Fetal and postnatal patterns of growth in a bi-ethnic sample of children
2015-06-24T10:23:42Z (GMT) by
Background: Substantial variation exists between ethnicities in both birth weight and the prevalence of obesity-related non-communicable diseases (OR-NCDs). South Asians, who display a reduced birth weight and increased risk of developing these OR-NCDS, have been the focus of much of the research into the developmental origins of health and disease (DOHaD) paradigm. However, little research utilising ultrasonically derived estimates of fetal growth has been conducted. The use of more direct measures of fetal growth may also enable the identification of relationships between patterns of fetal growth with patterns of postnatal growth, explicitly, whether periods of restricted or rapid growth lead to postnatal catch-up or down, respectively. The known differences in birth weight existing between South Asians and White British infants may also have implications for the assessment of neonatal health in these sub-groups when using a population derived birth weight chart, such as the UK-World Health Organisation (UK-WHO). Customised charts, which adjust for maternal variables including ethnicity, have been recommended for clinical practice, yet evidence for their efficacy is varied. Objectives: The aims of this thesis were to: 1) investigate whether fetal growth patterns differ between Pakistani and White British foetuses and determine whether maternal size and demographic variables mediate any such differences; 2) produce a birth weight chart adjusting for ethnicity and compare this to the UK-WHO and customised birth weight charts to determine which chart better identifies neonates at risk of the adverse delivery and neonatal outcomes associated with small-for-gestational-age (SGA) and large-for-gestational age (LGA); 3) identify whether there is evidence of weight growth tracking between fetal and infant periods and determine whether patterns of fetal growth predict patterns of postnatal growth. Methods: All data come from the Born in Bradford (BiB) birth cohort. Objective 1: Multilevel models and fractional polynomials were employed for the modelling of fetal weight, head circumference (HC) and abdominal circumference (AC) growth. Potential mediators of the effect of being of Pakistani origin were entered into the model and the effect on the ethnicity variable was assessed. Objective 2: Ethnic specific birth weight charts (BiB) were constructed using the LMS method. SGA and LGA were defined as a birth weight <10th and >90th relative to the BiB, the UK-WHO or the customised charts. Sensitivity, specificity, positive & negative predictive values and area-under-the curve were calculated for each of the three charts SGA and LGA cut-offs, to assess the predictive ability of each chart for a range of delivery and neonatal outcomes. Objective 3: Multilevel models were employed for the modelling of fetal and postnatal growth. Fitted values were produced at 20, 30, 40 prenatal weeks & 1, 3, 6, 9, 12, 24 postnatal months in both an internal reference and the sample population. Z scores were calculated and conditional Z scores were generated to account for regression to the mean. Growth tracking was defined as change in Z score ≤ 0.67 & ≥ -0.67. Restricted and rapid fetal growth were defined as a change in Z score in the fetal period of <-0.67 and >0.67, respectively. Catch-down and catch-up growth were defined in the same way, except in the postnatal period. ANOVAs were used to test for differences in size and growth by type of fetal growth. Furthermore, logistic regression and a sensitivity and specificity analysis were employed to examine the predictive ability of the type of fetal growth. Results: Objective 1: Pakistani fetuses were significantly smaller and lighter than White British fetuses, throughout gestation. In terms of weight, Pakistani fetuses were approximately 2.25% lighter at 20 weeks, 4.13% at 30 weeks and 5.94% at 40 weeks. The differences in size for AC and HC between the two groups were not as great, with the AC and HC of Pakistani fetuses being approximately 4.1% and 1.25% smaller, respectively, at 40 weeks. Despite these significant differences in size the pattern of growth for HC and weight was not significantly different between the two groups. There was a trend for Pakistani fetuses to display a greater deceleration of growth in the final trimester (figure 4-12). The biggest mediators of the effect of being of Pakistani origin were maternal height and weight. Objective 2: Classifying infants as SGA or LGA by the BiB, UK-WHO or customised charts had low predictive utility for the outcomes under investigation. Despite the fact that the BiB ethnic specific birth weight reference provided significantly better prediction for more outcomes than both the UK-WHO and customised charts in both White British and Pakistani infants, with the exception of shoulder dystocia, AUROC values for all three charts were all below 0.61. Objective 3: The prevalence of tracking within the same centile band from 20 weeks gestation to 2 years was 10.82%. Infants who experienced restricted fetal growth remained significantly lighter than those who had not, for the duration of infancy. In this group however, there was a pattern of greater growth than expected during infancy. This was opposite to the pattern observed in infants who had experienced rapid fetal growth, who exhibited less growth than expected during infancy. However, the ability of the type of fetal growth to predict the pattern of postnatal growth was minimal, with only rapid fetal growth being significantly associated with increased odds of catch-down growth in infancy. Conclusions: No ethnic difference in the pattern of growth was found in terms of the whole body (weight) or in HC. The trend for reduced growth of the AC in Pakistanis may be a result of a reduced growth of the visceral organs during the third trimester, which may lead to both an altered liver metabolism and impaired renal function in post-natal life. Although being small or large at birth may increase the risk of an adverse neonatal outcome, size alone is not sensitive or specific enough with current detection to be a useful clinical tool. The finding that neither restricted nor rapid fetal growth predicted postnatal catch-up growth may suggest that the timing of canalisation is outside of the fetal period. If infant catch-up and down growth are not associated with periods of restricted or rapid fetal growth, the definitions of these growth patterns may need revising.