Pediatric Critical Care Medicine最新文献

Objective: To identify factors associated with fluid overload (FO) and early outcomes in pediatric patients with moderate or severe traumatic brain injury (TBI).

Design: Retrospective cohort study using patient electronic medical records.

Setting: Hospital Universitario Fundación Valle del Lili, a tertiary care university hospital in Cali, Colombia.

Patients: Pediatric patients 1-16 years old treated in the PICU for moderate or severe TBI between 2011 and 2022.

Measurements and main results: We identified 158 pediatric patients who met study inclusion criteria. We recorded baseline clinical characteristics and interventions administered in the PICU. FO was defined as a cumulative fluid balance greater than or equal to 10% of body weight on the third day after trauma. Outcomes included mechanical ventilation (MV) duration, acute kidney injury (AKI), functional status, and mortality. Hypothesis tests and multivariable models assessed associations. FO occurred in 32 of 158 (20.2%) of patients at 72 hours and its presence was associated with lower weight ( p < 0.001) and age ( p < 0.001). On comparing those with and without FO, we failed to identify an association with mortality or AKI, respectively: 2 of 32 vs. 11 of 126 (mean difference 2.5% [95% CI, -11.9 to 10.2%], p = 0.65); 1 of 32 vs. 8 of 126 (mean difference 3.2% [95% CI, -9.8 to 9.4%], p = 0.49). However, FO compared with not, was associated with adjusted relative increases in MV duration (1.49 [95% CI, 1.08-2.04], p = 0.040) and PICU length of stay (1.44 [95% CI, 1.04-2.00], p < 0.001).

Conclusions: In our retrospective cohort from 2011 to 2022, FO occurred in approximately one in five pediatric TBI patients with moderate or severe injury. We also found that FO was associated with an adjusted relative increase in MV duration, but we failed to identify an association with other outcomes.

Objectives: U.S. multicenter pediatric data from 2001 to 2014 indicate that the prevalence of pediatric pulmonary embolism (PE) is rising, yet it remains rare at individual centers. Evaluation and management vary due to the low frequency of events. And, therefore, we aimed to develop, implement, and evaluate a clinical management pathway and PE response team (PERT) at our center.

Design: Single-center observational implementation study with pre- vs. post-implementation comparisons.

Setting: Quaternary care pediatric hospital.

Patients: Pediatric patients younger than 21 years old with acute PE who presented for care between January 2005 and August 2022 (pre-implementation) and between September 2022 and August 2024 (post-implementation).

Interventions: Implementation of a PE clinical pathway and PERT.

Measurements and main results: A PE clinical pathway and PERT were developed by a multidisciplinary team of pharmacists, quality improvement experts, and physicians from six clinical specialties. Electronic medical record tools were created to support the implementation of this pathway and PERT, and research and quality databases were created for ongoing evaluation. PE evaluation was standardized with the implementation of the PE clinical pathway. In the 23 months post-implementation, there have been 33 patients with acute PE (24 low-risk, seven intermediate-risk, and two high-risk PE), which we compared with our pre-pathway experience of 175 episodes. The proportion of patients who had laboratory testing with brain natriuretic peptide and troponin, or underwent echocardiogram during their admission increased, pre-vs.-post-implementation, respectively (50/175 vs. 30/33; p < 0.01; 55/175 vs. 30/33; p < 0.01; and 142/175 vs. 32/33; p < 0.02). The multidisciplinary PERT was successfully activated for seven patients with intermediate- or high-risk PE.

Conclusions: Implementation of a pediatric PE clinical pathway and PERT improved standardization of PE evaluation and provided rapid multidisciplinary care for intermediate and high-risk PE. Future reports will evaluate both short- and long-term clinical outcomes in these patients.

Objectives: To explore the association between geospatial determinants of health and prevalence of delirium in the PICU.

Design: Nonprespecified secondary analysis of an observational study dataset.

Setting: Urban academic tertiary care PICU.

Patients: All children admitted over a 12-month period in 2014-2015.

Interventions: None.

Measurements and main results: Of 1547 admissions, there were individual 1264 patients, with race and ethnicity subgroup recorded for 961 (76%), insurance status for 1240 (98%), and language for 1227 (97%). Child Opportunity Index (COI) was determined by a patient's 2010 Census Tract and was available for 1246 (98%). Data were grouped from 1 to 5 (very low to very high). PICU delirium (PD) was present in 193 patients during PICU admission, with PD rates highest in Asian/Pacific Islander, Black non-Hispanic, and Hispanic (respectively, 26, 29, and 39 patients) as compared with White non-Hispanic children ( n = 45; p < 0.001). PD was higher for patients with public insurance (113/573 [20%]) compared with those with private insurance (67/583 [12%]) and those with "other" (8/84 [10%]) forms of support ( p < 0.001). We failed to identify a difference in delirium rates by language preference. A higher prevalence of delirium was present among patients in the lowest COI groups (1-3) as compared with highest (4-5; 135/748 [18%] vs. 55/494 [11%]; p = 0.003). In multivariable analysis, COI was associated with greater adjusted odds of delirium (after adjusting for other demographic and clinical predictors of delirium, including age, developmental disability, severity of illness at admission, need for invasive mechanical ventilation, depth of sedation, and medication exposures), with adjusted odds ratio of 1.55 (95% CI, 1.05-2.3; p = 0.028).

