Weaning off mechanical ventilation: much less an art, but not yet a science

How do I wean a patient with acute hypercapnic respiratory failure from noninvasive ventilation?

Noninvasive ventilation (NIV) has been increasingly used for the management of different etiologies of acute hypercapnic respiratory failure (AHRF). Although NIV implementation has been framed well by the guidelines, limited number of studies evaluated the NIV weaning strategies, including a gradual decrease in the level of ventilator support and/or duration of NIV as well as abrupt discontinuation, once respiratory acidosis and distress have resolved. None of the methods have yet been established to be superior to the other in terms of the success rate of weaning and duration of NIV; as well as mortality, length of stay (LOS) in hospital, respiratory ICU (RICU), and ICU.  Patient-derived factors, such as etiology of AHRF, disease severity, history of prior NIV use, and clinical status can help to predict NIV weaning outcome and eventually choose the best method for each individual. In this paper, we have described the strategies for weaning a patient with AHRF from NIV and provided a quick guide for implementation of these data into daily practice based on our experience in and the current scientific evidence.

An artificial intelligence system to predict the optimal timing for mechanical ventilation weaning for intensive care unit patients: A two-stage prediction approach

Background

For the intensivists, accurate assessment of the ideal timing for successful weaning from the mechanical ventilation (MV) in the intensive care unit (ICU) is very challenging.

Purpose

Using artificial intelligence (AI) approach to build two-stage predictive models, namely, the try-weaning stage and weaning MV stage to determine the optimal timing of weaning from MV for ICU intubated patients, and implement into practice for assisting clinical decision making.

Methods

AI and machine learning (ML) technologies were used to establish the predictive models in the stages. Each stage comprised 11 prediction time points with 11 prediction models. Twenty-five features were used for the first-stage models while 20 features were used for the second-stage models. The optimal models for each time point were selected for further practical implementation in a digital dashboard style. Seven machine learning algorithms including Logistic Regression (LR), Random Forest (RF), Support Vector Machines (SVM), K Nearest Neighbor (KNN), lightGBM, XGBoost, and Multilayer Perception (MLP) were used. The electronic medical records of the intubated ICU patients of Chi Mei Medical Center (CMMC) from 2016 to 2019 were included for modeling. Models with the highest area under the receiver operating characteristic curve (AUC) were regarded as optimal models and used to develop the prediction system accordingly.

Results

A total of 5,873 cases were included in machine learning modeling for Stage 1 with the AUCs of optimal models ranging from 0.843 to 0.953. Further, 4,172 cases were included for Stage 2 with the AUCs of optimal models ranging from 0.889 to 0.944. A prediction system (dashboard) with the optimal models of the two stages was developed and deployed in the ICU setting. Respiratory care members expressed high recognition of the AI dashboard assisting ventilator weaning decisions. Also, the impact analysis of with- and without-AI assistance revealed that our AI models could shorten the patients' intubation time by 21 hours, besides gaining the benefit of substantial consistency between these two decision-making strategies.

Conclusion

We noticed that the two-stage AI prediction models could effectively and precisely predict the optimal timing to wean intubated patients in the ICU from ventilator use. This could reduce patient discomfort, improve medical quality, and lower medical costs. This AI-assisted prediction system is beneficial for clinicians to cope with a high demand for ventilators during the COVID-19 pandemic.

Author\'s Response to Diaphragm Evaluation and Lung Ultrasound Score during Weaning

How to cite this article: Amara V, Chaudhuri S. Author's Response to Diaphragm Evaluation and Lung Ultrasound Score during Weaning. Indian J Crit Care Med 2022;26(9):1056–1057.

New integrated weaning indices from mechanical ventilation: A derivation-validation observational multicenter study

Background

To develop ten new integrated weaning indices that can predict the weaning outcome better than the traditional indices.

Methods

This retrospective-prospective derivation-validation observational multicenter clinical trial (Clinical Trial.Gov, NCT 01779297), was conducted on 1,175 adult patients admitted at 9 academic affiliated intensive care units (ICUs; 4 surgical and 5 medical), from Jan 2013 to Dec 2018. All patients, intubated and mechanically ventilated for at least 24 h and ready for weaning were enrolled. The study had two phases: at first, the threshold values of each index that best discriminate between a successful and an unsuccessful weaning outcome was determined among 208 patients in the derivation group. In the second phase, the predictive performance of these values was prospectively tested in 967 patients in the validation group. In the prospective-validation set we used Bayes’ theorem to assess the probability of each test in predicting weaning.

