Skip to main content
PMC home page
Human Reproduction (Oxford, England) logoLink to Human Reproduction (Oxford, England)
. 2016 May 10;31(7):1621–1628. doi: 10.1093/humrep/dew106

Infertility treatment and children's longitudinal growth between birth and 3 years of age

EH Yeung 1,*, R Sundaram 1, EM Bell 2,3, C Druschel 2,4, C Kus 5, Y Xie 1, GM Buck Louis 1
PMCID: PMC4901884  PMID: 27165624

Abstract

STUDY QUESTION

Does early childhood growth from birth through to 3 years of age differ by mode of conception?

SUMMARY ANSWER

Findings suggest early childhood growth was comparable for children irrespective of infertility treatment, but twins conceived with ovulation induction with or without intrauterine insemination (OI/IUI) were slightly smaller than twins conceived without treatment.

WHAT IS KNOWN ALREADY

Although studies have found that babies conceived with infertility treatment are born lighter and earlier than infants conceived without treatment, little research especially for non-assisted reproductive technology (ART) treatments has focused on their continued growth during early childhood.

STUDY DESIGN, SIZE, DURATION

Upstate KIDS recruited infants born (2008–2010) to resident upstate New York mothers. Infants were sampled based on birth certificate indication of infertility treatment; specifically, for every singleton conceived by infertility treatment, three singletons without infertility treatment were recruited and matched on region of birth. All multiple births irrespective of treatment were also recruited. Children were prospectively followed, returning questionnaires every 4–6 months until 3 years of age. In total, 3905 singletons, 1129 sets of multiples (96% of whom were twins) enrolled into the study. Analyses included 3440 (88%) singletons (969 conceived with treatment; specifically, 433 with ART and 535 with OI/IUI) and 991 (88%) sets of multiples (439 conceived with treatment; specifically 233 with ART and 206 with OI/IUI) with growth data available.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Mothers reported infertility treatment use at baseline and children's height and weight from pediatric visits. Self-reported use of ART was previously verified by linkage with the US Society for Assisted Reproductive Technology Clinic Outcome Reporting System (SART CORS) database. Mixed linear models with cubic splines accounting for age and age–gender interactions were used to estimate mean differences in growth from birth to 3 years by infertility treatment status and adjusting for maternal age, race, education, private insurance, smoking status during pregnancy, maternal pre-pregnancy and paternal body mass indices (BMI).

MAIN RESULTS AND THE ROLE OF CHANCE

Compared with singletons conceived without treatment (n = 2471), singletons conceived by infertility treatment (433 by assisted reproductive technologies (ART), 535 by OI/IUI and 1 unknown specific type) did not differ in growth. Compared with twins not conceived with treatment (n = 1076), twins conceived with OI/IUI (n = 368) weighed slightly less over follow-up (122 g). They were also proportionally smaller for their length (−0.17 weight-for-length z-score units). No differences in mean size over the 3 years were observed for twins conceived by ART, though some evidence of rapid weight gain from birth to 4 months (adjusted OR 1.08; 95% CI: 1.00–1.16) suggestive of catch up growth was observed.

LIMITATIONS, REASONS FOR CAUTION

Participants from upstate New York may not be representative of US infants. Although accounted for in statistical analysis, attrition during follow-up may have limited power to detect small differences.

WIDER IMPLICATIONS OF THE FINDINGS

This study is the first to prospectively track the growth of children conceived with and without infertility treatment in the USA, including a substantial number of twins. Our findings are similar to what was previously observed in the ART literature outside of the states.

STUDY FUNDING/COMPETING INTEREST(S)

Supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD; contracts #HHSN275201200005C, #HHSN267200700019C). Authors have no competing interests to declare.

TRIAL REGISTRATION NUMBER

Not applicable.

Keywords: assisted reproductive technologies, infertility treatment, growth, infant weight gain, ovulation induction

Introduction

With the rising global use of infertility treatments (Kupka et al., 2014; Sunderam et al., 2014), many children are now conceived with assisted reproductive technologies (ART) or fertility enhancing drugs such as clomiphene citrate (Schieve et al., 2009). Arising from the concerns observed in animal models of overgrowth syndromes (Feuer et al., 2013) and the higher risk of low birthweight and preterm delivery found in quantitative meta-analyses (McDonald et al., 2009, 2010), the growth of children conceived with infertility treatment remains controversial in light of its implications for children's health (Yeung and Druschel, 2013). Particularly, impaired fetal growth can precede rapid catch-up growth with potential long-term cardiovascular risk later in childhood (Kelishadi et al., 2015). Most recently, evidence has surfaced that catch-up may increase risk of atherosclerosis and fat deposition even at 3–6 years of age. It is hypothesized therefore that children conceived by infertility treatment may grow differently early in life (i.e. continue to be smaller and experience catch-up in growth) compared with children not conceived by treatment. Although unproven, possible underlying mechanisms include type of hormonal treatment, media culture (Kleijkers et al., 2014) or other aspects of treatment such as ICSI that may induce epigenetic changes during sensitive windows of embryonic development.

Although studies from outside of the U.S. have generally found no difference in growth by infertility treatment exposure (Yeung and Druschel, 2013), crucial data gaps remain. To date, most studies have only included singleton and not multiple-birth infants. Moreover, much of the research has only followed children conceived by in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) with little to no attention to the children conceived with infertility treatments not requiring ART. In addition, despite the genetic contribution of body mass index (BMI) and its association with risk of infertility, parental BMIs are inconsistently adjusted for in analyses investigating childhood growth. Lastly, risk of rapid weight gain during infancy or childhood has also been rarely examined.

Collectively, these findings and data gaps provided the impetus for us to design and implement the Upstate KIDS Study. Specifically using a population-based US sampling framework, we sought to delineate the relation between conception mode (use or not of infertility treatment) and children's growth from birth through to 3 years of age.

