OBJECTIVES: To investigate the abnormal characteristics of nailfold microcirculation (NMC) in patients with rheumatoid arthritis (RA) and to further analyze the associations between these abnormalities and common systemic inflammatory indicators or the key immunoregulatory cell subpopulation Th17/Treg cells. METHODS: This cross-sectional observation study enrolled 120 RA patients as the RA group and 60 gender- and age-matched healthy individuals as the healthy control (HC) group. A NMC microscope was used to assess the NMC status and clinical data were collected. Spearman correlation was used to evaluate the relationships between NMC abnormalities and inflammatory or immunological indicators. Lasso regression and random forest model were used to screen key predictors. Receiver operating characteristic (ROC) curves were used to assess the diagnostic efficacy of the Th17/Treg ratio for moderate-to-severe NMC abnormalities, and the goodness-of-fit and clinical utility were validated using the Hosmer-Lemeshow test, calibration curves, and decision curve analysis (DCA). Bootstrap method was used for internal validation. RESULTS: Compared to the HC group, the RA group possessed higher number of crossed capillary loops, morphology score, flow pattern score, perivascular score, and total NMC score but lower flow velocity. Spearman correlation analysis demonstrated that the total NMC score was positively correlated with the platelet-to-lymphocyte ratio (PLR) (p = 0.028), platelet-to-red blood cell ratio (PAR) (p = 0.037), systemic immune-inflammation index (SII) (p = 0.009), and interleukin-2R (IL-2R) (p = 0.046). A total NMC score >4 was defined as moderate-to-severe NMC abnormality. Lasso regression and random forest model identified Th17/Treg ratio was the only independent associated factor for this subgroup and the area under the curve (AUC) was 0.708. The Hosmer-Lemeshow test and calibration curve confirmed favorable goodness-of-fit, and the DCA curve confirmed considerable clinical benefit of this model. The mean AUC from internal validation was 0.706, indicating satisfactory model stability. CONCLUSIONS: Our study illustrated the severity of NMC abnormality was significantly associated with systemic inflammatory burdens (PLR, PAR, SII) and core indicators of T-cell activation and dysfunction (IL-2R and Th17/Treg ratio).