The Diagnostic Role of Normalized and Mean Apparent Diffusion Coefficient in Differentiation between Pancreatic Lesions

Abstract

Background: Diffusion-weighted magnetic resonance imaging (DWI) with quantification of apparent diffusion coefficient (ADC) values is well established in the diagnosis of a variety of abdominal abnormalities. Regarding pancreatic disease, several investigators have shown that DWI with ADC measurement helps detect and characterize focal pancreatic lesions, as well as assess the severity of other pancreatic conditions.
Objective: The study aimed to evaluate the diagnostic role of both normalized and mean apparent diffusion coefficient in discrimination between pancreatic lesions.
Patients and Methods: Thirty-one participants presented with pancreatic lesions using clinical examination and ultrasound. Dynamic contrast MRI abdomen with diffusion-weighted MR imaging (DWI) on a 1.5-Tesla MRI machine was done. Mean ADC and normalized ADC (as the ratio of ADC of the lesion to the adjacent normal pancreas) were measured and compared.
Results: Our study reported a mean ADC cut value of ≤1.47 while the cut-off value for normalized ADC was ≤ 0.96. Normalized ADC revealed a higher sensitivity 92.31%, specificity 88.89%, PPV 85.71, NPV 94.12%, accuracy 90.32%, positive likelihood ratio 8.31 and negative likelihood ratio 0.09 as compared to mean ADC, which revealed sensitivity 84.62%, specificity 77.78%, PPV 73.33%, NPV 87.5%, accuracy 80.65%, positive likelihood ratio 3.81 and negative likelihood ratio 0.2 respectively.
Conclusion: Measuring the mean and normalized ADC value in pancreatic focal lesions can significantly differentiate between benign and malignant pancreatic lesions. However, normalized ADC has a higher sensitivity, specificity, PPV, and NPV than mean ADC value and could be used to differentiate between pancreatic lesions with higher accuracy than mean ADC.