Development and implementation of explicit computerized protocols for mechanical ventilation in children

Clinical Challenges in Pediatric Ventilation Liberation: A Meta-Narrative Review

OBJECTIVES

To map the evidence for ventilation liberation practices in pediatric respiratory failure using the Realist And MEta-narrative Evidence Syntheses: Evolving Standards publication standards.

DATA SOURCES

CINAHL, MEDLINE, COCHRANE, and EMBASE. Trial registers included the following: ClinicalTrials.gov, European Union clinical trials register, International Standardized Randomized Controlled Trial Number register.

STUDY SELECTION

Abstracts were screened followed by review of full text. Articles published in English language incorporating a heterogeneous population of both infants and older children were assessed.

DATA EXTRACTION

None.

DATA SYNTHESIS

Weaning can be considered as the process by which positive pressure is decreased and the patient becomes increasingly responsible for generating the energy necessary for effective gas exchange. With the growing use of noninvasive respiratory support, extubation can lie in the middle of the weaning process if some additional positive pressure is used after extubation, while for some extubation may constitute the end of weaning. Testing for extubation readiness is a key component of the weaning process as it allows the critical care practitioner to assess the capability and endurance of the patient's respiratory system to resume unassisted ventilation. Spontaneous breathing trials (SBTs) are often seen as extubation readiness testing (ERT), but the SBT is used to determine if the patient can maintain adequate spontaneous ventilation with minimal ventilatory support, whereas ERT implies the patient is ready for extubation.

CONCLUSIONS

Current literature suggests using a structured approach that includes a daily assessment of patient's readiness to extubate may reduce total ventilation time. Increasing evidence indicates that such daily assessments needs to include SBTs without added pressure support. Measures of elevated load as well as measures of impaired respiratory muscle capacity are independently associated with extubation failure in children, indicating that these should also be assessed as part of ERT.

The role of computer-based clinical decision support systems to deliver protective mechanical ventilation

PURPOSE OF REVIEW

Mechanical ventilation of adults and children with acute respiratory failure necessitates balancing lung and diaphragm protective ventilation. Computerized decision support (CDS) offers advantages in circumstances where complex decisions need to be made to weigh potentially competing risks, depending on the physiologic state of the patient.

RECENT FINDINGS

Significant variability in how ventilator protocols are applied still exists and clinical data show that there continues to be wide variability in ventilator management. We have developed a CDS, which we are currently testing in a Phase II randomized controlled trial. The CDS is called Real-time Effort Driven ventilator management (REDvent). We will describe the rationale and methods for development of CDS for lung and diaphragm protective ventilation, using the REDvent CDS as an exemplar.

SUMMARY

Goals for achieving compliance and physiologic objectives can be met when CDS instructions are simple and explicit, provide the clinician with the underlying rule set, permit acceptable reasons for declining and allow for iterative adjustments.

Using machine learning models to predict oxygen saturation following ventilator support adjustment in critically ill children: A single center pilot study

BACKGROUND

In an intensive care units, experts in mechanical ventilation are not continuously at patient's bedside to adjust ventilation settings and to analyze the impact of these adjustments on gas exchange. The development of clinical decision support systems analyzing patients' data in real time offers an opportunity to fill this gap.

OBJECTIVE

The objective of this study was to determine whether a machine learning predictive model could be trained on a set of clinical data and used to predict transcutaneous hemoglobin oxygen saturation 5 min (5min SpO2) after a ventilator setting change.

DATA SOURCES

Data of mechanically ventilated children admitted between May 2015 and April 2017 were included and extracted from a high-resolution research database. More than 776,727 data rows were obtained from 610 patients, discretized into 3 class labels (< 84%, 85% to 91% and c92% to 100%).

PERFORMANCE METRICS OF PREDICTIVE MODELS

Due to data imbalance, four different data balancing processes were applied. Then, two machine learning models (artificial neural network and Bootstrap aggregation of complex decision trees) were trained and tested on these four different balanced datasets. The best model predicted SpO2 with area under the curves < 0.75.

