Pediatric Critical Care Medicine最新文献

Objectives: We aimed to implement a post-cardiac arrest targeted temperature management (TTM) bundle to reduce the percent of time with a fever from 7% to 3.5%.

Design: A prospective, quality improvement (QI) initiative utilizing the Method for Improvement. The pre-intervention historical control period was February 2019 to March 2021, and the intervention test period was April 2021 to June 2022.

Setting: The PICU of a freestanding, tertiary children's hospital, in the United States.

Patients: Pediatric patients 2 days old or older to 18 young or younger than years old who experienced cardiac arrest, received greater than or equal to 2 minutes of chest compressions, required invasive mechanical ventilation post-resuscitation, and had no documented limitations of care.

Interventions: We developed and implemented a TTM bundle that included standard temperature goals, instructions and training on cooling blanket use, scheduled prescription of antipyretics, an algorithm for managing shivering, and standardized orders in our electronic health record.

Measurements and results: We reviewed data from 29 patients in the pre-intervention period and studied 46 in the intervention period. In comparison with historical controls, the reduction in median (interquartile range [IQR]) percentage of febrile (> 38°C) time per patient associated with the TTM bundle was 0% (IQR, 0-3%) vs. 7% (IQR, 0-13%; p < 0.001). The intervention period, vs. pre-intervention, was associated with fewer patients with fever at any time (16/46 vs. 21/29; mean reduction, 37%; 95% CI, 13.8-54.8%; p = 0.002). We failed to identify an association between the intervention period, vs. pre-intervention, and the development of hypothermia (< 35°C; 8/46 vs. 3/29; mean change, 7%; 95% CI, -10.9% to 21.8%; p = 0.40).

Conclusions: In this QI project, we have demonstrated that implementation of a TTM bundle is associated with reduced duration and frequency of fever in patients who survive cardiac arrest.

Objectives: Pediatric patients with advanced heart disease (AHD) often receive high intensity medical care at the end of life (EOL). In this study, we aimed to determine whether receipt and timing of pediatric palliative care (PPC) consultation was associated with EOL care intensity of pediatric patients with AHD.

Design: Retrospective cohort study.

Setting: Single-center, 16-bed cardiac ICU (CICU) in a children's hospital in the Mountain West.

Patients: Pediatric patients (0-21 yr) with AHD treated in the CICU and subsequently died from January 2014 to December 2022.

Interventions: None.

Measurements and main results: We compared demographics, EOL characteristics, and care, including medical interventions and mortality characteristics for patients by receipt and timing of PPC (i.e., ≥ 30 d from [early] or < 30 d of death [late]) using chi-square and Wilcoxon rank-sum tests. Of 218 patients, 78 (36%) did not receive PPC, 76 received early PPC (35%), and 64 received late PPC (29%). Compared with patients who did not receive PPC, patients receiving PPC had lower EOL care intensity (77% vs. 96%; p = 0.004) and fewer invasive interventions within 14 days of death (74% vs. 92%; p = 0.004). Receipt of PPC, vs. not, was associated with lower rate of death during cardiopulmonary resuscitation (12% vs. 32%; p = 0.004) and more use of comfort care (23% vs. 3%; p = 0.004). Among patients receiving PPC, early PPC was associated with fewer invasive interventions within 14 days of death (65% vs. 85%; p = 0.033). Care intensity was high for patients with early and late PPC.

Conclusions: Early PPC was associated with fewer invasive interventions within 14 days of death, yet the care intensity at EOL remained high. With early PPC, families likely receive timely psychosocial and advance care planning support without significantly altering goals of care.

Objectives: To determine the reliability of the Bleeding Assessment Scale in critically Ill Children (BASIC) definition of bleeding severity in a diverse cohort of critically ill children.

Design: Prospective cohort study.

Setting: Eight mixed PICUs in the Netherlands, Israel, and the United States.

Subjects: Children ages 0-18 years admitted to participating PICUs from January 1, 2020, to December 31, 2022, with bleeding noted by bedside nurse.

Interventions: None.

Measurements and main results: The bleeding events were classified as minimal, moderate, or severe, according to the BASIC definition, by two independent physicians at two different time points. Patient demographic data, laboratory values, and clinical outcomes were collected. Three hundred twenty-eight patients were enrolled. The overall inter-rater reliability was substantial (weighted kappa coefficient, 0.736; 95% CI, 0.683-0.789), and the intra-rater reliability was "almost-perfect" (weighted kappa coefficient, 0.816; 95% CI, 0.769-0.863). The platelet count ( p = 0.008), prothrombin time ( p = 0.004), activated partial thromboplastin time ( p = 0.025), and fibrinogen levels ( p = 0.035) were associated with the bleeding severity, but the international normalized ratio was not ( p = 0.195). Patients were transfused blood components in response to any bleeding in 31% of cases and received hemostatic medications in 9% of cases. More severe bleeding was associated with increased 28-day mortality, longer hospital length of stay, and more days receiving inotropic support.

