Diffusion-weighted magnetic resonance imaging of parotid glands in diabetes mellitus: A preliminary study

Hirotaka Muraoka, Takashi Kaneda, Takumi Kondo, Shunya Okada, Eri Sawada, Satoshi Tokunaga

Research output: Contribution to journalArticlepeer-review

Abstract

Objective: This retrospective study aimed to assess the parotid glands in diabetes mellitus (DM) using diffusion-weighted magnetic resonance imaging (DW-MRI). Methods: We investigated 54 patients aged between 38 and 83 years, with DM (DM group) and without DM (control group), who underwent magnetic resonance imaging (MRI) in the facility between November 2006 and January 2021. We calculated the median apparent diffusion coefficients (ADCs) of the parotid glands in the control group and DM group. These groups were compared using the Mann–Whitney U tests. A receiver operating characteristic (ROC) curve analysis was performed to assess the ability of the ADC to predict DM. Statistical significance was set at p < 0.05. Results: The median ADCs of parotid grands in the control group and DM group were 0.83 × 10−3 mm2/s and 0.74 × 10−3 mm2/s, respectively (p < 0.01). ROC analysis revealed a cut-off ADC value of 0.81 to distinguish between the control and DM group. Sensitivity, specificity, accuracy, and area under the ROC curve were 0.7, 0.67, 0.69, and 0.71, respectively. Conclusion: The ADC was significantly higher in DM groups than in the control group. This suggests that DW-MRI may be able to quantitatively distinguish parotid glands between normal and DM patients.

Original languageEnglish
Pages (from-to)106-111
Number of pages6
JournalOral Science International
Volume21
Issue number1
DOIs
Publication statusPublished - Jan 2024

Keywords

  • apparent diffusion coefficient
  • diabetes mellitus
  • diffusion-weighted imaging
  • magnetic resonance imaging
  • salivary glands

Fingerprint

Dive into the research topics of 'Diffusion-weighted magnetic resonance imaging of parotid glands in diabetes mellitus: A preliminary study'. Together they form a unique fingerprint.

Cite this