Table 2.
Diagnostic accuracies of the included studies for TWEAK to predict active LN from lupus patients.
| Author, year | Sample type | Measurement method | TP | FN | FP | TN | TWEAK cutoff | Sensitivity | Specificity | AUC (95% CI) |
|---|---|---|---|---|---|---|---|---|---|---|
| Schwartz, et al. 2009 | Urine | ELISA | 15 | 15 | 5 | 44 | 13 pg/mg Cr | 50% | 90% | 0.724 (NR) |
| Tan, et al. 2009 | Urine | ELISA | 24 | 10 | 1 | 11 | 4.34 pg/mg Cr | 69.70% | 92.31% | 0.790 (NR) |
| El-shehaby, et al. 2011 | Urine | ELISA | 45 | 5 | 10 | 13 | 9.1 pg/mg Cr | 89% | 56% | 0.816 (NR) |
| Choe, et al. 2016b | Serum | ELISA | 26 | 6 | 20 | 18 | 395.0 pg/mL | 81.3% | 47.5% | 0.648 (0.52–0.78) |
| Martínez, et al. 2017 | Urine | ELISA | 9 | 2 | 3 | 8 | 4.91 pg/mg Cr | 81% | 75% | 0.876 (0.75–0.99) |
| Dong, et al. 2018 | Urine | ELISA | 30 | 18 | 1 | 21 | 12.53 pg/mg Cr | 62.22% | 93.33% | 0.815 (0.699–0.930) |
| Dong, et al. 2018a | Urine | ELISA | 14 | 25 | 2 | 18 | 26.95 pg/mg Cr | 36.7% | 88.9% | 0.626 (0.427–0.825) |
| Salem, et al. 2018 | Urine | ELISA | 14 | 0 | 6 | 24 | 8.22 pg/mg Cr | 100% | 80% | 0.96 (NR) |
| Mirioglu, et al. 2020b | Serum | ELISA | 12 | 3 | 3 | 12 | 1542.2 ng/mL | 80% | 80% | 0.796 (0.622–0.969) |
aThis study compared the level of uTWEAK with urine albumin/creatinine ratio in proteinuria detection in patients with LN. Since proteinuria is a component of rSLEDAI scoring system, the study was also included.
buTWEAK levels were also measured in the indicated studies. However, there was lack of mandatory indexes regarding uTWEAK in distinguishing patients with active LN in the studies. TWEAK: TNF-like weak inducer of apoptosis; LN: lupus nephritis; TP: true positive; FN: false negative; FP: false positive; TN: true negative; AUC: area under the receiver operating characteristic curve; CI: confidence interval; ELISA: enzyme-linked immunosorbent assay; Cr: creatinine; NR: not reported.