INTRODUCTION — Selective serotonin reuptake inhibitors (SSRIs) were introduced in the late 1980s for the treatment of depression. They are generally much safer in overdose than tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs). Commonly used SSRIs include fluoxetine, sertraline, paroxetine, fluvoxamine, citalopram, and escitalopram.
In 2017, an SSRI was mentioned in 57,254 single-substance toxic exposures reported to United States poison centers, 167 major outcomes, and only five fatalities [1]. The incidence of SSRI ingestions has steadily increased and mirrors the increase in SSRI prescriptions [2]. Isolated SSRI ingestions generally produce mild symptoms, although SSRI intoxication can infrequently produce serotonin syndrome, seizures, central nervous system (CNS) depression, or various cardiac abnormalities. Citalopram and escitalopram are structurally different from the other SSRIs and potentially more toxic in overdose.
The toxicity and management of acute SSRI overdose will be discussed here. A summary table to facilitate emergency management of SSRI poisoning is provided (table 1). Discussions of the therapeutic use of these medications and of serotonin syndrome are found elsewhere. (See "Serotonin-norepinephrine reuptake inhibitors: Pharmacology, administration, and side effects" and "Serotonin syndrome (serotonin toxicity)".)
PHARMACOLOGY AND CELLULAR TOXICOLOGY — Serotonin is produced from tryptophan, while norepinephrine and dopamine are produced from tyrosine, through similar chemical pathways in nerve terminals. These neurotransmitters are stored in cytoplasmic vesicles of presynaptic neurons and released into the synapse with nerve depolarization. Transport proteins specific for serotonin, norepinephrine, or dopamine are embedded in the presynaptic terminal cell membrane and are responsible for pumping these neurotransmitters out of the synapse and back into the cytoplasm. Once in the cytoplasm, the neurotransmitters are either repackaged into vesicles for reuse or degraded by monoamine oxidase. (See "Selective serotonin reuptake inhibitors: Pharmacology, administration, and side effects", section on 'Pharmacology'.)
Approximately 98 percent of the body's serotonin is found outside the central nervous system (CNS). The majority of serotonin is synthesized and stored in the enterochromaffin cells of the gut, where it induces contraction of gastrointestinal smooth muscle. There are at least seven families of 5-HT receptors (5-HT1 to 5-HT7), some of which have multiple subtypes [3]. In the CNS, serotonergic neurons are found in the brainstem, mostly in the raphe nucleus. They have a role in regulating mood, personality, appetite, wakefulness, emesis, temperature regulation, and sexual behavior [4]. Peripherally, serotonin aids in regulating vascular tone, peristalsis, and platelet activation.
Selective serotonin reuptake inhibitors (SSRIs) inhibit the reuptake mechanism of serotonin in CNS neurons, as well as peripherally and in platelets, thereby increasing the stimulation of serotonin receptors. Compared with tricyclic antidepressants (TCAs) and other antidepressants, SSRIs are less effective at blocking reuptake of norepinephrine and likely less able to antagonize muscarinic, histaminergic, and adrenergic receptors.
KINETICS — Selective serotonin reuptake inhibitors (SSRIs) are rapidly absorbed and reach a peak serum concentration within six hours (table 2). Most SSRIs demonstrate a high degree of serum protein binding, and most have a long elimination half-life, with sustained biological activity due to active metabolites. Paroxetine and fluvoxamine are unusual in that they have no active metabolites. In overdose, SSRI elimination times are further prolonged by the saturation of metabolic enzymes and by delayed absorption, especially for sustained-release preparations.
Due to fluoxetine's active metabolite norfluoxetine and to the slow elimination of both, significant accumulation of these drugs and delays in reaching a steady state can occur. Significant serum concentrations can persist even four to five weeks after discontinuation of fluoxetine [5]. This can result in adverse drug interactions and toxicity if a new antidepressant or serotonergic medication is introduced for up to a month after fluoxetine is stopped (table 2).
All SSRIs are metabolized in the liver by the cytochrome P450 system; most metabolites are renally excreted.
CLINICAL FEATURES OF OVERDOSE
History — Clinicians should seek answers to the following questions:
●Which drugs were ingested (including all coingestants and drug formulations, eg, sustained release)?
●What amount was ingested?
●When were the drugs ingested?
Every effort should be made to obtain this information, particularly if the patient is unable to provide a history due to alterations in mental status. (See "General approach to drug poisoning in adults", section on 'Initial evaluation and treatment'.)
Patients with intentional ingestions often develop greater toxicity than those with accidental exposures, so it is important to ask the patient directly about his or her intentions. Information obtained from the overdose patient can be unreliable, and confirmation should be sought if available. Additional history, including comorbid conditions and signs and symptoms, should be elicited from the patient, family members, friends, emergency medical services (EMS) personnel, and police.
