The Association Between Acetaminophen and Asthma
The Association Between Acetaminophen and Asthma
Although experts' opinion is one of the humblest sources of evidence, it is worth reviewing how field experts have assessed available evidence. Like any other human beings, experts have their own myths and prejudices and are as scientific-fashion-susceptible as anybody else. These myths, prejudices and fashion trends can then influence their scientific opinions, even in the most evidence-based background. In this regard, experts can be classified, by means of their attitude to the acetaminophen hypothesis, into believers and nonbelievers. Believers are those who have accepted the large body of evidence, mostly epidemiological, but also with one large well-designed clinical trial and consider that some sort of limiting intervention is warranted. JT McBride, a pediatric pulmonologist from Youngstown, OH, USA, produced an excellent review paper published in Pediatrics in 2011 in which, after a thorough review of the available evidence, concluded that he "…recommend avoidance of acetaminophen by all children with asthma or those at risk for asthma and will work to make patients, parents and primary care providers aware of the possibility that acetaminophen is detrimental to children with asthma". ST Holgate wrote an editorial published in 2011 which showed a general acceptance of the hypothesis. In the final paragraph, he stated that "there is now overwhelming evidence establishing a link between acetaminophen and asthma…' However, he did not cross the line of overtly recommending limiting acetaminophen exposure because "… despite one trial of acetaminophen versus ibuprofen in respiratory fever in children with asthma reporting a difference in favor of increased asthma in the acetaminophen arm, the absence of a placebo arm precludes one for determining whether the difference was due to reduced asthma with ibuprofen…" The authors have already commented on the practical and ethical difficulties for a placebo arm in this setting and we can add that, not having any other alternative, the question as to whether acetaminophen increases the risk or ibuprofen decreases the risk is clinically irrelevant.
On the other hand, nonbelievers argue that the entire epidemiological association between acetaminophen and asthma is due to bias, as already commented, and that there is no need to recommend any changes to the current clinical practice. Lowe et al. support this point of view in their own study, the Melbourne Atopy Cohort Study, already discussed. Many of the epidemiological studies showing this epidemiological association have been accompanied with letters to the editors explaining these potential sources of bias. One French journal, Prescrire International, specializing in independent drug reviews, have recently published two reviews with assertive titles stating the absence of safety concerns of acetaminophen use in pregnancy and childhood. In these reviews, authored collectively by the editorial staff, all the available evidence is dismissed based on the heterogeneity of epidemiological studies and the potential sources of bias and the final conclusion is that the safety profile of acetaminophen is as good as it used to be. Heintze et al., in a recent review, also considered that the epidemiological association between acetaminophen and asthma is fully explained by confounding by indication and concluded that there is no sound reason to replace acetaminophen as the preferred pain relief and fever medication in this age group.
It is clear that the traditional, extremely good acetaminophen safety profile is acting now like a huge prejudice preventing pediatricians to accept the hypothesis that acetaminophen exposure may increase wheezing disorders prevalence and morbidity. This hypothesis was completely unexpected and the pediatric community is being somehow reluctant to accept it. In general, epidemiological and clinical results can be interpreted in two ways: the rightful and straightforward way, in which main results, analyzed as previously established in the study protocol are accepted as they are, not considering how unexpected or disappointing they may be; and what we call the desperate way, in which, not having obtained the desired or expected results, authors begin a data torturing procedure, performing a large array of different analyses not previously included in the study protocol in order to obtain some desired results. It is worth considering these two types of data interpretation in the epidemiology of wheezing disorders studies. A large number of very large well-designed epidemiological studies looking for a genetic signal explaining the familial aggregation of wheezing disorders have consistently found that the signals are very weak, inconsistent and not reproducible. This is the rightful interpretation of their results. However, due to these results being so disappointing, many desperate analyses have been proposed through extremely complicated mathematical methods in order to maintain what scientific facts stubbornly discard: that genetic factors are important in the epidemiology of asthma. These desperate genetic analyses have been very well received by the scientific community with the hope that they will eventually fulfill our prejudices about the role of genetics in asthma. The authors wish they were true.
In the same way, the results of the epidemiological and clinical studies on the effect of acetaminophen exposure have only one rightful and straightforward interpretation: acetaminophen exposure is associated with an increase in wheezing disorders prevalence and morbidity. However, these results being so unexpected and contrary to previously-held beliefs, both clinical and research pediatricians are very prone to look for and to accept alternative or, what we call, desperate analyses that discard the whole body of evidence against acetaminophen, based on the possibility of bias. We are very likely to accept these desperate analyses because, as in the case of genetic studies, we do not like what we found because it is too challenging to our previous beliefs. However, it is important to remember that, apart from tobacco smoke exposure, no other genetic or environmental factors, including genes, allergens, infections and bacterial substances, has shown the stubborn and consistent association with wheezing disorders prevalence as acetaminophen has done.
However, believers and nonbelievers all agree that more focused clinical trials are needed to assess the efficacy of limiting acetaminophen exposure interventions in the general pediatric population and in wheezing children. We all agree but only one research group has taken action to date for a new clinical trial.
