Experimental And Clinical Studies Of Multi-slice Spiral CT Perfusion Image In Hepatic Cirrhosis

Experimental and Clinical Studies of Multi-slice Spiral CT Perfusion Image in Hepatic CirrhosisPurposeImaging diagnosis of early-stage hepatic cirrhosis has been difficult because early morphological changes are not obvious. Hemodynamic study already showed blood flow changes in arterial and portal system of the liver in this stage. In this experiment, multi-slice spiral CT liver perfusion and scanning and quantitative analysis of early hemodynamic changes were carried out in early hepatic cirrhosis model in rabbit. The results of these studies facilitate detection of early hepatic cirrhosis before morphological changes and provide theoretical basis for early diagnosis and estimation of therapeutic effect and prognosis.This study explored the correlation of perfusion parameters of multi-slice spiral CT perfusion image and HIF-1α, VEGF expression, and thereby provide theoretical basis for early diagnosis and estimation of therapeutic effect and prognosis and for gene-target treatment of liver disease.The correlation of different-degree patients' CT perfusion parameters and different hepatic cirrhosis degrees were studied in order to establish quantitative diagnostic criteria of different-degree hepatic cirrhosis; at the same time, the correlation between hemodynamic changes, hepatic volume changes, and Child-Pugh staging of liver function was studied for better estimation of liver function reserve.Experimental Materials and Methods1. Animal experiment(1) Model preparation of test animal 80 New Zealand albino rabbit, both sexes, weighting 2.0-3.1kg, were randomly divided into two groups, 50 in experimental group and 30 in control group. In the experimental group, subcutaneous injection of arachis oil- diluted 50% carbon tetrachloride was given, twice weekly, with each dose of 1.5ml/kg in the first week, each dose of 2ml/kg in the second week, each dose of 2.5ml/kg from the third week, the overall time was 12 weeks. In the control group the same dosage of normal saline was given by the same route.(2) CT scanning and examination techniqueSpiral CT perfusion scanning was carried out in experimental group before drug use and every second week after drug use starting from the second week in five experimental rabbits and five controls. The scanner used was Light Speed 16MSCT from GE corporation. Scanning mode: cine-scanning 0.5s/eircle, tube voltage 120KV, tube current 60mA, slice thickness 5mm, slice distance 5mm, matrix 512x512,. Before scanning ultravist(300mgI/ml) was injected into the ear edge vein with high pressure injector, at rate of 2ml/s and dose of 1ml/kg. At the beginning of contrast medium(ultravist ) bolus injection, rapid dynamicscanning was started immediately and lasted for 50s.(3) Analysis and calculation of perfusion parametersData processing was made by use of software of deconvolution model liver perfusion in perfusion 3. By processing with this software the following perfusion parameters were obtained: hepatic blood flow(HBF), mean transit time(MTT), hepatic arterial flow(HAF) and analog computation of hepatic arterial perfusion(HAP). HAP=HBFxHAF. HPP(hepatic portal perfusion)=HBFx(1-HAF). The changes of blood flow perfusion parameters were analyzed in the period from the end of the 2nd week to the end of 12th week.(4) Dynamic observation of biochemical index of bloodBefore experimental model preparation and in every second week after the preparation, blood was taken from the ear edge vein for examination of ALT, AST, ALB.(5) Sample preparation.The animal was immediately sacrificed after CT perfusion and scanning, and liver parenchyma corresponding to scan slices were taken. Part of the taken tissue was placed in 10% formalin fixation solution for immunohistochemistry test, the other part was placed in liquid nitrogen for RT-PCR.(6) Routine pathological examination and immunohistochemistry test.In routine pathological examination, HE stain was used. The expressions of HIF-1αand VEGF in liver tissue were tested, with the assessment criteria referring the methods of Birner, et al in the former and referring the methods of Park, et al in the latter.(7) Test of content of HIF- 1αmRNA, VEGFmRNA in liver tissueAfter resuscitation of 0.1g frozen liver tissue, total RNA was extracted according to Trizol kit, and total RNA concentration was determined by use of ultraviolet spectrophotometer. After ultraviolet spectrophotometry, RNA was adjusted as 1μg/μl in various groups, to be used as template in making RT-PCR reaction solution. Two step method was used for RT-PCR,β-actin as inner reference. IOD of HIF-1α/β-actin, VEGF/β-actin was used to express the relative expression levels of HIF-1αmRNA, VEGFmRNA.2. Clinical research(1) 22 healthy volunteers, 15 males and 7 females, with ages of 25 to 70 years. 