Below is an excerpt of the full MAOI toolkit, available for free download.
References are located at the bottom of the full toolkit PDF.
MAOIs are generally well-tolerated, but there is a potential for adverse effects which may become severe enough to warrant intervention (e.g., dose change or use of an additional medication), a switch to another MAOI, or complete discontinuation. Rabkin and colleagues evaluated the occurrence of significant side effects in patients receiving tranylcypromine or phenelzine compared to imipramine and placebo (108,109). Most adverse effects occurred with phenelzine, although they didn’t cause more frequent discontinuations than with tranylcypromine. Table 8 shows the results of this investigation. Moclobemide was associated with nausea, dizziness, and insomnia in placebo-controlled trials. With long-term treatment, moclobemide was associated with insomnia, nausea, dizziness, dry mouth, epigastric discomfort, tachycardia, and palpitations in 2 – 5% of patients (30).
Significant but usually asymptomatic and transient blood pressure increases (20 mm Hg or more in either systolic or diastolic pressures) may occur in approximately 50% of patients immediately after they take either phenelzine or tranylcypromine. Blood pressures return to normal after 1 – 2 hours (110). A small number of cases of MAOI-related spontaneous, acute, severe hypertensive reactions that were apparently not related to ingestion of other drugs or TYR-containing foods have been reported. In some cases, the reactions were severe enough to warrant discontinuation of the MAOI (67). At least one case of severe hypertension with headache and bradycardia occurred after a patient’s first dose of 10 mg of tranylcypromine(111). If patients are monitoring their blood pressure, they should be educated about this effect and reassured that it is usually not harmful, but to report to their prescriber if severe or symptomatic hypertensive episodes occur. This hypertensive response has not been associated with isocarboxazid, transdermal selegiline, or moclobemide. FDA-approved labeling for isocarboxazid states that it is contraindicated in patients with hypertension (as well as known/suspected cerebrovascular defect, cardiovascular disease, or a history of headache) (17). It is not clear why these contraindications are listed, although it may be based on concern regarding possible hypertensive reactions to dietary tyramine. Labeling for phenelzine and tranylcypromine contains less strongly-worded warnings about use of these agents in patients with hypertension; labeling for transdermal selegiline does not contraindicate its use in hypertensive patients. There are no warnings regarding the use of moclobemide in hypertensive patients in that product’s labeling (16,18-22). Based on this inconsistent labeling and the fact that MAOIs were used in the past as antihypertensive agents, avoidance of MAOIs in patients with hypertension may not be necessary. In patients with reasonably well-controlled blood pressure, a MAOI could be used with careful blood pressure monitoring; in some patients, downward adjustment of antihypertensive medication dosage may be necessary. At least in the case of isocarboxazid, documentation of informed consent would be prudent.
(N = 141)
(N = 41)
(N = 65)
(N = 43)
|Hypertensive crisis*||11 (8)||1 (2)||0 (0)||0 (0)|
|Passing Out/Falling (Orthostasis)||16 (11)||7 (17)||6 (9)||0 (0)|
|Hypomania||14 (10)||3 (7)||0 (0)||0 (0)|
|Weight gain (≥ 15 pounds)**||11 (8)||0 (0)||0 (0)||0 (0)|
|Convulsion||0 (0)||0 (0)||1 (2)||0 (0)|
|Rash**||1 (1)||1 (2)||1 (2)||1 (2)|
|Edema**||6 (4)||0 (0)||0 (0)||0 (0)|
|Drowsiness**||4 (3)||0 (0)||0 (0)||0 (0)|
|Anorgasmia/Impotence||31 (22)||1 (2)||3 (5)||0 (0)|
|Paresthesias||6 (4)||2 (5)||0 (0)||0 (0)|
|Urinary Retention||7 (5)||1 (2)||2 (3)||0 (0)|
|Other**||21 (15)||4 (10)||4 (6)||1 (2)|
|None||56 (40)||23 (56)||46 (71)||41 (96)|
*Hypertensive crises resulted from non-adherence to tyramine-restricted diet.