Physical examination — Examination begins with assessment and stabilization of the airway, breathing, and circulation. Next, the clinician should attempt to identify any possible toxidromes (table 3). Key elements of the physical examination include vital signs, mental status, pupil size, skin dryness and color, and bowel sounds. (See "General approach to drug poisoning in adults", section on 'Initial evaluation and treatment'.)
Clinicians should look specifically for evidence of serotonin toxicity, including agitated delirium, hyperthermia, diaphoresis, hypertonia, hyperreflexia, tremor, ankle clonus, and slow, continuous, horizontal eye movements (referred to as ocular clonus). Neuromuscular findings are generally more pronounced in the lower extremities [4]. (See "Serotonin syndrome (serotonin toxicity)".)
Overview of SSRI toxicity — Many studies comparing the ingestion of selective serotonin reuptake inhibitors (SSRIs) with other antidepressants have found that SSRIs rarely cause fatalities or serious sequelae [6-10]. SSRIs are significantly less toxic than tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs). As an example, 31 to 32 percent of TCA ingestions require intubation compared with 4 to 6 percent of SSRI ingestions. Similarly, 31 to 35 percent of TCA ingestions require intensive care unit admission compared with 0 to 6 percent of SSRI ingestions [6,8,11]. One study found an incidence of 1.6 deaths per million SSRI prescriptions compared with 34.8 for TCAs and 20 for MAOIs [10].
It is also important to note the potential disparity in toxicity between SSRIs and serotonin and norepinephrine reuptake inhibitors (SNRIs). SNRIs such as venlafaxine have been associated with greater risk of significant toxicity and mortality in overdose. (See "Serotonin-norepinephrine reuptake inhibitors: Pharmacology, administration, and side effects".)
SSRIs (except for citalopram and escitalopram) have a wide therapeutic window: ingestion of up to 30 times the daily dose typically produces minor or no symptoms, while ingestion of 50 to 75 times the daily dose can cause vomiting, mild central nervous system (CNS) depression, or tremor. Most fatalities are reported with either extremely large doses (greater than 150 times the daily dose) or with the presence of coingestants such as ethanol or benzodiazepines [12]. (See "Ethanol intoxication in adults" and "Benzodiazepine poisoning and withdrawal".)
Serotonin syndrome rarely occurs after isolated SSRI ingestions, but severe episodes can occur following either a mixed serotonergic ingestion or changes made in therapeutic SSRI dosing. Serotonin syndrome caused by isolated SSRI ingestion tends not be severe. (See "Serotonin syndrome (serotonin toxicity)".)
SSRI ingestions are unlikely to cause seizures or severe CNS depression [6,11]. Seizures have been noted to occur in 1 to 2 percent of SSRI overdoses, but they are typically short and self-limited [11]. However, based on United States poison center data, seizure is a relatively common consequence of isolated citalopram overdose [13].
SSRIs do not have significant cardiotoxicity. Unlike the TCAs, they are typically not associated with QRS prolongation or ventricular dysrhythmias. In a review of 469 isolated SSRI ingestions, bradycardia occurred in 8 percent of cases and tachycardia in 15 percent [11]. Citalopram, which is structurally unlike the other SSRIs, is the most cardiotoxic: ingestion is associated with prolongation of the corrected QT (QTc) interval, which can predispose a patient to torsade de pointes (see 'Citalopram' below). Escitalopram ingestions too can cause similar cardiotoxicity. (See 'Escitalopram' below.)
Since SSRIs have little or no antagonism of muscarinic, histaminic, or adrenergic receptors, they do not typically cause anticholinergic symptoms, significant sedation, or hypotension.
In therapeutic dosing, SSRIs (specifically paroxetine) may cause hyponatremia, most likely via the syndrome of inappropriate antidiuretic hormone secretion (SIADH) [14-16]. Older adults appear to be at greatest risk of this complication. (See "Pathophysiology and etiology of the syndrome of inappropriate antidiuretic hormone secretion (SIADH)".)
Serotonin syndrome — Serotonin syndrome or toxicity is a clinical constellation of autonomic instability, mental status change, and increased neuromuscular tone caused by excessive stimulation of 5-HT1A and 5-HT2A receptors. The term "serotonin toxicity" implies excess serotonin activation, which manifests itself as a serotonin toxidrome. The term "serotonin syndrome" is also used to describe moderate to severe serotonin toxicity.
It is typically caused by combining two or more serotonergic medications. Historically, MAOIs have caused most cases, but SSRI interactions have become far more prevalent.
Serotonin syndrome often occurs several days to one week after the addition of a new serotonergic agent or after the dose of a serotonergic agent is increased [4,17]. Patients with SSRI overdose have a 10 to 14 percent incidence of serotonin syndrome, but many of these are mild presentations [6,11]. In a study of 469 isolated SSRI ingestions, sertraline, paroxetine, and fluvoxamine were most likely to cause a serotonin syndrome [11].
