Miles Berger1, Sarada S Eleswarpu1, Mary Cooter Wright1, Anna M Ray2, Sarah A Wingfield3, Mitchell T Heflin4, Shahrukh Bengali5, Ankeet D Udani1. 1. From the Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina. 2. Department of Anesthesiology, Brigham and Women's Hospital, Boston, Massachusetts. 3. Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas. 4. Department of Medicine, Geriatrics Division, Duke University Medical Center, Durham, North Carolina. 5. Department of Anesthesiology, University of Texas Southwest Medical Center, Dallas, Texas.
Abstract
BACKGROUND: Different anesthetic drugs and patient factors yield unique electroencephalogram (EEG) patterns. Yet, it is unclear how best to teach trainees to interpret EEG time series data and the corresponding spectral information for intraoperative anesthetic titration, or what effect this might have on outcomes. METHODS: We developed an electronic learning curriculum (ELC) that covered EEG spectrogram interpretation and its use in anesthetic titration. Anesthesiology residents at a single academic center were randomized to receive this ELC and given spectrogram monitors for intraoperative use versus standard residency curriculum alone without intraoperative spectrogram monitors. We hypothesized that this intervention would result in lower inhaled anesthetic administration (measured by age-adjusted total minimal alveolar concentration [MAC] fraction and age-adjusted minimal alveolar concentration [aaMAC]) to patients ≥60 old during the postintervention period (the primary study outcome). To study this effect and to determine whether the 2 groups were administering similar anesthetic doses pre- versus postintervention, we compared aaMAC between control versus intervention group residents both before and after the intervention. To measure efficacy in the postintervention period, we included only those cases in the intervention group when the monitor was actually used. Multivariable linear mixed-effects modeling was performed for aaMAC fraction and hospital length of stay (LOS; a non-prespecified secondary outcome), with a random effect for individual resident. A multivariable linear mixed-effects model was also used in a sensitivity analysis to determine if there was a group (intervention versus control group) by time period (post- versus preintervention) interaction for aaMAC. Resident EEG knowledge difference (a prespecified secondary outcome) was compared with a 2-sided 2-group paired t test. RESULTS: Postintervention, there was no significant aaMAC difference in patients cared for by the ELC group (n = 159 patients) versus control group (N = 325 patients; aaMAC difference = -0.03; 95% confidence interval [CI], -0.09 to 0.03; P =.32). In a multivariable mixed model, the interaction of time period (post- versus preintervention) and group (intervention versus control) led to a nonsignificant reduction of -0.05 aaMAC (95% CI, -0.11 to 0.01; P = .102). ELC group residents (N = 19) showed a greater increase in EEG knowledge test scores than control residents (N = 20) from before to after the ELC intervention (6-point increase; 95% CI, 3.50-8.88; P < .001). Patients cared for by the ELC group versus control group had a reduced hospital LOS (median, 2.48 vs 3.86 days, respectively; P = .024). CONCLUSIONS: Although there was no effect on mean aaMAC, these results demonstrate that this EEG-ELC intervention increased resident knowledge and raise the possibility that it may reduce hospital LOS.
BACKGROUND: Different anesthetic drugs and patient factors yield unique electroencephalogram (EEG) patterns. Yet, it is unclear how best to teach trainees to interpret EEG time series data and the corresponding spectral information for intraoperative anesthetic titration, or what effect this might have on outcomes. METHODS: We developed an electronic learning curriculum (ELC) that covered EEG spectrogram interpretation and its use in anesthetic titration. Anesthesiology residents at a single academic center were randomized to receive this ELC and given spectrogram monitors for intraoperative use versus standard residency curriculum alone without intraoperative spectrogram monitors. We hypothesized that this intervention would result in lower inhaled anesthetic administration (measured by age-adjusted total minimal alveolar concentration [MAC] fraction and age-adjusted minimal alveolar concentration [aaMAC]) to patients ≥60 old during the postintervention period (the primary study outcome). To study this effect and to determine whether the 2 groups were administering similar anesthetic doses pre- versus postintervention, we compared aaMAC between control versus intervention group residents both before and after the intervention. To measure efficacy in the postintervention period, we included only those cases in the intervention group when the monitor was actually used. Multivariable linear mixed-effects modeling was performed for aaMAC fraction and hospital length of stay (LOS; a non-prespecified secondary outcome), with a random effect for individual resident. A multivariable linear mixed-effects model was also used in a sensitivity analysis to determine if there was a group (intervention versus control group) by time period (post- versus preintervention) interaction for aaMAC. Resident EEG knowledge difference (a prespecified secondary outcome) was compared with a 2-sided 2-group paired t test. RESULTS: Postintervention, there was no significant aaMAC difference in patients cared for by the ELC group (n = 159 patients) versus control group (N = 325 patients; aaMAC difference = -0.03; 95% confidence interval [CI], -0.09 to 0.03; P =.32). In a multivariable mixed model, the interaction of time period (post- versus preintervention) and group (intervention versus control) led to a nonsignificant reduction of -0.05 aaMAC (95% CI, -0.11 to 0.01; P = .102). ELC group residents (N = 19) showed a greater increase in EEG knowledge test scores than control residents (N = 20) from before to after the ELC intervention (6-point increase; 95% CI, 3.50-8.88; P < .001). Patients cared for by the ELC group versus control group had a reduced hospital LOS (median, 2.48 vs 3.86 days, respectively; P = .024). CONCLUSIONS: Although there was no effect on mean aaMAC, these results demonstrate that this EEG-ELC intervention increased resident knowledge and raise the possibility that it may reduce hospital LOS.
Authors: Oluwaseun Akeju; Andrew H Song; Allison E Hamilos; Kara J Pavone; Francisco J Flores; Emery N Brown; Patrick L Purdon Journal: Clin Neurophysiol Date: 2016-03-16 Impact factor: 3.708
Authors: Elizabeth L Whitlock; Alexander J Villafranca; Nan Lin; Ben J Palanca; Eric Jacobsohn; Kevin J Finkel; Lini Zhang; Beth A Burnside; Heiko A Kaiser; Alex S Evers; Michael S Avidan Journal: Anesthesiology Date: 2011-12 Impact factor: 7.892
Authors: Mary Cooter; Katherine Ni; Jake Thomas; Dhanesh K Gupta; Thomas J Hopkins; Timothy E Miller; Michael L James; Miklos D Kertai; Miles Berger Journal: Br J Anaesth Date: 2019-11-01 Impact factor: 9.166
Authors: Sandeep C Manyam; Dhanesh K Gupta; Ken B Johnson; Julia L White; Nathan L Pace; Dwayne R Westenskow; Talmage D Egan Journal: Anesthesiology Date: 2007-03 Impact factor: 7.892
Authors: Mary Cooter Wright; Thomas Bunning; Sarada S Eleswarpu; Mitchell T Heflin; Shelley R McDonald; Sandhya Lagoo-Deenadalayan; Heather E Whitson; Pablo Martinez-Camblor; Stacie G Deiner; Miles Berger Journal: Anesth Analg Date: 2022-01-01 Impact factor: 5.108
Authors: Yu Raymond Shao; Pegah Kahali; Timothy T Houle; Hao Deng; Christopher Colvin; Bradford C Dickerson; Emery N Brown; Patrick L Purdon Journal: Anesth Analg Date: 2020-11 Impact factor: 6.627