24-HOUR MH HOTLINE
FOR EMERGENCIES ONLY
By Paul E. Keck Jr., MD
The serotonin syndrome is an uncommon but potentially fatal complication of treatment with serotoninergic agents.This syndrome is distinguished from more common side effects of these agents by the severity, duration, and co-occurrence of a constellation of specific serotoninergically mediated systemic and central nervous system (CNS) effects (Lane and Baldwin 1997).
Insel et al. (1982) first introduced the term serotonin syndrome in the scientific literature. Sternbach (1991) conducted the first comprehensive clinical review, which summarized 12 clinical reports involving 38 patients. The publication of this review, along with the increasingly widespread use of serotoninergic medications, heightened clinical awareness of and stimulated research interest in the syndrome. For example, 133 reports involving 168 patients with suspected serotonin syndrome have been published since Sternbach’s (1991) review.
There is no estimate of the incidence of the serotonin syndrome. This is not surprising because its incidence is likely to be dependent on the criteria used to define the syndrome, specific medications and combinations of medications, pharmacokinetics, dose, and drug interactions. For example, the risk of the serotonin syndrome is likely to be substantially higher in patients receiving MAOIs in combination with other medications that enhance serotoninergic neurotransmission.
To date, seven studies have estimated the incidence of the serotonin syndrome in association with specific medications or combinations of medications (Ebert et al. 1997; Hegerl et al. 1998; Kudo et al. 1997; Lejoyeux et al. 1993; MacKay et al. 1999; Richard et al. 1997; A.L. Weiner 1999). However, the available surveys have significant methodological limitations. A blinded, prospective, naturalistic survey with patients not receiving nonserotoninergic agents as control subjects is needed to better define the incidence of serotonin syndrome in a large clinical population. Nevertheless, the results of the surveys to date suggest that serotonin syndrome is a relatively uncommon complication of treatment with serotoninergic agents. The most important risk for developing the syndrome, as reviewed later, may lie in specific medications and combinations of medications that appear to account for a greater proportion of cases described in the literature and for particularly severe and lethal manifestations of the syndrome.
Nearly all medications that enhance CNS serotoninergic neurotransmission have been reported in association with cases of the serotonin syndrome (Table 3-1). Since the cases reviewed by Sternbach (1991), the most commonly implicated agents reported in subsequent case reports and case series have been combinations of MAOIs (reversible and irreversible) and tricyclics (N=18) and MAOIs and SRIs (N=34). These combinations also have been involved in 11 (61.1%) of the 18 fatal cases of serotonin syndrome. Serotonin syndrome has been reported when an MAOI was used in combination with L-tryptophan, dextromethorphan, or clonazepam. Tricyclics combined with alprazolam, amoxapine, SRIs, lithium, SRIs and lithium, trazodone and lithium, nefazodone and thioridazine, and m-chlorophenylpiperazine (m-CPP) also have been associated with the syndrome. SSRIs in combination with lithium, venlafaxine, buspirone, dextromethorphan, carbamazepine, clonazepam, mirtazapine, m-CPP, nefazodone, trazodone, risperidone, sumatriptan, dihydroergotamine and tramadol have been reported to produce serotonin syndrome. Combinations of nefazodone with trazodone, fluoxetine, valproate, and dothiepin have been reported in association with the syndrome. Similarly, serotonin syndrome has been reported with trazodone alone and a combination of trazodone and buspirone. The syndrome also has been described in association with venlafaxine overdose and when venlafaxine was combined with lithium. Other cases have been associated with monotherapy with SRIs, clomipramine, moclobemide, m-CPP, and sumatriptan. Of nontherapeutic agents, 3, 4-methylenedioxy-methamphetamine (MDMA or “Ecstasy”) has been implicated in at least 14 reports involving 14 patients (Bedford). Five of these cases were fatal. The effect of MDMA on serotonin neurotransmission is discussed later in the section “Pathogenesis”(1). The onset of the serotonin syndrome in most cases described above was within several hours to several days of administration of serotoninergic medications or dosage increase.
