Performance of magnetic resonance elastography and diffusion-weighted imaging for the staging of hepatic fibrosis: A meta-analysis

Abstract

A meta-analysis was performed to assess and compare the accuracies of magnetic resonance elastography (MRE) and diffusion-weighted imaging (DWI) for the staging of hepatic fibrosis. Online journal databases and a manual search from January 2000 to May 2011 were used. We identified 41 studies, but only 14 met the criteria to perform a meta-analysis assessing MRE (five trials) or DWI (10 trials). Fibrosis was categorized by redistribution into five stages according to histopathological description. A bivariate binomial model was used to combine the sensitivity and specificity and their 95% confidence intervals (CIs), from which diagnostic odds ratio (DOR), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and summary receiver operating characteristic (sROC) were derived to indicate the diagnostic accuracy of imaging modalities. With MRE, the sensitivity, specificity, DOR, PLR, NLR, and area under sROC curve (with 95% CIs) for staging F0 ∼ F1 versus F2 ∼ F4 and F0 ∼ F2 versus F3 ∼ F4 were 0.94 (0.81-0.98), 0.95 (0.87-0.98), 20 (7-57), 0.06 (0.02-0.22), 317 (55-1,796), 0.98 (0.97-0.99) and 0.92 (0.85-0.96), 0.96 (0.91-0.98), 21 (10-45), 0.08 (0.04-0.16), 251 (103-609), and 0.98 (0.96-0.99), respectively; and with DWI, these values were 0.77 (0.71-0.82), 0.78 (0.69-0.85), 3 (2-5), 0.30 (0.22-0.40), 12 (6-21), 0.83 (0.79-0.86) and 0.72 (0.60-0.81), 0.84 (0.77-0.89), 5 (3-7), 0.34 (0.23-0.50), 13 (6-29), and 0.86 (0.83-0.89), respectively. A z test demonstrated that MRE had a significantly higher accuracy than DWI in those indicators (P < 0.05). MRE is more reliable for staging hepatic fibrosis, compared with DWI, with a high combination of sensitivity, specificity, likelihood ratios, DOR, and area under sROC curve.