ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH Vol. 41, No.

5
May 2017

Prevalence of Prenatal Alcohol Exposure in the State of

Texas as Assessed by Phosphatidylethanol in Newborn
Dried Blood Spot Specimens
Ludmila N. Bakhireva , Janet Sharkis, Shikhar Shrestha, Tristan J. Miranda-Sohrabji,
Sonnie Williams, and Rajesh C. Miranda

Background: While 2 to 5% of school-aged children in the United States are estimated to be affected
by fetal alcohol spectrum disorders (FASD), the prevalence of prenatal alcohol exposure (PAE) might
be substantially underreported. Our objective was to systematically estimate the prevalence of PAE in
Texas by measuring a direct ethanol metabolite, phosphatidylethanol (PEth), in 1,000 infant residual
dried blood spots (irDBSs) in the Texas Newborn Screening Repository.
Methods: All public health regions (PHRs) were represented proportional to their 2014 birth rate
(~0.25% of total births). A cross-sectional study design (unit of observation: individual irDBS cards/
infants) with additional ecologic subanalysis (unit of observation: aggregate measures for each Texas
PHR) was utilized. The study used PEth-irDBS to estimate the prevalence of PAE within 1 month
before delivery for the state of Texas and each Texas PHR. The ecologic subanalysis compared different
geographical regions’ aggregate prevalence of PAE with (i) retail liquor licenses, (ii) median household
income by PHR, and (iii) prevalence of birth outcomes commonly associated with FASD.
Results: The sample included an equal number of males and females; 47.8% non-Hispanic White,
40.8% Hispanic, 6.6% African American, and 4.8% Asian infants. In the entire sample, 8.4% of
irDBSs were positive for PEth (>20 ng/ml) indicative of PAE within approximately 1 month before
delivery. Large regional differences were observed with mostly urban, high median-income regions
demonstrating the highest prevalence.
Conclusions: Results of this first systematic statewide PAE prevalence study demonstrate that PAE
might be more prevalent than previously thought. Active case ascertainment efforts for FASD coupled
with systematic objective assessment of PAE should expand to the national level to better estimate
public health needs required to provide adequate services for children affected by PAE.
Key Words: Alcohol, Prevalence, Infants, Biomarkers, Phosphatidylethanol, Surveillance, Dried
Blood Spots.