**Drug discontinued by some patients
Adapted from Rabkin et al (108)
As shown in Table 8, orthostatic hypotension severe enough to result in syncope is common with MAOIs and may be seen in 10 – 20 % of treated patients (108). Less severe orthostasis is even more common, and, to some degree, is probably present in all patients taking MAOIs (at one time, an MAOI, pargyline, was used as an antihypertensive agent). It has been suggested that the presence of orthostatic hypotension, usually asymptomatic (i.e., a 10 – 15 mm Hg decrease), is a marker that an effective dose of the MAOI has been achieved (47). Most patients will develop tolerance to orthostasis after a few weeks of treatment, however, temporary dose reduction or other interventions may be needed. The mechanism for induction of orthostasis is not established. It has been theorized that it results from accumulation of trace amines such as octopamine in adrenergic neurons. It appears more likely that orthostasis is caused by increased NE-related central alpha-adrenergic activity (similar to the antihypertensive action of clonidine) (112). Moclobemide and transdermal selegiline seem to be much less likely to cause orthostasis (19,30). When MAOI treatment is initiated, patients should be educated about the possibility of orthostatic hypotension and told to be aware of dizziness. It has been suggested that all patients taking an MAOI should self-monitor their blood pressure (one sitting and two standing measurements) from the beginning of treatment in order to identify orthostasis as a marker of therapeutic dose and to identify potentially serious orthostasis (47). Routine blood pressure monitoring may be especially important for elderly patients or others for whom falls may be dangerous. It is not clear that this is a common practice by clinicians. When orthostasis is severe enough to warrant intervention, suggested approaches to management are shown in Table 9 (108,109,113).
|Can be considered as a stepped approach|
|Identify and, if possible, correct contributing factors other than the MAOI:|
|Alter dosing of MAOI|
|Consider support hosiery|
|Consider ancillary treatment to expand intravascular volume|
|Salt &/or fludrocortisone can usually be eliminated after 4 – 6 weeks|
|Consider an alternative MAOI|
Phenelzine has been associated with depletion of pyridoxine (B6) (14,114). Reductions in B6 are thought to result from phenelzine, as a hydrazine compound, binding with pyridoxine to create a compound that inhibits pyridoxal kinase, which in turn prevents B6 from being activated to pyridoxal-5`-phosphate (PLP) and inhibits several reactions for which PLP is a coenzyme. This adverse effect is probably similar to B6 deficiency that results in neuropathy in patients treated with isoniazid for tuberculosis. In addition to peripheral sensory neuropathy, untreated B6 deficiency may result in other neurologic symptoms including ataxia, hyperacusis, hyperirritability, muscle tension, muscle twitching, and, rarely, seizures. Edema may also result from B6 depletion. While isocarboxazid (also a hydrazine MAOI) has not been reported to cause peripheral neuropathy, it does cause muscle twitching and has been reported to cause edema responsive to B6 treatment (115,116). Patients in reported cases of peripheral neuropathy had lowered B6 plasma levels, although not always abnormally low. Patients’ symptoms responded within a few weeks after administration of 150 – 300 mg/day of B6 (114,116). These doses are significantly higher than those recommended for treatment of isoniazid-induced neuropathy (100 mg/day) (117); it is not known if such doses would be necessary. Patients treated with phenelzine and isocarboxazid should be educated about the possibility of this adverse effect. It has also been recommended that pre-treatment B6 plasma levels be measured and then repeated at 1 month and 3 months (14). It is not known if there is benefit to routine addition of prophylactic B6 to phenelzine therapy (114). Prophylactic B6 (25 – 50 mg/day) is recommended with isoniazid treatment in patients at increased risk for peripheral neuropathy (elderly, persons with diabetes, poorly nourished, alcohol use disorder, persons infected with human immunodeficiency virus, persons with chronic renal failure) (117). If such persons were treated with phenelzine or isocarboxazid, prophylactic B6 might be considered.
Impotence or anorgasmia, most commonly in males, were the most common significant adverse effects seen with phenelzine; these were far less commonly seen with tranylcypromine (Table 8) (108,109). Moclobemide and transdermal selegiline are associated with a very low rates of sexual dysfunction (30,118). In assessing patient-reported sexual dysfunction, it is important to know their baseline, as some types of sexual dysfunction may be related to the underlying depression. In some affected patients, sexual dysfunction will improve over time; dose reduction may also help. Switching to a MAOI with a lower risk of sexual dysfunction may also be a strategy. Cyproheptadine 1 – 4 mg at bedtime has been reported to be helpful for anorgasmia (119). Bethanechol at doses up to 50 mg/day has also been reported to be helpful for impotence in males and anorgasmia in females (109).
