Objective The aim of this study was to investigate the prevalence of xenobiotic substances, such as caffeine, nicotine and illicit drugs (eg, cannabis and cocaine), in blood samples from first-trimester Danish pregnant women unaware of the screening.
Design A cross-sectional study examined 436 anonymised residual blood samples obtained during 2014 as part of the nationwide prenatal first-trimester screening programme. The samples were analysed by ultra performance liquid chromatography with high-resolution time-of-flight mass spectrometry.
Setting An antenatal clinic in a Danish city with 62 000 inhabitants, where >95% of pregnant women joined the screening programme.
Primary and secondary outcome measures The prevalence and patterns of caffeine, nicotine, medication and illicit drug intake during the first trimester of pregnancy.
Results The prevalence of prescription and over-the-counter drug detection was 17.9%, including acetaminophen (8.9%) and antidepressants (3.0%), of which citalopram (0.9%) was the most frequent. The prevalence of illegal drugs, indicators of smoking (nicotine/cotinine) and caffeine was 0.9%, 9.9%, and 76.4%, respectively. Only 17.4% of women had no substance identified in their sample.
Conclusions This study emphasises the need for further translational studies investigating lifestyle habits during pregnancy, as well as the underlying molecular mechanisms through which xenobiotic substances may affect placental function and fetal development.
- drug abuse
This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Statistics from Altmetric.com
Strengths and limitations of the study
The pregnant women were unaware of the screening, eliminating information and recall bias as potential sources of error, as the blood samples analysed were anonymised surplus blood samples from the prenatal screening programme.
Ultra performance liquid chromatography with high-resolution time-of-flight mass spectrometry is a broad targeting method that has been validated for the 225 most toxicologically relevant drugs and metabolites that also supplies patterns of co-exposure, as all substances are detected at once.
As the study was based on anonymous blood samples, we had no opportunity to follow-up the children prospectively.
We only had a single blood sample from the first trimester and no samples from late pregnancy, which would have provided further information by confirming exposure patterns and/or identifying changes in exposure during pregnancy.
Since the thalidomide catastrophe of the 1960s,1 in which a seemingly safe antiemetic drug caused major congenital malformations in children worldwide, focus on avoiding teratogenic exposures during pregnancy has increased. However, with the Developmental Origins of Health and Disease Hypothesis (DOHaD),2 3 the need for a deeper understanding of less-recognised threats to fetal health has become more apparent. According to the DOHaD, early pregnancy constitutes a highly sensitive period during which environmental factors can affect fetal development. The intrauterine environment is thus believed to ‘programme’ the fetus and placenta through subtle molecular changes, that is, changes of the epigenome that alter gene regulation and affect disease risk later in life.2–4 Such epigenetic alterations have already been linked to the development of many chronic diseases, including type 2 diabetes5 and cardiovascular illnesses.6 To unravel potential links between disease development later in life and epigenetic changes, a better understanding of the exposures currently affecting pregnant women is needed. In this respect, attention should be focused on all xenobiotic substances (substances not normally found in the human body, ranging from medications to nicotine or caffeine), which might constitute a potential health risk during pregnancy or later in the life of the child.
