Phosphatidylethanol in Maternal or Neonatal Blood to Detect Alcohol Exposure during Pregnancy: A Systematic Review

Background: Alcohol consumption during pregnancy, even at low doses, may damage the fetus. Pregnant women tend to underreport their alcohol consumption generating the need for sensitive and specific biomarkers, among which PEth has emerged due to its high specificity and possibility to be measured in both maternal and neonatal blood. The aim of this study is to systematically review the latest 20 years of literature for depicting the state of the art, the limitations, and the prospects of PEth for estimating alcohol consumption during pregnancy. Materials and methods: A systematic search, adhering to PRISMA guidelines, of the latest 20 years of literature through “MeSH” and “free-text” protocols in the databases PubMed, SCOPUS, and Web of Science, with time limits 1 January 2002–1 March 2022, was performed. The inclusion criteria were as follows: PEth used for detecting alcohol consumption during pregnancy, quantified in blood through liquid chromatography coupled to mass spectrometry, and full texts in the English language. Opinion papers, editorials, and narrative reviews were excluded. Results: Sixteen (16) papers were included in the present review (0.81% of total retrieved records). All the included records were original articles, of which there were seven prospective cohort/longitudinal studies, six cross-sectional studies, two observational-descriptive studies, and one retrospective study. All studies assayed PEth in at least one biological matrix; seven (7) studies quantified PEth in maternal blood, seven studies in newborn blood, and only two studies in both maternal and neonatal blood. In several included papers, PEth proved more sensitive than self-reports for identifying pregnant women with an active alcohol intake with the diagnostic efficiency improving with the increase of the maternal alcohol intake. Conclusions: Further studies, performed on wider and well-stratified populations, are needed to drive any definitive conclusion. PEth is a promising marker for monitoring alcohol use in pregnancy; however, at the present time, its use is still limited mainly by the absence of a globally agreed interpretative cut-off, the paucity of data regarding its specificity/sensitivity, and the lack of standardization on the diagnostic efficiency of the different isoforms.


Introduction
Prenatal alcohol exposure (PAE) can cause birth defects and lifelong neurocognitive deficits in affected children collectively termed "Fetal Alcohol Spectrum Disorders" (FASD) [1,2]. One of the ongoing challenges for an early and accurate diagnosis of FASD is the difficulty of assessing whether a mother drank alcohol during her pregnancy [3].
Alcohol can cross the placenta, with the fetal blood alcohol level tending to become similar to the maternal one, with a prolonged exposure due to slower elimination and PubMed and Web of Science: ("Phosphatidylethanol" [Supplementary Concept] OR PEth OR Phosphatidylethanol) AND (forensic OR legal OR biomarker OR marker OR alcohol abuse OR abstinence OR monitoring"), while a slightly modified search string was used for Scopus: "ALL ((Phosphatidylethanol OR PEth) AND (forensic OR legal OR biomarker OR marker OR alcohol AND abuse OR abstinence OR monitoring))". Paper selection was conducted independently by three authors (LF, AC and MP), based on titles and abstracts. The following inclusion and exclusion criteria were adopted.
Inclusion criteria A. Titles and abstracts available in the English language. B. PEth used for detecting alcohol consumption during pregnancy. C. PEth quantified in liquid human blood or in dried blood spots through liquid chromatography coupled to mass spectrometry. D. Full text in the English language.
Exclusion criteria E. Opinion papers, editorials, and narrative reviews without novel data.
Papers not fulfilling at least one of the A-D criteria and fulfilling the exclusion criterium E were excluded. In case of doubtful classification based on title and abstract, the full text was retrieved. Any discrepancy in paper selection process was settled by consensus discussion performed by four authors (LF, AC, MP and GV).
Data extraction was conducted independently by three authors (LF, CG and GC) and the data extracted from the studies were collected in a table by two authors (AC and MP), while another author (GV) verified the accuracy of the data extraction process, in order to minimize subjective judgment. The following items were collected from each study: authors, journal and year, features of the study (type of study, subjects involved, main aims, clinical setting, and inclusion and exclusion criteria), characteristics of the investigated population (numbers of subjects and race/ethnicity), methods for estimating alcohol use, analytical method used for PEth analysis, type of measured PEth and concentration, type of collected sample, timing of collection, and main results obtained (sensitivity, specificity, positive predictive value, and negative predictive value). Any discrepancies in the data extraction process were settled by consensus discussion performed by four authors (LF, AC, MP and GV).