Experts' Opinion & Attitudes to the Acetaminophen as a Wheezing Disorders Promoter Hypothesis
Although experts' opinion is one of the humblest sources of evidence, it is worth reviewing how field experts have assessed available evidence. Like any other human beings, experts have their own myths and prejudices and are as scientific-fashion-susceptible as anybody else. These myths, prejudices and fashion trends can then influence their scientific opinions, even in the most evidence-based background. In this regard, experts can be classified, by means of their attitude to the acetaminophen hypothesis, into believers and nonbelievers. Believers are those who have accepted the large body of evidence, mostly epidemiological, but also with one large well-designed clinical trial and consider that some sort of limiting intervention is warranted. JT McBride, a pediatric pulmonologist from Youngstown, OH, USA, produced an excellent review paper published in Pediatrics in 2011 in which, after a thorough review of the available evidence, concluded that he "…recommend avoidance of acetaminophen by all children with asthma or those at risk for asthma and will work to make patients, parents and primary care providers aware of the possibility that acetaminophen is detrimental to children with asthma". ST Holgate wrote an editorial published in 2011 which showed a general acceptance of the hypothesis. In the final paragraph, he stated that "there is now overwhelming evidence establishing a link between acetaminophen and asthma…' However, he did not cross the line of overtly recommending limiting acetaminophen exposure because "… despite one trial of acetaminophen versus ibuprofen in respiratory fever in children with asthma reporting a difference in favor of increased asthma in the acetaminophen arm, the absence of a placebo arm precludes one for determining whether the difference was due to reduced asthma with ibuprofen…" The authors have already commented on the practical and ethical difficulties for a placebo arm in this setting and we can add that, not having any other alternative, the question as to whether acetaminophen increases the risk or ibuprofen decreases the risk is clinically irrelevant.
On the other hand, nonbelievers argue that the entire epidemiological association between acetaminophen and asthma is due to bias, as already commented, and that there is no need to recommend any changes to the current clinical practice. Lowe et al. support this point of view in their own study, the Melbourne Atopy Cohort Study, already discussed. Many of the epidemiological studies showing this epidemiological association have been accompanied with letters to the editors explaining these potential sources of bias. One French journal, Prescrire International, specializing in independent drug reviews, have recently published two reviews with assertive titles stating the absence of safety concerns of acetaminophen use in pregnancy and childhood. In these reviews, authored collectively by the editorial staff, all the available evidence is dismissed based on the heterogeneity of epidemiological studies and the potential sources of bias and the final conclusion is that the safety profile of acetaminophen is as good as it used to be. Heintze et al., in a recent review, also considered that the epidemiological association between acetaminophen and asthma is fully explained by confounding by indication and concluded that there is no sound reason to replace acetaminophen as the preferred pain relief and fever medication in this age group.
It is clear that the traditional, extremely good acetaminophen safety profile is acting now like a huge prejudice preventing pediatricians to accept the hypothesis that acetaminophen exposure may increase wheezing disorders prevalence and morbidity. This hypothesis was completely unexpected and the pediatric community is being somehow reluctant to accept it. In general, epidemiological and clinical results can be interpreted in two ways: the rightful and straightforward way, in which main results, analyzed as previously established in the study protocol are accepted as they are, not considering how unexpected or disappointing they may be; and what we call the desperate way, in which, not having obtained the desired or expected results, authors begin a data torturing procedure, performing a large array of different analyses not previously included in the study protocol in order to obtain some desired results. It is worth considering these two types of data interpretation in the epidemiology of wheezing disorders studies. A large number of very large well-designed epidemiological studies looking for a genetic signal explaining the familial aggregation of wheezing disorders have consistently found that the signals are very weak, inconsistent and not reproducible. This is the rightful interpretation of their results. However, due to these results being so disappointing, many desperate analyses have been proposed through extremely complicated mathematical methods in order to maintain what scientific facts stubbornly discard: that genetic factors are important in the epidemiology of asthma. These desperate genetic analyses have been very well received by the scientific community with the hope that they will eventually fulfill our prejudices about the role of genetics in asthma. The authors wish they were true.
In the same way, the results of the epidemiological and clinical studies on the effect of acetaminophen exposure have only one rightful and straightforward interpretation: acetaminophen exposure is associated with an increase in wheezing disorders prevalence and morbidity. However, these results being so unexpected and contrary to previously-held beliefs, both clinical and research pediatricians are very prone to look for and to accept alternative or, what we call, desperate analyses that discard the whole body of evidence against acetaminophen, based on the possibility of bias. We are very likely to accept these desperate analyses because, as in the case of genetic studies, we do not like what we found because it is too challenging to our previous beliefs. However, it is important to remember that, apart from tobacco smoke exposure, no other genetic or environmental factors, including genes, allergens, infections and bacterial substances, has shown the stubborn and consistent association with wheezing disorders prevalence as acetaminophen has done.
However, believers and nonbelievers all agree that more focused clinical trials are needed to assess the efficacy of limiting acetaminophen exposure interventions in the general pediatric population and in wheezing children. We all agree but only one research group has taken action to date for a new clinical trial.