A matched follow-up study design was used to test the hypothesis that pregnancy rates following assisted reproduction procedures do not differ between women with or without intramural or sub-serosal uterine leiomyomas. Women undergoing their first in-vitro fertilization (IVF)-embryo transfer or zygote intra-Fallopian transfer (ZIFT) cycle between January 1993 and June 1995 were included. Cases (women with leiomyomas) were matched 1:1 with the next consecutive control (women without leiomyomas) according to age, number of embryos transferred, embryo grade, and the route of embryo transfer (uterine or Fallopian). Assisted reproduction cycles were performed in an identical manner, independent of the presence or absence of uterine leiomyomas. The main outcomes measured were clinical pregnancy and delivery rates. A total of 182 cycles was evaluated. Of the 91 assisted reproduction cycles performed in the leiomyoma group, there were 34 clinical pregnancies (37%) and 30 deliveries (33%). Of the 91 assisted reproduction cycles in the control group, there were 48 clinical pregnancies (53%) and 44 deliveries (48%). The Mantel-Haenszel estimate of relative risk indicated that the presence of a uterine leiomyoma significantly reduced the chance for a clinical pregnancy or delivery. These findings suggest that leiomyomas are associated with a reduction in the efficacy of assisted reproduction cycles.