Objective
In vitro fertilization (IVF) is considered a generally safe procedure, although associated with a higher incidence of preterm birth. The literature is inconsistent about the psychological impact of IVF, and we found no reports about outcome in late preterm (LPT) children. Our objective was to study neuropsychological and behavioral outcomes in a cohort of preschoolers born LPT between 2004 and 2007.
Study Design
Participants were 397 LPT children (mean age, 3.8 years) conceived assisted by IVF (n = 105) or non-IVF (n = 292). Standardized performance-based tests of general conceptual ability (intelligence quotient), executive function, focused/selective attention, visual-spatial perception, visual-motor skill, manual dexterity, learning, and memory were administered. Parents completed behavioral and executive function questionnaires.
Results
IVF group characteristics included older maternal age ( P < .001), lower birthweight ( P < .001), and higher maternal education ( P < .001). No main-effect significant group differences were found for any variable after controlling for these variables. However, sex differences were demonstrated for the neuropsychological variables in copying ( P > .001), nonverbal reasoning ( P = .001), manual dexterity ( P = .001), and inhibitory capacity ( P = .006), all favoring girls.
Conclusion
Birth following IVF-assisted conception did not increase the risk of intellectual, neuropsychological, or behavioral deficit in LPT preschoolers. As shown in earlier gestational-age participants, girls have selective advantages. These findings should be reassuring for parents who conceive through IVF and deliver infants 1-3 weeks before term gestational age. Future study of these children at elementary school age may detect subtle impairments not yet apparent at age 3 years.
For Editors’ Commentary, see Contents
Following the first successful in vitro fertilization (IVF) birth in 1978, the incidence of IVF birth has increased to 1% of all births in the United States. Successful live birth rates have risen to 41.7% for women under age 35 years and 12.5% for women over age 42 years. IVF is a viable option in cases of infertility caused by endometriosis, fallopian tube damage, partner sterility, or advanced maternal age, and when other less expensive methods of assisted reproductive technology have failed. Although generally considered to be a safe procedure, potential birth defects, adverse perinatal health outcomes, low birthweight, and preterm birth have been associated with children conceived by IVF. A 14% prevalence rate of preterm birth (<37 weeks) has also been reported in IVF children compared with children spontaneously conceived.
The literature regarding neuropsychological and behavioral outcomes of children born following IVF has been inconsistent. Outcomes are either reported to be advantageous, adverse, or neither. Some investigators have reported good outcomes as demonstrated by higher scores on standardized achievement tests and by maternal perceptions of fewer behavior problems in children conceived by IVF compared with those born following non-IVF, spontaneous conception. Some studies found IVF to be associated with an increased risk of neurological problems and worse behavioral outcomes in the first year of life, whereas others found no significant differences in development between IVF and non-IVF groups.
Two studies included 2 additional infertile control groups and found that children born after IVF were at no greater risk of psychomotor, neurodevelopmental, neuromotor, cognitive, language, or behavioral impairments at age 3 years than children born after ovarian hyperstimulation, to subfertile naturally conceiving parents, or spontaneously. IVF and non-IVF adolescents did not differ on neuropsychological measures of information processing, attention, or visual-motor integration. However, none of these studies confined their cohort to children born at late preterm (LPT: 34-36 weeks) gestational age, a population at higher risk compared with those born at term.
The research concerning outcomes of preterm birth by sex indicates that very preterm boys (<25 weeks) are more vulnerable to long-term neurological and motor impairments than girls born at the same gestation. An additional study found that school-aged boys born LPT demonstrated worse visuospatial reasoning, attentional control, and executive functioning than LPT girls. Although these studies support the likelihood of a neuropsychological sex differential, none addressed sex differences for preterm children born after IVF.
Thus, contradictory results characterize the literature on IVF and psychological outcome in children born at term gestational age, sparse data exist about the effects of IVF and sex on children born preterm after IVF, and we found no reported study of IVF in children born late preterm (LPT). Notably, some born LPT will experience associated subtle neuropsychological and behavioral deficits, such as as hyperkinetic disorder.
This study’s objective was to study how IVF conception influences the cognitive, neuropsychological, and behavioral outcomes of preschool-aged children born LPT. We hypothesized that preschoolers conceived by IVF and born LPT would have worse cognitive, neuropsychological, and behavioral outcomes compared with LPT preschoolers conceived non-IVF. We also included sex as a supplemental area of inquiry and hypothesized that girls would perform more favorably than boys on all variables.
Materials and Methods
Participants
The Institutional Review Board of Inova Children’s Hospital (Falls Church, VA) approved the protocol and procedures described in the following text for this observational, single-center, cohort study, part of the ongoing Prematurity’s Effects on Toddlers, Infants, and Teens (PETIT) Study. Parents provided written informed consent prior to their child’s participation, which was conducted by well-trained research examiners blind to participant group and preterm status.