CONCLUSION

This single center pilot study using machine learning predictive model resulted in an algorithm with poor accuracy. The comparison of machine learning models showed that bagged complex trees was a promising approach. However, there is a need to improve these models before incorporating them into a clinical decision support systems. One potentially solution for improving predictive model, would be to increase the amount of data available to limit over-fitting that is potentially one of the cause for poor classification performances for 2 of the three class labels.

Creating a High-Frequency Electronic Database in the PICU: The Perpetual Patient

OBJECTIVE

Our objective was to construct a prospective high-quality and high-frequency database combining patient therapeutics and clinical variables in real time, automatically fed by the information system and network architecture available through fully electronic charting in our PICU. The purpose of this article is to describe the data acquisition process from bedside to the research electronic database.

DESIGN

Descriptive report and analysis of a prospective database.

SETTING

A 24-bed PICU, medical ICU, surgical ICU, and cardiac ICU in a tertiary care free-standing maternal child health center in Canada.

PATIENTS

All patients less than 18 years old were included at admission to the PICU.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Between May 21, 2015, and December 31, 2016, 1,386 consecutive PICU stays from 1,194 patients were recorded in the database. Data were prospectively collected from admission to discharge, every 5 seconds from monitors and every 30 seconds from mechanical ventilators and infusion pumps. These data were linked to the patient's electronic medical record. The database total volume was 241 GB. The patients' median age was 2.0 years (interquartile range, 0.0-9.0). Data were available for all mechanically ventilated patients (n = 511; recorded duration, 77,678 hr), and respiratory failure was the most frequent reason for admission (n = 360). The complete pharmacologic profile was synched to database for all PICU stays. Following this implementation, a validation phase is in process and several research projects are ongoing using this high-fidelity database.

CONCLUSIONS

Using the existing bedside information system and network architecture of our PICU, we implemented an ongoing high-fidelity prospectively collected electronic database, preventing the continuous loss of scientific information. This offers the opportunity to develop research on clinical decision support systems and computational models of cardiorespiratory physiology for example.

Mechanical Ventilation and Decision Support in Pediatric Intensive Care

Respiratory support is required in most children in the pediatric intensive care unit. Decision-support tools (paper or electronic) have been shown to improve the quality of medical care, reduce errors, and improve outcomes. Computers can assist clinicians by standardizing descriptors and procedures, consistently performing calculations, incorporating complex rules with patient data, and capturing relevant data. This article discusses computer decision-support tools to assist clinicians in making flexible but consistent, evidence-based decisions for equivalent patient states.

Automated Weaning from Mechanical Ventilation after Off-Pump Coronary Artery Bypass Grafting

BACKGROUND

The discontinuation of mechanical ventilation after coronary surgery may prolong and significantly increase the load on intensive care unit personnel. We hypothesized that automated mode using INTELLiVENT-ASV can decrease duration of postoperative mechanical ventilation, reduce workload on medical staff, and provide safe ventilation after off-pump coronary artery bypass grafting (OPCAB). The primary endpoint of our study was to assess the duration of postoperative mechanical ventilation during different modes of weaning from respiratory support (RS) after OPCAB. The secondary endpoint was to assess safety of the automated weaning mode and the number of manual interventions to the ventilator settings during the weaning process in comparison with the protocolized weaning mode.

MATERIALS AND METHODS

Forty adult patients undergoing elective OPCAB were enrolled into a prospective single-center study. Patients were randomized into two groups: automated weaning (n = 20) using INTELLiVENT-ASV mode with quick-wean option; and protocolized weaning (n = 20), using conventional synchronized intermittent mandatory ventilation (SIMV) + pressure support (PS) mode. We assessed the duration of postoperative ventilation, incidence and duration of unacceptable RS, and the load on medical staff. We also performed the retrospective analysis of 102 patients (standard weaning) who were weaned from ventilator with SIMV + PS mode based on physician's experience without prearranged algorithm.

