Vol. 76 - No. 11 MINERVA ANESES IOLOGICA 929 REVIEW ABS R A C Te results of meta-analyses on the effectiveness of high positive end-expiratory pressure (PEEP) and prone posi- tioning in acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) are not consistent. In addition, the meta-analyses on the activated protein C in patients with sepsis combine trials with discordant results. Terefore, the aim of this paper was to give a critical review of these meta-analyses. All relevant meta-analyses were identified by a computerized search of PubMed using combinations of the following terms: acute lung injury, acute respiratory distress syndrome, positive end-expiratory pressure, mechanical ventilation, prone position, drotrecogin, activated protein C, sepsis, and septic patients. A high level of PEEP and prone ventilation was shown to reduce the mortal- ity in patients with severe acute hypoxemic respiratory failure. Although the evidence for the efficacy of activated protein C is not conclusive, it should be considered in patients that are at a high risk for death without any contrain- dications related to bleeding risk. Meta-analysis model s can be very useful for clinical decisions if they include all of the similar papers on a medical topic and are correct from the methodological point of view; however , these results must be checked by a careful and well-informed reader. (Minerva Anestesiol 2010;76:929-36) Key words: Positive End-Expiratory Pressure - Activated protein C - Meta-analysis [Publication ype]. M eta-analyses are research models that com- bine similar individual studies using rig- orous scientific methods and formal statistical techniques to obtain global, quantitative, and unbiased effect estimates. Tus, meta-analyses of randomized clinical trials (RCs) provide a strong level of evidence concerning the effectiveness of treatments and, according to evidence-based medicine para- digm (EBM), 1 are very relevant tools for gain- ing clinical knowledge and making clinical de- cisions. In addition, as recently suggested by Moher et al., 2 the meta-analysis validity has to be evaluat- ed according to preferred reporting items for sys- tematic reviews and meta-analyses (PRISMA). Among the main criti cal topics of the meta- anal- ysis model, the following aspects are very impor- tant: 1) an exhaustive collection of the published and non-published papers on the medical topic of interest; 2) a careful assessment of their validity and their clinically and methodological heteroge- neity to sensibly combine them; 3) the appropriate statistical analysis according to a fixed or random model; 4) an assessment of the publication bias and of the heterogeneity of the results; and 5) the presence of appropriate conclusions, especially if they are supported by a sensitivity analysis. Positive end-expiratory pressure, prone positioning, and activated protein C: a critical review of meta-analyses B. M. CESANA 1 , P. ANONELLI 1 , D. CHIUMELLO 2 , L. GAINONI 2 1 Medical Statistics and Biometry Unit, Department of Biomedical Sciences and Biotechnologies, University of Brescia, Brescia, Italy; 2 Department of Anesthesia and Intensive Care, Fondazione IRCCS Ca’ Granda – Ospedale Maggiore Policlinico, University of Milan, Milan, Italy