Other than the risks associated with MAOI-tricyclic and MAOI-SRI combinations, there are no other known risk factors for the development of the serotonin syndrome. In addition to pharmacodynamic interactions, pharmacokinetic interactions appear to have contributed to the development of the syndrome in several cases. Specifically, these reports suggest that the administration of serotoninergic agents with other medications (with or without direct effects on serotoninergic activity) that substantially inhibit their metabolism may have produced robust increases in the plasma concentrations.
Most cases of the syndrome have occurred in patients with psychiatric disorders for which serotoninergic agents are commonly prescribed (e.g., mood disorders, obsessive-compulsive disorder, panic disorder, and generalized anxiety disorder). The syndrome also has been reported in patients with migraine, Parkinson’s disease and MDMA intoxication and in unintentional overdoses in children. Of the cases reported in the scientific literature since 1991, the mean (SD) age was 43 (12) (range =13 months –
82 years); 53 (31.5%) of 168 cases were in men. This disproportionate sex ratio is likelyto be due to the higher prevalence of mood disorder and migraine in women. In contrast to NMS, no evidence to date suggests that the physiological state of the patient may contribute to the development of the serotonin syndrome.
Table 3-1. Drugs potentiating serotonin in the central nervous system
Inhibition of serotonin reuptake
Inhibition of serotonin catabolism
Enhancement of serotonin release
Enhancement of serotonin synthesis
Source. Adapted from Keck PE Jr., Arnold LM: “Serotonin Syndrome”. Psychiatric Annals 30:333-343, 2000. Used with permission.
The clinical features of the serotonin syndrome include alterations in mental status and behavior, hyperthermia, and signs and symptoms of neurological, gastrointestinal, and autonomic nervous system dysfunction (Table 3-2). In general, substantial heterogeneity in the signs and symptoms of the serotonin syndrome is found across cases reported. This heterogeneity is likely a result of the degree of detail provided in cases reported and differences in severity among cases. Altered mental status was reported in all but 25 (14.9%) of the 168 cases reported in the literature since 1991. Mental status changes included alterations in level of consciousness (e.g., confusion, delirium, and coma) and mood (e.g., hypomania, irritability, anxiety, and dysphoria). Behavioral abnormalities were also common, with restlessness described in 55 (32.7%) cases and agitation in 43 (25.6%) others. The emergence and severity of autonomic nervous system disturbances appeared to parallel the severity of the syndrome. These signs and symptoms included tachycardia, labile blood pressure changes, diaphoresis, shivering, tachypnea, mydriasis, and sialorrhea. Hyperthermia also was associated with severity of the syndrome and was reported in 53 (31.5%) cases. The high proportion of patients with hyperthermia may reflect a reporting bias toward more severe cases. Neurological signs of serotonin syndrome were tremor, myoclonus, andle clonus, muscle rigidity, hyperreflexia, ataxia, and incoordination. Interestingly, gastrointestinal signs and symptoms were not commonly observed.
No specific laboratory abnormalities have been identified in association with the serotonin syndrome. Those that have been reported have been either nonspecific (e.g., leukocytosis) or secondary to complications of the syndrome (e.g., azotemia, thrombocytopenia). However, a thorough laboratory evaluation is often needed to rule out other causes of the clinical features associated with serotonin syndrome, especially in moderate to severe cases. Among the 168 cases described since 1991, leukocytosis was reported in 14 (8.3%) cases. Rhabdomyolysis with associated elevations in creatine phosphokinasc (CPK) levels was reported in 45 (26.8%) cases; 7 (4.2%) led to myoglobinuria-induced acute renal failure and death. Elevations of CPK were modest in most cases but were increased 1,000-fold in 12 (7.1%) cases. Findings indicative of disseminated intravascular coagulation (DIC) were reported in 10 (6.0%) cases; 6 were fatal. Similarly, clinically significant hepatic transaminase elevations were reported in only 7 (4.2%) cases, but 5 were fatal. Hyponatremia, hypomagnesemia and hypercalcemia also have been reported in isolated cases. These abnormalities are likely to be secondary to fluid and electrolyte disturbances resulting from the syndrome.
