A systematic review and meta-analysis of ketamine in pediatric status epilepticus.

OBJECTIVE: Status epilepticus (SE) is a common neurological medical emergency in the pediatric population, with 10%-40% of cases progressing to refractory SE (RSE), requiring treatment with anesthetic infusions. We present a systematic review and meta-analysis of the use of ketamine for the treatment of pediatric SE and its potential advantages over other anesthetic infusions. METHODS: This review follows the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement. Electronic databases, including PubMed, Cochrane Library, Ovid, Embase, and Google Scholar, were searched with the keywords "pediatrics," "status epilepticus," and "ketamine treatment." Randomized trials, prospective and retrospective cohort studies, and case reports were considered for inclusion. RESULTS: Eighteen publications met the initial inclusion criteria. The 18 publications comprise 11 case reports, one nonconclusive clinical trial, two case series, and four retrospective cohorts. After excluding the case reports because of reporting bias, only the six case series and cohorts were included in the final analysis. There were 172 patients in the six included studies. The weighted age was 9.93 (SD = 10.29) years. The weighted maximum dose was 7.44 (SD = 9.39) mg/kg/h. SE cessation was attained in 51% (95% confidence interval = 43-59) of cases with the addition of ketamine. The heterogeneity was I2 = 0%, t2 = 0, χ2 (5) = 3.39 (p = .64). SIGNIFICANCE: Pediatric RSE is difficult to treat, resulting in increased morbidity and mortality. Without strong recommendations and evidence regarding preferred agents, this review provides evidence that ketamine may be considered in managing SE in the pediatric population.

Caring for Critically Ill Children With the ICU Liberation Bundle (ABCDEF): Results of the Pediatric Collaborative.

OBJECTIVES: Assess clinical outcomes following PICU Liberation ABCDEF Bundle utilization. DESIGN: Prospective, multicenter, cohort study. SETTING: Eight academic PICUs. PATIENTS: Children greater than 2 months with expected PICU stay greater than 2 days and need for mechanical ventilation (MV). INTERVENTIONS: ABCDEF Bundle implementation. MEASUREMENT AND MAIN RESULTS: Over an 11-month period (3-mo baseline, 8-mo implementation), Bundle utilization was measured for 622 patients totaling 5,017 PICU days. Risk of mortality was quantified for 532 patients (4,275 PICU days) for correlation between Bundle utilization and MV duration, PICU length of stay (LOS), delirium incidence, and mortality. Utilization was analyzed as subject-specific (entire PICU stay) and day-specific (single PICU day). Median overall subject-specific utilization increased from 50% during the 3-month baseline to 63.9% during the last four implementation months ( p < 0.001). Subject-specific utilization for elements A and C did not change; utilization improved for B (0-12.5%; p = 0.007), D (22.2-61.1%; p < 0.001), E (17.7-50%; p = 0.003), and F (50-79.2%; p = 0.001). We observed no association between Bundle utilization and MV duration, PICU LOS, or delirium incidence. In contrast, on adjusted analysis, every 10% increase in subject-specific utilization correlated with mortality odds ratio (OR) reduction of 34%, p < 0.001; every 10% increase in day-specific utilization correlated with a mortality OR reduction of 1.4% ( p = 0.006). CONCLUSIONS: ABCDEF Bundle is applicable to children. Although enhanced Bundle utilization correlated with decreased mortality, increased utilization did not correlate with duration of MV, PICU LOS, or delirium incidence. Additional research in the domains of comparative effectiveness, implementation science, and human factors engineering is required to understand this clinical inconsistency and optimize PICU Liberation concept integration into clinical practice.

Pediatric Chronic Critical Illness: Validation, Prevalence, and Impact in a Children's Hospital.

