Abstract
Dear Sir We are grateful for the letter by Evans in response to our report on a case of serotonin syndrome induced by concomitant use of fluoxetine, hypericum and eletriptan, since it gives us the opportunity to clarify a few crucial points (1). The Author is concerned about the diagnosis of serotonin syndrome because he believes that the reported case would not have fulfilled the current criteria. Although several diagnostic criteria and algorithms have recently been developed (2, 3), we adopted the original criteria by Sternbach, which had also been extensively validated in conditions induced by combined use of serotonergic medications and not in selective serotonin re-uptake inhibitor (SSRI)-alone overdose.
Sternbach's criteria require the presence of three or more of the 10 most common clinical features detected in the first case series of serotonin syndrome that developed at the time of the addition of or an increase in a known serotonergic agent, in the absence of other causes (4). Evans agrees that the clinical features of our case report, i.e. tremor, fever and mental status changes, were in keeping with those known to be induced by serotonergic toxicity, but was concerned that we did not rule out other aetiologies. The patient had no history of substance abuse or withdrawal, and no serum metabolic alterations were detected in the diagnostic work-up at admission. The possibility of an infective aetiology was considered unlikely since the clinical history was negative for symptoms of upper respiratory tract or gastrointestinal infections, the blood count examination and serum indices for an infective disease were unremarkable, and neurological examination revealed no signs of meningitis. The increase in body temperature was mild and transient with spontaneous remission in 12 h. The headache presented with characteristics similar to those reported in the past clinical history, met the criteria for migraine, and was responsive to triptan administration. In addition, EEG did not show slow wave activity and was not suggestive of encephalitis. On the basis of these considerations, we deemed it unnecessary to perform a lumbar puncture or other investigations to rule out an infection. Furthermore, the likelihood that the rhabdomyolysis could have been due to a viral infection such as Coxsackie B virus infection is remote. In the paper quoted by Evans, Coxsackie-induced rhabdomyolysis is reported as a rare condition associated with classical symptoms of a viral infection such as malaise, sore throat, enduring fever and serum leukocytosis—features suggestive of a viral infection which were lacking in our patient (5). Conversely, increased creatine phosphokinase levels have been described in 12% of serotonin syndrome cases, followed by renal failure in 4% (3).
In our opinion, the patient meets Sternbach's criteria for serotonin syndrome, considering the recent addition of a serotonergic agent in a background of a combined assumption of herbal and drug serotonergic tone modulators, and that the clinical symptoms and signs present at admission were in keeping with those described in serotonergic toxicity and were in remission as soon as these agents were withdrawn. Other aetiologies such as metabolic disorders, drug intoxication or cerebrovascular disease were ruled out; the hypothesis of a meningoencephalitis was unlikely considering the clinical history, the neurological examination and the serum laboratory findings, even though a CSF study had not been performed.
Another point raised by Evans is the debatable contribution of triptans to the development of a serotonin syndrome. We are aware that administration of triptans alone has not been associated with clinical features of serotonin toxicity. However, a few case reports of serotonin syndrome induced by drug combinations including triptans have been published, arguing against the hypothesis that triptans may cause serotonin toxicity when used alone, but in favour of potential pharmacodynamic and/or pharmacokinetic interactions in multidrug regimens involving serotonergic agents. In our case, for example, hypericum may have increased the serotonergic tone by stimulating the release of serotonin, thus amplifying the effect of high-dose fluoxetine (60 mg/day), but like fluoxetine it is also a inhibitor of the enzyme P450 CYP3A4, involved in the metabolism of eletriptan (6, 7). In such complex drug–drug and drug–herbal interactions, the temporal relationship between medication changes and symptoms course may help to distinguish the contribution of the single serotonergic agent. Symptoms onset was within 3 days of triptan addition and discontinuation of both fluoxetine and triptan was effective, whereas immediate hypericum withdrawal did not prevent the development of rhabdomyolysis.
Finally, considering the limited number of case reports on serotonin toxicity induced by triptans and SSRI or SNRI, we share with Evans the view that caution should be taken in advising patients of this potential complication in order to avoid generating unnecessary alarm. However, mild atypical cases are very likely to be ignored or mistaken and, for these reasons, under-reported in the literature. In our opinion it would be important also to draw the attention of physicians treating migraine to serotonin syndrome, thus increasing awareness and consequently the ability to recognize the variety of its presentations.