Multivariate analysis of factors predictive of successful live births in in vitro fertilization (IVF) suggests strategies to improve IVF outcome.

Abstract

Our purpose was (1) to identify characteristics correlated with pregnancy outcome, (2) to use these characteristics to predict in vitro fertilization (IVF) outcome, and (3) to develop strategies that might improve IVF success. Maternal age, cause for IVF, donor insemination, rank of attempt, serum estradiol and luteinizing hormone levels on the day of human chorionic gonadotropin administration, flexible vs rigid catheter, number of embryos transferred of each morphologic type, and cell number were analyzed by logistic regression. Variables positively correlated with success are as follows: (1) for pregnancy, endometriosis and 2-, 3-, and 4-cell good and 4-cell excellent embryos; (2) for live births, 2-, 3-, and 4-cell good and 4-cell excellent embryos and donor insemination; and (3) for multiple births, 2- and 4-cell good and 4-cell excellent embryos. Maternal age was negatively correlated with live births. Embryos derived from IVF have different potentials for implantation, live births, and multiple births. Transferring one additional good-quality embryo for each 5 years of incremental increase in maternal age is predicated to improve live birth rates without increasing multiple births.