Allergic Reactions and Angioedema
Allergic angioedema is a medical emergency that necessitates airway protection and treatment with antihistamines or possibly steroids. Psychotropic drugs that cause angioedema should be immediately discontinued, and rechallenges generally should not be undertaken.
Adverse cutaneous reactions have been reported to occur in up to 5% of individuals who receive antipsychotic medications. True allergic reactions are IgE–mediated phenomena. Angioedema refers to the relatively rapid development of facial edema with swelling of oropharyngeal mucosal membranes and possible airway constriction. Allergic angioedema is similar to anaphylactic shock and may involve urticarial eruptions. In the presence of stridor or other signs of respiratory distress, allergic angioedema constitutes a medical emergency that may require intubation to maintain an intact airway.
Case reports have described the occurrence of urticaria and angioedema with asenapine, bupropion, lurasidone, oxcarbazepine (oxcarbazepine-related angioedema has been determined to arise in 9.8 per 1 million pediatric cases; Knudsen et al. 2007), paroxetine, risperidone (with uneventful rechallenge), and ziprasidone.
Among patients with allergies to sulfonamide antibiotics such as sulfamethoxazole-trimethoprim (so-called sulfa allergies), there has been debate and uncertainty about the potential for allergic cross-reactivity to non-antibiotic sulfonamide drugs, including certain anticonvulsants (e.g., zonisamide) or carbonic anhydrase inhibitors (e.g., topiramate) (Wulf and Matuszewski 2013). Most experts argue that in patients with known sulfa antibiotic allergies, type I immediate hypersensitivity reactions to nonantibiotic sulfonamides would be unlikely because of stereospecific differences in molecular structure between sulfonamide antibiotics and nonantibiotics.
Antiepileptic Hypersensitivity Reactions
Hypersensitivity reactions should be suspected in anticonvulsant recipients who develop systemic, multiorgan system disturbances, regardless of the presence or absence of a skin rash. Hypersensitivity reactions are potentially life threatening, and the suspected causal agent should be immediately discontinued. Prompt medical attention involves supportive measures and a possible role for steroids. The suspected causal agent should not subsequently be reintroduced.
Certain anticonvulsants (and a select number of other agents) have been associated with rare idiosyncratic hypersensitivity reactions (often referred to in the literature as drug rash [or reaction] with eosinophilia and systemic symptoms [DRESS] syndrome), which have been described most extensively with carbamazepine, lamotrigine, phenytoin, and phenobarbital. Through mechanisms that are not well understood, such reactions are thought to involve an immunological response to toxic effects of the parent compound or a metabolite. When hypersensitivity reactions occur, they usually do so within the first 8 weeks of treatment. Characteristic features involve fever, multiorgan system involvement, and possible skin eruptions. Associated phenomena may include rhabdomyolysis (identifiable by myoglobinuria on urinalysis), which in itself should prompt a review of all medications that are known to cause this phenomenon, including statins and antiparkinsonian agents.
In August 2010, the FDA issued a warning that aseptic meningitis may arise as a type of rare hypersensitivity reaction to lamotrigine, particularly during the first few months after treatment initiation. This warning was based on 40 reported postmarketing cases from December 1994 through November 2009. Symptoms typically arose within days 1–42 after drug initiation (mean=16 days). Accordingly, the onset of meningeal signs (headache, fever, nausea, vomiting, nuchal rigidity, photophobia, and myalgias) during this time frame likely warrants cessation of therapy. Symptoms typically resolve after drug discontinuation.
Body Temperature Dysregulation
Patients who take FGAs or SGAs should be counseled about the potential disruption to maintaining their core body temperature when they are exposed to ambient extremes.
Both hypo- and hyperthermia can result from the use of FGAs and SGAs, most often arising near the time of treatment initiation or dosage increases. Antipsychotics disrupt the medullary chemoreceptor trigger zone, which among other functions, maintains homeostatic core body temperature. Pharmacological mechanisms thought to impair thermoregulation and thereby cause poikolothermia include the blockade of D1 and D2 dopamine receptors, as well as possible disruption of compensatory peripheral vasoconstriction due to α1-adrenergic blockade by most antipsychotics. Antipsychotic-induced temperature dysregulation may lead to medical hospitalization (nearly 70% of cases) or mortality (~4% of cases) (van Marum et al. 2007).