Materials and Methods

Study design and population

The Upstate KIDS Study is an ongoing birth cohort, which recruited 5034 women who delivered in New York State (excluding the 5 boroughs of New York City) between 2008 and 2010 (Buck Louis et al., 2014). Parents of all infants conceived by infertility treatment as indicated on birth certificates were invited to participate at approximately 4 months of age. Singleton infants not conceived with infertility treatment were recruited and frequency matched on region of birth to singleton infants whose conceptions were aided by treatment at a 3:1 ratio (Buck Louis et al., 2014). Response rates were slightly higher among infants conceived with infertility treatment (32%) than not (27%) but were otherwise similar by plurality (Buck Louis et al., 2014). Birthweight and gestational age of participants were also similar to non-participants after accounting for plurality and exposure status by birth certificate indication suggesting no bias in participation by these perinatal factors (data not shown). All multiples were also invited to participate. In total, 3905 singletons, 2132 twins and 145 higher-order multiples enrolled. The primary cohort comprised all singletons and one randomly selected multiple of each set.

Ethical approval

The New York State Department of Health (NYSDOH) and the University at Albany (State University of New York) Institutional Review Boards approved the Study and entered into a Reliance Agreement with the National Institutes of Health. All parents provided written informed consent prior to data collection.

Infertility treatment exposure

Using a standardized questionnaire, mothers were queried about specific infertility treatment at 4 months post-partum (Buck Louis et al., 2014). ART included any combination of IVF, ICSI, assisted hatching, frozen embryo transfer, zygote intrafallopian transfer (ZIFT), gamete intrafallopian transfer (GIFT) and/or donor eggs or embryos. Ovulation induction included any combination of oral or injectable medications with or without intrauterine insemination (OI/IUI). We previously compared maternal report of ART use against the US clinical gold standard by linking our cohort with the Society for Assisted Reproductive Technology—Clinical Outcome Reporting System (SART-CORS), and found sensitivity (93%) and specificity (99%) to be excellent (Buck Louis et al., 2015). For this analysis, ART was defined by maternal report, as no registry exists for non-ART infertility treatments. Where maternal self-report was missing (3%), birth certificate information was used to complete exposure status. Hence, no child in the cohort was missing infertility treatment exposure information.

Growth measures

In their invitation packets, mothers received child health journals that were designed to record children's health through to 3 years of age. Mothers were encouraged to take journals to children's regular medical visits for the recording of height and weight performed by healthcare providers. Standardized questionnaires were mailed to mothers every 4–6 months to capture growth information. In each questionnaire, mothers were given a grid to complete with multiple entries corresponding to each time that their children were measured. For example, a mother completing the 4-month questionnaire could report on weight and length at 6 weeks, 2 months etc. up to 4 entries. Mothers indicated the age and date of visit. Weight was elicited in grams/pounds and height in inches/centimeters and converted into SI units.

Birthweight was from birth certificates whereas birth length was from maternal report (Yeung et al., 2015). Observations two interquartile ranges (IQR) below the 25th percentile or 2 IQR above the 75th percentile (Tukey, 1977) for grouped age categories were removed as necessary (<1%). Observations of a child's length decreasing by 1 cm or more were also removed (<1%). To better standardize the responses and to evaluate weight relative to length, z-scores were calculated using the World Health Organization Child Growth Standards (de Onis et al., 2006).

In addition to analyses of growth trajectories, rapid infant weight gain was examined among singletons and twins. Infants were classified as having experienced rapid weight gain (yes/no) at 4, 9 and 12 months, respectively. Infant weight gain since birth was calculated at three different time points by taking the differences between weights predicted at 4, 9 and 12 months from our statistical models and birthweight. Predicted weights were used because children did not always visit the pediatrician at those exact time points. Weights before and after those time points (e.g. 3 and 5 months to predict 4 months) then informed the predictive models for estimating weight at those time points. We then calculated standard deviation scores (SDS) for each child at each time point by taking the difference between the child's weight gain and average weight gain among singletons in our sample, divided by their standard deviation (Breij et al., 2014). A child was classified as having experienced rapid weight gain if his/her SDS was above 0.5 for 4 or 9 months or above 0.67 for 12 months (Breij et al., 2014).

Statistical analysis

Twelve percent (n = 751) of children were missing information on weight after birth and were excluded from analyses. An additional 344 (5.5%) children were excluded due to missing data for length analyses. Missing data were more likely in the unexposed group than the exposed group with no material differences by plurality (Supplementary data, Fig. S1). Comparisons were made in baseline characteristics by exclusion and infertility status using χ2 or t-test of difference for categorical and continuous data, respectively. Baseline comparisons by infertility treatment exposure were made using similar statistical tests.

Mixed linear models with random coefficient cubic splines for age and age–gender interactions were used to estimate mean differences in growth from birth through to 3 years by infertility treatment exposure status. Furthermore, models included maternal-level random effects and nested infant-level random effects to account for the correlation between repeated measures and between siblings (i.e. for multiples) (Molenberghs, 2006). Analyses which included singletons and one child per family (i.e. the primary cohort) included an infant-level random effect. Longitudinal analyses were first limited to observations before 2 years of age, as pediatricians would begin measuring standing height rather than supine length, per se. However, the full longitudinal association with growth to 3 years was then repeated to evaluate whether observations changed. Logistic regression was used to estimate odds ratios (95% CI) for rapid infant weight gain and for risk of childhood overweight/obesity at 36 months. To account for the correlation within families, generalized estimating equations were used for the analysis of multiples.

All analyses were adjusted for maternal age (years), race (white versus non-white), education (college education or not), married or living as such (yes/no), private health insurance (yes/no), smoking during pregnancy (yes/no), maternal pre-pregnancy BMI (kg/m2) and paternal BMI (kg/m2). Both maternal and paternal BMI were retained in statistical models as they were not highly correlated (Pearson's correlation = 0.34) but both were significantly positively associated with growth, as measured by weight and length/height. Analyses were repeated stratifying on plurality. Sampling weights were derived based on vital records data on the births occurring during the recruitment period for infertility treatment, plurality and region of birth. They were applied to all analyses to account for the oversampling of infertility treatment and twins by design. Missing covariate data were imputed using the most common response for those with few missing (n < 10) and with multiple imputations for all others (i.e. 3.5% missing marital status, 6.7% missing paternal BMI). All analyses were conducted in SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).