The spectrum of clinical findings to diagnose serotonin toxicity include spontaneous or inducible muscle clonus, ocular clonus, agitation, diaphoresis, tremor, hyperreflexia, muscle rigidity, and hyperthermia. The syndrome can manifest a wide range of toxicity, and prognosis correlates with clinical findings. Patients with severe disease may develop acute hyperthermia (core temperature >41.1°C), hypertension, tachycardia, agitated delirium, and muscle rigidity. Unless aggressive intervention is instituted promptly, these patients can go on to develop rhabdomyolysis, metabolic acidosis, renal failure, seizures, disseminated intravascular coagulation, and shock. Many of the signs and symptoms of excessive serotonin activation are also present in other conditions or ingestions, and several are sufficiently specific (eg, leg hyperreflexia and clonus) for serotonin toxicity. A full discussion of serotonin syndrome is found separately. (See "Serotonin syndrome (serotonin toxicity)".)
TOXICITY OF SPECIFIC AGENTS
Fluoxetine — Fluoxetine has been available longer than the other selective serotonin reuptake inhibitors (SSRIs), and many studies and reports exist supporting its safety in overdose [12,18-21]. Approximately half of fluoxetine ingestions remain asymptomatic, while the most frequently reported symptoms are nausea, vomiting, somnolence, tremor, agitation, and tachycardia. Most symptoms are benign and resolve quickly. Serious toxicity, such as dysrhythmias, seizures, central nervous system (CNS) depression, and respiratory arrest, is rare but can occur, especially when coingestants such as ethanol are present [12,20]. Serious toxicity is typically attributed to the coingestant and not to fluoxetine.
A prospective study assessing toxicity found fluoxetine safer than tricyclic antidepressants (TCAs) [22]. Two deaths occurred out of 124 patients in the TCA group; no deaths occurred out of 16 patients in the fluoxetine group. Moreover, no fluoxetine patient developed seizures, QRS prolongation, hypotension, or serotonin syndrome; or required intubation. Mild CNS depression was seen in five of the fluoxetine overdoses.
Fluoxetine overdose is well tolerated, even in very large amounts. Acute ingestions of 2 to 3 g typically cause only minimal symptoms [19,23]. There are several reports of death attributed to fluoxetine overdose, but they involved ingestions of 6 to 12 g [24-26]. Fluoxetine serum concentrations were 2300 to 6000 mcg/L in these patients, which is greater than most previously reported levels and significantly greater than the typical steady-state therapeutic concentration of 250 mcg/L. Fluoxetine has been associated with a 1 percent incidence of developing serotonin syndrome in isolated ingestion, but rarely are cases severe [11]. Most fatalities associated with fluoxetine overdose involve coingestants [24].
Case reports suggest that seizure should be considered a possible, albeit rare, complication of fluoxetine overdose [11,19,24,27-30]. In two cases, the onset of seizure was delayed 9 to 16 hours after the ingestion [28,30]. However, in most cases, proconvulsant coingestants were either present or could not be excluded, and seizures tended to be brief and self-limited. During clinical trials, seizures rarely occurred in patients given therapeutic doses of fluoxetine [19,31]. In three prospective or cohort studies of fluoxetine ingestions, only four seizures were observed in a pooled sample of 325 patients [11,20,22].
Fluoxetine has minimal cardiotoxicity. Prolongation of the corrected QT (QTc) interval of no greater than 450 msec was noted in 62 percent of fluoxetine overdose patients in one study, but none developed a ventricular dysrhythmia [22]. A prospective study of 32 fluoxetine ingestions found no electrocardiogram (ECG) abnormalities [21]. Rare reports exist of sinus bradycardia, supraventricular tachycardia, atrial fibrillation, atrial flutter, ST segment depression, and ventricular tachycardia associated with fluoxetine [11,29,32-36]. There is one report of QT prolongation to 600 msec and torsade de pointes requiring cardioversion in a 74-year-old woman three weeks after starting therapeutic fluoxetine [37]. In another report, a patient who ingested 2.4 g of fluoxetine developed a TCA-like QRS interval prolongation that responded to sodium bicarbonate administration [30]. A patient who ingested 12 g of fluoxetine developed bradycardia, which progressed to ventricular fibrillation and asystole [24]. Nevertheless, in both therapeutic use and overdose, fluoxetine-associated cardiotoxicity remains rare.
