Neurocognitive Disorders After Major Surgery in Elderly (POEGEA)

Effect of Electroencephalography-guided Anesthesia on Neurocognitive Disorders in Elderly Patients Undergoing Major Non-cardiac Surgery: a Randomized Clinical Trial

The objective of the study is to investigate, in patients aged 70 years and over undergoing major non-cardiac surgery, the effect of electroencephalic (EEG)-guided anesthesia on postoperative neurocognitive disorders when controlling for intraoperative nociception, personalized blood pressure targets and using full information provided by the processed EEG monitor (including burst suppression ratio, density spectral array and raw EEG waveform). This prospective, randomized, controlled trial will be conducted in a single Canadian university hospital. Patients aged 70 years and over, undergoing elective major non-cardiac surgery will be included. The administration of sevoflurane will be adjusted to maintain a BIS value between 40 and 60, a suppression Ratio at 0%, a direct EEG display without any suppression time and a spectrogram with most of the EEG wave frequency within the alpha, theta and delta frequencies in the EEG-guided group. In the control group sevoflurane will be administered to achieve an age-adjusted minimum alveolar concentration of [0.8-1.2]. A nociception monitor will guide intraoperative opioids' infusion and individual blood pressure targets will be personalized in both groups. The primary endpoint is the incidence of neurocognitive disorder (NCD) at postoperative day 1 evaluated by the Montréal Cognitive Assessment. Secondary endpoints include the incidence of postoperative neurocognitive disorder at different timepoints and the evaluation of cognitive trajectories among EEG-guided and control groups.

The main objective of the study is to investigate the effect of EEG-guided anesthesia aiming at reducing anesthetic administration and minimizing burst suppression on the EEG during general anesthesia on the incidence of NCD at postoperative day 1 in elderly patients (> 70 years old) undergoing major (expected duration > 1h) non-cardiac surgery compared to standard care. Secondary objectives are to investigate the effect of EEG-guided anesthesia compared to standard care on: neurocognitive disorders at postoperative day 2, 7, 15, 30 and 90, perioperative cognitive trajectories, postoperative delirium, intraoperative consumption of volatile anesthetics, opioids and vasopressors, intraoperative hypotension (number of interventions to treat hypotensive events), cerebral hypoxemia (assessed by cerebral oximetry), cumulative burst suppression duration and cumulative low processed EEG values duration during anesthesia, awareness postoperative surgical and quality of life / recovery outcomes. Tertiary objectives are to: explore the effect of EEG-guided anesthesia compared to standard care across and within subgroups including different categories of surgical, duration of surgery, preoperative frailty, preoperative presence of depressive symptoms, age and patients with preoperative neurocognitive disorder. explore the effect of a possible imbalance in risk factors for postoperative NCD between the two groups on the incidence of postoperative NCD. Methods Participants 314 patients 70 years of age or older scheduled for elective major gynecologic, abdominal, urologic, thoracic or orthopedic surgeries via laparoscopy or laparotomy under general anesthesia and an anesthesia time of more than 60 minutes at the institution HMR-CEMTL, who are seen at the preoperative clinic (CIEPC) by internal medicine and/or anesthesiology. Exclusion criteria: patients with known diagnosis of dementia or other neurological, psychiatric, developmental or medical condition that resulted in severe documented cognitive impairment, emergency surgery, significant auditory or visual impairment that precludes participation in cognitive testing, known allergy or intolerance or other medical condition that precludes the use of prescribed general anesthesia protocol for this study, inability to communicate in French or English will not be included in the present study. Anesthesia Protocol All patients will undergo general anesthesia with the following: induction with IV slow boluses of lidocaine to numb the vein, propofol 1.5mg.kg-1, remifentanil 1µg.kg-1, rocuronium 0.8mg.kg-1. The maintenance of anesthesia will be based on sevoflurane to reach [0.8-1.2] minimal alveolar concentration (MAC adjusted to age) in the control group, and to achieve a Bispectral index (BIS) of [40-60] in the EEG-guided group (see below). The Nociception level (NOL) index (PMD200™ device, Medasense Biometrics Ltd, Ramat Gan, Israel) and the bilateral cerebral regional oximetry (rSO2) (Invos™, Medtronic, Canada) will be placed and available for both groups throughout the entire anesthesia. Remifentanil infusion will be set between 0.02 and 0.3 µg.kg-1.min-1 to achieve a NOL index of [5-25]. Phenylephrine will be started at 0.2 µg.kg-1.min-1 and adjusted to maintain +/- 20% of the baseline values of the pre-anesthesia mean arterial pressure (MAP). No benzodiazepines or ketamine will be used intraoperatively. An epidural may be placed at the discretion of the anesthesiologist in charge of the patient in the OR. Patients will receive standard prophylaxis against postoperative nausea and vomiting. Hydromorphone PCA (or hydromorphone SC or PO if patient is not a candidate for PCA) for postoperative pain scores < 4/10, or Patient controlled epidural analgesia if an epidural is started in Postoperative anesthesia care unit (PACU). All anesthesia related side effects and quality of analgesia and rehabilitation will be evaluated for 48h. Intervention Patients will be randomized (1:1) to receive EEG guided anesthesia versus standard of care. Information provided by the BIS (Medtronic, Canada) monitor will guide the volatile anesthetic administration in the EEG-guided group to maintain a BIS value between 40 and 60, a Suppression Ratio (SR; % of time with suppressed brain electrical activity) at 0% or the closest, a direct EEG display without any suppression time and a spectrogram (DSA or density spectral array) with most of the EEG wave frequency within the Alpha (8-12Hz), Theta (4-8Hz) and Delta (0.5-4Hz) frequencies. In the standard of care group, the age-adjusted Minimum Alveolar Concentration (MAC-age) of sevoflurane will be kept at [0.8-1.2] MAC. Cognitive Assessment Participants will undergo cognitive assessment preoperatively to establish a baseline and then postoperatively to assess change from that baseline at postoperative days 1, 2, 7, 15, 30 and 90. The MoCA or the telephone version of the MoCA (T-MoCA) will be administered at each time point depending on patient location (hospital / home). Verbal fluency including phonemic and categorical fluency will be administered at each time point The Confusion Assessment Method (CAM) will be used to diagnose postoperative delirium at postoperative days 1 and 2. The Quality of Recovery-15 (QoR-15) questionnaire will be used at postoperative day 30 and 90 to explore quality of life after anesthesia and surgery. Significance/Importance This study was designed to explore the effect of EEG-guided anesthesia on perioperative neurocognitive disorders with tight control of intraoperative blood pressure and nociception level. Understanding EEG patterns of anesthesia to individualize titration of hypnotic drugs may help in reducing the incidence of perioperative neurocognitive disorders, particularly in the elderly. Study Design Prospective monocentric randomized controlled trial. Subject Population Patients ≥ 70 years old scheduled for major non-cardiac surgery (expected duration ≥1h). Sample Size Three hundred and fourteen (314 total; 157 per group, 2 groups) will be included in the present study. Study Duration 2 years. Study Center Single centre study, at Maisonneuve-Rosemont Hospital (HMR), Centre intégré universitaire de santé et de services sociaux (CIUSSS) de l'Est de l'Ile de Montréal (CEMTL), Montréal, Québec, Canada. Adverse Events Non expected.

