Patterns and prevalence of alcohol consumption in pregnancy using infant biomarkers

Henderson, Elizabeth (2021) Patterns and prevalence of alcohol consumption in pregnancy using infant biomarkers. MD thesis, University of Glasgow.

Abstract

Ethanol is an intoxicant widely available within the UK which has implications for the behaviour capabilities of persons who have consumed even modest amounts, as well as significant longer-term health issues. Although the sale and consumption of alcohol is subject to legislative restrictions, consumption remains widespread in the UK.

Ethanol is recognised as being teratogenic to the developing fetus; while awareness of the potential adverse consequences of alcohol consumption in pregnancy is increasing, a significant number of pregnant women continue to consume alcohol despite national advice to the contrary.

This thesis explores the known effects of in utero alcohol exposure upon the unborn child, examines methods of detecting alcohol consumption in pregnancy and describes the pattern and prevalence of prenatal alcohol consumption in the West of Scotland as measured by ethanol biomarkers in the newborn and by confidential postnatal maternal interview. The thesis also explores the potential use of routinely stored newborn blood spot cards for retrospective ascertainment of prenatal alcohol exposure (PAE).

A total of 840 mothers consented to participate in an anonymised study of randomly selected participants, 92.5% of those approached. 827 mothers completed a confidential postnatal health questionnaire, including details of alcohol consumption in pregnancy and were asked to retain their infant’s first stool (meconium) for analysis for alcohol biomarkers. Additionally, if mothers agreed and the baby’s heel bled sufficiently, an extra blood spot card was obtained on day five, coincident with routine newborn blood spot screening. In total 740 meconium samples and 668 dried blood spot cards were returned, of which 712 meconium samples and 502 dried blood spots cards were analysable. 463 recruited mothers had previously delivered a baby and 17.8% of all participants had smoked in pregnancy. 46.4% of mothers declared some alcohol consumption in pregnancy, the majority of whom did not drink after realising that they were pregnant. 114 (13.8%) mothers declared some alcohol consumption after 20 weeks’ gestation; in only three cases did this include five or more units of alcohol. Mothers who declared alcohol consumption in later pregnancy were more likely to be aged > 35 years and to identify as white British (p < 0.05).

Fatty acid ethyl esters (FAEEs) were identified in all meconium samples with a concentration ≥600 ng/g in 39.6%. There was no relationship between maternal age, body mass index, socioeconomic status as determined by postcode of residence, parity or ethnicity and the likelihood of the infant’s meconium being either negative or positive for FAEEs, as defined by a concentration ≥600 ng/g. Mothers tended to have been less likely to smoke during pregnancy within the group whose infant’s meconium was positive for FAEEs (14 vs 20%, p=0.071). Correlation between FAEEs in meconium and self-reported alcohol consumption in pregnancy was poor; of the eight mothers that reported drinking at least three units of alcohol on any one occasion beyond 20 weeks of gestation, only three had an infant whose meconium was positive for FAEEs.

Ethylglucuronide (EtG) was detected in 41.1% of meconium samples and the concentration was ≥30 ng/g in 14.5%. There was no relationship between maternal age, body mass index, socioeconomic status as determined by postcode of residence or parity and the likelihood of the infant’s meconium being either negative or positive for EtG, as defined by a concentration ≥30 ng/g. Infants whose meconium sample was positive for EtG were less likely to have a mother who identified as white British (71.3 vs 81.8%, p=0.028). When infant meconium was positive for EtG the mother tended to have been less likely to have smoked in pregnancy (13.6 vs 18.5%); this difference was not significant. Maternal self-report of any alcohol consumption in later pregnancy did not predict an infant meconium EtG concentration of >30 ng/g. When meconium samples positive for both FAEEs (≥600 ng/g) and EtG (≥30 ng/g) (n=51) were considered, there was a weakly positive correlation between the two biomarkers (Pearson’s coefficient= 0.283, p=0.044).

Phosphatidylethanol (PEth) was detectable in 262 (52%) dried blood spot samples, with concentrations ranging from 2.4 to 3991.6 ng/ml. When the infant blood spot card contained ≥8 ng/ml PEth, the mother was more likely to self-identify as white British (86.6 vs 79.1%, p=0.028) and to have smoked during pregnancy (p=0.047). Mothers of infants whose blood spot card contained ≥ 20 ng/ml PEth were younger (p=0.023), and their BMI was greater (p=0.038). They were also more likely to have smoked in pregnancy (p=0.21). There was no correlation between PEth measured in dried blood spot cards and either FAEEs or EtG measured in meconium.

Compared to self-report of alcohol consumption in pregnancy, the sensitivity of the individual biomarkers ranged from 11.6% for EtG >30 ng/g to 47.8% for a PEth concentration of >8 ng/ml, and specificities ranged from 57.1% (PEth >8 ng/ml) to 84.8% (EtG >30 ng/g). Using a combination of total FAEEs >600 ng/g and EtG >30 ng/g had a very high specificity for alcohol consumption either ever in pregnancy or only after 20 weeks’ gestation, but a very low sensitivity of 6.7 and 5.3% respectively. A combination of FAEEs <600 ng/g and EtG <30 ng/g had a positive predictive value of 87% for no alcohol consumption after 20 weeks of gestation.

It is concluded that one in seven mothers delivering in the west of Scotland has consumed alcohol in pregnancy beyond twenty weeks’ gestation. Reported amounts of alcohol consumed are likely underestimated.

This study indicates that routine use of meconium biomarkers for detection of alcohol consumption in pregnancy is not currently justified. Similarly, retrospective testing of stored dried blood spot samples taken on day five of life cannot be recommended as a measure of assessing PAE. Further work is required to explore relationships between alcohol biomarkers in meconium and longer term neurodevelopmental outcomes, as well as the role of measuring PEth in the newborn.

• Item Type: Thesis (MD)
• Qualification Level: Doctoral
• Additional Information: From research conducted in the Princess Royal Maternity, Glasgow.
• Subjects: R Medicine > RA Public aspects of medicine; R Medicine > RG Gynecology and obstetrics
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing
• Supervisor's Name: McMillan, Professor Donald, Mactier, Dr. Helen and Tappin, Professor David
• Date of Award: 2021
• Depositing User: Theses Team
• Unique ID: glathesis:2021-82316
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 09 Jul 2021 10:10
• Last Modified: 09 Jul 2021 10:32
• Thesis DOI: 10.5525/gla.thesis.82316
• URI: https://theses.gla.ac.uk/id/eprint/82316