Sertraline — Clinical trial and poison center data, various case reports, and several small studies suggest that sertraline is as safe in overdose as fluoxetine [12,38-41]. In a prospective poison center study of sertraline ingestions, 10 of 17 isolated overdoses had no signs or symptoms [41]. Another poison center study found that 34 out of 52 isolated sertraline ingestions were asymptomatic [40]. In these studies, the most common symptoms were nausea, tremor, and lethargy, while less common findings were agitation, vomiting, tachycardia, and bradycardia. Most symptoms resolved spontaneously within six hours [40,41]. Rare reports exist of brief, self-limited seizures and QTc prolongation [11,12]. Sertraline has been associated with a 20 percent incidence of developing serotonin syndrome in isolated ingestion, but rarely are cases severe [11].
Paroxetine — As with the other SSRIs, paroxetine is generally safe in overdose [12,42]. A poison center review noted that 8 out of 35 isolated paroxetine ingestions had no symptoms, while common symptoms included vomiting, drowsiness, tremors, dizziness, and sinus tachycardia [43]. Paroxetine has been associated rarely with brief, self-limited seizures and with an 18 percent incidence of developing serotonin syndrome in isolated ingestion, but rarely are cases severe [11,12].
Of note, among SSRIs, paroxetine has the highest rate of discontinuation syndrome because of its short half-life and lack of active metabolites. Symptoms are usually mild, lasting one to two weeks; they include nausea, dizziness, bad dreams, paresthesia, and a flu-like illness. Severe symptoms can occur and persist for prolonged periods and may require restarting paroxetine [44]. (See "Switching antidepressant medications in adults".)
Fluvoxamine — Fluvoxamine is generally safe in overdose. In a poison center study of 299 fluvoxamine ingestions, toxicity was benign when the dose ingested was less than 1000 mg [45]. Common symptoms of mild to moderate ingestions were drowsiness, tremor, diarrhea, vomiting, abdominal pain, dizziness, mydriasis, and sinus tachycardia. At higher doses, there were five seizures reported, as well as deeper levels of sedation, hypokalemia, hypotension, sinus bradycardia (never requiring treatment), and repolarization disorders. Thirteen deaths were reported, but they could all be attributed to coingestants or other medical problems. Among the 55 patients in whom fluvoxamine was the sole ingestant, only mild to moderate symptoms were evident, and the authors concluded that isolated overdoses of fluvoxamine rarely cause severe toxicity. In another study, fluvoxamine was associated with brief, self-limited seizures and with a 17 percent incidence of developing serotonin syndrome in isolated ingestion [11]. Only rarely were these episodes of serotonin syndrome severe.
Fluvoxamine is associated with a discontinuation syndrome; symptoms can include dizziness, headache, nausea, and irritability [46]. (See "Switching antidepressant medications in adults".)
Citalopram — Citalopram has the greatest potential for serious toxicity compared with other SSRIs. Citalopram is a racemic bicyclic phthalane derivative that is structurally different from the other SSRIs. It acts as a highly selective serotonin reuptake inhibitor but, unlike other SSRIs, also has antihistaminergic properties [47].
In citalopram ingestions of less than 600 mg, symptoms tend to be mild and include nausea, dizziness, tachycardia, tremor, and somnolence [48,49]. With larger ingestions, citalopram can cause serious cardiac and neurologic toxicity. It has been associated with a 9 percent incidence of serotonin syndrome in isolated ingestion, but rarely are cases severe [11]. In rare instances, citalopram ingestions have been associated with metabolic acidosis [50].
Cardiotoxicity — Anecdotal evidence and cohort analysis suggest that citalopram is the most cardiotoxic SSRI. With moderate (approximately 600 mg) to large (approximately 1900 mg) ingestions, citalopram has been associated with a variety of cardiac conduction disturbances.
Citalopram is hepatically metabolized to desmethylcitalopram (DCT) and the likely cardiotoxic metabolite didesmethylcitalopram (DDCT). Animal studies have shown a correlation between high serum levels of DDCT and both prolonged QTc and death [51].
In therapeutic dosing, citalopram has not been found to prolong the QTc interval, nor was there a relationship noted between DDCT concentration and the QTc interval [52]. However, in this study, the highest recorded serum DDCT concentration was approximately one-hundredth (1/100) of the levels measured in overdose patients who developed QTc prolongation [53]. Therefore, it is difficult to extrapolate these results to poisoned patients.
A poison center review of citalopram ingestions in Swedish hospitals identified 44 cases without coingestants [48]. Of these, 6 of 18 patients who ingested greater than 600 mg developed a widened QRS complex, while all five patients who ingested greater than 1900 mg had either QRS widening, a seizure, or both. QRS changes were not seen immediately and resolved within 24 hours. None of the patients had a clinically significant dysrhythmia. A follow-up to this study included 108 cases, in which 25 percent of patients had either ST segment changes or widening of ECG intervals, but again, none had a clinically significant dysrhythmia [49]. Additional reviews have noted similar findings [53,54].
