Objective: To determine if early serum hCG levels are predictive of implantation outcome in patients undergoing IVF-ET.
Design: Retrospective study of IVF cycles using receiver operator characteristic curve (ROC) analysis.
Setting: Tertiary-care, university hospital-affiliated IVF program.
Patients: Three hundred fifty-one conception cycles were studied.
Interventions: None.
Main outcome measure: Implantation failure, defined as chemical pregnancies, ectopic gestations, and first trimester abortions, or implantation success, defined as delivered singleton and multiple pregnancies, and second trimester abortions.
Results: For each post-ET day 14 to 20, mean hCG levels of the implantation success group were significantly greater than implantation failure outcomes (P < 0.0001). Using ROC curve analysis, hCG cutoff values for each post-ET day were calculated for optimal discrimination of implantation failure from implantation success cycles. A patient with an hCG measurement greater than the calculated cutoff value had a > or = 90% chance of having an implantation success after IVF-ET.
Conclusion: Discriminatory hCG cutoff values may be useful in predicting implantation outcome in IVF-ET cycles and may guide clinicians in identifying those pregnancies at risk for adverse outcomes and instituting more intensive surveillance in this population. This information also may be useful in providing counseling to IVF patients regarding pregnancy prognosis and result in cost savings.