Decreased clinical pregnancy and live birth rates after short interval from delivery to subsequent assisted reproductive treatment cycle.

STUDY QUESTION: Does the interval from delivery to initiation of a subsequent ART treatment cycle impact clinical pregnancy or live birth rates? SUMMARY ANSWER: An interval from delivery to treatment start of < 6 months or >= 24 months is associated with decreased likelihood of clinical pregnancy and live birth. WHAT IS KNOWN ALREADY: Short interpregnancy intervals are associated with poor obstetric outcomes in the naturally conceiving population prompting birth spacing recommendations of 18-24 months from international organizations. Deferring a subsequent pregnancy attempt means a woman will age in the interval with an attendant decline in her fertility. STUDY DESIGN, SIZE, DURATION: Retrospective analysis of the Society for Assisted Reproductive Technology Clinical Outcome Reporting System (SARTCORS) cohort containing 61 686 ART cycles from 2004 to 2013. PARTICIPANTS/MATERIALS, SETTING, METHODS: The delivery-to-cycle interval (DCI) was calculated for patients from SARTCORS with a history of live birth from ART who returned to the same clinic for a first subsequent treatment cycle. Generalized linear models were fit to determine the risk of clinical pregnancy and live birth by DCI with subsequent adjustment for factors associated with outcomes of interest. Predicted probabilities of clinical pregnancy and live birth were generated from each model. MAIN RESULTS AND THE ROLE OF CHANCE: A DCI of < 6 months was associated with a 5.6% reduction in probability of clinical pregnancy (40.1 +/- 1.9 versus 45.7 +/- 0.6%, P = 0.009) and 6.8% reduction in live birth (31.6 +/- 1.7 versus 38.4 +/- 0.6%, P = 0.001) per cycle start compared to a DCI of 12 to < 18 months. A DCI of >= 24 months was associated with a 5.1% reduction in probability of clinical pregnancy (40.6 +/- 0.5 versus 45.7 +/- 0.6%, P < 0.001) and 5.7% reduction in live birth (32.7 +/- 0.5 versus 38.4 +/- 0.6%, P < 0.001) compared to 12 to < 18 months. LIMITATIONS, REASONS FOR CAUTION: The SART database is reliant upon self-report of many variables of interest including live birth. It remains unclear whether poorer outcomes are a result of residual confounding from factors inherent to the population with a very short or long DCI or the interval itself. WIDER IMPLICATIONS OF THE FINDINGS: Birth spacing recommendations for naturally conceiving populations may not be generally applicable to patients with a history of infertility. Patients planning ART treatment should wait a minimum of 6 months, but not more than 24 months, from a live birth for optimization of clinical pregnancy and live birth rates.