Background:A growing evidence supports the role of microvasculopathy as a primary pathogenic event in systemic sclerosis (SSc). The most commonly used imaging technique to identify microangiopathy in SSc is high magnification videocapillaroscopy (NVC), and reduced capillary density and/or capillary loss, which is a typical feature of “scleroderma microangiopathy”, easily identified by NVC, has been associated with digital ulcers (DUs). Different approaches have been proposed to measure capillary density or capillary loss. Some of these were qualitative methods, others semi-quantitative, others only concerned a limited nailfold area, without ever evaluating the overall density, which is more suitable for quantitative estimate.Objectives:To assess the association between the extent of different values of nailfold capillary density and the presence of DUs and to identify the risk of developing DUs, based on quantitative parameters.Methods:The study involved 54 SSc selected patients (47 women and 7 men, mean age 59.5 years, 50 with limited and 4 with diffuse). The study population came from an ongoing database, that includes clinical and laboratory data of patients with definite SSc. A videocapillaroscope (VideoCap® 3.0, DS Medica, Milan, Italy) with a 200x optical probe was used. During examination, eight fingers (fingers 2–5 of each hand), 4 fields per finger, according to the standard literature were assessed. For each patient, a total of 32 images were collected, then classified as having either “normal”, “non-specific” or the “scleroderma pattern” (SP). Capillary density was defined as the number of capillaries/mm in the distal row, regardless of its shape and morphology. Avascular areas were defined by the absence of loops within a width/area extending over more than 500 microns. For each patient, the SP images were further graded with no/slight reduction of the capillary density (7-9 loops/mm) (NOR), with a well-defined reduction of capillary density (6-4 loops/mm) (RED) and with loss of capillaries (<4) plus avascular areas (AA). Then, the overall percentages were calculated (the number with SP, the number with NOR, with RED and with AA, respect to 32), thus obtaining the quantitative measures. All data were analysed using the MedCalc® version 18.6; 64-bit (MedCalc Software, Mariakerke, Belgium).Results:A total of 1728 images were analyzed. Patients with DUs were 16/54 (29.6%). All patients had a SP, but only five patients showed a SP along the entire nailfold. A comparison between patients with or without DUs showed a significant difference both for the overall extent of AA (p=0.032), and particularly for the overall extent of RED (p<0.001). No significant difference was found regarding the overall extent of the SP (p=0.085). Factor significantly associated with DUs in multivariate analysis was the overall extent of RED (p=0.0286). The ROC curve was very effective at discriminating the capillary feature able to distinguish patients with DUs from patients without DUs. The discriminatory power of the overall extent of RED was very good, with an AUC of 0.948 (95 % CI 0.852 ± 0.990). Then, we calculated the cut-off values of the overall extent of RED for presence/absence of DUs with the highest combination of sensitivity and specificity. The resulting cut-off value (Yourden index of 0.825) was >68.7 (sensitivity 92.31 %; specificity 90.24 %) with a LR+ of 9.46.Conclusion:Our data strongly support that the capillary density between 4 and 6 loops/mm is the best capillaroscopic quantitative measure associated with DUs and able to discriminate the probability of having DUs. If all SSc-specific antibodies and/or other laboratory/clinical parameters are not yet available, the overall capillary density can allow physicians to assess SSc patients easily, regarding DUs and risk for developing DUs.Disclosure of Interests:None declared