Association of Fluid Overload with Escalation of Respiratory Support and Endotracheal Intubation in Acute Bronchiolitis Patients.

Fluid overload has been associated with increased oxygen requirement, prolonged duration of mechanical ventilation, and longer length of hospital stay in children hospitalized with pulmonary diseases. Critically ill infants with bronchiolitis admitted to the pediatric intensive care unit (PICU) also tend to develop fluid overload and there is limited information of its role on noninvasive respiratory support. Thus, our primary objective was to study the association of fluid overload in patients with bronchiolitis admitted to the PICU with respiratory support escalation (RSE) and need for endotracheal intubation (ETI). Infants ≤24 months of age with bronchiolitis and admitted to the PICU between 9/2009 and 6/2015 were retrospectively studied. Demographic variables, clinical characteristics including type of respiratory support and need for ETI were evaluated. Fluid overload as assessed by net fluid intake and output (net fluid balance), cumulative fluid balance (CFB) (mL/kg), and percentage fluid overload (FO%), was compared between patients requiring and not requiring RSE and among patients requiring ETI and not requiring ETI at 0 (PICU admission), 12, 24, 36, 48, 72, 96, and 120 hours. One-hundred sixty four of 283 patients with bronchiolitis admitted to the PICU qualified for our study. Thirty-four of 164 (21%) patients required escalation of respiratory support within 5 days of PICU admission and of these 34 patients, 11 patients required ETI. Univariate analysis by Kruskal-Wallis test of fluid overload as assessed by net fluid balance, CFB, and FO% between 34 patients requiring and 130 patients not requiring RSE and among 11 patients requiring ETI and 153 patients not requiring ETI, at 0, 12, 24, 36, 48, 72, 96 and 120 hours did not reveal any significant difference ( p >0.05) at any time interval. Multivariable logistic regression analysis revealed higher PRISM score (odds ratio [OR]: 4.95, 95% confidence interval [95% CI]: 1.79-13.66; p = 0.002), longer hours on high flow nasal cannula (OR: 4.86, 95% CI: 1.68-14.03; p = 0.003) and longer hours on noninvasive ventilation (OR: 11.16, 95% CI: 3.36-36.98; p < 0.001) were associated with RSE. Fluid overload as assessed by net fluid balance, CFB, and FO% was not associated with RSE or need for ETI in critically ill bronchiolitis patients admitted to the PICU. Further prospective studies involving larger number of patients with bronchiolitis are needed to corroborate our findings.

Early Post-operative ECG Changes as a Predictor of Post-pericardiotomy Syndrome Following Atrial Septal Defect Repair.

To identify risk factors associated with post-pericardiotomy syndrome (PPS) in patients undergoing surgical repair of atrial septal defects (ASD). A single-center retrospective study. Tertiary academic hospital. Included were patients of all ages who underwent surgical ASD repair, while exclusion criteria included the absence of post-operative electrocardiogram (ECG), lack of follow-up post-discharge and factors hindering ECG interpretation. Demographic and clinical data, including ECG changes indicative of pericardial inflammation, were collected. The primary outcome measure was the development of PPS, determined based on the standardized European Society of Cardiology (ESC) criteria. Among 190 patients who underwent surgical ASD repair, 154 (81%) met the inclusion criteria. Of these, 25 (16%)in total developed PPS, of which 60% were ≥ 18 years of age and 56% female. Significant associations relating both early ECG changes and pre-discharge pericardial effusion with subsequent occurrence of PPS were found in both univariate and multivariate analyses. The study establishes correlations of both early post-operative ECG changes indicative of inflammation and pre-discharge pericardial effusion with subsequent occurrence of PPS in patients undergoing surgical ASD repair. Both utilizing the standardized ESC definition of PPS and incorporating a physician-validated ECG evaluation strengthened the methodologic approach in establishing these relationships. The results also highlight the importance of considering age as a potential risk factor for PPS. Further research is needed to validate these findings and explore additional risk factors predicting early identification and management of patients at high risk for PPS following surgical ASD repairs.

