Table 4.
Prior Studies of DWI in Chronic Liver Disease
| Study | Population | b-values | ADC | D | F | D* | Inflammation Assessed? | Other Pertinent Features |
|---|---|---|---|---|---|---|---|---|
| Animal models of NAFLD: | ||||||||
| Anderson et al (10) | - 3 mice with normal diet - 14 mice with MCD diet |
21, 301, 601 | ↓ with histologic steatosis and inflammation. | – | – | – | Yes, but multiple linear regression not performed. | Imaging performed after animal sacrifice, therefore no perfusion effects. |
| Joo et al (9) | - 27 rabbits fed fatty diet, with range of NAFLD activity scores (NAS): - 6 with NAS 0; 4 with NAS 1–2; 6 with NAS 3–4; 11 with NAS ≥ 5. |
0, 50, 100, 150, 200, 400, 800 | NS | NS | ↓ with NAS and steatosis. No effect of lobular inf. or hep. ballooning in multiple linear regression. | NS | Yes, and multiple linear regression was performed. | maging performed prior to animal sacrifice. |
| Human studies of NAFLD: | ||||||||
| Poyraz et al (12) | - 78 DEI-FF < 5% (average 1%) - 42 DEI-FF > 5% (average 13%) |
0, 1000 | ↓ in steatosis | – | – | – | No | – |
| Hansman et al (11) | - 27 MRS-FF < 5.56% (average 2%) - 16 MRS-FF > 5.56% (average 13%) |
0, 500 | ↓ correlation with MRS-FF, but not different between two groups. | – | – | – | No | – |
| Eatesem et al (13) | - 12 normals - 9 biopsy-proven NAFLD - 7 HCV + HIV |
0, 600 |
↓ in NAFLD ↓ in HCV/HIV |
– | – | – | No | – |
| Guiu et al (14) | - 68 diabetics, MRS-FF < 5.56% - 40 diabetics, MRS-FF > 5.56% |
0, 5, 15, 25, 35, 50, 100, 200, 400, 600, 800 | ↓. with steatosis | ↑ with steatosis | ↓ with steatosis | No | Excluded patients with fibrosis ≥ F2 by FibroTest. | |
| Leitao et al (15) | - 14 normals - 5 histologically-proven steatosis |
0, 10, 20, 30, 40, 50, 75, 100, 150, 300, 500 | NS | ↓. with steatosis | NS | NS | No | Excluded patients with histologic fibrosis or inflammation |
| Human studies of Fibrosis/Cirrhosis: | ||||||||
| Taouli et al (16) | - 13 normals - 31 chronic liver disease from viral hepatitis or alcohol, 1 with NAFLD |
0, 500 |
↓ with fibrosis ↓ with inflammation |
– | – | – | Yes, but multiple linear regression not performed | Fibrosis and inflammation graded histologically in patients. |
| Luciani et al (18) | - 25 healthy - 12 cirrhosis from alcohol or viral hepatitis, diagnosed by biopsy |
0, 10, 20, 30, 50, 80, 100, 200, 400, 800 | ↓. with cirrhosis | NS | NS | ↓ with cirrhosis | No | Excluded fatty liver patients |
| Patel et al (19) | - 16 normals - 14 with cirrhosis primarily from viral hepatitis or alcohol diagnosed in a variety of ways |
0, 50, 100, 150, 200, 300, 500, 700, 1000 | ↓. with cirrhosis | ↓. with cirrhosis | ↓. with cirrhosis | ↓ with cirrhosis | No | – |
Abbreviations: “-” = not assessed or reported; NS = no statistically significant difference between groups; MCD = methionine and choline deficient; HCV = hepatitis C virus; HIV = human immunodeficiency virus; NAS = NAFLD activity score (sum of steatosis, inflammation, and ballooning scores); DEI-FF = dual echo imaging fat fraction; PDFF = proton density fat fraction; MRS-FF = magnetic resonance spectroscopy fat fraction.