Critical Care Medicine

Outcome Prediction in Postanoxic Coma With Deep Learning Objectives: Visual assessment of the electroencephalogram by experienced clinical neurophysiologists allows reliable outcome prediction of approximately half of all comatose patients after cardiac arrest. Deep neural networks hold promise to achieve similar or even better performance, being more objective and consistent. Design: Prospective cohort study. Setting: Medical ICU of five teaching hospitals in the Netherlands. Patients: Eight-hundred ninety-five consecutive comatose patients after cardiac arrest. Interventions: None. Measurements and Main Results: Continuous electroencephalogram was recorded during the first 3 days after cardiac arrest. Functional outcome at 6 months was classified as good (Cerebral Performance Category 1–2) or poor (Cerebral Performance Category 3–5). We trained a convolutional neural network, with a VGG architecture (introduced by the Oxford Visual Geometry Group), to predict neurologic outcome at 12 and 24 hours after cardiac arrest using electroencephalogram epochs and outcome labels as inputs. Output of the network was the probability of good outcome. Data from two hospitals were used for training and internal validation (n = 661). Eighty percent of these data was used for training and cross-validation, the remaining 20% for independent internal validation. Data from the other three hospitals were used for external validation (n = 234). Prediction of poor outcome was most accurate at 12 hours, with a sensitivity in the external validation set of 58% (95% CI, 51–65%) at false positive rate of 0% (CI, 0–7%). Good outcome could be predicted at 12 hours with a sensitivity of 48% (CI, 45–51%) at a false positive rate of 5% (CI, 0–15%) in the external validation set. Conclusions: Deep learning of electroencephalogram signals outperforms any previously reported outcome predictor of coma after cardiac arrest, including visual electroencephalogram assessment by trained electroencephalogram experts. Our approach offers the potential for objective and real time, bedside insight in the neurologic prognosis of comatose patients after cardiac arrest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http:/journals.lww.com/ccmjournal). Dr. van Putten is co-founder of Clinical Science Systems, a supplier of electroencephalogram systems for Medisch Spectrum Twente. The remaining authors have disclosed that they do not have any conflicts of interest. This work was performed in Medisch Spectrum Twente, Rijnstate Hospital, St. Antonius Hospital, University Medical Center Groningen and VieCuri Medical Center, The Netherlands. For information regarding this article, E-mail: m.tjepkema-cloostermans@mst.nl Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

XueBiJing Injection Versus Placebo for Critically Ill Patients With Severe Community-Acquired Pneumonia: A Randomized Controlled Trial Objectives: To investigate whether XueBiJing injection improves clinical outcomes in critically ill patients with severe community-acquired pneumonia. Design: Prospective, randomized, controlled study. Setting: Thirty-three hospitals in China. Patients: A total of 710 adults 18–75 years old with severe community-acquired pneumonia. Interventions: Participants in the XueBiJing group received XueBiJing, 100 mL, q12 hours, and the control group received a visually indistinguishable placebo. Measurements and Main Results: The primary outcome was 8-day improvement in the pneumonia severity index risk rating. Secondary outcomes were 28-day mortality rate, duration of mechanical ventilation and total duration of ICU stay. Improvement in the pneumonia severity index risk rating, from a previously defined endpoint, occurred in 203 (60.78%) participants receiving XueBiJing and in 158 (46.33%) participants receiving placebo (between-group difference [95% CI], 14.4% [6.9–21.8%]; p < 0.001). Fifty-three (15.87%) XueBiJing recipients and 84 (24.63%) placebo recipients (8.8% [2.4–15.2%]; p = 0.006) died within 28 days. XueBiJing administration also decreased the mechanical ventilation time and the total ICU stay duration. The median mechanical ventilation time was 11.0 versus 16.5 days for the XueBiJing and placebo groups, respectively (p = 0.012). The total duration of ICU stay was 12 days for XueBiJing recipients versus 16 days for placebo recipients (p = 0.004). A total of 256 patients experienced adverse events (119 [35.63%] vs 137 [40.18%] in the XueBiJing and placebo groups, respectively [p = 0.235]). Conclusions: In critically ill patients with severe community-acquired pneumonia, XueBiJing injection led to a statistically significant improvement in the primary endpoint of the pneumonia severity index as well a significant improvement in the secondary clinical outcomes of mortality, duration of mechanical ventilation and duration of ICU stay. