| Objective To compare respiratory-triggered (RT) and breath-hold (BH) diffusion-weighted imaging (DWI) in assessing therapeutic effect of transarterial chemoembolization (TACE) for hepatocellular carcinoma.Methods Institutional review board approved this study, and all informed consent were obtained.24 clinically-diagnosed hepatocellular carcinomas (HCCs) (mean diameter,6.6cm; range,1.7-13.1cm) of 18 patients (17men,1 woman; mean age, 53.5 years; range,35-74 years) from August 2010 to January 2011 were enrolled. All patients underwent RT-DWI and BH-DWI with b values of 0,500 s/mm2 one week after TACE. The differences of ADC values, contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR) and lesion-to-liver relative contrast ration between RT-DWI and BH-DWI were tested using Paired-Samples T Test. Receiver operating characteristic curve (ROC) analysis was performed for the comparison of viable tumor and necrosis tissue.Results There were no statistical differences regards to the ADC values of viable tumor tissue between RT-DWI and BH-DWI one week after TACE, and so do the necrotic tissue and normal liver and spleen parenchyma. The CNR of necrotic and viable tumor tissue respectively in RT-DWI were higher than that in BH-DWI; and there were significant differences (P=0.000 and P=0.001). The SNR of the normal liver was significantly better on RT-DWI than on BH-DWI (P=0.000). The relative contrast ratios of necrotic tumor tissue to liver in RT-DWI were higher than that in BH-DWI (P=0.011); no significant difference was noted in the relative contrast ratio of viable tumor tissue to liver between the two techniques (P=0.075>0.050). ADC value obtained with b value of 0,500 s/mm2 can effectively distinguish viable tumor tissue from necrosis tissue on both RT-DWI and BH-DWI. For BH-DWI, the area under ROC curve (AUC) was 0.833, and a threshold value of 1.99×10-3mm2/s permitted distinction with sensitivity of 95.8% and specificity of 80%; while for RT-DWI, the AUC was 0.915, and a threshold value of 2.02×10-3mm2/s permitted distinction with sensitivity of 79.2% and specificity of 90%.Conclusion Respiratory-triggered DWI should be preferred over breath-hold DWI for the early evaluation of tumor response after chemoembolization of hepatocellular carcinoma, because it provides better image quality and can identify viable and necrotic tissue in tumor more clearly. Objective To evaluate the feasibility and capability of respiratory-triggered diffusion-weighted imaging in follow-up after transarterial chemoembolization for hepatocellular carcinoma; and to determine whether the pretreatment apparent diffusion coefficients (ADCs) of hepatocellular carcinoma are predictive of response to transarterial chemoembolization and to compare the ADCs of tumors before and after transarterial chemoembolization.Methods Institutional review board approved this study, and all informed consent were obtained.21 clinically-diagnosed hepatocellular carcinomas (HCCs) (mean diameter,6.6cm; range,1.7-13.1cm) of 16 patients (15 men,1 woman; mean age, 54.9 years; range,35-74 years) from August 2010 to January 2011 were enrolled. All patients underwent RT-DWI and BH-DWI with b values of 0,500s/mm2 prior to TACE, and one week and one month after TACE. The differences of ADC values among pretreatment, one week and one month after TACE were analyzed by One-Way ANOVA. The ADCs between non-responding and responding before treatment was analyzed by Mann-Whitney U. Receiver operating characteristic curve (ROC) analysis was performed for analyzing the pretreatment apparent diffusion coefficients (ADCs) of hepatocellular carcinoma to predict response to transarterial chemoembolization and for analyzing the percentage of apparent diffusion coefficients (ADCs) to evaluate therapeutic effect of transarterial chemoembolization for hepatocellular carcinoma.Results The ADCs of necrotic tumor tissue after TACE had a significant increas (one week post-treatment (2.29±0.28)×10-3 mm2/s;one month post-treatment (2.43±0.25)×10-3mm2/s;pretreatment(1.16±0.15)×10-3mm2/s;F=187.349, P=0.000).There was no significant difference of the ADCs of residual tumor issue post-treatment (F=0.836, P=0.446).Before treatment, a significant statistical deference of the ADCs between non-responding lesions and responding lesions was found (Z=-2.289, P=0.022). The results of receiver operator characteristic (ROC) analysis for differentiation of non-responding and responding lesions with mean apparent diffusion coefficients for images showed that threshold ADC value of 1.225×10-3mm2/s had 77.8% sensitivity and 81% specificity. Responding lesions had a significant increase in %ADC values than did non-responding lesions (one week post-treatment Z=-3.507,P=0.000; one month post-treatment Z=-3.960,P=0.000).The results of receiver operator characteristic (ROC) analysis for differentiation of non-responding and responding lesions with mean ADCs changes for images before TACE and one week after TACE showed that threshold %ADC value of 73.02% had 81% sensitivity and 88.9% specificity for identification of responding lesions;that of before TACE and one month after TACE showed that threshold %ADC value of 72.52% had 90.50% sensitivity and 88.9% specificity for identification of responding lesions. No significant ADC value change was observed in non-treatment liver parenchyma (F=0.041, P=0.726) and spleen (F=1.013, P=1.000) after TACE.Conclusion RT-DWI is useful in the follow-up imaging after TACE. A significant increase in mean ADCs was observed in lesions that responded to chemoembolization. High pretreatment mean ADC of HCC was predictive of poor response to chemoembolization.
|
PDF Full Text Request