Background: Nailfold capillaroscopy is a non-invasive, cost-effective, and well-established examination used to diagnose various systemic autoimmune diseases and support patient follow-up. Although its clinical significance is acknowledged, the challenge of subjectivity remains a notable obstacle, impeding progress in both research and diagnosis. The categorisation of the three scleroderma patterns established by Cutolo et al adds objectivity to the analysis, although the boundaries between different patterns can be unclear. Addressing this, the pending publication of the CAPI-SCORE algorithm, a fully quantitative system with four straightforward steps and validated through Capillary.io software (end-to-end capillaroscopy analysis system), offers an objective classification for any capillaroscopy. This study presents an external validation utilising an independent cohort, further establishing the algorithm’s robustness and potential impact on enhancing capillaroscopic assessments.

Objectives: To validate the CAPI-SCORE algorithm for detecting capillaroscopic patterns using a wide patient cohort.

Methods: Seven hundred and fifty capillaroscopies were analysed from patients evaluated for Raynaud’s phenomenon (with an average of 27 images and 48 mm analysed per capillaroscopy). The procedure was performed blindly by three expert capillaroscopists to obtain a gold standard. The comprehensive consensus patterns, unanimously agreed upon by all observers, were compared with the CAPI-SCORE algorithm to assess both sensitivity and specificity.

Results: Of all cases, 38% of them had scleroderma patterns and 34% had nonspecific pattern and 28% were normal. Within the scleroderma patterns, 40% of cases belonged to the early pattern, 32.5% to the active pattern, and 27.5% to the late pattern.

In distinguishing between the scleroderma pattern and the non-scleroderma pattern, the accuracy demonstrated by CAPI-SCORE was 91.25%. The accuracy for differentiating between normal and nonspecific patterns was 90%. Within the scleroderma patterns, the overall accuracy was 86.25%.

Conclusion: The CAPI-SCORE algorithm stands out as a straightforward, objective, and easily applicable tool for the accurate identification of capillaroscopic patterns. Its simplicity and effectiveness render it a valuable asset for standardising the interpretation of capillaroscopic images.

REFERENCES: [1] Ng SA, Tan WH, Saffari SE, Low AHL. Evaluation of Nailfold Capillaroscopy Online Training Using the Fast Track Algorithm. J Rheumatol 2023;50:368-372.

[2] Gracia Tello BC, Ramos Ibañez E, Saez Comet L, Guillén Del Castillo A, Simeón Aznar CP, Selva-O’Callaghan A, et al. External clinical validation of automated software to identify structural abnormalities and microhaemorrhages in nailfold videocapillaroscopy images. Clin Exp Rheumatol 2023;41:1605-1611.

[3] Garaiman A, Nooralahzadeh F, Mihai C, et al. Vision transformer assisting rheumatologists in screening for capillaroscopy changes in systemic sclerosis: an artificial intelligence model. Rheumatology 2022;keac541.

[4] Smith V, Vanhaecke A, Herrick AL, Distler O, Guerra MG, Denton CP, et al; EULAR Study Group on Microcirculation in Rheumatic Diseases. Fast track algorithm: How to differentiate a “scleroderma pattern” from a “non-scleroderma pattern”. Autoimmun Rev 2019;18:102394.

Acknowledgements: SPANISH SOCIETY OF INTERNAL MEDICINE (SEMI), SPANISH MULTIDISCIPLINARY SOCIETY OF SYSTEMIC AUTOIMMUNE DISEASES (SEMAIS)

Disclosure of Interests: None declared.