Results

Exclusions based on missing growth data were associated with several baseline characteristics. Mothers of children included in the analysis (88%) were more likely to have higher socioeconomic status (i.e. college or more education, private insurance, married/cohabitating), were non-Hispanic White, of older age, less likely to smoke and were nulliparous compared with mothers of children without sufficient weight information (12%) (data not shown). After accounting for these sociodemographic differences, exclusions for missing data were not significantly associated with infertility treatment status (P = 0.30).

Infertility treatment was associated with multiple birth, non-Hispanic white race and higher socioeconomic status (Table I). Maternal pre-pregnancy BMI and paternal BMI were higher among those who underwent infertility treatment, particularly for parents who used OI/IUI. The average age at last reported weight and height was 3–4 months older for the infertility treatment group than the group not conceived by treatment. Birthweight and gestational age did not differ by treatment exposure groups after stratifying by plurality (data not shown).

Table I.

Baseline characteristics of the primary cohort by infertility treatment status, Upstate KIDS.

Characteristic No treatment (n = 3023)
 Infertility treatment (n = 1408)
 OI/IUI (n = 741)
 ART (n = 666)
n % n % n % n %
Singleton†a,b 2471 82 969 69 535 72 433 65
Whitea,b 2402 80 1333 95 667 90 566 85
College or morea,b 1358 44 1040 74 516 70 523 79
Private insurancea,b 2213 70 1336 95 693 94 643 97
Nulliparousa 1238 41 836 60 442 60 394 60
Pregnancy smokera 504 17 52 4 32 4 20 3
Marrieda 2545 87 1295 95 682 95 613 96
Maternal obesityb 767 25 389 28 251 34 138 21
Gestational Hypertensionb 309 10 178 13 90 12 88 13
Gestational Diabetesb 255 8 177 13 94 13 83 12
Mean SD Mean SD Mean SD Mean SD
Maternal age (years)a,b 29.3 5.73 34.0 5.1 32.5 4.6 35.8 5.1
Pre-pregnancy BMI (kg/m2)a,b 26.9 6.8 27.3 6.9 28.3 7.5 26.3 6.0
Paternal BMI (kg/m2)a,b 28.0 5.4 28.9 5.4 29.4 5.8 28.5 5.0
Birthweight (g)a,b 3236 662 3049 755 3088 768 3006 739
Gestational age (weeks)a,b 38.3 2.3 37.6 2.9 37.7 2.9 37.4 2.8
Age at last weight (months)a 19.4 12.5 23.4 12.3 23.4 12.5 23.6 12.1
Age at last length (months)a 18.0 12.4 21.5 12.3 21.3 12.2 21.3 12.2
Open in a new tab

†Among multiples, there were 3 sets of higher-order multiples (i.e. triplets and quadruplets) conceived without treatment and 34 sets with treatment.

Missing data: 281 (6.3%) length, 295 (6.7%) paternal BMI, 156 (3.5%) marital status, 1 missing specific treatment exposure (OI/IUI vs. ART).

aP < 0.05 in comparison between infertility treatment and no treatment.

bP < 0.05 in comparison between ART and OI/IUI.

Among all children (including twins and higher-order multiples), infertility treatment was significantly associated with smaller size from birth through to 3 years irrespective of growth measure (Table II). However, these differences were largely due to the greater percentage of multiples (31 versus 18%) among children conceived by infertility treatment. In fact, no differences were observed among singletons for any treatment or for the two specific types of treatment (OI/IUI or ART). Twins conceived by infertility treatment were on average smaller than twins not conceived with treatment, with most of the difference observed for twins conceived with OI/IUI rather than with ART. Twins conceived with OI/IUI weighed on average 122 g less, had lower weight for age (−0.18 z-score units; 95% CI: −0.33, −0.03), weight for length (−0.17 z-score units; 95% CI: −0.33, −0.01), BMI (−0.28 kg/m2; 95% CI: −0.50, −0.05) and BMI z-score (−0.19 units; 95% CI: −0.35, −0.03) than twins not conceived by any infertility treatment throughout early childhood up to 3 years of age. No differences in growth measures were observed for twins conceived by ART compared with twins that were not conceived with any treatment.

Table II.

Adjusted mean differences in growth from birth through 3 years of age by infertility treatment exposure, Upstate KIDS Study.

