To analyze and compare the efficacy of ADNEX model, Copenhagen index (CPH-I), and the combination of them in the diagnosis of benign and malignant ovarian tumors, and to identify the independent factors for the prediction of ovarian malignancy.

A total of 186 patients with 190 ovarian tumors who underwent surgical treatment at the Department of Gynaecology of Changzhou No.2 People's Hospital from January 2018 to December 2019 were enrolled in this retrospective analysis. ADNEX model analysis and CPH-I calculation were performed to estimate the possibility of malignancy. Using the pathologic results as the gold standard, ROC curve analysis was performed to calculate the sensitivity, specificity, and Youden index of ADNEX model, CPH-I, and the combination of both. The correlation between the variables and ovarian malignancies was also calculated. The independent sample t test was used to compare the difference in age and the maximum diameter of tumor between the benign and malignant groups, the rank sum test was used to compare the difference in carbohydrate antigen 125 (CA125), human epididymis protein 4 (HE4), and the largest diameter of the solid part of the tumor between groups, and the Fisher exact test and correction χ² test were used to compare the number of compartments, number of papillae, acoustic shadow, and ascites between groups. Single factor analysis and multivariate binary Logistic regression analysis were carried out to identify the predictive factors for ovarian malignancy.

The sensitivities of ADNEX model, CPH-I, and the combination of both in the differential diagnosis of benign and malignant ovarian tumors were 65%, 65%, and 74%, the specificities were 99%, 94%, and 94%, and the areas under the ROC curves were 0.82, 0.80, and 0.84, respectively. Single factor analysis showed that the differences of age [(41.99±14.75) years vs (54.74±13.39) years, t=-4.64, P<0.001], CA125 [(18.39 (12.11~36.98) U/ml vs 124.05 (41.27~1121.35) U/ml), Z=-5.90, P<0.001], HE4 [40.20 (32.93~50.55) pmol/Lvs 78.65 (48.38~639.80) pmol/L, Z=-6.32, P<0.001], maximum diameter of tumor [(6.89±3.34) cm vs (8.91±3.91) cm, t=-3.11, P=0.002], the largest diameter of the solid part of tumor [0.00 (0.00~0.00) cm vs 4.75 (2.50~7.70) cm, Z=-9.80, P<0.001], more than three papillae (1.92% vs 26.47%, P<0.001), and ascites (0 vs 32.35%, χ²=47.80, P<0.001) were statistically significant between the benign and malignant groups. Multivariate analysis showed that age, HE4, the largest diameter of the solid part of tumor, and more than three papillae were independent risk factors for malignant ovarian tumors (OR [odds ratio]=1.059, P=0.002; OR=1.003, P=0.004; OR=1.533, P<0.001; OR=60.930, P<0.001).

Both ADNEX model and CPH-I have important clinical value in the differentiation of benign and malignant ovarian tumors, and the combination of them has a higher sensitivity. Age, HE4, maximum diameter of the solid part of tumor, and number of papillae more than 3 are independent risk factors for malignant ovarian tumors.