Acetaminophen is generally considered safe during pregnancy. However, recent studies seem to link its use with some risks for the unborn. The signal has been highlighted by researchers of the University of Bristol and Cardiff, whom analyzed the long-term results of a cohort enrolled at birth (7,796 pregnancies in 1991-1992) and followed beyond school age (2015-2016) within the Avon Longitudinal Study of Parents and Children.
The expectant mothers had to answer a questionnaire to assess the exposure to acetaminophen at the 18th week of pregnancy (n=4,415, 53%), at the 32nd week of pregnancy (n=3,381, 42%) and, as a control, at the 61st months of age of the child from the mother (n=6,916, 89%) or the father (n=3,454, 84%). Information about infections, articular pains or headaches were collected, but not about the doses or the duration of the therapy. When the children reached age 7, the mothers were asked to report any behavioural problems based on the Strengths and Difficulties Questionnaire (SDQ) which consists of 5 sections (symptoms, behavioural problems, attention deficit/ hyperactivity symptoms, relation problems). Prenatal exposure during the first or second trimester was linked to an increased risk of behavioural disorders (relative risk = 1.42, 95% CI, 1.25-1.62) and attention deficit/hyperactivity (RR 1.31, 95% CI, 1.16-1.49). Exposure during the 2nd trimester was also linked to an increased risk of emotional symptoms (RR 1.29, 95% CI, 1.09-1.53) and general difficulties (RR 1.46, 95% CI, 1.21-1.77). Acetaminophen administration after birth from one of the parents did not appear to have any relevant association.
Even though acetaminophen is listed among the drugs safe to use in pregnancy, these results suggest that caution should be adopted until new evidences emerge. The results of the study allow to hypothesise a specific effect of the drug in prenatal phase,1 in accordance with what had already been found by two other cohort studies that will be described below.
A Danish study enrolled two cohorts (from 1996 to 2002), the first one to identify diagnosis of hyperkinetic disorders (HKD) and/or drugs used for attention deficit/hyperactivity (ADHD), whilst the other cohort, smaller, to identify behaviours linked to ADHD, with the mothers answering the SDQ when the child had reached age 7. The children whose mothers had been exposed to acetaminophen (n=36,187 in the first cohort and n=22,687 in the second cohort) presented an increased risk of HKD (hazard ratio 1.37, 95% CI, 1.19-1.59), of use of ADHD drugs (HR 1.29, 95% CI, 1.15-1.44) and of behaviours linkable to ADHD (RR 1.13, 95% CI, 1.01-1.27). The duration (>1 trimester) and the frequency of use of acetaminophen seem to be correlated to an increase of risk.2
A study conducted in New Zealand evaluated the association between drugs commonly used in pregnancy and the diagnosis of ADHD. Data about the usage were collected by interviewing the mothers just after their child’s birth. ADHD symptoms were evaluated through the SDQ questionnaire and the Conners’ Behavioural Rating Scale (CRS), respectively administered to parents when the child was between 7 and 11 years old and to the child at age 11. Acetaminophen was used by 49.8% of the mothers (anti-inflammatory drugs 1.3%, acetylsalicylic acid 5.3%, antacids 17.4% and antibiotics 23.5%).
No other statistically significant difference associated to any other drug emerged from the SDQ score (anti-inflammatories: -1.3, 95% CI, -3.5-0.9; acetylsalicylic acid: 0.5, 95% CI, -1.5-2.4; antacids: -0.1, 95% CI, -1.0-0.7; antibiotics: 0.5, 95% CI, -0.4-1.4) at age 7. Negative values should be interpreted as higher SDQ score for the children not exposed to anti-inflammatories and antacids during their fetal life. Children whose mothers had used acetaminophen during the pregnancy had a higher risk of ADHD, but not statistically significant at age 7 (OR 2.1, 95% CI, 0.0-5.0) and smaller at age 11, both when the questionnaire had been administered to the parents (OR 1.2, 95% CI, 0.6-2.5, and when the children answered it directly (OR 1.0, 95% CI, 0.6-1.6).3
Neither the univariate or the multivariate analysis of the SDQ score differences among the children exposed to acetaminophen have showed any statistically significant difference for any of the 5 sub scales, at age 7 or 11. In the multivariate analysis, on the other side, the score differences relative to the 3 sub scales of the CRS questionnaire resulted higher and statistically significant in regard to emotional lability: 3.0 (95% CI, 1.3-4.7) and general behavioural problems: 2.6 (95% CI, 1.0-4.2).
In another study, sub-analysis of the Norwegian Mother and Child Cohort Study, the authors evaluated the long-term effects on the neurological development after in utero exposure to acetaminophen. In particular, they observed prospectively a potential association between the prenatal exposure to acetaminophen and the psychomotor development (communication problems and motor development) and behavioural alterations in 3 years old children. Children were classified for short-term exposure (from 1 to 27 days) or long-term exposure (≥28 days). Ibuprofen was used as control drug. Principal indications reported for the long-term exposure were: headache or migraine (63.4%); backache and pelvic pain (19.5%); fever (19.5%); flu or cold (12.2%). Long-term prenatal exposure to acetaminophen has been associated to an increased risk of delayed walking (risk difference 0.26, 95% CI, 0.06-0.45), of delayed motor development (0.24, 95% CI, 0.12-0.51), of communication alterations (0.20, 95% CI, 0.01 a 0.39), behavioural alterations (0.24, 95% CI, 0.12-0.37), behavioural problems (0.14, 95% CI, 0.01-0.28).
Short term exposure has been associated to scarce motor development and delay in walking, but the effects are smaller than with long-term use. Relative risks were the following: 1.69 for behavioural problems, 1.67 for psychomotor problems and 1.51 speech impairments. The strengths of this study were: the design, the sample dimension, the adjustments for the indications (fever, infections, muscular pains, headache) and for other possible confounding factors (for example co-therapies with other drugs, maternal depression, alcohol consume). It is interesting to notice that ibuprofen has not been associated to deficit in the psychomotor development, therefore suggesting a specific effect of acetaminophen. However, usually ibuprofen is not recommended in pregnancy and is contraindicated in the third trimester, therefore it is possible for the average exposure to be shorter than for acetaminophen.4
Although the described researches hold the limits due to the observational and retrospective nature of the studies, altogether they seem to suggest that acetaminophen can actually influence the child development. It should be noticed that the increased risk was not observed for drugs of the same class, an aspect that, besides supporting the good quality of the method used, seems to suggest a drug-specific effect. The effect seems to be correlated to the duration of the exposure, therefore if the use of acetaminophen cannot be avoided during pregnancy, it is opportune to limit the therapy to the bare minimum. Further studies are however needed to confirm the evidences currently available.
Monitoring Unit of Adverse Reactions to Drugs, University Hospital of Pisa, Pharmacovigilance Centre Toscana
- JAMA Pediatr 2016;170:964-970. CDI
- JAMA Pediatr 2014;168:313-20. CDI
- PLoS One 2014;9:e108210. CDI
- Int J Epidemiol 2013;42:1702-13. CDI