OBJECTIVES: Large populations of chronically critically ill patients test the critical care system's resource utilization ability. Defining and tracking this group is necessary for census predictions. DESIGN: Retrospective cohort analysis. SETTINGS: Tertiary academic center in United States. PATIENTS: Patients admitted to PICU or neonatal ICU. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Demographics and resource utilization variables were sampled in PICU, intermediate care unit, cardiovascular ICU, and neonatal ICU on 3 random days in 3 consecutive months. The attendings' perception of pediatric chronic critical illness was contrasted to Shapiro's definition of chronic critical care criteria. Each unit's active and maximal capacity census was computed: the occurrence rate of pediatric chronic critical illness was 34%, the prevalence was 44.5%, and the tolerance, or percentage pediatric chronic critical illness patients to all available beds, was 36.8%. The median length of stay for the nonpatients with pediatric critical care illness to patients with pediatric critical care illness was 9 versus 46 days (1/5.1). The attending's decision was 58 times more concordant with the criteria. Pediatric chronic critical illness bed occupancy was 40.6% in PICU, 97.2% in intermediate care unit, 47.8% in cardiovascular ICU, and 33.9% in neonatal ICU. CONCLUSIONS: Pediatric chronic critical illness patients occupied more than one third of the ICU beds and have five times longer stay. This mounting load needs to be uniformly defined, addressed at regional and national levels, and considered in the current pandemic planning.

SNAP II index: an alternative to the COMFORT scale in assessing the level of sedation in mechanically ventilated pediatric patients.

Sedation monitoring is essential in pediatric patients on ventilatory support to achieve comfort and safety. The COMFORT scale was designed and validated to assess the level of sedation in intubated pediatric patients. However, it remains unreliable in pharmacologically paralyzed patients. The SNAP II index is calculated using an algorithm that incorporates high-frequency (80-420 Hz) electroencephalogram (EEG) components, known to be useful in discriminating between awake and unconscious states, unlike other measurements that only include low-frequency EEG segments such as the bispectral index score. Previous studies suggested that the SNAP II index is a reliable and sensitive indicator of the level of consciousness in adult patients. Despite its potential, no data are currently available in the pediatric critically ill population on ventilatory support. This is the first pilot study assessing the potential application of the SNAP II index in critically ill pediatric patients by comparing it to the commonly used COMFORT scale.

Prospective assessment of novice learners in a simulation-based extracorporeal membrane oxygenation (ECMO) education program.

This study aimed to assess the impact of integrating a simulation-based education module into an extracorporeal membrane oxygenation (ECMO) curriculum on novice learners and to test the duration of time that skills obtained during this training exercise were retained. The authors hypothesized that multidisciplinary, simulation-based ECMO training would improve comfort and confidence levels among participants. An ECMO training curriculum was developed that incorporated in situ simulation modules to train multidisciplinary health care professionals involved in the management of patients receiving ECMO in the pediatric cardiac intensive care unit (PCICU). During the simulation, a team was assembled similar to the one that would staff the PCICU during a routine workday. Pre- and postparticipation questionnaires were used to determine the effects on the knowledge, ability, and confidence level of the participants. The participants were required to repeat the simulation test within 6-8 months. The study enrolled 26 providers (10 fellow physicians, 12 nurses and nurse practitioners, 4 respiratory therapists). All except one had no previous training in the management of ECMO. Of the 26 participants, 24 passed the initial written and practical tests. One participant failed the written test, whereas another failed the practical test. All the responding participants scored the didactic and scenarios education as useful, at 4 or higher (5 = very useful), in improving their perception of their overall knowledge and their ability to perform the required critical performance criteria on simulated ECMO. The 20 participants who appeared for the 6 month follow-up visit to assess maintenance of competency skills demonstrated success with simulated ECMO emergencies. All four questionnaires were completed by 18 participants. Simulation-based training is an effective method of improving knowledge, ability, and confidence levels among novice ECMO specialists and physician trainees. Further research is needed to assess real-time demonstration of skills retention during ECMO emergencies.

Peripheral vascular access as exclusive access mode in pediatric intensive care unit.

