BackgroundThe SQUEEZE trial is a multicentred randomized controlled trial which seeks to determine the optimal approach to fluid resuscitation in paediatric septic shock. SQUEEZE also includes a nested translational study, SQUEEZE-D, investigating the value of plasma cell-free DNA for prediction of clinical outcomes. ObjectiveTo present a pre-specified statistical analysis plan (SAP) for the SQUEEZE trial prior to finalizing the trial data set and prior to commencing data analysis. DesignSQUEEZE is a pragmatic, two-arm, open-label, prospective multicentre randomized controlled trial. SettingCanadian paediatric tertiary care centres. ParticipantsPaediatric patients with suspected sepsis and persistent signs of shock in need of ongoing resuscitation. Sample size target: 400 participants. InterventionsThe trial is designed to compare a fluid-sparing resuscitation strategy to usual care. Main outcome measuresThe primary outcome for the SQUEEZE trial is the time to shock reversal (in hours). The primary outcome analysis will assess the difference in time to shock reversal between the intervention and control groups, reported as point estimate with 95% confidence intervals. The statistical test for the primary analysis will be a two-sided t-test. Secondary outcome measures include clinical outcomes and adverse events including measures of organ dysfunction and mortality outcomes. ResultsThe SAP presented here is reflective of and where necessary clarifies in detail the analysis plan as presented in the trial protocol. The SAP includes a mock CONSORT diagram, figures and tables. Data collection methods are summarized, primary and secondary outcomes are defined, and outcome analyses are described. ConclusionsWe have developed a statistical analysis plan for the SQUEEZE Trial for transparency and to align with best practices. Analysis of SQUEEZE Trial data will adhere to the SAP to reduce the risk of bias. RegistrationClinicalTrials.gov identifiers: Definitive trial NCT03080038; Registered Feb 28, 2017. Pilot Trial NCT 01973907; Registered Oct 27, 2013.
Read full abstract