All children
Mean difference (95% CI) na Any infertility treatment P-value OI/IUI P-value ART P-value
Weight (g) 5420 −151 (−217, −84) <0.0001 −180 (−262, −97) <0.0001 −116 (−204, −27) 0.01
Weight for age (z-score) 5420 −0.25 (−0.32, −0.18) <0.0001 −0.27 (−0.36, −0.18) <0.0001 −0.23 (−0.33, −0.14) <0.0001
Length (cm) 5076 −0.61 (−0.83, −0.40) <0.0001 −0.61 (−0.88, −0.34) <0.0001 −0.63 (−0.91, −0.34) 0.0001
Length for age (z-score) 5076 −0.29 (−0.39, −0.20) <0.0001 −0.29 (−0.40, −0.17) <0.0001 −0.30 (−0.43, −0.18) <0.0001
Weight for length (z-score) 4986 −0.11 (−0.20, −0.02) 0.02 −0.14 (−0.26, −0.03) 0.02 −0.08 (−0.20, 0.04) 0.20
BMI (kg/m2) 5042 −0.29 (−0.43, −0.17) <0.0001 −0.33 (−0.49, −0.17) 0.0003 −0.26 (−0.44, −0.09) 0.005
BMI (z-score) 5042 −0.21 (−0.30, −0.13) <0.0001 −0.23 (−0.34, −0.12) 0.0002 −0.19 (−0.30, −0.07) 0.003
Singletons
Mean difference (95% CI) n Any infertility treatment P-value OI/IUI P-value ART P-value
Weight (g) 3440 −8.81 (−40.90, 58.52) 0.72 15.74 (−44.54, 76.03) 0.60 −4.12 (−85.72, 77.48) 0.92
Weight for age (z-score) 3440 −0.002 (−0.05, 0.06) 0.94 −0.0005 (−0.06, 0.07) 0.99 0.04 (−0.09, 0.10) 0.88
Length (cm) 3223 −0.16 (−0.34, 0.02) 0.08 −0.13 (−0.35, 0.09) 0.25 −0.22 (−0.52, 0.07) 0.13
Length for age (z-score) 3223 −0.08 (−0.16, −0.001) 0.05 −0.07 (−0.17, 0.03) 0.14 −0.10 (−0.23, 0.003) 0.14
Weight for length (z-score) 3182 0.06 (−0.02, 0.14) 0.15 0.06 (−0.04, 0.16) 0.22 0.05 (−0.08, 0.18) 0.42
BMI (kg/m2) 3196 −0.01 (−0.10, 0.13) 0.83 −0.03 (−0.11, 0.16) 0.71 −0.01 (−0.20, 0.17) 0.89
BMI (z-score) 3196 −0.02 (−0.06, 0.09) 0.61 −0.02 (−0.07, 0.11) 0.61 0.01 (−0.11, 0.13) 0.85
Twins
Mean difference (95% CI) nb Any infertility Treatment P-value OI/IUI P-value ART P-value
Weight (g) 1872 −81 (−185, 24) 0.12 −122 (−251, 7) 0.06 −42 (−168, 5) 0.50
Weight for age (z-score) 1872 −0.12 (−0.24, 0.0003) 0.05 −0.18 (−0.33, −0.03) 0.02 −0.06 (−0.20, 0.08) 0.40
Length (cm) 1747 −0.13 (−0.47, 0.22) 0.45 −0.26 (−0.69, 0.16) 0.22 −0.004 (−0.42, 0.41) 0.98
Length for age (z-score) 1747 −0.03 (−0.18, 0.12) 0.68 −0.10 (−0.28, 0.08) 0.26 0.04 (−0.14, 0.22) 0.67
Weight for length (z-score) 1708 −0.14 (−0.28, −0.01) 0.03 −0.17 (−0.33, −0.01) 0.03 −0.12 (−0.27, 0.03) 0.12
BMI (kg/m2) 1742 −0.22 (−0.40, −0.03) 0.02 −0.28 (−0.50, −0.05) 0.02 −0.16 (−0.38, 0.06) 0.15
BMI (z-score) 1742 −0.14 (−0.27, −0.01) 0.04 −0.19 (−0.35, −0.03) 0.02 −0.09 (−0.25, 0.07) 0.24
Open in a new tab

Model adjusted for infant age, gender and their interactions, maternal age (years), maternal race (White, Black, Asian, Hispanic, other), education (less than HS, HS, some college, college, advanced), pregnancy smoking (yes/no), private insurance (yes/no), maternal pre-pregnancy BMI (kg/m2) and paternal BMI (kg/m2).

aAll children including 108 higher-order multiples.

bAll twins counting each sibling separately. There were 954, 902, 888 and 900 twins, respectively, from the primary cohort.

With regard to rapid infant weight gain at 4, 9 and 12 months of age, approximately 19% of children at 4 months of age, 25% at 9 months and 22% at 12 months were estimated to have been affected. The odds of rapid weight gain was 2–3 times greater among infants conceived with than without infertility treatment, but only earlier during childhood (i.e. 4 and 9 months) (Table III). Twins conceived by infertility treatment, and particularly by ART, had slightly greater odds of rapid infant weight gain by 4 months (adjusted OR 1.08; 95% CI: 1.00–1.16). No significant differences were observed for singletons.

Table III.

Odds ratios (95% CI) for risk of rapid infant weight gain by infertility treatment status in the primary cohort of Upstate KIDS.

Age at measurement Primary cohorta
 Singletons
 Twins
n Unadjusted Model 1 Model 1 + plurality n Model 1 n Model 1
4-Months
 Any treatment 1408 2.00 (1.41, 2.84) 2.59 (1.80, 3.72) 1.17 (0.76, 1.81) 969 1.24 (0.72, 2.15) 784 1.06 (1.00, 1.13)
  OI/IUI 741 1.73 (1.04, 2.87) 2.24 (1.33, 3.76) 1.08 (0.59, 1.96) 535 1.21 (0.57, 2.58) 361 1.05 (0.98, 1.12)
  ART 666 2.30 (1.43, 3.70) 2.98 (1.83, 4.86) 1.27 (0.71, 2.26) 433 1.28 (0.59, 2.80) 423 1.08 (1.00, 1.16)
9-Months
 Any treatment 1408 1.19 (0.83, 1.70) 1.53 (1.06, 2.22) 1.02 (0.66, 1.54) 969 0.98 (0.55, 1.72) 784 1.02 (0.97, 1.08)
  OI/IUI 741 1.02 (0.60, 1.73) 1.26 (0.74, 2.17) 0.88 (0.50, 1.56) 535 0.85 (0.38, 1.88) 361 1.01 (0.94, 1.08)
  ART 666 1.36 (0.84, 2.23) 1.83 (1.11, 3.02) 1.18 (0.69, 2.03) 433 1.13 (0.52, 2.49) 423 1.04 (0.97, 1.11)
12-Months
 Any treatment 1408 0.87 (0.58, 1.32) 1.17 (0.77, 1.78) 0.91 (0.58, 1.43) 969 0.78 (0.41, 1.48) 784 1.01 (0.96, 1.07)
  OI/IUI 741 0.78 (0.42, 1.42) 0.99 (0.53, 1.82) 0.79 (0.42, 1.49) 535 0.66 (0.26, 1.67) 361 0.99 (0.93, 1.05)
  ART 666 0.98 (0.56, 1.70) 1.37 (0.78, 2.41) 1.04 (0.57, 1.90) 433 0.92 (0.38, 2.23) 423 1.03 (0.96, 1.10)
Open in a new tab

aPrimary cohort includes singletons and one randomly selected multiple of a set.