Insomnia occurs relatively commonly with MAOIs. Tranylcypromine is structurally similar to amphetamine and can exert stimulant effects, especially at higher doses. Selegiline is metabolized to l-amphetamine and l-methamphetamine, and these metabolites are believed responsible for some CNS stimulation (7). Moclobemide causes insomnia in a small number of patients (30). For patients troubled by insomnia, adjusting the timing of doses to no later than mid-afternoon may be helpful. Trazodone 50 – 200 mg/day has been safely and effectively used to manage MAOI-associated insomnia (96,120). However, it would seem prudent to first use other sedative-hypnotics (e.g., benzodiazepine receptor agonists, melatonin, hydroxyzine) which are not known to adversely interact with MAOIs in order to minimize risk of possible interactions.
Occasionally, a small number of patients taking phenelzine or tranylcypromine experience marked daytime drowsiness. Patients should be educated about the possibility of daytime sedation and cautioned about driving until they learn what to expect from their medication. Phenelzine is more likely to cause general drowsiness, and this can be managed by focusing more of the daily dose in the evening.
Iproniazid, a hydrazine compound, was one of the original MAOIs approved for the treatment of depression. Soon after it came into use, cases of hepatic necrosis related to its use were reported. Approximately 1% of patients developed hepatic injury; fatality resulted in as many as 20% of cases of hepatic injury. Iproniazid was withdrawn from use in most of the world (121). The other two hydrazine MAOIs still in use, phenelzine and isocarboxazid, are far less likely to cause hepatic injury; no cases appear to have been reported with isocarboxazid, and less than 5 cases linked to phenelzine appear in the literature (122,123). As of 1987, the manufacturer of phenelzine had had 14 cases of hepatotoxicity reported in the preceding 8 years; all were associated with risk/confounding factors such as alcohol use or viral infection (124). While hepatotoxicity is usually associated with hydrazine MAOIs, there has apparently been one case of hepatic damage linked to tranylcypromine (121). The National Center for Biotechnology Information resource, LiverTox, gives phenelzine a hepatotoxicity likelihood score of C (probable rare cause of clinically apparent liver injury); isocarboxazid is given a likelihood score of E* (unproven, but suspected rare cause of clinically apparent liver injury); tranylcypromine is given a likelihood score of D (possible rare cause of clinically apparent liver injury) (125-127). Transdermal selegiline and moclobemide do not appear to be associated with hepatotoxicity.
The mechanism of phenelzine-related hepatotoxicity is not known. It has been proposed to be a similar mechanism to isoniazid-induced hepatic damage, and there has been a controversial association with slow acetylator status (122). Reported cases have not appeared to result from hypersensitivity reactions. The liver damage takes the form of hepatitis with some possible cholestatic features. At least one case resulted in cirrhosis (122), and 2 cases resulted in fulminant hepatic failure requiring liver transplantation (123). There is no information to suggest a relationship to dosage.
Phenelzine-related hepatic damage occurs early in treatment (6 months or less). There is gradual onset with symptoms such as anorexia, weakness, malaise, and jaundice. Dramatic elevations of ALT and AST (up to 100 times the upper limit of normal or more) are seen; alkaline phosphatase is less dramatically elevated. Jaundice may be severe and reflect the severity of the hepatic damage. While not described in specific literature case reports, there have apparently been fatalities related to phenelzine-induced liver toxicity (121).
While hepatotoxicity with phenelzine appears to be quite rare, baseline pretreatment liver function tests should probably be obtained, and patients should be monitored for potential symptoms during early therapy. If symptoms should appear, liver function tests should be repeated. Transient, modest increases in AST and ALT are apparently seen during phenelzine treatment and are not associated with significant liver damage (124). Therefore, routine, repetitive testing of liver function in the absence of symptoms is probably not warranted. If evidence of significant liver damage appears, phenelzine should be discontinued immediately.
Application-site reactions occurred in 24% of patients receiving transdermal selegiline; these reactions led to discontinuation in 2% of treated patients. This was the most common adverse effect leading to discontinuation (19). Education of patients about proper application of selegiline patches and rotation of application sites may help minimize this side effect. The patch should be applied to intact, dry skin of the upper torso (above the waist and below the neck), the outer surface of the upper arm, or the upper thigh.