RESULTS AND DISCUSSION

Realization of the automated weaning protocol required change in respiratory settings in 2 patients vs. 7 (5-9) adjustments per patient in the protocolized weaning group. Both incidence and duration of unacceptable RS were reduced significantly by means of the automated weaning approach. The FiO₂ during spontaneous breathing trials was significantly lower in the automated weaning group: 30 (30-35) vs. 40 (40-45) % in the protocolized weaning group (p < 0.01). The average time until tracheal extubation did not differ in the automated weaning and the protocolized weaning groups: 193 (115-309) and 197 (158-253) min, respectively, but increased to 290 (210-411) min in the standard weaning group.

CONCLUSION

The automated weaning system after off-pump coronary surgery might provide postoperative ventilation in a more protective way, reduces the workload on medical staff, and does not prolong the duration of weaning from ventilator. The use of automated or protocolized weaning can reduce the duration of postoperative mechanical ventilation in comparison with non-protocolized weaning based on the physician's decision.

Noninvasive Estimation of Arterial CO2 From End-Tidal CO2 in Mechanically Ventilated Children: The GRAeDIENT Pilot Study

OBJECTIVES

The aim of our pilot study was to develop a model to better predict Paco2 in mechanically ventilated children using noninvasive parameters including volumetric capnography.

DESIGN

Prospective clinical pilot study.

SETTING

Level III PICU.

PATIENTS

Sixty-five mechanically ventilated children.

INTERVENTIONS

None.

MATERIALS AND METHODS

We conducted a prospective clinical pilot study that included all children admitted to the PICU (< 18 yr; weight, > 3 kg; mechanically ventilated, > 6 hr; with an arterial line). A predictive model for PaCO2 was developed using linear multivariable regression. Among the data collected in PICU patients, candidate predictors of PaCO2 were defined by a panel of experts and included end-tidal partial pressure of carbon dioxide, ventilation parameters, and data resulting from the analysis of volumetric capnogram recorded 5 minutes before an arterial blood gas. Children with tidal volume less than 30 mL were excluded because of technical limits.

RESULTS

A total of 65 children (43 boys, 65%) (65 [21-150] mo old) were analyzed. By linear multivariable regression, the best model included the mean airway pressure, end-tidal partial pressure of carbon dioxide, FIO2, and the capnographic index with an R equal to 0.90, p value less than 0.001. After correction, 95% (n = 62) of children had an estimated PaCO2 at ± 5 mm Hg.

CONCLUSION

Our model developed provides an accurate estimation of the PaCO2 using end-tidal CO2 and noninvasive variables. Studies are needed to validate the equation in PICUs.

A novel simulator for mechanical ventilation in newborns: MEchatronic REspiratory System SImulator for Neonatal Applications

Respiratory problems are among the main causes of mortality for preterm newborns with pulmonary diseases; mechanical ventilation provides standard care, but long-term complications are still largely reported. In this framework, continuous medical education is mandatory to correctly manage assistance devices. However, commercially available neonatal respiratory simulators are rarely suitable for representing anatomical and physiological conditions; a step toward high-fidelity simulation, therefore, is essential for nurses and neonatologists to acquire the practice needed without any risk. An innovative multi-compartmental infant respirator simulator based on a five-lobe model was developed to reproduce different physio-pathological conditions in infants and to simulate many different kinds of clinical scenarios. The work consisted of three phases: (1) a theoretical study and modeling phase, (2) a prototyping phase, and (3) testing of the simulation software during training courses. The neonatal pulmonary simulator produced allows the replication and evaluation of different mechanical ventilation modalities in infants suffering from many different kinds of respiratory physio-pathological conditions. In particular, the system provides variable compliances for each lobe in an independent manner and different resistance levels for the airway branches; moreover, it allows the trainer to simulate both autonomous and mechanically assisted respiratory cycles in newborns. The developed and tested simulator is a significant contribution to the field of medical simulation in neonatology, as it makes it possible to choose the best ventilation strategy and to perform fully aware management of ventilation parameters.

