Presentation: Diagnostic Validity in Detecting Different Capillaroscopic Patterns by Qualitative versus Quantitative Scoring Methods (2007)

11 Diagnostic Validity in Detecting Different Capillaroscopic Patterns by Qualitative versus Quantitative Scoring Methods

Objective. Nailfold video capillaroscopy (NVC) has became an establish test used in the assessment of patients with Raynaud’s phenomenon (RP) and connective tissue diseases (CTDs). NVC offers the potential of monitoring microvascular disease progression and/or treatment response. Cutolo et al. have developed a semiquantitative scoring method that has shown the ability to assess the extent of microvascular damage in patients with Systemic Sclerosis, but the advantages of this method to assess different capillaroscopic patterns has not been studied so far. This study aims to evaluate if this scoring method is able to distinguish the different capillaroscopic patterns.
Methods. One hundred and ninety two consecutive adult patients who attended the Department of Rheumatology at Istituto Gaetano Pini (Milano, Italy) to perform NVC were included in this study. In all cases, we captured all the areas of the nailfold in which there was good capillary visibility, and the evaluation of capillaroscopic images was based on qualitative (normal, minor abnormalities [non specific findings], major abnormalities [other findings] and scleroderma-type findings) and semiquantitative scales.
Results. A total of 751 nailfold images were analyzed. Scoring the diagnostic and progression parameters together with a cut-off value ≥5 is the best way to discriminate between normal patterns (normal + minor abnormalities) and pathologic patterns (major abnormalities + scleroderma pattern). The area under the Receiver Operating Characteristic (ROC) curve is 0.969, the 95% confidence interval (C.I.) is 0.933 to 0.988, with a sensitivity of 92.1, specificity of 90.3, and positive likelihood ratio of 9.49. Furthermore, scoring the diagnostic parameters separately with a cut-off value ≥4 is the best way to discriminate between the scleroderma pattern and all the other capillaroscopic patterns. The area under the ROC curve is 0.977, the 95% C.I. is 0.944 to 0.993, with a sensitivity of 93.3, specificity of 91.2, and positive likelihood ratio of 10.5.
Conclusions. Our findings demonstrate that the semiquantitative capillaroscopic scoring method is a useful tool to discriminate between different capillaroscopic patterns. Particularly, scleroderma pattern can be identified when the overall score of three diagnostic parameters is ≥4. Whereas pathologic patterns (major abnormalities and scleroderma pattern) can be identified when the overall score of diagnostic and progression parameters is ≥5. This grading system is an interesting method to depict and quantify microvascular involvement in patients with RP and CTDs.

 F. Ingegnoli, None.