Predicting pediatric cardiac surgery-associated acute kidney injury using machine learning.

BACKGROUND: Prediction of cardiac surgery-associated acute kidney injury (CS-AKI) in pediatric patients is crucial to improve outcomes and guide clinical decision-making. This study aimed to develop a supervised machine learning (ML) model for predicting moderate to severe CS-AKI at postoperative day 2 (POD2). METHODS: This retrospective cohort study analyzed data from 402 pediatric patients who underwent cardiac surgery at a university-affiliated children's hospital, who were separated into an 80%-20% train-test split. The ML model utilized demographic, preoperative, intraoperative, and POD0 clinical and laboratory data to predict moderate to severe AKI categorized by Kidney Disease: Improving Global Outcomes (KDIGO) stage 2 or 3 at POD2. Input feature importance was assessed by SHapley Additive exPlanations (SHAP) values. Model performance was evaluated using accuracy, area under the receiver operating curve (AUROC), precision, recall, area under the precision-recall curve (AUPRC), F1-score, and Brier score. RESULTS: Overall, 13.7% of children in the test set experienced moderate to severe AKI. The ML model achieved promising performance, with accuracy of 0.91 (95% CI: 0.82-1.00), AUROC of 0.88 (95% CI: 0.72-1.00), precision of 0.92 (95% CI: 0.70-1.00), recall of 0.63 (95% CI: 0.32-0.96), AUPRC of 0.81 (95% CI: 0.61-1.00), F1-score of 0.73 (95% CI: 0.46-0.99), and Brier score loss of 0.09 (95% CI: 0.00-0.17). The top ten most important features assessed by SHAP analyses in this model were preoperative serum creatinine, surgery duration, POD0 serum pH, POD0 lactate, cardiopulmonary bypass duration, POD0 vasoactive inotropic score, sex, POD0 hematocrit, preoperative weight, and POD0 serum creatinine. CONCLUSIONS: A supervised ML model utilizing demographic, preoperative, intraoperative, and immediate postoperative clinical and laboratory data showed promising performance in predicting moderate to severe CS-AKI at POD2 in pediatric patients.

The Pediatric Data Science and Analytics Subgroup of the Pediatric Acute Lung Injury and Sepsis Investigators Network: Use of Supervised Machine Learning Applications in Pediatric Critical Care Medicine Research.

OBJECTIVE: Perform a scoping review of supervised machine learning in pediatric critical care to identify published applications, methodologies, and implementation frequency to inform best practices for the development, validation, and reporting of predictive models in pediatric critical care. DESIGN: Scoping review and expert opinion. SETTING: We queried CINAHL Plus with Full Text (EBSCO), Cochrane Library (Wiley), Embase (Elsevier), Ovid Medline, and PubMed for articles published between 2000 and 2022 related to machine learning concepts and pediatric critical illness. Articles were excluded if the majority of patients were adults or neonates, if unsupervised machine learning was the primary methodology, or if information related to the development, validation, and/or implementation of the model was not reported. Article selection and data extraction were performed using dual review in the Covidence tool, with discrepancies resolved by consensus. SUBJECTS: Articles reporting on the development, validation, or implementation of supervised machine learning models in the field of pediatric critical care medicine. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of 5075 identified studies, 141 articles were included. Studies were primarily (57%) performed at a single site. The majority took place in the United States (70%). Most were retrospective observational cohort studies. More than three-quarters of the articles were published between 2018 and 2022. The most common algorithms included logistic regression and random forest. Predicted events were most commonly death, transfer to ICU, and sepsis. Only 14% of articles reported external validation, and only a single model was implemented at publication. Reporting of validation methods, performance assessments, and implementation varied widely. Follow-up with authors suggests that implementation remains uncommon after model publication. CONCLUSIONS: Publication of supervised machine learning models to address clinical challenges in pediatric critical care medicine has increased dramatically in the last 5 years. While these approaches have the potential to benefit children with critical illness, the literature demonstrates incomplete reporting, absence of external validation, and infrequent clinical implementation.