Model 1 adjusted for infant age, gender and their interactions, maternal age (years), maternal race (White, Black, Asian, Hispanic, other), education (less than HS, HS, some college, college, advanced), pregnancy smoking (yes/no), private insurance (yes/no), maternal pre-pregnancy BMI (kg/m2) and paternal BMI (kg/m2). Bold value indicates significance at P < 0.05.

Discussion

To our knowledge, Upstate KIDS is the first US study to longitudinally track the growth of children specifically by mode of conception. We observed that children were smaller through to 3 years of age primarily due to the greater frequency of twins and high order multiples. These differences may not be a direct function of having been conceived with infertility treatment as singletons' growth during early childhood did not differ by treatment status. Contrarily, twins conceived with OI/IUI relative to twins conceived without any treatment were smaller, though the absolute differences were small (e.g. ∼122 g in weight). However, no significant differences in growth through to 3 years of age were observed among twins born by ART compared with twins not conceived with treatment. Rapid infant weight gain was evident among twins conceived by ART only at 4-months and the risk was small.

Our study builds upon previous research while being more inclusive in important regards. To our knowledge, this is the first population-based cohort of US children prospectively followed through to 3 years of age (Yeung and Druschel, 2013). Similar to many previous studies (Brandes et al., 1992; Ron-El et al., 1994; Saunders et al., 1996; Bowen et al., 1998; Wennerholm et al., 1998; Koivurova et al., 2003; Place and Englert, 2003; Kai et al., 2006; Ludwig et al., 2009; Lee et al., 2010; Woldringh et al., 2011) on ART and childhood growth, we evaluated growth early in childhood (≤6 years old), though several other studies have included older children (Belva et al., 2007; Miles et al., 2007; Makhoul et al., 2009) or adolescents (Ceelen et al., 2007; Belva et al., 2012). To date, much of the available data is cross-sectional in nature, with growth measured at one time point rather than evaluating longitudinal trajectories, as we have done (Yeung and Druschel, 2013). Our cohort also includes multiples that have typically been excluded from previous analyses. Some studies also excluded singletons not meeting a gestational age threshold (e.g. ≥32 weeks gestation). We included a substantially larger number of children conceived by ART than almost all studies with the exception of one combining children from four European countries (Bonduelle et al., 2005). Our findings suggest that they are indeed smaller over the first 3 years of life compared with those not conceived by treatment. However, as growth of twins and higher-order multiples differ so greatly, we also stratified results by plurality. Despite the myriad of differences in study design, population and statistical methods, overall we find no evidence suggesting altered growth in early childhood for children conceived with ART, and no evidence of either over- or under-growth after accounting for plurality.

Compared with the few studies (Makhoul et al., 2009; Savage et al., 2012) that tracked growth of children conceived by non-ART treatments, our study found that twins conceived by OI/IUI were significantly smaller from birth through to age 3 years than twins not conceived by infertility treatment. Although not completely comparable, Savage et al. (2012) also found that pre-pubertal children (3–11 years old) conceived by ovarian stimulation were more likely to be of shorter stature than children who were not conceived by treatment. Their study was restricted to singletons born appropriate for gestational age (Savage et al., 2012). We did not limit our sample by birth size because we wanted to evaluate the total effect of infertility treatment, which may be in part due to mediating effects on size for gestational age (Schisterman et al., 2009). We were also careful to account for parental BMI in our analyses, as some evidence has shown maternal and paternal obesity to be associated with infertility treatment (Campbell et al., 2015) and may confound findings. Savage's study adjusted for variation in mid-parental height as a measure of genetic growth potential (Savage et al., 2012), though it is unclear whether height is a true confounder given the lack of data supporting an association with infertility or its treatment. To address this consideration, we adjusted for both parents' BMI that were positively associated with children's weight and height.

The focus of research on mode of conception and childhood health has traditionally been on ART procedures, where use of in vitro culture with or without ICSI may have implications for children's growth. We were unable to account for the type of culture media in our analysis, given that we did not have this data available. When available in previous research, the findings relative to fetal growth have been inconsistent (Nelissen et al., 2013; Lemmen et al., 2014; Wunder et al., 2014). One study on post-natal growth comparing Vitrolife and Cook culture media used in IVF found that singletons conceived differed in growth persisting to 2 years (Kleijkers et al., 2014). Specifically, Cook media was associated with a lower trajectory of weight and height among singletons than Vitrolife without any differences in growth velocity (Kleijkers et al., 2014). Hence, culture data may be necessary to examine the potential for differences in growth.

The rationale for why children conceived by ovulation induction or ovarian stimulation may be smaller than children conceived without this treatment remains unclear (Savage et al., 2012). Although our findings remain to be confirmed in both singletons and twins conceived by OI/IUI, some potential mechanisms for why hormonal medications may lead to differences in growth have been hypothesized (Wennerholm et al., 1998). Our OI/IUI group included women who used oral (e.g. Clomid©) as well as injectable medications (e.g. Ovidrel©, Profasi©). One hypothesis is that stimulation leads to ovulation of abnormal oocytes by developing them too rapidly or developing oocytes that would have otherwise perished (Sato et al., 2007; Fortier et al., 2008). Alternatively, stimulation may impact the endometrium by exposure to supra-physiological levels of estradiol, progesterone and other hormones (Weinerman and Mainigi, 2014). Estradiol, for example, is increased after application of gonadotrophins due to the higher number of follicles, as compared with only a dominant follicle, being present. Estradiol levels have also been found to be elevated after clomiphene citrate (Maxson et al., 1984). Elevations of estradiol and other hormones can then impact endometrial receptivity (as demonstrated in histological studies) and potential adaptations made to accommodate for successful placentation and survival have been hypothesized to play a role in fetal growth (Weinerman and Mainigi, 2014) and, therefore, potentially influence later growth. Some evidence from epidemiological studies showing better birth outcomes after cryopreservation has supported this latter hypothesis (Weinerman and Mainigi, 2014). However, due to the few number of our participants reporting frozen embryo transfer in the cycle of the index birth (n = 124) (Stern et al., 2016), we were unable to disentangle this aspect of treatment. That only twins were found to differ requires replication for a more complete understanding of these associations.

