Ischemic Stroke in the Cardiac Surgery Intensive Care Unit: A Quality Improvement Study.

OBJECTIVE: To investigate the frequency of stroke and code stroke activation and the factors influencing code stroke management in postoperative cardiac surgical patients. DESIGN: A retrospective quality improvement study was conducted between January 1, 2016, and December 31, 2021. SETTING: The Cardiac Surgery Recovery Unit (CSRU) at London Health Sciences Centre in London, Ontario, Canada. PARTICIPANTS: Postcardiac surgery patients aged 18 years or older who developed ischemic stroke during their admission to the CSRU. INTERVENTIONS: No specific interventions were administered as part of this study. Code stroke activation mobilizes a specialized team. The objectives include assessment by a physician within 10 minutes, obtaining neuroimaging and interpretation within 45 minutes, and beginning treatment within 60 minutes. MEASUREMENTS AND MAIN RESULTS: The incidence rate of stroke in the CSRU was 1.3%, and 34% of these patients had code stroke activated. The time since the last known well status was 11 ± 8 hours. The most common reasons for not activating code stroke were not meeting both timing and clinical criteria. The average time for computed tomography (CT) scan was 36 ± 22 minutes. Among patients who had code stroke activated, 24% had large- vessel occlusion (LVO), and 67% of those with LVO had an established stroke on their initial CT. CONCLUSION: Code stroke was activated in only one-third of patients who experienced a stroke following cardiac surgery. Additionally, out of those who had code stroke activated, only one-fourth were diagnosed with LVO. Among those with LVO, two-thirds were found to have a well-established stroke on noncontrast CT scans and were deemed ineligible for intervention.

Point-of-care lung and cardiac ultrasound (LUCAS) study in hip fracture patients: a prospective cohort study.

PURPOSE: A clinical conflict often presented with hip fracture patients is whether to proceed with timely surgery or delay surgery until a formal echocardiogram is conducted. This study aimed to assess the impact of incorporating point-of-care lung and cardiac ultrasound (LUCAS) scans as part of the preoperative assessment for hip fracture patients. METHODS: We recruited 225 consecutive adult patients booked for urgent hip arthroplasty surgery. A LUCAS scan was performed for each patient. The anesthesiologists were asked to provide their anesthetic plans before and after acknowledging the results of the LUCAS scans. The primary endpoint was a composite outcome of changes to the anesthetic plan. The secondary outcomes included anesthesiologists' opinions of the LUCAS scans. RESULTS: One-hundred-ninety-eight patients were included. The majority of LUCAS findings were not severe. A common abnormal finding was hypovolemia (31%). One-hundred-and-six anesthetic management decisions were changed, with 59 of these changes being an escalation of the anesthetic plan, and 47 of these changes being a de-escalation. Eighty-three percent of anesthesiologists agreed that LUCAS affirmed their anesthetic plans and should be an integral part of the perioperative assessment. CONCLUSION: This study found that LUCAS scans did not significantly alter the anesthetic plan for hip fracture patients. Nevertheless, LUCAS scans can rule out severe cardiopulmonary conditions and allow for both escalation and de-escalation of care. In the setting of early hip surgery, LUCAS presents a viable option in selected patients to address the unmet need to allow for both timely surgery and comprehensive patient evaluation. STUDY REGISTRATION: ClinicalTrials.gov (NCT03275129); registered 8 July 2018.

