Kirk C. Mills
Serotonin syndrome is an important drug-related complication of psychopharmacologic therapy for depression, bipolar affective disorder, obsessive-compulsive disorder and Parkinson’s disease. It is characterized by variable alterations in cognition and behavior, autonomic nervous system function and neuromuscular activity[1,2] (Table 1). Serotonin syndrome is an iatrogenic disorder: it only occurs in the setting of drug therapy that has a net effect of augmenting brain serotonin neurotransmission. Serotonin syndrome most commonly occurs when two or more serotonergic drugs are given concurrently. However, the syndrome has also been reported with single drug exposure in both therapeutic settings and overdose settings.
Signs and Symptoms of Serotonin
Syndrome (Review of 100 Cases)(*)
In general serotonin receptors are categorized into four separate classes designated as 5-[HT.sub.1], 5-[HT.sub.2], 5-[HT.sub.3] and 5-[HT.sub.4] (Figure 1). The 5-[HT.sub.1] class is the largest and most studied class of serotonin receptors. Some of these receptor classes also have different subclasses of receptors. In animal models, serotonin syndrome results from excessive stimulation of specific postsynaptic serotonin receptors located in the lower brainstem (pons, medulla) and spinal cord regions.[1,6] The 5-[HT.sub.1A] receptor subtype predominates in this region of the brainstem. In the past, the 5-[HT.sub.1A] postsynaptic receptor was believed to mediate the entire syndrome, but recent studies have demonstrated that the 5-[HT.sub.2] postsynaptic receptor is at least partially involved in the expression of this syndrome. In addition, other neurotransmitter receptors (dopamine, beta-adrenergic, etc.) may modulate the expression of the serotonin syndrome.
The diagnosis of serotonin syndrome is based entirely on a strong clinical suspicion and the exclusion of other medical and psychiatric conditions. The “typical” patient with serotonin syndrome demonstrates altered behavior-cognition ability, autonomic nervous system dysfunction and neuromuscular abnormalities. The frequency with which specific signs and symptoms were reported in 100 cases of serotonin syndrome reported in the literature is listed in Table 1.
Some patients report recurrent mild symptoms days to weeks before symptoms become more severe. In a few cases, patients rapidly progress to multiple organ failure and death. Serotonin syndrome has been reported in pediatric patients.[2,4]
Abnormalities of the neuromuscular system are the most commonly reported findings in patients with serotonin syndrome. Many of these symptoms are quite unusual and may cause the evaluating physician to mistakenly attribute the symptoms to malingering or a primary neurologic disorder (Table 3). Neuromuscular complaints include restlessness, involuntary myoclonic jerks (even when sleeping), resting extremity tremor, teeth chattering and difficulty walking. Generalized muscular rigidity develops in all severe cases. Sustained muscle contraction predisposes patients to hyperthermia, metabolic acidosis, rhabdomyolysis and impaired respiratory function. Isolated bilateral lower extremity rigidity is fairly specific for serotonin syndrome.
The initial cognitive-behavioral changes of serotonin syndrome are often overlooked. These alterations include anxiety, confusion, agitation, hypomania, headache and insomnia. A search of the literature volume 52, number 5 revealed that coma occurred in over 25 percent of all patients. Seizures are usually generalized, but focal seizures have been reported. Visual and auditory hallucinations are unusual but have also been reported.
Dysfunction of the autonomic nervous system commonly occurs in serotonin syndrome. Diaphoresis, hyperthermia, sinus tachycardia, hypertension and tachypnea are commonly reported. In addition, pupillary dilatation was noted in 28 percent of cases found in the literature. Twenty percent of patients had pupils unreactive to light. Other symptoms related to autonomic nervous system dysfunction include nausea, vomiting, diarrhea, skin flushing and abdominal cramps. Hypotension usually occurs as a preterminal event. Ventricular tachycardia has been reported but is unusual.
No specific tests are available for the diagnosis of serotonin syndrome. Laboratory tests are best used to identify complications (e.g., rhabdomyolysis) and assist in overall patient care. Elevated serotonergic drug levels (e.g., lithium, fluoxetine, clomipramine) are not required to produce serotonin syndrome, and in over 90 percent of the cases in which drug levels (and metabolites) were measured, drug levels of serotonergic agents were within accepted therapeutic limits.