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). Supported, in part, by a Tianjin Science and Technology committee grant (14ZXLJSY00230) and National Natural Science Foundation of China (81630001,81490533). Drs. X. Yu and Zhi Liu disclosed work for hire. Dr. B. Zhang disclosed government work. The remaining authors have disclosed that they do not have any potential conflicts of interest. Clinical Trial Registration: http://www.chictr.org.cn/index.aspx. Unique identifier: ChiCTR-TRC-13003534. For information regarding this article, E-mail: bai.chunxue@zs-hospital.sh.cn; shanghongcai@126.com Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Intracranial Hypertension and Cerebral Perfusion Pressure Insults in Adult Hypertensive Intraventricular Hemorrhage: Occurrence and Associations With Outcome Objectives: Elevated intracranial pressure and inadequate cerebral perfusion pressure may contribute to poor outcomes in hypertensive intraventricular hemorrhage. We characterized the occurrence of elevated intracranial pressure and low cerebral perfusion pressure in obstructive intraventricular hemorrhage requiring extraventricular drainage. Design: Prospective observational cohort. Setting: ICUs of 73 academic hospitals. Patients: Four hundred ninety-nine patients enrolled in the CLEAR III trial, a multicenter, randomized study to determine if extraventricular drainage plus intraventricular alteplase improved outcome versus extraventricular drainage plus saline. Interventions: Intracranial pressure and cerebral perfusion pressure were recorded every 4 hours, analyzed over a range of thresholds, as single readings or spans (≥ 2) of readings after adjustment for intracerebral hemorrhage severity. Impact on 30- and 180-days modified Rankin Scale scores was assessed, and receiver operating curves were analyzed to identify optimal thresholds. Measurements and Main Results: Of 21,954 intracranial pressure readings, median interquartile range 12 mm Hg (8–16), 9.7% were greater than 20 mm Hg and 1.8% were greater than 30 mm Hg. Proportion of intracranial pressure readings from greater than 18 to greater than 30 mm Hg and combined intracranial pressure greater than 20 plus cerebral perfusion pressure less than 70 mm Hg were associated with day-30 mortality and partially mitigated by intraventricular alteplase. Proportion of cerebral perfusion pressure readings from less than 65 to less than 90 mm Hg and intracranial pressure greater than 20 mm Hg in spans were associated with both 30-day mortality and 180-day mortality. Proportion of cerebral perfusion pressure readings from less than 65 to less than 90 mm Hg and combined intracranial pressure greater than 20 plus cerebral perfusion pressure less than 60 mm Hg were associated with poor day-30 modified Rankin Scale, whereas cerebral perfusion pressure less than 65 and less than 75 mm Hg were associated with poor day-180 modified Rankin Scale. Conclusions: Elevated intracranial pressure and inadequate cerebral perfusion pressure are not infrequent during extraventricular drainage for severe intraventricular hemorrhage, and level and duration predict higher short-term mortality and long-term mortality. Burden of low cerebral perfusion pressure was also associated with poor short- and long-term outcomes and may be more significant than intracranial pressure. Adverse consequences of intracranial pressure-time burden and cerebral perfusion pressure-time burden should be tested prospectively as potential thresholds for therapeutic intervention. A full list of the Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage (CLEAR III). Investigators are available in Appendix 1 (Supplemental Digital Content 9, http://links.lww.com/CCM/E727). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). Supported, in part, by the National Institutes of Health grants 5U01NS062851 for Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage (CLEAR III). Genentech assisted by donating the study drug. Dr. Ziai received funding from C. R. Bard, Headsense, and NINDS. Drs. Ziai, Thompson, and Mayo, Ms. McBee, Dr. Ullman, Ms. Lane, and Drs. Awad and Hanley received support for article research from National Institutes of Health (NIH). Drs. Thompson's and Award's institutions received funding from the NIH. Ms. McBee's, Ms. Lane's, and Dr. Hanley's institutions received funding from NIH/NINDS 5U01NS062851, NIH/NINDS 1U01NS08082, and Genentech (drug donation). Ms. McBee, Dr. Ullman, Ms. Lane, and Drs. Awad and Hanley disclosed off-label product use of alteplase. Dr. Ullman's institution received funding from NIH/NINDS. Dr. Awad received funding from expert reviews and medicolegal opinions. Dr. Hanley received funding from BrainScope, Neurotrope, Portola Pharmaceuticals, Op2Lysis, HeadSense, and Medtronic. Dr. Ziai, Ms. McBee, Ms. Lane, and Drs. Awad and Hanley are supported by grant 5U01NS062851. The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: weziai@jhmi.edu Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Limiting Treatment in Intensive Care: Contributions and Limits of Ethics Consultation No abstract available