P RENATAL ALCOHOL EXPOSURE (PAE) is the

leading nongenetic cause for neurocognitive disabilities
in the United States and worldwide (Riley et al., 2011).
PAE-related disabilities are collectively termed fetal alcohol
spectrum disorders (FASD). A recent meta-analysis esti-
mated that the worldwide prevalence of FASD is approxi-
mately 22.7 per 1,000 (Roozen et al., 2016). Active case
ascertainment studies in the United States estimate that 1.1
From the Department of Pharmacy Practice and Administrative
Sciences (LNB, SS, SW), University of New Mexico College of Phar-
macy, Albuquerque, New Mexico; Department of Family and Community
Medicine (LNB), University of New Mexico, Albuquerque, New
Mexico; Texas Office for Prevention of Developmental Disabilities (JS,
TJM-S), Austin, Texas; and Department of Neuroscience and Experi-
mental Therapeutics (RCM), Texas A&M Health Science Center,
College of Medicine, Bryan, Texas.
Received for publication December 29, 2016; accepted March 7, 2017.
Reprint requests: Ludmila Bakhireva, MD, PhD, MPH, University of
New Mexico College of Pharmacy, 1 University of New Mexico, MSC 09
5360, Albuquerque, NM 87131; Tel.: 505-272-2545; Fax: 505-272-6749;
E-mail: lbakhireva@salud.unm.edu
Copyright © 2017 by the Research Society on Alcoholism.
DOI: 10.1111/acer.13375
1004
to 4.8% of school-aged children may have FASD (May
et al., 2009, 2014, 2015). Although FASD has received much
less attention compared to other neurodevelopmental dis-
abilities, early detection of PAE must be treated as a signifi-
cant public health interest, as the primary and secondary
behavioral (Baer et al., 2003; Howell et al., 2006), mental
(Famy et al., 1998; O’Connor and Paley, 2009), and physical
(Chen and Nyomba, 2004; Dobson et al., 2012; Gray et al.,
2008, 2010; Parkington et al., 2014) health consequences of
PAE impose substantial healthcare and economic costs on
families and society (Lupton et al., 2004; Popova et al.,
2012). A recent study demonstrated that as many as 86.5%
of adolescents with FASD were undiagnosed or misdiag-
nosed prior to evaluation (Chasnoff et al., 2015). This high-
lights a sizable missed opportunity to provide early
interventions that prevent secondary conditions associated
with undiagnosed or untreated FASD (Thanh et al., 2013).
The challenges with accurate early FASD diagnosis are mul-
tifactorial, but broadly fall under 3 main categories: (i) lack
of reliable information on PAE; (ii) lack of physical features
in many individuals with FASD; and (iii) lack of healthcare
professionals trained in FASD assessment. For children
without the physical features of full or partial fetal alcohol
Alcohol Clin Exp Res, Vol 41, No 5, 2017: pp 1004–1011
SYSTEMATIC ASSESSMENT OF PAE IN TEXAS
syndrome, confirmation of PAE is required to make a diag-
nosis (Hoyme et al., 2016). However, this is often difficult to
obtain due to the stigmatization of alcohol use in pregnancy,
recall bias, and the paucity of information about the biologi-
cal mother among children in the welfare system. While some
self-reported measures of PAE, such as the Timeline Follow-
back procedure (Sobell and Sobell, 1992), are known to pro-
vide good estimates of PAE, they are typically administered
only in research settings and are time-consuming. An alter-
native approach to obtaining an objective measure of PAE
focuses on analysis of ethanol (EtOH) biomarkers in biologi-
cal tissues of either a pregnant woman or a child (Bakhireva
and Savage, 2011).
Biomarker assays for exposure focus on measuring the
presence of unique metabolites of alcohol, such as fatty
acid ethyl esters in meconium (Peterson et al., 2008), ethyl
glucuronide in placenta (Matlow et al., 2012), urine, nails,
or hair samples (Berger et al., 2014; Gutierrez et al., 2015;
Walsham and Sherwood, 2014), and phosphatidylethanol
(PEth) sampled in infant blood (Bakhireva et al., 2013).
While placental, meconium, hair, and nail samples provide
the advantage of longer detection windows, obtaining such
specimens for statewide assessments of PAE is technically
challenging and expensive. However, infant blood from
every live birth is routinely collected on a Guthrie card in
the United States. The infant residual dried blood spots
(irDBSs) are often stored in state repositories (storage
duration varies by state) and, therefore, can be used for
statewide assessments of prenatal exposures and disease
burdens.
PEth, a lipid metabolite of alcohol localized to erythro-
cyte cell membranes in blood, is recognized as a specific
and persistent blood biomarker for alcohol exposure and
is used to monitor compliance with treatment programs
for adults with alcohol use disorders (Isaksson et al.,
2011). PEth is not affected by liver disorders (Stewart
et al., 2009) or postcollection synthesis in dried specimens
after exposure to EtOH vapor in laboratory settings (Jones
et al., 2011). We have demonstrated the feasibility and
cost-effectiveness of measuring PEth in irDBS (Bakhireva
et al., 2013), high specificity of this biomarker (Bakhireva
et al., 2014), and its acceptable stability in banked DBS
cards (Bakhireva et al., 2016). We are aware of only 2
prior studies that have attempted to assess prevalence of
PAE in specific populations by PEth-irDBS (Bakhireva
et al., 2013; Baldwin et al., 2015). Both studies had rela-
tively small sample sizes (201 to 250 irDBSs), and sam-
pling was not representative of the statewide population.
The goal of this study is to estimate prevalence of PAE in
Texas, which in 2014 accounted for ~10% of recorded
births in the United States (Hamilton et al., 2015), and
assess variability in PAE prevalence among the 11 Texas
public health regions (PHRs). Additionally, the association
between demographic factors, regional ecological factors
(median income, availability of alcohol, prevalence of pre-
term births, and low birthweight [LBW]), and PAE
1005
prevalence was evaluated to provide additional insight and
context for the study.
MATERIALS AND METHODS
Study Design
This study began in 2015 as a collaboration between the
University of New Mexico, the Texas Office for Prevention of
Developmental Disabilities (TOPDD), the Texas Department of
State Health Services (TxDSHS), and Texas A&M University.
All research activities were approved by the Institutional Review
Boards at each of these organizations (except for the TOPDD,
which defers to the review board of the TxDSHS). All informa-
tion associated with the irDBS samples was deidentified; no
protected health information was obtained. A cross-sectional
study design (unit of observation: individual irDBS cards/in-
fants) with additional ecologic subanalysis (unit of observation:
aggregate measures for each Texas PHR) was utilized. The
study used PEth-irDBS to estimate the prevalence of PAE for
the state of Texas and each Texas PHR. The ecologic subanaly-
sis compared different geographical regions’ aggregate preva-
lence of PAE with (i) retail liquor licenses (TABC, 2015), (ii)
median household income by PHR (Texas Association of Coun-
ties, 2014), and (iii) prevalence of birth outcomes commonly
associated with FASD, such as LBW and preterm birth. To
this end, we studied the number of LBW versus total number
of live births in 2014, and the prevalence of preterm births
(number of preterm births vs. total live births in 2014). Data
on births by PHR were obtained from TxDSHS (TxDSHS,
2014).
Stratified random sampling was used to select 1,000 irDBS cards
(1 spot per infant) from the Texas Newborn Screening Repository.
The target sample size of 1,000 irDBSs constituted ~0.25% of the
total births in Texas based on the 2014 year (TxDSHS, 2014). The
number of irDBS cards from each PHR was proportionate to the
2014 birth rate in each region. The following eligibility criteria were
applied to irDBS samples: (i) from the newborn screening (first
48 hours of life); (ii) stored for ≤2 months (at room temperature) to
minimize potential degradation of PEth; (iii) from diverse racial/eth-
nic groups in the state and both sexes; and (iv) from each PHR, pro-
portional to the birth rate. Each sample had a unique study ID and
was accompanied by the following information: infant sex, race/eth-
nicity, and PHR of the birth hospital. Cards were shipped in batches
of 50 on dry ice to the U.S. Drug Testing Laboratory (USDTL; Des
Plaines, IL) for PEth analysis.
PEth Analysis
Levels of the 16:0/18:1 isoform of PEth in irDBS were assessed at
a contract laboratory by standard liquid chromatography–tandem
mass spectrometry (LC-MS/MS) protocols, as previously described
(Bakhireva et al., 2013; Jones et al., 2011). Briefly, 3 standardized
punches were obtained from each irDBS. PEth was extracted with
methanol, reconstituted into a mobile phase, and analyzed by LC-
MS/MS (Agilent Zorbax Eclipse Plus C-8 column, with an Agilent
Technologies 6460 tandem mass spectrometric detector and electro-
spray ionization in the negative mode; Agilent Technologies, Santa
Clara, CA). Sample PEth content was quantified relative to a stan-
dard curve of known amounts of PEth in blood spots. As previously
published, the limit of detection (LOD; lowest quantity of a sub-
stance that can be detected) was set at 2.0 ng/ml, and the limit of
quantitation (LOQ; lowest quantity of the analyte that can be reli-
ably quantified) was set at 8.0 ng/ml. All intra- and interassay accu-
racy determinations were within 10.8% of target concentration, and
the intra- and interassay precision calculations were <8.7% (Jones
1006 BAKHIREVA ET AL.