RéSUMé: OBJECTIF: Une interrogation clinique qui existe souvent avec les patient·es se présentant avec une fracture de hanche est de savoir s'il faut procéder à une intervention chirurgicale rapidement ou retarder la chirurgie jusqu'à ce qu'un échocardiogramme formel soit réalisé. Cette étude visait à évaluer l'impact de l'intégration de l'échographie ciblée pulmonaire et cardiaque (LUCAS, Lung and Cardiac ultrasound) dans le cadre de l'évaluation préopératoire des personnes ayant subi une fracture de la hanche. MéTHODE: Nous avons recruté 225 patient·es adultes consécutif·ves devant bénéficier d'une arthroplastie urgente de la hanche. Une échographie de type LUCAS a été réalisée pour chaque patient·e. On a demandé aux anesthésiologistes de fournir leurs plans anesthésiques avant et après avoir pris connaissance des résultats des échographies de type LUCAS. Le critère d'évaluation principal était un résultat composite des modifications apportées au plan anesthésique. Les critères d'évaluation secondaires comprenaient les opinions des anesthésiologistes sur les échographies de type LUCAS. RéSULTATS: Cent quatre-vingt-dix-huit patient·es ont été inclus·es. La majorité des résultats de l'échographie de type LUCAS n'étaient pas graves. Un résultat anormal courant était l'hypovolémie (31 %). Cent six décisions de prise en charge anesthésique ont été modifiées, 59 de ces changements étant une escalade du plan anesthésique et 47 de ces changements étant une réduction. Quatre-vingt-trois pour cent des anesthésiologistes ont convenu que l'échographie de type LUCAS confirmait leurs plans anesthésiques et devrait faire partie intégrante de l'évaluation périopératoire. CONCLUSION: Cette étude a révélé que les échographies de type LUCAS ne modifiaient pas de manière significative le plan anesthésique pour les patient·es se présentant avec une fracture de hanche. Néanmoins, les échographies de type LUCAS peuvent exclure des affections cardiopulmonaires graves et permettre à la fois d'augmenter ou de réduire les soins périopératoires. Dans le cadre d'une chirurgie précoce de la hanche, l'échographie de type LUCAS présente une option viable chez une patientèle sélectionnée pour répondre à un besoin non satisfait afin de permettre à la fois une chirurgie rapide et une évaluation complète des patient·es. ENREGISTREMENT DE L'éTUDE: www.clinicaltrials.gov (NCT03275129); enregistrée le 8 juillet 2018.

Canadian recommendations for training and performance in basic perioperative point-of-care ultrasound: recommendations from a consensus of Canadian anesthesiology academic centres.

Point-of-care ultrasound (POCUS) uses ultrasound at the bedside to aid decision-making in acute clinical scenarios. The increased use of ultrasound for regional anesthesia and vascular cannulation, together with more anesthesiologists trained in transesophageal echocardiography have contributed to the widespread use of POCUS in perioperative care. Despite the support of international experts, the practice of POCUS in perioperative care is variable as Canadian guidelines for anesthesiologists do not currently exist. Using a Delphi process of online surveys and a face-to-face national Canadian meeting, we developed a consensus statement for basic POCUS (bPOCUS) performance and training with a group of national experts from all Canadian universities. The group of experts consisted of 55 anesthesiologists from 12 Canadian universities considered local leaders in the field. An initial exploratory online survey of 47 statements was conducted. These statements were derived from previous generic guidelines or consensus conferences, or were based on current literature. Fourteen statements reached full consensus, 19 had 90-100% agreement, and 14 had less than 90% agreement. Eight new statements were proposed during the national meeting, and all statements without full agreement were discussed. A second online survey included 42 modified or new statements. From this second survey, 16 statements obtained full consensus, 39 had very good agreement, and one had good agreement. The final document includes 56 statements that define the scope of practice and necessary training for perioperative bPOCUS. The statements include five bPOCUS domains: cardiac, lung, airway, gastric, and abdomen. The use of bPOCUS is evolving and will play a significant role in perioperative medicine. This consensus statement aims to define a Canadian national standard on which curricula may be based. It also provides a framework to allow further development of bPOCUS in the perioperative setting.

RéSUMé: L'échographie ciblée (POCUS) utilise l'échographie au chevet des patients pour faciliter la prise de décisions dans les situations cliniques urgentes. L'utilisation accrue de l'échographie pour l'anesthésie régionale et la cannulation vasculaire, ainsi que l'augmentation du nombre d'anesthésologistes formés à l'échocardiographie transesophagienne, ont contribué à l'utilisation généralisée de l'échographie ciblée dans les soins périopératoires. Malgré son endossement par des experts internationaux, la pratique de l'échographie ciblée en soins périopératoires est variable, car il n'existe pas, à l'heure actuelle, de lignes directrices canadiennes destinées aux anesthésiologistes. À l'aide d'un processus Delphi de sondages en ligne et d'une réunion nationale canadienne en personne, un groupe d'experts nationaux provenant de toutes les universités canadiennes a élaboré une déclaration consensuelle pour la formation de base en et l'exécution de l'échographie ciblée (bPOCUS). Le groupe d'experts était composé de 55 anesthésiologistes issus de 12 universités canadiennes considérés comme des chefs de file locaux dans le domaine. Un premier sondage exploratoire en ligne comportant 47 énoncés a été réalisé. Ces énoncés étaient dérivés de lignes directrices antérieures ou de conférences consensuelles, ou étaient fondés sur la littérature actuelle. Quatorze énoncés ont obtenu un consensus complet, 19 ont atteint un taux de 90 à 100 %, et 14 ont obtenu moins de 90 % d'accord. Huit nouveaux énoncés ont été proposés au cours de la réunion nationale, et tous les énoncés n'ayant pas obtenu d'accord complet ont été discutés. Un deuxième sondage en ligne comprenait 42 énoncés modifiés ou nouveaux. Dans ce deuxième sondage, 16 énoncés ont obtenu un consensus total, 39 un très bon accord et un énoncé un bon accord. Le document final comporte 56 énoncés qui définissent le champ de pratique et la formation nécessaire pour l'échographie ciblée périopératoire de base. Les énoncés portent sur cinq domaines de l'échographie ciblée de base : échographie cardiaque, pulmonaire, des voies respiratoires, gastrique et abdominale. L'utilisation de l'échographie ciblée de base évolue et jouera un rôle important en médecine périopératoire. Cette déclaration consensuelle vise à définir une norme nationale canadienne sur laquelle les programmes d'études pourront s'appuyer. Elle fournit également un cadre pour encourager le développement ultérieur de l'échographie ciblée de base dans un contexte périopératoire.

