Abstract / Summary
Daily sedation interruption (DSI) aims to limit drug accumulation and promote a more awake state, thereby reducing the duration of mechanical ventilation and its associated complications. This is an update of a Cochrane review first published in 2014.
To assess the effects of DSI versus sedation management without DSI on the duration of invasive mechanical ventilation, mortality, intensive care unit (ICU) length of stay, adverse events related to under- or oversedation (including accidental removal of endotracheal tube and tracheostomy), total doses of sedative and analgesic drug administered, and health-related quality of life, for critically ill people requiring intravenous sedation.
We used CENTRAL, MEDLINE All, Embase Classic+Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science Core Collection, and two trial registers, together with reference checking, citation searching, and contact with study authors to identify the studies included in the review. The latest search date was 8 October 2025.
We included randomised controlled trials (RCTs) comparing DSI with sedation strategies that did not include DSI (protocolised sedation, non-protocolised usual care (clinician discretion), or analgesia first/no sedation) in mechanically ventilated, critically ill people (adults and infants or children under 18 years of age).
Our critical outcomes were duration of mechanical ventilation (from randomisation to successful extubation or death), mortality, and ICU length of stay (LOS). Our important outcomes included hospital LOS, adverse events related to under- or oversedation (e.g. accidental removal of endotracheal tube or other lines or catheters, tracheostomy, new onset of delirium occurrence, use of physical restraint, and cardiac events), total drug doses, and health-related quality of life (HRQoL).
We used the original Cochrane tool, RoB 1, to assess bias in the RCTs.
We synthesised results for each outcome using meta-analysis (random-effects modelling). We conducted subgroup and sensitivity analyses according to pre-defined criteria. We used GRADE to assess the certainty of evidence.
This update included 23 trials (one cluster-randomised trial; 18 RCTs in adults, five in children) with 5987 (4910 adjusted) participants (3238 adults, 2750 (1673 adjusted) children). Nine trials used a sedation protocol comparator, nine used 'usual sedation practices' with sedative and analgesic drugs at the clinical team's discretion, and five used analgesia-first or no-sedation comparators.