| The morbidity and mortality of rectal cancer is very high,and the accurate diagnosis is very important for the optimal choice of treatments and improving prognosis.To evaluate the TNM staging of malignant rectal tumors accurately before surgery is very significant for the selection of surgical methods.The patients with T1-2 staging can directly accept surgery.However,patients with T3-4 staging and lymphatic metastasis theoretically need to accept radiotherapy or chemotherapy firstly to downstage the lesions to go on surgery.So,before surgery,it is recommended that T3-T4 to receive neoadjuvant chemoradiotherapy.MRI has a very good imaging effect for soft tissue.It not only shows the lesions clearly,but also has functional imaging methods.IVIM is a non-related voxel movement,which can provide the diffusion condition of tissues and can be expressed in the form of variables,including D,D*and f.The malignant lesions with advanced growth due to the rapid proliferation of tumor cells,result in limited diffusion of water molecules.Low advanced malignancy or benign lesions have less or no limitation of water molecules diffusion.So,IVIM can be used as a noninvasive quantitative way to evaluate the malignant degree,pathological stage,lymph node invasion and pathological changes of cancers.The natural state of rectum is being of atresia,which means rectal lesions,normal intestinal tissue and luminal contents are close to each other,making the lesion boundary difficult to distinguish.The mix of tissue lead to the result that the IVIM measurement in the delineation of lesions is not accurate,in which way affecting the stability of the numerical results of the IVIM.In addition,the gas in the rectal lumen would produce significant magnetic sensitive artifacts in the IVIM sequence,causing serious disruption to the inspection and measurement results.At the same time,when choosing the IVIM ROI,different methods will also affect the final results.All factors mentioned above would lead to variability of IVIM measurement parameters.Angiogenesis plays an important role in the growth,invasion and metastasis of tumors,and affects the biological behavior and prognosis of tumors.Quantitative assessment of tumor angiogenesis includes MVD and VEGF.Determination of tumor angiogenesis cannot only predict the probability of tumor recurrence and metastasis,but also provide evidence for active adjuvant chemotherapy or radiotherapy.However,both MVD and VEGF are pathological indicators,which cannot provide useful clinical information before operations.It is of great significance to find the preoperative imaging parameters of MVD and VEGF to improve the treatment and the prognosis.Therefore,the Purposes of this study include:1.Compare the repeatability of IVIM parameters ether in natural state of the rectum or in filling state with rectum filled with ultrasound coupling agent,combining two kinds of ROI selection methods,including whole measurement and small solid sample,and find the optimal rectal condition to accept IVIM exam.2.To explore the application of IVIM in the determination of T staging of rectal cancer.3.To explore the difference of IVIM parameters between colorectal cancer patients with lymph node metastasis and non-metastasis groups.4.Of all the IVIM parameters,D*and F were related to tissue perfusion.Therefore,we explored the correlation between F value and D*and MVD and VEGF,in order to obtain a preoperative noninvasive imaging indexed to improve the treatment and prognosis of rectal cancers.Material and Methods60 patients(39 males and 21 females,average age 57.3 years old)from 2015.10 to 2016.10 who accepted rectal cancers surgery in our hospital were included in this study.A 3.0T MR(Discovery 750,GE Medical Systems)with an 8-channel-phased-array was used.T2-weighted sequence in the coronal,sagittal and axial HR-T2WI,T1WI were performed to define the tumor location utilizing the following parameters.Subsequently,axial IVIM was acquired using a single-shot echo-planar imaging(EPI)pulse sequence with b factors of 0,20,50,100,150,200,400,600,800,1000,1500s/mm2.Used 100ml disposable enema tube and ultrasonic coupling agent to fill the rectums before the second exam of IVIM scanning.The injection depth of the tube and the volume of the coupling agent was decided by the location of the tumor(known by the rectal examination,colonoscopy or conventional sagittal T2.The depth of the tube,taking the anal as starting spot,was 5cm for low and mid rectal tumors,and 10 for the high rectal tumors.The volumes of coupling agent were 60-70、70-80、80-100ml for the low,mid and high rectal tumors respectively.The IVIM imaging data were postprocessed on a workstation 4.5(GE Medical Systems).D,D*and f values were obtained by multiple ADC measurement tools.For the ROI,two kinds of ROI selection were usd,including[a]’whole measurement’and[b]’Solid tumour samples’.For the whole measurement method,a free hand ROI was drawn along the border of the tumour,containing the largest available tumour area.For the second method,ROI was calculated from a sample of three round/oval-shaped ROIs that were placed within the most solid tumour part(as identified on T2WI)of three independent tumour-containing slices.The specimens were sectioned for MVD and VEGF measurements.The sections were analyzed by immunohistochemistry.(1)MVD:using mouse anti human CD34 staining,in accordance with the standard of Weider,any brown stained by CD34 antibody isolated endothelial cells or cell clusters,as long as separate and adjacent capillaries,tumor cells or other connective tissue,was taken as a microvascular structure;as long as is the branch structure can also be used as a microvessel count.(2)VEGF:using VEGF monoclonal antibody staining,in according to the Volm criteria,the cells with cytoplasm or cell membrane were brownish yellow as VEGF positive cells.SPSS 22 and MedCalc were used.(1)used the intraclass correlation coefficient(ICC)method to analyze the stability of IVIM parameters under different rectal conditions and ROI.(2)After determining the most stable rectal status and ROI selecting methods,used the t test or t’ test to analyze the differences of IVIM parameters(D,D*,f)between T1+T2 and T3+T4,T2 and T3 stages.