Fluid Challenge in Intensive Care The FENICE II Study

Primary aim: The primary aim is to describe the modality of fluid administration during the first 5 days of ICU stay considering 1) the overall fluid balance; 2) the characteristics of the fluids given; 3) the modality of fluid administration. Secondary aims: * To explore any association between fluid administration characteristics and clinical outcomes (see further) * To evaluate factors potentially associated with the respective proportion of the different modalities of fluid administration * To characterize FC administration modality in a large cohort of ICU patients. Statistical Analysis: Data will be described as median and interquartile range (IQR) or number and percentage. Categorical variables were compared using Fisher's exact test and continuous variables using the nonparametric Wilcoxon test, Mann-Whitney test, or Kruskal-Wallis test. Volume of fluid and fluid balance (primary outcome) will be reported as median \[IQR\] from day 1 to day 5. This volume, fluid balance and respective volume of bolus and continuous fluid administration will be reported as distinct alluvial plots reporting median of fluid at day 1 to 5 per quartile along with outcome. In way to assess relationship between volume of fluid received, characteristics and outcome (secondary outcome), longitudinal cluster modelling will be used to identify clusters of patients with similar patterns of fluid administration profile. Longitudinal k-mean will be used to assess change in fluid received each day from day 1 to day 3. In way to avoid misinterpretation of findings due to time dependent competing events such death or ICU discharge, this analysis will be performed on patients alive and still in the ICU at day 3. A sensitivity analysis will be performed during first 5 days on the subgroup of patients alive and not discharged during first 5 ICU-days. In way to assess impact of fluid administration modality/strategy on outcome (secondary endpoint), factors associated with in-hospital mortality, including identified cluster of fluid administration, will be assessed using mixed logistic regression where in-hospital mortality will be event of interest. Center effect will be included as a random effect against the intercept. For secondary outcomes, and in particular for day-30 mortality, number of days alive without vasopressors, mechanical ventilation or renal replacement therapy, will be assessed using survival analysis. All tests will be two-sided, and P-value less than 0.05 will be considered statistically significant.