Association of culture medium with growth, weight and cardiovascular development of IVF children at the age of 9 years

STUDY QUESTION: Is embryo culture media used during an IVF/ICSI treatment associated with differences in growth, body composition and cardiovascular development as determined in 9-year-old singleton IVF children? SUMMARY ANSWER: The choice of in vitro culture medium for human embryos is associated with differences in body weight, BMI, truncal adiposity, waist circumference and waist/hip ratio at the age of 9, while no significant differences were observed in cardiovascular development. WHAT IS KNOWN ALREADY: Children born after IVF/ICSI have an increased risk of low birthweight, which is correlated with a higher risk of cardiovascular diseases. Some studies show that IVF children exhibit a significantly higher systolic and diastolic blood pressure and higher fasting glucose levels compared to naturally conceived children. After alternating assignment to G1 (TM) Version 3 (Vitrolife) or K-SICM (Cook) embryo culture media, birthweight of the resulting children was significantly higher in the Vitrolife group and they remained heavier during the first 2 years of life. STUDY DESIGN, SIZE, DURATION: In this observational cohort study (MEDIUM-KIDS), parents of singletons from a previous study were approached for further follow-up after the ninth birthday of their child. The singletons were born after fresh embryo transfer of cleavage stage embryos resulting from an IVF/ICSI treatment performed between July 2003 and December 2006 in our clinic, when two different culture media were used alternately: either G1 (TM) Version 3 (Vitrolife) or K-SICM (Cook). Follow-up measurements were performed between March 2014 and December 2016. PARTICIPANT/MATERIALS, SETTINGS, METHODS: Parents were invited to attend our clinic with their child for a single visit lasting similar to 2.5 h. Two experienced clinicians performed all measurements as part of the MEDIUM-KIDS study in a standardized way. Height and weight of the child was measured using calibrated scales, 4-point skinfold thickness measurements were measured in triplicate and waist and hip circumference were measured using a tape measure. The following cardiovascular parameters were measured in a standardized way: blood pressure, heart rate and endothelial function by skin laser-Doppler with iontophoresis using vasodilatory drugs. Cortisol and cortisone concentrations in hair were measured. A blood sample was taken after an overnight fast for insulin, glucose, TSH and lipid analysis. Blood samples of the IVF children were compared with a non-IVF control group. Differences between culture medium groups were analysed by Student's t-test and effects of confounders were analysed using multivariable regression analysis. MAIN RESULTS AND THE ROLE OF CHANCE: Of the 294 eligible children (168 Vitrolife and 126 Cook), 136 children (75 Vitrolife and 61 Cook) participated in the study. Baseline characteristics of the participating children from the Vitrolife and Cook group were similar. Birthweight was higher in the Vitrolife group, in keeping with the full cohort. After correction for confounders, the difference in weight and BMI attributable to culture medium was 1.58 kg (95% CI: 0.01-3.14) and 0.84 kg/m(2) (95% CI: 0.02-1.67), respectively, with the Vitrolife children being heavier. Height and height corrected for age and gender (SDS scores) were similar in both groups. Furthermore, waist circumference was significantly higher in the Vitrolife group with a corrected difference of 3.21 cm (95% CI: 0.60-5.81) leading to a 0.03 increase (95% CI: 0.01-0.05) in waist/hip ratio. Subscapular skinfolds combined with suprailiacal skinfolds (defined as truncal adiposity), was also significantly higher in Vitrolife children (adjusted difference 3.44 cm [95% CI: 0.27-6.62]). Both systolic (adj. beta 0.364 [95% CI: -2.129 to 2.856],) and diastolic (adj. beta 0.275 [95% CI: -2.105 to 2.654]) blood pressures (mmHg) were comparable for the two groups. After an overnight fast, cholesterol, glucose, insulin, low and high-density lipoprotein, triglycerides and TSH were normal and similar in the two groups. Endothelial function in the microcirculation was compared by using maximum perfusion units corrected for the baseline value as a measure for vasodilatory capacity. There were no significant differences between the two groups. Cortisol and cortisone concentration in hair samples were comparable. LIMITATIONS, REASONS FOR CAUTION: A limitation of the original study was its pseudo-randomized design. This and the dwindling enthusiasm of families for participation (47.7% after 9 years) prevent us from drawing robust causal conclusions from the observed association. Nevertheless, to date this is oldest cohort of IVF/ICSI children where culture medium was allocated alternatingly and used in a blinded setting, to be studied. We believe that our participants are representative for the full cohort. The current number of participants was sufficient to rule out differences as little as 3 mmHg in systolic and diastolic blood pressures. WIDER IMPLICATIONS OF THE FINDINGS: This study underlines the importance of structured follow-up of IVF/ICSI children to further elucidate possible long-term health effects. Health professionals and culture medium manufacturers should be aware that small changes in culture conditions and culture medium composition for the early embryo can have long-term health effects. The similar cardiovascular results for the two groups are reassuring but the children may still be too young to detect differences in cardiovascular development. Prolonged follow-up and structured investigations up until adulthood are necessary to gain more insight and reassurance in the cardiovascular development of IVF offspring, although long-term follow-up will become more complicated by confounding life-style and environmental factors possibly influencing development. STUDY FUNDING/COMPETING INTEREST(S): The study was financially supported by the March of Dimes (Grant number #6-FY13-153). The sponsor of the study had no role in study design, data collection, data analysis, data interpretation or writing of the report. The authors have no conflicts of interest to declare.