DESIGN: A systematic review of the literature followed by a consensus-based voting process.

SETTING: A web-based international consensus conference.

PARTICIPANTS: Two hundred fifty-one physicians from 46 countries.

INTERVENTIONS: The authors performed a systematic literature search and identified all randomized controlled trials (RCTs) showing a significant increase in unadjusted landmark mortality among surgical or critically ill patients. The authors reviewed such studies during a meeting by a core group of experts. Studies selected after such review advanced to web-based voting by clinicians in relation to agreement, clinical practice, and willingness to include each intervention in international guidelines.

MEASUREMENTS AND MAIN RESULTS: The authors selected 12 RCTs dealing with 12 interventions increasing mortality: diaspirin-crosslinked hemoglobin (92% of agreement among web voters), overfeeding, nitric oxide synthase inhibitor in septic shock, human growth hormone, thyroxin in acute kidney injury, intravenous salbutamol in acute respiratory distress syndrome, plasma-derived protein C concentrate, aprotinin in high-risk cardiac surgery, cysteine prodrug, hypothermia in meningitis, methylprednisolone in traumatic brain injury, and albumin in traumatic brain injury (72% of agreement). Overall, a high consistency (ranging from 80% to 90%) between agreement and clinical practice was observed.

DESIGN: Single blinded, international, multicenter randomized controlled trial with 1:1 allocation ratio.

SETTING: Tertiary and University hospitals.

INTERVENTIONS: Patients (n=10,600) undergoing coronary artery bypass graft will be randomized to receive either volatile anesthetic as part of the anesthetic plan, or total intravenous anesthesia.

MEASUREMENTS AND MAIN RESULTS: The primary end point of the study will be one-year mortality (any cause). Secondary endpoints will be 30-day mortality; 30-day death or non-fatal myocardial infarction (composite endpoint); cardiac mortality at 30day and at one year; incidence of hospital re-admission during the one year follow-up period and duration of intensive care unit, and hospital stay. The sample size is based on the hypothesis that volatile anesthetics will reduce 1-year unadjusted mortality from 3% to 2%, using a two-sided alpha error of 0.05, and a power of 0.9.

METHODS: We conducted a pragmatic, multicenter, single-blind, controlled trial at 36 centers in 13 countries. Patients scheduled to undergo elective CABG were randomly assigned to an intraoperative anesthetic regimen that included a volatile anesthetic (desflurane, isoflurane, or sevoflurane) or to total intravenous anesthesia. The primary outcome was death from any cause at 1 year.

RESULTS: A total of 5400 patients were randomly assigned: 2709 to the volatile anesthetics group and 2691 to the total intravenous anesthesia group. On-pump CABG was performed in 64% of patients, with a mean duration of cardiopulmonary bypass of 79 minutes. The two groups were similar with respect to demographic and clinical characteristics at baseline, the duration of cardiopulmonary bypass, and the number of grafts. At the time of the second interim analysis, the data and safety monitoring board advised that the trial should be stopped for futility. No significant difference between the groups with respect to deaths from any cause was seen at 1 year (2.8% in the volatile anesthetics group and 3.0% in the total intravenous anesthesia group; relative risk, 0.94; 95% confidence interval [CI], 0.69 to 1.29; P = 0.71), with data available for 5353 patients (99.1%), or at 30 days (1.4% and 1.3%, respectively; relative risk, 1.11; 95% CI, 0.70 to 1.76), with data available for 5398 patients (99.9%). There were no significant differences between the groups in any of the secondary outcomes or in the incidence of prespecified adverse events, including myocardial infarction.