Rapid infant weight gain has been linked with adverse long-term health outcomes including higher risks of childhood obesity and metabolic dysfunction (Breij et al., 2014). Hence, we evaluated rapid infant weight gain as a way to better understand whether the velocity of growth differs by infertility treatment status, especially given indications that children may have lower birthweight. Our observation that twins conceived with ART experience a slightly higher odds of rapid infant weight gain suggest that treatment may not be a strong risk factor. Other literature for infant weight gain remains equivocal. A recent study using data from the Danish National Birth Cohort included children conceived by OI/IUI in their fertility treatment group along with ART and showed no differences in clinically measured anthropometry at 5 years of age despite smaller birth size, suggestive of catch up growth (Bay et al., 2014). Greater change in weight at 3 months and 1 year was also observed in one study of IVF children compared with spontaneously conceived children (Ceelen et al., 2009). However, another study in Europe did not observe the same (Basatemur et al., 2010). Our study differs from previous work by defining rapid catch-up growth at multiple time points defined by cut-points previously found to be associated with higher risks of childhood fatty liver (Breij et al., 2014) rather than purely change in weight at different time points. We used such cut-points to evaluate rapid growth that may be of concern to future health.

The strengths of our study include having maternal reported ART exposure validated against SART-CORS data (Buck Louis et al., 2015), a substantially large number of children conceived by infertility treatment compared with previous studies, and longitudinal data on growth from birth through to 3 years of age. However, we relied on maternal report of pediatrician-measured values and we were not able to validate those measures. We recognize there may be error in the measurement of length; however, we are unaware of any empirical data suggesting that it systematically varies by mode of conception that would bias findings. Nevertheless, as there is concern that clinically measured length may be reduced compared with research measured values, we also implemented an adjustment previously published (Rifas-Shiman et al., 2005) for clinically measured lengths (up until 18 months) and observed no difference in our findings (data not shown). Also, we re-ran our analyses restricting to measurements up to the first 18 months of age to distinguish length from height measures and observed similar findings (data not shown). Although we were also limited by loss to follow-up in the study, we used generalized linear mixed effects modeling with the modeling assumption of missing at random (Molenberghs, 2006). The extent to which this assumption is true remains unknown. We cannot entirely eliminate residual confounding in light of the study's observational design. The small differences identified may not be clinically meaningful, but we here rule out large differences that may be more relevant.

In conclusion, we found that children grew similarly irrespective of mode of conception from birth through to 3 years of age. While twins conceived by OI/IUI were slightly smaller than twins conceived without such treatment, the differences were minor (absolute difference in weight of 121 g and in length of 0.26 cm). These findings offer reassurance for couples using infertility treatment to aid conception, and provide evidence that children conceived by infertility treatment are not stunted or over-grown in early childhood.

Supplementary data

Supplementary data are available at http://humrep.oxfordjournals.org/.

Authors’ roles

E.H.Y., R.S. and Y.X. had full access to the study data and takes responsibility for the integrity of its analysis and statistical analysis; G.M.B.L.: study concept and design; E.M.B., C.D., C.K.: oversight for implementation of the protocol and linkage with New York State registries; E.M.B., G.M.B.L., C.K., E.H.Y.: data interpretation; E.H.Y.: drafting the manuscript; E.M.B., G.M.B.L., C.K., Y.X., E.H.Y.: critical revision; G.M.B.L.: Obtained funding; E.H.Y., C.D., E.M.B.: study supervision.

Funding

Supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD; contracts #HHSN275201200005C, #HHSN267200700019C).

Conflict of interest

None declared.

Supplementary Material

Supplementary Data
supp_31_7_1621__index.html (885B, html)

Acknowledgements

The authors thank Drs Barbara Luke and Judy E. Stern for their important contributions in linking Upstate KIDS with SART CORS data. We also wish to thank those NY State clinics who submitted their data to SART CORS without whose efforts this linkage would not have been possible. The authors also thank all the Upstate KIDS participants and staff for their important contributions.

