The influence of the b‐value combination on apparent diffusion coefficient based differentiation between malignant and benign tissue in cervical cancer

Abstract

To analyze the influence of different b-value combinations on apparent diffusion coefficient (ADC)-based differentiation of known malignant and benign tissue in cervical cancer patients. A total of 35 patients with stage IB1, IB2, IIA cervical cancer underwent a 3.0T MRI scan prior to radical hysterectomy and pelvic lymph node dissection. Conventional T1- and T2-weighted sequences and a diffusion-weighted sequence (b = 0, 150, 500, 1000 seconds/mm(2)) were performed. Regions-of-interest (ROI) were drawn on ADC maps derived from five different b-value combinations (0, 500; 0, 150, 500; 0, 1000; 0, 150, 500, 1000; 150, 500, 1000 seconds/mm(2)). The influence of the b-value combination on ADC-based differentiation of benign and malignant tissue was analyzed using receiver-operating-characteristics curves. For all b-value combinations, ADCs were significantly lower (P < 0.001) in cervical malignancies (1.15 +/- 0.21.10(-3); 1.10 +/- 0.21.10(-3); 0.97 +/- 0.18.10(-3); 0.97 +/- 0.23.10(-3) and 0.85 +/- 0.18.10(-3) mm(2)/second respectively to the aforementioned b-value combinations) than in benign cervix (2.08 +/- 0.31.10(-3); 2.00 +/- 0.29.10(-3); 1.62 +/- 0.23.10(-3); 1.54 +/- 0.21.10(-3) and 1.42 +/- 0.22.10(-3) mm(2)/second respectively). The diagnostic accuracy was high for all b-value combinations and without statistical differences between the combinations. ADC-based differentiation of benign from malignant cervical tissue is independent of the tested b-value combinations. The results support the inclusion and possible pooling of studies using different b-value combinations in meta-analyses on ADC-based tissue differentiation in cervical cancer.