difference observed between pre and postgame respiratory mechanics (P>0.05). Model-based respiratory mechanics estimation has been used widely in the treatment of the critically ill in intensive care. However, the application of a ventilator to assess the respiratory mechanics of healthy human beings is limited. CONCLUSIONS This method adapted from ICU mechanical ventilation can be used to provide insight to respiratory mechanics of healthy participants that can be used as a more precise measure of lung inflammation/injury that avoids invasive procedures. This is the first study to conceptualize the assessment of respiratory mechanics in healthy athletes as a means to monitor postexercise stress and therefore manage recovery.

Mechanical ventilation is a life-saving intervention, which, despite being routinely used in ICUs, poses the risk of causing further damage to the lung tissue if the ventilator is set inappropriately. Medical decision support systems may help in optimizing ventilator settings according to therapy goals given by the clinician. Before using the decision support algorithms in commercially available systems, extensive tests are necessary to ensure patient safety and correct decision making. Model-based patient simulators can assist in evaluating such decision support systems by creating different clinical scenarios. We propose a new Java based patient simulator that implements various models of respiratory mechanics, gas exchange and cardiovascular dynamics to form a complex patient model. The implemented models interact with one another to allow simulation of the ventilators influence on various physiological processes. Model simulations are running in real-time and simulation results can be extracted via multiple interfaces. Each of the implemented models has been validated to exhibit physiologically correct behavior. Results of the combined model system also showed to be physiologically plausible.

Zusammenfassung Die Automatisierung der klinischen Intensivtherapie ist eine Entwicklung mit enormem ökonomischem Potential. Die immer komplexer werdenden Algorithmen für eine solche Automatisierung müssen während der Entwicklung aber auch zur Evaluierung und Validierung systematisch getestet werden. Besonders in der Entwicklungsphase eignen sich hierzu Simulationssysteme, die die physiologischen Reaktionen des Patienten abbilden und eine realitätsnahe Evaluierung ermöglichen. Im Folgenden soll ein solcher Patientensimulator vorgestellt werden, der einen mechanisch beatmeten Patienten abbilden kann. Die Ergebnisse der Patientensimulation zeigen in ihrer Gesamtheit physiologisch plausibles Verhalten und können in Echtzeit oder schneller berechnet werden.

Mechanical ventilation (MV) therapy partially or fully replaces the work of breathing in patients with respiratory failure. Respiratory mechanics during pressure controlled (PC) or pressure support (PS) are often not estimated due to variability induced by patient’s spontaneous breathing effort (SB) or asynchronous events (AEs). Proposed is an algorithm which allows for the improvement of respiratory system mechanics estimation during pressure controlled ventilation. For testing, 10 retrospective airway pressure and flow data samples were obtained from 6 MV patients, with each data sample containing 450-500 breaths. All data samples with AE present experienced a decrease in 5th to 95th range (Range90) and mean absolute deviation (MAD) for the estimated respiratory system elastance after reconstruction. These results suggested improved in respiratory mechanics estimation during pressure controlled ventilation. The median [maximum (max), minimum (min)] decrease in MAD was 29.4

Patient breathing efforts occurring during controlled ventilation causes perturbations in pressure data, which cause erroneous parameter estimation in conventional models of respiratory mechanics. A polynomial model of patient effort can be used to capture breath-specific effort and underlying lung condition. An iterative multiple linear regression is used to identify the model in clinical volume controlled data. The polynomial model has lower fitting error and more stable estimates of respiratory elastance and resistance in the presence of patient effort than the conventional single compartment model. However, the polynomial model can converge to poor parameter estimation when patient efforts occur very early in the breath, or for long duration. The model of patient effort can provide clinical benefits by providing accurate respiratory mechanics estimation and monitoring of breath-to-breath patient effort, which can be used by clinicians to guide treatment.

Mechanical ventilation patients may breathe spontaneously during ventilator supported breaths, altering airway pressure waveforms and hindering identification of true, underlying respiratory mechanics. This study aims to assess and identify respiratory mechanics for breathing cycles masked by spontaneous breathing (SB) effort using a pressure reconstruction method. The performance of the method is compared to parameters identified using a single-compartment model. Data from two patients (N=6305 breaths) experiencing SB and subsequent periods of muscle paralysis without SB were used for analysis. Patients are their own control and are assessed by breath-to-breath variation using coefficient of variation (CV) of respiratory elastance. Pressure reconstruction successfully estimates more consistent respiratory mechanics during SB by reducing CV up to 78% compared to conventional identification (p<0.05). Pressure reconstruction is comparable (p>0.05) to conventional identification during paralysis, and generally performs better as paralysis weakens (p<0.05). Pressure reconstruction provides less-affected pressure waveforms, ameliorating the effect of SB, resulting in more accurate respiratory mechanics identification.

Asynchronous Events (AEs) during mechanical ventilation (MV) result in increased work of breathing and potential poor patient outcomes. Thus, it is important to automate AE detection. In this study, an AE detection method, Automated Logging of Inspiratory and Expiratory Non-synchronized breathing (ALIEN) was developed and compared between standard manual detection in 11 MV patients. A total of 5701 breaths were analyzed (median [IQR]: 500 [469-573] per patient). The Asynchrony Index (AI) was 51% [28-78]%. The AE detection yielded sensitivity of 90.3% and specificity of 88.3%. Automated AE detection methods can potentially provide clinicians with real-time information on patient-ventilator interaction.

The power of a study can be established with estimation of total effective sample size (Ntotal). In this study, the impact of the length of mechanical ventilation (LoMV) distribution shape in intensive care on the estimated Ntotal is investigated. This study provides an overview on the study design involving LoMV, the resulting potential limitations, and the criteria for a ‘successful’ design. Data from mechanical ventilated patients in a single-center intensive care unit were used in this study. The Ntotal was estimated using two methods: 1) Model-based Altman’s nomogram (a standard); and 2) Monte-Carlo simulation. Using Monte-Carlo simulation, a patient selection criteria is imposed to estimate Ntotal from ‘realistic’ patient cohorts. The Altman nomogram shows that the Ntotal to detect a 25% change in LoMV (∆LoMV) at power of 0.8 is ≥1000 patients. For the Monte-Carlo simulation, a Ntotal ≥260 patients is needed to detect similar changes. It is important to consider the LoMV distribution shape and variability, particularly relative to target patient groups who might benefit from the intervention. Assessment of ∆LoMV in response to treatment should be carefully considered to avoid an under-powered studies. The Monte-Carlo simulation combined with objective patient selection provides better design of such studies.

Estimation of effective sample size (N/arm) is important to ensure power to detect significant treatment effects. However, traditional parametric sample size estimations depend upon restrictive assumptions that often do not hold in real data. This study estimates N to detect changes in length of mechanical ventilation (LoMV) using Monte-Carlo simulation (MCS) and mechanical ventilation (MV) data to better simulate the cohort.