The use of prescription medications in the first trimester of pregnancy has increased >60% over the past three decades.7 Previous international studies have found that up to 94% of all pregnant women take some form of medication at least once during their pregnancy,7–9 most frequently over-the-counter drugs (66.9%), but also treatments for long-term or chronic disorders (as much as 17% of women).8 Cultural differences in the use of medicines and other potential harmful exposures are evident. In the USA, it is estimated that 17%–25% of young women of fertile age are habitual smokers.10–12 However, recent data from the Danish Health Protection Agency found that only 12% of Danish pregnant women smoke daily.13 Many adverse effects of smoking are well known,14–17 but the patterns of co-exposure between smoking and other xenobiotics are not well-described, making it difficult to track harmful combinations and describe the mechanisms involved. Additionally, the prevalence of harmful exposures such as illicit drug use during pregnancy is relatively unknown in the Danish setting, despite the fact that numerous adverse pregnancy outcomes are related to drug abuse.12 18 The latest Danish population survey from 2013 found that 46% of young Danes under the age of 35 had experimented with cannabis and approximately 9% of them had at least once tried other illegal drugs in addition to cannabis, with cocaine and amphetamines being the most commonly reported.19 An increasing use of certain forms of psychostimulants such as methylphenidate (eg, Ritalin) has also been reported among Danish pregnant women.20 Epidemiological studies have linked intrauterine exposure to attention-deficit/hyperactivity disorder (ADHD) medications (methylphenidate or atomoxetine) with lower Apgar scores at birth and an increased risk of miscarriage.21 This underscores the need for more knowledge of illicit drug use during pregnancy.
The majority of previous studies investigating the use of medication during pregnancy have been based on retrospective self-reporting or searching in national prescription registries, and may thus be affected by recall and selection bias. Hence, the aim of this study was to characterise the use of prescription-type medicine, over-the-counter medicine and psychoactive substances in the blood of first-trimester pregnant Danish women who were unaware of the screening. We aimed to investigate both the prevalence and pattern of xenobiotic substance use, such as medicines, caffeine, nicotine and illicit drugs (eg, cannabis and cocaine), in pregnant women. We also aimed to include the use of over-the-counter medications, which has not been otherwise described when using the national prescription registries.
Materials and methods
Design and study group
The study was designed as a cross-sectional study, in which the study population consisted of the pregnant women from the municipality of Randers, Denmark who had participated in the national prenatal screening programme between gestational week 8–13. As part of the screening programme, blood samples (for the double test) are always drawn and the unused portion of the samples is stored in a biobank at −80°C22 for quality control purposes. All women with residual blood in the biobank were theoretically eligible to participate in the study. However, our inclusion criterion was that the original double test analysis must have occurred during the first 8 days of each month in 2014. Each woman contributed one sample. There were no specific exclusion criteria.
All blood samples were drawn either by family general practitioners or in outpatient settings connected to Randers Regional Hospital. Serum was isolated and samples were shipped to the Department of Clinical Biochemistry, Aarhus University Hospital, within 4 hours, recentrifuged and stored at 4°C until the double test analysis could be performed, 12–36 hours after sample collection. The unused portions of the samples were stored in the biobank at −80°C22 until our analysis was performed in the spring of 2016.
Pregnant women who had participated in the prenatal screening programme and attended the first-trimester scan with risk assessment in 2014 were identified using a treatment code. Their blood samples were subsequently identified in the biobank through barcode numbers, anonymised and delivered for analysis. A total of 436 samples fitting the inclusion criterion were analysed in the present study. This corresponds to 23.5% of the pregnant women who underwent first trimester nuchal translucent scans at Randers Regional Hospital in 2014.
All data were handled and stored in a de-identified manner.
Sample and data analysis
Serum samples were thawed, and 300 µL of each serum sample were transferred to an Eppendorf plate and analysed for the presence of substances using a forensic analytical method as described by Telving et al.23 Approximately 500 different substances comprising toxic compounds, illegal and legal drugs and several of their metabolites could be identified using protein precipitation followed by ultra performance liquid chromatography with high-resolution time-of-flight mass spectrometry (UPLC-HR-TOFMS analysis).23 A list of the substances included in this method can be found in the supplementary section of the methods in the article by Telving et al.23 This method has been validated for the 225 most toxicologically relevant drugs and metabolites. The supplementary section23 also summarises the cut-off values for these substances in antemortem whole blood. The method cannot detect alcohol or all types of medications used by pregnant women, that is, only certain antiemetic drugs (eg, metoclopramide and ondansetron) and the most frequently used antihypertensive drugs (eg, amlodipine, furosemide, diltiazem and metoprolol). Furthermore, antibiotics and thyroid medications, such as levothyroxine, could not be assessed by the UPLC-HR-TOFMS analysis.
