OBJECTIVES: In the Conservative vs. Liberal Approach to Fluid Therapy of Septic Shock in Intensive Care (CLASSIC) trial, restriction of IV fluid volumes led to similar overall mortality in ICU patients with septic shock. We assessed if variation in standard IV fluid treatment intensity across sites impacted the effects of fluid restriction. DESIGN: Secondary analysis of randomized clinical trial. SETTING: ICU. PATIENTS: The CLASSIC trial enrolled adult ICU patients with septic shock. We included 1366 participants from 19 sites, representing 88% of the full trial population. All sites with greater than or equal to 15 participants in the standard-fluid group were included in this study. INTERVENTIONS: Restrictive vs. standard IV fluid therapy. MEASUREMENTS AND MAIN RESULTS: We used machine learning (eXtreme Gradient Boosting) to predict the IV fluid volumes in the first 24 hours in the standard-fluid group while accounting for participant characteristics that could contribute to treatment variations. We then classified sites into intensity subgroups based on the mean differences between predicted and administered IV fluid volumes in the first 24 hours in the standard-fluid group. We assessed the intervention effects on mortality, serious adverse events and reactions, days alive without life support, and days alive out of hospital at day 90 across these intensity subgroups, using hierarchical Bayesian models with weakly informative priors. Sensitivity analyses evaluated intervention effects separately in each site. In the standard-fluid group, the median absolute difference between administered and predicted IV fluid volumes was -118 mL (interquartile range, -1,341 to 1,731 mL; full range, -5,873 to 11,761 mL). Sites were categorized into five intensity subgroups. The absolute differences in mortality across these subgroups ranged from -2.7% point to 1.4% point. We found similar effects of restrictive vs. standard IV fluid treatment on all outcomes within the intensity subgroups. Results were similar in the sensitivity analyses. CONCLUSIONS: Among adult ICU patients with septic shock, variation in standard IV fluid volumes across sites did not substantially impact the effects of fluid restriction on outcomes after accounting for patient characteristics.

1st Department of Anaesthesiology and Intensive Therapy Medical University of Lublin Lublin Poland

Biomedical Sciences Department Humanitas University Pieve Emanuele Milan Italy

Collaboration for Research in Intensive Care Copenhagen Denmark

Department of Anaesthesia and Intensive Care IRCCS Humanitas Research Hospital Milan Italy

Department of Anaesthesia and Intensive Care Lillebælt Hospital Kolding Denmark

Department of Anaesthesiology and Intensive Care Medicine Division of Emergencies and Critical Care Oslo University Hospital Oslo Norway

Department of Cardiothoracic Anaesthesia and Intensive Care The Heart Center Copenhagen University Hospital Rigshospitalet Copenhagen Denmark

Department of Clinical Medicine Faculty of Health and Medical Science University of Copenhagen Copenhagen Denmark

Department of Clinical Sciences Danderyd Hospital Karolinska Institutet Stockholm Sweden

Department of Intensive Care Copenhagen University Hospital Rigshospitalet Copenhagen Denmark

Department of Intensive Care Guy's and St Thomas' Hospital London United Kingdom

Department of Internal Medicine Intensive Care Unit University Hospital Pilsen Pilsen Czech Republic

Department of Public Health Section of Biostatistics University of Copenhagen Copenhagen Denmark

University of Bern Bern Switzerland

Citace poskytuje Crossref.org