Physical growth and intelligence development of discordant dizygotic twins from birth to preschool age: a prospective cohort study

Subjects

The study included all neonatal discordance dizygotic twins born at The Second Affiliated Hospital of Wenzhou Medical University (WMU) from June 2012 to June 2013 and survived to discharge. Dizygotic twins whose birth weight differed by ≥ 20% were classified as discordant twins. Informed consent was obtained from the parents before enrollment in the study. The Second Affiliated Hospital of WMU Ethics Committee approved the study. The exclusion criteria were infants with major congenital malformations, such as major cardiac malformations, chromosomal anomalies, and skeletal, renal, brain, and lung malformations. According to the birth weight in each twin pair, the smaller one was categorized into Group Twin-Low birthweight (Group Twin-L), and the larger one was assigned to Group Twin-High birthweight (Group Twin-H).

Procedure

Information related to maternal (maternal ethnicity, weight, height, age, marital status, education, family income, health insurance status, social status, reproductive history, and medical diseases before pregnancy) and neonatal (including pregnancy systolic/diastolic peak (S/D) value, birth weight, length, head circumference, gestational age, gender, malformation, Apgar score, respiratory distress, NICU entry, and feeding option) characteristics were collected from the medical files of the mothers and infants or an in‐person interview.

The height and weight parameters of each child were measured by physical examinations or collected from the Wenzhou Maternal and Child Healthcare system at birth and 1-, 3-, 6-, 8-, 12-, 18-, 30-, 36-, 48-, 60-, and 72-months-old. The electronic baby scale and clinostatic body length standard bed were designed to help monitor the weight and body length (measured lying flat with a single suit) of children < 36-months-old. A uniform height and weight scale were designed to help monitor the height and weight gain (measured standing without shoes) for children aged ≥ 36 months. All measurements were performed by the same experienced physician.

Bayley Scales of Infant Development (BSID), including intelligence and exercise scales, was used to test the psychomotor and mental development of 1-year-old infants. The intelligence and exercise scales were expressed by mental development index (MDI) and psychomotor development index (PDI), respectively. After the infants gained familiarity and were cooperative in a quiet environment, the evaluation process was carried out item by item to estimate the MDI and PDI. The MDI and PDI scores were defined as follows: low (< 70), critical (70–89), moderate (90–109), and high (≥ 110).

Wechsler Intelligence Scale for Children-IV (WISC-IV), including verbal test and operational test, was used to assess the intelligence development of 6-year-old children. After obtaining full familiarity and cooperation from the children in a quiet environment, the evaluation process was carried out item by item to deduce the full-scale IQ, including verbal comprehension index (VCI), perceptual reasoning index (PRI), working memory index (WMI), and processing speed index (PSI). According to the scores of the full-scale IQ, children were distributed into two groups: Group IQ-Normal (IQ-N) with scores ≥ 90 and Group IQ-Low (IQ-L) with scores < 90.

Cognitive performance tests were carried out by expert psychologists (who had a degree of Master of Arts in Psychology) at the Institute of Cognitive Sciences of the Second Affiliated Hospital of WMU. The examiners were fully unaware of the status of the twins. The evaluation was repeated after 15 days if the children did not cooperate satisfactorily.

Statistical analysis

The smooth spline function of R 3.5.3 was used to fit the sample data with three smoothing splines. Firstly, the parameter “CV (cross-validation) = true” was set to select the appropriate degree of freedom for the group with a small sample size using the leave-one-out cross-validation. Then, the fitting curves of the physical indicators of each group were drawn based on this degree of freedom. The data were analyzed using SPSS 19.0 software. Paired t-test was used to compare the measured data between normal distribution datasets, expressed as mean ± standard deviation, while paired Wilcoxon rank-sum test was used to compare the non-normal data sets, expressed as (minimum, maximum). The chi-square test was used for enumeration data, expressed as a ratio or constituent ratio. In univariate analysis, the t-test was used to compare the normal distribution variables, the Mann–Whitney U test for skewed distribution variables, and Fisher’s exact test was utilized for classified variables. In multivariate analysis, the linear regression method was applied. P < 0.05 indicated a statistically significant difference. The predictive performance of dependent variables to predict the degree of IQ scores was quantified using the area under the curve (AUC) of the receiver operating characteristic (ROC) analysis. An AUC of 0.5 represents a variable with no predictive ability, and the closer the AUC to 1.0, the higher the authenticity of the variable.

留言 (0)

沒有登入
gif