Table 3.
Statistical performance of biomarker combinations to discriminate primary Sjögren’s syndrome (pSS) from rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE).
| Low BDNF | HighI-TAC | Low sCD163 | Low Fractalkine | pSS patients (n) | No-pSS patients (n) | PPV % (pSS) | Se % (pSS) | Spe % (pSS) | ||
|---|---|---|---|---|---|---|---|---|---|---|
| 2 BMKs | X | X | 2 | 3 | 40.0 | 4.8 | 94.3 | |||
| X | X | 3 | 4 | 42.9 | 7.1 | 92.5 | ||||
| X | X | 8 | 2 | 80.0 | 19.0 | 96.2 | ||||
| X | X | 3 | 1 | 75.0 | 7.1 | 98.1 | ||||
| X | X | 1 | 3 | 25.0 | 2.4 | 94.3 | ||||
| X | X | 2 | 6 | 25.0 | 4.8 | 88.7 | ||||
| 3 BMKs | X | X | X | 4 | 2 | 66.7 | 9.5 | 96.2 | ||
| X | X | X | 1 | 2 | 33.3 | 2.4 | 96.2 | |||
| X | X | X | 4 | 8 | 33.3 | 9.5 | 84.9 | |||
| X | X | X | 0 | 0 | – | – | – | |||
| 4 BMKs | X | X | X | X | 0 | 1 | 0 | 0 | 98.1 | |
Estimation of the positive predictive value (PPV), sensitivity (Se), and specificity (Spe) of different biomarker (BMKs) combinations to discriminate pSS (primary Sjogren Syndrome) from RA (Rheumatoid Arthritis) and SLE (Systemic Lupus Erythematosus). In contrast to other biomarkers, BDNF, I-TAC, sCD163 and Fractalkine serum concentrations were significantly associated with pSS diagnosis compared to RA or SLE (OR < 0.8 or > 1.2). Concentrations of these biomarkers were considered high or low relative to their median concentration in the whole cohort.