In a logistic regression analysis of 57 citalopram ingestions compared with 240 other SSRI ingestions, citalopram was five times more likely to prolong the QTc interval beyond 440 msec [11]. In the same study, the QTc was prolonged beyond 440 msec in 68 percent of cases and beyond 500 msec in 12 percent. QTc interval prolongation has been seen in patients who ingested as little as 400 mg [55,56]. In one of these reports, QTc prolongation was noted two hours after overdose, but it did not reach its peak until 13 hours [55]. There are multiple reports of citalopram ingestions and therapeutic dosing resulting in torsade de pointes [57-63]. In one of these reports, a patient developed intermittent torsades requiring intervention 36 hours after an ingestion of 1000 mg of citalopram [58].
Other cardiac changes noted with citalopram ingestions include left bundle branch block [11,49,56,64], right bundle branch block [49], ventricular extra beats [48], sinus bradycardia [65], sinus tachycardia [53], supraventricular tachycardia [66], ST segment changes [48,53], and several cases of ventricular fibrillation [49,67].
The clinical significance of citalopram's effects on cardiac conduction is unclear. Multiple fatalities have been attributed to overdoses of citalopram, but most involved coingestants [12,47,68]. In one report, a patient suffered a seizure and torsades de pointe and then expired from a pulseless electrical activity (PEA) arrest after an isolated citalopram ingestion [63]. Although citalopram ingestions can cause a wide range of cardiac disturbances, patients typically do well with minimal intervention. Most patients with QRS or QTc prolongation recover within 24 hours with either supportive care or with bicarbonate infusion (to treat QRS prolongation).
Neurotoxicity — In clinical trials of therapeutic dosing, only 0.3 percent of patients treated with citalopram had a seizure, compared with 0.5 percent of patients taking placebo [51]. There are many reports, however, of citalopram ingestions causing seizures [11,48,49,53,54,56,63,64,66]. Although one head-to-head comparison suggested that seizures are no more likely following ingestions of citalopram versus other SSRIs [11], other reviews are contradictory and placed the incidence of citalopram-induced seizures at 6 to 14 percent [49,54]. Citalopram is a common cause of seizures in single-agent ingestions reported to poison centers [13]. The risk of seizure appears to increase with the ingested dose. In a review of 108 patients, seizures were seen in 18 percent of patients who ingested 600 mg to 1900 mg but in 47 percent of patients who ingested over 1900 mg [49]. In another case series, four out of five patients who ingested greater than 1400 mg developed seizures [53]. The majority of citalopram-induced seizures are brief, self-limited, and unlikely to result in neurologic sequelae, although recurrent seizures may occur occasionally [69].
Escitalopram — Citalopram is a racemic mixture of R- and S-citalopram; escitalopram is the pharmacologically active S-enantiomer and a more potent inhibitor of serotonin reuptake [70,71]. Escitalopram toxicity appears to mimic that of citalopram and consists primarily of serotonin syndrome and QTc interval prolongation. Compared with citalopram, escitalopram seems to have greater serotonergic toxicity and less cardiac toxicity [71]. (See 'Citalopram' above.)
No serious toxicities or fatalities due to overdose were reported in clinical trials of escitalopram efficacy [72,73]. However, postmarketing surveillance data includes reports of seizures, QTc prolongation, ventricular dysrhythmias, and serotonin syndrome, albeit without demonstration of a causal relationship [72]. There is a case report of an escitalopram ingestion causing both QRS and QTc widening that responded to bicarbonate and magnesium infusions [74]. Escitalopram appears to cause fewer seizures after overdose than does citalopram [75].
In overdose, escitalopram may cause serotonin toxicity and QTc prolongation, but there are no reports of torsades de pointes [72,76-82]. The best data related to overdose comes from a retrospective review of 79 escitalopram exposures. The authors report that significant serotonin toxicity (eg, hyperthermia) occurred in 15 percent of patients who ingested escitalopram only, while isolated serotonergic neuromuscular findings (eg, hyperreflexia) developed in as many as 46 percent. Tachycardia (42 percent), bradycardia (14 percent), and QTc interval abnormalities (14 percent) were also noted. A review of poison center data noted similar toxicity in addition to somnolence, vomiting and tremors [75].
DIAGNOSIS — The diagnosis of selective serotonin reuptake inhibitor (SSRI) poisoning is made clinically on the basis of the history and clinical presentation. Serum drug concentrations are generally not helpful in assessing SSRI poisoning and are not routinely available in most hospital clinical laboratories. Isolated SSRI ingestions generally produce only mild symptoms, although SSRI intoxication can infrequently produce serotonin syndrome, seizures, central nervous system (CNS) depression, or cardiac conduction abnormalities.
