Treatment of Refractory Status Epilepticus With Continuous Intravenous Anesthetic Drugs: A Systematic Review.

Importance: Multiple continuous intravenous anesthetic drugs (CIVADs) are available for the treatment of refractory status epilepticus (RSE). There is a paucity of data comparing the different types of CIVADs used for RSE. Objective: To systematically review and compare outcome measures associated with the initial CIVAD choice in RSE in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Evidence Review: Data sources included English and non-English articles using Embase, MEDLINE, PubMed, and Web of Science (January 1994-June 2023) as well as manual search. Study selection included peer-reviewed studies of 5 or more patients and at least 1 patient older than 12 years with status epilepticus refractory to a benzodiazepine and at least 1 standard antiseizure medication, treated with continuously infused midazolam, ketamine, propofol, pentobarbital, or thiopental. Independent extraction of articles was performed using prespecified data items. The association between outcome variables and CIVAD was examined with an analysis of variance or χ2 test where appropriate. Binary logistic regressions were used to examine the association between outcome variables and CIVAD with etiology, change in mortality over time, electroencephalography (EEG) monitoring (continuous vs intermittent), and treatment goal (seizure vs burst suppression) included as covariates. Risk of bias was addressed by listing the population and type of each study. Findings: A total of 66 studies with 1637 patients were included. Significant differences among CIVAD groups in short-term failure, hypotension, and CIVAD substitution during treatment were observed. Non-epilepsy-related RSE (vs epilepsy-related RSE) was associated with a higher rate of CIVAD substitution (60 of 120 [50.0%] vs 11 of 43 [25.6%]; odds ratio [OR], 3.11; 95% CI, 1.44-7.11; P = .006) and mortality (98 of 227 [43.2%] vs 7 of 63 [11.1%]; OR, 17.0; 95% CI, 4.71-109.35; P < .001). Seizure suppression was associated with mortality (OR, 7.72; 95% CI, 1.77-39.23; P = .005), but only a small subgroup was available for analysis (seizure suppression: 17 of 22 [77.3%] from 3 publications vs burst suppression: 25 of 98 [25.5%] from 12 publications). CIVAD choice and EEG type were not predictors of mortality. Earlier publication year was associated with mortality, although the observation was no longer statistically significant after adjusting SEs for clustering. Conclusions and Relevance: Epilepsy-related RSE was associated with lower mortality compared with other RSE etiologies. A trend of decreasing mortality over time was observed, which may suggest an effect of advances in neurocritical care. The overall data are heterogeneous, which limits definitive conclusions on the choice of optimal initial CIVAD in RSE treatment.

Neurological Critical Care Services' Influence Following Large Hemispheric Infarction and Their Impact on Resource Utilization.

INTRODUCTION: Acute ischemic stroke (AIS) is the fourth leading cause of death in the US. Numerous studies have demonstrated the use of comprehensive stroke units and neurological intensive care units (NICU) in improving outcomes after stroke. We hypothesized that an expanded neurocritical care (NCC) service would decrease resource utilization in patients with LHI. METHODS: Retrospective data from consecutive admissions of large hemispheric infarction (LHI) patients requiring mechanical ventilation were acquired from the hospital medical records. Between 2011-2013, there were 187 consecutive patients admitted to the Jefferson Hospital for Neuroscience (Philadelphia, USA) with AIS and acute respiratory failure. Our intention was to determine the number of tracheostomies done over time. The primary outcome measure was the number of tracheostomies over time. Secondary outcomes were, ventilator-free days (Vfd), total hospital charges, intensive care unit length of stay (ICU-LOS), and total hospital length of stay (hospital-LOS), including ICU LOS. Hospital charges were log-transformed to meet assumptions of normality and homoscedasticity of residual variance terms. Generalized Linear Models were used and ORs and 95% CIs calculated. The significance level was set at α = 0.05. RESULTS: Of the 73 patients included in this analysis, 33% required a tracheostomy. There was a decrease in the number of tracheostomies undertaken since 2011. (OR 0.8; 95% CI 0.6-0.9: p=0.02).Lower Vfd were seen in tracheostomized patients (OR 0.11; 95%CI 0.1-0.26: p<0.0001). The log-hospital charges decreased over time but not significantly (OR 0.9; 95%CI 0.78-1.07: p=0.2) and (OR 0.99; 95%CI 0.85-1.16: p=0.8) from 2012 to 2013 respectively.The ICU-LOS at 23 days vs 10 days (p=0.01) and hospital-LOS at 33 days vs 11 days (p=0.008) were higher in tracheostomized patients. CONCLUSION: The data suggest that in LHI-patients requiring mechanical ventilation, a dedicated NCC service reduces the overall need for tracheostomy, increases Vfd, and decreases ICU and hospital-LOS.