Conclusions: The BASIC definition is a reliable tool for identifying and classifying bleeding in critically ill children. Implementing this definition into clinical and research practice may provide a consistent and reliable evaluation of bleeding.

Objectives: To determine if a priori standardization of outcome hemostatic definitions alone was adequate to enable useful comparison between two cohorts of pediatric extracorporeal membrane oxygenation (ECMO) patients, managed according to local practice and protocol.

Design: Comparison of two separate prospective cohort studies performed at different centers with standardized outcome definitions agreed upon a priori.

Setting: General and cardiac PICUs at the Royal Children's Hospital (RCH) in Melbourne, Australia, and the Sophia Children's Hospital (SCH) in Rotterdam, The Netherlands.

Patients: Children (0-18 yr old) undergoing ECMO.

Interventions: None.

Measurements and main results: Although outcome definitions were standardized a priori, the interpretation of surgical interventions varied. The SCH study included 47 ECMO runs (September 2019 to April 2023), and the RCH study included 97 ECMO runs (September 2016 to Jan 2022). Significant differences in patient populations were noted. RCH patients biased toward frequent cardiac ECMO indications, central cannulation, and cardiopulmonary bypass before ECMO. The frequency of outcome ascertainment was not standardized.

Conclusions: This international comparison shows that standardizing hemostatic outcome definitions alone is insufficient for sensible comparison. Uniform interpretation of definitions, consistent frequency of outcome ascertainment, and stratification based on patient populations and ECMO practices are required. Our results highlight the granularity of detail needed for cross-center comparison of hemostatic outcomes in pediatric ECMO. Further work is needed as we move toward potential multicenter trials of pediatric ECMO.

Objectives: Pediatric sepsis results in significant morbidity and mortality worldwide. There is an urgent need to investigate adjunctive therapies that can be administered early. We hypothesize that using vitamin C combined with hydrocortisone increases survival free of inotropes/vasopressors support until day 7 compared with standard care. Here we describe the Resuscitation in Paediatric Septic Shock using Vitamin C and Hydrocortisone (RESPOND) trial protocol, which aims to address this hypothesis.

Design: Randomized, open label, controlled, parallel-group, three-arm trial with integrated economic evaluation.

Setting: Nine Australia and New Zealand PICUs, with interest from additional international sites.

Patients: Children between 7 days and younger than 18 years old who are treated for suspected or confirmed sepsis and receiving inotropes/vasopressors for greater than 1 hour.

Interventions: IV vitamin C (100 mg/kg [maximum 5 g] every 6 hr) and hydrocortisone (1 mg/kg [maximum 50 mg] every 6 hr), or IV hydrocortisone alone (1 mg/kg [maximum 50 mg] every 6 hr) or standard care.

Measurements and main results: Three hundred eighty-four children will be randomly assigned to receive the interventions, or standard care in a 1:1:1 ratio with stratification by steroid administration pre-randomization and hospital site. The primary outcome is time alive and free of inotropes/vasopressors, censored at 7 days. Secondary outcomes include 28-day mortality, survival free of organ support, PICU length of stay, quality of life, functional status and neurodevelopmental vulnerability at 6 months post-enrollment, and hospitalization-related costs. Statistical analysis will be based on an intention-to-treat principle. The study has ethical approval (HREC/20/QCHQ/69922, dated December 21, 2020), is registered in the Australian New Zealand Clinical Trials Registry (ACTRN12621000247875), commenced recruitment on December 8, 2021, and is expected to finish recruitment by mid-2026.

Conclusions: Dissemination of the results will occur through publication in peer-reviewed journals, presentations at international conferences, and additional consumer-informed pathways.

Objectives: To examine the association between family presence at the PICU bedside and daily positive delirium screening scores.

Design: Retrospective cohort study.

Setting: Tertiary children's hospital PICU.

Subjects: Children younger than 18 years old with PICU length of stay greater than 36 hours enrolled in the Seattle Children's Hospital Outcomes Assessment Program from 2014 to 2017.

Interventions: None.

Measurements and main results: In the dataset, delirium screening had been performed bid using the Cornell Assessment of Pediatric Delirium, with scores greater than or equal to 9 classified as positive. Family presence was documented every 2 hours. Among 224 patients, 55% (n = 124/224) had positive delirium screening on 44% (n = 408/930) of PICU days. Family presence at the bedside during PICU stay (< 90% compared with ≥ 90%) was associated with higher proportion of ever (as opposed to never) being screened positive for delirium (26/37 vs. 98/187; difference, 17.9% [95% CI, 0.4-32.1%]; p = 0.046). On univariate analysis, each additional decile of increasing family presence was associated with lower odds of positive delirium screening on the same day (odds ratio [OR], 0.87 [95% CI, 0.77-0.97]) and subsequent day (OR, 0.84 [95% CI, 0.75-0.94]). On multivariable analysis after adjustments, including baseline Pediatric Cerebral Performance Category (PCPC), higher family presence was associated with lower odds of subsequent-day positive delirium screening (OR, 0.89 [95% CI, 0.81-0.98]). Among patients with PCPC less than or equal to 2, each additional decile of increasing family presence was independently associated with lower odds of both same-day (OR, 0.90 [95% CI, 0.81-0.99]) and subsequent-day (OR, 0.85 [95% CI, 0.76-0.95]) positive delirium screening.

