Managing Pregnancies of Unknown Location With the M4 Prediction Model or the NICE Algorithm: A Randomised Controlled...

Area of Science:

• Reproductive Medicine
• Clinical Diagnostics
• Women's Health

Background:

• Pregnancy of unknown location (PUL) presents a diagnostic challenge in early pregnancy management.
• Accurate diagnosis is crucial to differentiate between ectopic pregnancy (EP) and non-ectopic pregnancies.
• Existing algorithms, such as the NICE algorithm, guide the management of women with PUL.

Purpose of the Study:

• To evaluate the diagnostic performance of the M4 prediction model against the NICE algorithm for women with PUL.
• To assess the clinical utility and outcomes associated with each management strategy.

Main Methods:

• A superiority trial design was employed, focusing on specificity for non-ectopic pregnancy for the M4 model.
• 595 women with PUL were randomized to either the M4 model or the NICE algorithm.
• Diagnostic performance was assessed using two serum human chorionic gonadotropin (hCG) levels, categorizing women into high or low risk for EP.

Main Results:

• Sensitivity for EP was 79% for M4 versus 85% for NICE (p=0.1496).
• Specificity for non-ectopic pregnancies was 67% for M4 versus 74% for NICE (p=0.0003).
• Clinical outcomes were similar between the M4 and NICE algorithm groups.

Conclusions:

• The M4 prediction model is non-inferior to the NICE algorithm in terms of sensitivity for detecting ectopic pregnancy.
• The NICE algorithm exhibited superior specificity in identifying non-ectopic pregnancies.
• No significant differences in clinical outcomes were observed between the management strategies.