Dynamic tracking of microvascular hemoglobin content for continuous perfusion monitoring in the intensive care unit: pilot feasibility study.

PURPOSE: There is a need for bedside methods to monitor oxygen delivery in the microcirculation. Near-infrared spectroscopy commonly measures tissue oxygen saturation, but does not reflect the time-dependent variability of microvascular hemoglobin content (MHC) that attempts to match oxygen supply with demand. The objective of this study is to determine the feasibility of MHC monitoring in critically ill patients using high-resolution near-infrared spectroscopy to assess perfusion in the peripheral microcirculation. METHODS: Prospective observational cohort of 36 patients admitted within 48 h at a tertiary intensive care unit. Perfusion was measured on the quadriceps, biceps, and/or deltoid, using the temporal change in optical density at the isosbestic wavelength of hemoglobin (798 nm). Continuous wavelet transform was applied to the hemoglobin signal to delineate frequency ranges corresponding to physiological oscillations in the cardiovascular system. RESULTS: 31/36 patients had adequate signal quality for analysis, most commonly affected by motion artifacts. MHC signal demonstrates inter-subject heterogeneity in the cohort, indicated by different patterns of variability and frequency composition. Signal characteristics were concordant between muscle groups in the same patient, and correlated with systemic hemoglobin levels and oxygen saturation. Signal power was lower for patients receiving vasopressors, but not correlated with mean arterial pressure. Mechanical ventilation directly impacts MHC in peripheral tissue. CONCLUSION: MHC can be measured continuously in the ICU with high-resolution near-infrared spectroscopy, and reflects the dynamic variability of hemoglobin distribution in the microcirculation. Results suggest this novel hemodynamic metric should be further evaluated for diagnosing microvascular dysfunction and monitoring peripheral perfusion.

The Effects of Positioning on the Volume/Location of the Internal Jugular Vein Using 2-Dimensional Tracked Ultrasound.

OBJECTIVE: To investigate the effects of different positioning on the volume/location of the internal jugular vein (IJV) using 2-dimensional (2D) tracked ultrasound. DESIGN: This was a prospective, observational study. SETTING: Local research institute. PARTICIPANTS: Healthy volunteers. INTERVENTIONS: Twenty healthy volunteers were scanned in the following 6 positions: (1) supine with head neutral, rotated 15 and 30 degrees to the left and (2) 5-, 10-, and 15-degree Trendelenburg position with head neutral. In each position the volunteer's neck was scanned using a 2D ultrasound probe tracked with a magnetic tracker. These spatially tracked 2D images were collected and reconstructed into a 3D volume of the IJV and carotid artery. This 3D ultrasound volume then was segmented to obtain a 3D surface on which measurements and calculations were performed. MEASUREMENTS AND MAIN RESULTS: The measurements included average cross-section area (CSA), CSA along the length of IJV, and average overlap rate. CSA (mm2) in the supine and 5-, 10-, and 15-degree Trendelenburg positions were as follows: 86.7 ± 44.8, 104.3 ± 54.5, 119.1 ± 58.6, and 133.7 ± 53.3 (p < 0.0001). CSA enlarged with the increase of Trendelenburg degree. Neither Trendelenburg position nor head rotation showed a correlation with overlap rate. CONCLUSIONS: Trendelenburg position significantly increased the CSA of the IJV, thus facilitating IJV cannulation. This new 3D reconstruction method permits the creation of a 3D volume through a tracked 2D ultrasound scanning system with image acquisition and integration and may prove useful in providing the user with a "road map" of the vascular anatomy of a patient's neck or other anatomic structures.