Table 3-2. Clinical features of the serotonin syndrome
Autonomic nervous system
Ataxia and incoodination
Source. Adapted from Keck PE Jr., Arnold LM: “serotonin Syndrome.” Psychiatric Annals 30:333-343, 2000. Used with permission.
Few data suggest that elevated ambient temperature or other environmental factors posed significant risks in the development of serotonin syndrome among reported cases. However, J.A. Henry et al. (1992) speculated that MDMA-induced serotonin syndrome may be linked to vigorous physical activity, elevated ambient temperature, and inadequate fluid intake occurring during “rave” parties.
Three criteria sets have been proposed to date to define the serotonin syndrome. No formal consensus has yet been achieved regarding the criteria for the syndrome. Sternbach (1991) proposed the first operational criteria based on his review of 38 cases.
Table 3-3. Operational criteria for serotonin syndrome
1. Coincident with the addition of or increase in a known serotoninergic agent to an established medication regimen, at least three of the following clinical features are present:
· Mental status changes (e.g., confusion or hypomania)
2. Other etiologies have been ruled out.
3. A neuroleptic had not been started or increased in dosage prior to the onset of the signs and symptoms listed above.
Source. Adapted from Sternbach H: “The Serotonin Syndrome.” American Journal of Psychiatry 148:705-713, 1991. Used with permission. Copyright 1991 American Psychiatric Association.
The signs and symptoms of serotonin syndrome reported in the 168 cases since 1991 are comparable with those proposed by Sternbach (1991). This is not surprising because most reports used these criteria to define the syndrome. Nevertheless, of the 10 core features of the syndrome identified by Sternbach, only diarrhea occurred with a low prevalence rate (8%). Other features were reported commonly, including autonomic nervous system dysfunction (e.g., tachycardia, 37%; labile blood pressure changes, 24%; mydriasis, 25%) and muscle rigidity (33%). These data suggest that muscle rigidity and signs of autonomic nervous system instability should be included as criteria for serotonin syndrome and diarrhea omitted. In short, these modified criteria would include:
1. Mental status changes
2. Agitaion of restlessness
9. Autonomic nervous system dysfunction
11. Muscle rigidity
The phenomenological overlap of serotonin syndrome with NMS is apparent in the last three criteria.
The diagnosis of serotonin syndrome involves differentiating this condition from other disorders associated with cognitive and behavioral, neuromuscular, and autonomic nervous system dysfunction with or without hyperthermia. Thus, the differential diagnosis of serotonin syndrome is virturally identical to that of and includes NMS
Table 3-4. Features of serotonin syndrome and neuroleptic malignant syndrome
|Feature||Serotonin syndrome||Neuroleptic malignant syndrome|
|Neurological||Muscle rigidity variable||Muscle rigidity|
|Hyper-or hypotension||Hyper-or hypotension|
|Laboratory||Elevations uncommon:||Elevations uncommon:|
|Creatine phosphokinase||Creatine phosphokinase|
|White blood cell count||White blood cell count|
|Liver function tests||Liver function tests|
Source. Adapted from Keck PE Jr, Arnold LM: “Serotonin Syndrome”. Psychiatric Annals 30:333-343, 2000. Used with permission.
