Prediction of Premalignant and Malignant Endometrial Polyps by Clinical and Hysteroscopic Features.

STUDY OBJECTIVE

To investigate whether hysteroscopic features can contribute to the diagnosis of malignancy in endometrial polyps.

DESIGN

Retrospective review.

SETTING

Obstetrics and gynecology department.

PATIENTS

All women who underwent operative hysteroscopy for the removal of endometrial polyps between January 2012 and September 2017. Their medical records were reviewed, and information on medical, surgical, and obstetric history and hysteroscopic findings (including the number, size, and vascular appearance of the polyps) were abstracted.

INTERVENTIONS

Operative hysteroscopy with resection or biopsy of endometrial polyps.

MEASUREMENTS AND MAIN RESULTS

Five hundred fifty-six women were included in the study. Their mean age was 55.4 ± 12.4 years, and 322 (57.9%) were menopausal. Endometrial carcinoma was found in 26 (4.7%) cases, whereas endometrial hyperplasia was found in 5 (0.9%) cases. Endometrial carcinoma or hyperplasia was significantly associated with patients' age, menopausal status, increased polyp vascularity on hysteroscopy, and the presence of 3 or more polyps on hysteroscopy (p <.01 for all comparisons). However, the size of the largest polyp was not associated with endometrial carcinoma or hyperplasia. On logistic regression analysis, only increased polyp vascularity was associated with endometrial carcinoma or hyperplasia (odds ratio =13.5; 95% confidence interval, 5.6-32.3; p <.001). The sensitivity, specificity, positive predictive value, and negative predictive value of polyp vascularity for the diagnosis of polyps of nonbenign pathology were 51.6%, 94.3%, 34.8%, and 97.1%, respectively.

CONCLUSION

Hysteroscopic findings of increased vascularity of endometrial polyps and numerous endometrial polyps may suggest the diagnosis of malignant polyps, in addition to demographic parameters such as age and menopausal status.