Abstract
To investigate the supplementary value of the neutrophil-to-lymphocyte ratio (NLR) in diagnosing rheumatoid arthritis (RA) and assessing disease activity, particularly in seronegative RA. This retrospective single-center study included 304 patients from the Department of Rheumatology and Immunology at a tertiary hospital between February 2021 and February 2024. The RA group consisted of 201 patients, and the non-RA control group had 103 patients. Demographic characteristics, blood tests, inflammatory markers (C-reactive protein, erythrocyte sedimentation rate), autoantibodies (rheumatoid factor, anti-cyclic citrullinated peptide antibody), and clinical evaluation data (Disease Activity Score in 28 Joints score) were collected. Multivariate logistic and ordinal logistic regression analyses were performed to identify RA-related factors. The combined diagnostic benefit of NLR was assessed using area under the receiver operating characteristic curve (AUC), net reclassification improvement, and integrated discrimination improvement. Among the 304 subjects, 201 were RA patients (mean age 52.3 ± 11.8 yr, 78.6% female) and 103 were non-RA controls (mean age 50.6 ± 10.2 yr, 73.8% female). The NLR in the RA group was significantly higher than in the non-RA group (3.2 [2.1-4.8] vs 1.8 [1.2-2.3], P < .001). Multivariate regression analysis showed that NLR was an independent predictor for RA diagnosis (odds ratio = 2.15, 95% confidence interval: 1.62-2.85, P < .001), with better diagnostic performance than C-reactive protein and erythrocyte sedimentation rate. In seronegative RA patients, NLR remained significantly diagnostic (odds ratio = 2.01, 95% confidence interval: 1.20-3.37, P = .008). NLR positively correlated with disease activity and was higher in moderate-to-high activity patients (4.2 [3.2-5.5], P < .001). Incremental analysis showed that NLR improved model performance: in the RA population, AUC increased from 0.89 to 0.94 (ΔAUC = 0.05, P < .001), and in seronegative RA, AUC increased from 0.72 to 0.82 (P = .002). For disease activity prediction, AUC increased from 0.85 to 0.91 (ΔAUC = 0.06, P = .004), with significant improvements in net reclassification improvement and integrated discrimination improvement (0.18 and 0.07, respectively). NLR is an independent predictor for diagnosing RA and assessing disease activity, particularly in seronegative RA. It enhances diagnostic sensitivity and model discriminative ability and should be promoted as a key inflammatory biomarker for early RA diagnosis.