Eur J Obstet Gynecol Reprod Biol. 2026 Sep 6;327:115418. doi: 10.1016/j.ejogrb.2026.115418. Online ahead of print.

ABSTRACT

RESEARCH QUESTION: What is the prognostic value of first complete IVF cycle outcomes on live birth rates in the second cycle, and does this differ by age group in contemporary IVF practice?

DESIGN: Retrospective cohort study using Swiss national IVF registry data. Included were 1,860 women who completed two complete consecutive IVF cycles between 2018 and 2023. An historical cohort comprising 2,162 women whose first and second complete IVF cycles were both initiated between 2014 and 2017 was analysed using identical methods to compare the contemporary findings with an earlier treatment period. Outcomes of the first complete cycle (no pregnancy, biochemical pregnancy, clinical miscarriage, live birth) were used to predict live birth in the second complete cycle, analyzed with multivariable logistic regression adjusted for age and ovarian response and grouped by age (<38 vs. ≥ 38 years).

RESULTS: Second-cycle live birth showed a graded association: 31.8 % (no pregnancy), 36.6 % (biochemical), 43.1 % (miscarriage), 42.7 % (live birth; p < 0.001). Clinical pregnancy/live birth in first cycle increased odds vs. no pregnancy (aOR 1.41, 95 %CI 1.07-1.86, p = 0.010), with a stronger effect in women < 38 years; age was the predominant prognostic factor in those ≥ 38 years. Comparison with the historical 2014-2017 cohort demonstrated stronger prognostic effects.

CONCLUSIONS: First complete IVF cycle outcomes stratify prognosis for subsequent success, particularly in women < 38 years, while age overtakes as predominant predictor in older women. Biochemical pregnancy was not confirmed as an independent predictor in the main multivariable model. These findings support cautious, individualised counselling after the first complete IVF cycle in contemporary practice, while underscoring that prognostic estimates are conditional on completing a second cycle and should be validated in larger, independent cohorts.

PMID:42710464 | DOI:10.1016/j.ejogrb.2026.115418