Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2026

Résumé fourni par la source

Sepsis, life-threatening acute organ dysfunction due to infection (1), is a global health priority (2,3) with approximately 49 million cases and 13 million sepsis-related deaths each year (4–6). Beyond being acutely deadly, sepsis contributes to new and worsened physical, cognitive, and mental health problems in many survivors (7,8). Early identification and treatment are critical to improving outcomes. The Surviving Sepsis Campaign (SSC) guidelines are intended to support clinicians caring for adult patients with sepsis, focusing on management in the hospital, the immediate prehospital setting, and the immediate post-hospital setting. These guidelines incorporate principles of antimicrobial stewardship through responsible antimicrobial use, proper diagnostic strategies, and de-escalation of antimicrobial therapy. The recommendations reflect evidence-based best practice, distilling a large body of research into actionable recommendations. They empower individuals and health systems to make informed choices about care and support improvements in management and outcomes of sepsis (Table 1).TABLE 1.: Table of StatementsMETHODS Details on the guidelines scope, relationship to SSC bundles, history, and sponsorship; committee selection, characteristics, and organization; conflicts of interest management; patient, intervention, comparator, outcome (PICO) question selection; outcomes prioritization; evidence synthesis; certainty of evidence assessment; recommendation formulation; consensus voting; and implications of strong vs. conditional recommendations are presented in Supplemental Digital Content 1 (https://links.lww.com/CCM/H904). We highlight key aspects of the methodology here. The SSC guidelines were fully funded by Society of Critical Care Medicine (SCCM) and European Society of Intensive Care Medicine (ESICM), with methodological support from the Guidelines in Intensive Care Medicine, Development, and Evaluation group. Sponsoring professional societies supported the participation of their representatives. There was no industry support. The 69-person guidelines committee had broad geographic diversity, representing 23 countries, with 38% of panelists currently or previously practicing in a low- or middle-income country (LMIC). The guidelines committee also had broad diversity of clinical professions and disciplines, as detailed in Supplemental Digital Content 1 (https://links.lww.com/CCM/H904). We convened a patient and family advisory panel who worked with the committee throughout the guidelines development process to ensure incorporation of patients’ values and preferences. The guidelines committee was organized into six subgroups based on clinical domain: screening and early management; infection; hemodynamics; respiratory support; adjunctive and additional therapies; and goals of care, transitions of care, and long-term outcomes, as well as a usability workgroup that developed material to support dissemination and uptake of the guidelines. Each subgroup had at least one panelist currently practicing in a LMIC. Evidence Synthesis The process for selecting PICO questions and completing evidence synthesis is detailed in Supplemental Digital Content 1 (https://links.lww.com/CCM/H904). Prioritized patient-centered outcomes were selected a priori for each question. PICO questions addressed in this guideline update are presented in Supplemental Digital Content 2 (https://links.lww.com/CCM/H905). Evidence synthesis followed the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) methodology. Evidence profiles for each PICO question are presented in Supplemental Digital Content 3 (https://links.lww.com/CCM/H906), Supplemental Digital Content 4 (https://links.lww.com/CCM/H907), Supplemental Digital Content 5 (https://links.lww.com/CCM/H908), Supplemental Digital Content 6 (https://links.lww.com/CCM/H909), Supplemental Digital Content 7 (https://links.lww.com/CCM/H910), and Supplemental Digital Content 8 (https://links.lww.com/CCM/H911). The certainty of evidence was graded as high, medium, low, or very low according to GRADE methodology. We used the Evidence to Decision (EtD) framework (9) to support consistent, transparent, and structured formulation of statements across the guideline. The EtD framework considers the balance of effects, the certainty of evidence, as well as patient values, resource intensity, equity, and cost effectiveness (9). The EtD summary of judgements for each PICO question is presented in Supplemental Digital Content 3 (https://links.lww.com/CCM/H906), Supplemental Digital Content 4 (https://links.lww.com/CCM/H907), Supplemental Digital Content 5 (https://links.lww.com/CCM/H908), Supplemental Digital Content 6 (https://links.lww.com/CCM/H909), Supplemental Digital Content 7 (https://links.lww.com/CCM/H910), and Supplemental Digital Content 8 (https://links.lww.com/CCM/H911). We used standardized language to summarize the findings of evidence syntheses based on effect size (i.e., point estimate) and certainty of the evidence, as recommended by GRADE methodology (10). Process for Determining the Type of Statement Figure 1 summarizes the stepwise process we used to determine the type of statement for each PICO question. In contrast to many other guidelines, we always provided a graded recommendation when there was at least low certainty evidence to inform a recommendation. For most PICO questions, we relied most heavily on the balance of desirable and undesirable effects to inform recommendations. However, when the balance of effects was equivalent, we relied on other domains in the EtD framework to determine the type of recommendation and used “either/or” statements when the balance of all EtD domains for two different approaches or therapies were equivalent. However, for PICOs where the comparator was usual care alone, we used the language of “we make no recommendation due to equal balance of effects” since suggestions to use or not use a particular therapy were viewed as potentially confusing and unhelpful.Figure 1.: Process for selecting the type of statement.Types of Graded Recommendations Using the GRADE approach, we classified each graded recommendation as either “strong” or “conditional” (referred to as “weak” recommendations in prior SSC guidelines). We used the language “we recommend” for strong recommendations and “we suggest” for conditional recommendations, consistent with GRADE guidance. Implications of Graded Recommendations A strong recommendation indicates that most, if not all, individuals in the relevant clinical situation should receive, or avoid, the intervention. In contrast, a conditional recommendation acknowledges that the balance between desirable and undesirable may vary depending on patient values, clinical circumstances, or resource availability. Conditional recommendations may not be universally implementable and are less likely to be suitable for rigid performance metrics or enforcement. Flexibility and local context should guide their adaptation into policy. Further details on implication of strong vs. conditional recommendations are presented in Table 2 and Supplemental Digital Content 1 (https://links.lww.com/CCM/H904). TABLE 2. - Implications of Strong Vs. Conditional Recommendations Stakeholder Strong Recommendation Conditional Recommendation Patients and families All or most would want the intervention. Most would want the intervention, a would The should be to all or most There is for for to The should be for most clinicians should clinical and for or performance be less for or performance are on other research may be and may recommendations. We followed the GRADE to statements that were classified as statements to as in prior SSC guidelines These statements that not on a evidence We used this in where evidence was as a a recommendation had to all of the was and addressed a practice, the was the evidence was or to summarize the was and GRADE were not suitable to certain

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.