Results

In the prospective validation group, sensitivity, specificity, diagnostic accuracy, positive and negative predictive values, and finally area under the receiver operator characteristic curves and standard errors for each index (ten formulae) were calculated. Statistical values of ten formulae for aforesaid variables were higher than 87% (0.87–0.99).

Conclusion

The new indices can be used for hospitalized patients in intensive care settings for accurate prediction of the weaning outcome.

Association between fluid intake and extubation failure in intensive care unit patients with negative fluid balance: a retrospective observational study

BACKGROUND

Negative fluid balance (NFB) is associated with reduced extubation failure. However, whether achieving more NFB can further improve extubation outcome has not been investigated. This study aimed to investigate whether more NFB and restricted fluid intake were associated with extubation success.

METHODS

We performed a retrospective study of adult patients with mechanical ventilation (MV) admitted to Medical Information Mart for Intensive Care (MIMIC-III) from 2001 to 2012. Patients with duration of MV over 24 hours and NFB within 24 hours before extubation were included for analysis. The primary outcome was extubation failure, defined as reintubation within 72 hours after extubation. Association between fluid balance or fluid intake and extubation outcome were investigated with multivariable logistic models.

RESULTS

A total of 3433 extubation events were recorded. 1803 with NFB were included for the final analysis, of which 201(11.1%) were extubation failure. Compared with slight NFB (- 20 to 0 ml/kg), more NFB were not associated improved extubation outcome. Compared with moderate fluid intake (30 to 60 ml/kg), lower (< 30 ml/kg, OR 0.75, 95% CI [0.54, 1.05], p = 0.088) or higher (> 60 ml/kg, OR 1.63, 95% CI [0.73, 3.35], p = 0.206) fluid intake was not associated with extubation outcome. Duration of MV, chronic obstructive pulmonary disease (COPD), hypercapnia, use of diuretics, and SAPSIIscore were associated with extubation failure.

CONCLUSIONS

More NFB or restricted fluid intake were not associated with reduced extubation failure in patients with NFB. However, for COPD patients, restricted fluid intake was associated with extubation success.

Effect of sequential high-flow nasal cannula oxygen therapy and non-invasive positive-pressure ventilation in patients with difficult weaning from mechanical ventilation after extubation on respiratory mechanics

Background

Patients with difficult weaning who undergo mechanical ventilation are more likely to be at risk of reintubation and the sequential use of oxygen therapy after extubation is a concern for clinicians. Therefore, the aim of the present study was to compare the effects of transnasal high-flow nasal cannula (HFNC) oxygen therapy and non-invasive positive-pressure ventilation (NIV) on respiratory mechanics in patients with difficult weaning.

Methods

The present study was a single-center, retrospective, observational study. Twenty-nine patients with difficult weaning off invasive mechanical ventilation from the Department of Critical Care Medicine, The First Affiliated Hospital of Guangzhou Medical University, from December 2018 to April 2021, were included. Within 48 h after extubation, alternate respiratory support with HFNC and NIV was provided. Relevant indicators were recorded after each support mode had been maintained for at least 60 min. These included esophageal pressure (Pes), gastric pressure (Pga), transdiaphragmatic pressure (Pdi), pressure-time product of Pes (PTPes), pressure-time product of Pga (PTPga), pressure-time product of Pdi (PTPdi), ratio of the PTPdi to the PTPes (PTPdi/PTPes), and ratio of the Pes to the Pdi (Pes/Pdi), diaphragmatic electromyogram (EMGdi), percentage of esophageal pressure coefficient of variation (CVes%),diaphragmatic electromyogram coefficient of variation (CVEMG),inspiratory time (Ti), expiratory time (Te) and respiratory cycle time (Ttot).

Results

Of the 29 patients included, 22 were males and 7 were females [age: 63.97±15.34 years, Acute Physiological and Chronic Health Estimation II (APACHE II) score: 18.00±5.63]. The CVes% and the Pes/Pdi were significantly higher in patients with NIV than HFNC using 40 L/min, CVes%: 9 (−6, 20) vs. −7 (−23, 6) and Pes/Pdi: 0.17 (−0.1, 0.53), vs. −0.12 (−0.43, 0.08) (P<0.05). The remaining indicators were not statistically different.

