A comparative study of multimodal magnetic resonance imaging parameters for distinguishing lacrimal gland lymphoma from immune globulin G4-related lacrimal disease and lacrimal gland inflammatory pseudotumor

Immune globulin G4-related lacrimal disease (IgG4-RLD) and lacrimal gland inflammatory pseudotumor (LGIP) may be misdiagnosed as lacrimal gland lymphoma (LGL) by conventional magnetic resonance imaging (MRI) alone. Thus, we aimed to evaluate the effectiveness of multiple MRI parameters in distinguishing LGL from IgG4-RLD and LGIP. The MRI data, including conventional MRI, diffusion weighted imaging (DWI), and dynamic contrast-enhanced MRI (DCE-MRI), of nine, eight, and nine patients with LGL, IgG4-RLD, and LGIP, respectively, confirmed by histopathology, were collected from October 2010 to August 2021. We compared among the three groups using variance analysis, least significant difference, and rank sum tests. We performed multivariate logistic regression analysis including statistically significant MRI parameters to establish a model and evaluated the diagnostic efficacy using the receiver operating characteristic curve. The apparent diffusion coefficient (ADC) ratios were significantly different among the patients with LGL, IgG4-RLD, and LGIP (0.88 ± 0.11 and 1.14 ± 0.29, P = 0.036; 0.88 ± 0.11 and 1.40 ± 0.27, P < 0.001; 1.40 ± 0.27 and 1.14 ± 0.29, P = 0.030). Additionally, DWI signal intensity (DWI-SI), ADC-SI, and extravascular extracellular space volume per unit tissue volume (Ve) values significantly differed between patients with LGL and LGIP. DWI-SI and ADC ratios were the most valuable parameters for predicting and diagnosing LGL. The areas under the curve of the DWI-SI and ADC ratio for predicting LGL potency were 0.83 and 0.87, respectively. The ADC ratio might aid in the differential diagnosis of LGL, IgG4-RLD, and LGIP. DWI-SI, ADC-SI, and Ve values might facilitate the differential diagnosis between LGL and LGIP. A logistic regression model based on DWI-SI and the ADC ratio showed promise for accurate clinical diagnosis of LGL.