Impact of Critical Care Point-of-Care Ultrasound Short-Courses on Trainee Competence Objectives: Competence in point-of-care ultrasound is recommended/mandated by several critical care specialties. Although doctors commonly attend point-of-care ultrasound short-courses for introductory training, there is little follow-up data on whether they eventually attain competence. This study was done to determine the impact of point-of-care ultrasound short-courses on point-of-care ultrasound competence. Design: Web-based survey. Setting: Follow-up after point-of-care ultrasound short-courses in the Asia-Pacific region. Subjects: Doctors who attended a point-of-care ultrasound short-course between December 2015 and February 2018. Interventions: Each subject was emailed a questionnaire on or after 6 months following their short-course. They were asked if they had performed at least 30 structured point-of-care ultrasound scans and/or reached point-of-care ultrasound competence and their perceived reasons/challenges/barriers. They were also asked if they used point-of-care ultrasound as a clinical diagnostic aid. Measurements and Main Results: The response rate was 74.9% (182/243). Among the 182 respondents, only 12 (6.6%) had attained competence in their chosen point-of-care ultrasound modality, attributing their success to self-motivation and time management. For the remaining doctors who did not attain competence (170/182, 93.4%), the common reasons were lack of time, change of priorities, and less commonly, difficulties in accessing an ultrasound machine/supervisor. Common suggestions to improve short-courses included requests for scanning practice on acutely ill ICU patients and prior information on the challenges regarding point-of-care ultrasound competence. Suggestions to improve competence pathways included regular supervision and protected learning time. All 12 credentialled doctors regularly used point-of-care ultrasound as a clinical diagnostic aid. Of the 170 noncredentialled doctors, 123 (72.4%) reported performing unsupervised point-of-care ultrasound for clinical management, either sporadically (42/170, 24.7%) or regularly (81/170, 47.7%). Conclusions: In this survey of doctors attending point-of-care ultrasound short-courses in Australasia, the majority of doctors did not attain competence. However, the practice of unsupervised point-of-care ultrasound use by noncredentialled doctors was common. Further research into effective strategies to improve point-of-care ultrasound competence is required. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: rrarvind@hotmail.com Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Why Understanding Sepsis Endotypes Is Important for Steroid Trials in Septic Shock? No abstract available

Fluid Overload Associates With Major Adverse Kidney Events in Critically Ill Patients With Acute Kidney Injury Requiring Continuous Renal Replacement Therapy Objectives: We examined the association between fluid overload and major adverse kidney events in critically ill patients requiring continuous renal replacement therapy for acute kidney injury. Design: Retrospective cohort study. Setting: ICU in a tertiary medical center. Patients: Four-hundred eighty-one critically ill adults requiring continuous renal replacement therapy for acute kidney injury. Interventions: None. Measurements and Main Results: Fluid overload was assessed as fluid balance from admission to continuous renal replacement therapy initiation, adjusted for body weight. Major adverse kidney events were defined as a composite of mortality, renal replacement therapy-dependence or inability to recover 50% of baseline estimated glomerular filtration rate (if not on renal replacement therapy) evaluated up to 90 days after discharge. Patients with fluid overload less than or equal to 10% were less likely to experience major adverse kidney events than those with fluid overload greater than 10% (71.6% vs 79.4%; p = 0.047). Multivariable logistic regression showed that fluid overload greater than 10% was associated with a 58% increased odds of major adverse kidney events (p = 0.046), even after adjusting for timing of continuous renal replacement therapy initiation. There was also a 2.7% increased odds of major adverse kidney events for every 1 day increase from ICU admission to continuous renal replacement therapy initiation (p = 0.024). Fluid overload greater than 10% was also found to be independently associated with an 82% increased odds of hospital mortality (p = 0.004) and 2.5 fewer ventilator-free days (p = 0.044), compared with fluid overload less than or equal to 10%. Conclusions: In critically ill patients with acute kidney injury requiring continuous renal replacement therapy, greater than 10% fluid overload was associated with higher risk of 90-day major adverse kidney events, including mortality and decreased renal recovery. Increased time between ICU admission and continuous renal replacement therapy initiation was also associated with decreased renal recovery. Fluid overload represents a potentially modifiable risk factor, independent of timing of continuous renal replacement therapy initiation, that should be further examined in interventional studies. Dr. Woodward designed the project, managed the database, gathered the data, prepared the figures, and wrote and submitted the article. Dr. Lambert and Ms. Li managed the database, analyzed the statistical data, and prepared the figures. Drs. Ortiz-Soriano, Bissell, Adams, Yessayan, and Morris designed the project and edited the article. Dr. Ruiz-Conejo gathered data and wrote the article. Mr. Kelly designed the project, managed the database, and extracted data. Dr. Neyra designed and supervised the project, and wrote and submitted the article. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). Dr. Lambert received support for article research from the National Institutes of Health. Dr. Neyra is currently supported by an Early Career Pilot Grant from the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1TR001998. The remaining authors have disclosed that they do not have any potential conflicts of interest. This work was performed at the University of Kentucky, Lexington, KY. For information regarding this article, E-mail: javier.neyra@uky.edu Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

A Progressive Early Mobilization Program Is Significantly Associated With Clinical and Economic Improvement: A Single-Center Quality Comparison Study Objectives: To determine whether a progressive early mobilization protocol improves patient outcomes, including in-hospital mortality and total hospital costs. Design: Retrospective preintervention and postintervention quality comparison study. Settings: Single tertiary community hospital with a 12-bed closed-mixed ICU. Patients: All consecutive patients 18 years old or older were eligible. Patients who met exclusion criteria or were discharged from the ICU within 48 hours were excluded. Patients from January 2014 to May 2015 were defined as the preintervention group (group A) and from June 2015 to December 2016 was the postintervention group (group B). Intervention: Maebashi early mobilization protocol. Measurements and Main Results: Group A included 204 patients and group B included 187 patients. Baseline characteristics evaluated include age, severity, mechanical ventilation, and extracorporeal membrane oxygenation, and in group B additional comorbidities and use of steroids. Hospital mortality was reduced in group B (adjusted hazard ratio, 0.25; 95% CI, 0.13–0.49; p < 0.01). This early mobilization protocol is significantly associated with decreased mortality, even after adjusting for baseline characteristics such as sedation. Total hospital costs decreased from $29,220 to $22,706. The decrease occurred soon after initiating the intervention and this effect was sustained. The estimated effect was $–5,167 per patient, a 27% reduction. Reductions in ICU and hospital lengths of stay, time on mechanical ventilation, and improvement in physical function at hospital discharge were also seen. The change in Sequential Organ Failure Assessment score and Sequential Organ Failure Assessment score at ICU discharge were significantly reduced after the intervention, despite a similar Sequential Organ Failure Assessment score at admission and at maximum. Conclusions: In-hospital mortality and total hospital costs are reduced after the introduction of a progressive early mobilization program, which is significantly associated with decreased mortality. Cost savings were realized early after the intervention and sustained. Further prospective studies to investigate causality are warranted. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). Dr. Oosaki disclosed work for hire. The remaining authors have disclosed that they do not have any potential conflicts of interest. This study was conducted in the ICU of the Japan Red Cross Maebashi Hospital. For information regarding this article, E-mail: keiliu0406@gmail.com Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Impact of Telemonitoring of Critically Ill Emergency Department Patients Awaiting Intensive Care Unit Transfer Objectives: Because of overcrowding and limited critical care resources, critically ill patients in the emergency department may spend hours to days awaiting transfer to the ICU. In these patients, often termed "ICU boarders," delayed ICU transfer is associated with poor outcomes. We implemented an emergency department–based, electronic ICU monitoring system for ICU boarders. Our aim was to investigate the effect of this initiative on morbidity, mortality, and ICU usage. Design: Single-center, retrospective cohort study. Setting: Nonprofit, tertiary care, teaching hospital with greater than 100,000 emergency department visits per year. Patients: Emergency department patients with admission orders for the medical ICU, who spent more than 2 hours boarding in the emergency department after being accepted for admission to the medical ICU, were included in the study. Interventions: None. Measurements and Main Results: During the study period, a total of 314 patients were admitted to the medical ICU from the emergency department, 214 of whom were considered ICU boarders with a delay in medical ICU transfer over 2 hours. Of ICU boarders, 115 (53.7%) were enrolled in electronic ICU telemonitoring (electronic ICU care), and the rest received usual emergency department care (emergency department care). Age, mean illness severity (Acute Physiology and Chronic Health Evaluation IVa scores), and admitting diagnoses did not differ significantly between ICU boarders receiving electronic ICU care and emergency department