25 patients(17 males, 8 females, age 35～75 years)were diagnosed as liver cirrhotics by clinical, biochemical and imageological examinations. According to the Child-Pugh grading method for the severity of liver diseases, the cirrhotics were graded:11 cirrhotics as Child-Pugh A level; 10 as Child-Pugh B level ;4 as Child-Pugh C level.(2) 22 healthy volunteers and 25 cirrhotics were examined by multi-slice spiral CT liver perfusion and scan. Before scanning ultravist(300mgI/ml) was injected into the ulnar vein using high pressure injector, at rate of 4.0～4.5ml/s and dose of 1ml/kg. The rapid dynamic scanning was started 5 seconds after contrast medium bolus injection with the other technique details being the same as in the animal experiment.(3) Analysis and calculation of perfusion parametersAs in animal experiment,(4) Liver volume determinationPlain scan was made in all the cirrhotics. The obtained CT images and slice-by-slice manual mouse description were used to measure the area of various slices, each area was multiplied by its slice thickness, and thereby the liver volume was made out by accumulation.3. Statistical analysisSPSS 11.5 statistical software was used for statistical analysis of the results. For measurement data of normal distribution and homoscedasticity, t test or pairing t test was used in two-group comparison. Single-factor analysis of variance was used in multi-group comparison and then SNK-q test was used for multiple comparison between various means. The obtained data are expressed by mean±standard deviation. Pearson method was used for inter-parameter correlation analysis, rank correlation for graded data.Results1. The changes of rabbit liver functionThere appeared obvious changes in rabbit serum AST and ALT parameters between pre- and post- drug injection, that is, AST, ALT evidently rose after injection, reaching the peak at the end of the fourth week and then slowly lowered. The changes of ALT are more marked than that of AST with the peak value of 191U/L. The change of ALB was not evident, with slight lowering at the ends of the 10th to 12th week.2. Results of CT perfusion and scanStatistical significance was not found in comparison between CT perfusion parameters of the various control groups(P＞0.05). In the experimental groups, from the end of the 2nd week to the end of the 12th week, hepatic arterial perfusion(HAP)showed gradual rising. Between the control rabbits and the test rabbits, there was no statistical significance at the end of the 2nd week(P＞0.05), but there was statistical significance between the control rabbits and the test rabbits at the ends of the 4th, 6th, 8th ,10th and 12th weeks(P＜0.05). Hepatic portal perfusion(HPP) and hepatic blood flow(HBF) showed lowering tendency. Between the control rabbits and the test rabbits, HPP was no statistical significance at the end of the 2nd week(P＞0.05), but there was statistical significance between the control rabbits and the test rabbits at the ends of the 4th, 6th, 8th,10th and 12th weeks(P＜0.05). HBF was no statistical significance at the end of the 2nd, 4th, 6th week(P＞0.05), but there was statistical significance between the control rabbits and the test rabbits at the ends of the 8th ,10th and 12th weeks(P＜0.05). Mean transit time(MTT) was no statistical significance in comparison between the control and experimental rabbit groups at the end of the 2nd,4th weeks(P＞0.05), whereas there was statistical significance at the ends of 6th,8th, 10th, 12th weeks(P＜0.05).3. Pathological resultsAccording to the pathological results obtained with HE staining, the pathological changes of liver disease in experimental group were graded as hepatitis stage, hepatic fibrosis stage, and early cirrhosis stage: 2-4th weeks was the period of hepatitis; 6-8th weeks was that of hepatic fibrosis; 10-12th weeks was that of early hepatic cirrhosis.4. Expression of HIF- 1αin liver tissue HIF-1αis located mainly in the cytoplasm of liver cells, has expression also in part of cell nucleus. Positive expression is brown-yellow granules. In early hepatic cirrhosis (the ends of 10-12th weeks), HIF-1αhad positive expression in 10 rabbits; in hepatic fibrosis (the ends of 6-8th weeks), HIF-1αhad positive expression in 9 rabbits, and in hepatitis period (the ends of 2-4th weeks), had positive expression in four rabbits.5. Expression of VEGF in liver tissueThe positive site of immunal histochemistry stain of VEGF is in the cytoplasma, nucleus of liver cells and in the vascular endothelial cell. Positive expression is brown-yellow granules. Positive expression of VEGF was