Introduction: The type of vascular access (central or peripheral) in pediatric critical care depends on several factors, including the duration of treatment, the properties of the medication (osmolarity or vesicant), and the need for central pressure monitoring. The utilization of peripheral intravascular catheters (PIVCs) has shown a notable increase in the number of patients being treated. Extended dwell or midline catheters are another peripheral access option in addition to PIVCs. However, there are currently no established guidelines on their placement. Objectives: The aim of this study is to estimate the duration of dwell time for PIVCs, analyze the specific parameters affecting it, and develop recommendations for switching to extended dwell and midline catheter placement as an alternative to peripheral access. Methods: The study enrolled patients aged 0-18 years admitted to the pediatric intensive care unit (PICU) for over 24 h and managed with peripheral access only over 2 years (2019-2021). Measurements and main results: A total of 484 patients met the specified criteria. Patients who had peripheral access exhibited a lower PRISM score and a shorter length of stay in the PICU, with mean values of 18 (SD: 8.5) and 9.5 (SD: 6.4) days, respectively, compared with patients who had central access with mean values of 8.9 (SD: 5.9) and 5.7 (SD: 3.6) days, respectively. The PIVC dwell time was found to be 50.1 h (SD: 65.3) and required an average of 1.6 insertion attempts. Patients with three or more insertions exhibited an increased odds ratio of 5.2 (95% CI: 3.1-8.5) for receiving an extended dwell or midline insertion. Increased dwell time was associated with female gender, 59.5 h (P < 0.001), first attempt insertion, 53.5 h (P < 0.001), use of 24 Ga bore, 56.3 h (P = 0.04), left-sided insertions, 54.9 (P = 0.07), less agitation, 54.8 h (P = 0.02), and less edema, 61.6 (P < 0.001). Decreased dwell time was associated with the use of vancomycin infusion at 24.2 h (P < 0.001) and blood transfusions at 29.3 h (P < 0.001). Conclusions: Extended catheters last longer than PIVCs in PICU patients. Extended catheter placement requires consideration of the length of treatment, as well as the overall body edema, the level of the patient's restlessness, and the need for vancomycin infusion or blood transfusions, as these factors reduce PIVC dwell time and expose the patients to painful insertions. For such cases, an extended dwell catheter may be a better option, even if the projected treatment time is less than 6 days.

Ketamine inhibits tumor necrosis factor secretion by RAW264.7 murine macrophages stimulated with antibiotic-exposed strains of community-associated, methicillin-resistant Staphylococcus aureus.

BACKGROUND: Infections caused by community-associated strains of methicillin-resistant Staphylococcus aureus (CA-MRSA) are associated with a marked and prolonged host inflammatory response. In a sepsis simulation model, we tested whether the anesthetic ketamine inhibits the macrophage TNF response to antibiotic-exposed CA-MRSA bacteria via its antagonism of N-methyl-D-aspartate (NMDA) receptors. RAW264.7 cells were stimulated for 18 hrs with 105 to 107 CFU/mL inocula of either of two prototypical CA-MRSA isolates, USA300 strain LAC and USA400 strain MW2, in the presence of either vancomycin or daptomycin. One hour before bacterial stimulation, ketamine was added with or without MK-801 (dizocilpine, a chemically unrelated non-competitive NMDA receptor antagonist), APV (D-2-amino-5-phosphono-valerate, a competitive NMDA receptor antagonist), NMDA, or combinations of these agents. Supernatants were collected and assayed for TNF concentration by ELISA. RESULTS: RAW264.7 cells exposed to either LAC or MW2 in the presence of daptomycin secreted less TNF than in the presence of vancomycin. The addition of ketamine inhibited macrophage TNF secretion after stimulation with either of the CA-MRSA isolates (LAC, MW2) in the presence of either antibiotic. The NMDA inhibitors, MK-801 and APV, also suppressed macrophage TNF secretion after stimulation with either of the antibiotic-exposed CA-MRSA isolates, and the effect was not additive or synergistic with ketamine. The addition of NMDA substrate augmented TNF secretion in response to the CA-MRSA bacteria, and the addition of APV suppressed the effect of NMDA in a dose-dependent fashion. CONCLUSIONS: Ketamine inhibits TNF secretion by MRSA-stimulated RAW264.7 macrophages and the mechanism likely involves NMDA receptor antagonism. These findings may have therapeutic significance in MRSA sepsis.