The 3 year old study group included 397 LPT participants born between 2004 and 2007: 105 born assisted by IVF (52 male/53 female) and 292 non-IVF (160 male/132 female). LPT participants were recruited through mailing of institutional review board–approved letters to the last known addresses of randomly selected eligible children born at our hospital, through flyers posted in local pediatricians’ offices, and through parent community groups. A priori exclusion criteria were genetic disorder, severe sensorineural impairment, brain tumor, or the child not speaking English. Tables 1 and 2 describe participant sociodemographic and medical characteristics by group, obtained from medical records. Participant chronological age at testing ranged from 3 years 6 months to 3 years 11 months.
| IVF (n = 105) | Non-IVF ( n = 292) | d | Total (n = 397) | t | P value | ||||
|---|---|---|---|---|---|---|---|---|---|
| Participant characteristics | Mean | SD | Mean | SD | Mean | SD | |||
| Gestational age, wks | 34.82 | 0.76 | 35.03 | 0.80 | 0.27 | 34.98 | 0.79 | —2.39 | .020 |
| Age at testing, y | 3.09 | 0.28 | 3.02 | 0.15 | 0.33 | 3.04 | 0.20 | 2.80 | .010 |
| Birthweight, g | 2322.40 | 384.35 | 2535.74 | 537.31 | 0.46 | 2479.22 | 508.50 | —3.74 | < .001 a |
| Length of hospital stay, d | 10.65 | 6.25 | 9.26 | 6.69 | 0.21 | 9.61 | 6.59 | 1.86 | .060 |
| Maternal education, y | 16.91 | 2.05 | 16.02 | 2.16 | 0.42 | 16.24 | 2.17 | 3.70 | < .001 a |
| Maternal age, y | 37.73 | 4.72 | 33.22 | 5.04 | 0.92 | 34.40 | 5.33 | 8.00 | < .001 a |
| Participant characteristics | IVF (n = 105) | Non-IVF (n = 292) | Total (n = 397) | |||
|---|---|---|---|---|---|---|
| n | % | n | % | n | % | |
| Cesarean delivery | 93 | 87.6 | 195 | 63.4 | 288 | 69.7 |
| Twin delivery | 54 | 51.4 | 42 | 14.4 | 96 | 24.2 |
| Male sex | 52 | 49.5 | 162 | 54.8 | 214 | 53.6 |
| Right-handedness | 84 | 79.0 | 227 | 77.7 | 311 | 77.9 |
| Race/ethnicity | ||||||
| African American | 0 | 0.0 | 19 | 6.2 | 19 | 4.8 |
| White | 84 | 80.0 | 188 | 64.4 | 272 | 68.2 |
| Hispanic | 4 | 3.8 | 25 | 8.6 | 29 | 7.3 |
| Other | 17 | 16.0 | 64 | 21.2 | 81 | 19.8 |
| English as second language | 17 | 16.0 | 87 | 29.8 | 104 | 26.3 |
Test instruments
Selected tests were developmentally appropriate measures of general conceptual ability (GCA), executive function, attention, memory, language, motor and visual-motor skill, and behavioral symptoms. These domains were chosen to provide a broad representation of emergent neuropsychological and behavioral functioning at age 3 years. Test instruments and parental behavioral report questionnaires included the following parameters.
Differential Ability Scales, second edition
This multisubtest battery provides GCA, verbal cluster, nonverbal reasoning, and spatial cluster standard scores, each with a mean of 100 and SD of 15 and subtest scores with a mean of 10 and SD of 3. Two core subtests (verbal comprehension and naming vocabulary) comprise the verbal cluster, the nonverbal cluster (picture similarities and matrices), and the spatial cluster (pattern construction and copying). Three additional diagnostic subtests were administered (recognition of pictures, recall of digits forward, and early number concepts).
Developmental Test of Visual-Motor Integration, fifth edition
This is an untimed paper-and-pencil test of graphomotor proficiency requiring copying of increasingly complex geometric designs. Standardized scores have a mean of 100 and SD of 15.
Modified-Hopkins Board
This is a task that requires naming line drawings (initial naming score) printed on individual cards, placement of the card in their correct respective locations over learning trials (trials-to-criterion and errors-to-criterion scores), and delayed recall of the picture names (delayed item naming score) and spatial locations (delayed location recall score).
Purdue pegboard Test of Manual Dexterity
This is a timed test of motor dexterity requiring unilateral and bilateral placement of keyhole shaped pegs into a form board, yielding 3 raw scores: number of pegs placed by the dominant hand, nondominant hand, and both hands working simultaneously.
Noun fluency
This is a measure of verbal word retrieval of animal exemplars within 60 seconds, providing raw scores for a total number of correct words generated.
Action-verb fluency
The nonstandardized task (Baron, personal communication) requires verbal retrieval of words describing things people do within a 60 second time span, providing raw scores for a total number of correct words generated.
Behavior Rating Inventory of Executive Function: Preschool Parent Form (BRIEF-P)
This parental questionnaire provides 5 subscale T scores (inhibit, shift, emotional control, working memory, and plan/organize), composite T scores for inhibitory self-control, flexibility, and emergent metacognition, and an overall global executive composite.
Behavioral Assessment Scales for Children-2: Preschool Parent Form
The parental questionnaire provides 12 subscale T scores (hyperactivity, aggression, anxiety, depression, somatization, attention problems, atypicality, withdrawal, adaptability, social skills, activities of daily living, and functional communication) and index T scores for externalizing problems, internalizing problems, behavioral symptoms, and adaptive skills.
Analysis
Group differences (LPT/IVF vs LPT/non-IVF) on cognitive, neuropsychological, and behavioral variables were examined by conducting Student t tests. Analysis of covariance was conducted to examine differences between sexes (boy vs girl) and group (IVF vs non-IVF) for all variables while controlling for birthweight, maternal education, and maternal age. Effect sizes were calculated using Cohen’s d , with a small effect size defined as 0.2, moderate as 0.5, and large as 0.8. A conservative alpha of .01 was set to control for type I error. The following assumptions were tested: independence of observations, homogeneity of variance, normality of the distribution of the dependent variable, and homogeneity of slopes.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
