Risk of malignancy on suspicion of polyps in menopausal women

Improved performances in gynaecological ultrasonography have enabled an increasing number of often asymptomatic endometrial polyps to be detected. Most of these polyps are removed surgically, as a precautionary measure, so as not to miss a case of endometrial cancer. Nonetheless, this management strategy is based solely on the sonographer's judgement and a number of these operations, which are probably of no benefit, could be avoided. In order to do so, risk factors for malignancy need to be identified. Objective: Estimate the prevalence of lesions in menopausal patients with a pre-operative diagnosis of endometrial polyp. Establish risk factors for malignancy. Study design: This is a single-centre retrospective study. Enrolment criteria were menopausal patients aged over 45 who had undergone hysteroscopic resection of a polyp. Pre-op diagnosis was made either by ultrasonography or diagnostic hysteroscopy. Malignant lesions included cancers and atypical hyperplasia. Benign lesions consisted of simple polyps, non-atypical simple hyperplasia and non-atypical complex hyperplasia. Risk factors studied were existing abnormal uterine bleeding, endometrial thickness, personal or first-degree family history of gynaecological cancer (breast, cervix, endometrium, ovary) and age on diagnosis. Results: 631 patients were enrolled of whom 30 presented a malignant disorder (4.75%); 579 patients (91.76%) presented a simple polyp, 11 a non-atypical simple hyperplasia (1.74%) and 11 a non-atypical complex hyperplasia (1.74%). On univariate analysis age alone proved to be statistically significant (OR 1.05; 95%Cl = [1.02-1.09] p < 0.01), with a threshold of 59 years of age on the ROC curve.On multivariate analysis, factors predictive of a malignant lesion were age (OR = 1.06; 95% CI [1.02-1.10]), existence of AUB (OR = 2.4; 95% CI [1.07-5.42]) and family history (OR = 2.88; 95%CI [1.08-7.67]). Neither the univariate nor multivariate model was able to demonstrate a statistically significant relationship with respect to endometrial thickness. The risk of malignancy was 12.3% in patients aged over 59 presenting AUB. For all other subgroups, the risk varied between 2.31 and 3.78%. Conclusion: The risk of a malignant lesion appears to be high (12%) in menopausal patients aged over 59 presenting an endometrial polyp detected when there is pre-existing AUB. In this situation, hysteroscopic resection of endometrial polyps should therefore be routinely proposed. For other patients, the risk of a malignant lesion is low but not insignificant, standing at about 3%. Each patient record should therefore be discussed on an individual case basis, taking into consideration the patient's pre-existing conditions, after providing clear and appropriate information. (C) 2017 Elsevier B.V. All rights reserved.