Role of bacterial components in macrophage activation by the LAC and MW2 strains of community-associated, methicillin-resistant Staphylococcus aureus.

We tested the contribution of four staphylococcal components - PSM-α, PSM-ß, δ-toxin, and PVL - in triggering macrophage secretion of tumor necrosis factor (TNF) and interleukins 6 (IL-6) and 12 (IL-12) by two prominent, circulating strains of community-associated, methicillin-resistant Staphylococcus aureus (CA-MRSA): LAC, USA300; MW2, USA400. RAW 264.7 murine macrophages were stimulated with live, antibiotic-exposed bacteria, and cytokine secretion was quantitated in supernatants. Deletion of PSM-α expression in LAC led to >50% reduction in macrophage TNF and IL-6 secretion and a 20% reduction in IL-12 secretion, while PSM-α deletion in MW2 did not significantly reduce macrophage TNF secretion but resulted in a 15-20% reduction in IL-6 and IL-12 secretion. Deletion of δ-toxin in either strain led to more than 50% reduction in macrophage IL-6 secretion and smaller reductions in macrophage TNF and IL-12 secretion (8-25%). Our data implicate both PSM-α and δ-toxin in stimulating macrophage cytokine responses to CA-MRSA bacteria.

Impact of Different Cardiopulmonary Bypass Strategies on Renal Injury After Pediatric Heart Surgery.

BACKGROUND: The purpose of this study is to compare the incidence and severity of acute kidney injury (AKI) after open heart surgery in neonates and infants for two different cardiopulmonary bypass (CPB) strategies. METHODS: In all, 151 infants undergoing cardiac surgery were prospectively enrolled between June 2017 and June 2018 at two centers, one using conventional CPB (2.4 L · min-1 · m-2, 150 mL · kg-1 · min-1) with reduction of flow rates with moderate hypothermia and with a targeted hematocrit greater than 25% (center 1, n = 91), and the other using higher bypass flow rates (175 to 200 mL · kg-1 · min-1) and higher minimum hematocrit (greater than 32%) CPB (center 2, n = 60). The primary endpoint was the incidence of postoperative AKI as defined by Acute Kidney Injury Network criteria and risk factors associated with AKI. RESULTS: Preoperative characteristics and complexity of surgery were comparable between centers. The overall incidence of early postoperative AKI was 10.6% (16 of 151), with 15.4% (14 of 91) in center 1 and 3.3% (2 of 60) in center 2 (P = .02). Mean lowest flow rates on CPB were 78 mL · kg-1 · min-1 vs 118 mL · kg-1 · min-1 and mean highest hematocrit on separation from CPB were 33% vs 43% at center 1 and 2, respectively (P < .001). Center 1 used less packed red blood cells but more fresh frozen plasma than center 2 (P = .001). By multivariate analysis, only lower flows on CPB (78 vs 96 mL · kg-1 · min-1, P = .043) and lower hematocrit at the end of CPB (33% vs 37%, P = .007) were associated with AKI. CONCLUSIONS: In this contemporary comparative study, higher flow rates and higher hematocrit during cardiopulmonary bypass were associated with better preservation of renal function.

Correlation of intraocular pressure with intracranial pressure in children with severe head injuries.