Patients who take FGAs or SGAs should be warned that their body temperature could rise significantly under conditions such as strenuous exercise or exposure to hot climates and should be instructed to monitor their body temperature periodically in such circumstances, as well as in the setting of infection. Management of hyperthermia typically involves hydration, acetaminophen, and cooling blankets or ice packs if necessary, along with removing the patient from exposure to high ambient temperature. Antipsychotic-induced hypothermia is rare and routine screening is considered unnecessary, although the possible presence of hypothermia should be considered in antipsychotic recipients who develop subjective coldness along with acute mental status changes, bradycardia, fatigue, and focal neurological signs, such as ataxia.
Cancer and Excess Mortality
Several large retrospective studies have drawn media attention after reporting an increased risk for cancer (e.g., Kao et al. 2012) or premature death (odds ratio=3.32 after adjustment for confounding factors; Weich et al. 2014) among patients taking benzodiazepines. In the largest of such studies, Kao et al. (2012) compared a database of 59,647 Taiwanese adults (mean age=48 years) who took benzodiazepines with age-matched controls; an overall 19% increased risk for developing cancer, particularly liver, prostate, bladder, and kidney cancers, was found among benzodiazepine recipients. Despite the fact that confounders such as age, smoking, and medical morbidity were controlled for, it is nevertheless difficult to draw definitive causal inferences from such retrospective database reviews given the inherent increased mortality associated with major psychiatric disorders such as schizophrenia, bipolar disorder, and major depression. In some studies, absence of information about causes of death (e.g., accidents, suicides) further limits the ability to discern plausible mechanistic explanations for increased mortality. Moreover, the nonrandomized designs of such chart reviews preclude knowledge about chronology and causality; in other words, while it is possible that benzodiazepines may cause cancer, it is no less plausible that cancer patients are especially likely to take benzodiazepines.
Withdrawal symptoms from abrupt cessation of most short-acting SSRIs and all SNRIs warrant gradual tapers that may require many days to weeks. Antidepressant withdrawal symptoms are medically benign but uncomfortable and can be managed either by extending the duration of tapering off the medication or by providing supportive pharmacotherapies as indicated (e.g., trimethobenzamide or promethazine for nausea; acetaminophen or ibuprofen for headaches or myalgias). Augmentation or switching an SNRI or short-acting SSRI to fluoxetine, followed by discontinuation of fluoxetine, may also help diminish the potential for symptoms of discontinuation syndrome.
In the middle to late 1990s, withdrawal syndromes were first described in the setting of abrupt cessation of short-acting SSRIs (see Table 2–2, Chapter 2, “Pharmacokinetics, Pharmacodynamics, and Pharmacogenomics”), with features including dizziness, insomnia, nervousness, nausea, or agitation; fluoxetine, by virtue of the long elimination half-life of its metabolite norfluoxetine, appears significantly less likely to cause this phenomenon than are other SSRIs (Rosenbaum et al. 1998). A review by Fava and colleagues (2015) found high variability in reported incident rates of SSRI discontinuation symptoms ranging from 6% to 47% of patients, with lower frequencies associated with longer-half-life SSRIs such as fluoxetine. No demographic or clinical characteristics appear to predict the likelihood of developing withdrawal symptoms after SSRI discontinuation, and withdrawal phenomena have been reported to persist for up to 3 weeks or longer following drug cessation, although symptoms usually resolve within 24 hours of SSRI resumption. Withdrawal features from the abrupt cessation (or sometimes even a few missed doses) of SNRIs can be even more substantial and protracted than occurs with SSRIs.
Two main strategies are usually advocated for managing SSRI or SNRI discontinuation syndromes. The first involves resuming either a discontinued agent or a dose that had been reduced and then undertaking a protracted taper of the existing antidepressant (per manufacturers’ recommendations)—in some instances, over the course of several weeks or longer, with supportive management of emergent withdrawal symptoms using antinausea drugs (e.g., trimethobenzamide, promethazine, prochlorperazine), meclezine for vertigo, and analgesics (e.g., acetaminophen, ibuprofen) as needed for headaches and myalgias. However, Fava and colleagues (2015) observed that gradual SSRI tapers did not reliably diminish the chances for discontinuation symptoms to occur. The second approach is to augment the initial antidepressant with fluoxetine for several days, then discontinue the original antidepressant entirely and capitalize on the long half-life of fluoxetine plus its metabolite, norfluoxetine, and then discontinue fluoxetine after several days with a lesser likelihood of withdrawal phenomena. Conceivably, a similar approach could be taken by switching an existing short t½ serotonergic antidepressant to vortioxetine, given its relatively long t½ of 66 hours—although the actual implementation of this strategy has not yet been reported in the literature.
