Purpose
To explore the diagnostic performance of delayed post gadolinium enhancement MRI (DEMRI) in the diagnosis of Meniere's disease (MD), and to establish an effective MRI diagnostic model.
Materials and methods
This retrospective multicenter study evaluated DEMRI descriptors of patients with Ménièriform symptoms examined consecutively from May 2022 to May 2024. A total of 162 ears (95 MD ears, 67 control ears) were enrolled in this study. Each ear was randomly allocated to a training set (n = 98) and a validation set (n = 64). Logistic regression determined three models for the diagnosis of MD in the training cohort. AUC was applied to evaluate the diagnostic performance of different models. Delong test was used to compare the AUC estimates between the different diagnostic models.
Results
The proposed DEMRI diagnostic model demonstrated good diagnostic performance in both the training (AUC, 0.907) and the validation cohort (AUC, 0.887), outperforming the clinical diagnostic model (Z = 2.503, p = 0.01231; 95%CI:0.033–0.269) in the validation cohort. The AUC value of DEMRI model was higher than combined DEMRI-clinical model in the validation cohort (AUC, 0.796) as well, but there was no statistically significant difference (Z = -1.9291, p = 0.05372). In the training set, the sensitivity, specificity, and accuracy of the DEMRI model were 78.9%, 88.5% and 82.8%, respectively.
Conclusion
A diagnosis model based on DEMRI features for MD diagnosis efficiency was higher than that of clinical variables alone. Therefore, DEMRI should be recommended when MD is suspected because of its significant potential in the diagnosis of MD.