Table 2.
Selected cut-off values of LSM and APRI to predict patients with significant hepatic fibrosis (≥F2), advanced hepatic fibrosis (≥F3), and cirrhosis (F4)
| Significant Hepatic Fibrosis (≥F2) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| LSM (kPa) | Patients Tested, n (%)c |
Actual Fibrosis, n (%)d |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
| All (n = 284) | ≥F2 (n = 101) | <F2 (n = 183) | ||||||||
| 7.1 | 63 (22) | 56 (55) | 7 (4) | 55 | 96 | 89 | 80 | 14.50 | 0.46 | 82 |
| 8.8 | 30 (11) | 30 (30) | 0 (0) | 30 | 100 | 100 | 72 | —e | 0.70 | 75 |
| 5.3a | 116 (41) | 94 (93) | 22 (12) | 93 | 88 | 82 | 96 | 8.11 | 0.08 | 90 |
| APRIb | Patients Tested, n (%) |
Actual Fibrosis, n (%) |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
|---|---|---|---|---|---|---|---|---|---|---|
| All (n = 289) | ≥F2 (n = 103) | <F2 (n = 186) | ||||||||
| 0.50 | 137 (47) | 81 (79) | 56 (30) | 79 | 70 | 60 | 86 | 2.66 | 0.30 | 73 |
| 1.50 | 3 (1) | 3 (3) | 0 (0) | 3 | 100 | 100 | 65 | —e | 0.97 | 65 |
| 0.55a | 109 (38) | 76 (74) | 33 (18) | 74 | 82 | 69 | 85 | 4.04 | 0.32 | 79 |
| Advanced Hepatic Fibrosis (≥F3) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| LSM (kPa) | Patients Tested, n (%) |
Actual Fibrosis, n (%) |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
| All (n = 284) | ≥F3 (n = 40) | <F3 (n = 244) | ||||||||
| 9.5 | 20 (7) | 18 (45) | 2 (1) | 45 | 99 | 95 | 92 | 109.80 | 0.55 | 92 |
| 9.6 | 20 (7) | 18 (45) | 2 (1) | 45 | 99 | 95 | 92 | 109.80 | 0.55 | 92 |
| 8.3a | 40 (14) | 38 (95) | 2 (1) | 95 | 99 | 93 | 99 | 77.30 | 0.10 | 99 |
| APRIb | Patients Tested, n (%) |
Actual Fibrosis, n (%) |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
|---|---|---|---|---|---|---|---|---|---|---|
| All (n = 289) | ≥F3 (n = 41) | <F3 (n = 248) | ||||||||
| 0.75 | 63 (22) | 38 (93) | 25 (10) | 93 | 90 | 61 | 99 | 9.60 | 0.08 | 90 |
| 1.75 | 3 (1) | 1 (2) | 2 (1) | 2 | 99 | 33 | 86 | 3.02 | 0.98 | 85 |
| 0.75a | 63 (22) | 38 (93) | 25 (10) | 93 | 90 | 61 | 99 | 9.60 | 0.08 | 90 |
| Cirrhosis (F4) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| LSM (kPa) | Patients Tested, n (%) |
Actual Fibrosis, n (%) |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
| All (n = 284) | =F4 (n = 14) | <F4 (n = 270) | ||||||||
| 12.5 | 5 (2) | 5 (36) | 0 (0) | 36 | 100 | 100 | 97 | —e | 0.64 | 97 |
| 14.6 | 4 (1) | 4 (29) | 0 (0) | 29 | 100 | 100 | 96 | —e | 0.71 | 96 |
| 9.2a | 25 (9) | 14 (100) | 11 (4) | 100 | 96 | 58 | 100 | 27.00 | 0.00 | 96 |
| APRIb | Patients Tested, n (%) |
Actual Fibrosis, n (%) |
Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR + | LR − | DA (%) | |
|---|---|---|---|---|---|---|---|---|---|---|
| All (n = 289) | =F4 (n = 14) | <F4 (n = 275) | ||||||||
| 1.00 | 28 (10) | 6 (43) | 22 (8) | 43 | 92 | 22 | 97 | 5.61 | 0.62 | 90 |
| 2.00 | 3 (1) | 0 (0) | 3 (1) | 0 | 99 | 0 | 95 | 0.00 | 1.00 | 94 |
| 0.80a | 58 (20) | 14 (100) | 44 (16) | 100 | 85 | 26 | 100 | 7.05 | 0.00 | 85 |
PPV, positive predictive value; NPV, negative predictive value; DA, diagnostic accuracy.
The optimized cut-off value was defined as the one with the minimal (1 − sensitivity)2 + (1 − specificity)2 value and the maximal Youden index (sensitivity + specificity − 1).
The total patient number to evaluate APRI was 289 (including 284 patients included in the study and another five patients who received liver biopsy and APRI but who failed LSM).
Patient number and percentage with LSM or APRI levels ≥ the selected cut-off values.
Patient number and percentage with LSM or APRI levels ≥ the selected cut-off values in defined stages of hepatic fibrosis.
Not applicable because the denominator for the likelihood ratio calculation is zero.