Neurocognitive outcome of monochorionic twins with different birth weights

Abstract

Background: Although the long term effect of intrauterine growth restriction has been assessed in a number of singleton studies, they all suffer from multiple confounding effects. A model that utilises monozygotic twins may markedly reduce the effect of confounders as monochorionic twins share the same gestational age length, family background, gender and genetic influences on growth and cognition. Comparison of monochorionic twins with birth weight discordance of 20% or more could be used as a model of in utero growth constraint. This model will still involve certain limitations and assumptions nevertheless; we used this to determine the level of cognitive function of in-utero growth discordant monochorionic twins in later childhood along with any differences in auxology and behavioural problems. Methods: This was a retrospective cohort study. Eligible twins were identified from the Northern Survey of Twins and Multiple Pregnancies register. Cognitive function was assessed by a single observer using the British Ability Scales 2 to measure the general conceptual ability. Strength and Difficulties Questionnaire was used to identify behavioral problems. Height, weight, mid arm circumference, waist measurement and head circumference were also collected. Generalised estimating equations were used to determine the effect of birth weight on general conceptual ability scores. Statistical analyses were performed using SPSS v19. Results: Between 2000 and 2004, a total of 51 twin pairs were assessed (n=23 female) with mean birth weight discordance 664gm and mean gestational age 34.7 weeks. The mean difference in the general conceptual ability score between the heavier and lighter twins was 3 points. Significant association between within pair differences in birth weight and general conceptual ability scores was found. Increasing birth weight discordance was not associated with a decrease of general conceptual ability scores. The differences in the size seen at birth between the twins were still detectable at the age of 5-8 years. There was a trend to increased prevalence of behavioural problems in the lighter twin compared to the heavier twin as reported by both teachers and parents but this result was not statistically significant. Conclusions: The smaller twin of a monochorionic growth discrepant pair was statistically significantly more likely to have a lower cognitive score compared to their co-twin at 5-8 years of age. This suggests that growth restriction in-utero is associated with lower cognitive scores in later childhood.