Serotonin syndrome is not associated with abnormalities in cerebrospinal fluid analysis, radiographic studies (including computed tomographic scanning), electroencephalography (excluding seizures) or routine serum electrolyte determinations. When laboratory abnormalities occur, they are almost always secondary to complications. These complications include hypoxia from respiratory muscle rigidity or coma, rhabdomyolysis from diffuse muscle rigidity, disseminated intravascular coagulation from multiple organ failure, metabolic acidosis secondary to seizures or ventricular tachycardia, and aspiration pneumonia from decreased consciousness.
Because serotonin syndrome most commonly occurs when two or more serotonergic drugs are used in combination, it becomes imperative to recognize and avoid potential serotonergic drug interactions, as well as to understand safe drug therapy alternatives.
Meperidine (Demerol) and dextromethorphan are potent inhibitors of serotonin uptake and are notorious for precipitating acute serotonin syndrome, especially in patients taking MAOIs, including selegiline (Eldepryl).[7,8] In general, it is best avoid prescribing either meperidine or dextromethorphan for any patient taking a serotonergic agent. Morphine, fentanyl, salicylates, acetaminophen and nonsteroidal anti-inflammatory drugs are considered safe analgesic alternatives to meperidine. The safety of codeine as an alternative antitussive to dextromethorphan remains unproved.
All three SSRIs can produce serotonin syndrome, most commonly when taken concomitantly with other serotonergic drugs. The unique pharmacokinetics of fluoxetine make it especially prone to cause serotonin syndrome. The half-life of fluoxetine ranges from one to four days, and its active metabolite has a half-life of seven to 14 days. A five-week abstinence period is commonly recommended once fluoxetine therapy has been stopped before initiating any other serotonergic agent. Paroxetine and sertraline (Zoloft) have half-lives of approximately 24 hours. Patients receiving these drugs require only a two-week abstinence period before starting new medications. Three new, recently released drugs also inhibit serotonin uptake: venlafaxine (Effexor), nefazodone (Serzone) and fluvoxamine (Luvox). Until proven otherwise, these three drugs should be viewed as having the potential to produce serotonin syndrome.
All currently available MAOIs produce irreversible inhibition of MAO enzyme activity. Return of MAO enzyme activity to an acceptable level occurs approximately two weeks after discontinuation of any of the irreversible MAOIs. This delay is due to the gradual production of uninhibited MAO enzyme. An abstinence period of at least two weeks is required before drugs with serotonergic activity can be administered after discontinuation of any of the MAOIs. Most MAOIs are used to treat depression, but selegiline is used as adjunctive therapy in Parkinson’s disease because of its ability to inhibit MAO-B isoenzymes and thus secondarily increase dopamine availability. This selectivity in MAO-B inhibition is easily lost with dosages that are slightly higher than therapeutic. Therefore, selegiline should be viewed as having the same potential to precipitate serotonin syndrome as the other MAOIs.
Drugs that increase central nervous system dopamine concentrations, such as levodopa (Sinemet), bromocriptine (Parlodel) and selegiline, have the potential to precipitate serotonin syndrome by means of their ability to cause indirect serotonin release.2 l0 These medications should be used cautiously, if at all, with other serotonin agonists.
Only four drugs are currently available that directly interact with serotonin receptors. These drugs are buspirone (Buspar), sumatriptan (Imitrex), ondansetron (Zofran) and granisetron (Kytril). Buspirone is an anxiolytic agent that has partial agonist activity at 5-[HT.sub.1A] receptors. In humans, buspirone has been reported to induce myoclonus but not the complete serotonin syndrome. Sumatriptan is an antimigraine medication that stimulates both 5-[HT.sub.1D] and 5-[HT.sub.1A] receptors. Interestingly, sumatriptan poorly penetrates the blood-brain barrier and, therefore, it is highly unlikely to cause serotonin syndrome. Ondansetron and granisetron are antiemetics that selectively antagonize 5-HT3 receptors. They are incapable of producing serotonin syndrome.
