TR-FLUID: Evaluation of Intravenous Fluid Administration Practices in Turkish Intensive Care Units: A Multicenter, Prospective, Point-Prevalence Observational Study

TR-FLUID is a multicenter, prospective, point-prevalence observational study of intravenous (IV) fluid practice in adult intensive care units (ICUs) in Turkey. The design follows the methodology of the international SAFE-TRIPS and Fluid-TRIPS point-prevalence studies but extends it in three ways: (1) it captures all four indications for fluid therapy-resuscitation, maintenance, replacement, and fluid creep (hidden fluids such as drug diluents and line-patency fluids)-rather than resuscitation alone; (2) it classifies each patient by the phase of fluid therapy using the ROSE conceptual model (Resuscitation, Optimization, Stabilization, Evacuation); and (3) it collects detailed center-level contextual data, including the locally available fluid portfolio, the presence of a written fluid protocol, intensivist coverage, and who orders the fluids.

Sampling and data window: Each participating ICU selects one of four pre-specified alternative study days, distributed across calendar quarters and across weekdays/weekends to balance seasonal and weekly effects. The index time is defined as 08:00 local time on the chosen day. All eligible patients present in the ICU at the index time are screened and enrolled. For each patient, every IV fluid administered during the 24-hour prospective observation window (08:00 to 08:00 the next day) is recorded, including generic/brand name, volume, route, indication category, and start/stop time. Cumulative fluid balance and cumulative albumin received from ICU admission up to the index time are recorded retrospectively from the hospital information system. The sodium, chloride, and glucose load of each fluid is calculated from standard composition tables.

Multicenter organization: The study is coordinated from Ondokuz Mayıs University. Following coordinating-center ethics approval, an open call is issued to adult ICUs across Turkey through scientific societies and direct contact. Each participating center obtains its own local institutional approval before collecting any data; no center collects data without institutional permission. Each center designates a local investigator responsible for screening, the written informed consent process, and data entry. Written informed consent is obtained from each patient or, for patients unable to provide consent, from a legal representative.

Data collection and standardization: Data are entered through two standardized electronic instruments-a Center Information Form (completed once per center) and a Patient Case Report Form (completed per patient). Before data collection begins, all centers attend an online training session and receive a written operations manual and operational definitions covering fluid-type and indication categorization, ROSE phase assignment, norepinephrine-equivalent dose calculation, and KDIGO acute kidney injury staging. A one-day pilot is conducted at the coordinating center to refine the instruments.

Data management and analysis: Each patient is assigned a center-based unique study identifier; no patient identifiers are transmitted to the coordinating center. Data are stored in an access-restricted, encrypted database subject to automated range and logical-consistency checks and manual verification, with data queries returned to centers for missing or implausible values. Data management complies with the Turkish Personal Data Protection Law (KVKK) and the Declaration of Helsinki, and the study is reported in accordance with the STROBE statement. Statistical analyses are performed using IBM SPSS Statistics. Descriptive results are presented with 95% confidence intervals; determinants of fluid choice are explored using mixed-effects models with center as a random effect, and between-center variability is assessed using the intraclass correlation coefficient and median odds ratio. All association analyses are hypothesis-generating and make no causal claims.