Application of diffusion-weighted MR imaging with ADC measurement for distinguishing between the histopathological types of sinonasal neoplasms

Abstract

PurposeTo evaluate the potential contribution of quantitative DWI parameters including ADCmean and ADCratio values to help in distinguishing the histopathological types of sinonasal neoplasms. MethodsThis retrospective study included 83 patients (50 males, 33 females; mean age 61 years) with pathologically proven untreated sinonasal neoplasms who have undergone diffusion-weighted MRI imaging from February 2010 to August 2017. Diffusion-weighted MRI was performed on a 3 T unit with b factors of 0 and 1000 s/mm2, and ADC maps were generated. Mean ADC values of sinonasal tumors and ADC ratios (ADCmean of the tumor to ADCmean of pterygoid muscles) were compared with the histopathological diagnosis by utilizing the Kruskal-Wallis non-parametric test. ResultsMean ADCmean and ADCratio were 0.8 (SD, ±0.4) × (10−3 mm2/s) and 1.2 (SD, ±0.5), respectively, and each parameter was significantly different between histopathological types (p < 0.05). Mean ADCmean and ADCratio were higher in adenoid cystic carcinoma (ACC) than in SCC, lymphoma, neuroendocrine carcinoma and sinonasal undifferentiated carcinoma (SNUC) (p < 0.05). Optimized ADCmean thresholds of 0.79, 0.81, 0.74 and 0.78 (10−3 mm2/s) achieved maximal discriminatory accuracies of 100%, 79%, 100% and 89% for ACC/SNUC, ACC/SCC, ACC/neuroendocrine carcinoma, and ACC/lymphoma, respectively. ConclusionsThe optimized ADCmean threshold of 0.80 (10−3 mm2/s) could be used to differentiate ACC from non-ACC sinonasal neoplasms with maximal discriminatory accuracy (82%) and sensitivity of 100%. However, there is considerable overlapping of the ADCmean and ADCratio values among non-ACC sinonasal neoplasms hence surgical biopsy is still needed.