Lithium is a nonspecific serotonin agonist. Many cases of serotonin syndrome have occurred when lithium is combined with another serotonergic agent. The serotonergic effect of lithium is present even when blood lithium levels are within the therapeutic range.
The majority of patients with serotonin syndrome can be effectively managed using four basic principles. First, provide necessary supportive care. Second, discontinue all serotonergic drugs. Third, consider use of antiserotonergic medications. Fourth, make necessary adjustments in the medication regimen before reinstituting serotonergic medications.
Serotonin syndrome is associated with a good prognosis. The review of the literature indicated that approximately 70 percent of cases show complete resolution within 24 hours after symptom onset. Even in severe cases, some clinical improvement is usually noted within the first 24 hours. Symptoms rarely last longer than 72 to 96 hours in the absence of complications. Despite these favorable statistics, many patients are severely ill on initial presentation. Forty percent of the cases culled from the literature required admission to an intensive care unit, and 25 percent required artificial ventilation. At least 11 deaths were attributed to serotonin syndrome.
Hyperthermia is usually associated with generalized muscle rigidity. Efforts to lower the hyperthermia will only be successful once muscle rigidity resolves. Benzodiazepines are considered the first drugs of choice to treat muscle rigidity. Muscle paralysis, achieved through the use of nondepolarizing agents, is indicated when rigidity remains refractory to benzodiazepines. Dantrolene (Dantrium) has also been used but is generally unnecessary.
Attempts at postsynaptic serotonin receptor blockade may be appropriate when symptoms persist or become particularly severe. It appears that nonspecific (5-[HT.sub.1], 5-[HT.sub.2]) receptor antagonists are the most effective agents.[1,6] Animal studies have shown that propranolol, methysergide and cyproheptadine are effective in preventing symptoms of serotonin syndrome.46 Any potential benefit from the use of these agents is entirely speculative, because no controlled human studies of drug therapy in serotonin syndrome have been performed. Clinical experience with serotonin antagonists is limited to less than a dozen case reports[4,10,11,13-17]
Cyproheptadine is consistently the most effective agent in humans.[11,15,17] The initial dose of cyproheptadine is usually 4 to 8 mg orally. This dose can be followed by repeat doses of 4 mg every two to four hours, up to a total of 0.5 mg per kg per day. Because cyproheptadine also antagonizes muscarinic acetylcholine receptors, anticholinergic symptoms (e.g., urinary retention) may develop following large doses.
Methysergide appears to be the second most effective agent in humans.[10,16] The dose of methysergide ranges from 2 to 6 mg, for a total daily dose of up to 6 mg. Propranolol was effective in one case, but had no effect in other cases.[14,15] Many phenothiazines are relatively potent blockers of 5-[HT.sub.2] receptors. Chlorpromazine (Thorazine) has been associated with reports of mixed effectiveness in reversing symptoms of serotonin syndrome.[22,23]l A significant concern with chlorpromazine is the potential lowering of the seizure threshold. This is an important concern in serotonin syndrome since 12 percent of patients have seizures.