(3)used the t test or t’ test to determine the difference of IVIM parameters(D,D*,f)between metastatic lymph nodes and non-metastatic lymph node lesions of colorectal cancers.Then,evaluate the diagnostic efficacy and significance of each parameter by sensitivity,specificity and ROC curve and area under curve(AUC).Youden index is used to determine the optimal threshold for IVIM parameters.(4)Correlation between D*and f and MVD and VEGF was determined using the Spearman rank correlation analysis.P<0.05 was as the criterion of statistical difference.Results1.In the comparison of the stability of IVIM parameters,the results are as follows:For D value,the ICC of natural state+round ROI,filling state+round ROI,natural state+whole measurement ROI and filling state+whole measurement ROI were 0.886、0.894、0.892、0.919,respectively.And the 95%CI were 0.778-0.946,0.794-0.950,0.789-0.949 and 0.843-0.962.For D*value,the ICC of natural state+round ROI,filling state+round ROI,natural state+ whole measurement ROI and filling state+ whole measurement ROI were 0.405,0.824,0.663,0.916 respectively.And the 95%CI were 0.156-0.719,0.657-0.917,0.345-0.841,0.836-0.960.For f value,the ICC of natural state+round ROI,filling state+round ROI,natural state+ whole measurement ROI and filling state+ whole measurement ROI were 0.682,0.684,0.808,0.835,respectively.And the 95%CI were 0.381-0.849,0.386-0.851,0.627-0.909,0.680-0.922.Therefore,the filling state+ whole measurement ROI had the beat stability of each parameter of IVIM,the ICC of D,D*and F were 0.919,0.916,and 0.835,respectively.2.For the T stage,the D and f of T1+T2 were both higher than T3+T4,while the D*was lower than T3+T4.The differences of D、D*and f had statistical significance(p<0.05).Single factor analysis of variance showed the D value had the best diagnostic efficacy.When the cutoff selected as 0.877×10-3,the AUC,specificity and sensitivity were 0.711,83.33%and 55.56%.For the f value,When the cutoff selected as 0.280,the AUC,specificity and sensitivity were 0.704,72.22%and 64.29%.For the D*value,When the cutoff selected as 8.02×10-3,the AUC,specificity and sensitivity were 0.631,88.89%and 47.62%.Multiple logistics analysis of variance showed the combination of all three IVIM parameters had the best diagnostic performance,with the AUC,specificity and sensitivity were 0.806,66.67%and 83.33%.3.For the T stage,the D of T2 was higher than T3,while the D*was lower than T3.The differences of D、D*had statistical significance(p<0.05),while the f value had no statistical differences between two groups.Single factor analysis of variance showed the D value had the best diagnostic efficacy.When the cutoff selected as 0.862×10-3,the AUC,specificity and sensitivity were 0.779,86.67%and 63.64%.For the D*value,When the cutoff selected as 8.02×10-3,the AUC,specificity and sensitivity were 0.657,86.67%and 48.48%.Multiple logistics analysis of variance showed the combination of all the two IVIM parameters had the best diagnostic performance,with the AUC,specificity and sensitivity were 0.897,86.67%and 81.82%.4.For the lymph node metastasis,the D value of without non-metastatic lymph node was higher than with metastatic lymph node and the differences had staistical significance(p<0.05).When the cutoff selected as 0.79×10-3,the AUC,sensitivity and specificity were 0.629,84.85%and 48.15%.5.The Spearman rank correlation coefficient between VEGF and D*was 0.50466,the p value was 0.0073,the Spearman rank correlation coefficient between VEGF and F was-0.39358,and the p value was 0.0422.There was statistical significance of correlation coefficient test.The Spearman rank correlation coefficient between MVD and D*was-0.03115,the p value was 0.8774,the correlation coefficient between VEGF and Spearman was0.11634,and the p value was 0.5634.There was no statistical significance of correlation coefficient test.Conclusions1.When the rectum was filled with ultrasound coupling agent,and use whole measurement method to get the ROI can get stable results of IVIM,which is very meaningful for the popularization and application in the clinic.2.IVIM can help in the diagnosis of T staging of rectal cancer,especially between T2 and T3,which is the difficulty of clinical diagnosis.3.To some extent,IVIM can be used to indicate the state of lymph node metastasis in rectal cancer,which is very significant for the selection and formulation of treatment.4.VEGF has a positive correlation with D*value,and has a negative correlation with the f value.So IVIM can be used as a noninvasive imaging index to reflex VEGF.SignificanceAfter bowel preparation,filling agent coupling parameters and used whole measurement ROI,we can obtain stable IVIM,and apply it to colorectal cancer in routine MR examination to evaluate the rectal cancers comprehensively,including T staging,lymph node metastasis,and in a certain extent,replace the VEGF to improve the treatment and prognosis before surgery.Innovations1.Explored the effect of different rectal states and ROI methods to the Repeatability of IVIM results,which was very important to make a standard scanning programme of IVIM.2.Explored the use of IVIM to diagnose the T stage quantitatively,which was helpful in the diagnosing of T2 and T3 stages.3.Explored the use of IVIM to judge if the patients with rectal cancers have had the lymph node metastasis or not,which was helpful in the determination of treatments.4.Explored the use of noninvasive imaging methods to prognose the results of postoperative pathology to give useful clinical information before surgery,which can have a more accurate and comprehensive evaluation of the disease.Defects1.The number of samples was not enough,and the samples of different T periods are not equal,which may lead to the bias of the results.2.Rectal filling coupling agent would lead to an extension of the overall inspection time,while some patients may reject the filling of agent.3.Rectal filling coupling agent led to the dilatation of the rectal cavity,which may influence the circumferential margin and the location of the lesion. |
PDF Full Text Request