Administration of Normal Saline versus Ringers Lactate in Critically Ill Patients: A Protocol and Statistical Analysis Plan for a Secondary Analysis of the FLUID trial (FLUID-ICU)

IntroductionRecent evidence from randomized controlled trials and meta-analyses in the critically ill suggest small but clinically important differences in death and requirement for renal replacement therapy that favor Ringers lactate as compared to normal saline. To futher contribute to this evidence base, we will perform a secondary analysis of the FLUID trial with a focus on critically ill patients (FLUID - ICU). The FLUID trial was hospital wide cluster randomized cross-over trial that compared normal saline to Ringers lactate. ObjectivesThe primary objectives of FLUID-ICU are to examine the effect of Ringers lactate compared to normal saline compared to on the primary composite outcome and secondary outcomes in critically ill patients. As secondary exploratory objectives, we will evaluate the primary composite outcome and secondary outcomes within prespecified subgroups defined by age, sex, case mix group, Elixhauser Co-morbidity Index, presence of infection with organ dysfunction, traumatic brain injury, and trauma with an injury severity score of >= 12. MethodsFLUID was an open-label two-period, two-sequence, cross sectional, cluster-randomized cross over trial that was conducted in 7 hospitals in the province of Ontario, Canada. Hospitals were assigned to either Ringers lactate or normal saline for a period of 12 weeks and after a washout period, switched to the other fluid at 12 weeks. The primary composite outcome was death or re-admission to hospital within 90 days after the index admission. Secondary outcomes were individual components of the primary outcome as well as length of stay in hospital, an incident emergency department visit post hospital discharge, initiation of dialysis within 90 days after the index admission, and discharge to a facility other than home. All data were obtained from health administrative databases. In FLUID-ICU, we will examine the primary and secondary outcomes in a cohort of critically ill patients, defined as patients admitted to the ICU within 1 day after admission to hospital. A sensitivity analysis will relax this definition to 2 days after admission to hospital. The analyses will be performed at the individual patient level using random effects logistic regression with adjustment for prespecified covariates. Ethics and DisseminationThe results of FLUID-ICU will be published in a peer-reviewed journal. As a secondary analysis of anonymized data, this study does not require research ethics board approval. All data is obtained from a population based health administrative database (ICES) in the province of Ontario, in aggregate and without patient identifiers.