DIFFERENTIAL DIAGNOSIS — Significant toxicity in an isolated selective serotonin reuptake inhibitor (SSRI) ingestion is unlikely. If a patient appears sick or comatose, poisoning with a coingestant or an underlying non-toxicologic condition is more likely to be responsible. Diagnosis of the poisoned patient and comatose patient are reviewed separately. (See "General approach to drug poisoning in adults" and "Approach to the child with occult toxic exposure" and "Initial management of the critically ill adult with an unknown overdose" and "Stupor and coma in adults" and "Evaluation of stupor and coma in children".)
LABORATORY EVALUATION — Specific serum drug concentrations are generally not helpful in assessing or managing selective serotonin reuptake inhibitor (SSRI) ingestions. As examples, one study of antidepressant overdose found no correlation between fluoxetine levels and mental status, while a study of fluvoxamine overdose found no correspondence between drug levels and symptom severity [22,45].
In the setting of SSRI ingestion, routine laboratory evaluation should be targeted at possible coingestants and should include the following:
●Fingerstick glucose to rule out hypoglycemia as the cause of any alteration in mental status
●Acetaminophen and salicylate concentrations to rule out these common coingestions
●Ethanol concentration; ethanol coingestion is associated with more severe toxicity
●Electrocardiogram (ECG) to rule out conduction system poisoning by drugs that effect the QRS or corrected QT (QTc) intervals
●Serum bicarbonate to screen for metabolic acidosis
●Pregnancy test in women of childbearing age
In patients who develop serotonin syndrome with increased muscular tone or hyperthermia, the following laboratory information should be obtained:
●Creatine kinase and urine myoglobin to evaluate for rhabdomyolysis (see "Rhabdomyolysis: Clinical manifestations and diagnosis")
●Serum creatinine to assess for kidney damage
●Serum aminotransferase concentrations to assess for liver injury
●Coagulation studies (ie, activated partial thromboplastin time [aPTT], prothrombin time [PT], international normalized ratio [INR], platelet count, D-dimer) to assess for disseminated intravascular coagulation (see "Evaluation and management of disseminated intravascular coagulation (DIC) in adults")
●Venous or arterial blood gas to assess for metabolic acidosis
MANAGEMENT — A summary table to facilitate emergency management of selective serotonin reuptake inhibitor (SSRI) poisoning is provided (table 1). Since most SSRI ingestions develop minimal or no toxicity, the broad goal of management should be to provide supportive care and to minimize potentially unneeded and harmful interventions (see "General approach to drug poisoning in adults"). Serotonin agonists, such as meperidine, must be avoided in patients with SSRI ingestions or in those who have serotonin syndrome so as not to precipitate worse toxicity. Most patients with SSRI ingestions have good outcomes with supportive care alone. Victims of minor, accidental exposures that are asymptomatic can be observed at home.
Airway, breathing, and circulation — The airway should be secured with endotracheal intubation in patients who are obtunded and cannot protect their airway. Hypoventilation, hypoperfusion, and unstable dysrhythmias must be addressed early according to standard advanced cardiac life support (ACLS) protocols (see "Advanced cardiac life support (ACLS) in adults"). For wide QRS complex tachycardia, sodium bicarbonate boluses can be given, followed by a bicarbonate infusion if there is subsequent narrowing of the QRS complex. (See "Tricyclic antidepressant poisoning", section on 'Sodium bicarbonate for cardiac toxicity'.)
Seizures — Seizures typically respond to treatment with benzodiazepines such as lorazepam or diazepam.
Decontamination — A single dose of activated charcoal should be administered for gastrointestinal decontamination if the patient presents promptly after ingestion and has no contraindications, such as altered mental status or vomiting. The typical dose is 50 g in adults and 1 g/kg in children. The greatest benefit from charcoal decontamination occurs if given within one to two hours following the ingestion [83]. In a pharmacokinetic study of sertraline overdose, a single dose of activated charcoal was found to increase clearance even up to four hours after ingestion [84]. Given the typical benign course of SSRI ingestions, activated charcoal should not be forced into a patient (such as via nasogastric tube) if they are refusing to drink it.
In retrospective observational studies, a single dose of activated charcoal was associated with a lower rate of corrected QT (QTc) interval prolongation in citalopram and escitalopram overdose [85,86].
Due to the rapid absorption of SSRIs, there is no role for multiple-dose charcoal or charcoal additives such as magnesium citrate or sorbitol. Charcoal should be withheld in patients who are sedated and may not be able to protect their airway, unless endotracheal intubation is performed first. However, endotracheal intubation should not be performed solely for the purpose of giving charcoal. (See "Gastrointestinal decontamination of the poisoned patient".)
There is no role for extracorporeal removal techniques such as hemodialysis because SSRIs are highly bound to serum proteins.