A review of diagnostic accuracy and clinical impact from the focused use of perioperative ultrasound.

PURPOSE: To perform a narrative review of the current trials examining the use of perioperative ultrasound to diagnose common issues related to the heart, lungs, stomach, and airway. METHOD: A review of the current literature was conducted in June 2017 on all trials involving ultrasound, including both surface and transesophageal ultrasound, in the perioperative period. The search included the terms 'ultrasonography', 'perioperative care', 'point-of-care', and 'bedside'. Trials were limited to human subjects with no language or time restrictions being applied. The results were then collected and a narrative review was completed with the available information. RESULTS: In total 1,176 reports of original investigation or systematic reviews were collected and reviewed. Of those 1,176 reports and reviews, a total of 80 original articles met the inclusion criteria for this review. Topics were broadly defined based on common themes emerging from the literature including cardiac disease, lung pathology (pneumothorax, pleural effusion, pulmonary edema, and pulmonary consolidation), volume and contents of the stomach, confirmation of endotracheal tube position, confirmation of lung isolation, and the application of ultrasound for guiding cricothyroidotomy. Where possible, the sensitivity and specificity of the trials are presented. Few trials reported on patient outcomes, although several discussed provider outcomes such as a change in anesthesia practice. In addition, trials reporting outcomes, although few in number, were included. CONCLUSION: Perioperative point-of-care ultrasound is a useful method for the diagnosis of many important perioperative conditions. The impact of this diagnostic approach on patient outcomes however remains to be determined.

The use of point-of-care ultrasound to diagnose patent foramen ovale in elective hip and knee arthroplasty patients and its association with postoperative delirium.

PURPOSE: Hip and knee arthroplasty surgeries are associated with embolism of materials such as air, cement, and fat. Patent foramen ovale (PFO) is a common congenital cardiac condition that has been reported to lead to paradoxical embolism. This observational study aimed to investigate if the presence of a PFO was associated with an increased risk of postoperative delirium in patients undergoing primary elective hip or knee arthroplasties. METHOD: This was a prospective cohort study at a tertiary teaching hospital. We enrolled patients undergoing primary elective hip or knee arthroplasty who did not have any risk factors for embolism or delirium. Bedside transthoracic echocardiography (TTE) with a bubble study was performed on all patients to detect the presence of PFO. The primary outcome was postoperative delirium as assessed by the standardized Confusion Assessment Method. Secondary outcomes included the ease of performing a TTE bubble study in the perioperative setting, the quality of the TTE images, length of stay, major cardiovascular and neurologic complications, and effects of anesthetic or analgesic management techniques on delirium. RESULTS: Two hundred two patients completed the study. The median [interquartile range] duration of stay was 2 [2-3] days. Only 16 patients (8%) had a positive bubble study. Postoperative delirium was observed in only one patient. Major adverse events were not seen. The inter-rater reliability for the TTE image quality scores was fair (kappa statistic = 0.22). CONCLUSION: Given the very low incidence of PFO and postoperative delirium in this study, we could not form any conclusions regarding the impact of a PFO on important outcomes including delirium or other major adverse events. No recommendation can be made regarding screening for PFO in patients scheduled for lower extremity arthroplasty surgery. TRIAL REGISTRATION: ClinicalTrials.gov (NCT02400892). Registered 27 March 2015.

Modeling Patient-Specific Deformable Mitral Valves.

Medical imaging has advanced enormously over the last few decades, revolutionizing patient diagnostics and care. At the same time, additive manufacturing has emerged as a means of reproducing physical shapes and models previously not possible. In combination, they have given rise to 3-dimensional (3D) modeling, an entirely new technology for physicians. In an era in which 3D imaging has become a standard for aiding in the diagnosis and treatment of cardiac disease, this visualization now can be taken further by bringing the patient's anatomy into physical reality as a model. The authors describe the generalized process of creating a model of cardiac anatomy from patient images and their experience creating patient-specific dynamic mitral valve models. This involves a combination of image processing software and 3D printing technology. In this article, the complexity of 3D modeling is described and the decision-making process for cardiac anesthesiologists is summarized. The management of cardiac disease has been altered with the emergence of 3D echocardiography, and 3D modeling represents the next paradigm shift.