Results and Discussion
As reported in the PRISMA flow-chart (Figure 1), the combined search on the databases PubMed, Web of Science, and Scopus retrieved 1969 records, 562 of which were eliminated being duplicates. Of the 1407 records screened by title and abstract, 1226 were excluded based on criteria A and B. The remaining 181 papers were analyzed in full text with 165 records excluded based on criteria C, D, and E. Sixteen (16) papers (0.81% of the total records) were included in the present review.
The data extracted from the sixteen (16) included papers are presented in detail in Table 1.
Although the search was extended to the last 20 years, only papers published in the last decade fulfilled all the inclusion criteria and were therefore analyzed in this review; in particular, it was noticed that half of the included articles (8 out of 16) were published in the last 4 years [15,17,18,[22][23][24]26,27]. This result reflects the growing interest of the scientific community in PEth as a potential marker of alcohol use during pregnancy. The data extracted from the sixteen (16) included papers are presented in detai Table 1.
Although the search was extended to the last 20 years, only papers published in last decade fulfilled all the inclusion criteria and were therefore analyzed in this revi in particular, it was noticed that half of the included articles (8 out of 16) were publis in the last 4 years [15,17,18,[22][23][24]26,27]. This result reflects the growing interest of scientific community in PEth as a potential marker of alcohol use during pregnancy.

Main Aims of the Included Studies
All the included studies, although with slightly different secondary objectives, had a main aim of the identification of alcohol consumption during pregnancy. Eight out of the sixteen included papers assessed maternal alcohol consumption during pregnancy; Di Battista et al. [26], for example, analyzed PEth in neonatal blood without any comparison with the mother's declaration on alcohol intake during pregnancy.

Reported Alcohol Intake
In three included studies, the reported daily alcohol intake of the mother was compared not only with the maternal blood PEth concentration, but also with the neonatal blood PEth concentration after birth [1,7,22].
In 5 of the 16 included records, the assessment of alcohol consumption during pregnancy was performed using a standardized questionnaire. Four different types of questionnaires were employed:  [17].
In three included records, only self-reporting, without any structured questionnaires, has been employed for reconstructing maternal alcohol exposure during pregnancy [13,18,20].    Although misreporting or under-reporting due to shame or other social reasons always remains an issue, employing standardized questionnaires should help in minimizing these problems, and could increase the comparability of the collected data. The high heterogeneity encountered in the included records suggests an urgent need for an international effort of standardization, with the hopeful creation of a dedicated structured questionnaire for collecting alcohol use/abuse data during pregnancy.

Isoforms of PEth
As well-known, numerous molecular species of PEth can be detected and quantified in blood by liquid chromatography coupled to mass spectrometry with a lack of consensus in the scientific community on their diagnostic efficiency in different clinical and forensic settings. In recent years, PEth 16:0/18:1 and PEth 16:0/18:0 have been increasingly used for the identification of chronic alcohol abuse or dependence and for monitoring abstinence.
In the included records (Figure 2), the following major isoforms have been quantified:

Units of Measurement
Regarding the different units of measurement of PEth molecular species, the 19 2010 literature generally used moles, whereas more recent studies (2010-2022) favored use of fractions of weight over volume (e.g., ng/mL or µg/L).

Interpretative Cut-Offs and Sensitivity/Specificity of PEth
Another issue regards the identification of an interpretative cut-off or thresh capable of discriminating with high selectivity an active alcohol intake during pregna (i.e., ingestion of alcoholic beverages) from an unintentional exposure to minute amou of ethanol contained, for example, in food (e.g., cakes with liquor and fermentation fruit) [29][30][31].
In the included studies, most authors interpreted total PEth concentration basing

Units of Measurement
Regarding the different units of measurement of PEth molecular species, the 1990-2010 literature generally used moles, whereas more recent studies (2010-2022) favored the use of fractions of weight over volume (e.g., ng/mL or µg/L).
In the present review, it was noticed that 10 included studies expressed PEth concentrations in weight over volume fractions [1,[15][16][17][18][21][22][23][24][25], whereas 6 papers used moles [7,13,20,[26][27][28]. The latter unit of measurement requires the knowledge of the molecular weight of each quantified PEth isoform, making it difficult to compare studies that utilize different molecular species and impossible to convert total PEth concentration without detailed information on the concentration of each isoform.

