Individualized anticoagulation monitoring in critically ill patients: study protocol for a prospective observational comparison of Anti-Xa, viscoelastic testing, and thrombin generation

Introduction: Critically ill patients experience high rates of thromboembolic and bleeding events despite the use of pharmacological thromboprophylaxis or weight-based anticoagulation. Laboratory assays for drug monitoring, as Anti-factor Xa (Anti-Xa), overcome the limitations of activated partial thromboplastin time (aPTT) but only quantify drug effects and do not reflect global haemostasis. Viscoelastic testing (VET) may provide additional functional information. Robust clinical data in the ICU setting are limited. Objective: Primary objective is to assess the association between VET clotting time (CT) in Russell viper venom (RVV) assay and Anti-Xa activity. Secondary objectives include assessing whether VET or Anti-Xa aligns more closely with global haemostatic function, defined by thrombin generation assay (TGA) variables. Exploratory analyses aim to characterise associations among VET parameters, standard laboratory tests, and biomarkers; model dose-exposure-response relationships; and identify clinical modifiers as sepsis or renal dysfunction. Sepsis status will be incorporated as predefined effect modifier in mixed-effects models. Study design: This single-center, prospective, observational cohort study enrols critically ill adults receiving unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH). Sampling is standardized to pharmacokinetics: 4 h post-dose for subcutaneous LMWH and 4-6 h post-start for i.v. UFH. At each time point, Anti-Xa, VET on ClotPro® (RVV, EX, and RVH assays), standard laboratory tests, and advanced haemostatic markers (factor VIII, factor XIII, PF1 + 2, TGA, PAP, PAI, TFPI, TAT) are collected. Two repeated measurements per patient capture intra-individual variability. Expected results: We hypothesize that CT-RVV demonstrates a clinically interpretable monotonic relationship with Anti-Xa activity under ICU conditions. We further anticipate discordance between Anti-Xa and global haemostatic markers in a relevant subset of patients, particularly in sepsis, supporting the hypothesis that VET aligns more closely with functional haemostasis than Anti-Xa alone. These findings are expected to delineate when fixed dosing is insufficient, identify predictors of under- or over-anticoagulation, and guide personalised anticoagulation strategies. If RVV-CT consistently correlates with Anti-Xa across clinical states, VET could streamline monitoring where laboratory turnaround time limits timely dose adjustment; if not, the results will clarify the boundaries of VET applicability and highlight scenarios requiring laboratory confirmation or TGA. Registration: Ethics approval was obtained from the responsible committee of Technical University Dresden (BO-EK-338082024). The study is registered with the German Clinical Trials Register (DRKS00037385), registration date: 10 July 2025, https://drks.de/search/de/trial/DRKS00037385.