Serotonin syndrome — Patients who develop serotonin syndrome are treated with supportive care, discontinuation of the offending agent, and, depending upon the severity of illness, a serotonin antagonist such as cyproheptadine. Severe cases of serotonin syndrome that develop muscular rigidity and a core temperature greater than 41ºC must be treated aggressively with neuromuscular paralysis, endotracheal intubation, and external cooling to prevent complications. A complete discussion of serotonin syndrome is found elsewhere. (See "Serotonin syndrome (serotonin toxicity)".)
QTc interval prolongation — All patients with SSRI ingestion and a prolonged QTc interval should be placed on a cardiac monitor and have serial electrocardiograms (ECGs) to be certain the interval is not continuing to lengthen. A reasonable approach is to perform a repeat ECG after one hour if the ingested agent may cause QRS prolongation, and again after four to six hours if concern for QTc prolongation persists. During this time, the patient should remain on a cardiac monitor to exclude the development of dysrhythmias, as ECGs only help determine the predisposition for a dysrhythmia.
A prolonged QTc interval predisposes to torsade de pointes, an intermittent ventricular dysrhythmia that can, albeit rarely, degenerate into potentially life-threatening monomorphic ventricular tachycardia or fibrillation. If on any subsequent ECG the QTc is longer than previous, the patient should remain on a cardiac monitor and another ECG performed in four to six hours. This is especially important for agents that have cardiotoxic metabolites, such as citalopram. (See 'Cardiotoxicity' above.)
If the QTc interval is significantly prolonged (greater than 560 msec) and other concerning signs are present (eg, frequent premature ventricular contractions) or magnesium therapy is instituted, serial ECGs should be continued. As the QTc interval narrows, ECGs can be obtained every four to eight hours until the interval normalizes (less than 440 msec). Patients who will not be admitted should have a final ECG prior to discharge that demonstrates normal intervals.
Warning signs indicating an increased risk of torsade de pointes (torsades) include: a QTc interval greater than 560 msec, previous history of torsades, bradycardia, increased frequency and complexity of premature ventricular complexes/contractions (PVCs; also referred to as premature ventricular beats or premature ventricular depolarizations), or a PVC falling on the T-wave [87]. Magnesium therapy to prevent torsades should be given to patients with any of these findings. The dose of magnesium sulfate in adults is 2 g over two minutes, which can be repeated after 10 to 15 minutes if the rhythm is not terminated. If there is clinical effect, a magnesium infusion at a rate of 2 to 10 mg/minute may be helpful until the QTc interval is less than 500 msec.
Observation — Patients can be referred to psychiatry after a four- to six-hour observation period if they are asymptomatic and have not developed any ECG changes, particularly prolongation of the QTc. Massive ingestions (ie, those greater than 150 times the daily dose of the SSRI) should be admitted for 24 hours of cardiac monitoring, since these are at greatest risk for severe toxicity.
Because citalopram and escitalopram potentially have a cardiotoxic metabolite, patients with these ingestions should receive an ECG at least six hours after their ingestion. If these patients have any signs of QTc or QRS prolongation, have an increasing QTc interval compared with the initial ECG, or have experienced a dysrhythmia, they should be admitted for cardiac monitoring and observation until their ECG intervals normalize.
PEDIATRIC CONSIDERATIONS — Pediatric ingestions of selective serotonin reuptake inhibitors (SSRIs) are generally well tolerated [18]. Fluoxetine caused either no or minimal symptoms (sleepiness, hyperactivity, and diarrhea) in 20 pediatric ingestions in one prospective study [20]. A retrospective poison center review identified 120 cases of isolated fluoxetine ingestions in children less than six years of age; 116 were asymptomatic, three had emesis, and one was sedated [88]. Although most fluoxetine ingestions are reported to be benign, intentional large ingestions can lead to serotonin toxicity [89]. Sertraline and paroxetine also produce minimal toxicity. Several studies noted that most patients are asymptomatic; those that are symptomatic most commonly manifest central nervous system (CNS) stimulation, drowsiness, dizziness, vomiting, or sinus tachycardia [40,43,90,91].
Few data exist on the safety of fluvoxamine in pediatric ingestions. There is one case report of a child of four years requiring intubation for CNS depression and vasopressor support for hypotension following a 400 mg ingestion [92]. There are several reports of sertraline and fluvoxamine ingestions causing serotonin syndrome in children [93-95].
Case reports suggest that cyproheptadine can be used to treat serotonin syndrome in children [93]. The pediatric dose of cyproheptadine is 0.25 mg/kg per day divided twice or three times daily up to a maximum dose of 12 mg [96].
The relationship between SSRIs and increased risk of suicide among pediatric patients is discussed elsewhere. (See "Effect of antidepressants on suicide risk in children and adolescents".)
