POS0025 COMBINED NAILFOLD CAPILLAROSCOPY AND ULTRASONOGRAPHY OF THE NAIL-ENTHESIS COMPLEX TO DISCRIMINATE PSORIATIC DISEASE FROM RHEUMATOID ARTHRITIS PATIENTS AND HEALTHY SUBJECTS

• Published in: Annals of the rheumatic diseases Vol. 82; no. Suppl 1; pp. 218 - 219
• Main Authors: Cafaro, G., Venerito, V., Valentini, V., Bursi, R., Perricone, C., Gerli, R., Bartoloni, E.
• Format: Journal Article
• Language: English
• Published: Kidlington BMJ Publishing Group Ltd and European League Against Rheumatism 01.06.2023; Elsevier B.V; Elsevier Limited
• Subjects: Angiogenesis; Artificial Intelligence; Blood vessels; Capillaries; Elbow; Factor analysis; Fingernail; Pilot projects; Psoriasis; Psoriatic arthritis; Rheumatoid arthritis; Scientific Abstracts; Synovitis; Synovium; Ultrasound; Ultrasound; Artificial Intelligence; Psoriatic arthritis
• ISSN: 0003-4967; 1468-2060
• DOI: 10.1136/annrheumdis-2023-eular.4700

BackgroundThe distal interphalangeal joint (DIP), the extensor tendon enthesis and fingernail, in psoriatic arthritis (PsA) represent a very interesting site where the dermatologic, synovial, and entheseal features of the disease merge. The nail-enthesis complex (NEC) can be well analysed with ultrasound (US), and the nailfold is the ideal site to assess angiogenesis, another typical feature of psoriatic synovitis, with both power Doppler (PD) and nailfold videocapillaroscopy (NVC). Studies have shown that PsA patients may have differences in the structure of the nail complex and capillaries compared with healthy individuals. No study has combined the two techniques.ObjectivesTo evaluate the ability of combined US examination of the NEC and NVC to differentiate PsA from psoriasis (PSO), rheumatoid arthritis (RA), and healthy subjects (HC).MethodsTwenty age- and sex-matched subjects per group were consecutively enrolled. Each subject was blindly assessed and underwent US of the NEC, NVC, and clinical examination. For the US examination, the nail plate, matrix, and bed were assessed in grayscale (GS) and with PD according to the Brown University Nail Enthesis Scale (BUNES)[1]. For the nail plate, the Wortsman[2] classification was also evaluated. The thickness of the nail plate, matrix, and nail bed were measured. Mean values of 10 digits were used for the analysis. Capillary density, number of microhaemorrhages, tortuous capillaries, ectasia, and ramified capillaries were assessed at NVC. Mean values were calculated for II-V digits in both hands. After excluding subjects with missing values, differences among groups were analysed with Kruskal-Wallis test. US and NVC variables were then included in a K-Means clustering model. The number of clusters was determined using the elbow method. Factor analysis was performed to explain the clustering results. The number of factors was selected using the Scree plot method.ResultsThe BUNES GS total score, the BUNES GS score for the plate, and the Wortsman score were significantly lower in HC compared to all other groups. BUNES GS for the plate was higher in PSO than in HC. The number of tortuous capillaries was higher in PsA than in PSO and HC, but not in RA. The clustering model identified five clusters (Figure 1A), showing good separation of PsA and HC. PSO were distributed from cluster 2 to cluster 5 and RA among all clusters. At factor analysis (Figure 1B), all US GS variables loaded on factor 1, US PD variables loaded on factor 2, matrix and bed thickness loaded on factor 3, and number of tortuous capillaries and ectasias loaded on factor 4. Loadings of Factor 5 and 6 were below 0.5.ConclusionThe results of this pilot study confirm that PsA patients have significant changes in the structure of the NEC compared with healthy subjects. However, most of these changes do not appear to be specific to PsA, as they are also found in PSO and, most surprisingly, in RA where DIP joint involvement is not present. Interestingly, PsA patients appear to have higher numbers of tortuous capillaries compared with PSO and HC, consistent with increased angiogenesis in the synovium. The combined evaluation seems to provide a good distinction between PsA and HC. Not surprisingly, PSO patients clustered closer to PsA (although with significant overlap with HC), which may be due to subclinical involvement. RA patients are distributed among all clusters and have alterations similar to those of PsA. This study does not provide a clear explanation for this phenomenon, and further studies are needed. This pilot study allowed us to identify US and NVC features that can be selected and applied in a prospective study that includes early undifferentiated arthritides and new-onset PSO, to understand whether the observed differences occur early in the disease and whether changes occur during the course of the disease. A larger sample size could also improve the accuracy of the clustering model.References[1]Cunha JS et al. doi:10.3899/jrheum.170146[2]Wortsman X et al. doi:10.1016/j.det.2006.03.014Figure 1.Acknowledgements:NIL.Disclosure of InterestsNone Declared.