| ObjectiveIn this study, we aims at discussing the different performance of diffusion tensor imaging (DTI) and dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in differentiation of nasopharyngeal carcinoma recurrence from postheraputic changes, comparing the sensitivity, specificity and accuracy of the two modalities and their various parameters, and defining a threshold value of both high sensitivity and specificity for each of the parameters. After all, the ultimate goal is to find an effective modality as well as one or more parameters to differentiate nasopharyngeal carcinoma from posttherapetutic changes, which are available and feasible for clinical practice to detect the early stage of recurrence nasopharyngeal carcinomas.Materials and methods47 patients with nasopharyngeal carcinoma had radiotherapy and/or chemotherapy at least six months ago, and presented abnormal signals in the nasopharynx or skull base on both T1-weighed imagings and T2-weightd imagings were included in this perspective study. Three of them who had failed in the dynamic contrast enhanced magnetic resonance imaging were ruled out. Finally,44 patients,34 males and 10 females, were included in this study. They had a mean age of 54.2±10.9, ranging from 34 to 79. The examinations of DCE and DTI were added.We estimated the ability of DCE and DTI modalities in the differerntation of nasopharyngeal carcinoma recurrence from posttheraputic changes, and measured the values of the parameters. The parameters of DTI include appear diffusion coefficient (ADC), factional anisotropy (FA), trace diffusion-weighted image (TraceW) and relative anisotropy (RA), while the parameters of DCE include time-intensity curve (TIC), maximum contrast enhancement ratio (ERmax), maximum slope (Slopemax). Two senior radiologists with rich experience in head and neck imagings processed the imagings and made decisions whether the abnormal signals indicating malignant or benign lesions. Then we divided these patients into two groups, recurrence group and postthreputic changes group, according to the pathological findings and the clinical follow-ups. First, we calculated the sensitivity, specificity and accuracy of DCE, DTI and DCE+DTI, respectively, by chi-square test. Then we compared the parameters of the two modalities through independent sample T test or Mann-Whitney U test, and defined threshold values of both high sensitivity and specificity for each of the parameters by means of the receiver operating characteristic curve (ROC).ResultsThe recurrence group included 28 patients and the postthreputic changes group 16 patients. The sensitivity, specificity and accuracy of DTI, DCE, DCE+DTI were 100%、88.2%、95.6%、100%、100%、100%、100%、100%、100%, respectively. Among them, DCE and DCE+DTI both had satisfied sensitivity, specificity and accuracy. According the results, the application of DCE alone is enough for the clinical practice, but the feasibility of the DCE modality is not as well as DTI.In our study, the TIC of recurrence group includes three types:rapid early enhancement followed by sustained delayed enhancement (10 patients), rapid early enhancement followed by plateau phase (9 patients), and rapid early enhancement followed by slowly washout (9 patients). The value of Slopemax and ERraax were 1128.9±335 and 90.3±26.8 percent, respectively. The TIC of postthreputic changes group includes two types:gradual increase of enhancement (15 patients) and rapid early enhancement followed by sustained delayed enhancement (1 patient). The mean value of Slopemax and ERmax were 1128.9±335 and 90.3±26.8 percent, respectively. The value of Slopemax and ERmax of the recurrence group were much higher than the value of the postthrepatutic changes group, the difference is statistically significance (P<0.001). We defined a value of 53.2 percent of ERmax as a threshold in differentiation nasopharyngeal carcinoma recurrence from postthreputic changes, which has a sensitivity and specificity of 100%, both. The threshold value of Slopemax was 666.6, which had a sensitivity and specificity as high as 100%, too.The mean values of ADC, FA, TraceW and RA in recurrence group and postthreputic changes group were (0.941±0.279)×10-3s/mm2(0.219±0.042)×10-3s/mm2、 (0.113±0.039) 10-3s/mm2、(0.189±0.040) 10-3s/mm2; (0.195±0.329) × 10-3/mm2、(0.173±0.062)×10-3s/mm2、(0.068.0±0.032) 10-3s/mm2、(0.153 ±0.063) 10"3s/mm2, respectively. The mean values of ADC and RA were higher in the recurrence group than the postthreputic changes group, and the values of FA and TraceW in recurrence group were much lower than the postthreputic changes group. The difference was statistically significant of all parameters (P<0.05), except for RA. We defined a value of 1.244 ×10-3s/mm2of ADC as a threshold in differentiation nasopharyngeal carcinoma from postthreputic changes, which has a sensitivity and specificity of 100%,96.5%, respectively. The threshold value of FA was 0.194X 10-3s/mm2, with a sensitivity and specificity of 82.1%,75%, respectively. The threshold value of TraceW was 0.194×10-3s/mm2, with a sensitivity and specificity of 82.1%,75%, respectively.ConclusionBoth of the two modalities of DCE and DTI could differentia nasopharyngeal carcinoma recurrence from postthreputic changes. Because of DCE had a satisfied sensitivity, specificity and accuracy, the application of DCE alone can fulfill the duty of differentiation. As the feasibility of the DCE modality is no better than DTI, and there is also some contraindications during the performance of DCE, thus the modality of DTI can be a helpful assisted examination in the differentiation of nasopharyngeal carcinoma recurrence and posttheraputic changes. Among all the parameters, ERmax and Slopemax had the most satisfied sensitivity and specificity as high as 100%, so they could be defined as the standard parameters in clinical practice. In the study of DTI, the sensitivity and specificity of ADC value is better than FA value, TraceW value and RA value. The parameters of FA and TraceW did not show any preponderance in differentiation of nasopharyngeal carcinoma recurrence from postthreputic changes. |
PDF Full Text Request