Information provided by the BIS (Medtronic, Canada) monitor will guide the volatile anesthetic administration in the EEG-guided group to maintain a BIS value between 40 and 60, a Suppression Ratio (SR; % of time with suppressed brain electrical activity) at 0% or the closest, a direct EEG display without any suppression time and a spectrogram (DSA or density spectral array) with most of the EEG wave frequency within the Alpha (8-12Hz), Theta (4-8Hz) and Delta (0.5-4Hz) frequencies.

In the standard care group, the age-adjusted Minimum Alveolar Concentration (MAC-age) of sevoflurane will be kept at [0.8-1.2] MAC.

Information provided by the BIS (Medtronic, Canada) monitor will guide the volatile anesthetic administration in the EEG-guided group to maintain a BIS value between 40 and 60, a Suppression Ratio (SR; % of time with suppressed brain electrical activity) at 0% or the closest, a direct EEG display without any suppression time and a spectrogram (DSA or density spectral array) with most of the EEG wave frequency within the Alpha (8-12Hz), Theta (4-8Hz) and Delta (0.5-4Hz) frequencies.

In the standard care group, the age-adjusted Minimum Alveolar Concentration (MAC-age) of sevoflurane will be kept at [0.8-1.2] MAC.

The presence of postoperative neurocognitive disorder will be evaluated using the Montreal Cognitive Assessment The Montréal Cognitive Assessment range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 1 (primary endpoint)

The presence of postoperative neurocognitive disorder will be evaluated using the Montreal Cognitive Assessment (MoCA) The Montréal Cognitive Assessment range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 2

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (T-MoCA) The Telephone MoCA range from 0 to 22, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 7

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (T-MoCA) The Telephone MoCA range from 0 to 22, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 15

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (T-MoCA) The Telephone MoCA range from 0 to 22, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 30

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (T-MoCA) The Telephone MoCA range from 0 to 22, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 90

The verbal fluency test consists of asking a patient to report as many word as possible in 1 minutes for a given letter (ex: as many "F" words as possible in 1 minutes) The verbal fluency test has no unit. It starts from 0 and has no maximum value. Higher values corresponding to The preoperative mean and standard deviation of the Verbal fluency will be calculated for the overall population. An individual reduction of the verbal fluency test ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder evaluated on verbal fluency (Yes/No). Baseline cognitive testing will be performed before surgery, then repeated on postoperative day 1,2,7,15,30 and 90

The Telephone version of the Montréal Cognitive Assessment is a short form of the Montreal cognitive assessment The Telephone MoCA ranges from 0 to 22, higher values corresponding to better cognitive performances. The MoCA ranges from 0 to 30, higher values corresponding to better cognitive performances. The T-MoCA is included in the MoCA T-MoCA part of the MoCA test will be used at day 0 and postoperative days 1 and 2 T-MoCA will be used at postoperative Day 7,15, 30 and 90 The evolution of these scores will establish cognitive trajectories.