Moclobemide and transdermal selegiline appear to be free of unpleasant symptoms occurring after rapid discontinuation (19,30); however, unless clinical circumstances dictate rapid withdrawal, these two MAOIs should probably be tapered. Withdrawal effects are described in the FDA labeling for isocarboxazid and phenelzine (16,17). Isocarboxazid labeling describes reports of dependency associated with use of excessive doses; some cases occurred in patients with a history of substance use disorders. The labeling describes isocarboxazid withdrawal as possibly causing restlessness, anxiety, depression, confusion, hallucinations, weakness, headache, and/or diarrhea (17). Reports of isocarboxazid dependency/abuse do not appear in the literature. It is not clear whether withdrawal symptoms would be likely with discontinuation of therapeutic isocarboxazid doses. Phenelzine’s labeling describes an uncommon, infrequently reported withdrawal syndrome occurring 24 – 72 hours following rapid discontinuation. Symptoms may range from vivid nightmares with agitation to frank psychosis; convulsions may occur (16). Patients treated with phenelzine for narcolepsy who had the drug rapidly discontinued experienced total insomnia with continuous, frightening visual hallucinations when in the dark; they also experienced depression, severe anxiety, and suicidal thoughts. These symptoms disappeared with re-initiation of phenelzine. Tapering phenelzine over 1 – 3 weeks prevented these symptoms (31). The literature contains a report of 2 patients who developed shivering and feelings of intense cold within 24 hours of phenelzine discontinuation. In one patient these symptoms were accompanied by a severe frontal headache lasting for days; in the other, they were accompanied by depression and irritability (128). There are also a small number of published case reports of acute psychosis almost immediately following rapid withdrawal of phenelzine 60 – 90 mg/day. Psychotic symptoms included hostility, hallucinations (auditory, visual, gustatory, and tactile), paranoia and aggression. Confusion and disorientation were also present. One patient became catatonic. Notable physical symptoms included miosis, hyperreflexia, myoclonus, athetosis, and ataxia. The withdrawal-related psychotic states lasted for several days and, in at least one patient, full recovery required 6 weeks. Reported patients were hospitalized and treated symptomatically with sedatives and brief antipsychotic drug therapy (129).
Both FDA labeling and several literature reports describe significant withdrawal symptoms following rapid discontinuation or dosage decreases of tranylcypromine. The large majority of reported cases, but not all, were associated with abuse of tranylcypromine at large doses (as high as 750 mg/day) (18,130-133). Symptoms develop within 1 – 3 days of decreasing dosage or discontinuation and consist of return of depression (often more severe than that which led to treatment with tranylcypromine), anxiety, impaired cognition, severe sleep disturbances (insomnia, hypersomnia, and intense nightmares), hyperphagia, tremor, and myoclonic movements. Withdrawal from high doses of tranylcypromine is also associated with frank delirium, psychosis, thrombocytopenia, and occasional mild, reversible liver damage (133,134). Most cases involving these more severe symptoms have resulted from overuse/abuse of tranylcypromine at high doses and the patient subsequently ran out of the drug. Many experts have pointed out that the symptoms associated with tranylcypromine withdrawal closely resemble those of amphetamine withdrawal (129,135). As discussed above, tranylcypromine does appear to have amphetamine-like activity. Abuse and addiction to tranylcypromine have been well-described (131,136). Patients with a history of substance use disorder and, possibly, personality disorder appear to be at higher risk for tranylcypromine abuse, and such patients may not be good candidates for treatment with this agent.
Weight gain associated with antidepressant treatment may be difficult to assess because of the improvement in appetite associated with successful treatment of depression. For this reason, long-term observation (more than six to eight weeks) may be required to adequately differentiate drug-induced weight gain from that associated with treatment response (137). Most clinical trials are not long enough to conclusively evaluate this relationship. Based on literature reports and manufacturer labeling, transdermal selegiline, moclobemide, and tranylcypromine appear to be weight-neutral (6,18-22,30,137). In the case of tranylcypromine, its weight-neutrality has been hypothesized to be related to its amphetamine-like structure as well as to its ability to promote NE release and possibly block reuptake of 5HT, NE, and DA (137,138).
Phenelzine has been associated with cases of significant weight gain during long-term treatment (longer than 16 weeks); some patients gained over 40 pounds. Phenelzine influences gluconeogenesis and potentiates hypoglycemia; this is related to its hydrazine structure. It also is a potent inhibitor of adipocyte lipid storage and differentiation (14,137). There is less information about isocarboxazid, but, because it is also a hydrazine MAOI, it may have effects similar to phenelzine. Isocarboxazid appears to be intermediate between phenelzine and tranylcypromine in its tendency to cause weight gain (137). Patients who are to be treated with phenelzine and, possibly, isocarboxazid should be educated regarding the potential for weight gain and helped with dietary modification should weight appear to be increasing.