care. Forty-one electronic ICU care patients (36%) were ultimately transitioned to a less intensive level of care in lieu of ICU admission while still in the emergency department, compared with zero patients in the emergency department care group. Among all ICU boarders transferred to the ICU, in-hospital mortality was lower in the electronic ICU care cohort when compared with the emergency department care cohort (5.4% vs 20.0%; adjusted odds ratio, 0.08). Conclusions: In critically ill patients awaiting transfer from the emergency department to the medical ICU, electronic ICU care was associated with decreased mortality and lower ICU resource utilization. Current address for Dr. Kadar: 222 East Pearson St, Apt 1601, Chicago, IL 60611. Dr. Kadar, Ms. Hesse, and Dr. Bonder's institutions received funding from Be the Change Grant award from the Emergency Medicine Residents' Association. Mr. Amici was supported by National Institutes of Health T32GM008152. The remaining authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: rachel.burt.kadar@gmail.com Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Circulating Histones Are Major Mediators of Multiple Organ Dysfunction Syndrome in Acute Critical Illnesses Objectives: Multiple organ dysfunction syndrome is characterized by simultaneous multiple organ failure, which is the leading cause of death in acute critically ill patients. However, what mediates multiple organ dysfunction syndrome is not fully understood. The discovery of toxic effects by extracellular histones on different individual organs strongly suggests their involvement in multiple organ dysfunction syndrome. In this study, we investigate whether circulating histones are major mediators of multiple organ dysfunction syndrome in acute critical illnesses. Design: Combination of retrospective clinical studies and animal models with intervention. Setting: ICU in a tertiary hospital and research laboratories. Patients: Four hundred and twenty ICU patients, including sepsis (140), severe trauma (63), severe pancreatitis (89), and other admission diagnoses (128). Laboratory Investigation: Cells from major organs are treated with calf thymus histones or histone-containing sera. Animal models for sepsis, trauma, and acute pancreatitis are treated with antihistone reagents. Intervention: Antihistone reagents in in vitro, ex vivo, and animal models. Measurement and Main Results: Retrospective analysis of a prospectively recruited ICU cohort demonstrated a strong correlation between circulating histones and organ injury markers and Sequential Organ Failure Assessment scores. Ex vivo experiments showed that patient sera containing high histone levels were toxic to cultured cells from different origins, suggesting their universal toxicity to multiple organs. Animal models of sepsis, trauma, and pancreatitis further demonstrated a temporal correlation between histone levels and disease severity and multiple organ injury. Importantly, antihistone reagents, that is, antihistone single-chain variable fragment and nonanticoagulant heparin, could dramatically reduce multiple organ injury, particularly of the heart and lungs, and improve survival in mouse models. Conclusions: High levels of circulating histones are major mediators of multiple organ dysfunction syndrome. Our results indicate that monitoring upon ICU admission could inform on disease severity and developing antihistone therapy holds great potential of reducing multiple organ dysfunction syndrome and improving survival of critically ill patients. Dr. Cheng performed animal experiments and sample analysis. Dr. Cheng, Dr. Abrams, and Mr. Toh performed the cell viability assay. Dr. Abrams analyzed the clinical data and performed statistical analysis as well as figure and article editing. Dr. Alhamdi collected part of the clinical data. Drs. Yu, Wang, and Toh supervised the work and wrote, edited, and reviewed the article and figures. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website (http://journals.lww.com/ccmjournal). This work was supported by the Medical Research Council (G0501641), British Heart Foundation (PG/14/19/30751 and PG/16/65/32313), and National Institute of Health Research (II-FS-0110-14061)and received scholarships from China Scholarship Council (CSC:20160609156) and Southeast University (YBJJ1740) to Dr. Cheng. Dr. Cheng, Dr. Abrams, Mr. Toh, Dr. Yu, Dr. Wang, and Dr. Toh's institutions received funding from Medical Research Council (G0501641), British Heart Foundation (PG/14/19/30751 and PG/16/65/32313), and National Institute of Health Research (II-FS-0110-14061). Drs. Cheng, Abrams, Yu, Wang, and Toh received support for article research from Research Councils UK (RCUK). Drs. Cheng, Yu, and Wang disclosed government work. Dr. Alhamdi has disclosed that he does not have any potential conflicts of interest. For information regarding this article, correspondence via E-mail to: wpylg@hotmail.com; wangg@liverpool.ac.uk; toh@liverpool.ac.uk Copyright © by 2019 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Alexandros Sfakianakis
Anapafseos 5 . Agios Nikolaos
Crete.Greece.72100
2841026182

6948891480

alsfakia@gmail.com