| Hepatocellular carcinoma (HCC) is one of the most common malignant tumor in the world. With a very low5year overall survival rate and poor prognosis, HCC is one of the most threatening disease globally. Complete surgical resection is a potential curative treatment for HCC. Early diagnosis plays a crucial role in the management and improvement of the prognosis for patients with HCC. More than80%of patients with HCC have underlying cirrhosis with varied etiology. Hepatitis-fibrosis-cirrhosis-HCC is the most widely recognized pathological progression pattern for hepatocarcinogenesis. It has also been believed that cirrhosis usually accompanied by regenerative nodule (RN) and dysplastic nodule (DN), which can predispose small HCC(less than2cm). The diethylnitrosamine (DEN)-induced HCC in rat shares the similar biological behavior as human being’s, which has been considered as an appropriate animal model to investigate HCC of human being. From the first to fourth week after administration of DEN, the animal may experience hepatitis, while the fifth to eighth week fibrosis represents, and the ninth to twelfth week cirrhosis. Nodule formation in liver can be observed at the beginning of the eighth week following the administration DEN.Sufficient blood flow is essential for the liver to maintain its basic function. Decline in blood perfusion may lead to hepatocellular dysmetabolism and predispose to damage in hepatic function. The hemodynamic parameters of liver have been regarded as one of the important biomarkers reflecting its function. It is valuable to monitor the hemodynamic changes in liver for the diagnosis and prognosis evaluation of HCC, as the angiogenesis is necessary for tumor formation, progression and metastasis. With the progression of hepatic disorder and the abnormally of hepatic tissue, there is a significant decline in blood perfusion for hepatic tissue at the early stage of cirrhosis compared to hepatitis and fibrosis. RN is supplied mainly by portal vein. Undergoing malignant transformation, blood from portal vein supplying nodules decreased while blood from hepatic artery increased gradually. HCC is mainly supplied by hepatic artery. Blood flow in HCC is raised compared to the normal hepatic tissue, while it declined in DN. However, the vessels in small DN are as more as HCC.Clinically, liver biopsy is the most commonly employed to diagnose the early stage of hepatic nodules. As there are usually multiple lesions in hepatic tissue, missed diagnosis is not uncommon, and it could not possibly have all of the nodules examined with liver biopsy. Even more, biopsy is an invasive technique which has a risk of complication. It is not only very important to monitor the hemodynamic changes and angiogenesis of the nodules in cirrhosis with noninvasive methods, but also helpful in the early diagnosis and prognosis evaluation. Conventional post-enhanced CT and MRI can be used to detect angiogenesis of tumor, but the need of contrast material administration hampered them the most appropriate candidates for disease follow-up. With the advancement of medical imaging equipment and post-processing software, functional magnetic resonance imaging is widely used, such as dynamic contrast-enhanced MRI (DCE-MRI), diffusion weighted imaging (DWI), blood oxygen level dependent MRI (BOLD MRI), and magnetic resonance spectroscopy (MRS). By quantification of water diffusion in vivo, DWI can evaluate the function of human body noninvasively. DWI is regarded very sensitive in detecting molecular movement, and can reflect the freedom of water molecular diffusion, which can determine the composition of the organization in the micro-molecular level. The tumor containing dense cells may limit water molecular diffusion, DWI showed high signal and the apparent diffusion coefficient (ADC) attenuation. DWI can select more than two b values, contributing to detection of the focal lesions, especially small focal liver lesions.The blood oxygen level dependent (BOLD) MR imaging is an alternative imaging technique that involves the use of deoxyhemoglobin as an endogenous contrast mechanism to reflect alterations in blood oxygenation, blood flow,and blood volume. When deoxygenated hemoglobin in one region of the organization increased, the signal on T2*WI decreased. BOLD MR can provide quantitative information of the oxygen level within the organization, and accurately reflect tumor angiogenesis, which is expected to observe the transformation process from dysplastic nodule to early HCC, and contribute to early diagnosis and therapeutic assessment of HCC. BOLD MRI is widely used to investigate functional activation of the brain, they are increasingly being used for functional imaging of tumors, cardiac, renal tissue, skeletalmuscle, and the liver. BOLD MRI in the liver is mainly used for acute