The Diagnostic Value Of Dual-phase Imaging Of ¹⁸F-FDG PET/CT In Benign And Maligant Colorectal Lesions

Objective: To explore the diagnosis value in 86 patients with colorectal diseases being abnormal high metabolism image in ¹⁸F–FDG PET/CT via the early-imaging and delay-imaging.Methodes:1 Analyzing 86 patients with colorectal diseases being abnormal high metabolism image in¹⁸F–FDG PET/CT retrospectively.All the patients have been mixing PET and CT image together using computer tecnology,using egion of interest?ROI?technology measruing maximum standardized uptake value?SUV max?of colorectal lesions and taking the delay-imaging for the abnormal high metabolism lesions.The early-imaging SUVmax can be marked as SUVearly and the delay-imaging SUVmax can be marked as SUVdelayed.Then caculating the change of SUVmax?SUVchange rate?,SUVchange rate=?SUVdelayed – SUVearly?/ SUVearly x100%.2 Taking comparative analysis for the differences between the benign diseases and the malignant diseases of SUVearly,SUVdelayed and SUVchange rate,respectively.3 To find the best diagnosis cut-off point and caculate the sensitivity,accuracy,specificity,positive predictive value and negative predictive value in colorectal maligant lesions of ¹⁸F-FDG PET/CT respectively.4 To find the best method to differentiate malignant from benign colorectal diseases by comparing the sensitivity,accuracy and specificity of SUVearly,SUVdelayed and SUVchange rate respectively.5 To decide the PET/CT clinical diagnosis value in colorectal lesions by comparing the sensitivity,accuracy,specificity,positive predictive value and negative predictive value with endoscope and CT.6 All the data have been decided by pathological diagnostic results andfollow-up results?follow-up time>6months?.7 Statistical method: using SPSS statistical sofeware 19.0 for data analysis.The measurement data are presented as mean±standard deviation.The number of samples were compared with Levene test between group variance.If meet the homogeneity of variance,independent samples t test was used,if it does not meet the homogeneity of variance,using the approximate t test.The receiver operating characteristic?ROC?threshold curves are used for calculating the best differential diagnosis cut-off point of SUV in colorectal cancer.Caculating the sensitivity,specificity,accuracy,positive predictive value and negative predictive value of different check methods.Two different diagnosis results are compared using paired x2 test four data table?McNemar test?.Results:1 The SUVearly,SUVdelayed and SUVchange rate of the maligant disease were all significant higher than the benign one.2 The SUVdelayed were significant higher than the SUVearly in maligant disease.The maximum SUVearly was 45.0,the minimum SUVearly was 2.0,the mean SUVearly was 10.17±1.04;the maximum SUVdelayed was 50.0,the minimum SUVdelayed was 2.8,the mean SUVearly was 13.51±1.30.The difference had the statistic significance?t=5.050,P=0.000?.The SUVdelayed were significant lower than the SUVearly in benign disease.The maximum SUVearly was 13.0,the minimum SUVearly was 2.4,the mean SUVearly was 5.8±0.52;the maximum SUVdelayed was 11.0,the minimum SUVdelayed was 1.7,the mean SUVearly was 4.09±0.52.The difference had the statistic significance?t=2.951,P=0.007?.3 The best differential diagnosis cut-off point of SUVearly,SUVdelayed and SUV change rate 4.0,4.0 and-4% respectively.Took the SUVearly of more than 4.0 as positive diagnosis criteria,the sensitivity was 88.5%,specificity was 48%,accuracy was 76.7%,positive predictive value was80.6% and negative predictive value was 63.4%.Took the SUVdelayed ofmore than 4.0 as positive diagnosis criteria,the sensitivity was 96.7%,specificity was 72.0%,accuracy was 89.5%,positive predictive value was89.4% and negative predictive value was 90.0%.Took the SUVchange rate of more than-4% as positive diagnosis criteria,the sensitivity was 96.7%,specificity was 80%,accuracy was 91.9%,positive predictive value was92.2% and negative predictive value was 90.9%.4 The diagnosis value of CT in colorectal benign and maligant lesions:the sensitivity was 82.0%?50/61?,specificity was 68.0%?17/25?,accuracy was77.9%?67/86?,positive predictive value was 86.2%?50/58?and negative predictive value was 60.7%?17/28?.Conclusions:1 There were all significant difference of the SUVearly and SUVdelayed between the malignant disease and the benign one.The SUVdelayed were significant lower than the SUVearly in benign disease.The SUVdelayed were significant higher than the SUVearly in maligant disease.The SUVearly and SUVdelayed can be used for differentiating maligant from benign colorectal diseases.2 The best differential dagnosis cut-off point of SUVearly,SUVdelayed and SUVchange rate were 4.0?4.0?-4% respectively.Took them as positive diagnosis criteria,the sensitivity and specificity were the highese.3 SUVchange rate was significant higher than CT in the aspect of sensitivity and accuracy.SUVchange rate was significant superior to CT in differential diagnosis.4 The¹⁸F-FDG PET/CT scan is not fully substituted for endoscopic pathology,and the corresponding intestinal endoscopy should be performed for patients who are highly suspicious of intestinal malignancy.