Computerized Decision Support Systems for Mechanical Ventilation in Children

Mechanical ventilation is an effective treatment in the ICU but can have significant adverse effects. Approaches from adult research have been adopted in pediatric critical care despite known differences in respiratory physiology and ICU processes. There continues to be considerable variation in how ventilators are managed. Computerized decision support systems implement explicit protocols, and are designed to make mechanical ventilation management safer, more consistent, and more lung protective. Variable results and low or unknown compliance with protocols and CDSS tools have been reported. To date, there has been limited research regarding CDSS for mechanical ventilation in children.

Smart Care™ versus respiratory physiotherapy-driven manual weaning for critically ill adult patients: a randomized controlled trial

Introduction

A recent meta-analysis showed that weaning with SmartCare™ (Dräger, Lübeck, Germany) significantly decreased weaning time in critically ill patients. However, its utility compared with respiratory physiotherapist–protocolized weaning is still a matter of debate. We hypothesized that weaning with SmartCare™ would be as effective as respiratory physiotherapy–driven weaning in critically ill patients.

Methods

Adult critically ill patients mechanically ventilated for more than 24 hours in the adult intensive care unit of the Albert Einstein Hospital, São Paulo, Brazil, were randomly assigned to be weaned either by progressive discontinuation of pressure support ventilation (PSV) with SmartCare™. Demographic data, respiratory function parameters, level of PSV, tidal volume (VT), positive end-expiratory pressure (PEEP), inspired oxygen fraction (FiO₂), peripheral oxygen saturation (SpO₂), end-tidal carbon dioxide concentration (EtCO₂) and airway occlusion pressure at 0.1 second (P_0.1) were recorded at the beginning of the weaning process and before extubation. Mechanical ventilation time, weaning duration and rate of extubation failure were compared.

Results

Seventy patients were enrolled 35 in each group. There was no difference between the two groups concerning age, sex or diagnosis at study entry. There was no difference in maximal inspiratory pressure, maximal expiratory pressure, forced vital capacity or rapid shallow breathing index at the beginning of the weaning trial. PEEP, VT, FiO₂, SpO₂, respiratory rate, EtCO₂ and P_0.1 were similar between the two groups, but PSV was not (median: 8 vs. 10 cmH₂O; p =0.007). When the patients were ready for extubation, PSV (8 vs. 5 cmH₂O; p =0.015) and PEEP (8 vs. 5 cmH₂O; p <0.001) were significantly higher in the respiratory physiotherapy–driven weaning group. Total duration of mechanical ventilation (3.5 [2.0–7.3] days vs. 4.1 [2.7-7.1] days; p =0.467) and extubation failure (2 vs. 2; p =1.00) were similar between the two groups. Weaning duration was shorter in the respiratory physiotherapy–driven weaning group (60 [50–80] minutes vs. 110 [80–130] minutes; p <0.001).

Conclusion

A respiratory physiotherapy–driven weaning protocol can decrease weaning time compared with an automatic system, as it takes into account individual weaning difficulties.

Trial registration

Clinicaltrials.gov Identifier: NCT02122016. Date of Registration: 27 August 2013.

Understanding the global epidemiology of pediatric critical illness: the power, pitfalls, and practicalities of point prevalence studies

OBJECTIVE

The point prevalence methodology is a valuable epidemiological study design that can optimize patient enrollment, prospectively gather individual-level data, and measure practice variability across a large number of geographic regions and healthcare settings. The objective of this article is to review the design, implementation, and analysis of recent point prevalence studies investigating the global epidemiology of pediatric critical illness.

DATA SOURCES

Literature review and primary datasets.

STUDY SELECTION

Multicenter, international point prevalence studies performed in PICUs since 2007.

DATA EXTRACTION

Study topic, number of sites, number of study days, patients screened, prevalence of disease, use of specified therapies, and outcomes.

