Development and multicentric validation of GRAVITS (GRavity of Venous Insufficiency in post-Thrombotic Syndrome scoring): a novel imaging scoring system for post-thrombotic syndrome. A study by the French Society of Cardiovascular Imaging
摘要
A standardized grading system for venous post-thrombotic syndrome (PTS) in cross-sectional imaging is lacking. This study aimed to develop and validate GRAVITS (GRavity of Venous Insufficiency in post-Thrombotic Syndrome scoring), a novel MRI-Venography/CT-Venography-based scoring system for venous obstruction, using a Delphi consensus.
Materials and methodsThe Delphi consensus involved a survey distributed to 22 experts from the French Society of Cardiovascular Imaging. Data was reviewed and discussed in a consensus meeting, where a structured voting process established the GRAVITS system. A retrospective cohort of 40 PTS patients from three centers was analyzed. Twelve venous imaging specialists evaluated venous sequelae in axial MRV/CTV slices using GRAVITS. Inter-observer reproducibility was measured using Krippendorff’s alpha (α) and tested across vein types, imaging types, and vein pathological reference status. Intra-observer reproducibility was measured using weighted Kappa (κw). Correlations between GRAVITS scores and clinical scores (Villalta, CIVIQ-20) were assessed using Spearman tests.
ResultsGRAVITS, defined as an ordinal score with four levels (“A: normal vein” to “D: totally occluded vein”), was evaluated in 851 venous segments (40 patients, median age 43.4, 55% women). Inter-observer reproducibility was substantial (α = 0.607), with the highest reproducibility for the external iliac vein (α = 0.693) and the lowest for the popliteal veins (α = 0.364). TrueFISP and BFFE had the highest reproducibility (α = 0.679), while direct CTV was lower (αK = 0.548). κw for two radiologists was calculated using a random sample of 168 venous segments, yielding values of 0.777 and 0.882 (both p < 0.0001). Average GRAVITS scores were significantly correlated with CIVIQ-20 (rho = 0.515, p = 0.0042).
ConclusionGRAVITS demonstrated substantial reproducibility and significant correlation with CIVIQ-20, supporting its potential as a reliable tool for grading venous sequelae in PTS. Further validation is warranted to refine its clinical utility.
Key Points