Three Different GHDT ( Goal Hemodynamic Directed Therapy) Strategies for Intraoperative Fluid Management Optimization During Major Abdominal Surgery: A Randomized Controlled Trial

Hemodynamic

GHDT, Hypotension prediction, Assisted Fluid Management

Major oncological surgery is among the most complex procedures, involving patients with a combination of high-risk factors that can significantly influence immediate postoperative outcomes and quality of life. The intraoperative hemodynamic management of these patients represents a crucial challenge: maintaining cardiovascular stability and fluid balance during the surgery is associated with reduced complications, including acute kidney injury, myocardial ischemia, and sepsis. Literature has shown that intraoperative fluid administration guided by specific algorithms can reduce complications and improve patient outcomes. In recent years, innovations in artificial intelligence (AI) have profoundly changed how hemodynamic variables are managed during surgery. AI enables real-time clinical data processing and offers the possibility to predict imminent hypotension episodes, allowing the medical team to intervene proactively. An example of such technologies is the Hypotension Prediction Index (HPI), which uses a machine learning algorithm to analyze hemodynamic data and predict the risk of hypotension with up to 80% accuracy, up to 10 minutes before it occurs. Therefore, softwares that integrate fluid administration volumes with parameters derived from pulse contour systems are used currently, enabling an analysis of the efficacy of administration of fluid boluses. For example, the Assisted Fluid Management (AFM) software helps the clinician in choosing the timing of fluid administration, determining its effectiveness in terms of fluid responsiveness. This allows to reduce complications related to improper intraoperative fluid management, such as organ damage, and optimize the use of fluids and vasopressor drugs. Despite the growing use of AI in surgery, the clinical and economic impact of such technologies is still under study. Advanced intraoperative hemodynamic management tools have been shown to reduce the duration of hypotensive episodes and improve hemodynamic stability. The clinical impact of such monitoring, in terms of complications and length of postoperative stay, could be crucial to recommend their use in high-risk patient cohorts. This aligns with medical literature showing that postoperative complications increase patient-related hospitalization costs. This study aims to explore the utility of combining a Goal-Directed Hemodynamic Therapy (GDHT) protocol with AI software in three different scenarios. The primary objective of the study is to evaluate if there is a significant difference in intraoperative fluid administration volumes across three different protocols of GDHT supported by AI, in patients undergoing major abdominal oncological surgery. The study's secondary objectives include: * Assess the rate of hypotensive episodes in terms of Time-Weighted Average Hypotension (TWAH) across the three groups. * Analyze the rate of postoperative complications and hospital mortality across the three groups. * Evaluate the total hospital stay duration and/or the number of days spent in intensive care across the three groups. The study aims to provide evidence on the clinical efficacy of haemodynamic monitoring technologies currently present in daily practice. The results will allow us to define an optimization of intraoperative haemodynamic management, improving clinical outcomes and optimizing the use of healthcare resources.

ASSISTED FLUID MANAGEMENT TECHNOLOGY SOFTWARE The Assisted Fluid Management (AFM) technology, developed by Edwards Lifesciences, is an advanced software designed to assist clinicians in optimizing fluid administration during non-cardiac surgeries. Using arterial pressure waveform analysis, the software evaluates in real-time the patient's response to administered fluids and provides personalized recommendations to achieve specific stroke volume variation targets. Thanks to an algorithm that learns from prior data and the patient's current conditions, the system can predict the effectiveness of a fluid bolus and suggest whether and when to administer it, leaving the clinician with final decision-making control. This combination of automated analysis and clinical flexibility makes it a potentially valuable tool for improving intraoperative fluid management and reducing the complexity of therapeutic decisions. Will be a randomized controlled trial including patients undergoing major oncological surgery in tertiary care hospitals, with fluid management aligned with the GDHT philosophy. The analysis will compare three groups: * FLOTRAC Group: Patients managed using the FLOTRAC technology for intraoperative fluid monitoring and management. * HPI Group: Patients managed using the Hypotension Prediction Index (HPI) technology. * HPI-AFM Group: Patients managed using both HPI and AFM technologies for intraoperative fluid management. The study will be multicentric, involving tertiary care hospitals. This will allow us to collect a sufficiently large and representative sample to ensure statistical validity and generalizability of the results. Inclusion Criteria: * Age ≥ 18 years. * ASA physical status II-III-IV. * Patients undergoing elective major abdominal oncological surgery. * Revised Cardiac Index Score ≥ 2. * Plan to perform the procedure with invasive arterial monitoring. * Expected surgical time greater than 120 minutes. Exclusion Criteria: * Emergency or urgent surgeries. * Severe chronic renal failure (creatinine clearance \< 30 ml/min). * Chronic heart failure (NYHA Class IV). * Pregnant women. * Contraindications to pulse contour hemodynamic monitoring. * Liver surgery. * Patient refusal. Data will be collected in three phases: * Preoperative: basic demographic and clinical data. * Intraoperative: hemodynamic parameters and fluid management. * Postoperative: complications, length of stay, and clinical outcomes. Data will be analyzed using parametric and non-parametric tests based on the distribution. A multivariate regression model will be used to control for confounding factors. The study will be conducted in compliance with GDPR regulations. An informed consent will be required from all participants.

No related papers found yet.

Italy