DATA SYNTHESIS

Since 2007, five-point prevalence studies have been performed on acute lung injury, neurologic disease, thromboprophylaxis, fluid resuscitation, and sepsis in PICUs. These studies were performed in 59-120 sites in 7-28 countries. All studies accounted for seasonal variation in pediatric disease by collecting data over multiple study days. Studies screened up to 6,317 patients and reported data on prevalence and therapeutic variability. Three studies also reported short-term outcomes, a valuable but atypical data element in point prevalence studies. Using these five studies as examples, the advantages and disadvantages and approach to designing, implementing, and analyzing point prevalence studies are reviewed.

CONCLUSIONS

Point prevalence studies in pediatric critical care can efficiently provide valuable insight on the global epidemiology of disease and practice patterns for critically ill children.

Closed loop mechanical ventilation

Mechanical ventilation is a sophisticated technique with very narrow therapeutic ranges i.e. highly efficient and able to keep alive the most severe patients, but with considerable side effects and unwanted complications if not properly and timely used. Computerized protocols, closed loop systems, decision support, all terms which need to be defined, may help making mechanical ventilation safer and more efficient. The present paper will provide a short overview on technical and engineering considerations regarding closed loop controlled ventilation as well as tangible clinical evidences supporting the previous statement.

Simulations for mechanical ventilation in children: review and future prospects

Mechanical ventilation is a very effective therapy, but with many complications. Simulators are used in many fields, including medicine, to enhance safety issues. In the intensive care unit, they are used for teaching cardiorespiratory physiology and ventilation, for testing ventilator performance, for forecasting the effect of ventilatory support, and to determine optimal ventilatory management. They are also used in research and development of clinical decision support systems (CDSSs) and explicit computerized protocols in closed loop. For all those reasons, cardiorespiratory simulators are one of the tools that help to decrease mechanical ventilation duration and complications. This paper describes the different types of simulators described in the literature for physiologic simulation and modeling of the respiratory system, including a new simulator (SimulResp), and proposes a validation process for these simulators.

A pilot prospective study on closed loop controlled ventilation and oxygenation in ventilated children during the weaning phase

Introduction

The present study is a pilot prospective safety evaluation of a new closed loop computerised protocol on ventilation and oxygenation in stable, spontaneously breathing children weighing more than 7 kg, during the weaning phase of mechanical ventilation.

Methods

Mechanically ventilated children ready to start the weaning process were ventilated for five periods of 60 minutes in the following order: pressure support ventilation, adaptive support ventilation (ASV), ASV plus a ventilation controller (ASV-CO₂), ASV-CO₂ plus an oxygenation controller (ASV-CO₂-O₂) and pressure support ventilation again. Based on breath-by-breath analysis, the percentage of time with normal ventilation as defined by a respiratory rate between 10 and 40 breaths/minute, tidal volume > 5 ml/kg predicted body weight and end-tidal CO₂ between 25 and 55 mmHg was determined. The number of manipulations and changes on the ventilator were also recorded.

Results

Fifteen children, median aged 45 months, were investigated. No adverse event and no premature protocol termination were reported. ASV-CO₂ and ASV-CO₂-O₂ kept the patients within normal ventilation for, respectively, 94% (91 to 96%) and 94% (87 to 96%) of the time. The tidal volume, respiratory rate, peak inspiratory airway pressure and minute ventilation were equivalent for all modalities, although there were more automatic setting changes in ASV-CO₂ and ASV-CO₂-O₂. Positive end-expiratory pressure modifications by ASV-CO₂-O₂ require further investigation.

Conclusion

Over the short study period and in this specific population, ASV-CO₂ and ASV-CO₂-O₂ were safe and kept the patient under normal ventilation most of the time. Further research is needed, especially for positive end-expiratory pressure modifications by ASV-CO₂-O₂.

Trial registration

ClinicalTrials.gov: NCT01095406