| Number of studies | Pooled sensitivity (95% CI) | (%) | Pooled specificity (95% CI) | (%) | Pooled PLR (95% CI) | Pooled NLR (95% CI) | Pooled DOR (95% CI) | AUC |
| Part 1: the diagnostic accuracy for uNGAL to identify LN | All (9) | 0.84 (0.71, 0.91) | 86.20 | 0.91 (0.70, 0.98) | 88.93 | 9.08 (2.31, 35.69) | 0.18 (0.09, 0.35) | 50.51 (8.15, 313.03) | 0.92 | subgroup (4) | 081 (0.81, 0.91) | 89.90 | 0.95 (0.89, 0.98) | 62.70 | 10.75 (2.65, 43.57) | 0.21 (0.07, 0.67) | 60.96 (6.09, 610.22) | 1.00 | Part 2: the diagnostic accuracy for uNGAL to identify active LN | All (10) | 0.72 (0.56, 0.84) | 91.17 | 0.71 (0.51, 0.84) | 94.24 | 2.45 (1.32, 4.54) | 0.39 (0.22, 0.70) | 6.24 (2.08, 18.68) | 0.77 | Cross-sectional subgroup (4) | 0.87 (0.71, 0.95) | 84.56 | 0.82 (0.57, 0.94) | 84.04 | 4.89 (1.63, 14.67) | 0.16 (0.06, 0.43) | 30.60 (4.30, 217.68) | 0.92 | Prospective cohort subgroup (6) | 0.57 (0.37, 0.75) | 89.83 | 0.61 (0.40, 0.79) | 94.92 | 1.46 (1.12, 1.91) | 0.71 (0.56, 0.88) | 2.07 (1.39, 3.08) | 0.62 | Part 3: the diagnostic accuracy for uNGAL to predict renal flare | All (6) | 0.80 (0.57, 0.92) | 72.50 | 0.67 (0.58, 0.75) | 66.15 | 2.41 (1.57, 3.72) | 0.30 (0.11, 0.79) | 8.08 (2.02, 32.35) | 0.74 | R-SLEDAI subgroup (3) | 0.90 (0.80, 0.96) | 55.40 | 0.74 (0.66, 0.81) | 21.70 | 3.41 (2.31, 5.03) | 0.17 (0.07, 0.40) | 20.75 (6.13, 70.29) | | Five studies without Elewa 2015 | 0.73 (0.54, 0.86) | 65.05 | 0.63 (0.56, 0.70) | 43.96 | 1.97 (1.41, 2.77) | 0.43 (0.23, 0.82) | 4.55 (1.74, 11.87) | 0.68 | Part 4: the diagnostic accuracy for uNGAL to identify proliferative LN | All (2) | 0.87 (0.66, 0.97) | | 0.69 (0.39, 0.91) | | 2.89 (1.26, 6.61) | 0.20 (0.06, 0.65) | 16.42 (2.56, 105.37) | |
|
|