The aim of this conference was to identify all the interventions influencing mortality in the perioperative period, including only evidence derived from RCTs and meta-analyses of RCTs.

For this purpose, two different search strategies were proposed: one to identify all the randomized trials based on the largely validated research strategy proposed by Robinson et al. [4] and the other a wider search strategy for meta-analyses identification (Table 2.1). To increase the sensitivity of this phase of research, additional strategies were adopted. Experts in the perioperative area were contacted to obtain suggestions on further trials, along with more than 100 international anesthesiological and surgical societies and more than 50 patients and nurses associations. Moreover, the authors contacted all editorial offices of peer-reviewed journals with impact factors relative to perioperative medicine, including surgery, anesthesiology, critical care medicine, and cardiac or cardiovascular systems. Proposal of new articles was encouraged and allowed until the consensus meeting held in Milan.

All studies identified were reviewed at an abstract/title level by a trained team of physicians, to identify all the papers with all the following characteristics:

Only articles fulfilling all these criteria were included in the next step of the consensus process.

Two separate sessions of Web-based polling were proposed, allowing 1,090 participants from 77 countries to participate (Fig. 2.2). Participants covered a large area of medical specialties. The upcoming consensus was advertised via Web, e-mail and by information spreading through scientific networks.

From June 1 to June 8, 2011, participants voted through a Web-poll on the beneficial or detrimental effect on mortality of the interventions proposed in the previous phase. For each intervention, participants were asked to express their agreement, choosing from a nominal scale, and were actively encouraged to express their opinion on a separate open panel. All opinions were collected and reported to the consensus meeting.

Web participants were required to disclose any potential conflict of interest for each proposed intervention, and they were invited to suggest new articles until the date of the consensus meeting.

A meeting was held on June 8, 2011, at the Vita-Salute University of Milan among a task force composed by anesthesiologists, intensivists, surgeons, cardiologists, and epidemiologists. During the meeting, new participant-proposed articles were discussed for inclusion. Every new or former article was double-checked for adherence to inclusion criteria.

Each consensus topic was presented by a reporter and a co-reporter. This was followed by a discussion through consensus building. A simple vote was taken if at least one participant was against the inclusion of the paper in the major topics at the end of the discussion, and a position statement was approved describing the reasons for the inclusion of this topic and its effect on mortality. Final topics [5–18] are reported in Table 2.2, while major excluded topics are reported in Table 2.3. Reasons for exclusion were the following: