2D PC-MRI And T1mapping In Predicting Portal Hyperperfusion And Liver Regeneration After ALPPS

PartⅠApplication of 2D PC-MRI in Monitoring Portal Blood Flow and Regeneration of Remaining Liver Before and After ALPPSObjective To investigate the role of 2D PC-MRI in monitoring the changes of portal blood flow and liver regeneration before and after ALPPS.Method Thirty-eight patients diagnosed with HCC and undergoing ALPPS surgery who were admitted to the First Affiliated Hospital of Guangxi Medical University from January 2018 to December 2021 were included.The 2D PC-MRI factors were used to calculate the mean blood flow(MF_PV and MF_LPV),maximum blood flow(Max F_PV and Max F_LPV),mean flow velocity(MV_PV and MV_LPV),maximum flow velocity of the portal vein and left portal vein before and after surgery(Max V_PV and Max V_LPV)and（i PVF and i LPVF）.At the same time,the 6 indexes were compared with ultrasound LPV blood flow indexes（us-MF and us-i LPVF）;Finally,the correlation between the above 6 blood flow indexes and the remaining liver regeneration volume（FLR）was analyzed in our study.Results The PV flow increased from an average of 0.69±0.19L/min to0.75±0.20L/min before and after stage-I,but the difference between them was not statistically significant（t=0.6967,P=0.4907）.We compared the following 6factors,average flow（t=5.099,P=0.0081）,peak flow（t=3.834,P=0.0254）,average velocity（t=3.579,P=0.00321）,peak velocity（t=3.385,P=0.0384）and pressure gradient（Z=7.886,P=0.0194）in different periods,there were statistically significant differences between the means of all six factors.MF increased from 0.449±0.21 L/min to 0.580±0.15 L/min,and the difference was statistically significant（P=0.0017）;i LPVF increased from 121.70±68.71ml/min/100g to 154.0±48.17ml/min/100g,the difference was statistically significant（t=2.322,P=0.0231）.The changes of LPV blood flow measured by MRI and ultrasound were consistent,and the parameter i LPVF was positively correlated with US-ILPVF（r=0.5719,P=0.0100）.A negative association was observed among all the 4 flow indexes between MRI indexes and FLR,including i PVF（R=-0.4886,P=0.0022）,i LPVF（R=-0.4301,P=0.0079）,post-i PVF（r=-0.0079）0.3457,P=0.00342)and post-i LPVF（r=-0.5302,P=0.0009）.A negative association was observed between FLR volume and US-ILPVF（r=-0.4982,P=0.0029）Conclusions（1）Phase contrast magnetic resonance imaging（PC-MRI）is a noninvasive approach that can quantify flow-related parameters such as blood flow on the patients of ALPPS.i LPVF exhibit a significant correlation with US parameters US-ILPVF.（2）Both PV and LPV increased after stage I,and the level of the 2 factors remained the same or decreased after stage II;there were significant differences in MRI between i PVF and i LPVF before and after stage I;（3）The index ILPVF was negatively correlated with residual liver hyperplasia.Part II Application of 2D PC-MRI in Predicting Portal Hyperperfusion Syndrome After ALPPSObjective To evaluate the application value of 2D PC-MRI technology in portal hyperperfusion syndrome after ALPPS.Method Thirty-eight patients diagnosed with HCC and underwent ALPPS surgery who were admitted to the First Affiliated Hospital of Guangxi Medical University from August 2018 to December 2021 were included in the study.They were divided into Portal Hyperperfusion Syndrome（PHPS）group（n=9）and Non-Portal Hyperperfusion Syndrome（Non-PHPS）group（n=29）,according to clinical data.The blood flow parameters before and after operation were measured by MRI,ultrasound and actual measurement,and the correlations between the three measurement parameters were compared.Subgroup analysis was used to investigate changes in blood flow parameters at each stage.Univariate analysis was performed on all the above blood flow indexes and pathological liver fibrosis indexes to obtain the index with the highest diagnostic efficiency.Cox and logistic regression were used for multivariate analysis to find out the independent influencing factors of liver hyperplasia.Results There were statistically significant differences among the following factors before and after surgery,such as LPV blood flow parameters(MF_LPV,Max F_LPv,MV_PV,Max V_LPV,LPV pressure gradient and i LPVF)（P values were 0.0017,0.0028,0.0005,0.0039,0.0006 and 0.0022,respectively）.Compared with the Non-PHPS group,the i LPVF,postoperative mean left portal vein flow(MV_PV),left portal vein flow change(DMF_LPV)and left portal flow change rate(DMF_LPV%)were significantly different between the PHPS group and the Non-PHPS group（P values