Article Text

FRI0258 A Novel MRI-Based Longitudinal Scoring System for Arterial Involvement in Large-Vessel Vasculitis
  1. E. Tombetti1,
  2. A. Zia2,
  3. D. Gopalan2,
  4. A. Kiprianos1,
  5. K. Bechmam1,
  6. B. Ariff2,
  7. J.C. Mason1
  1. 1Rheumatology
  2. 2Radiology, Imperial College London, London, United Kingdom


Background Although prevention of arterial progression is a hard therapeutic goal in large-vessel vasculitides (LVV), there is neither an agreed definition of progression nor a proposed method to globally assess arterial injury. This lack of validated clinimetry limits the feasibility of clinical trials. Although distinguishing disease activity from damage in Takayasu arteritis (TA) is difficult, current damage indices lack imaging data. Outside LVV, the only validated quantitative methods to assess arterial damage are NASCET and ECST, both designed for carotid atherosclerosis.

Objectives Starting from NASCET and ECST, we sought to develop an objective widely applicable magnetic resonance (MR)-based system for scoring large artery injury in TA and large vessel-giant cell arteritis (LV-GCA).

Methods NASCET was selected as a starting point as it only requires intraluminal data which are: i) more widely accessible in routine clinical practice, ii) require less acquisition time than arterial wall sequences, iii) more directly describe the haemodynamic derangement in LVVs, where wall thickening may occur in the absence of demonstrable luminal changes. NASCET was modified to account for the intrinsic characteristics of LVV by: i) defining a core set of arteries to be evaluated, with the final score determined by summing individual artery scores, ii) measuring the reference diameter for every artery in a plane judged “uninvolved”; iii) including the length of the stenotic segment in the algorithm, iv) scoring aneurysmal disease similarly to stenotic disease, based on the percent diameter dilatation and length of the involved segment (Fig. 1A). A cohort of 60 patients with TA or LV-GCA was selected, and those with vascular progression or improvement identified based on the radiologist's judgement. Cross-sectional and longitudinal analysis by two independent observers was performed. The K-inter-observer variability and the sensitivity of the score in distinguishing patients with stable or progressive vascular involvement was evaluated.

Results Although more time-consuming than conventional assessment, our initial impression is that this new index is more precise. To date, we have evaluated 41 scans from 21 patients and scoring of the remaining images will be concluded in 3 months. Arterial involvement was classified as type IIa, IIb, III and V in 2, 1, 1 and 17 patients respectively. The median vascular score of the 21 pts was 16 (range 7-48). Analysis of Type V disease identified pts with both minimal and extensive arterial artery injury (median 24, range 9-42), demonstrating the precision of the index. 12 scans were judged to be stable, 6 improved and 2 worsened. This correlated with changes in the arterial score (median from baseline scan of -1.5 (range -5 to -1) for improved scans, of 0 (range 0 to 1) for stable scans and 7.5 (range 1 to 14) for worsening scans (p<0.001 stable versus improved, and 0.022 stable versus progression, Fig. 1B).

Conclusions We are evaluating a new index for objectively quantifying arterial injury in LVV. Initial data suggest this approach precisely assesses arterial injury and reflects its evolution over time. This novel index might represent an important clinical measure of arterial disease also suitable for inclusion in LVV damage indices and hence a possible outcome measure for new clinical trials.

Disclosure of Interest None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.