The presence of a substance and all related metabolites from each blood sample were counted as one exposure in a given sample, for example, nicotine and/or cotinine as indicators of smoking. In several cases, substances were identified solely based on the presence of their metabolites (eg, benzoylecgonine, a metabolite of cocaine).
The analysing laboratory is accredited by an external independent organisation, DANAK (Danish Accreditation), and participates in various screening and quantification proficiency tests. All results from the UPLC-HR-TOFMS analysis were evaluated by two experienced persons at the Section for Forensic Chemistry, as described by Telving et al.23
We found the prevalence of xenobiotic substances (medications, caffeine, nicotine and illicit drugs) in first-trimester Danish pregnant women to be 82.6%. Thus, only in 17.4% of the samples there was no trace of any exogenous substance detected. However, in 62% of samples, only one substance was identified, whereas 16.2% contained two substances and 3.6% of the samples contained evidence of three or more substances. Table 1 shows all the identified substances. As several of the women had traces of more than one substance, these women are counted more than once in table 1.
Over-the-counter and prescription medicine
In 17.9% (n=78) of the samples, we identified traces of medicine; a total of 27 different forms of medications were identified (see table 1 for details). The prevalence of over-the-counter medications (acetaminophen, ibuprofen, aspirin, analgesic bandages or antihistamines) was 11.9% (n=52), and the prevalence of medication available only with a prescription in Denmark was 7.1% (n=34). Samples with two or more drugs accounted for 2.1% (n=9) of all samples. A combination of over-the-counter and prescription medicine was found in 1.1% (n=5) of our samples, and the combination of medication and caffeine was present in 13.5% (n=59) of all samples. The frequencies of the different medications are given in table 1, with analgesics (11.7%) including acetaminophen (9.4%) and codeine (1.1%) being the most frequently used drug category.
Psychoactive medications (defined as antidepressants, antipsychotics, anxiolytics and methylphenidate) were identified in 3.7% of samples, with antidepressants (3.0%) being the most frequent type and citalopram (0.9%) being the drug most frequently used. Antipsychotic and anxiolytic drugs were identified in 0.7% and 0.2% of the samples, respectively. In addition, 0.5% of the samples contained antiepileptic drugs.
Antihistamines, antidiabetic drugs and the asthma medication salbutamol were also identified in 1.8%, 1.8%, and 0.2% of all samples, respectively.
We identified illicit drugs (including cannabis) and/or their metabolites in 0.9% (n=4) of the serum samples (table 1). Cannabis was identified in all four illicit drug-positive samples (0.9%). Moreover, amphetamine and benzoylecgonine (a cocaine metabolite) were also present in one of these samples.
Caffeine and nicotine
Caffeine was the most frequent substance used, identified in a total of 76.4% of the samples. Indicators of smoking (nicotine and cotinine, a metabolite of nicotine) were found in 9.9% of samples. Caffeine, nicotine and cotinine have psychoactive properties and, if considered as psychoactive substances alongside psychoactive medications and illicit drugs, as many as 79.4% of the pregnant women had used some form of psychoactive substance.
In this study, we found a high prevalence (82.6%) of xenobiotics (medicine, caffeine, nicotine and illicit drugs) in blood samples from first-trimester Danish pregnant women. To the best of our knowledge, this is the first study investigating the prevalence and pattern of xenobiotic substance use in early pregnancy in women unaware of the screening. This study was strengthened by the inclusion of samples from Danish pregnant women who generally had a very high attendance rate (95%) in prenatal screening programmes.24 Furthermore, this is the first study using this type of broad targeted substance screening on serum samples using UPLC-HR-TOFMS analysis, which can identify most relevant psychoactive medications, over-the-counter medications and illicit drugs found in the Danish population, although not the use of alcohol.23
Importantly, the cross-sectional study design only reflects the substance use within a short period of time, and it should be considered that habits may change during the pregnancy. Moreover, differences in the pharmacokinetics of various xenobiotics could also affect the likelihood of identifying a given substance. However, as the first trimester is the most sensitive and critical period in fetal development, it is critical to note that only 17.4% of pregnant women analysed in this study had no evidence of xenobiotic substances in their blood, indicating that early intervention is needed to prevent or regulate the use of medications during pregnancy. Due to ethical considerations, this analysis of samples from women unaware of the screening required anonymisation of the samples; hence, the present study is limited by the prevention of inclusion of perinatal outcomes in the analysis, which would very likely have been highly relevant in this context.