Baseline mean arterial blood pressure (MAP) will be defined as the average of 3 consecutive values taken 1 min apart and determined before the induction of general anesthesia. personalized objectives of intraoperative MAP will be defined as Baseline MAP +/- 20%. Total hypotension time will be defined as the cumulative time below Baseline MAP - 20% from induction of anesthesia until tracheal extubation.

Baseline cerebral oxygen saturation (rSO2) will be defined as the average of 3 consecutive values taken 1 min apart and determined before the induction of general anesthesia. a decrease of 20% of the baseline rSO2 will define cerebral hypoxemia Total cerebral hypoxemia time will be defined as the cumulative time in cerebral hypoxemia from induction of anesthesia until tracheal extubation

Processed BIS EEG monitor provides a suppression time value corresponding to the cumulative time spent in burst suppression. Total burst suppression time will be measured from induction of anesthesia until tracheal extubation

Low BIS values will be defined as a BIS value under 40 (no unit for the BIS value) Total low BIS values duration will be defined as the cumulative time under 40 from anesthesia induction until tracheal extubation

Postoperative neurocognitive disorder day 1 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Montreal Cognitive Assessment (MoCA) The MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Postoperative neurocognitive disorder day 2 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Montreal Cognitive Assessment (MoCA) The MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the Montréal Cognitive assessment will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Postoperative neurocognitive disorder day 7 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (MoCA) The T-MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the T-MoCA will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Postoperative neurocognitive disorder day 15 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (MoCA) The T-MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the T-MoCA will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Postoperative neurocognitive disorder day 30 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (MoCA) The T-MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the T-MoCA will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Postoperative neurocognitive disorder day 90 (Yes/No); dichotomic result; among prespecified subgroups of patients

The presence of postoperative neurocognitive disorder will be evaluated using the Telephone Montreal Cognitive Assessment (MoCA) The T-MoCA range from 0 to 30, higher values corresponding to better cognitive performances. The preoperative mean and standard deviation of the T-MoCA will be calculated. An individual reduction of the MoCA score ≥ 1.96 standard deviation calculated on overall population preoperative mean will define postoperative neurocognitive disorder (Yes/No). Prespecified subgroups: different surgical types (orthopedic, abdominal, gynecologic, urologic, thoracic), different surgical duration ([1-2 hours], [2-3h], > 3h), patients with preoperative neurocognitive disorder (defined as a MoCA score at baseline < 26), frail patients (defined as a Clinical Frailty Scale ≥ 5 evaluated at baseline) depressive patients (defined as a Patient Health Questionnaire 9 ≥ 10 evaluated at baseline) patients ≥ 80 years old

Risk difference between groups of Postoperative neurocognitive disorder adjusted for prespecified risk factors (no unit)

In a case the investigators observe important quantitative imbalance between prognostic factors between the treatment groups, the investigators will conduct an adjusted analysis of the primary outcome. The investigators will estimate a standardized marginal average treatment effect (risk difference) model adjusted from 10 strata based on an estimated propensity score from neurocognitive disorder's risk factors (age, sex, preoperative MoCA score, Clinical frailty scale (CFS), Patient Health Questionnaire (PHQ) -9 and education level). The marginal risk difference will be reported with 95% confidence intervals, estimated by non-parametric bootstrap (10 000 replications).

Inclusion Criteria: Patients 70 years of age or older, Major gynecologic, abdominal, urologic, thoracic or orthopedic surgery via laparoscopy or laparotomy under general anesthesia - with or without concomitant use of regional or neuraxial anesthesia -, Expected anesthesia time of more than 60 minutes, Seen for assessment by internal medicine and/or anesthesiology at the preoperative clinic (CIEPC) Exclusion Criteria: Known diagnosis of dementia or other neurological, psychiatric, developmental or medical condition that resulted in documented severe cognitive impairment, Emergency surgery, Significant auditory or visual impairment that precludes participation in cognitive testing, Known allergy or intolerance or other medical condition that precludes the use of prescribed general anesthesia protocol for this study, Inability to communicate in French or English.

Morisson L, Laferriere-Langlois P, Carrier FM, Page G, Godbout C, Fortier LP, Ogez D, Letourneau G, Jarry S, Denault A, Fortier A, Guertin MC, Verdonck O, Richebe P. Effect of electroencephalography-guided anesthesia on neurocognitive disorders in elderly patients undergoing major non-cardiac surgery: A trial protocol The POEGEA trial (POncd Elderly GEneral Anesthesia). PLoS One. 2021 Aug 10;16(8):e0255852. doi: 10.1371/journal.pone.0255852. eCollection 2021.