Epistaxis as an Unexpected Side Effect of Aripiprazole and Risperidone Treatment in Two Children with Two Different Psychiatric Diagnosis

To the Editor:

Risperidone, one of the most prescribed atypic antipsychotic (AAP), is a potent antagonist of both D2 and 5HT2A receptors and aripiprazole has a unique mechanism of action with partial agonist at the D2 and 5HT1A receptors and an antagonist at the 5HT2A receptor (Correll 2015). Both of these medications have been shown to be effective in treatment of disruptive behavior disorders, irritability, and aggressive behaviors associated with autistic spectrum disorder (ASD) and mood disorders (Correll 2015). Even though these two drugs are generally well tolerated, they have side effects, including somnolence, headache, increased appetite, weight gain, and extrapyramidal symptoms (Correll 2015). Risperidone-related epistaxis was reported rarely in adults (Harrison-Woolrych and Clark 2004) and children (Coskun and Mukaddes 2008), whereas there is no evidence for aripiprazole-related epistaxis. In this report, we describe two children who had epistaxis during aripiprazole and risperidone treatment and discuss the possible mechanism related to this adverse effect.

Case I

A 5-year-old male with a diagnosis of ASD was admitted to our outpatient clinic because of severe hyperactivity and impulsive behaviors. He was found to have attention-deficit/hyperactivity disorder (ADHD). He was started on risperidone 0.25 mg once a day and was slowly titrated to 0.5 mg/b.i.d. in 2 weeks. Although he was taking risperidone 0.5 mg/b.i.d., his mother reported epistaxis during sleep on the fifth day. His parents discontinued the medication because of the ongoing epistaxis. He was reexamined after 2 months of discontinuation and the parents stated that he had no epistaxis during this time. Behavior symptoms got worse during this period, and the parents agreed on restarting risperidone medication with the same dosage. Four days after initiating 0.5 mg/b.i.d. risperidone, he had two to three episodes of epistaxis. Physical examinations and investigations were normal (complete blood count [CBC], protrombin time [PT], activated parsial tromboplastin time [aPTT]). During the 2 weeks medication-free period, he had no epistaxis. Then we started aripiprazole 0.5 mg once a day and slowly titrated to 2 mg in 2 weeks. On the fourth day of 2 mg dose, he had two to three episodes of epistaxis again and we stopped medication. With the approval of his parents, we started him on short-acting methylphenidate, he is on this medication during follow-up and no epistaxis has been reported since then.

Case 2

A 7-year-old boy with severe behavioral problems was directed to psychiatric evaluation by his school and was found to have both disruptive mood dysregulation disorder (DMDD) and ADHD. Owing to his parents' unwillingness for use of medication, CBT was started. When it came to the point of disposal from the school, parents agreed on medication. Risperidone was started 0.125 mg/day and was titrated up to 1 mg/b.i.d. in a month. On the second week while he was on 2 mg dosage, he had spontaneous epistaxis and had one to two episodes of epistaxis for the next 3 days. Physical examinations and blood tests results were normal (CBT, PT, APTT). His parents refused medication and CBT was the only treatment that he got for the next 3 months. Depressive symptoms were added to severe behavioral problems during his parents' divorce period but he never fulfilled any depressive spectrum disorders DSM-5 criteria except DMDD. Aripiprazole was started 0.5 mg and was titrated to 0.5 mg twice a day in a week. On the second week, he had epistaxis with 1.5 mg/day aripiprazole and we stopped medication. His treatment continues with CBT without any epistaxis episodes.

In these cases with different diagnosis, using risperidone and aripiprazole provided moderate outcome of their symptomatology but caused unspecified epistaxis episodes. A review of literature showed a few risperidone-related epistaxis in adult (Harrison-Woolrych and Clark 2004) and children (Coskun and Mukaddes 2008) but there were no data regarding aripiprazole-related epistaxis. Only in a study evaluating effects of aripiprazole in pediatric bipolar disorder it was reported that one of the children was removed from the study because of epistaxis (Findling et al. 2011). Harrison-Woolrych and Clark described two adults having epistaxis during risperidone treatment and also mentioned a report of the World Health Organization's international drug monitoring database. In this report, it has been informed that 54 patients had risperidone-related epistaxis (Harrison-Woolrych and Clark 2004). Epistaxis and gastrointestinal bleeding have also been reported in two children who were treated with risperidone (Coskun and Mukaddes 2008).

A variety of mechanisms might explain this adverse bleeding effect of risperidone and aripiprazole. A rare side effect thrombocytopenia may occur because of the APP drugs (Correll 2015), but none of our cases had thrombocytopenia.

Recent studies have reported that Sargogrelate, which is a cardiovascular drug, inhibits platelet aggregation through 5HT2A antagonism and reduces vasoconstriction in coronary arteries (Ayme-Dietrich et al. 2017). If risperidone and aripiprazole are thought to be potent 5HT2A receptor antagonists, they may cause bleeding in nasal vessels by mimicking antithrombotic action in the microcirculation with a similar mechanism of action. In addition, aripiprazole is an inverse agonist of the 5HT2B receptor that produces a vasodilating response in the arteries (Ayme-Dietrich et al. 2017). However, its affinity for the 5HT2B receptor is very low than its 5HT2A receptor affinity (Shapiro et al. 2003). For this reason, it can be considered that it reduces vasoconstriction and increases the tendency to bleed.

Although the epistaxis observed in the 2–3 weeks of treatment in the cases was similar to the onset of bleeding in the patients of the WHO report (Harrison-Woolrych and Clark 2004), it varied in the previous case reports (Harrison-Woolrych and Clark 2004; Coskun and Mukaddes 2008). The reason for the differences in the emergence of this side effect is unclear, but it may be because of individual differences.

There is a need for more detailed evaluation of drug effect in the occurrence of epistaxis which is chronologically related to use of these drugs.