Prediction models in women with postmenopausal bleeding: a systematic review

Area of Science:

• Gynecology
• Oncology
• Medical Diagnostics

Background:

• Postmenopausal bleeding is a key symptom associated with an increased risk of endometrial cancer.
• Accurate risk stratification is crucial for timely diagnosis and management of endometrial carcinoma.

Purpose of the Study:

• To review and assess existing prediction models for endometrial cancer in women experiencing postmenopausal bleeding.
• To identify key predictor variables and evaluate the performance of these models.

Main Methods:

• Systematic literature search to identify prognostic studies on endometrial cancer prediction.
• Quality assessment, development phase evaluation, and performance analysis of identified models.
• Identification of significant predictor variables from the reviewed studies.

Main Results:

• Nine prognostic studies were included, with varying quality and development phases.
• Models incorporating power Doppler demonstrated the best internal validation performance.
• No models underwent external validation or impact analysis, limiting generalizability.

Conclusions:

• Distinguishing between low-risk and high-risk patients with postmenopausal bleeding is a critical first step.
• Further (invasive) testing is recommended for patients identified as high-risk for endometrial carcinoma.
• Accessible and validated prediction models are needed for improved clinical decision-making.