Pharmacological compounds available as over-the-counter drugs were the most frequent finding in this study. Acetaminophen was identified in a total of 9.4% of the samples. This finding raises concerns, as the use of acetaminophen during pregnancy might be associated with an increased risk of asthma,25 cryptorchidism,26 27 autism, hyperkinetic disorders and ADHD-like behaviour in children,28 29 despite not being associated with major birth defects.30 Thus, the markedly high prevalence of acetaminophen use in this and previous studies27–30emphasises the need for further research concerning the more subtle effects of acetaminophen in the human placenta and early development.
The prevalence of ibuprofen and other over-the counter drugs of the non-steroidal anti-inflammatory drug family (NSAIDs) associated with adverse outcomes, such as increased risk of miscarriage31 32 and congenital cardiac defects,33 was low (0.5%). Similarly, the frequency of antihistamine use was only 1.8%, which we consider surprisingly low as they are recommended for common conditions such as hyperemesis and allergies and have previously been reported to be widely used.34
In the present study, prescription drugs were found in 7.1% of the samples; other European studies have found a 79% prevalence of prescription drug use during pregnancy, with up to 48% of all pregnant women receiving a prescription in first trimester.7 35 This value likely reflects the design of this study, showing only a snapshot of the use of medicine among Danish pregnant women. The fact that certain types of drugs, such as antibiotics and antiemetic drugs, are not detected in the analytical method used28 also contributes to the differences between these and previously reported results. The analytical method used includes substances that are relevant in forensic cases.
The most frequent group of prescription drugs identified in our study was antidepressants, with a frequency of 3%. However, this was considerably lower than in a US study, which found that 6.6% of pregnant women used antidepressants.36 It remains unclear whether prenatal exposure to antidepressants and other psychoactive substances can impair brain development, cause postnatal neurobehavioural differences21 37 or increase the risk of miscarriage,38–40 major malformation,41–43 stillbirth and preterm birth.44
Among the prescription drugs identified, several are suspected of having adverse effects on fetal development and are thus not recommended for pregnant women (table 1). However, many of these drugs are also vital in the treatment of chronic diseases, highlighting the complexity of the safe use of medications in early pregnancy. Studies have shown that insufficient treatment of conditions such as depression,39 psychosis,45 epilepsy46 or inflammatory bowel47 disease during pregnancy can have negative effects on both the mother and child. Thus, when treating patients with psychiatric disorder or patients with other chronic diseases who are pregnant, it is important to consider the potential teratogenicity with the risk associated with the untreated diseases.39 45 47 However, a deeper understanding of lifestyle patterns and the intake of non-prescription medications is needed to limit the risks to necessary medical treatments, just as emphasis should be placed on avoiding unnecessary multidrug treatments. The use of medications among the pregnant population calls for continued optimisation of obstetric guidelines. Reflecting the use of such guidelines, metformin was only identified in 1.8% of our samples, which is in accordance with the Danish guidelines advocating discontinuation during pregnancy when used for treatment of diabetes or PCOS.
