Eva K Ritzl1,2,3, Sung-Min Cho4,5, Jaeho Hwang1, Jay Bronder1,2, Nirma Carballido Martinez2, Romergryko Geocadin1,3, Bo Soo Kim6, Errol Bush7, Glenn Whitman8, Chun Woo Choi8. 1. Department of Neurology, Johns Hopkins Hospital, Baltimore, MD, USA. 2. Division of Epilepsy, Department of Neurology, Johns Hopkins Hospital, Baltimore, MD, USA. 3. Division of Neurosciences Critical Care, Department of Anesthesiology and Critical Care Medicine, Johns Hopkins Hospital, 600 N. Wolfe Street, Phipps 455, Baltimore, MD, 21287, USA. 4. Department of Neurology, Johns Hopkins Hospital, Baltimore, MD, USA. csmfisher@gmail.com. 5. Division of Neurosciences Critical Care, Department of Anesthesiology and Critical Care Medicine, Johns Hopkins Hospital, 600 N. Wolfe Street, Phipps 455, Baltimore, MD, 21287, USA. csmfisher@gmail.com. 6. Department of Pulmonary and Critical Care Medicine, Johns Hopkins Hospital, Baltimore, MD, USA. 7. Division of General Thoracic Surgery, Department of Surgery, Johns Hopkins Hospital, Baltimore, MD, USA. 8. Division of Cardiac Surgery, Department of Surgery, Johns Hopkins Hospital, Baltimore, MD, USA.
Abstract
BACKGROUND: We aimed to identify continuous electroencephalogram (cEEG) markers associated with survival and death in patients with extracorporeal membrane oxygenation (ECMO) support under standardized sedation cessation protocol. METHODS: Prospectively collected records of adult patients (age ≥ 18 years) who were started on ECMO support in July 2016 to December 2020 at a single tertiary center were analyzed. cEEGs were performed on patients on the basis of inclusion and exclusion criteria. Patients receiving sedation that affect cEEG reactivity at the start of cEEG recording, including propofol, ketamine, or benzodiazepines, were excluded. We allowed fentanyl and dexmedetomidine during cEEG monitoring. cEEGs were evaluated for frequency, amplitude, variability, reactivity, and state changes. RESULTS: Of 290 patients, 40 underwent cEEG in the absence of confounding sedation (median age 60 years, 85% venoarterial-ECMO, 15% venovenous-ECMO). The median length of ECMO support and analyzable cEEG were 143 h and 24 h, respectively. A total of 27 patients underwent withdrawal of life-sustaining therapies (WOLST) during ECMO support. Of the 13 who weaned off ECMO, 9 underwent WOLST later in the hospitalization and 4 survived at hospital discharge. Decisions of WOLST were not influenced by cEEG features' results. Proportions of present EEG reactivity, present state changes, and fair/good variability were significantly higher in patients who survived compared with those who died (odds ratios infinity, infinity, and 13.57, respectively; p values < 0.001, < 0.001, and 0.0299, respectively). Sensitivity and specificity for survival at discharge were 100% and 91.67% for intact reactivity, 100% and 97.20% for present state changes, and 75% and 83.3% for fair/good variability. CONCLUSIONS: Although future multicenter studies with larger patient cohorts are certainly warranted, we were able to validate the feasibility of protocolized sedation cessation and cEEG analyses in the absence of a confounding effect from sedating medications. Moreover, we demonstrate some evidence that cEEG features of intact reactivity, present state changes, and fair/good variability in comatose patients on ECMO may be associated with survival at hospital discharge.
BACKGROUND: We aimed to identify continuous electroencephalogram (cEEG) markers associated with survival and death in patients with extracorporeal membrane oxygenation (ECMO) support under standardized sedation cessation protocol. METHODS: Prospectively collected records of adult patients (age ≥ 18 years) who were started on ECMO support in July 2016 to December 2020 at a single tertiary center were analyzed. cEEGs were performed on patients on the basis of inclusion and exclusion criteria. Patients receiving sedation that affect cEEG reactivity at the start of cEEG recording, including propofol, ketamine, or benzodiazepines, were excluded. We allowed fentanyl and dexmedetomidine during cEEG monitoring. cEEGs were evaluated for frequency, amplitude, variability, reactivity, and state changes. RESULTS: Of 290 patients, 40 underwent cEEG in the absence of confounding sedation (median age 60 years, 85% venoarterial-ECMO, 15% venovenous-ECMO). The median length of ECMO support and analyzable cEEG were 143 h and 24 h, respectively. A total of 27 patients underwent withdrawal of life-sustaining therapies (WOLST) during ECMO support. Of the 13 who weaned off ECMO, 9 underwent WOLST later in the hospitalization and 4 survived at hospital discharge. Decisions of WOLST were not influenced by cEEG features' results. Proportions of present EEG reactivity, present state changes, and fair/good variability were significantly higher in patients who survived compared with those who died (odds ratios infinity, infinity, and 13.57, respectively; p values < 0.001, < 0.001, and 0.0299, respectively). Sensitivity and specificity for survival at discharge were 100% and 91.67% for intact reactivity, 100% and 97.20% for present state changes, and 75% and 83.3% for fair/good variability. CONCLUSIONS: Although future multicenter studies with larger patient cohorts are certainly warranted, we were able to validate the feasibility of protocolized sedation cessation and cEEG analyses in the absence of a confounding effect from sedating medications. Moreover, we demonstrate some evidence that cEEG features of intact reactivity, present state changes, and fair/good variability in comatose patients on ECMO may be associated with survival at hospital discharge.