were 0.0215,0.0425,0.0268,0.0115,respectively）.The intra-group analysis of PHPS showed that there were significant differences in i LPVF,US-LPVF,US-ILPVF and intraoperative PVP before and after portal vein ligation（P=0.0022,0.0003,P<0.0001 and 0.0124,respectively）.In the Non-PHPS group,there was no significant difference in all blood flow indexes before and after portal vein ligation.Subgroup analysis showed that there was no significant difference between the I-PHPS group and the I-Non-PHPS group before stage I surgery.The differences among the flow i LPVF,d MF_LPV and d MF_LPV%were statistically significant（P values were 0.0468,0.0371 and 0.0124,respectively）.There was statistically significant difference in all MR indices(MF_LPV、Max F_LPV、ILPVF、DMF_LPV、DMF_LPV%、Ishak-fibrosis)between the PHPS and Non-PHPS groups（P=0.0215,0.0340,0.0425,0.0268,0.0115 and 0.0307,respectively）.The index DMF_LPV had the highest diagnostic efficacy（AUC 0.808,95%CI 0.662-0.955）.DMF_LPV was an independent factor for PHPS（COX P=0.007,RR=17.424,95%CI 2.169-139.946）.The cutoff value was 0.195 L/min.Conclusions MRI blood flow parameters can monitor the changes of blood flow after ALPPS,and the measurement results are related to the results of ultrasound blood flow measurement.PHPS is related to many parameters such as MFLPV,Max F_LPV,ILPVF,DMF_LPV,DMF_LPV%,Ishak fibrosis grade,among which DMF_LPV is an independent influencing factor of PHPS.Part III 2D PC-MRI and T1mapping Factors Predict Liver Hyperplasia After ALPPSObjective Application value of 2D PC-MRI combined with T1mapping sequence in predicting liver dysplasia after ALPPS.Methods The clinical data of patients who were diagnosed with HCC and underwent ALPPS surgery between August 2018 and December 2021 were collected,including 58 patients who completed T1mapping sequence scanning,and 35 patients who completed both T1mapping and PC-flow sequence at the same time.Among the 58 patients who completed the T1mapping scan,45patients were in the good liver hyperplasia group（proliferation rate≥30%）and13 patients were in the poor dysplasia group（proliferation rate<30%）.All patients underwent Gd-EOB-DTPA-enhanced scanning,and T1_pre,T1_5min,T1_10min,T1_20min and△T1 changes were recorded before surgery.The total functional liver volume and functional residual liver volume was calculated;the correlation between the T1mapping and clinical factors and the liver hyperplasia rate（FLR%）after stage I surgery was analyzed.Combined with the T1mapping parameters for the diagnosis of liver dysplasia,the functional volume and the MR portal blood flow and clinical parameters determined in the first part of this study were used for diagnostic tests,and a ROC curve of each parameter was obtained.Results Correlation analysis showed that the residual liver hyperplasia rate（FLR%）was negatively correlated with Ishak liver fibrosis score（r=-0.3893,P=0.0043）,negatively correlated with inflammation score（r=-0.3437,P=0.0126）,and PLT The levels were positively correlated（r=0.3489,P=0.0073）.There was a negative correlation between T1pre and FLR%（r=-0.2682,P=0.0418）,but there were no statistical correlation between FLR%and T1_5min,T1_10min,T1_20min,△T1and△T1%（P<0.05）.There was a moderate negative correlation between FLR%and functional residual liver volume RLV×△T1%（r=-0.5234,P<0.0001）,but no statistical correlation with functional total liver volume TLV×△T1%（r=0.3179,P=0.0369）.The differences of RLV×△T1%,△T1%,PLT,and Ishak inflammation scores between the better hyperplasia group and the less hyperplasia group were statistically significant（statistical values:t=3.395,P=0.0013;Z=1.995,P=0.0456;t=2.315,P=0.0243 and Z=1.523,P=0.0458）.Cox regression showed that the remaining functional liver volume RLV×△T1%was an independent factor affecting liver hyperplasia（RR=11.649,95%CI,2.575,52.688）.The cut off value is 293.62.Logistic regression results showed that both RLV×△T1%and PLT were independent predictors of liver dysplasia,and the cut-off value of PLT was 215.45.The combination of RLV×△T1%,ILPVF and GRWR had the highest diagnostic efficiency（AUC=0.924,95%）CI 0.839,1.000,P=0.000).Conclusion（1）Among all the factors,the factors liver fibrosis,inflammation degree and RLV×△T1%,had correlation with FLR%.（2）RLV×△T1%and PLT were independent influencing factors for predicting liver dysplasia;（3）Gd-EOB-DTPA-enhanced T1mapping,functional volume parameters and PC-MRI blood flow parameters can be used as image-functional markers for preoperative prediction of future liver hyperplasia.