Drug-induced
yawning: A review of the french
pharmacovigilance database
Sommet A, Desplas M, Lapeyre-Mestre M,
Montastruc JL.
The French Network of
Pharmacovigilance Centers.
Background Yawning is a complex
stereotyped behaviour, the physiological
function of which remains unknown. It can occur
in different physiological (i.e. sedation,
hunger, hypoglycaemia) or pathological
(neurological, infectious metabolic,
psychiatric) circumstances. Several
neurotransmitters are involved, such as
dopamine, acetyicholine, serotonin or peptides.
However, drugs that induce such behaviour in
humans remain badly identified. In this article,
we investigate the observations of 'yawning' as
an adverse drug reaction (ADR), reported in the
French Pharmacovigilance Database-up to December
2004.
Methods The French Pharmacovigilance
System, established in 1973, consists of a
network of 31 regional centres of
pharmacovigilance. The French Pharmacovigilance
Database was established in 1985 to register
spontaneous reporting of ADRs. Drug causality
assessment is carried out by the French
imputability method according to the time to
onset and course of reaction, risk factors and
screening for other causes. The score of
imputability is classified into five levels:
'unlikely', 'possible', 'probable', 'likely' and
'certain'.
Results From 1985 to December 2004,
28 reports of drug-induced yawning were
registered in the French Pharmacovigilance
Database from a total of 173 055 reports. The
sex ratio of the patients documented in the
reports was 1.5 and the mean age was 46.2 years
(range 2-78 years). Several underlying diseases
were observed: depression and Parkinson's
disease, psychosis, epilepsy, pharmacodependance
and arterial hypertension. For each observation
others reasons for yawning (hypoglycaemia,
metabolic or neurological diseases) were
excluded.
The 28 reports involved a total of 38 drugs.
Twenty-six different drugs were involved. The
majority (n = 12) of reports involved serotonin
reuptake inhibitors, followed by dopaniinergic
agents (n = 7), opioids (n = 4) benzodiazepines
(n = 4) and sodium channel inhibitors (n, = 4)
drugs. Concerning opioids, one observation
reported yawning occurring after morphine
withdrawal, another case occurrèdafter an
abuse of buprefibrphine in a patient already
treated by methadone and another case was
observed after intoxication in a young child
caused by a cough mixture, which contained
dextromethorphan.
The ADR was 'unexpected' for most of the
drugs (n = 29). Imputability was rated
'possible' for 35 drugs, 'probable' for
zuclopenthixol and lidocaine and 'likely' for
only one drug (sertraline).
Four ADRs were rated as 'serious', leading
to hospitalisation (one patient taking
fluoxetine and metoclopramide, one taking
dextromethorphan, one taking lidocaine and one
taking methylergometrine).
The delay of occurrence largely varied from
30 minutes to several months after the
introduction of the drug. For example, several
months were necessary between introduction of
dopaniinergic drugs and the occurrence of
yawning. This delay was longer for
benzodiazepines: 1 month to several years.
When the suspected drugs were stopped, a
positive dechallenge was observed most of the
time: this was observed in 18 cases for 21 drugs
withdrawn. The ADR disappeared without the
implicated drug being stopped in five
observations; the implicated drugs here were
dopaniinergic agents (levodopa (1), selegiline
(1), ropinirole (1), methadone (1) and
paroxetine (1).
Discussion Drug-induced yawning is
rarely mentioned in pharmacological textbooks.
Antidepressants, in particular serotonin
reuptake inhibitors, are drugs that have been
implicated in excessive yawning. This ADR is
mentioned in the summary product characteristic
of several drugs from this pharmacological
class. It has also been reported when starting a
dopamine receptor agonist (like apomorphine,
pergolide or ropinirole). Other drugs known to
induce excessive yawning are
acetyicholinesterase inhibitors, corticotrophin,
ovulation inductors and sodium vaiproate.