Interpretative Cut-Offs and Sensitivity/Specificity of PEth
Another issue regards the identification of an interpretative cut-off or threshold capable of discriminating with high selectivity an active alcohol intake during pregnancy (i.e., ingestion of alcoholic beverages) from an unintentional exposure to minute amounts of ethanol contained, for example, in food (e.g., cakes with liquor and fermentation of fruit) [29][30][31].
Therefore, in contrast to the general population where several authors proposed different PEth thresholds to differentiate teetotalers from social or heavy drinkers, in this specific field of research there is still a lack of studies examining interpretative PEth cut-offs to discriminate if and how much a woman drinks ethanol during pregnancy.
Consequently, the diagnostic efficiency in terms of sensitivity and specificity for detecting and quantifying alcohol use during pregnancy has still to be systematically assessed. In the included studies, only Bakhireva et al. [1] reported that PEth exhibited low sensitivity (22.2%) and high specificity (100%) even for low alcohol exposures during pregnancy.
Hence, to better analyze the extracted data, the included records were divided into three categories:
In six [7,13,16,18,20,27] out of seven papers, PEth quantification was performed in maternal blood collected from pregnant women attending to an antenatal care center; differently, Raggio et al. [15] examined pregnant and unpregnant women referring for HIV therapy.
Blood samples were not collected at the same time of pregnancy in all the included records: • in three records sample collection was performed during the first trimester [13,16,20]; • in one record before the 15th week of gestational age [18]; • in one record between the 16th-18th week of gestational age [7]; • in one record two blood withdrawals were performed [27]; the first at the 12th week of gestational age and the second one at the 24th week of gestational age; • in one record before the 34th week of gestational age [15].
Only in two out of the seven records analyzed here, populations were stratified into subgroups on the basis of the women's self-reported alcohol consumption before and during pregnancy [7,20]. In detail, Comasco et al. [7] administered the same questionnaire (AUDIT) in two different time of pregnancy: the first regarded pre-pregnancy alcohol consumption, administered between the 16th and the 18th gestational week, and the second one regarded alcohol consumption during pregnancy, being administered at the 32nd gestational week. In all cases, PEth was under the cut-off of 0.1 µmol/L, while the AUDIT score suggested that a significant number of women continued consuming alcohol during pregnancy. Another type of stratification, based on weekly alcohol self-report (A.U./week), was performed by Kwak et al. [20], where the population involved was divided into teetotalers (0 A.U./week), light drinkers (≤3 A.U./week), moderate drinkers (3-7 A.U./week), and heavy drinkers (≥7 A.U./week). The authors concluded that all women (100%) who declared a heavy alcohol consumption (≥7 A.U./week) had PEth concentrations above the LOQ (limit of quantification); PEth sensitivity proportionally decreased with the reduction of the declared alcohol intake.
Breunis et al. [18] found that 44.4% of the cases with a positive blood PEth result exhibited a PEth concentration above the lower limit of detection (LOD) but below the lower limit of quantitation (LOQ). These results may suggest that a further improvement of the analytical performance of the methods used for blood PEth quantitation could favor an improvement of the diagnostic sensitivity of PEth in maternal blood for detecting an active alcohol exposure during pregnancy.
Nowadays, despite that the employment of PEth for monitoring alcohol consumption in pregnancy still has numerous limitations and the data obtained until now are incomplete, some authors have already begun to use PEth as a screening tool in the clinical practice; in detail, a study performed in Norway [27] analyzed blood PEth concentration in the same samples collected for Rhesus typing, which is routinely performed in all pregnancies around the 12th gestational week. In the above study, the authors found out that 1.4% of the included cases had a positive PEth sample around the 12th gestational week, whereas only 0.4% of the included cases had a positive sample around the 24th gestational week.