Conclusions: In our 2014-2017 retrospective cohort, greater family presence was associated with lower odds of delirium in PICU patients. Family presence is a modifiable factor that may mitigate the burden of pediatric delirium, and future studies should explore barriers and facilitators of family presence in the PICU.

Objectives: To evaluate contextual factors relevant to implementing pediatric ventilator liberation guidelines and to develop an implementation strategy for a multicenter collaborative.

Design: Cross-sectional qualitative analysis of a 2023/2024 survey.

Setting: International, multicenter Ventilation Liberation for Kids (VentLib4Kids) collaborative.

Subjects: Physicians, advanced practice providers, respiratory therapists, and nurses.

Interventions: None.

Measurements and main results: The survey was distributed to 26 PICUs representing 18 unique centers (17 in North American)-14 general medical/surgical, eight cardiac, and four mixed (1935 solicitations). All 409 responses were analyzed (prescribers 39.8%, nursing 32.8%, and respiratory therapists 27.4%). Three implementation tiers were identified based on perceptions of evidence, feasibility, positive impact, and favorability constructs. Tier A (≥ 80% agreement for all constructs) included extubation readiness testing (ERT) screening, ERT bundle, spontaneous breathing trials (SBTs), upper airway obstruction (UAO) risk mitigation, and risk stratified noninvasive respiratory support (NRS). Tier B (50-79% agreement) included standard risk SBT method, risk stratified SBT duration, and UAO risk assessment. Tier C (< 50% agreement) included high-risk SBT method, respiratory muscle strength testing, and infant NRS. The smallest perceived practice gaps were noted in tier A and the largest in tier C. The smallest practice gap was risk stratified NRS (88% agreement). The largest practice gap was respiratory muscle strength (18% agreement). In regression analysis, independently significant differences in perceptions based on role and unit type for multiple constructs were identified for UAO risk assessment, UAO risk mitigation, risk stratified NRS, and infant NRS.

Conclusions: This survey study of the VentLib4Kids collaborative lays the foundation for phased implementation of the 2023 pediatric ventilator liberation guidelines. Early phases should focus on the best implementation profiles and smallest practice gaps. Later phases should address those that are more challenging. Unit- and role-based tailoring of differences should be considered for some recommendations more than others.

Objectives: To describe frequency of, and risk factors, for change in caregiver employment among critically ill children with acute respiratory failure.

Design: Preplanned secondary analysis of prospective cohort dataset, 2018-2021.

Setting: Quaternary Children's Hospital PICU.

Patients: Children who required greater than or equal to 3 days of invasive ventilation, survived hospitalization, and completed greater than or equal to 1 post-discharge survey.

Interventions: None.

Measurements and main results: We measured change in caregiver employment 1 and 12 months post-discharge relative to pre-admission and, when present, change in caregiver identity defined by relationship to the patient. Data were collected by survey. We used logistic regression to identify factors associated with these changes. We evaluated 130 children, median age 6.4 years (interquartile range, 1.10-13.3 yr), 40 (30.8%) with a complex chronic condition (CCC), and 99 (76.2%) with normal pre-illness Functional Status Scale scores. Of 123 with 1-month post-discharge data, 25 of 123 (20.3%) experienced a change in caregiver employment and an additional 14 of 123 (11.4%) had a change in caregiver(s). Of 115 with 12-month post-discharge data, 33 of 115 (28.7%) experienced a change in caregiver employment and an additional 16 of 115 (13.9%) had a change in caregiver(s). After controlling for age, CCC, baseline caregiver employment, new morbidity at discharge, and social and economic index; higher maximum Pediatric Logistic Organ Dysfunction-2 score (odds ratio [OR], 1.19 [95% CI, 1.01-1.41]) and government insurance (OR, 3.85 [95% CI, 1.33-11.11]) were associated with the composite outcome of change in caregiver employment or caregiver(s) at 1-month post-discharge.

Conclusions: At 1 and 12 months post-discharge, more than one-in-five children who survived greater than or equal to 3 days of invasive ventilation had a change in caregiver employment and one-in-ten had a change in caregiver(s). Identification of risk factors, such as illness severity and social determinants of health, associated with a significant family change may improve our support of these families.