Background: Although several scoring systems have been published to evaluate the pregnancy rate after ICSI-IVF in infertile patients, none of them are applicable for patients with deep infiltrating endometriosis (DIE) nor can they evaluate the chances of pregnancy for individual patients. The aim of this study was to develop a nomogram based on an association of patients' characteristics to predict the clinical pregnancy rate in patients with endometriosis.
Methods: This prospective longitudinal study was conducted from January 2007 to June 2010. The nomogram was built from a training cohort of 94 consecutive patients (141 ICSI-IVF cycles) and tested on an independent validation cohort of 48 patients (83 ICSI-IVF cycles). DIE was confirmed in all participants.
Results: The pregnancy rate (per patient) in women with and without DIE was 58 and 83%, respectively (P = 0.03). Increased patient age (P = 0.04), serum anti-Mullerian hormone (AMH) level ≤ 1 ng/ml (P = 0.03) and increased number of ICSI-IVF cycles (P = 0.03) were associated with a decreased clinical pregnancy rate. The presence of DIE was the strongest determinant factor of the clinical pregnancy rate in our model [odds ratio = 0.26, 95% confidence interval (CI): 0.07-0.9 (P = 0.006)], which also included patient age, serum AMH level and number of attempts at ICSI-IVF. The nomogram showed an area under the curve (AUC) of 0.76 for the training cohort (95% CI: 0.7-0.8) and was well calibrated. The AUC for the validation cohort was 0.68 (95% CI: 0.6-0.75) and calibration was good.
Conclusions: Our nomogram provides realistic and precise information about ICSI-IVF success and can be used to guide couples and practitioners.