Table 5.
Performance of proton density fat fraction compared with liver biopsy for the detection of steatosis in patients with NAFLD
| Design | Comparison with controlled attenuation parameter | Steatosis grade | Cut-off value (%) | AUROC | Se | Sp | PPV | NPV | References |
|---|---|---|---|---|---|---|---|---|---|
| Cross-sectional prospective, single center (N = 51) | No | Grade ≥ 1 | 8.9 | ND | ND | ND | ND | ND | Permutt et al. Aliment Pharmacol Ther. 2012 [124] |
| Grade ≥ 2 | 16.3 | ND | ND | ND | ND | ND | |||
| Grade 3 | 25.02 | ND | ND | ND | ND | ND | |||
| Cross-sectional prospective, single center(N = 77) | No | Grade ≥ 1 | 6.4 | 0.989 | 0.97 | 1.00 | 1.00 | 0.71 | Tang et al. Radiology. 2014 [125] |
| Grade ≥ 2 | 17.4 | 0.825 | 0.61 | 0.90 | 0.90 | 0.61 | |||
| Grade 3 | 22.1 | 0.893 | 0.68 | 0.91 | 0.72 | 0.90 | |||
| Cross-sectional prospective, single center (N = 142) | Yes vs. VCTE (M probe) | Grade ≥ 1 | 5.2 | 0.96 | 0.900 | 0.933 | 0.892 | 0.519 | Imajo et al. Gastroenterology. 2016 [62] |
| Grade ≥ 2 | 11.3 | 0.90 | 0.789 | 0.841 | 0.845 | 0.784 | |||
| Grade 3 | 17.1 | 0.79 | 0.737 | 0.810 | 0.632 | 0.953 | |||
| Cross-sectional prospective, single center (N = 27), Child | No | Grade ≥ 1 | 3.5 | ND | 0.890 | 0.880 | ND | ND | Di Martino M et al. World J Gastroenterol. 2016 [126] |
| Grade ≥ 2 | ND | ND | ND | ND | ND | ND | |||
| Grade 3 | ND | ND | ND | ND | ND | ND | |||
| Cross-sectional prospective, single center(N = 104) | Yes vs. VCTE (M and XL probe) | Grade ≥ 1 | 3.71 | 0.99 | 0.958 | 1.000 | 1.000 | 0.700 | Park et al. Gastroenterology. 2017 [63] |
| Grade ≥ 2 | 13.03 | 0.90 | 0.800 | 0.833 | 0.750 | 0.870 | |||
| Grade 3 | 16.37 | 0.92 | 0.818 | 0.836 | 0.450 | 0.966 |
AUROC Area under the receiver-operating characteristic, NPV negative predictive value, PPV positive predictive value, Se sensitivity, Sp specificity