Intent to treat analysis of in vitro fertilization and preimplantation genetic screening versus expectant management in patients with recurrent pregnancy loss

Abstract STUDY QUESTION In an intent to treat analysis, are clinical outcomes improved in recurrent pregnancy loss (RPL) patients undergoing IVF and preimplantation genetic screening (PGS) compared with patients who are expectantly managed (EM)? SUMMARY ANSWER Among all attempts at PGS or EM among RPL patients, clinical outcomes including pregnancy rate, live birth (LB) rate and clinical miscarriage (CM) rate were similar. WHAT IS KNOWN ALREADY The standard of care for management of patients with RPL is EM. Due to the prevalence of aneuploidy in CM, PGS has been proposed as an alternate strategy for reducing CM rates and improving LB rates. STUDY DESIGN, SIZE, DURATION Retrospective cohort study of 300 RPL patients treated between 2009 and 2014. PARTICIPANTS/MATERIALS, SETTING, METHODS Among two academic fertility centers, 112 RPL patients desired PGS and 188 patients chose EM. Main outcomes measured were pregnancy rate and LB per attempt and CM rate per pregnancy. One attempt was defined as an IVF cycle followed by a fresh embryo transfer or a frozen embryo transfer (PGS group) and 6 months trying to conceive (EM group). MAIN RESULTS AND THE ROLE OF CHANCE In the IVF group, 168 retrievals were performed and 38 cycles canceled their planned PGS. Cycles in which PGS was intended but cancelled had a significantly lower LB rate (15 versus 36%, P = 0.01) and higher CM rate (50 versus 14%, P < 0.01) compared with cycles that completed PGS despite similar maternal ages. Of the 130 completed PGS cycles, 74% (n = 96) yielded at least one euploid embryo. Clinical pregnancy rate per euploid embryo transfer was 72% and LB rate per euploid embryo transfer was 57%. Among all attempts at PGS or EM, clinical outcomes were similar. Median time to pregnancy was 6.5 months in the PGS group and 3.0 months in the EM group. LIMITATIONS, REASONS FOR CAUTION The largest limitation is the retrospective study design, in which patients who elected for IVF/PGS may have had different clinical prognoses than patients who elected for expectant management. In addition, the definition of one attempt at conception for PGS and EM groups was different between the groups and can introduce potential confounders. For example, it was not confirmed that patients in the EM group were trying to conceive for each month of the 6-month period. WIDER IMPLICATIONS OF THE FINDING Success rates with PGS are limited by the high incidence of cycles that intend but cancel PGS or cycles that do not reach tra