Abrupt cessation of MAOIs has been associated with discontinuation phenomena that may include hallucinations, anxiety, agitation, paranoia, and delirium. It is therefore recommended that MAOIs be tapered over at least several days or more rather than being abruptly stopped, except in the setting of palpitations or frequent headaches that may be thought to reflect a hypertensive crisis.
Drug-Induced Lupus Erythematosus
Drug-induced lupus erythematosus (DILE) is rarely caused by psychotropic drugs. In suspected cases, symptoms mirror those seen with systemic lupus erythematosus (SLE) and involve flulike symptoms, myalgias, arthralgias, and fever. The causal agent should be discontinued and not subsequently reintroduced. Symptoms typically resolve within 1–2 weeks either spontaneously or with the use of NSAIDs or (more infrequently) oral steroids.
DILE is an autoimmune phenomenon with clinical features and laboratory findings similar to those of SLE. Symptoms may include flulike symptoms, myalgias, arthralgias, and fever. Rashes or other skin lesions (e.g., oral ulcers) are less common with DILE than with SLE. More severe cases of DILE can involve cardiac or pulmonary inflammation. By definition, DILE is triggered by the use of certain medications, including several psychotropic agents—each considered by the Lupus Foundation of America (www.lupus.org) to have a low or very low risk. From among the nearly 40 known drugs associated with DILE, cases most commonly result from the use of procainamide, hydralazine, and quinidine. Psychotropic agents that have rarely been associated with DILE include carbamazepine, oxcarbazepine, lithium, clonidine, pindolol, chlorpromazine, perphenazine, and phenelzine. Limited risk factors have been identified for developing DILE, including chronicity of drug therapy, male sex, age >50, and the presence of the so-called “slow acetylator” phenotype (see the section “Pharmacokinetics and Pharmacodynamics” in Chapter 2, “Pharmacokinetics, Pharmacodynamics, and Pharmacogenomics”).
Diagnosis of DILE depends on clinical presentation as well as corroboration from laboratory indices (notably, antinuclear antibody and antihistone antibodies). Treatment involves discontinuing the causal agent alongside supplemental NSAIDs or oral steroids, corticosteroid creams to treat skin rashes, and hydroxychloroquine to treat arthralgias. More rarely, immunosuppressants such as azathioprine or cyclophosphamide are also necessary. Reintroduction of a causal agent is generally not advised. DILE is an iatrogenic phenomenon that is fundamentally different from SLE. Long-term consequences or recurrences of DILE would not be expected unless the suspected causal agent was reintroduced.
Neuroleptic Malignant Syndrome
NMS can occur with any FGAs or SGAs and should be considered in any patient receiving an antipsychotic who develops a fever. Treatment involves immediate cessation of dopamine antagonists followed by supportive measures (e.g., hydration and autonomic monitoring).
NMS is a relatively rare adverse systemic reaction to any antipsychotic drugs (including antiemetic phenothiazines such as prochlorperazine and related compounds such as metoclopramide) as well as other non-neuroleptic drugs with antidopaminergic effects, such as phenelzine, some TCAs (e.g., desipramine, trimipramine), and lithium. NMS rarely arises beyond 1 month after initiation of an antidopaminergic drug. (Two-thirds of cases occur within the first 7 days of starting an antidopaminergic drug.) Although clear risk factors have not been empirically determined, some authors have observed that NMS may be more likely to occur in patients with psychomotor agitation or dehydration, in those who receive high doses of antipsychotics, and in recipients of frequent intramuscular injections of FGAs. Key symptoms include fever, muscle rigidity, acute mental status changes (e.g., delirium), autonomic instability, elevation of serum CK, tremor, and leukocytosis. Serum CK levels typically exceed 1,000 IU/L and in some cases may be as high as 100,000 IU/L (Levenson 1985). However, NMS is a clinical diagnosis that can manifest in various ways, and no one symptom is pathognomonic. The clinician must also recognize nonpsychotropic drugs that may cause myalgias with possible rhabdomyolysis (e.g., statins), which could be mistaken for the muscle rigidity of NMS. Indeed, it is critical for practitioners to differentiate NMS from other systemic drug reactions (e.g., serotonin syndrome [see the section “Serotonin Syndrome” below] or anticholinergic delirium) as well as other forms of delirium or encephalopathy, infectious etiologies, or heat stroke, and also to avoid mistaking it for manifestations of primary psychopathology (including catatonia).