In most reports, serotonin syndrome has been mild to moderate in severity without medical complications. However, a distinct minority of cases, particularly those associated with MAOI-tricyclic and MAOI-SRI combinations and MDMA, were severe and some fatal. Overall, medical complications appear to be less common than in NMS. Rhabdomyolysis was the most common serious medical complication of serotonin syndrome, occurring in 45 (26.8%) cases. Myoglobinuria and renal failure were reported in 8 (4.8%) patients. Generalized seizures were reported in 18 (10.7%) cases, including 7 fatalities. Similarly, DIC occurred in 8 patients, 5 of whom died. Thus, although medical complications were not common in patients with serotonin syndrome, they were associated with a substantial mortality risk.
No systematic studies of the treatment of serotonin syndrome have been done. In most of the cases reported, the syndrome was self-limited and usually resolved quickly after discontinuation of serotoninergic medications. In more severe cases, other interventions were necessary. These included respiratory and cardiovascular monitoring and support, the use of a cooling blanket for hyperthermia, intravenous hydration to prevent renal failure, and anticonvulsants for myoclonus or seizures. Drug discontinuation and supportive measures were associated with resolution of the syndrome in 74 (44.0%) of the 168 cases reported since those reviewed by Sternbach (1991). In most of these cases, patients had mild to moderate signs and symptoms of the syndrome.
The use of specific pharmacological treatments is anecdotal but can be considered in severe cases of serotonin syndrome or in patients who do not respond to discontinuation of serotoninergic agents and supportive measures. The nonspecific serotonin antagonist cyproheptadine (4-24 mg/day) was the most consistently effective treatment in case reports, with rapid resolution occurring in 19 (90.5%) of 21 patients. Benzodiazepines were administered in 33 cases and appeared to be effective in 5, of equivocal benefit in 19, and of no benefit in 9. Low-potency typical antipsychotics were tried in 6 cases, all with equivocal benefit. However, the use of neuroleptics in a patient with suspected serotonin syndrome could complicate the subsequent course of illness and pose a theoretical risk of exacerbating hyperthermia.
Several other agents, including propranolol (L-isomer is a serotonin antagonist), ketanserin (a 5-HT2 antgonist), and mirtazapine produced equivocal improvement in a few cases. Mirtazapine also was suspected of contributing to the development of serotonin syndrome. Nitroglycerin was reported effective in one case. Dantolene, a muscle relaxant thought to be effective in treating NMS was administered in 10 patients (9 severe); 4 (40%) of the 10 patients appeared to improve markedly. In contrast to the clinical data and based on an animal model of serotonin syndrome, Nisijima et al. (2001) argued that potent 5-HT2A receptor antagonists (including atypical antipsychotics) represent the most effective drugs for treating this condition. This view derives from the role of 5-HT2A receptors in mediating hyperthermia and the association of hyperthermia with severity of the syndrome.
The serotonin syndrome is an uncommon adverse event usually produced by combinations or high doses of serotoninergic medications. The features of the syndrome include mental status and mood changes, behavioral and neuromuscular abnormalities, autonomic nervous system dysfunction, and hyperthermia. The onset of the syndrome is usually abrupt following the addition of or increase in the dose of a seroteninergic agent. In many cases, the syndrome is mild and abates when serotoninergic agents are discontinued. The most severe cases have been associated with MAOI-tricyclic and MAOI-SRI combinations and MDMA abuse. In severe cases, rapid intervention with supportive measures may be required, and nonspecific serotonin antagonists may be helpful. The development of a 5-HT1A antagonist could provide a more specific pharmacological intervention, and the utility of 5-HT2A antagonists merits further investigation. Despite heightened clinical awareness in recent years, recent surveys suggest that the syndrome still may be under-diagnosed. Greater awareness of the drug combinations commonly associated with the syndrome should lead to better prevention and early recognition and treatment.
Reprinted with permission from Neuroleptic Malignant Syndrome and Related Conditions, (Copyright 2003) on American Psychiatric Publishing, Inc.
(1) This section is deleted from this report due to condensing for space. It may be seen in the complete chapter on Serotonin Syndrome in Neuroleptic Malignant Syndrome and Related Conditions.