et al., 2011). The contract laboratory, USDTL, was blinded to sam-

ple identity.

Statistical Analyses
Descriptive statistics (means, frequencies) were used to summa-
rize the data. PEth > 20 ng/ml was categorized as “positive” for
PAE, as suggested by the USDTL and used in previous research
(Bakhireva et al., 2014). A more conservative cutoff concentration
of PEth > 28 ng/ml was also utilized, as proposed in other studies
(Helander et al., 2012). Overall state and PHR-specific prevalence
rates were estimated. Association between infant characteristics
(sex, race/ethnicity, PHR of the birth hospital) and positive PEth
results was examined by chi-square test and logistic regression in
univariate and multivariable analyses, respectively. For analysis
purposes, PHRs 2, 4, 5N, 5S, 9, and 10 were categorized as “mostly
rural”; 3, 6, 7, and 8 as “mostly urban”; and 1 and 11 as “mixed
urban and rural” as per guidelines provided by the TOPDD.
For the ecologic analysis, 11 PHRs were taken as the unit of anal-
ysis. Median income, preterm birth rate, rate of LBW infants, and
number of liquor licenses per 10,000 people were taken as ecological
indicators. The proportion of positive samples in each PHR was
determined and merged with the ecological data. Correlation and
regression analyses were performed with the proportion of positive
PEth samples (using the cutoff >20 ng/ml) as the dependent variable
and the ecological indicators as the independent variables; the
regression line was fitted, and the proportion of variability in the
dependent variable explained by the independent variables (r2) was Fig. 1. PEth distribution in the sample (n = 1,000).
estimated. All analyses were performed in SAS 9.3 (Cary, NC) and
STATA (College Station, TX).
estimated at 8.4%. Using a more stringent cutoff of >28 ng/
ml (Helander et al., 2012), PAE prevalence was 6.3%.
RESULTS Approximately 1.7% of the total sample showed levels of
The sample included an equal number of male and female PEth greater than 100 ng/ml indicative of heavy PAE.
infants (50%/50%) and had representation from major The highest prevalence of positive PEth samples (17.7%
racial/ethnic groups. The majority of irDBSs were obtained that exceeded the 20 ng/ml, and 13.8% that exceeded the
from birth hospitals located in mostly urban areas (72.2%), 28 ng/ml criterion) was observed in PHR-3 (which includes
while 27.8% were from mostly rural or mixed regions the Dallas metropolitan area). The second highest preva-
(Table 1). The distribution of PEth values in the overall sam- lence, 8.4% at PEth > 20 and 5.6% at PEth > 28 criteria,
ple is shown in Fig. 1. The mean PEth concentration was was observed in PHR-6, which encompasses the Houston
10.2  37.1 ng/ml (range: <LOD to 711 ng/ml; median: metropolitan area. In contrast, another relatively high popu-
<LOD). In this sample, 44.5 and 24.7% had PEth values lation density area, PHR-11, which encompasses Corpus
above the LOD and LOQ, respectively. Using a cutoff con- Christi and Laredo, showed a much lower prevalence (2.0%
centration of >20 ng/ml (i.e., >10-fold above the LOD for at PEth > 20 and 2.0% at PEth > 28). In some low–popula-
the assay, as suggested by USDTL), prevalence of PAE was tion-density regions, such as PHR-9, none of the proportion-
ally sampled irDBSs met the criterion for being identified as
PEth positive (Fig. 2).
Table 1. Demographic Characteristics of Newborns in the Sample In univariate analyses, no association was observed
(n = 1,000) between positive PEth results (>20 ng/ml) and infant sex
(p = 0.17). However, a significant association was observed
Characteristics N (%) between race/ethnicity and positive PEth results (p = 0.002).
Gender Specifically, a higher proportion of Hispanic (12.3%) and
Male 500 (50.0) African American (9.1%) infants were positive compared to
Female 500 (50.0)
Race/ethnicity
White (5.7%) and Asian (2.1%) infants. In addition, a much
White, non-Hispanic 478 (47.8) higher prevalence of positive PEth results was observed in
African American 66 (6.6) cards obtained from predominantly urban areas (10.4%)
Hispanic 408 (40.8)
Asian 48 (4.8)
compared to predominantly rural (4.2%) or mixed (2.2%)
Location regions (p = 0.001). In multivariable analyses, participants
Mostly urban 722 (72.2) with Hispanic ethnicity were 3 times more likely to have posi-
Mixed 135 (13.5)
Mostly rural 143 (14.3)
tive PEth results (OR = 3.03; 95% CI: 1.84 to 4.99) after
adjusting for infant sex and PHR. Participants in mostly
SYSTEMATIC ASSESSMENT OF PAE IN TEXAS 1007

