Optimization of b value in diffusion-weighted MRI for characterization of benign and malignant gynecological lesions

J Magn Reson Imaging. 2012 Mar;35(3):650-9. doi: 10.1002/jmri.22871. Epub 2011 Nov 8.

Abstract

Purpose: To explore the optimal b value in diffusion-weighted (DW)-MRI for differentiation of benign and malignant gynecological lesions.

Materials and methods: Consecutive 58 patients (66 lesions) with pathologically confirmed diagnosis of gynecological disease were included in the study. Routine pelvic MRI sequences were used for defining the lesions and reviewed independently for benignity/ malignity. Single-shot echoplanar imaging (SH-EPI) DW-MRI with eight b values and nine apparent diffusion coefficient (ADC) maps were obtained. The lesions were analyzed qualitatively on DW-MRI for benignity/malignity on a five-point-scale and quantitatively by measurement of apparent diffusion coefficient (ADC) values. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic accuracy of ADC values for differentiating between benign and malignant lesions. Pathology results were the reference standard.

Results: Differentiation between benign and malignant gynecological lesions using visual scoring was found to be successful with b values of 600, 800, or 1000 s/mm(2) . The mean ADC values of malignant lesions were significantly lower than those of benign lesions for all b value (P < 0.005). The ADCs with b = 0 and 600, 0 and 1000 s/mm(2) , 0, 600, 800 and 1000 s/mm(2) , and all b values were more effective for distinguishing malignant from benign gynecological lesions (Az = 0.851, 0.847, 0.848, 0.849, respectively). Using ADC with b = 0, 600, 800, and 1000 s/mm(2) , a threshold value of 1.20 × 10(-3) mm(2) /s permitted this distinction with a sensitivity of 83%, a specificity of 81%.

Conclusion: DW-MRI is an important method, and the optimal b values are between 600 and 1000 s/mm(2) for differentiation between benign and malignant gynecological lesions.

MeSH terms

  • Adult
  • Aged
  • Contrast Media
  • Diagnosis, Differential
  • Female
  • Genital Neoplasms, Female / diagnosis*
  • Genital Neoplasms, Female / pathology
  • Humans
  • Image Interpretation, Computer-Assisted / methods
  • Middle Aged
  • Prospective Studies
  • ROC Curve
  • Statistics, Nonparametric

Substances

  • Contrast Media