hemorrhagic, liver after partial hepatectomy, and liver fibrosis. By comparison of the signal change before and after transcatheter arterial embolization of liver tumors the assessment of liver tumors after transcatheter arterial embolization oxygenation expected to become a non-invasive evaluation of liver tumors involved in the curative effect of the technology. BOLD MR imaging can depict hepatic hemodynamic alterations during the progression of fibrosis and has the potential to serve as a noninvasive, nonenhanced imaging method for liver fibrosis diagnosis and staging. However, the report of BOLD MRI for the liver nodules is still relatively rare.ObjectiveTo observe the imaging performance of focal hepatic nodules and investigate the value of the ADC value and BOLD MRI to differentiate the nodules in a rat model of DEN-induced irreversible hepatocellular carcinoma.MethodsEighty male Wistar rats were randomly divided into treatment group (64rats) and control group (16rats). The rats in treatment group were given water joined by DEN (0.99mg/ml) in for14weeks. The rats in control group were given clean water. After DEN administration for5-20weeks, MR examinations were performed in four rats of the treatment group every week. Two healthy rats performed the MRI scans every two weeks. MR scanning sequences included axial and coronal T2WI and axial T1WI, DWI and BOLD. Ten b values were used during the scanning (Os/mm2,50s/mm2,100s/mm2,150s/mm2,200s/mm2,300s/mm2,400s/mm2,600s/mm2,800s/mm2,1000s/mm2). BOLD scanning methods:after anesthesia was induced, the rats were administered room air (78%N2,20%O2) and carbogen gas (95%O2,5%CO2) separately through a nose cone for10minutes to acquire the multiple-gradient-echo MR images. The animal was killed after MRI. The liver was removed and cut sequentially into3-mm sections in the transverse plane that corresponded as closely as possible to the MRI plane. The tissuse was fixed in10%buffered formaldehyde solution and embedded in paraffin for hematoxylin eosin (HE) staining. The ROI was placed in the nodules of the solid part on the ADC maps and T2*WI maps for3-4times, avoiding the blood vessels and bile duct as far as possible. Then we compared the ADC value and the R2*values of the liver nodules before and after the inhaling of carbogen gas.Statistical methods:The data was analyzed by SPSS13.0statistic software.Values are presented as number and mean±SD. If variance to be equalized when two sets data compared with each other, the data were analyzed with independent sampler t test. If variance to be not equalized, data were analyzed with Welt t test. Comparisons among groups of the ADC value, R2*air, R2*carb, and ΔR2*among groups were performed by one way ANOVA analysis. If the variance to be equalized, two sets data were analyzed with LSD test. If the variance to be not equalized, two sets data were analyzed with Dunnett’s T3test. A ROC (receiver operating characteristic) curve was used to analyze the sensitivity and specificity of the ADC value, R2*air and R2*carb, and also the stansard error and95%confidence interval were given. A p≤0.05was considered significant.ResultsNine rats died before the MR scan during the experiment. During the hepatoma formation procession, liver fibrosis was the first occurred, and then cirrhosis, liver regenerative nodule formation to a dysplastic nodule formation of HCC. A total of78 nodules included in the study, including10RN,16DN,52HCC.1. T1WI and T2WI performanceOn T1WI, RN appeared isointensity (70%) or hyointensity (20%), and mainly hyperintensity on T2WI (60%). The DN was isointensity (68.75%,11/16) or hypointensity (25%,4/16) on T1WI. On T2WI, it showed hypo-, hyper-or iso-intensity (18.75%,18.75%,62.5%). Most HCC showed hypointensity (57.69%,30/52) or isointensity (36.54%,19/52) on T,WI, and hyperintensity on T2WI (96.15%,50/52).2. DWI and the ADC valuesOn DWI, with the b value increased, the liver lesions showed much clearer. RN mainly on DWI images showed hyperintensity, of which6was hyperintensity, and1hypointensity,3isointensity.68.75%of DN showed hyperintensity (11/16), and18.75%showed isointensity (3/16),12.5%for hypointensity (2/16). HCC on DWI showed high signal accounting for94.23%(49/52), and isointensity for1.92%(1/52), hyointensity for3.85%(2/52). This set of experiments selected10different b values. With the b value increased, the surrounding liver background was suppressed gradually, and the signal of the lesions increased significantly. The mean ADC values of RN, DN, HCC, was (0.860±0.099)×10-3mm2/s,(0.853±0.141)×10-3mm2/s,(0.716±0.147)×10-3mm2/s, respectively. The ADC value of the control group was (0.887±0.125)×10-3mm2/s. The ADC value of four groups was statistically significant (F=9.447, P=0.000). After the LSD, the ADC value of RN between DN and the control group had no significant difference (P=0.902vs P=0.631). The ADC value of RN was statistically different from HCC (P=0.003).There was no statistically significant difference between DN and the control group (P=0.492). There were statistically significant difference between DN and HCC, HCC and the control group (P=0.001vs P=0.000). A cutoff ADC value being equal to or more than0.765×10-3mm2/s was used to differentitate RN, DN from HCC with a sensitivity of80.8%and specificity of73.1%. Using a cutoff ADC value equal to or more than0.782×10-3mm2/s for differentitating DN from HCC, the sensitivity was75%and specificity was76.9%. 