Table 2. Predictors of Positive Phosphatidylethanol (>20 ng/ml) Test

Results

Predictors OR (95% CI) p

Female versus male 0.74 (0.46 to 1.14) 0.21
Ethnicity
White 1.00 (–) –
African American 1.55 (0.61 to 3.92) 0.36
Hispanic 3.03 (1.84 to 4.99) <0.001
Asian 0.33 (0.04 to 2.49) 0.283
Public health region
Mostly rural 1.00 (–)
Mostly urban 3.04 (1.29 to 7.20) 0.011
Mixed urban and rural 0.39 (0.09 to 1.60) 0.19

Fig. 2. Prevalence of positive PEth (>20 ng/ml) by Texas public health

regions (n = 1,000).

urban regions were also 3 times more likely to have positive

PEth results (odds ratio [OR] = 3.04; 95% confidence inter-
val [CI]: 1.29 to 7.20) after adjusting for race/ethnicity and
sex (Table 2).
Results of ecologic analyses on aggregate data per PHR
demonstrated a weak nonsignificant positive association
between median income and positive PEth (r = 0.28,
r2 = 0.08, p = 0.41; Fig. 3). No association was observed
between the number of liquor licenses per PHR, adjusted for
population density, and positive PEth (r = 0.04; p = 0.912;
data not shown). Similarly, no association was observed
between the prevalence of preterm delivery (<37 weeks) per
PHR and positive PEth (r = 0.34; p = 0.298) or LBW and
PEth (r = 0.16; p = 0.64; data not shown).
DISCUSSION
To our knowledge, this is the first study to objectively
assess PAE across a large, multiethnic, high–birth-rate state,
with proportional representation by region and demographic
composition using a unique metabolite of EtOH, PEth, in
infant blood samples. Our results reveal that, depending on
the criterion used, 6.3 to 8.4% of irDBSs were positive for
PEth, indicative of PAE within approximately 1 month
before delivery. It should be noted that the overall prevalence
of PAE is likely to be much higher than these estimates as
PEth-irDBS can only detect exposure during approximately
1 month before delivery and many pregnant women substan-
tially decrease their alcohol consumption upon pregnancy
recognition (Ethen et al., 2009).
These prevalence estimates are in accord with our earlier
study of irDBS specimens, obtained from a deidentified sam-
ple of infants born at the University of New Mexico hospital
in Albuquerque, NM, where 6.5% of infants tested positive
(>20 ng/ml) for PEth (Bakhireva et al., 2013). However, a
Fig. 3. Association between median income and positive PEth results
by public health region (PHR): ecologic analysis
study of 250 irDBS cards obtained from a Midwestern state
revealed much lower prevalence (4% of irDBS with
PEth ≥ 8 ng/ml and 1.6% with PEth > 20 ng/ml) (Baldwin
et al., 2015). A variety of technical reasons related to
extended sample storage periods (Bakhireva et al., 2016;
Baldwin et al., 2015), and possible small-sample sampling
biases may have contributed to lower estimates in the Mid-
western state. However, these data also leave open the possi-
bility for wide regional variations in the risk for PAE,
emphasizing the need for systematic state surveillance using
multiple objective measures. Our data, coupled with recent
active case ascertainment FASD reports (May et al., 2009,
2014, 2015), highlight the possibility that PAE/FASD preva-
lence may equal or exceed that of other common develop-
mental disabilities, including autism spectrum disorders
(CDC, 2014).
To interpret our results, a logical question is, “what level
of alcohol consumption can PEth detect, and what is the
temporal window of detectability (i.e., the interval between
the last drinking episode and elevated PEth levels)?” We
should acknowledge that a precise answer to this question is
not feasible given the ethical considerations preventing a ran-
domized controlled trial in pregnant women. Thus,
1008
extrapolations need to be made from observational studies
or randomized controlled trials involving healthy, nonpreg-
nant populations. A small randomized controlled trial
among 27 healthy volunteers, who received either 0.25 or
0.5 g/kg of EtOH in experimental fashion, demonstrated
that 81.5% of subjects had detectable PEth levels (Javors,
2016). These findings suggest that PEth in liquid blood speci-
mens can detect consumption of 1 to 3 standard drinks in
most subjects. The half-life of PEth was estimated at
4.6  3.5 days. In another study with 16 volunteers who
ingested a single dose of alcohol leading to blood alcohol
concentration of 1 g/kg, PEth was detectable for up to
12 days, with PEth concentrations ranging from 37.2 to
122 ng/ml (Schr€ ock et al., 2016). Given that metabolism of
PEth in newborns is largely unknown, the uncertainties
regarding PEth pharmacokinetics/dynamics after lower
levels of alcohol consumption, and the between-subject vari-
ability observed with adult populations, we suggest that our
estimated prevalence rates should be interpreted as being
indicative of “any” alcohol exposure approximately a month
within delivery, rather than a specific level of PAE.
Statewide self-report-based estimates on the prevalence of
“any” drinking during pregnancy range from 8.5%
(SAMHSA, 2013), 10.2% (Tan et al., 2015), to as high as
