Intensive glucose control in critically ill adults. A protocol for a systematic review and individual patient data meta-analysis

Objective: The optimal target for blood glucose concentration in critically ill patients is unclear. We will perform a systematic review and meta-analysis with aggregated and individual patient data from randomized controlled trials, comparing intensive glucose control with liberal glucose control in critically ill adults. Data sources: MEDLINE®, Embase, the Cochrane Central Register of Clinical Trials, and clinical trials registries (World Health Organization, clinical trials.gov). The authors of eligible trials will be invited to provide individual patient data. Published trial-level data from eligible trials that are not at high risk of bias will be included in an aggregated data meta-analysis if individual patient data are not available. Methods: Inclusion criteria: randomized controlled trials that recruited adult patients, targeting a blood glucose of ≤ 120mg/dL (≤ 6.6mmol/L) compared to a higher blood glucose concentration target using intravenous insulin in both groups. Excluded studies: those with an upper limit blood glucose target in the intervention group of > 120mg/dL (> 6.6mmol/L), or where intensive glucose control was only performed in the intraoperative period, and those where loss to follow-up exceeded 10% by hospital discharge. Primary endpoint: In-hospital mortality during index hospital admission. Secondary endpoints: mortality and survival at other timepoints, duration of invasive mechanical ventilation, vasoactive agents, and renal replacement therapy. A random effect Bayesian meta-analysis and hierarchical Bayesian models for individual patient data will be used. Discussion: This systematic review with aggregate and individual patient data will address the clinical question, 'what is the best blood glucose target for critically ill patients overall?