This article is more than 5 years old. Information may no longer be current.
MRE more accurate than diffusion-weighted imaging in hepatic fibrosis staging
Click Here to Manage Email Alerts
Magnetic resonance elastography is a more reliable and accurate method of staging hepatic fibrosis, according to data from a recent study.
In a meta-analysis of 14 trials conducted from January 2000 to May 2011 and including 972 patients, researchers evaluated the accuracy of magnetic resonance elastography (MRE) and diffusion-weighted imaging (DWI) in the staging of hepatic fibrosis. Diagnostic odds ratio (DOR), positive and negative likelihood ratios (PLR and NLR) and summary receiver operating characteristic (sROC) were calculated to determine accuracy for both methods across five fibrosis stages.
Staging F0-F1 vs. F2-F4, MRE and DWI resulted in the following pooled values:
- Sensitivity: 0.94 (95% CI, 0.81-0.98) for MRE vs. 0.77 (95% CI, 0.71-0.82) for DWI
- Specificity: 0.95 (95% CI, 0.87-0.98) for MRE vs. 0.78 (95% CI, 0.69-0.85) for DWI
- DOR: 20 (95% CI, 7-57) for MRE vs. 3 (95% CI, 2-5) for DWI
- PLR: 0.06 (95% CI, 0.02-0.22) for MRE vs. 0.30 (95% CI, 0.22-0.40) for DWI
- NLR: 317 (95% CI, 55-1,796) for MRE vs. 12 (95% CI, 6-21) for DWI
- Area under sROC curve: 0.98 (95% CI, 0.97-0.99) for MRE vs. 0.83 (95% CI, 0.79-0.86) for DWI
Staging F0-F2 vs. F3-F4, MRE and DWI resulted in the following pooled values:
- Sensitivity: 0.92 (95% CI, 0.85-0.96) for MRE vs. 0.72 (95% CI, 0.60-0.81) for DWI
- Specificity: 0.96 (95% CI, 0.91-0.98) for MRE vs. 0.84 (95% CI, 0.77-0.89) for DWI
- DOR: 21 (95% CI, 10-45) for MRE vs. 5 (95% CI, 3-7) for DWI
- PLR: 0.08 (95% CI, 0.04-0.16) for MRE vs. 0.34 (95% CI, 0.23-0.50) for DWI
- NLR: 251 (95% CI, 103-609) for MRE vs. 13 (95% CI, 6-29) for DWI
- Area under sROC curve: 0.98 (95% CI, 0.96-0.99) for MRE vs. 0.86 (95% CI, 0.83-0.89) for DWI
Investigators found through a z-test that MRE was significantly more accurate than DWI according to these indicators (P<.05).
“MRE is an accurate, noninvasive method of staging hepatic fibrosis and can replace liver biopsy for monitoring patients,” the researchers concluded. “DWI has a moderate accuracy in distinguishing fibrosis, but is easily integrated with conventional MRI without any extra hardware. Further studies on this issue should focus on the standardization of the parameters for both imaging modalities to make them more feasible in clinical practice.”