References

  1. Basatemur E, Shevlin M, Sutcliffe A. Growth of children conceived by IVF and ICSI up to 12 years of age. Reprod Biomed Online 2010;20:144–149. [DOI] [PubMed] [Google Scholar]
  2. Bay B, Mortensen EL, Kesmodel US. Is subfertility or fertility treatment associated with long-term growth in the offspring? A cohort study. Fertil Steril 2014;102:1117–1123. [DOI] [PubMed] [Google Scholar]
  3. Belva F, Henriet S, Liebaers I, Van SA, Celestin-Westreich S, Bonduelle M. Medical outcome of 8-year-old singleton ICSI children (born > or = 32 weeks’ gestation) and a spontaneously conceived comparison group. Hum Reprod 2007;22:506–515. [DOI] [PubMed] [Google Scholar]
  4. Belva F, Painter R, Bonduelle M, Roelants M, Devroey P, De SJ. Are ICSI adolescents at risk for increased adiposity? Hum Reprod 2012;27:257–264. [DOI] [PubMed] [Google Scholar]
  5. Bonduelle M, Wennerholm UB, Loft A, Tarlatzis BC, Peters C, Henriet S, Mau C, Victorin-Cederquist A, Van SA, Balaska A et al. . A multi-centre cohort study of the physical health of 5-year-old children conceived after intracytoplasmic sperm injection, in vitro fertilization and natural conception. Hum Reprod 2005;20:413–419. [DOI] [PubMed] [Google Scholar]
  6. Bowen JR, Gibson FL, Leslie GI, Saunders DM. Medical and developmental outcome at 1 year for children conceived by intracytoplasmic sperm injection. Lancet 1998;351:1529–1534. [DOI] [PubMed] [Google Scholar]
  7. Brandes JM, Scher A, Itzkovits J, Thaler I, Sarid M, Gershoni-Baruch R. Growth and development of children conceived by in vitro fertilization. Pediatrics 1992;90:424–429. [PubMed] [Google Scholar]
  8. Breij LM, Kerkhof GF, Hokken-Koelega AC. Accelerated infant weight gain and risk for nonalcoholic fatty liver disease in early adulthood. J Clin Endocrinol Metab 2014;99:1189–1195. [DOI] [PubMed] [Google Scholar]
  9. Buck Louis GM, Hediger ML, Bell EM, Kus CA, Sundaram R, McLain AC, Yeung E, Hills EA, Thoma ME, Druschel CM. Methodology for establishing a population-based birth cohort focusing on couple fertility and children's development, the Upstate KIDS Study. Paediatr Perinat Epidemiol 2014;28:191–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Buck Louis GM, Druschel C, Bell E, Stern JE, Luke B, McLain A, Sundaram R, Yeung E. Use of assisted reproductive technology treatment as reported by mothers in comparison with registry data: the Upstate KIDS Study. Fertil Steril 2015;103:1461–1468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Campbell JM, Lane M, Owens JA, Bakos HW. Paternal obesity negatively affects male fertility and assisted reproduction outcomes: a systematic review and meta-analysis. Reprod Biomed Online 2015;31:593–604. [DOI] [PubMed] [Google Scholar]
  12. Ceelen M, van Weissenbruch MM, Roos JC, Vermeiden JP, van Leeuwen FE, Delemarre-van de Waal HA. Body composition in children and adolescents born after in vitro fertilization or spontaneous conception. J Clin Endocrinol Metab 2007;92:3417–3423. [DOI] [PubMed] [Google Scholar]
  13. Ceelen M, van Weissenbruch MM, Prein J, Smit JJ, Vermeiden JP, Spreeuwenberg M, van Leeuwen FE, Delemarre-van de Waal HA. Growth during infancy and early childhood in relation to blood pressure and body fat measures at age 8–18 years of IVF children and spontaneously conceived controls born to subfertile parents. Hum Reprod 2009;24:2788–2795. [DOI] [PubMed] [Google Scholar]
  14. de Onis M, Onyango AW, Borghi E, Garza C, Yang H, Group WHOMGRS. Comparison of the World Health Organization (WHO) Child Growth Standards and the National Center for Health Statistics/WHO international growth reference: implications for child health programmes. Public Health Nutr 2006;9:942–947. [DOI] [PubMed] [Google Scholar]
  15. Feuer SK, Camarano L, Rinaudo PF. ART and health: clinical outcomes and insights on molecular mechanisms from rodent studies. Mol Hum Reprod 2013;19:189–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fortier AL, Lopes FL, Darricarrere N, Martel J, Trasler JM. Superovulation alters the expression of imprinted genes in the midgestation mouse placenta. Hum Mol Genet 2008;17:1653–1665. [DOI] [PubMed] [Google Scholar]
  17. Kai CM, Main KM, Andersen AN, Loft A, Chellakooty M, Skakkebaek NE, Juul A. Serum insulin-like growth factor-I (IGF-I) and growth in children born after assisted reproduction. J Clin Endocrinol Metab 2006;91:4352–4360. [DOI] [PubMed] [Google Scholar]
  18. Kelishadi R, Haghdoost AA, Jamshidi F, Aliramezany M, Moosazadeh M. Low birthweight or rapid catch-up growth: which is more associated with cardiovascular disease and its risk factors in later life? A systematic review and cryptanalysis. Paediatr Int Child Health 2015;35:110–123. [DOI] [PubMed] [Google Scholar]
  19. Kleijkers SH, van Montfoort AP, Smits LJ, Viechtbauer W, Roseboom TJ, Nelissen EC, Coonen E, Derhaag JG, Bastings L, Schreurs IE et al. . IVF culture medium affects post-natal weight in humans during the first 2 years of life. Hum Reprod 2014;29:661–669. [DOI] [PubMed] [Google Scholar]
  20. Koivurova S, Hartikainen AL, Sovio U, Gissler M, Hemminki E, Jarvelin MR. Growth, psychomotor development and morbidity up to 3 years of age in children born after IVF. Hum Reprod 2003;18:2328–2336. [DOI] [PubMed] [Google Scholar]
  21. Kupka MS, Ferraretti AP, de Mouzon J, Erb K, D'Hooghe T, Castilla JA, Calhaz-Jorge C, De Geyter C, Goossens V, European Ivf-Monitoring Consortium ftESoHR et al. Assisted reproductive technology in Europe, 2010: results generated from European registers by ESHREdagger. Hum Reprod 2014;29:2099–2113.25069504 [Google Scholar]
  22. Lee SH, Lee MY, Chiang TL, Lee MS, Lee MC. Child growth from birth to 18 months old born after assisted reproductive technology–results of a national birth cohort study. Int J Nurs Stud 2010;47:1159–1166. [DOI] [PubMed] [Google Scholar]
  23. Lemmen JG, Pinborg A, Rasmussen S, Ziebe S. Birthweight distribution in ART singletons resulting from embryo culture in two different culture media compared with the national population. Hum Reprod 2014;29:2326–2332. [DOI] [PubMed] [Google Scholar]