From 1985 to December 2004, only 28 cases of
yawning as ADR were reported in the French
Pharmacovigilance Database. Evaluation of
spontaneous reporting does not permit
quantitative evaluation of the frequency of
yawning as an ADR. To determine the real
incidence of this ADR, the number of patients
taking each drug involved would need to be
known. Unfortunately, this information is not
available in the French Pharmacovigilance
Database. However, spontaneous reporting systems
remain useful tools for gaining a better
understanding of drugs and for detecting new
ADRs.
The small number of observations in our
study could reflect both a low incidence of
yawning in clinical practice and under-reporting
of this reaction. It is well known that
under-reporting is an important factor in a
spontaneous reporting system and the extent of
under-reporting depends on the characteristics
of the particular ADR (seriousness, expected or
unexpected).'3"41 Yawning is rarely a 'serious'
ADR: in our study, only four patients were
hospitalised and in one case this was because
the yawning induced a jaw luxation. Moreover,
yawning may not be reported to the
pharmacovigilance system by health professionals
because it is not frequently listed as an ADR in
the summary product characteristics of drugs.
This could explain why this ADR was rated as
'expected' in only 25% of our observations (n =
9). For example, yawning is mentioned as an ADR
in the summary product characteristic for only
one serotonin reuptake inhibitors among the
three implicated in our cases. For sodium
channel inhibitors, yawning is listed in the
summary product characteristics of lidocaine but
not flecainide, whereas two ADRs were registered
in our study with lidocaine and one ADR with
flecainide.
Although yawning was usually an 'unexpected'
ADR, its occurrence could be explained in most
of our observations by pharmacodynamic effects
of drugs involved. In our study, serotonin
reuptake inhibitors were the main
pharmacological class involved, followed by
dopaminergic drugs, underlining the excitatory
effects of serotonin and dopamine on yawning
described in experimental studies. Three
observations were available with opioids,
involving four dogs. The inhibitory role of
opioid system on yawning could be confirmed by
the case occurring after a morphine withdrawal.
However, the two other cases occurred quickly
after opioid abuse. Yawning occurring with
benzodiazepines can be expected because of the
sedative effect of these agents. Involvement of
follitropin is understandable because of
prefacilitatory action of sexual hormones on
yawning. Methylergometrine induced-yawning could
be explained by its lateral action as a dopamine
agonist. The observation of yawning occurring
with metoclopramide is more surprising since
this drug is believed to act as a dopamine
antagonist. However, this drug is also known to
activate some subtypes of serotoninergic
receptors (5-HT4). Thus, we suggest that
metoclopramide could induce yawning through a
serotoninergic activation. Moreover, this drug
is a procaine-like agent, which can contribute
to the effect of yawning. Domperidone should
inhibit yawning, because it is a dopamine
antagonist. In our study, one observation
described yawning with domperidone. However,
this could be explained by the fact that the
patient was also receiving a benzodiazepine
(lorazepam). The mechanism of drug-induced
yawning is still unclear for other drugs in our
study, for example atorvastatin and isotretinom,
which are not associated pharmacologically to
other drugs implicated in causing this ADR.
Conclusion The results of this study
do not in themselves provide evidence of the
mechanisms involved in drug-induced yawning, but
lend some support to findings in experimental
studies. Stimulation of central dopamine or
serotonin receptors elicits yawning in humans.
In contrast, the role of opioid systems seems
inhibitory. Despite the methodological drawbacks
of this study (mainly under-reporting), the
drugs found to be involved in excessive yawning
in our study can be compared with drugs already
known to induce such an effect. This study
underlines the role of several drugs in yawning
and shows that sometimes ADRs, which can be
explained by the pharmacodynamic properties of
drugs, remain unknown since they are not listed
in the summary 'product characteristics. This
study underlines the role of several drugs in
yawning and shows that this ADR is not
systematically listed in the summary product
characteristic even when it can be explained by
the pharmacodynamic properties of the
drugs.