Association of enteral feeds in critically ill bronchiolitis patients supported by high-flow nasal cannula with adverse events and outcomes.

To study association of enteral feeds in bronchiolitis patients supported by different levels of high flow nasal cannula (HFNC) with adverse events, nutritional goals, and clinical outcomes. Bronchiolitis patients ≤ 24 months of age treated with < 1 L/kg/min, 1-2 L/kg/min and > 2 L/kg/min of HFNC between January 2014 and December 2021 were studied retrospectively at a tertiary care children's hospital. Adverse events (aspiration pneumonia, emesis, and respiratory support escalation), nutritional goals (initiation of enteral feeds, achievement of nutritional goal volume and goal calories, percentage weight change during hospital stay) and clinical outcomes (HFNC duration, oxygen supplementation duration after HFNC, length of hospital stay following HFNC support, total length of hospital stay and follow-up for 1 month after hospital discharge) were compared between fed and non-fed patients on HFNC. Six hundred thirty-six (489 fed and 147 not-fed) bronchiolitis patients on HFNC studied. 260 patients, 317 patients and 59 patients were supported by < 1 L/kg/min, 1-2 L/kg/min and > 2 L/kg/min of HFNC, respectively. Enterally fed patients had significantly less adverse events (OR = 0.14, 95% CI 0.083 - 0.23, p < 0.001), significantly better nutritional goals: earlier initiation of enteral feeds by 65% in time (mean ratio = 0.35, 95% CI 0.28 - 0.43, p < 0.001), earlier achievement of goal volume and goal calorie needs by 14% in time (mean ratio = 0.86, 95% CI 0.78 to 0.96, p = 0.005) and significantly better clinical outcomes: shorter HFNC duration by 29.75 h (95% CI 20.19 -39.31, p < 0.001), shorter oxygen supplementation duration after HFNC by 12.14 h (95% CI 6.70 -17.59, p < 0.001), shorter length of hospital stay after HFNC support by 21.35 h (95% CI 14.71-27.98, p < 0.001) and shorter total length of hospital stay by 51.10 h (95% CI 38.65 -63.55, p < 0.001), as compared to non-fed patients, after adjusting for age, weight, prematurity, comorbidities, admission time, admission bronchiolitis score, admission respiratory rate, and HFNC levels. The number of revisits and readmissions at 7 and 30 days after hospital discharge were not significantly different (p > 0.05) between the fed and non-fed groups.    Conclusion: Enteral feeding of bronchiolitis patients supported by different levels of HFNC is associated with less adverse events and better nutrition goals and clinical outcomes. What is Known: •There is general apprehension to feed critically ill bronchiolitis patients supported by high flow nasal cannula. What is New: •Our study reveals that enteral feeding of critically ill bronchiolitis patients supported by different levels of high flow nasal cannula is associated with minimal adverse events, better nutritional goals and improved clinical outcomes as compared to non-fed patients.

Wearable Biosensors in Congenital Heart Disease: Needs to Advance the Field.

Traditional measures of clinical status and physiology have generally been based in health care settings, episodic, short in duration, and performed at rest. Wearable biosensors provide an opportunity to obtain continuous non-invasive physiologic data from patients with congenital heart disease (CHD) in the real-world setting, over longer durations, and across varying levels of activity. However, there are significant technical limitations to the use of wearable biosensors in CHD. Here, we review current applications of wearable biosensors in CHD; how clinical and research uses of wearable biosensors must consider various CHD physiologies; the technical challenges in developing wearable biosensors for CHD; and special considerations for digital biomarkers in CHD.

Clinical Informatics and Quality Improvement in the Pediatric Intensive Care Unit.

