Can contrast-enhanced ultrasound differentiate cervical lesions?

Objective

Assessment of whether contrast-enhanced ultrasound (CEUS) can be used to differentiate diverse cervical lesions.

Methods

A retrospective analysis of ultrasonographic reports was conducted for patients with different cervical lesions, including 18 cases of chronic cervicitis, 28 cases of cervical intraepithelial neoplasia grade I (CIN1), 46 cases of cervical intraepithelial neoplasia grade II (CIN2), 100 cases of cervical intraepithelial neoplasia grade III (CIN3), 7 cases of carcinoma in situ (CIS), and 26 cases of cervical cancer (CC). Timing began with the contrast agent injection via the elbow vein, recorded separately when the myometrium and cervix began to be enhanced. The intensity of enhancement of the cervix was observed and recorded as hyper-enhancement, iso-enhancement, or hypo-enhancement with respect to the myometrium. The rate of regression of the cervix enhancement was also analyzed and classified as fast, synchronous, or slow relative to the myometrium. Quantitative data were analyzed using either one-way ANOVA or the Kruskal-Wallis H test, while qualitative data were analyzed using Fisher’s exact test. Significant differences were further analyzed using post-hoc tests, logistic regression models, and ROC curves.

Results

Menopausal status affected longitudinal, anteroposterior, and transverse diameters of the cervix, in addition to cervical volume. Moreover, the time when the myometrium and cervix began to image was also different between the menopause group and the pre-menopause group (P < 0.05), whereas there was no significant variance of pulsatility index (PI), resistance index (RI), enhancement intensity of the cervix, or rate of the cervix fading. The study population was grouped according to their menopausal status, and then we discovered that in the pre-menopausal group, the cervical anteroposterior diameter (APD) differed in the six cervical lesion groups (P = 0.006), especially in CIN1 & CIN2, CIN1 & CIN3, and CIN1 & CC (P = 0.014, 0.045, 0.021, respectively). The enhancement start time (EST) of the cervix differed among the cervical lesion groups (P = 0.02). In particular, there was a significant difference in CC & CIN2 and CC & CIN3 (P = 0.04, 0.03, respectively). The subjects of the study were divided into two groups: those with cervical cancer and those with precancerous lesions. The discrepancy of cervical EST was still significant along with the sensitivity of 0.62, the specificity of 0.76 using a cutoff of 14.895 s, and the accuracy was 0.74.

Conclusion

The cervical EST can serve as an indicator of malignancy, and CEUS can be a complementary tool for cervical cancer screening. However, it should be noted that qualitative CEUS analysis alone was unable to differentiate among various precancerous lesions.