30.3% (Ethen et al., 2009). Similarly, research estimates
range from 5.3% (Derauf et al., 2003) to 21.3% (Slater et al.,
2011). Conventionally, it is thought that these high estimates
are driven by drinking in early gestation, prior to pregnancy
recognition. However, across the United States, an estimated
3.7% (SAMHSA, 2013) to 7.9% (CDC, 2011, Ethen et al.,
2009) of pregnant women report alcohol use in the third tri-
mester. In Texas, the CDC-sponsored PRAMS survey indi-
cated that 6.1% (CDC, 2010) of women reported consuming
alcohol during the third trimester. PAE estimates from
PEth-irDBS analyses in our study (6.3 to 8.4%) are some-
what higher than state estimates relying on maternal self-
report, but are generally comparable.
A surprising finding of this study was much higher preva-
lence of PAE in the Dallas metropolitan area (PHR-3).
Recent data indicate that low- and high-income levels, but
not middle-income levels, may be associated with a greater
likelihood of drinking during pregnancy (Lepper et al.,
2016). It is possible that regional differences in affluence play
a role in the variable PAE levels observed across Texas
PHRs. Our ecologic analysis suggests a trend for association
between higher income and PEth-positive results. Although
the association did not reach statistical significance over all
PHRs, PHR-3 had the highest median income in the state.
While the sample size was proportional to the birth rate in
each PHR, low prevalence of PAE in regions with a sample
size of <50 should be interpreted with caution. Future efforts
should focus on oversampling from those regions to generate
more accurate PAE estimates.
Higher prevalence of positive PEth was observed in mostly
urban regions and among Hispanic infants. It should be rec-
ognized that the Hispanic/Latino ethnic designation
BAKHIREVA ET AL.
comprises a very heterogeneous group, and prior research
(Bakhireva et al., 2009; Caetano et al., 2012; Vaeth et al.,
2012) indicates that acculturation is associated with riskier
alcohol consumption behavior. While our results should be
interpreted with caution and warrant further investigation,
they advance the possibility of a hitherto unsuspected vulner-
ability for excessive alcohol consumption among women of
childbearing age within this ethnic group. Additionally, we
recognize that this study had a limited number of available
patient characteristics (limited to gender, PHR, and race,
due to the deidentified nature of the data), and association
with these factors needs to be interpreted with caution. It is
important to note that the ecological factors described (in-
cluding other sociodemographic information) were not
drawn specifically from the sample, but from their general
geographical location, and should thus be viewed from this
perspective. Finally, our sample had a slightly lower propor-
tion of African Americans (6.6%) compared to the 2015
U.S. Census for Texas (12.5%), potentially limiting our abil-
ity to further examine racial/ethnic differences (United States
Census Bureau, 2010-2015).
The routine assessment of PAE by PEth-irDBS requires
consideration of the possible limitations associated with this
method. First, in legal and forensic contexts, PEth is most
often assessed in liquid whole blood; however, previous stud-
ies have shown high correlation in PEth measured in DBS
and liquid specimens (Faller et al., 2011), minimal influence
of hematocrit, minimal effect due to samples obtained from
venous versus capillary blood, and no significant differences
in PEth concentrations from peripherally versus centrally
located punches from the DBS (Kummer, 2016). Second, the
irDBS cards used for this analysis were stored for up to
2 months prior to analysis, which might result in a very small
amount of PEth degradation (Bakhireva et al., 2016). How-
ever, this would imply some potential underestimation of the
PAE prevalence. Advantages of PEth analysis in irDBS
include minimal invasiveness, ease of storage and transporta-
tion, lower cost associated with processing specimens
(Bakhireva et al., 2013), and lack of postcollection synthesis
of PEth in dried blood specimens minimizing the possibility
of false-positive results (Jones et al., 2011).
There is some uncertainty regarding cutoff concentrations
of PEth. While PEth is well established as a biomarker for
adult alcohol consumption, it is a relatively new biomarker
for PAE. The cutoff of 20 ng/ml, proposed by USDTL, rep-
resents a 10-fold higher value than the LOD and is 2.5-fold
higher than the LOQ for the assay (Armbruster and Pry,
2008). It is therefore viewed as a conservative cutoff. Further,
the cutoff of 0.10 lmol/l (roughly equivalent to 28 ng/ml)
has been proposed as even more conservative, for the identi-
fication of “regular drinking” (Helander et al., 2012). While
the majority of positive samples in this study recorded levels
of PEth < 100 ng/ml, 1.7% of the total sample exhibited
PEth concentrations greater than 100 ng/ml, likely indicat-
ing heavy and perhaps persistent PAE. Of note, a majority of