Clinical informatics can support quality improvement and patient safety in the pediatric intensive care unit (PICU) in several ways including data extraction, analysis, and decision support enabled by electronic health records (EHRs), and databases and registries. Clinical decision support (CDS), embedded in EHRs, now an integral part of the workflow in the PICU, includes several tools and is increasingly leveraging artificial intelligence (AI). Understanding the opportunities and challenges can improve the engagement of clinicians with the design, validation, and implementation of CDS, improve satisfaction with CDS, and improve patient safety, care quality, and value.

Stroke in pediatric ECMO patients: analysis of the National Inpatient Sample (NIS) database.

BACKGROUND: The rates, outcomes, and long-term trends of stroke complicating the use of extracorporeal membrane oxygenation (ECMO) have been inconsistently reported. We compared the outcomes of pediatric ECMO patients with and without stroke and described the frequency trends between 2000 and 2017. METHODS: Using the National Inpatient Sample (NIS) database, pediatric patients (age ≤18 years) who received ECMO were identified using ICD-9&10 codes. Binary, regression, and trend analyses were performed to compare patients with and without stroke. RESULTS: A total of 114,477,997 records were reviewed. Overall, 28,695 (0.025%) ECMO patients were identified of which 2982 (10.4%) had stroke, which were further classified as hemorrhagic (n = 1464), ischemic (n = 1280), or combined (n = 238). Mortality was higher in the hemorrhagic and combined groups compared to patients with ischemic stroke and patients without stroke. Length of stay (LOS) was significantly longer in stroke vs. no-stroke patients. Hypertension and septicemia were more encountered in the hemorrhagic group, whereas the combined group demonstrated higher frequency of cardiac arrest and seizures. CONCLUSIONS: Over the years, there is an apparent increase in the diagnosis of stroke. All types of stroke in ECMO patients are associated with increased LOS, although mortality is increased in hemorrhagic and combined stroke only. IMPACT: Stroke is a commonly seen complication in pediatric patients supported by ECMO. Understanding the trends will help in identifying modifiable risk factors that predict poor outcomes in this patient population.

Simplified Pediatric Index of Mortality 3 Score by Explainable Machine Learning Algorithm.

Pediatric Index of Mortality 3 is a validated tool including 11 variables for the assessment of mortality risk in PICU patients. With the recent advances in explainable machine learning algorithms, we aimed to assess feasibility of application of these machine learning models to simplify the Pediatric Index of Mortality 3 scoring system in order to decrease time and labor required for data collection and entry for Pediatric Index of Mortality 3. DESIGN: Single-center, retrospective cohort study. Data from the Virtual Pediatric Systems for patients admitted to Cleveland Clinic Children`s PICU between January 2008 and December 2019 was obtained. Light Gradient Boosting Machine Regressor (a gradient boosting decision tree algorithm) was used for building the machine learning models. Variable importance was analyzed by SHapley Additive exPlanations. All of the 11 Pediatric Index of Mortality 3 variables were used as input variables in the machine learning models to predict Pediatric Index of Mortality 3 risk of mortality as the outcome variable. Mean absolute error, root mean squared error, and R-squared were calculated for each of the 11 machine learning models as model performance parameters. SETTING: Quaternary children's hospital. PATIENTS: PICU patients. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Five-thousand sixty-eight patients were analyzed. The machine learning models were able to maintain similar predictive error until the number of input variables decreased to four. The machine learning model with five input variables (mechanical ventilation in the first hour of PICU admission, very-high-risk diagnosis, surgical recovery from a noncardiac procedure, low-risk diagnosis, and base excess) produced lowest mean root mean squared error of 1.49 (95% CI, 1.05-1.93) and highest R-squared of 0.73 (95% CI, 0.6-0.86) with mean absolute error of 0.43 (95% CI, 0.35-0.5) among all the 11 machine learning models. CONCLUSIONS: Explainable machine learning methods were feasible in simplifying the Pediatric Index of Mortality 3 scoring system with similar risk of mortality predictions compared to the original Pediatric Index of Mortality 3 model tested in a single-center dataset.