PEth in Neonatal Blood
In the United States, 2-5% of school-aged children are estimated to be affected by FASD, but the prevalence of PAE might be substantially underreported [32]; therefore, an accurate detection of PAE in newborns might offer the opportunity for an early identification of children at risk for future neurocognitive disorders, allowing for an early intervention to prevent or reduce long-term consequences.
Besides other direct ethanol metabolites measured in the meconium (e.g., FAEEs, EtG and EtS [33,34]), in recent years, PEth has been proposed as a very promising biomarker of PAE, being measurable both in whole venous blood and DBS.
All the included seven papers regarding PEth in neonatal blood have used DBS for blood collection. DBS can be obtained from venous or capillary blood, being minimally invasive, easy, and cheap to collect, storage, and transport, giving the opportunity of a potential integration with routine newborn screening for metabolic diseases [1,22].
In one of the included papers [23], PEth was quantified in blood collected from the umbilical cord immediately after birth, whereas the other six papers used DBS from heel capillary blood [17,21,[24][25][26]28].
Unfortunately, only two out of the seven papers belonging to this section provided data on maternal alcohol exposure; Baldwin et al. [21] did not correlate neonatal blood PEth concentration to the maternal alcohol intake during pregnancy, whereas Stevens et al. [17] highlighted that about 60% of the included neonates, which were born from women exposed to ethanol during pregnancy, had a capillary blood PEth concentration above the analytical threshold (8 ng/mL). Therefore, none of the included papers tested the diagnostic efficiency of PEth in neonatal blood to reconstruct the alcohol consumption of the mother during pregnancy.
Umer et al. [24] and Yang et al. [28] reported a potential correlation between high neonatal blood PEth levels and low birth weight, preterm birth, and increased risk of miscarriage, although these observations did not reach any statistical significance.
Stevens et al. [17] reported that, if the pregnancy is unplanned, there is an increased risk of moderate to heavy alcohol exposure in the early stages of pregnancy, as women became aware of their pregnancy later.

PEth in Both Maternal and Neonatal Blood
Two (2) out of the sixteen included papers quantified PEth in both maternal and neonatal blood.
Bakhireva et al. [1] divided the included pregnant women into two groups (i.e., moderate alcohol consumption versus low/absent alcohol intake) based on their AUDIT scores and on the in-depth TFLB calendar assessments and compared total PEth concentration in maternal venous blood to total PEth concentration in neonatal DBS. Neonatal PEth demonstrated a higher sensitivity (32.1%) than maternal PEth (22.2%) for detecting moderate alcohol consumption.
Baldwin et al. [22] included 611 pregnant women from Uruguay and 524 from Brazil and compared maternal to neonatal blood PEth concentrations.
The authors found out that the infants had significantly higher blood PEth concentrations than their mothers. This phenomenon has not yet been explained and more studies are needed to understand its molecular origin. A possible explanation could be that the ethanol ingested by the mother, crossing the placenta and reaching blood concentrations similar to the maternal ones, accumulates in the amniotic fluid (due to its slow turnover) with a consequent longer exposure of fetal red blood cells to alcohol and a higher production of PEth [4].
Only little evidence has been published on neonatal blood PEth concentrations and, at the present time, it is unclear if dosing PEth both in maternal and neonatal blood could enhance the diagnostic efficiency of the marker for detecting an alcohol exposure during pregnancy.

Conclusions
Alcohol consumption during pregnancy, even at low doses, may damage the fetus. Pregnant women tend to underreport their alcohol consumption out of shame or social stigma generating the need for sensitive and specific biomarkers capable of identifying any alcohol use during pregnancy in order to promote educational programs and counselling interventions.
In the recent literature, among the proposed biomarkers, PEth has emerged due to its high specificity and possibility to be quantified in both maternal and neonatal blood, also using dried blood spots (DBS) collected for routine screenings.
As reported in the present review, only few studies (16), where ethanol exposure during pregnancy was not always collected through structured questionnaires, and with a non-homogeneous stratification of the included populations, have been published on this topic until now.
In several included papers, PEth has proven more sensitive than self-reports for identifying pregnant women with an active alcohol intake with the diagnostic efficiency of the marker improving with the increase of the maternal alcohol intake. Probably, a further improvement of PEth diagnostic efficiency might be reached by improving the analytical performance of the methods used for its quantitation in blood (i.e., liquid chromatography coupled to mass spectrometry). Moreover, in order to implement PEth analysis in the assessment of PAE, more data about a worldwide established interpretative cut-off, its certain sensitivity and specificity, and the diagnostic efficiency of the different isoforms are still needed.
Further studies, performed on wider and well-stratified populations, are still needed to verify if PEth concentration in maternal and/or neonatal blood could be capable of identifying even minimal ethanol intakes during the first trimester.