Presented at the Neonatal Society 2009 Autumn Meeting.
Murgasova D1, Logan K1, Santhakumaran S1, Hyde MJ1, Thomas EL2, Bell JD2, Modi N1
1 Section of Neonatal Medicine, Division of Medicine, Chelsea and Westminster campus, Imperial College London, UK
2 Metabolic & Molecular Imaging Group, MRC Clinical Sciences Centre, Imperial College London, UK
Background: Intrahepatocellular lipid (IHCL) is a strong correlate of insulin resistance in adults. Our previous studies have shown wide variation in IHCL in healthy term newborns(1). This observation led us to consider the influence of maternal Body Mass Index (BMI) as a possible determinant. In addition preclinical studies from our group show piglets born by pre-labour Caesarean section (CS) to have significantly higher liver fat than those delivered vaginally (VD)(2). Overweight and obese mothers are also more likely to give birth by CS.
Aim: The aim of this observational study was to examine the effect of maternal BMI and mode of delivery on IHCL in healthy term babies, taking other potentially relevant variables (sex, birth weight, postnatal age, maternal diabetes) into account.
Methods: The study received Research Ethics approval. Infants were recruited from the postnatal wards of Chelsea and Westminster Hospital with written informed parental consent. Liver 1H magnetic resonance spectroscopy was carried out within 4 weeks of birth in natural postprandial sleep in accordance with our previously established protocol1. Data were analysed using SPSS and STATA. As IHCL distribution was positively skewed and included values of zero, these data were log-transformed after adding a calculated constant (c = 0.104).
Results: We have studied 62 mother-infant pairs. Maternal BMI ranged from 17 to 36 kg/m2. Infants were born by VD (n=40), in-labour CS (n=10) or pre-labour CS (n=12). IHCL ranged from 0 to 4.7 (median 0.41, IQR 0.06 to 0.76). As sex, birth weight and maternal diabetes (any type) had no significant effect on infant IHCL these variables were omitted from the final ANCOVA model. Postnatal age at time of assessment was significantly related to IHCL (p<0.002). The effect of maternal BMI depended on mode of delivery (effect of birth method p=0.025, effect of maternal BMI p=0.030, interaction between birth method and maternal BMI p=0.022). In infants not exposed to labour (born by pre-labour CS) there was a significant effect of maternal BMI on IHCL, with an increase of 32% (95% confidence interval 10% to 59%) per unit increase in maternal BMI. There was no evidence of an effect in infants born by in-labour CS (increase of 6% [95% CI -17% to 35%] per unit maternal BMI) or in infants born by VD (increase of 0.4% [95% CI -6% to 7%] per unit maternal BMI).
Conclusion: This study is in keeping with our previous animal studies and suggests that VD may activate pathways that protect against neonatal hepatic lipid accumulation that is driven by increasing maternal BMI. It is unknown if infant IHCL tracks though later life, though our preterm studies provide some evidence that this is the case. Our findings are of concern given the rising prevalence of overweight mothers and the rate of pre-labour CS. The positive association with postnatal age suggests that infant feeding may be also be relevant and we intend to address this possibility in future work.
References
1. Thomas et al, Arch Dis Child Fetal Neonatal Ed 2008; 93:F382-3
2. Hyde MJ et al, Clin Sci (Lond) 2010; 118:47-59