Aortic Valve Replacement in Young and Middle-Aged Adults: Current and Potential Roles of TAVR.

BACKGROUND: Contemporary practice patterns and outcomes for aortic valve replacement (AVR) among young and middle-aged adults are unknown given guideline modifications for surgical AVR (SAVR) and increasing transcatheter AVR (TAVR) acceptance. This study describes SAVR and TAVR use and outcomes using The Society of Thoracic Surgeons (STS) National Databases. METHODS: Adults 18 to 55 years of age in the Congenital Heart Surgery Database (CHSD) and the Adult Cardiac Surgery Database (ACSD) who underwent SAVR or TAVR from 2013 to 2018 were included. Perioperative characteristics and early outcomes were described by valve type. Multivariable regression identified determinants of death, length of hospital stay, and a composite end point of renal failure, persistent neurologic deficit, readmission, and reoperation. RESULTS: The study analyzed 1580 unique CHSD and 44,173 ACSD operations, 16% of which were performed in patients with congenital heart disease. Valve use included the following: TAVR, 1%; mechanical, 42%; bioprosthetic, 55%; autograft, 0.6%; homograft, 1.2%; and Ozaki, 0.4%. Over time, TAVR volumes increased by 167%. The 30-day mortality was as follows: TAVR, 3.8%; mechanical, 3.2%; bioprosthetic, 3.7%; autograft, 0.6%; homograft, 9%; and Ozaki, 3.4%. Stroke rate was lower for isolated SAVR vs isolated TAVR (0.9% vs 2.4%; P = .002). In multivariable analyses, mortality risk was lower with mechanical valves, congenital morbidity risk was higher with TAVR, and length of stay was shorter with TAVR. CONCLUSIONS: TAVR is increasingly used for adults younger than 55 years of age. Given the uniformly excellent results with SAVR, including both mortality and morbidity-particularly regarding stroke, our data favor SAVR in this population, but a prospective trial is needed. Ongoing efforts to harmonize variables and outcomes definitions between the ACSD and CHSD are valuable.

Outcomes of hypoplastic left heart syndrome: analysis of National Inpatient Sample Database 1998-2004 versus 2005-2014.

Neonates with hypoplastic left heart syndrome (HLHS) were identified from the National Inpatient Sample dataset for the years 1998-2014. These patients were stratified into two chronological groups, past group (1998-2005) and recent group (2006-2014). A total of 20,649 neonates with HLHS were identified. Of them, 9179 (44.5%) were born in the past group and 11,470 (55.5%) in the recent group. Median birth weight was significantly less in the recent group (2967 g vs. 3110 g, p = 0.005). The patients in the recent group had more patients with low birth weight ( < 2.5 kg) and prematurity (8.7% vs 7.6% and 12.7% vs. 4.3%., respectively). In addition, recent group had more comorbidities including chromosomal anomalies, total anomalous pulmonary venous return, and kidney anomalies (5.6% vs. 3.6%, 2.3% vs. 1.7%, and 5.6% vs. 3.6%, respectively, p < 0.001); these were associated with a higher rate of extracorporeal membrane oxygenation utilization (9.2% vs. 4.5%, p < 0.001). Consequently, median length of stay was longer in the recent group (8 vs. 6 days, p < 0.001).Conclusion: Despite the higher frequency of comorbidities in recent group, the mortality rates decreased by 20% (from 25.3% to 20.6%, p < 0.001). Balloon atrial septostomy was performed less frequently in the recent group (23.3% vs. 16.1%, p < 0.001).What is known:• Hypoplastic left heart syndrome has the highest mortality among congenital cardiac defects during the first year of life.• Limited studies on patients' comorbidities and mortality rates trends over last two decades.What is new:• The study utilized a national database to compare in-hospital mortality and length of stay between the two time periods 1998-2005 and 2006-2014.• The recent group had more comorbidities (prematurity, chromosomal anomalies, total anomalous pulmonary venous return, and kidney anomalies), and there was higher rate of ECMO and longer length of stay, while mortality rates decreased by 20%.

