Withholding antibiotics does not reduce clinical pregnancy outcomes of natural cycle frozen embryo transfers

In a retrospective study of 250 natural cycle frozen embryo transfers, withholding doxycycline did not reduce pregnancy rates compared with the antibiotic administration arm.

VOLUME 115, ISSUE 5, P1225-1231


Isabel Beshar, M.Phil. , Jasmyn K. Johal, M.D., M.Sc., Brindha Saravanabavanandhan, M.D., Amin A. Milki, M.D.



To assess the impact of withholding doxycycline on the success rate of natural cycle frozen embryo transfers (NC-FET).


Retrospective cohort study.


Single academic institution.


Women undergoing 250 NC-FET with euploid blastocysts performed by a single provider.


One hundred and twenty-five NC-FET cycles performed after January 2019 without antibiotic administration compared with 125 NC-FET cycles before January 2019 with doxycycline administration.

Main Outcome Measure(s)

Primary outcome: live birth (LB) or ongoing pregnancy rate (OPR, defined as pregnancies ≥13 weeks); secondary outcomes included positive β-human chorionic gonadotropin (β-hCG) level and clinical pregnancy rate (CPR, defined as the presence of fetal cardiac activity on ultrasound).


Each group of women comprised 125 NC-FET during the study period of March 2017 to March 2020. The women’s mean age was 36.3 years and mean body mass index was 24 kg/m2. Between the two groups, the baseline characteristics were similar, including age, body mass index, race, smoking status, parity, endometrial thickness, Society of Assisted Reproductive Technology diagnosis, and number of prior failed transfers. Comparing NC-FET with doxycycline administration versus without, we found no statistically significant difference in LB-OPR (64% vs. 62.6%), positive β-hCG (72.8% vs. 74.0%), or CPR (68% vs. 65.9%). After controlling for all variables in a logistic regression, doxycycline still had no effect on LB-OPR.


In this analysis of similar patients undergoing NC-FET by a single provider, withholding doxycycline does not reduce success rates. Given the risks of antibiotics, our findings support withholding their use in NC-FET.