this group, ~76%, were from densely populated PHR-3 and
SYSTEMATIC ASSESSMENT OF PAE IN TEXAS
PHR-8. Given that it is a relatively new biomarker, the
appropriate cutoff indicative of PAE needs to be addressed
in future studies. It should be emphasized, though, that any
detectable level of direct EtOH metabolites in a newborn is
concerning, as no amount of PAE is viewed as safe (Williams
and Smith, 2015).
A pertinent question, which, to our knowledge, has not
yet been systematically examined, is whether medications
containing EtOH, given during the first 24 to 48 hours after
birth, result in detectable PEth levels, and therefore false-
positive results. Studies with adult volunteers have shown
that PEth typically increases 8 hours after alcohol adminis-
tration and that the equivalent of 1 to 3 standard alcohol-
containing drinks needs to be consumed to result in
detectable PEth levels (Javors, 2016; Schr€ ock et al., 2016). It
is unlikely that medications containing alcohol would
achieve such blood alcohol concentrations in the newborn.
Furthermore, our ecologic analyses demonstrating the lack
of association between prevalence of preterm birth and LBW
(infants more likely to have medications administered early
in life) and PEth results is reassuring. Nevertheless, the possi-
ble contribution of perinatal medication regimens to PEth
levels in irDBS merits explicit examination in future studies.
Any type of screening for highly stigmatized conditions
is fraught with ethical challenges (Zizzo et al., 2013). Fac-
tors such as fear of prosecution, child care needs, financial
stress, and social stigma may present significant barriers
for even the most motivated woman struggling with alco-
hol dependence to seek treatment during or after preg-
nancy. The perception that alcohol use around pregnancy
is rare and that it is a “failure” of the mother may further
exacerbate this problem. While detailed analysis of ethical
issues surrounding PAE biomarkers is beyond the scope of
this report, this should be carefully considered from a mul-
tistakeholder perspective before PEth-irDBS analysis is
considered on a broader scale.
In summary, to our knowledge, this is the first report to
examine geographical differences in the statewide prevalence
of PAE using an objective assessment method and the largest
sample size to date. While prevalence of FASD approaches
or even surpasses other neurodevelopmental disorders,
FASD receives a much lower level of attention, potentially
due to a widely held perception that gestational alcohol use
is rare or occurs only early in pregnancy. Lack of systematic
objective data on PAE coupled with stigma associated with
alcohol use in pregnancy makes the topic appear of lower
public health significance than other pediatric conditions.
Emerging recent data on the prevalence of PAE and FASD,
including data presented in this report, indicate otherwise.
ACKNOWLEDGMENTS
We are thankful to Laura Garrison, MA, Steven Bishop,
MS, and Sandra Cano, MA, for their assistance with the
data collection, management, and analysis, and for coordi-
nating research activities. We are in debt to the staff at the
1009
Texas DSHS Newborn Screening Repository and to Leah
Davies, MS, at TOPDD for their assistance with the study
design and implementation. Authors do not have any con-
flict of interests to declare.
FUNDING
This research was supported by contracts from TOPDD
and TxDSHS (2016-048842-001).
DISCLAIMER
Opinions expressed in this manuscript do not necessarily
represent official positions of the Texas Department of State
Health Services (TxDSHS) or the Texas Office for Preven-
tion of Developmental Disabilities (TOPDD).
REFERENCES
Armbruster DA, Pry T (2008) Limit of blank, limit of detection and limit of
quantitation. Clin Biochem Rev 29(Suppl 1):S49–S52.
Baer JS, Sampson PD, Barr HM, Connor PD, Streissguth AP (2003) A 21-
year longitudinal analysis of the effects of prenatal alcohol exposure on
young adult drinking. Arch Gen Psychiatry 60:377–385.
Bakhireva LN, Leeman L, Savich RD, Cano S, Gutierrez H, Savage DD,
Rayburn WF (2014) The validity of phosphatidylethanol in dried blood
spots of newborns for the identification of prenatal alcohol exposure.
Alcohol Clin Exp Res 38:1078–1085.
Bakhireva LN, Savage DD (2011) Focus on: biomarkers of fetal alcohol
exposure and fetal alcohol effects. Alcohol Res Health 34:56–63.
Bakhireva LN, Savich RD, Raisch DW, Cano S, Annett RD, Leeman L,
Garg M, Goff C, Savage DD (2013) The feasibility and cost of neonatal
screening for prenatal alcohol exposure by measuring phosphatidylethanol
in dried blood spots. Alcohol Clin Exp Res 37:1008–1015.
Bakhireva LN, Shrestha S, Gutierrez HL, Berry M, Schmitt C, Sarangarm D
(2016) Stability of phosphatidylethanol in dry blood spot cards. Alcohol
Alcohol 51:275–280.
Bakhireva LN, Young BN, Dalen J, Phelan ST, Rayburn WF (2009) Peri-
conceptional binge drinking and acculturation among pregnant Latinas in
New Mexico. Alcohol 43:475–481.
Baldwin AE, Jones J, Jones M, Plate C, Lewis D (2015) Retrospective assess-