Transfusion Requirements in Cardiac Surgery III (TRICS III): Study Design of a Randomized Controlled Trial.

OBJECTIVES: To determine if a restrictive transfusion threshold is noninferior to a higher threshold as measured by a composite outcome of mortality and serious morbidity. DESIGN: Transfusion Requirements in Cardiac Surgery (TRICS) III was a multicenter, international, open-label randomized controlled trial of two commonly used transfusion strategies in patients having cardiac surgery using a noninferiority trial design (ClinicalTrials.gov number, NCT02042898). SETTING: Eligible patients were randomized prior to surgery in a 1:1 ratio. PARTICIPANTS: Potential participants were 18 years or older undergoing planned cardiac surgery using cardiopulmonary bypass (CPB) with a preoperative European System for Cardiac Operative Risk Evaluation (EuroSCORE I) of 6 or more. INTERVENTIONS: Five thousand patients; those allocated to a restrictive transfusion group received a red blood cell (RBC) transfusion if the hemoglobin concentration (Hb) was less than 7.5 g/dL intraoperatively and/or postoperatively. Patients allocated to a liberal transfusion strategy received RBC transfusion if the Hb was less than 9.5 g/dL intraoperatively or postoperatively in the intensive care unit or less than 8.5 g/dL on the ward. MEASUREMENTS AND MAIN RESULTS: The primary outcome was a composite of all-cause mortality, myocardial infarction, stroke, or new onset renal dysfunction requiring dialysis at hospital discharge or day 28, whichever comes first. The primary outcome was analyzed as a per-protocol analysis. The trial monitored adherence closely as adherence to the transfusion triggers is important in ensuring that measured outcomes reflect the transfusion strategy. CONCLUSION: By randomizing prior to surgery, the TRICS III trial captured the most acute reduction in hemoglobin during cardiopulmonary bypass.

A systematic review of transthoracic and transesophageal echocardiography in non-cardiac surgery: implications for point-of-care ultrasound education in the operating room.

PURPOSE: Point-of-care ultrasound (POCU) is an evolving field in anesthesia. Therefore a systematic review of common diagnoses made by POCU during non-cardiac surgery was conducted. The information obtained from the review may be used to develop POCU curricula for the perioperative setting during non-cardiac surgery. SOURCE: A systematic review was conducted for perioperative use of transthoracic /transesophageal echocardiography (TTE/TEE) in high-risk patients or in other patients experiencing periods of hemodynamic instability. The diagnoses included segmental wall motion abnormalities (SWMAs), low left ventricular ejection fraction (LVEF), hypovolemia, air embolism, cardiac/aortic thrombus, pulmonary embolus (PE), aortic valve disease, mitral valve disease, tricuspid valve disease, right ventricular (RV) failure, pericardial disease, and patent foramen ovale. PRINCIPAL FINDINGS: Three hundred twenty-one studies were found using our search terms, and thirteen studies were retained that met our inclusion criteria for review. The studies included 968 patients analyzed as either preoperative exams in high-risk patients (n = 568) or intraoperative exams during times of hemodynamic compromise/cardiac arrest (n = 400). The most common diagnoses in the preoperative exam group were low ejection fraction (25.4%), aortic valve disease (24.4%), mitral valve disease (20.0%), RV failure (6.6%), and hypovolemia (6.3%). In the intraoperative exam group, the most common diagnoses were hypovolemia (33.2%), low ejection fraction (20.5%), RV failure (13.1%), SWMAs (10.1%), and PE (5.8%). CONCLUSION: In this systematic review examining the use of TTE or TEE in non-cardiac surgery, the most frequent diagnoses were valvulopathy, low LVEF, hypovolemia, PE, SWMAs, and RV failure. This information should be used to inform evidence-based curricula for POCU in anesthesiology.

Comparison of isoflurane and sevoflurane in cardiac surgery: a randomized non-inferiority comparative effectiveness trial.