OBJECTIVE: To determine whether there was a correlation between tonometric measurements of the intraocular pressure and transducer measurements of the intracranial pressure in the acute setting, and whether intraocular pressure can be used as a surrogate measure of intracranial pressure. Children with traumatic brain injuries commonly develop increased intracranial pressure requiring surgical placement of a pressure transducer to measure the intracranial pressure during the acute recovery period. The increased intracranial pressure may cause engorgement of the orbital compartments via dilation of the episcleral veins and manifest as increased intraocular pressure. DESIGN: Prospective study. SETTING: Tertiary academic pediatric intensive care unit. PATIENTS: Children admitted with severe traumatic brain injury. INTERVENTIONS: Tonometric intraocular pressure measurements. MEASUREMENTS AND MAIN RESULTS: We performed an Institutional Review Board-approved, prospective study on 36 children (age range, 2.9-15.1 yrs) with traumatic brain injuries, requiring intracranial pressure monitoring. A total of 274 intraocular pressure measurements were made after placement of the pressure transducer, and concordance between the sites of injury and measurement was documented. The average age of the patients was 8.3 yrs. The mean intraocular pressure, intracranial pressure difference was -0.5 +/- 0.68 cm H2O, and the variance was 29.88 (sd, 5.47). The 95% confidence interval was between -11.22 and 10.22. With concordance between the sites of measurement and injury, the mean IOP, intracranial pressure difference was -0.02 +/- 0.61 cm H2O (variance, 23.28; sd, 4.82; 95% confidence interval, - 9.47 to 9.42). Concordance reduced the variance of the intraocular pressure, intracranial pressure discrepancy by 20.3%. The Pearson intraocular pressure-intracranial pressure regression coefficient and the Krippendorff's alpha reliability estimate analyses indicated good agreement. The patient's age or Paco2 did not influence the intraocular pressure, intracranial pressure difference. Using 20 cm H2O as a normal intracranial pressure cutoff, the intraocular pressure had a specificity of 0.7 and sensitivity of 0.97; with concordance, the values improved to 0.78 and 0.96, respectively. CONCLUSIONS: Tonometry is a useful screening surrogate measure of intracranial pressure in children with traumatic brain injuries, but seems to lack the accuracy necessary for close management of intracranial pressure in the acute posttraumatic period.

Brain tumor resection in children: neurointensive care unit course and resource utilization.

OBJECTIVE: To describe the pediatric intensive care unit (PICU) course and resource utilization for children with brain tumor resection and to identify factors predicting prolonged (>1 day) PICU length of stay. After craniotomy for brain tumor resection, children recover in the PICU. A few require critical care interventions and a >24-hr length of stay. DESIGN: We reviewed all brain tumor resection patients admitted to the PICU over 2 yrs. Preoperative, intraoperative, and postoperative variables and tumor characteristics were examined. The extracted variables were compared between two groups with a length of stay in the PICU of >1 or <1 day. SETTING: Pediatric intensive care unit in a tertiary academic children's medical center. PATIENTS: A total of 105 patients post brain tumor resection were admitted to the PICU over the study period and analyzed. INTERVENTIONS: Record review. MEASUREMENTS AND MAIN RESULTS: Thirty-two (31%) of 105 patients remained in the PICU for >1 day. The mean age of patients in the >1 day group was 5.0 ± 0.81 yrs and 8.78 ± 0.65 yrs in the <1 day group (p < .05). The estimated blood loss was 20 ± 2.37 mL/kg in the >1 day and 9 ± 0.92 mL/kg in the <1 day group (p < .05). Fifteen (14.3%) patients were mechanically ventilated on arrival in the PICU; these patients more often had a length of stay of >1 day (p < .05). The number of unexpected intensive care unit interventions were 0.7 per patient, were more common in the >1 day group, and included treatment of sodium abnormalities, new neurologic deficits, paresis, or seizures (p < .05). In a logistic regression model, estimated blood loss and intubation on arrival predicted longer lengths of stay in the PICU (odds ratio, 1.1; 95% confidence interval, 1.05-1.18; and odds ratio, 33; 95% confidence interval, 2.57-333, respectively), with a receiver operating characteristic curve of 0.86 and 95% confidence interval, 0.78-0.94. CONCLUSIONS: Large intraoperative estimated blood loss and intubation on arrival may be predictive of PICU lengths of stay of >1 day for children who have had a craniotomy for brain tumor resection. Intensive care unit interventions are more common in these children.