Non-infectious pulmonary complications of hematopoietic stem cell transplantation.

Noninfectious pulmonary complications of hematopoietic stem cell transplant are currently more prevalent than infectious complications. Unfortunately, the pathophysiology basis is not completely understood. However, there is a string association with graft-versus-host disease for many of them. Therefore, an important component of their pathophysiology is likely an allo-immune response. There is much research that needs to be conducted to improve the less than optimal outcomes for these disorders.

Evaluation report of pediatric intensive care units in Turkey.

BACKGROUND/AIM: To collect data from throughout Turkey in order to facilitate the organization of pediatric intensive care units (PICUs), and to develop short-term immediate action plans and draft long-term strategic plans. MATERIALS AND METHODS: A total of 35 specialists including 17 pediatric critical care (PCC) specialists, 9 PCC fellows in training, and 9 pediatricians working in PICUs evaluated PICUs and their infrastructures, mortality rates, appropriateness of indications for PICU admissions, PICU bed numbers, and utilization of those PICU beds. RESULTS: PICU bed numbers, PCC specialist numbers, and PICU nurse numbers are insufficient in Turkey. The high percentage of inappropriate and inefficient use of current PICU beds is also another problem. CONCLUSION: In the light of this report, it is obvious that pediatric intensive care services are successful and efficient only in the presence of PCC specialists in PICUs. Studies for improving the infrastructure of PICUs and the training of PCC specialists and other health personnel should be started immediately.

Assessing pulmonary arterial hypertension in infants with severe chronic lung disease of infancy: a role for a pulmonary artery catheter?

The treatment of pulmonary arterial hypertension (PAH) associated with chronic lung disease of infancy (CLDI) is becoming commonplace. However, an optimal approach to the monitoring of this treatment has not been clearly established, and data suggest that such therapy may not be without risk. This study assessed the feasibility and safety of pulmonary artery catheter (PAC) placement and its role in the management of PAH associated with CLDI. The medical records of 12 infants with CLDI requiring chronic mechanical ventilation who underwent PAC monitoring were reviewed. Data analyzed included demographics, hemodynamic data, PAH pharmacological therapy, respiratory support, echocardiographic data, sedation level, complications related to PAC use, and mortality. In this analysis, PAC placement and monitoring was found to be feasible, appeared safe, and was associated with the ability to wean inspired oxygen, decrease sedation, and titrate PAH therapy without untoward effect. However, no definitive conclusions can be drawn from this report given its small sample size and uncontrolled, retrospective design. It is hoped that these data will renew interest in PAC monitoring for CLDI and foster prospective study where its true value can be ascertained.

Scientific rationale for the use of alpha-adrenergic agonists and glucocorticoids in the therapy of pediatric stridor.

Purpose. The most common pharmacological therapies used in the treatment of stridor in children are glucocorticosteroids (GC) and alpha-adrenergic (αAR) agonists. Despite the long-standing reported efficacy of these medications, there is a paucity of data relating to their actual mechanisms of action in the upper airway. Summary. There is compelling scientific evidence supporting the use of αAR-agonists and GCs in pediatric stridor. αAR signaling and GCs regulate the vasomotor tone in the upper airway mucosa. The latter translates into better airflow dynamics, as delineated by human and nonhuman upper airway physiological models. In turn, clinical trials have demonstrated that GCs and the nonselective αAR agonist, epinephrine, improve respiratory distress scores and reduce the need for further medical care in children with stridor. Future research is needed to investigate the role of selective αAR agonists and the potential synergism of GCs and αAR-signaling in the treatment of upper airway obstruction and stridor.