Maternal fluoxetine and paroxetine use has been associated with a hyperserotonergic state in neonates and in breast-feeding infants [97,98]. In newborns, serotonergic symptoms can manifest as fever, somnolence, jitteriness, irritability, tremors, and a hyperactive Moro (ie, startle) reflex. (See "Antenatal use of antidepressants and the potential risk of teratogenicity and adverse pregnancy outcomes: Selective serotonin reuptake inhibitors", section on 'Selective serotonin reuptake inhibitors'.)
ADDITIONAL RESOURCES
Regional poison control centers — Regional poison control centers in the United States are available at all times for consultation on patients with known or suspected poisoning, and who may be critically ill, require admission, or have clinical pictures that are unclear (1-800-222-1222). In addition, some hospitals have medical toxicologists available for bedside consultation. Whenever available, these are invaluable resources to help in the diagnosis and management of ingestions or overdoses. Contact information for poison centers around the world is provided separately. (See "Society guideline links: Regional poison control centers".)
Society guideline links — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: General measures for acute poisoning treatment" and "Society guideline links: Treatment of acute poisoning caused by specific agents other than drugs of abuse".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Serotonin syndrome (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Overview – Selective serotonin reuptake inhibitors (SSRIs) are generally well tolerated in overdose and are much safer than other antidepressant medications such as tricyclic antidepressants (TCAs). Toxicity is most commonly due to co-ingestants. Victims of intentional ingestions often develop greater toxicity than accidental exposures. A summary table to facilitate the emergency management of SSRI poisoning is provided (table 1). (See 'Introduction' above.)
●Clinical features of overdose – Most SSRI ingestions develop minimal or no toxicity. SSRIs rarely produce fatality or serious sequelae, mainly due to their wide therapeutic window: Ingestion of up to 30 times the daily dose typically produces minor or no symptoms, while ingestion of 50 to 75 times the daily dose can cause vomiting, mild central nervous system (CNS) depression, or tremor. Most fatalities are reported with either extremely large doses (greater than 150 times the daily dose) or with the presence of co-ingestants such as ethanol, benzodiazepines, or TCAs. (See 'Clinical features of overdose' above.)
•Serotonin syndrome – This complication rarely occurs after isolated SSRI ingestions, but severe episodes can occur following either a mixed serotonergic ingestion or changes made in therapeutic SSRI dosing. Serotonin syndrome caused by isolated SSRI ingestion tends not be severe. (See "Serotonin syndrome (serotonin toxicity)".)
•Dysrhythmias and seizures – Although rare, SSRI ingestions can cause seizures or prolonged electrocardiogram (ECG) intervals, predisposing to dysrhythmias. Citalopram and escitalopram are most likely to cause abnormal cardiac conduction. (See 'Citalopram' above and 'Escitalopram' above.)
●Laboratory evaluation – Specific serum drug concentrations are generally not helpful in assessing or managing SSRI ingestions. Routine laboratory evaluation should be targeted at possible co-ingestants such as acetaminophen, salicylates, and ethanol. (See 'Laboratory evaluation' above.)
●Management – Assess and stabilize airway, breathing, and circulation. Since toxicity tends to be mild, the broad goal of management should be to provide supportive care and to minimize potentially unneeded and harmful interventions. A summary table to facilitate the emergency management of SSRI poisoning is provided (table 1). (See 'Management' above.)
•Decontamination – We suggest that a single dose of activated charcoal be administered for gastrointestinal decontamination (Grade 2B). The typical dose is 50 g in adults and 1 g/kg in children (see "Gastrointestinal decontamination of the poisoned patient"). The greatest benefit from charcoal decontamination occurs if given within one to two hours following the ingestion. Charcoal should be withheld in patients who are sedated and may not be able to protect their airway unless endotracheal intubation is performed first. Endotracheal intubation should not be performed solely for the purpose of giving charcoal. There is no role for extracorporeal removal techniques such as hemodialysis. (See 'Decontamination' above.)
Because of the potential risk of aspiration and the lack of benefit, we recommend that multiple-dose charcoal and charcoal additives such as magnesium citrate or sorbitol not be given (Grade 1C).
•Widened QRS – For wide QRS complex tachycardia associated with ingestion of these agents, we suggest treatment with boluses of sodium bicarbonate, followed by a bicarbonate infusion, if there is subsequent narrowing of the QRS complex (Grade 2C). (See "Tricyclic antidepressant poisoning", section on 'Sodium bicarbonate for cardiac toxicity'.).
•Seizures – Seizures typically respond to treatment with benzodiazepines such as lorazepam or diazepam.
•Widened QTc – All patients with SSRI ingestion and a prolonged corrected QT (QTc) interval should have serial ECGs to be certain the interval is not continuing to lengthen and be observed on a cardiac monitor. (See 'QTc interval prolongation' above.)
