| Objective: At present,there are still deficiencies in conventional ultrasound for the diagnosis of atypical focal adenomyosis.There is no consensus on the diagnostic efficacy of Contrast-Enhanced ultrasound(CEUS).The purpose of this study was to evaluate the CEUS features of atypical focal adenomyosis and the diagnostic efficacy of CEUS in differentiating atypical focal adenomyosis from uterine leiomyoma.Methods: Patients with atypical adenomyosis or uterine leiomyoma who underwent routine ultrasound examination in our hospital during April 2021 to July 2022 were included in the research.Then analyze the CEUS features of atypical focal adenomyosis and establish a CEUS model of the differential diagnosis of uterine leiomyoma.This study is diagnostic.Part I: A total of 74 patients with atypical focal adenomyosis who underwent CEUS examination in our hospital during April 2021 to June 2022 were collected and analyzed,those were,patients with adenomyosis difficult to diagnose by conventional ultrasound,and 50 cases were chosen with normal uterus as control group.The qualitative characteristics of CEUS were evaluated in the lesion area of atypical focal adenomyosis group and the abnormal lesion area not detected by conventional ultrasound.Then chose the regions of interest respectively and draw the time intensity curve(TIC)to analyze the quantitative indicators and the difference of lesion time,and compared with the control group with normal uterine.Part II: A total of 117 patients with atypical adenomyosis and uterine leiomyoma who underwent routine ultrasound examination in our hospital during April 2021 to July 2022 were collected and analyzed,including 78 patients in the adenomyosis group and 39 patients in the uterine leiomyoma group confirmed by pathology.Qualitative analysis data of adenomyosis group and uterine leiomyoma and quantitative analysis data of TIC curve were gathered,including the difference of contrast agent entering the lesion.Firstly,the patients were randomly divided into training group and validation group.There were90 patients in the training group,including 60 patients with adenomyosis and 30 patients with uterine leiomyoma.There were 27 patients in the validation group,including 18 patients with adenomyosis and 9 patients with uterine leiomyoma.The Logistic risk prediction model was established by analyzing each index in the training group,and the score was assigned according to the weight of each index in the prediction model.The CEUS scoring system for the differential diagnosis of atypical focal adenomyosis and uterine leiomyoma was established,and its diagnostic efficacy was further verified in the validation group.Secondly,the indicators included in the decision tree model were screened by single factor and multivariate analysis,then the decision tree models of CEUS in the differential diagnosis of atypical focal adenomyosis and uterine leiomyoma were established.Ten-fold cross validation was used to verify its reliability.The area under the receiver operating characteristic(ROC)curve was used to evaluate the diagnostic efficacy of the two models.Results: Firstly,the contrast ultrasound findings of the atypical focal adenomyosis group were analyzed: the lesions of adenomyosis mostly showed short linear enhancement(40/74),uneven(59/74)advanced enhancement(39/74)low enhancement(39/74),and multiple advanced dissection(37/74).The differences in AT,TTP and PI between adenomyosis group and normal uterine control group were of statistically significance(P<0.05).In the adenomyosis group,there was no statistical significance in the quantitative indexes of the lesions detected by CEUS and the area in which the lesion was not found under the conventional ultrasound(P > 0.05).There was statistical significance of the time difference between adenomyosis group and normal uterus control group(P <0.05).In the correlation analysis of lesion time difference and other features,lesion time phase difference was related to lesion location,lesion enhancement intensity and contrast agent distribution(P <0.05).Secondly,the differential diagnosis between atypical focal adenomyosis and uterine leiomyoma was analyzed.First,build the Logistic risk prediction model and incorporate the indicators selected from the univariate analysis into the multi-factor analysis.The Logistic risk prediction model was established as follows:The boundary was not clear after angiography [OR =4.499,95% confidence interval(CI):1.238-16.352],lesion time phase difference≥9.5s(OR =8.651,95% CI: 2.478 ～ 30.204),the uneven distribution of contrast agent in lesions(OR = 4.502,95% CI: 1.262 ～ 16.060)and | Δ AT | acuity 1.05 s(OR = 4.481,95% CI: 1.277 ～ 15.730)(P < 0.05).The AUC for the diagnosis of adenomyosis using this model was 0.898(95%CI: 0.826 ～ 0.970).When the model was converted into a scoring system,the AUC for differential diagnosis in the training group was 0.885(95%CI: 0.808-0.963).In the validation group,the AUC for the diagnosis of adenomyosis by the scoring system was 0.840(95%CI: 0.649～1.000).After the univariate analysis,the results of multivariate analysis showed that the unclear lesion boundary(OR= 7.096,95%CI = 1.973-25.522),the short linear enhancement pattern(OR=4.009,95%CI= 1.520-10.576)and high lesion time phase difference(OR= 1.288,95%CI= 1.131-1.467)were the independent predictors of adenomyosis(P < 0.05).The area under ROC curve of the final established decision tree model was 0.914(95%CI: 0.859 ～ 0.970),and the sensitivity,specificity,PPV and NPV were 89.74%,87.18%,93.33% and 80.95%,respectively.The misjudgment rate of the model is(16.20±3.40)%,and the accuracy of the model is 88.89%.Conclusion: In most cases,the lesions of adenomyosis show short linear enhancement,the whole lesion shows uneven enhancement,and the boundary is fuzzy through CEUS.The lesion will show an increase in time difference in contrast mode,and the above characteristics can be used as the characteristics of adenomyosis ultrasound and as a diagnostic basis.CEUS may provide an early detection imaging scheme for early small adenomyosis lesions.The model established by CEUS indicators provides new diagnostic ideas and methods for clinical application,and further guides clinical precision treatments. |
PDF Full Text Request