Objective: It is difficult to predict which mechanically ventilated patients will ultimately require a tracheostomy which further predisposes them to unnecessary spontaneous breathing trials, additional time on the ventilator, increased costs, and further ventilation-related complications such as subglottic stenosis. In this study, we aimed to develop a machine learning tool to predict which patients need a tracheostomy at the onset of admission to the intensive care unit (ICU).

Study Design: Retrospective Cohort Study.

Setting: Multicenter Study of 335 Intensive Care Units between 2014 and 2015.

Methods: The eICU Collaborative Research Database (eICU-CRD) was utilized to obtain the patient cohort. Inclusion criteria included: (1) Age >18 years and (2) ICU admission requiring mechanical ventilation. The primary outcome of interest included tracheostomy assessed via a binary classification model. Models included logistic regression (LR), random forest (RF), and Extreme Gradient Boosting (XGBoost).

Results: Of 38,508 invasively mechanically ventilated patients, 1605 patients underwent a tracheostomy. The XGBoost, RF, and LR models had fair performances at an AUROC 0.794, 0.780, and 0.775 respectively. Limiting the XGBoost model to 20 features out of 331, a minimal reduction in performance was observed with an AUROC of 0.778. Using Shapley Additive Explanations, the top features were an admission diagnosis of pneumonia or sepsis and comorbidity of chronic respiratory failure.

Conclusions: Our machine learning model accurately predicts the probability that a patient will eventually require a tracheostomy upon ICU admission, and upon prospective validation, we have the potential to institute earlier interventions and reduce the complications of prolonged ventilation.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11605030PMC
http://dx.doi.org/10.1002/ohn.919DOI Listing

Publication Analysis

Top Keywords

intensive care
12
machine learning
12
admission intensive
8
mechanically ventilated
8
ventilated patients
8
require tracheostomy
8
icu admission
8
admission
5
tracheostomy
5
predicting tracheostomy
4

Similar Publications

The hemodynamic definitions of pulmonary hypertension consider resistive loading (pulmonary vascular resistance [PVR]), but there are increasing evidence that pulsatile loading (pulmonary artery compliance [PAC]) has functional and prognostic importance. The aims of the present study on patients with left heart disease, were to evaluate a novel echocardiographic right ventricular (RV) afterload score and to investigate its relation to risk of mortality or implantation of a left ventricular assist device. Patients ( = 220) with left ventricular ejection fraction < 50% consecutively referred for heart transplant or heart failure workup underwent echocardiography and right heart catheterization.

View Article and Find Full Text PDF

This study documents the experience of implementing an adaptation of the Hospital-at-Home (HaH) model to alleviate the constraints in available hospital beds and manpower amid a surge in infection rates in Singapore during the Omicron and XBB COVID waves, addressing challenges and proposing insights for scalable implementation. HaH substitutes inpatient hospitalizations by leveraging existing community healthcare services and remote healthcare technologies. This HaH adaptation was designed to be activated in during surges and deactivated when bed and manpower demands stabilize, making it less intensive on hospital resources.

View Article and Find Full Text PDF

Operationalizing the New Global Definition of ARDS: A Retrospective Cohort Study From South Africa.

CHEST Crit Care

December 2024

Division of Pulmonary, Allergy, and Critical Care (G. L. A.), University of Pennsylvania Perelman School of Medicine, Philadelphia, PA; the Division of Pulmonary and Critical Care Medicine (S. M. S.), University of Rochester Medical Center, Rochester, NY; the Department of Anaesthesia and Critical Care (A. R., Z. F., and M. T. D. S.), Greys Hospital, KwaZulu-Natal Department of Health, the Department of Anaesthesia and Critical Care (J. I.), Harry Gwala Regional Hospital, KwaZulu-Natal Department of Health, Pietermaritzburg, the Department of Anaesthesia and Critical Care (R. D. W. and M. T. D. S.), School of Clinical Medicine, University of KwaZulu-Natal, Durban, South Africa; the Faculty Medicine and Pharmacy (R. D. W.), Vrije Universiteit Brussel (VUB), Brussels, Belgium; and the Department of Intensive Care (R. D. W.), John Radcliffe Hospital, Oxford University Trust Hospitals, Oxford, England.

Background: A proposed new global definition of ARDS seeks to update the Berlin definition and account for nonintubated ARDS and ARDS diagnoses in resource-variable settings.

Research Question: How do ARDS epidemiologic characteristics change with operationalizing the new global definition of ARDS in a resource-limited setting?

Study Design And Methods: We performed a real-use retrospective cohort study among adult patients meeting criteria for the Berlin definition of ARDS or the global definition of ARDS at ICU admission in two public hospitals in the KwaZulu-Natal Department of Health, South Africa, from January 2017 through June 2022.

Results: Among 5,760 adults (aged ≥ 18 years) admitted to the ICU, 2,027 patients (35.

View Article and Find Full Text PDF

Introduction: Central nervous system (CNS) tumors are the second most prevalent malignant neoplasms in childhood, with surgical resection as the primary therapeutic approach. The immediate postoperative period following CNS tumor resection requires intensive care to mitigate complications associated with high morbidity and mortality.

Objective: The primary aim of this study is to comprehensively describe the postoperative complications observed in pediatric patients who underwent primary CNS tumor resection and were subsequently admitted to the pediatric intensive care unit (PICU) at Hospital Universitario Fundación Valle del Lili in Colombia.

View Article and Find Full Text PDF

From Prone to Prepared: Airway Management in a Patient With Penetrating Thoracic Trauma.

Cureus

December 2024

Anesthesiology, Unidade Local de Saúde de São José, Lisbon, PRT.

Perioperative and critical care management following penetrating thoracic trauma represents a complex challenge. Those who survive the early trauma approach and reach the hospital alive often remain in critical condition, with cardiocirculatory complications and major pulmonary injuries. Additional difficulty arises from the presence of a weapon , particularly in a dorsal location, which limits patient positioning, and the safe manipulation of both the weapon and the patient.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!