Correlation of blood pressure and clinical presentation with arterial phenotype in renal fibromuscular dysplasia: towards a new classification of lesions in the feiri cohort

Objective: Currently, fibromuscular dysplasia (FMD) lesions are classified binary as either focal or multifocal. This classification does not account for the observed anatomical variety and severity of lesions. In this study, we developed a more detailed classification and explored its correlation with clinical parameters. Design and method: Available renal FMD images (CTA, MRA or angiography) with matching clinical data (+/- 1 year from imaging) were extracted from the European/International FMD Registry (FEIRI). Images were analyzed blinded from clinical data and assigned to 6 categories: focal FMD (group 1), multifocal FMD with typical string of beads with (group 2) or without (group 3) significant narrowing between beads, mild arterial irregularities (group 4), lesions limited to 1-3 beads (group 5), and irregular dilatations (group 6). Relevant clinical data were compared between groups. Results: 174 cases meeting the inclusion criteria were classified, predominantly group 1 (n=48), group 2 (n= 54) and group 3 (n=46). Groups 4-6 were not included in further analysis due to small sample sizes. A gradient of severity was documented across groups 1 to 3: compared to group 3, patients in group 1 and 2 were diagnosed younger with FMD (38.4 and 47.9 vs 53.5 years, p <0.001) and hypertension (32.9 and 37.0 vs 47.4 years, p<0.001), and had higher mean systolic (153.6 and 147.4 vs 136.0 mmHg, p=0.003) and diastolic BP (95.7 and 89.2 vs 83.1 mmHg, p=0.002), with similar number of antihypertensives. Revascularisation proportion was 87.5%, 59.3% and 19.6% in groups 1, 2 and 3 (p<0.001). In focal FMD, a significantly lower proportion of multivessel FMD and higher proportion of men were found compared to groups 2 and 3. Conclusions: Blinded analysis of the FEIRI imaging database confirms the differences between focal and multifocal FMD. More importantly, we introduce new subtypes in multifocal FMD, and show significant correlations between imaging phenotype and clinical parameters. This suggests that centralised analysis of images may predict the hemodynamic impact of lesions. After validation in a larger cohort, this classification may improve objective comparison of arterial lesions between and within FMD registries, as well as in clinical practice.