Conclusions

The sequential NIV and HFNC can be tolerated in patients with such difficult weaning off mechanical ventilation after extubation, and more patients tend to choose HFNC subjectively. Compared with HFNC, NIV reduces the work of adjunctive respiratory muscle, but the patient’s Pes dispersion is high when NIV is used, and it is necessary to pay attention to patient–ventilator coordination in clinical practice. We recommend alternating HFNC and NIV during the sequential respiratory therapy after extubation.

[A critical carol : Being an essay on anemia, suffocation, starvation, and other forms of intensive care, after the manner of Dickens]

Irgendwo in den USA – Einige Tage vor Heiligabend wird der angetrunkene Charlie Cratchit bei dem Versuch, eine Straße zu überqueren, von einem Bus angefahren und schwer verletzt: Rippenserienfraktur, Femur- und Fibulafraktur, Milz- und Pankreaslazeration, Darmrisse. In einem US-amerikanischen Krankenhaus der Maximalversorgung wird er operativ versorgt und anschließend auf die Intensivtherapiestation verlegt und dort kontinuierlich von einem namenlosen, sehr erfahrenen Arzt betreut. Vier Tage vor Heiligabend, erscheint am Patientenbett der Geist des berühmten britischen Physiologen Ernest Henry Starling. Er tritt in einen Dialog mit dem namenlosen Arzt, interessiert sich sehr für den Swan-Ganz-Katheter und verschwindet dann wieder. Die Besuche wiederholen sich in den kommenden 3 Nächten. Einmal kritisiert er Cratchits niedrigen Hämatokrit, beim nächsten Mal zeigt er sich unzufrieden mit der Respiratoreinstellung, und beim letzten Besuch ist er äußerst besorgt über den Ernährungszustand des Patienten. Der namenlose Arzt ist zunächst indigniert über des Geistes Kritik und Belehrungen, erkennt aber, dass darin der Schlüssel zu Cratchits Genesung liegt und handelt letztlich nach seinen Vorschlägen. Mit Erfolg: Nach der vom Geist Starlings angeregten Umstellung der maschinellen Ventilation, Gabe von 3 Erythrozytenkonzentraten und Aufnahme einer parenteralen Ernährung kann Charlie Cratchit am Weihnachtsabend extubiert und am Neujahrstag von der Intensivtherapiestation entlassen werden. In diesem Essay hat Robert Bartlett Charles Dickens’ „Weihnachtsgeschichte“ in die Welt der Intensivmedizin verlegt. Sie soll den Intensivmediziner anregen, therapeutische Interventionen wie maschinelle Ventilation, hämodynamische Interventionen und Gabe von Blutprodukten kritisch zu hinterfragen. Hintergrundinformationen und Kommentare zu den angesprochenen aktuellen Problemen der modernen Intensivmedizin ergänzen den Essay.

Prophylactic noninvasive positive pressure ventilation in the weaning of difficult-weaning tracheotomy patients

Background

Prophylactic noninvasive positive pressure ventilation (NPPV) reduces reintubation in endotracheal intubation patients. However, the efficacy of using the prophylactic NPPV in the weaning of tracheotomy patients is unclear.

Methods

We performed prophylactic NPPV in 11 tracheotomy patients who passed a spontaneous breathing trial (SBT), removed the tracheotomy tube, and closed the incision (intervention group). We matched another 11 tracheotomy patients who also passed an SBT but weaning and removing of tracheotomy tube were managed as conventional methods (control group).

Results

Patients in the control group had reinstitution of mechanical ventilation 36 times after the initial SBT success. Compared with the control group, the interventional group had fewer weaning days (3.0±2.1 vs. 11.3±9.2, P=0.01) from initial SBT success to successful weaning and shorter intensive care unit (ICU) length of stay (11.6±4.2 vs. 20.3±11.6, P=0.03) after initial SBT success. The interventional group had lower nosocomial pneumonia rates after initial SBT success (0/11 vs. 2/11), lower ICU mortality (0/11 vs. 2/11), lower hospital mortality (0/11 vs. 3/11), and higher successful weaning rate (11/11 vs. 8/11), but it didn’t reach significant difference. Also, there was no significant difference between groups in total duration of ventilation (25.5±13.3 vs. 34.7±24.2 days), hospital stay after initial SBT success (24.0±22.3 vs. 37.4±31.3 days), total ICU stay (35.7±15.3 vs. 45.0±29.5 days), and total hospital stay (48.7±33.1 vs. 68.6±52.6 days).

Conclusions

Prophylactic NPPV may be useful to accelerate weaning, and shorten ICU stay after initial SBT success in tracheotomy patients.