Associations of IVF singleton birthweight and gestation with clinical treatment and laboratory factors: a multicentre cohort study

Catherine M Castillo, Edward D Johnstone, Greg Horne, Deborah A Falconer, Stephen A Troup, Rachel Cutting, Vinay Sharma, Daniel R Brison, Stephen A Roberts

Research output: Contribution to journalArticlepeer-review


STUDY QUESTION: Do IVF treatment and laboratory factors affect singleton birthweight (BW)?

SUMMARY ANSWER: BWs of IVF-conceived singleton babies are increasing with time, but we cannot identify the specific treatment factors responsible.

WHAT IS KNOWN ALREADY: IVF-conceived singleton babies from fresh transfers have slightly lower BW than those conceived naturally, whilst those from frozen embryo transfer (FET) cycles are heavier and comparable to naturally conceived offspring. Our recent studies have shown that BW varies significantly between different IVF centres, and in a single centre, is also increasing with time, without a corresponding change in BWs of naturally conceived infants. Although it is likely that factors in the IVF treatment cycle, such as hormonal stimulation or embryo laboratory culture conditions, are associated with BW differences, our previous study designs were not able to confirm this.

STUDY DESIGN, SIZE, DURATION: Data relating to BW outcomes, IVF treatment and laboratory parameters were collated from pre-existing electronic records in five participating centres for all singleton babies conceived between August 2007 and December 2014.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Seven thousand, five hundred and eighty-eight births, 6207 from fresh and 1381 from FET. Infants with severe congenital abnormalities were excluded. The primary outcome of gestation-adjusted BW and secondary outcomes of unadjusted BW and gestation were analysed using multivariable regression models with robust standard errors to allow for the correlation between infants with the same mother. The models tested treatment factors allowing for confounding by centre, time and patient characteristics. A similar matched analysis of a subgroup of 379 sibling pairs was also performed.

MAIN RESULTS AND THE ROLE OF CHANCE: No significant associations of birth outcomes with IVF embryo culture parameters were seen independent of clinic or time, including embryo culture medium, incubator type or oxygen level, although small differences cannot be ruled out. We did not detect any significant differences associated with hormonal stimulation in fresh cycles or hormonal synchronization in FET cycles. Gestation-adjusted BW increased by 13.4 (95% CI 0.6-26.1) g per year over the period of the study, and babies born following FET were 92 (95% CI 57-128) g heavier on average than those from the fresh transfer.

LIMITATIONS, REASONS FOR CAUTION: Although no specific relationships have been identified independent of clinic and time, the confidence intervals remain large and do not exclude clinically relevant effect sizes. As this is an observational study, residual confounding may still be present.

WIDER IMPLICATIONS OF THE FINDINGS: This study demonstrates the potential for large scale analysis of routine data to address critical questions concerning the long-term implications of IVF treatment, in accordance with the Developmental Origins of Health and Disease hypothesis. However, much larger studies, at a national scale with sufficiently detailed data, are required to identify the treatment parameters associated with differences in BW or other relevant outcomes.

STUDY FUNDING/COMPETING INTEREST(S): This work was supported by the EU FP7 project grant, EpiHealthNet (FP7-PEOPLE-2012-ITN-317146). No competing interests were identified.


Original languageEnglish
Pages (from-to)2860-2870
Number of pages11
JournalHuman reproduction (Oxford, England)
Issue number12
Publication statusPublished - 1 Dec 2020


Dive into the research topics of 'Associations of IVF singleton birthweight and gestation with clinical treatment and laboratory factors: a multicentre cohort study'. Together they form a unique fingerprint.

Cite this