The presence of illicit drugs (0.9%) in our population appears relatively low compared with previous international studies, which have reported frequencies of up to 4.4%.12 18 A Danish study from 1998 found that 2.1% of Danish pregnant women had used cannabis and 0.2% had used other illicit drugs within the last 3 months before recognition of pregnancy.48 Naturally, some form of ‘healthy worker selection’ or change of abuse pattern prior to the doctor’s visit cannot be excluded, as cannabis is currently only approved for a severely limited number of conditions in Denmark and is presumably not used in the treatment of pregnant women. Thus, in other countries in which medical cannabis is legal, the prevalence rate among pregnant women may likely be higher.
Our findings show a high frequency of caffeine (76.4%), even though only 40% of Danish pregnant women reported drinking coffee in previous studies.49 50 This discrepancy may be explained by the intake of caffeine-containing beverages and energy drinks. Regardless of the source, caffeine may affect the fetus as it crosses the placenta51 and can cause miscarriage,52 53 fetal growth restriction54 55 and childhood leukaemia.56 Furthermore, animal studies indicate that caffeine exposure might potentiate the adverse effects of certain medications.57 58
We found evidence of (habitual) cigarette smoking in nearly 10% of the samples, despite the fact that smoking in pregnancy is a well-known risk factor59 associated with an increased risk of placenta dysfunction15 and adverse birth outcomes, such as low birth weight and preterm birth.14 60 As a previous report stated that 8% of pregnant women continue to smoke throughout pregnancy,61 this finding might reflect a continuous increase in smoking cessation during pregnancy. In recent years, prenatal exposure to cigarette smoke has also been associated with the development of diseases later in life, such as asthma and allergies,62 metabolic diseases63 and certain forms of cancer.14 Moreover, prenatal exposure to cigarette smoke has been associated with behavioural deficiencies and neuropsychiatric disorders.16 17 Many smokers reduce or quit smoking during pregnancy61; however, even short-term exposure during the first trimester may still result in adverse effects on the fetus. In this respect, first-trimester exposure to maternal cigarette smoke has been shown to reduce the number of gonadal cells in both male and female human fetuses.64 Thus, our findings underline the importance of maintaining focus on early smoking cessation during pregnancy.
With the recognised limitation of being a one-time-only analysis, this study provides unbiased knowledge regarding medication and substance abuse among Danish pregnant women seeking prenatal diagnostic advise at the end of the first trimester. The fact that almost one in five randomly selected Danish pregnant women displayed traces of medication in a single first-trimester blood sample in the present study underscores the critical need for early intervention to prevent or regulate the use of medication during pregnancy, as well as to address continuous exposure to caffeine and smoking in early pregnancy. The study emphasises the need for further studies on the mechanistic effects of xenobiotic exposures during pregnancy, as both legal and illegal drugs may be teratogenic or have adverse effects on placental function and ultimately have subtle effects on epigenetic alterations.
The authors are grateful for financial support from the Faculty of Health at Aarhus University, supporting the research year of Sissel Kramer Aagaard; the support from the Research Department at Randers Regional Hospital, the Central Denmark Region; and the financial support from The Linex Foundation, Gross L F Foght’s Foundation and Helga & Peter Korning’s Foundation.
Contributors Project design: NU, PB, IL, AL and MFA. Management of blood samples and data selection: NT. Blood analysis: SKA, MFA and RT. Data analysis: SKA under supervision of PB, IL, AL, MFA and NU. Drafting of the manuscript: SKA and ALV. Critical revision of the manuscript: PB, IL, AL, NU and MFA. All authors have commented on the manuscript.
Funding The study was supported by the Faculty of Health at Aarhus University, the Research Department at Randers Regional Hospital, The Linex Foundation, Gross L F Foght’s Foundation and Helga & Peter Korning’s Foundation.
Competing interests None declared.
Patient consent Detail has been removed from this case description/these case descriptions to ensure anonymity. The editors and reviewers have seen the detailed information available and are satisfied that the information backs up the case the authors are making.
Ethics approval The study was approved by the Central Denmark Region Committee on Biomedical Research Ethics (1-10-72-22-16) and the Danish Data Protection Agency (1-16-02-255-16).
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement We have no unpublished data. All data are presented in the article.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.