Application of apparent diffusion coefficient values derived from diffusion-weighted imaging for assessing different...

Lymph nodes metastasis is an important factor affecting survival rate and recurrence in cervical cancer patients. Currently, diagnosis of metastatic lymph nodes is mainly based on morphological changes on imaging. However, it is difficult to differentiate normal-sized metastatic lymph nodes with short axis of 5-10mm.

To assess the diagnostic value of apparent diffusion coefficient (ADC) for discriminating different-sized metastatic lymph nodes in patients with cervical cancers.

Pathologically confirmed cervical cancer patients were documented from January 2013 to July 2018 in our hospital. A total of 133 patients who underwent conventional MRI and diffusion-weighted imaging with complete pathology were finally enrolled. A total of 157 lymph nodes were harvested and analyzed. All lymph nodes were divided into three groups according to pathology and their short axis (S) measured on axial T2-weighted imaging: normal-sized (5 mmmean), minimum ADC (ADC_min), and maximum ADC (ADC_max) values of lymph nodes were analyzed and compared among the three groups.

ADC_mean of Groups 1 and 2 were significantly larger than those of Group 3 (P<0.001, P=0.005, respectively). ADC_min of Group 1 were significantly larger than those of Groups 2 and 3 (P<0.001, P<0.001, respectively). ADC_max was not statistically different among the three groups. ADC_mean had the relatively highest area under the curve (AUC) of 0.644 for assessing enlarged metastatic lymph nodes, with a sensitivity of 64.4% and specificity of 67.9%. ADC_min had the highest AUC of 0.758 for assessing normal-sized metastatic lymph nodes, with a sensitivity of 84.7% and specificity of 60.7%.

Diffusion-weighted imaging can be used to differentiate enlarged metastatic lymph nodes from benign lymph nodes, and ADC_min can be further used to identify micro-metastasis in normal-sized lymph nodes.