The Association Between Acetaminophen and Asthma

Should Its Pediatric Use Be Banned?

Antonio Martinez-Gimeno; Luis García-Marcos

Expert Rev Resp Med. 2013;7(2):113-122.

Weakness of the Case Against Acetaminophen as a Wheezing Disorders Promoter

Epidemiological association, however strong, does not guarantee a causal relationship. Moreover, there still remains the possibility that the whole epidemiological association between acetaminophen and wheezing disorders is just the product of large sources of bias, which could explain entirely the overwhelming epidemiological evidence against acetaminophen. Should this be the case, acetaminophen could be an innocent bystander falsely accused of something actually produced by some factor to which acetaminophen is associated. Three different types of explanatory bias have been proposed.

First, confounding by indication, in which the real culprit would be infections, mainly viral and respiratory in nature, which would trigger simultaneously the use of acetaminophen and the real causal relationship producing asthma. This source of bias has been defended, among others, by Lowe et al.,^[28] supported by their results in the Melbourne Atopy Cohort Study,^[4] in which the statistically significant OR against acetaminophen initially found in a first crude analysis disappeared after controlling for respiratory tract infections and stratifying for indication, not being significant for nonfever use of acetaminophen. The authors suggest that this would have happened in other studies if they had controlled for respiratory tract infections. Schnabel et al., in a prospective population-based birth cohort study, found that more than 95% of the participant children had been exposed to acetaminophen during the first year of life.^[29] Those with a diagnosis of asthma had had more months with a course of acetaminophen during the first year of life compared with those without the diagnosis of asthma, but this difference was explained by a larger number of respiratory infections. The number of acetaminophen courses due to nonrespiratory causes was similar in children with or without asthma. The authors concluded that respiratory infections and not acetaminophen use were associated with asthma development.

However, Kreiner-Møller et al., in a prospective birth cohort study of newborns from asthmatic mothers, found that the epidemiological association between infant acetaminophen use and early asthmatic symptoms remained after controlling for respiratory infections.^[30] Therefore, confounding by indication has not always been found when the research protocol has looked for it.

Second, confounding by reverse causation, asthma itself being the origin of symptoms that prompt acetaminophen use, creating a sort of vicious circle: asthma induces acetaminophen use, which in turn apparently (but falsely) induces asthma. This view is supported by some studies that have found that asthmatic children suffer more comorbidities prone to the use of acetaminophen, like headaches and low-back pain.^[31]

Finally, preferential use of acetaminophen by children at greater risk of asthma due to fear that ibuprofen might increase asthma symptoms. Table 2 offers an overview of the arguments in the case against acetaminophen as a promoter of wheezing disorders in children.

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Next Section

Abstract and Introduction
The Case Against Acetaminophen as a Wheezing Disorders Promoter
Epidemiological Evidence
Evidence From Randomized Clinical Trials
Biological Plausibility
Review of the Bradford Hill Criteria
Weakness of the Case Against Acetaminophen as a Wheezing Disorders Promoter
The Question of Acetaminophen & Wheezing Disorders in the Light of Evidence-based Medicine
Should the Pediatric Use of Acetaminophen Be Banned?
Theoretical Effect Size of Limiting Acetaminophen Exposure
Experts' Opinion & Attitudes to the Acetaminophen as a Wheezing Disorders Promoter Hypothesis
Conclusion
Expert Commentary
Five-year View
References
Sidebar

Table 1. Bradford Hill criteria applied to the hypothesis of acetaminophen as wheezing diseases promoter.
Criteria	Evidence
Strength of effect	Increased risk after acetaminophen exposure ≥1.5
Dose response	Present for acetaminophen exposure in utero, childhood and adult life
Consistency/coherence	Consistency among studies, age groups, sex, geography and culture
Exposure before response	Observed in prospective studies of acetaminophen exposure in utero, infancy and adult life Observed in the only prospective double-blind randomized clinical trial in children
Biological plausibility	Increased oxidant-induced airway inflammation Enhanced Th2 response Decreased immune response to rhinovirus infection
Removal of exposure prevents disease	Partially examined in the only double-blind randomized clinical trial on the subject published to date: using ibuprofen instead of acetaminophen in febrile children with concomitant use of anti-asthma drugs reduced the risk for an outpatient visit for asthma in the following month Effect of removal of exposure on prevalence to be determined
Specificity	No increased risk of asthma associated with aspirin or other NSAID
Analogy	Weak, protective effects of antioxidant diet in asthma

Adapted with permission from [17].

Table 2. Overview of arguments in the case against acetaminophen as a promoter of wheezing disorders in children.
Arguments that support a detrimental effect of acetaminophen in asthma	Arguments against a detrimental effect of acetaminophen in asthma
Epidemiologic association between acetaminophen exposure and asthma
Strength of the association Consistency of the association across geography, culture, age and studies Dose dependency of the epidemiologic association Coincidence in time of increase in asthma prevalence and increase in acetaminophen use Ecologic relationship among asthma prevalence and per-capita sales of acetaminophen Uniqueness of this association, more consistent than any other environmental factor	The whole epidemiologic association may be spurious, acetaminophen being an innocent bystander falsely accused due to several bias or confounding factors: Confounding by indication (real culprit, respiratory virus infections producing fever) Confounding by reverse causation (real culprit, asthma itself which induces acetaminophen use) Confounding by preferential use of acetaminophen by children at higher risk of asthma or already asthmatics
Clinical trials
A randomized clinical trial with 1879 children 6 months to 12 years old with asthma exacerbation and fever, showing a lower incidence of outpatient visits for asthma in those randomly assigned to ibuprofen compared with acetaminophen (adjusted relative risk: 0.56 [95% CI: 0.34–0.95] compared with those receiving acetaminophen)	No placebo group included. The lower relative risk for children who received ibuprofen could be due to a protective effect of ibuprofen instead of a detrimental effect of acetaminophen No clinical trials have been performed on the effect of acetaminophen on the prevalence of asthma
Biological plausibility
Acetaminophen depletes glutathione from the airways Acetaminophen reduces immune response to, and prolongs, rhinovirus infection