NMS is a medical emergency. Treatment hinges on the prompt discontinuation of antipsychotics and other potential antidopaminergic drugs, supportive treatment of hyperthermia (e.g., cooling blankets, ice packs), and hydration (usually intravenous, both for circulatory support and to minimize the potential for kidney damage due to myoglobinuria). Intravenous use of the muscle relaxant dantrolene (1–2.5 mg/kg initially, followed by 1 mg/kg every 6 hours) is typically reserved for extreme hyperthermia and the persistent abnormalities of marked vital signs despite supportive care. In patients not responding to the above treatments, electroconvulsive therapy (ECT) is sometimes advised based on the rationale of its known efficacy for malignant catatonia (reviewed by Davis et al. ). A history of prior NMS may increase the likelihood of future episodes of NMS. There is no contraindication to resuming antipsychotic drugs after the resolution of NMS, although recurrence happens in up to 30% of individuals; SGAs are preferred over FGAs, and clinicians should use the lowest possible dosages and monitor carefully for early signs of reemergent NMS.
Serotonin syndrome is a medically emergent toxicity state resulting from excessive serotonergic activity, usually caused by an interaction among drugs that increase serotonin through different mechanisms. Management involves cessation of serotonergic drugs and supportive measures that fundamentally include hydration and airway monitoring and protection.
A constellation of signs and symptoms related to serotonergic hyperstimulation has come to be known as serotonin syndrome, which may be one of a family of CNS toxicity states that also includes NMS. The criteria classically described by Sternbach (1991) for defining serotonin syndrome include the following:
At least three of the following: agitation, ataxia, diaphoresis, diarrhea, hyperreflexia (particularly in lower extremities), mental status changes (may include hypervigilance, psychosis, confusion, or agitation), myoclonus, shivering, tremor, or hyperthermia
Emergence of signs and symptoms temporally following either the addition of a serotonergic drug or a dosage increase of an existing serotonergic agent
No recent addition of an antipsychotic, or dosage increase of an existing antipsychotic
Features that are not better accounted for by other causes such as infection, intoxication, metabolic derangement, substance abuse, or substance withdrawal
Low specificity of the Sternbach criteria, coupled with observations that clonus appears to be more specific to serotonin syndrome than other drug toxicity states (e.g., NMS, anticholinergic delirium), prompted refinements that led to the Hunter Serotonin Toxicity Criteria (Dunkley et al. 2003), which emphasize the importance of clonus (inducible, spontaneous, or ocular), agitation, diaphoresis, tremor, and hyperreflexia for establishing diagnostic specificity.
Serotonin syndrome typically is associated with drug interactions that increase central serotonergic tone. Drugs that affect serotonin through varied mechanisms of action are thought to incur a greater risk for serotonin syndrome than are drugs with a single mechanism (as in the case of overdosing on a single SSRI; even the combination of an SSRI plus SNRI involves mechanistic redundancy rather than serotonergic novelty, and hence is unlikely to produce serotonin syndrome). Examples of potentially toxic interactions include the following:
Dextromethorphan (which blocks neuronal serotonin uptake)+ MAOIs
Buspirone+SSRIs or lithium
Linezolid (synthetic antibiotic that weakly inhibits MAO)+SSRIs or other serotonergic agents
Triptans+SSRIs or SNRIs (note that in 2006 the FDA issued an alert regarding this combination, although this relative contraindication remains controversial and unnecessarily overconservative in the opinions of some authorities, such as the American Headache Society [Evans et al. 2010])
MAOIs+amphetamines (which release serotonin)
3,4-Methylenedioxymethamphetamine (i.e., Ecstasy)
Tramadol+SSRIs or SNRIs
Patient-specific risk factors for serotonin syndrome have not been well identified, although some reports suggest that CYP2D6 poor metabolizers may be at greater risk when a second serotonergic drug is added to a first.
Treatment of serotonin syndrome depends on supportive treatment plus cessation of all serotonergic agents. Fundamentally, the condition is self-limited and eventually resolves spontaneously after discontinuation of the offending agent(s). Supportive measures may include intravenous hydration, clonazepam for myoclonus, cooling blankets or ice packs for centrally mediated hyperthermia, and airway protection (with the possible need for mechanical ventilation).
The symbol ■ is used in this chapter to indicate that the FDA has issued a boxed warning for a prescription medication that may cause serious adverse effects.