ment of prenatal alcohol exposure by detection of phosphatidylethanol in
stored dried blood spot cards: an objective method for determining preva-
lence rates of alcohol consumption during pregnancy. Int J Alcohol Drug
Res 4:131–137.
Berger L, Fendrich M, Jones J, Fuhrmann D, Plate C, Lewis D (2014) Ethyl
glucuronide in hair and fingernails as a long-term alcohol biomarker.
Addiction 109:425–431.
Caetano R, Mills B, Vaeth PA (2012) Alcohol consumption and binge drink-
ing among U.S.-Mexico border and non-border Mexican Americans.
Alcohol Clin Exp Res 36:677–685.
CDC (2010) PRAMStat | Reproductive Health | CDC. [online] Available at:
https://www.cdc.gov/prams/pramstat/index.html. Accessed November 28,
2016.
CDC (2011) PRAMStat | Reproductive Health | CDC. [online] Available at:
https://www.cdc.gov/prams/pramstat/index.html. Accessed November 28,
2016.
CDC (2014) Prevalence of autism spectrum disorder among children aged
8 years – autism and developmental disabilities monitoring network, 11
sites, United States, 2010. MMWR Surveill Summ 63:1–21.
Chasnoff IJ, Wells AM, King L (2015) Misdiagnosis and missed diagnoses in
foster and adopted children with prenatal alcohol exposure. Pediatrics
135:264–270.
1010
Chen L, Nyomba BL (2004) Whole body insulin resistance in rat offspring of
mothers consuming alcohol during pregnancy or lactation: comparing pre-
natal and postnatal exposure. J Appl Physiol (1985) 96:167–172.
Derauf C, Katz AR, Easa D (2003) Agreement between maternal self-
reported ethanol intake and tobacco use during pregnancy and meconium
assays for fatty acid ethyl esters and cotinine. Am J Epidemiol 158:705–
709.
Dobson CC, Mongillo DL, Brien DC, Stepita R, Poklewska-Koziell M,
Winterborn A, Holloway AC, Brien JF, Reynolds JN (2012) Chronic pre-
natal ethanol exposure increases adiposity and disrupts pancreatic mor-
phology in adult guinea pig offspring. Nutr Diabetes 2:e57.
Ethen MK, Ramadhani TA, Scheuerle AE, Canfield MA, Wyszynski DF,
Druschel CM, Romitti PA (2009) Alcohol consumption by women before
and during pregnancy. Matern Child Health J 13:274–285.
Faller A, Richter B, Kluge M, Koenig P, Seitz HK, Thierauf A, Gnann H,
Winkler M, Mattern R, Skopp G (2011) LC-MS/MS analysis of phos-
phatidylethanol in dried blood spots versus conventional blood specimens.
Anal Bioanal Chem 401:1163–1166.
Famy C, Streissguth AP, Unis AS (1998) Mental illness in adults with fetal
alcohol syndrome or fetal alcohol effects. Am J Psychiatry 155:552–554.
Gray SP, Denton KM, Cullen-Mcewen L, Bertram JF, Moritz KM (2010)
Prenatal exposure to alcohol reduces nephron number and raises blood
pressure in progeny. J Am Soc Nephrol 21:1891–1902.
Gray SP, Kenna K, Bertram JF, Hoy WE, Yan EB, Bocking AD, Brien JF,
Walker DW, Harding R, Moritz KM (2008) Repeated ethanol exposure
during late gestation decreases nephron endowment in fetal sheep. Am J
Physiol Regul Integr Comp Physiol 295:R568–R574.
Gutierrez HL, Hund L, Shrestha S, Rayburn WF, Leeman L, Savage DD,
Bakhireva LN (2015) Ethylglucuronide in maternal hair as a biomarker of
prenatal alcohol exposure. Alcohol 49:617–623.
Hamilton BE, Martin JA, Osterman MJ, Curtin SC, Matthews TJ (2015)
Births: final data for 2014. Natl Vital Stat Rep 64:1–64.
Helander A, Peter O, Zheng Y (2012) Monitoring of the alcohol biomarkers
PEth, CDT and EtG/EtS in an outpatient treatment setting. Alcohol Alco-
hol 47:552–557.
Howell KK, Lynch ME, Platzman KA, Smith GH, Coles CD (2006) Prena-
tal alcohol exposure and ability, academic achievement, and school func-
tioning in adolescence: a longitudinal follow-up. J Pediatr Psychol 31:116–
126.
Hoyme HE, Kalberg WO, Elliott AJ, Blankenship J, Buckley D, Marais
AS, Manning MA, Robinson LK, Adam MP, Abdul-Rahman O, Jew-
ett T, Coles CD, Chambers C, Jones KL, Adnams CM, Shah PE, Riley
EP, Charness ME, Warren KR, May PA (2016) Updated clinical guide-
lines for diagnosing fetal alcohol spectrum disorders. Pediatrics 138:
e20154256.
Isaksson A, Walther L, Hansson T, Andersson A, Alling C (2011) Phos-
phatidylethanol in blood (B-PEth): a marker for alcohol use and abuse.
Drug Test Anal 3:195–200.
Javors MA (2016) Characterization of the pharmacokinetics of phos-
phatidylethanol 16:0/18:1 and 16:0/18:2 in human whole blood after alco-
hol consumption in a clinical laboratory study. Alcohol Clin Exp Res
40:1228.
Jones J, Jones M, Plate C, Lewis D (2011) The detection of 1-palmitoyl-2-
oleoyl-sn-glycero-3-phosphoethanol in human dried blood spots. Anal
Methods 3:1101–1106.
Kummer N (2016) Quantification of phosphatidylethanol 16:0/18:1, 18:1/
18:1, and 16:0/16:0 in venous blood and venous and capillary dried blood
spots from patients in alcohol withdrawal and control volunteers. Anal
Bioanal Chem 408:825–838.
Lepper LET, Lluka A, Mayer A, Patel N, Salas J, Xaverius PK, Kramer B
(2016) Socioeconomic status, alcohol use, and pregnancy intention in a
national sample of women. Prev Sci 17:24–31.
Lupton C, Burd L, Harwood R (2004) Cost of fetal alcohol spectrum disor-
ders. Am J Med Genet C Semin Med Genet 127C:42–50.
Matlow JN, Aleksa K, Lubetsky A, Koren G (2012) The detection and quan-