found in 10 rabbits in early hepatic cirrhosis (the ends of 10-12th weeks); in 9 rabbits in hepatic fibrosis (ends of 6-8th weeks); and in 4 rabbits in hepatitis (ends of 2-4th weeks).6. Results of HIF- 1αmRNA testThe expression of HIF-1αmRNA showed successive rising level from the end of 2nd week to the end of 12th week.7. Results of VEGFmRNA testThe expression of VEGFmRNA showed successive rising level from the end of 2nd week to the end of 12th week.8. Correlation between CT perfusion parameters and the expression of HIF-1α, VEGFMTT, HAP showed positive correlation with HIF-1αand VEGF(r = 0.532, P＜0.0 ; r = 0.381, P＜0.055); HPP, HBF had negative correlation with HIF-1αand VEGF(r = -0.721, P＜0.05; r = -0.687,P＜0.05).9. The difference of HAP,HPP,HBF parameters had statistical significance between patients of various degrees of cirrhosis.In Child-Pugh A cirrhotics, HAP had no statistical significance in comparison with the control group(P＞0.05); in Child B-Pugh and C cirrhotics, HAP was significantly higher than controls(P＜0.05). But the increase of HAP did not show significantly difference between Child-Pugh A and Child-Pugh B patients(P＞0.05); there was statistical significance between Child-Pugh B cirrhotics and Child-Pugh C cirrhotics(P＜0.05).HPP showed no statistical significance between Child-Pugh A cirrhotics and control group(P＞0.05); HPP is of statistical significance when comparing Child-Pugh B, Child-Pugh C cirrhotics at the one hand with control groups and Child-Pugh A cirrhotics at the other hand(P＜0.05) in each comparison; HPP is of statistical significance between Child-Pugh B and Child-Pugh C cirrhotics(P＜0.05).HBF is of statistical significance(P＜0.05) among the control group, Child-Pugh A, B, C cirrhotics, HBF was of statistical significanc(P＜0.05) between Child-Pugh B and Child-Pugh A cirrhotics. Child-Pugh C cirrhotics showed significant different HBF with Child-Pugh A cirrhotics(P＜0.05), but showed no statistical significance(P＞0.05) with Child-Pugh B cirrhotics.MTT showed statistical significance(P＜0.05) only between Child-Pugh C cirrhotics and the control group, but showed no statistical significance(P＞0.05) with other degrees of cirrhotics.10. There was no statistical significance in comparison of CT perfusion parameters between the left and right liver lobes in cirrhotics.11.Correlation between cirrhosis-caused liver volume change and perfusion parameters.HAP, MTT and LV was of moderate negative correlation(r = -0.454, P＜0.05; r= -0.656, P＜0.05); HPP, HBF MTT showed high positive correlation with LV(r = 0.745, P＜0.05; r = 0.668, P＜0.05).Conclusion1. The model establishment of early hepatic cirrhosis in rabbits provided theoretical basis for clinical study of early hepatic cirrhosis.2. Perfusion parameter of liver CT can be used for quantitative determination of liver blood flow and for reflection of hemodynamic changes in early hepatic cirrhosis. Changes of perfusion of liver blood flow provides objective base for clinical diagnosis and treatment of early cirrhosis and for estimation of prognosis.3. The expression of HIF-1αand VEGF is successively increasing in the order of hepatitis, hepatic fibrosis, early hepatic cirrhosis and reaches highest positive level in early hepatic cirrhosis. 4. The expression levels of HIF-1αmRNA and VEGFmRNA are successively increasing from the end of the 2nd week to the end of the 12th week.5. The correlation between perfusion parameter and expression of HIF-1αand VEGF in early hepatic cirrhosis is as follows: moderate positive correlation of HAP with the expressions of HIF-1αand VEGF(r = 0.532,P＜0.05); high negative correlation of HPP and HBF in the one hand with the expressions of HIF-1αand VEGF in the other hand(r =-0.721,r = -0.687,P＜0.05,P＜0.05);no statistical significance in comparing MTT with the expressions of HIF-1αand VEGF(r = 0.231,P＞0.05). Blood perfusion parameters in MSCT perfusion image can reflect degree of hepatic cirrhosis.6. There was no statistical significance in comparison of CT perfusion parameters between the left and right liver lobes in cirrhotics(P＞0.05).7. Correlation between liver volume change and liver blood flow perfusion parameters in liver cirrhosis: moderate negative correlation between HAP and LV(r = -0.454,P＜0.05); positive correlation in comparing LV with HPP and HBF(r = 0.745,P＜0.05; r = 0.668,P＜0.05); no statistical significance in correlation between MTT and LV(r = 0.256, P＞0.05).8. Correlation between liver volume change and liver blood flow perfusion parameters in liver cirrhosis: moderate negative correlation between HAP, MTT and LV(r = -0.454, P＜0.05; r =- 0.656, P＜0.05); positive correlation in comparing LV with HPP and HBF (r = 0.745, P＜0.05; r = 0.668, P＜0.05).