To validate the EFI, a clinical tool that predicts the possibility of pregnancy in Filipino women with laparoscopically-diagnosed endometriosis-associated infertility.

Retrospective cohort

St. Luke's Medical Center

One hundred twenty-two infertile women with laparoscopically-diagnosed endometriosis.

Subjects were scored using the Endometriosis Fertility Index (EFI). Pregnancy rates were determined separately. Discriminative capacity of the EFI was assessed by Receiver Operator Characteristic (ROC) analysis. Sensitivity, specificity, positive and negative predictive values of the test at different thresholds were determined. Correlation of EFI scores and pregnancy rates was determined by logistic regression analysis.

EFI scores, pregnancy rates, ROC curve, Area under the curve (AUC), sensitivity, specificity, positive and negative predictive values, Logistic fit of pregnancy VS. EFI scores.

Twenty-two pregnancies occurred during the 12-month period with 13 pregnancies occurring within the first six months. Over-all pregnancy rate was 18.3%. The ROC curve straddled the line and showed an AUC of 67%. The sensitivity and specificity of the EFI for each threshold showed increasing sensitivity with lower scores and higher specificity with higher scores. Positive predictive values (PPV) were low, but Negative predictive values (NPVs) were high. Logistic regression analysis showed a good correlation between EFI scores and pregnancy (P < 0.05).

The individual scores correlate well with the chances of pregnancy and the negative predictive value of the test is high. This index is a newly validated, simple clinical tool that can predict pregnancy in infertile women with endometriosis. The discrimination Gapability of the EFI needs further analysis.