The final management issue concerns restarting patients on appropriate pharmacologic drug therapy after their recovery from serotonin syndrome.72 23 Very little information is available to guide clinicians in this area. Patients who have recovered from serotonin syndrome are probably at higher risk for recurrence on drug rechallenge. It seems appropriate to limit serotonergic medications in these patients. Some ways that this can be achieved are to discontinue any combination drug therapy, use lower potency serotonergic agents, decrease drug dosages and closely monitor patients for symptom development.[Figure 1 ILLUSTRATION OMITTED]
REFERENCES. Sternbach H. The serotonin syndrome. Am J Psychiatry 1991;148:705-13. [2.] Mills KC. Serotonin toxicity: a comprehensive review for emergency medicine. Top Emerg Med 1993;15(4):54-73. [3.] Lejoyeux M, Fineyre F, Ades J. The serotonin syndrome [Letter]. Am J Psychiatry 1992;149:1410-1. [4.] Kaminski CA, Robbins MS, Weibley RE. Sertraline intoxication in a child. Ann Emerg Med 1994; 23:1371-4. [5.] Mills KC. Monoamine oxidase inhibitor toxicity. Top Emerg Med 1993;15(3):58-71. [6.] Heal DJ, Luscombe GP, Martin KF. Pharmacological identification of 5-HT receptor subtypes using behavioral models. In: Marsden CA, Heal DJ, eds. Central serotonin receptors and psychotropic drugs. Boston: Blackwell Scientific, 1992:56-99. [7.] Browne B, Linter S. Monoamine oxidase inhibitors and narcotic analgesics. A critical review of the implications for treatment. Br J Psychiatry 1987; 151:210-2. [8.] Nierenberg DW, Semprebon M. The central nervous system serotonin syndrome. Clin Pharmacol Ther 1993;53:84-8. [9.] Leonard BE. Pharmacological differences of serotonin reuptake inhibitors and possible clinical relevance. Drugs 1992;43(Suppl 2):3-9. [10.] Sandyk R. L-dopa induced “serotonin syndrome” in a parkinsonian patient on bromocriptine [Letter]. J Clin Psychopharmacol 1986;6:194-5. [11.] Goldberg RJ, Huk M. Serotonin syndrome from trazodone and buspirone [Letter]. Psychosomatics 1992;33:235-6. [12.] Price LH, Charney DS, Delgado PL, Heninger GR. Lithium and serotonin function: implications for the serotonin hypothesis of depression. Psychopharmacology 1990;100:3-12. [13.] Guze BH, Baxter LR Jr. The serotonin syndrome: case responsive to propranolol [Letter]. J Clin Psychopharmacol 1986;6:119-20. [14.] Ruiz F. Fluoxetine and the serotonin syndrome. Ann Emerg Med 1994;24:983-5. [15.] Lappin RI, Auchincloss EL. Treatment of the serotonin syndrome with cyproheptadine [Letter]. N Engl J Med 1994;331:1021-2. [16.] Lieberman JA, Kane JM, Reife R. Neuromuscular effects of monoamine oxidase inhibitors. Adv Neurol 1986;43:231-49. [17.] Muly EC, McDonald W, Steffens D, Book S. Serotonin syndrome produced by a combination of fluoxetine and lithium [Letter]. Am J Psychiatry 1993;150:1565. [18.] Brodribb TR, Downey M, Gilbar PJ. Efficacy and adverse effects of moclobemide [Letter]. Lancet 1994;343:475. [19.] Graham PM, Potter JM, Paterson J. Combination monoamine oxidase inhibitor/tricyclic antidepressant interaction [Letter]. Lancet 1982;2(8295):440. [20.] Grantham J, Neel W, Brown RW. Reversal of imipramine-monoamine oxidase inhibitor induced toxicity with chlorpromazine. J Kans Med Soc 1964;65:279-80. [21.] Tackley RM, Tregaskis B. Fatal disseminated intravascular coagulation following a monoamine oxidase inhibitor/tricyclic interaction. Anaesthesia 1987;42:760 3. [22.] Brannan SK, Talley BJ, Bowden CL. Sertraline and isocarboxazid cause a serotonin syndrome [Letter]. J Clin Psychopharmacol 1994;14:144-5. [23.] Bodner RA, Lynch T, Lewis L, Kahn D. Serotonin syndrome. Neurology 1995;45:219-23.
RICHARD W. SLOAN, M.D., R.PH., coordinator of this series, is chairman and residency program director of the Department of Family Medicine at York (Pa.) Hospital and clinical associate professor in family and community medicine at the Milton S. Hershey Medical Center, Pennsylvania State University, Hershey, Pa.
KIRK C. MILLS, M.D. is currently assistant clinical professor of emergency medicine and associate director of medical toxicology at Wayne State University in Detroit, Mich., and associate director of the Detroit Poison Center. He formerly was assistant professor of emergency medicine and medical director of the Mid-America Poison Control Center at the University of Kansas Medical Center, Kansas City, Kan. He completed a fellowship in medical toxicology at Good Samaritan Medical Center in Phoenix. Dr. Mills is a graduate of the University of Kansas School of Medicine and completed a residency in emergency medicine at York (Pa.) Hospital.
Address correspondence to Kirk C. Mills, M.D., Detroit Receiving Hospital, Department of Emergency Medicine, 4201 Saint Antoine Street, Detroit, Ml 48201.
COPYRIGHT 1995 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group