Editor’s Choice Articles for March

March 2024 and another month of amazing content in Pediatric Critical Care Medicine (PCCM). Please take the time to read my three Editor’s Choice articles, each with editorials. First is an article about prognostic modeling in critically ill children in a low- and middle-income (LMIC) PICU in Cambodia (1,2). The second is a single-center analysis of noninvasive neurally adjusted ventilatory assist (NIV-NAVA) in infants with bronchiolitis (3,4). The third is a two-center PICU study about a machine learning model designed to improve the conventional clinical criteria to predict need for intubation in the PICU (5,6). WHAT IS THE BEST RISK STRATIFICATION MODEL FOR CHILDREN ADMITTED TO A PICU IN CAMBODIA? Chandna A, Keang S, Vorlark M, et al: A Prognostic Model for Critically Ill Children in Locations With Emerging Critical Care Capacity (1). My first editor’s choice article from Cambodia used a dataset of over 1,300 children (1,500 admission) in a PICU, 2018 to 2020. There were close to 100 deaths, and the authors examined the performance of nine existing severity of illness mortality prediction scores, and then derived their own prediction model for their resource constrained setting. The accompanying editorial provides an international perspective with a commentary on the various risk-prediction models available and what the study adds to the literature (2). This new work from Cambodia (1,2) is now the next piece of a contemporary narrative within PCCM focused on PICU practice in LMIC settings. For example, we have had articles about utility of Pediatric Index of Mortality scoring (7), resource inequities among facilities (8), pediatric acute respiratory distress syndrome diagnosis and prevalence (9,10), sepsis biomarkers (11,12), and sepsis definitions that are appropriate for children worldwide (13). Also look at the deeper insight provided by our PCCM editorial commentaries on LMIC settings about monitoring outcomes (14), development of services when resources are scarce (15), and centralization of practices (16). WHAT IS THE ASSOCIATED EVOLUTION IN RESPIRATORY EFFORT IN PICU PATIENTS AGED UNDER 2 YEARS WITH BRONCHIOLITIS? Lepage-Farrell A, Tabone L, Plante V, et al: Noninvasive Neurally Adjusted Ventilatory Assist in Infants With Bronchiolitis: Respiratory Outcomes in a Single-Center, Retrospective Cohort, 2016−2018 (3). My second editor’s choice article is from investigators at a PICU in Canada who report their experience of using NIV-NAVA in 64 of 205 bronchiolitis patients aged under 2 years. In this report, NIV-NAVA was used after failure of first-tier NIV support (i.e., continuous positive airway pressure or high-flow nasal oxygen [HFNO]) during the two winters, 2016−2018. Six of the NIV-NAVA patients deteriorated to the point of needing invasive mechanical ventilation (IMV). The researchers give a detailed account of respiratory effort physiology with quantitative electrical activity of the diaphragm (Edi) from 2 hours before to 2 hours after starting NIV-NAVA. This work extends two themes in PCCM: bronchiolitis and diaphragmatic electrophysiology. Regarding bronchiolitis respiratory support, by way of recalling what was published in 2023, we had a systematic review and network meta-analyses on HFNO and other NIV therapies in bronchiolitis (17); two quality improvement studies of “protocolized NIV” in bronchiolitis (18–20); and a multicenter, retrospective study of variations in early PICU management during IMV (21,22). Regarding diaphragmatic electrophysiology, in 2021 PCCM had a descriptive study of transcutaneous electromyography (23,24), and in 2023 there was a retrospective report about the range in Edi measurements in the PICU population (25,26) from the current researchers in Canada (3). Add to all this material the editorial that accompanies the new report (4). It gives a helpful discussion about bringing together bronchiolitis clinical care with diaphragmatic electrophysiology data in a potential protocolized trial (4) (n.b., elsewhere in PCCM we call these pragmatic trials (27,28)). CAN AN AUTOMATED MACHINE LEARNING PREDICTION MODEL HELP WITH EARLY IDENTIFICATION OF PATIENTS NEEDING ENDOTRACHEAL INTUBATION? Chanci D, Grunwell JR, Rafiel A, et al: Development and Validation of a Model for Endotracheal Intubation and Mechanical Ventilation Prediction in PICU Patients (5). My third editor’s choice article focuses on the problem of predicting need for endotracheal intubation and IMV in PICU patients. Here, the authors use large datasets to develop and validate an automated machine learning model for decision-support. This material is state-of-the-art for the PICU, so also read the accompanying editorial (6). There are two other editorials that have been part of the Journal’s narrative on machine learning: one gives details about evaluating machine learning models for clinical prediction problems (29); the other is about clinical deterioration detection using machine learning (30). These, together with this March’s editorial (6), serve as an education in this theme of research. In the April 2024 issue, the PEDAL (pediatric data science and analytics) subgroup of the PALISI (pediatric acute lung injury and sepsis investigators) network (31) have a scoping review as part of a Special Article on the use of supervised machine learning applications in PCCM research (32). This PEDAL subgroup position paper will be the standard for future PCCM articles on machine learning in the PICU. “PCCM CONNECTIONS” FOR READERS The PCCM Connections this month highlights two educational items. The first is in the new and improved Editorial Notes, Methods, and Statistics section article comments on the problem of measurement error in PCCM research (33). This commentary is very important for those reading and reporting research in PCCM as it describes the standard now required for considering error, precision, bias, noise, and differences between measurements and scales presented in our tables and figures. As an example, the authors write about data using point of care ultrasound (POCUS) measurements. They illustrate their material with one of the other studies published this month (34). Here, POCUS was used in under 5-year-olds to measure the laryngeal air column width around a cuffed endotracheal tube before extubation. These millimeter measurements (to 2 decimal places) were then related to risk of postextubation stridor. Finally, the second educational item highlighted in PCCM Connections is a Clinical Science commentary about the cold stress response in acute brain injury and critical illness (35). The authors from the Safar Center for Resuscitation Research, Pittsburgh, write an outstanding and beautifully illustrated commentary and, in PCCM’s 25th year, it shows how far the field has progressed since the Safar group’s 2000 (volume number 1) publication on secondary brain damage after traumatic injury (36).

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.