  24. Ludwig AK, Katalinic A, Thyen U, Sutcliffe AG, Diedrich K, Ludwig M. Physical health at 5.5 years of age of term-born singletons after intracytoplasmic sperm injection: results of a prospective, controlled, single-blinded study. Fertil Steril 2009;91:115–124. [DOI] [PubMed] [Google Scholar]
  25. Makhoul IR, Tamir A, Bader D, Rotschild A, Weintraub Z, Yurman S, Reich D, Bental Y, Jammalieh J, Smolkin T et al. . In vitro fertilisation and use of ovulation enhancers may both influence childhood height in very low birthweight infants. Arch Dis Child Fetal Neonatal Ed 2009;94:F355–F359. [DOI] [PubMed] [Google Scholar]
  26. Maxson WS, Pittaway DE, Herbert CM, Garner CH, Wentz AC. Antiestrogenic effect of clomiphene citrate: correlation with serum estradiol concentrations. Fertil Steril 1984;42:356–359. [PubMed] [Google Scholar]
  27. McDonald SD, Han Z, Mulla S, Murphy KE, Beyene J, Ohlsson A. Preterm birth and low birth weight among in vitro fertilization singletons: a systematic review and meta-analyses. Eur J Obstet Gynecol Reprod Biol 2009;146:138–148. [DOI] [PubMed] [Google Scholar]
  28. McDonald SD, Han Z, Mulla S, Ohlsson A, Beyene J, Murphy KE. Preterm birth and low birth weight among in vitro fertilization twins: a systematic review and meta-analyses. Eur J Obstet Gynecol Reprod Biol 2010;148:105–113. [DOI] [PubMed] [Google Scholar]
  29. Miles HL, Hofman PL, Peek J, Harris M, Wilson D, Robinson EM, Gluckman PD, Cutfield WS. In vitro fertilization improves childhood growth and metabolism. J Clin Endocrinol Metab 2007;92:3441–3445. [DOI] [PubMed] [Google Scholar]
  30. Molenberghs GV. Models for Discrete Longitudinal Data. New York, NY: Springer Science + Business Media, Inc., 2006. [Google Scholar]
  31. Nelissen EC, Van Montfoort AP, Smits LJ, Menheere PP, Evers JL, Coonen E, Derhaag JG, Peeters LL, Coumans AB, Dumoulin JC. IVF culture medium affects human intrauterine growth as early as the second trimester of pregnancy. Hum Reprod 2013;28:2067–2074. [DOI] [PubMed] [Google Scholar]
  32. Place I, Englert Y. A prospective longitudinal study of the physical, psychomotor, and intellectual development of singleton children up to 5 years who were conceived by intracytoplasmic sperm injection compared with children conceived spontaneously and by in vitro fertilization. Fertil Steril 2003;80:1388–1397. [DOI] [PubMed] [Google Scholar]
  33. Rifas-Shiman SL, Rich-Edwards JW, Scanlon KS, Kleinman KP, Gillman MW. Misdiagnosis of overweight and underweight children younger than 2 years of age due to length measurement bias. Med Gen Med 2005;7:56. [PMC free article] [PubMed] [Google Scholar]
  34. Ron-El R, Lahat E, Golan A, Lerman M, Bukovsky I, Herman A. Development of children born after ovarian superovulation induced by long-acting gonadotropin-releasing hormone agonist and menotropins, and by in vitro fertilization. J Pediatr 1994;125:734–737. [DOI] [PubMed] [Google Scholar]
  35. Sato A, Otsu E, Negishi H, Utsunomiya T, Arima T. Aberrant DNA methylation of imprinted loci in superovulated oocytes. Hum Reprod 2007;22:26–35. [DOI] [PubMed] [Google Scholar]
  36. Saunders K, Spensley J, Munro J, Halasz G. Growth and physical outcome of children conceived by in vitro fertilization. Pediatrics 1996;97:688–692. [PubMed] [Google Scholar]
  37. Savage T, Peek JC, Robinson EM, Green MP, Miles HL, Mouat F, Hofman PL, Cutfield WS. Ovarian stimulation leads to shorter stature in childhood. Hum Reprod 2012;27:3092–3099. [DOI] [PubMed] [Google Scholar]
  38. Schieve LA, Devine O, Boyle CA, Petrini JR, Warner L. Estimation of the contribution of non-assisted reproductive technology ovulation stimulation fertility treatments to US singleton and multiple births. Am J Epidemiol 2009;170:1396–1407. [DOI] [PubMed] [Google Scholar]
  39. Schisterman EF, Cole SR, Platt RW. Overadjustment bias and unnecessary adjustment in epidemiologic studies. Epidemiology 2009;20:488–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Stern JE, McLain AC, Buck Louis GM, Luke B, Yeung EH. Accuracy of self-reported survey data on assisted reproductive technology treatment parameters and reproductive history. Am J Obstet Gynecol 2016. doi:10.1016/j.ajog.2016.02.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sunderam S, Kissin DM, Crawford SB, Folger SG, Jamieson DJ, Barfield WD, Centers for Disease C, Prevention. Assisted reproductive technology surveillance–United States, 2011. MMWR Surveill Summ 2014;63:1–28. [PubMed] [Google Scholar]
  42. Tukey JW. Exploratory Data Analysis. Reading, PA: Addison-Wesley, 1977. [Google Scholar]
  43. Weinerman R, Mainigi M. Why we should transfer frozen instead of fresh embryos: the translational rationale. Fertil Steril 2014;102:10–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wennerholm UB, bertsson-Wikland K, Bergh C, Hamberger L, Niklasson A, Nilsson L, Thiringer K, Wennergren M, Wikland M, Borres MP. Postnatal growth and health in children born after cryopreservation as embryos. Lancet 1998;351:1085–1090. [DOI] [PubMed] [Google Scholar]
  45. Woldringh GH, Hendriks JC, van Klingeren J, van Buuren S, Kollee LA, Zielhuis GA, Kremer JA. Weight of in vitro fertilization and intracytoplasmic sperm injection singletons in early childhood. Fertil Steril 2011;95:2775–2777. [DOI] [PubMed] [Google Scholar]
  46. Wunder D, Ballabeni P, Roth-Kleiner M, Primi MP, Senn A, Chanson A, Germond M, Leyvraz C. Effect of embryo culture media on birthweight and length in singleton term infants after IVF-ICSI. Swiss Med Wkly 2014;144:w14038. [DOI] [PubMed] [Google Scholar]
  47. Yeung EH, Druschel C. Cardiometabolic health of children conceived by assisted reproductive technologies. Fertil Steril 2013;99:318–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yeung EH, McLain AC, Anderson N, Lawrence D, Boghossian NS, Druschel C, Bell E. Newborn adipokines and birth outcomes. Paediatr Perinatal Epidemiol 2015;29:317–325. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_31_7_1621__index.html (885B, html)
supp_dew106_dew106supp_fig1.pdf (175.9KB, pdf)

Articles from Human Reproduction (Oxford, England) are provided here courtesy of Oxford University Press

RESOURCES