DYNAMIC MONITORING OF ADAMTS 13 ACTIVITY FOR DIFFERENTIAL DIAGNOSIS ACROSS THE SPECTRUM OF SEPSIS ASSOCIATED WITH THROMBOTIC MICROANGIOPATIES

Background: In the ICU, distinguishing immune-mediated thrombotic thrombocytopenic purpura (iTTP) from sepsis-associated thrombotic microangiopathy (TMA) is time-critical. We tested whether serial ADAMTS-13 combined with targeted coagulation and inflammation markers improves iTTP risk stratification in a Sepsis-3 ICU cohort and whether a pragmatic rule-out is feasible. Methods: Prospective single-center study of adults meeting Sepsis-3 with thrombocytopenia and schistocytes ≥ 1% or LDH > 2× ULN within 24 h of ICU admission. ADAMTS-13 activity and VWF: Ag were assayed at 0/24/48/72 h alongside a thrombo-inflammatory panel. We derived a Dynamic ADAMTS-13 Index (DAI), a Coagulation Consumption Index (CCI) anchored to ISTH DIC and fibrinogen/antithrombin III, and an Inflammation Index (IL-6/HBP). The prespecified main rule-out required a ≥ 15% ADAMTS-13 rise by 48 h plus low CCI. A prespecified RCV-anchored sensitivity analysis required ≥ 35% relative rise or ≥ 10 absolute % points plus low consumption. For decision-making, pre-treatment (pre-plasma exchange, PEx) analyses are emphasized. Intent-to-diagnose (care-embedded) analyses are exploratory, and internal validation used a bootstrap optimism correction. Results: Of 1,274 screened, 330 were included (iTTP = 34). Discrimination improved from baseline ADAMTS-13 (AUROC 0.78) to DAI (0.93), with smaller gains after adding CCI (0.95) and the Inflammation Index (0.96). With the main rule-out (≥ 15% + low CCI) in the intent-to-diagnose analysis, sensitivity was 97.1%, specificity was 86.1%, and NPV was 99.6%. The RCV-anchored sensitivity analysis preserved 100.0% sensitivity and NPV with 76.0% specificity. A 72-h phenotype (ADAMTS-13 < 10% with high IL-6/HBP) was associated with higher 28-day mortality (adjusted HR 2.6). Conclusions: In Sepsis-3 ICU patients with TMA features, serial ADAMTS-13 testing, along with targeted coagulation/inflammation markers, enhances early iTTP risk stratification and supports a pragmatic rule-out framework. External validation and implementation studies remain essential. These findings also support investment in rapid/automated ADAMTS-13 activity assays and decision-support workflows to enable timely adoption beyond tertiary centers.