Accuracy of MRI R2*Mapping in Diagnosis of Liver Iron Concentration: A Meta-Analysis

While current guidelines recommend R2* method as the first-line method for liver iron concentration (LIC) measurement, its diagnostic accuracy is debatable. A prior meta-analysis suggested limited accuracy of R2* method for identifying patients with iron overload. However, substantial advances in R2* method over the past decade may have improved its diagnostic performance. The purpose of this study is to explore the accuracy of the R2* method in identifying patients at clinically relevant LIC thresholds (1.8, 3.2, 7.0, and 15.0 mg/g). This is a meta-analysis. On February 16, 2024, databases including PubMed, Medline, Embase, Web of Science, and the Cochrane Library were searched for studies using the R2* method to quantify LIC. R2* values were derived from gradient echo sequence, with LIC from the FerriScan (R) (R2) or biopsy as reference standard. The true positive, false positive, true negative, and false negative at each LIC threshold (1.8, 3.2, 7.0, and 15.0 mg/g) were extracted from each study by two reviewers. Summary receiver operating characteristic (SROC) curves were generated using the hierarchical SROC (HSROC) model. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Twenty-six studies with 1687 patients were included. In identifying patients at each LIC threshold (1.8, 3.2, 7.0, and 15.0 mg/g), the area under the SROC curves (AUCs) were 0.96 or higher, with pooled sensitivities of 0.95 or higher. Pooled specificities were 0.93 or higher; however, studies using biopsy as a reference showed a wide range 95% confidence interval (CI) (0.51-0.99) for identifying patients at the 1.8 mg/g threshold. The R2* method accurately identified patients at different clinically relevant LIC thresholds (1.8, 3.2, 7.0, and 15.0 mg/g). However, when biopsy is used as a reference, the method may be unstable for the 1.8 mg/g threshold (95% CI for pooled specificity, 0.51-0.99).Level of Evidence3Technical EfficacyStage 2