tification of ethyl glucuronide in placental tissue and placental perfusate
by headspace solid-phase microextraction coupled with gas
BAKHIREVA ET AL.
chromatography-mass spectrometry. J Popul Ther Clin Pharmacol 19:
e473–e482.
May PA, Baete A, Russo J, Elliott AJ, Blankenship J, Kalberg WO, Buckley
D, Brooks M, Hasken J, Abdul-Rahman O, Adam MP, Robinson LK,
Manning M, Hoyme HE (2014) Prevalence and characteristics of fetal
alcohol spectrum disorders. Pediatrics 134:855–866.
May PA, Gossage JP, Kalberg WO, Robinson LK, Buckley D, Manning M,
Hoyme HE (2009) Prevalence and epidemiologic characteristics of FASD
from various research methods with an emphasis on recent in-school stud-
ies. Dev Disabil Res Rev 15:176–192.
May PA, Keaster C, Bozeman R, Goodover J, Blankenship J, Kalberg WO,
Buckley D, Brooks M, Hasken J, Gossage JP, Robinson LK, Manning M,
Hoyme HE (2015) Prevalence and characteristics of fetal alcohol syn-
drome and partial fetal alcohol syndrome in a Rocky Mountain Region
City. Drug Alcohol Depend 155:118–127.
O’Connor MJ, Paley B (2009) Psychiatric conditions associated with prena-
tal alcohol exposure. Dev Disabil Res Rev 15:225–234.
Parkington HC, Kenna KR, Sozo F, Coleman HA, Bocking A, Brien JF,
Harding R, Walker DW, Morley R, Tare M (2014) Maternal alcohol con-
sumption in pregnancy enhances arterial stiffness and alters vasodilator
function that varies between vascular beds in fetal sheep. J Physiol
592:2591–2603.
Peterson J, Kirchner HL, Xue W, Minnes S, Singer LT, Bearer CF (2008)
Fatty acid ethyl esters in meconium are associated with poorer neurodevel-
opmental outcomes to two years of age. J Pediatr 152:788–792.
Popova S, Lange S, Burd L, Rehm J (2012) Health care burden and cost
associated with fetal alcohol syndrome: based on official Canadian data.
PLoS One 7:e43024.
Riley EP, Infante MA, Warren KR (2011) Fetal alcohol spectrum disorders:
an overview. Neuropsychol Rev 21:73–80.
Roozen S, Peters GJ, Kok G, Townend D, Nijhuis J, Curfs L (2016) World-
wide prevalence of fetal alcohol spectrum disorders: a systematic literature
review including meta-analysis. Alcohol Clin Exp Res 40:18–32.
SAMHSA (2013) Data spotlight: the NSDUH report. [online] Available at:
http://www.samhsa.gov/data/sites/default/files/spot123-pregnancy-alc
ohol-2013/spot123-pregnancy-alcohol-2013.pdf. Accessed November 28,
2016.
Schr€ock A, Thierauf-Emberger A, Sch€ urch S, Weinmann W (2017) Phos-
phatidylethanol (PEth) detected in blood for 3 to 12 days after single con-
sumption of alcohol—a drinking study with 16 volunteers. Int J Legal
Med 131:153–160.
Slater ME, Linabery AM, Blair CK, Spector LG, Heerema NA, Robison
LL, Ross JA (2011) Maternal prenatal cigarette, alcohol and illicit drug
use and risk of infant leukaemia: a report from the Children’s Oncology
Group. Paediatr Perinat Epidemiol 25:559–565.
Sobell LC, Sobell MB (1992) Timeline follow-back: a technique for assessing
self-reported ethanol consumption, in Measuring Alcohol Consumption:
Psychosocial and Biochemical Methods (Allen J, Litten RZ eds), pp. 41–72.
Humana Press, Totowa, NJ.
Stewart SH, Reuben A, Brzezinski WA, Koch DG, Basile J, Randall PK,
Miller PM (2009) Preliminary evaluation of phosphatidylethanol and alco-
hol consumption in patients with liver disease and hypertension. Alcohol
Alcohol 44:464–467.
TABC (2015) Roster. [online] Available at: http://www.tabc.texas.gov/Public
Inquiry/RosterSummary.aspx. Accessed December 5, 2016.
Tan CH, Denny CH, Cheal NE, Sniezek JE, Kanny D (2015) Alcohol use
and binge drinking among women of childbearing age – United States,
2011–2013. MMWR Morb Mortal Wkly Rep 64:1042–1046.
Texas Association of Counties (2014) Texas counties: median household
income. [online] Available at: http://www.txcip.org/tac/census/morecoun
tyinfo.php?MORE=1013. Accessed December 10, 2016.
Thanh NX, Moffatt J, Jacobs P, Chuck AW, Jonsson E (2013) Potential
impacts of the Alberta fetal alcohol spectrum disorder service networks on
secondary disabilities: a cost-benefit analysis. J Popul Ther Clin Pharmacol
20:e193–e200.
TxDSHS (2014) Vital statistics annual reports. [online] Available at: http://
www.dshs.texas.gov/chs/vstat/annrpts.shtm. Accessed December 5, 2016.
SYSTEMATIC ASSESSMENT OF PAE IN TEXAS
United States Census Bureau (2010-2015) Quick facts: Texas. [online] Avail-
able at: http://www.census.gov/quickfacts/table/PST045215/48. Accessed
December 1, 2016.
Vaeth PA, Caetano R, Rodriguez LA (2012) The Hispanic Americans Base-
line Alcohol Survey (HABLAS): the association between acculturation,
birthplace and alcohol consumption across Hispanic national groups.
Addict Behav 37:1029–1037.
1011
Walsham NE, Sherwood RA (2014) Ethyl glucuronide and ethyl sulfate.
Adv Clin Chem 67:47–71.
Williams JF, Smith VC (2015) Fetal alcohol spectrum disorders. Pediatrics
136:e1395–e1406.
Zizzo N, di Pietro N, Green C, Reynolds J, Bell E, Racine E (2013) Com-
ments and reflections on ethics in screening for biomarkers of prenatal
alcohol exposure. Alcohol Clin Exp Res 37:1451–1455.