3. The BOLD MRI findings3.1In the control group, the T2*air values of RN, DN and HCC before breathing carbogen gas were (12.813±2.889)ms,(14.908±4.862)ms, and (13.830±4.862)ms,(24.358±8.831)ms respectively, while the R2*air value were (81.750±18.003)/s, respectively,(72.928±20.596)/s (80.230±25.609)/s (47.541±20.640)/s respectively. The difference of T2*air the value between RN and DN (P=0.993), between RN and the control group (P=0.766) was not statistically significant, T2*air value difference between DN and the control group was not statistically significant (P=0.975). T2*air value of RN is greater than that of HCC (P=0.000), and T2*values of DN is greater than that of HCC (P=0.000), T2*value of HCC group is greater than that of the control group (P=0.000). R2*air value of RN, DN, HCC and the control group have statistical significance (P=0.000). After pairwise comparison, RN, DN, R2*air value difference between the control group was statistically significant (P=0.394, P=0.304), R2*air value between DN and the control group was statistically significant (P=0.839). R2*air value of RN is greater than that of HCC (P=0.001), R2*air the value of DN is greater than that of of HCC (P=0.000), the HCC group R2*air value is less than the control group (P=0.000).3.2After breathing carbogen gas in the control group, RN, DN of HCC T2*carb values, respectively (14.352±2.925) ms,(15.350±4.947) ms, and (15.485±6.015) ms,(27.059±12.503)ms. The R2*carb value were (72.467±15.114)/s, respectively,(70.030±17.572)/s,(71.775±24.462)/s,(43.471±17.144)/s.After breathing carbogen gas, RN, DN, HCC and the control group between the average R2*carb value difference was statistically significant (P=0.000). Through pairwise comparison, the R2*carb value difference between the RN and DN (P=0.814), RN with the control group (P=0.742) not statistically significant, the R2*carb value difference between the DN and the control group had no statistical significance (P=0.915); RN, DN, and the control group R2*carb values were greater than that of HCC (P=0.000). A cutoff R2*air value equal to or more than56.99/s was used to differentitate RN, DN from HCC with a sensitivity of84.6%and specificity of75.0%. Using a cutoff R2*carb value equal to or more than53.09/s for differentitating RN, DN from HCC, the sensitivity was80.8%and specificity was75.0%.3.3The ΔR2*values after breathing carbogen gas in decreasing order followed by the control group, DN HCC, RN. There were no significant difference among the groups (F=1.446, P=0.235).3.4The R2*air of liver tissue surrounding precancerous nudoles and HCC before breathing carbogen gas were (79.675±10.562)/s and (66.115±14.243)/s. The R2*air value after breathing carbogen gas were respectively (71.037±8.747)/s and (62.106±13.77)/s. The ΔR2*were respectively (8.637±7.287)/s and (4.009±5.762)/s. The difference of R2*air, R2*carb and ΔR2value between the two groups were statistically significant (P=0.003, P=0.026,P=0.048).Conclusion1. DWI is sensitive to molecular diffusion, which can help to detect the liver nodules. The ADC values of the normal liver parenchyma, RN, DN and HCC are gradually decreased. The ADC value of normal liver parenchyma, RN and DN were significantly higher than that of HCC. But there was no visible disparity between RN and DN. Using a cutoff ADC value being equal to or more than0.765x10-3mm2/s was used to differentitate RN, DN from HCC with a sensitivity of80.8%and specificity of73.1%. Combined DWI and ADC value has a potential value to differentiate hepatic focal lesions.2. T2*and R2*values between the precancerous nodules, RN and DN, and HCC are statistically significant before and after breathing carbogen gas. The R2*value of the surrounding liver tissue of precancerous nodules is greater then that of HCC before and after breathing carbogen gas. The T2*and R2*values can provide usefull information for identification of RN, DN, HCC and liver tissue surrounding the nodules.3. Before and after breathing carbogen gas, the ΔR2*values of RN, DN, HCC and the healthy liver parenchyma had no statistical difference. But the ΔR2*values of of the surrounding liver tissue of hepatic nodules could be used to provide information to predict the nodules. |
PDF Full Text Request