| Objective: To investigate the ultrasonic differential diagnosis criteria of benign and malignant cervical lesions as well as the manifestations of malignant cervical lesions of different stages and different pathological types through applications of color Doppler ultrasound and ultrasound elastography examination techniques.Method: A total of 146 patients with cervical diseases and treated in the Fourth Hospital of Hebei Medical University from Jan. 2014 to Dec. 2014 were selected as subjects and divided into: 1. the benign group consisted of 32 cases including 17 cases of cervical myoma and 15 cases of cervical polyp with an age range of 29-64 years and an average age of 47.63±7.88 years; 2. the malignant group consisted of 74 cases including 31 postmenopausal cases, 35 premenopausal cases and 8 cases with unconfirmed last menstrual period with an age range of 19-71 years and an average age of 48.73±11.04 years. Pathological types: there were 57 cases of cervical squamous cell carcinoma, 14 cases of cervical adenocarcinoma, 2 cases of cervical adenosquamous carcinoma, and 1 case of cervical clear-cell carcinoma. Clinical stages: there were cervical cancer patients of stage I 31 cases, stage II 27 cases, stage III 8 cases and stage IV 4 cases. The benign and malignant cervical lesion groups were confirmed by pathological results. 3. Forty contemporary normal individuals were screened out to form the control group with an age range of 24-65 years and an average age of 46.56±7.23 years. The group of normal cervix was confirmed by gynecological examinations and thinprep cytologic test(TCT) results.Color Doppler ultrasound examination: the blood flow distribution pattern of the cervical lesions was observed, and the section with the most abundant blood supply was selected to measure and record the blood flow parameters of the relatively larger vessels. Ultrasound elastography examination: the size and hardness of the lesions were observed, and the strain ratios of cervical lesion and normal cervical region(SRcervix), cervical lesion and anterior uterine wall(SRanterior-wall), and cervical lesion and posterior uterine wall(SRposterior-wall)were calculated.Statistical analysis: the measurement data of the strain ratios of the benign and the malignant lesions and the normal cervical region, as well as the strain ratios of the benign and the malignant lesions and the uterine wall underwent t-test, and statistical differences could be confirmed when P<0.05; the measurement data of the blood flow parameters of the normal cervix group, the benign group and the malignant group underwent analysis of variance, and statistical differences could be confirmed when P<0.05; the enumeration data underwent Wilcoxon Rank-Sum test and statistical differences could be confirmed when P<0.05. For those with statistical differences, the maximum value by adding the sensitivity value to the specificity value was determined as the dividing standard of optimal diagnosis, and the cut-off value was calculated. The pathological diagnosis was set as the gold standard for comparison, and the sensibility, specificity, positive predictive value(PPV), negative predictive value(NPV) and accuracy rate were evaluated. The receiver-operating curves(ROC) were applied to calculate the area under curve(AUC) and evaluate the accuracy of diagnosis.Results:1 In blood flow classification, most malignant cervical lesions were ≥ level 2 with relatively abundant blood stream while most benign cervical lesions were ≤ level 1 without abundant blood stream, and statistical difference existed with P<0.05.2 The comparisons between the blood flow parameters of benign and malignant cervical lesions and normal cervix presented statistical difference with P<0.05, but the comparison between the blood flow parameters of the benign and the malignant cervical lesions presented no statistical difference with P>0.05. In subjects of childbearing age, the blood flow parameters of the benign and the malignant cervical lesions were statistically different with P<0.05.3 The comparisons between the postmenopausal and premenopausal blood flow parameters of malignant cervical lesions presented statistical difference with P<0.05.4 There was a statistical difference of strain ratios among cervical myoma, cervical polyp and the normal cervical region with P<0.05.5 There was a statistical difference of SRcervix among benign and malignant cervical lesions with P<0.05.6 There was a statistical difference of SRanterior-wall among benign and malignant cervical lesions with P<0.05.7 There was a statistical difference of SRposterior-wall among benign and malignant cervical lesions with P<0.05.8 Taking the maximum value by adding the sensitivity value to the specificity value as the dividing standard of optimal diagnosis, the calculated cut-off value was blood flow classification level 2, SRcervix = 3.09, SRanterior-wall = 3.13 and SRposterior-wall = 3.11.9 Lesions with blood flow classification ≥level 2 are malignant and blood flow classification ≤ level 1 are benign. The sensitivity was 70.2%, specificity was 71.8%, PPV was 85.2%, NPV was 51.1%, and accuracy rate was 70.7%.10 Lesions with SRcervix < 3.09 are benign and with SRcervix ≥3.09 are malignant. The sensitivity was 82.8%, specificity was 87.5%, PPV was 93.5%, NPV was 70%, and accuracy rate was 84.3%.11 Lesions with SRanterior-wall < 3.13 are benign and with SRanterior-wall ≥ 3.13 are malignant. The sensitivity was 72.9%, specificity was 81.2%, PPV was 90%, NPV was 56.5%, and accuracy rate was 75.4%.12 Lesions with SRposterior-wall < 3.11 are benign and with SRposterior-wall ≥3.11 are malignant. The sensitivity was 74.3%, specificity was 78.1%, PPV was 88.7%, NPV was 56.8%, and accuracy rate was 75.4%.13 The comparisons of the SRcervix of cervical malignant lesions in different clinical stages were not statistically different with P>0.05.14 The comparisons of the SRcervix of malignant cervical lesions in different main pathologies were not statistically different with P>0.05.15 ROC analyses were performed for blood flow classification,SRcervix,SRanterior-wall and SRposterior-wall, and the results indicated middle degrees of accuracy and the AUCs were 0.71, 0.86,0.73 and 0.74. The results indicated SRcervix highest diagnostic accuracy.Conclusion: The color Doppler ultrasound and ultrasound elastography have a certain value in differential diagnosis of benign and malignant cervical lesions.Compared with color Doppler ultrasound, the ultrasound elastography possesses a higher accuracy in the differential diagnosis of benign and malignant cervical lesions.Ultrasound elastography is no value in different clinical stage, pathological classification in the differential diagnosis meaningless of cervical malignant lesions. |
PDF Full Text Request