Conclusions: In a 2014-2015 PICU dataset, we have identified an association between lower COI and greater adjusted odds of delirium. This finding calls for further study to investigate potential mediators of this relationship.

Objectives: We aimed to assess the reproducibility of mechanical power (MP) equations in comparison with the geometric method in critically ill children during respiratory support with invasive pressure-controlled ventilation (PCV).

Design: Prospective, exploratory research study.

Setting: Single-center, PICU in The Netherlands.

Patients: Children (< 18 yr old) admitted to the PICU receiving PCV.

Interventions: None.

Measurements and main results: MP was calculated in a cohort of 37 children, with a median (interquartile range [IQR]) age of 12 months (IQR, 2-60 mo). Three, previously proposed MP equations (simplified, comprehensive, and linear MP) were compared with the geometric mean ("gold standard"), measuring the area-under-the-pressure-volume-loop, and assessed using agreement (Bland-Altman) analysis and reliability (intraclass correlation coefficient [ICC]) analysis of parameters. The mean difference (95% CI) was as follows: simplified MP -0.02 J/min (95% CI, -1.02 to 0.99 J/min), comprehensive MP 0.03 J/min (95% CI, -0.94 to 1.00 J/min), and linear MP 0.16 J/min (95% CI, -0.76 to 1.08 J/min). The ICCs for all comparisons were excellent (i.e., > 0.99; p < 0.001).

Conclusions: In critically ill children undergoing invasive PCV, all three MP equations acceptably reproduce the geometric method for calculating MP in J/min.

Objectives: To test whether indexing stroke volume change (ΔSV%) to body size during the passive leg raising (PLR) test in spontanoeusly breathing children improves accuracy to detect fluid responsiveness (FR).

Design: Observational study.

Setting: Two pediatric hospitals.

Patients: Children age 2-16 years.

Interventions: None.

Measurements and main results: In study phase 1, we measured ΔSV% by echocardiography during PLR in healthy children. A positive PLR test was defined as a mean ΔSV% greater than or equal to 10%. The correlation between ΔSV% with body size parameters was assessed, and optimal body size indexation was generated. In study phase 2, the PLR was performed in acutely ill children before a fluid challenge of 20 mL/kg of normal saline. ΔSV% was measured at 10 and 20 mL/kg and FR was defined as ΔSV% greater than or equal to 10% or greater than or equal to 15% (four possible definitions of FR). The diagnostic performance of the PLR using nonindexed and indexed ΔSV% to identify FR was assessed using the area under the receiver operating characteristic curve (AUC) analyses. We recruited 133 and 87 children in phase 1 and 2, respectively. Mean ΔSV% and the proportion of positive PLR test increased with age tertiles both in healthy children and children receiving a fluid challenge ( p ≤ 0.01). ΔSV% positively correlated with body size. Indexing by height (i.e., [ΔSV%/0.0006] × height 2.493 ) removed the effect of body size. The AUC of the PLR for FR ranged from 0.745 to 0.802, depending on the FR definition applied. The use of height-indexed ΔSV% improved diagnostic performance (AUC range, 0.852-0.894) compared to non-indexed ΔSV%, although the result was significant only when FR was defined as ΔSV% greater than 15% after 20 mL/kg (DeLong test < 0.05).

Conclusions: The response in ΔSV% to a PLR is greatly influenced by body size. Indexing the value by height may improve the diagnostic performance of the PLR in children.

Objectives: To create a simple scoring system to evaluate the extent of brain injury on MRI after pediatric out-of-hospital cardiac arrest (OHCA).

Design: Two-center retrospective cohort from 2016 to 2020.

Setting: Two tertiary care children's hospital serving northern California.

Patients: Children older than 48 hours and younger than 18 years old at admission who experienced OHCA within 24 hours before admission and underwent brain MRI within 8 days following arrest.

Intervention: None.

Measurements and main results: Brain abnormalities on diffusion-weighted and T2/fluid-attenuated inversion recovery MRI in six brain regions were summed to quantify severity of injury (0/1 point for each region) in: deep structures (basal ganglia, thalamus, and/or posterior limb of the internal capsule), cortex, white matter, brainstem, hippocampus, and cerebellum. Unfavorable neurologic outcome was defined using Pediatric Cerebral Performance Category (PCPC) score greater than or equal to 1 point above baseline resulting in a hospital discharge PCPC score greater than or equal to 3. There were 58 children included and 41 (71%) had unfavorable outcome. In those with unfavorable outcome, four of 41 had no evidence of injury (MRI score 0), whereas 15 of 17 with favorable outcome had a MRI score of 0 ( p < 0.001). No patient with favorable outcome had evidence of injury in deep structures ( p < 0.001), brainstem ( p = 0.003), cerebellum ( p = 0.024), or hippocampus ( p < 0.001). There were 28 of 41 unfavorable outcome patients who had a MRI score greater than or equal to 3, whereas no children (0/17) with favorable outcome had a MRI score greater than or equal to 3. After adjusting for presence of bystander cardiopulmonary resuscitation, witnessed arrest, and estimated time to return of spontaneous circulation, the pre- to post-test probability of unfavorable outcome with a MRI score greater than or equal to 3 went from 71% to greater than 99% (0.996 [95% CI, 0.8-1.0]).

Conclusions: Our simple six-point MRI scoring system developed in a 2016-2020 cohort of pediatric OHCA cases managed in two centers shows an association with outcome, with a post-test probability of unfavorable outcome greater than 99% with a MRI score greater than or equal to 3.