PURPOSE: Volatile anesthetics possess cardioprotective properties, but it is unknown if the cardioprotective effects extend equally to all members of the class. Although sevoflurane is a relatively newer anesthetic than isoflurane, its introduction into practice was not preceded by a head-to-head comparison with isoflurane in a trial focusing on clinically important outcomes. Our objective was to determine whether sevoflurane was non-inferior to isoflurane on a clinically important primary outcome in a heterogeneous group of adults undergoing cardiac surgery. METHODS: This was a pragmatic randomized non-inferiority comparative effectiveness clinical trial in 464 adults having coronary artery bypass graft and/or single valve surgery during November 2011 to March 2014. The intervention was maintenance of anesthesia with sevoflurane (n = 231) or isoflurane (n = 233) administered at a dose of 0.5-2.0 MAC throughout the entire operation. All caregivers were blinded except for the anesthesiologist and perfusionist. The primary outcome was a composite of intensive care unit (ICU) length of stay ≥ 48 hr and all-cause 30-day mortality. We hypothesized that sevoflurane would be non-inferior to isoflurane (non-inferiority margin < 10% based on an expected event rate of 25%). Secondary outcomes included prolonged ICU stay, 30- and 365-day all-cause mortality, inotrope or vasopressor usage, new-onset hemodialysis or atrial fibrillation, stroke, and readmission to the ICU. RESULTS: No losses to follow-up occurred. The primary outcome occurred in 25% of sevoflurane patients and 30% of isoflurane patients (absolute difference, -5.4%; one-sided 95% confidence interval, 1.4), thus non-inferiority was declared. Sevoflurane was not superior to isoflurane for the primary outcome (P = 0.21) or for any secondary outcomes. CONCLUSION: Sevoflurane is non-inferior to isoflurane on a composite outcome of prolonged ICU stay and all-cause 30-day mortality. Sevoflurane is not superior to isoflurane on any other of the clinically important outcomes. This trial was registered at clinicaltrials.gov; NCT01477151.

RéSUMé: OBJECTIF: Les agents anesthésiques volatils possèdent des propriétés cardioprotectrices, mais nous ne savons pas si ces effets cardioprotecteurs sont équivalents pour tous les agents de cette classe. Bien que le sévoflurane soit un anesthésique plus récent que l'isoflurane, son introduction dans notre pratique n'a pas été précédée par une comparaison directe à l'isoflurane dans une étude s'intéressant à d'importants critères d'évaluation cliniques. Notre objectif était de déterminer si le sévoflurane était non inférieur à l'isoflurane en relation à un critère d'évaluation principal important d'un point de vue clinique dans un groupe hétérogène d'adultes subissant une chirurgie cardiaque. MéTHODE: Nous avons réalisé une étude clinique randomisée et pragmatique d'efficacité comparative et de non-infériorité auprès de 464 adultes subissant des pontages coronariens et/ou une chirurgie valvulaire unique entre novembre 2011 et mars 2014. L'intervention consistait en le maintien de l'anesthésie à l'aide de sévoflurane (n = 231) ou d'isoflurane (n = 233) administré à une dose de 0,5-2,0 MAC tout au long de l'opération. Aucun intervenant ne connaissait l'agent utilisé, à l'exception de l'anesthésiologiste et du perfusionniste. Le critère d'évaluation principal était une composée de la durée de séjour à l'unité de soins intensifs (USI) ≥ 48 h et de la mortalité, toutes causes confondues, à 30 jours. Nous avons émis l'hypothèse que le sévoflurane ne serait pas inférieur à l'isoflurane (marge de non-infériorité < 10 % sur la base d'un taux de complications attendu de 25 %). Les critères d'évaluation secondaires comprenaient un séjour prolongé à l'USI, la mortalité toutes causes confondues à 30 et à 365 jours, l'utilisation d'inotropes ou de vasopresseurs, une hémodialyse ou une fibrillation auriculaire nouvelles, un accident vasculaire cérébral et une réadmission à l'USI. RéSULTATS: Nous n'avons perdu aucun patient au suivi. Le critère d'évaluation principal est survenu chez 25 % des patients ayant reçu du sévoflurane et 30 % des patients ayant reçu de l'isoflurane (différence absolue, −5,4 %; intervalle de confiance unilatéral 95 %, 1,4): la non-infériorité a donc été déclarée. Le sévoflurane n'était pas supérieur à l'isoflurane en ce qui touchait au critère d'évaluation principal (P = 0,21) ou aux critères d'évaluation secondaires. CONCLUSION: Le sévoflurane n'est pas inférieur à l'isoflurane selon un critère d'évaluation composé d'une durée de séjour prolongée à l'USI et de la mortalité toutes causes confondues à 30 jours. Le sévoflurane n'est pas supérieur à l'isoflurane en ce qui touche à n'importe quel autre critère clinique important. Cette étude a été enregistrée au ClinicalTrials.gov, numéro NCT01477151.