Diffusion-weighted quantitative MRI to diagnose benign conditions from malignancies of the anterior mediastinum: Improvement of diagnostic accuracy by comparing perfusion-free to perfusion-sensitive measurements of the apparent diffusion coefficient

• Published in: Journal of magnetic resonance imaging Vol. 44; no. 3; pp. 758 - 769
• Main Authors: Priola, Adriano Massimiliano, Priola, Sandro Massimo, Gned, Dario, Piacibello, Edoardo, Sardo, Diego, Parvis, Guido, Torti, Davide, Ardissone, Francesco, Veltri, Andrea
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01.09.2016; Wiley Subscription Services, Inc
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; anterior mediastinum; apparent diffusion coefficient; b value; Diagnosis, Differential; Diffusion Magnetic Resonance Imaging - methods; diffusion-weighted magnetic resonance imaging; Female; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Angiography - methods; Magnetic resonance imaging; Male; Mediastinal Neoplasms - diagnostic imaging; Middle Aged; perfusion; Reproducibility of Results; Sensitivity and Specificity; Young Adult; apparent diffusion coefficient; b value; anterior mediastinum; perfusion; diffusion-weighted magnetic resonance imaging
• ISSN: 1053-1807; 1522-2586
• DOI: 10.1002/jmri.25203

Purpose To compare perfusion‐free to perfusion‐sensitive measurements of the apparent diffusion coefficient (ADC) to diagnose benign conditions from malignancies of the anterior mediastinum. Materials and Methods Seventy‐six subjects were divided into a “benign conditions” group (A, n = 44) and a “malignancies” group (B, n = 32), based on histological findings. diffusion‐weighted magnetic resonance imaging (DW‐MRI) was performed at b of 0/150/800 sec/mm2. The ADCs were obtained on an ADC map by including (perfusion‐sensitive = ADCb0‐800) and excluding (perfusion‐free = ADCb150‐800) the b = 0 sec/mm2. The Mann–Whitney U‐test was used to detect differences in ADCb0‐800 compared with ADCb150‐800 values between all cases, benign conditions, and malignancies. The same test was used to evaluate differences in ADCs between the two groups for each type of measurement (ADCb0‐800 and ADCb150‐800), and receiver‐operating characteristic (ROC) curves were obtained to evaluate discrimination abilities with comparison of areas‐under‐ROC‐curves (AUROC). Optimal cutpoints for discrimination between groups were determined by the Youden‐Index with computation of accuracy. Results The median ADCb0‐800 was significantly greater compared with ADCb150‐800 for all cases (P = 0.0014), benign conditions (P = 0.0412), and malignancies (P = 0.0001). The median percentage of increase was 5.30% for group‐A and 22.39% for group‐B (P < 0.0001). AUROC of ADC in discriminating between groups was significantly greater for ADCb150‐800 (0.932) compared with ADCb0‐800 (0.831) (P = 0.001). The optimal cutpoint for distinction between groups was 1.52 × 10‐3mm2/sec (sensitivity = 93.7%, specificity = 88.6%, accuracy = 90.8%) for ADCb150‐800 and 1.75 × 10‐3mm2/sec (sensitivity = 75.0%, specificity = 79.5%, accuracy = 77.6%) for ADCb0‐800. Conclusion The use of perfusion‐free ADC measurements significantly improves diagnostic accuracy of DW‐MRI in differentiating benign conditions from malignancies of the anterior mediastinum. J. Magn. Reson. Imaging 2016;44:758–769.