The story of monoamine oxidase inhibitors (MAOIs) goes back to isoniazid, a hydrazine derivative first synthesized in 1912 without knowing of its biological activity. Later, chemists at E. Hoffmann LaRoche, knowing that semicarbazones exerted antitubercular activity in animals, resynthesized isoniazid as an intermediary in preparing pyridine-4-aldehyde, a necessary precursor for synthesizing semicarbazones. Unexpectedly, biological screening discovered the marked tuberculostatic action of isoniazid in vivo. Subsequently, the drug was developed as a widely used antituberculosis drug.
Early on, reports appeared of other beneficial effects of isoniazid. Although better known for having recognized the psychiatric value of chlorpromazine, Delay1 suggested isoniazid might also have antidepressant properties. Noting that isoniazid’s mood benefits occurred even in patients whose tuberculosis did not improve, Delay argued isoniazid’s effect on mood was not due to its tuberculostatic ability.
In his very useful overview of isoniazid’s early history. Pletscher2 states, “In the early fifties numerous clinicians reported unexpected side effects of hydrazine derivatives, especially of iproniazid. In patients treated for tuberculosis with iproniazid, the drug not only showed tuberculostatic action, but also caused euphoria and psychostimulation, leading in some cases to psychomotor excitation, psychotic states, insomnia, increase of appetite, changes in sexual behaviour etc.” This description leaves it obscure as to whether this syndrome resembles mania or delirium, such as can be produced by high-dose psychostimulants, raising the question of whether isoniazid might more closely resemble an antidepressant medication or a psychostimulant.
Salzer and Lurie,3,4 addressed this question. Their detailed descriptions of isoniazid’s effect more closely resembled an antidepressant rather than a stimulant; isoniazid’s effect had a slower onset than a stimulant and enhanced sleep and appetite, effects opposite to those of stimulant medications. Their follow-up study quite convincing shows that isoniazid produces a substantial, maintained antidepressant effect without reported toxicity. Healy5 expanded on these findings, claiming that a useful psychotropic drug has been forgotten, primarily for lack of profitability.
Elucidation of the mechanism by which isoniazid produced its apparent antidepressant effects was confused by inexact chemical assays resulting in failures to replicate earlier findings. Zeller et al.,6 for example, claimed in 1952 that isoniazid is not an MAOI. However, they assayed the monoamine oxidase (MAO) of liver mitochondria, later shown by Robinson et al.7 to differ from the MAO in platelets and plasma. Assays of plasma MAO demonstrated isoniazid’s specific effect and explained Zeller et al.’s6 erroneous claim. This is interesting when considering the occasional report of a cheese reaction to isoniazid.8
Not all reports on isoniazid’s psychiatric effects were positive. Lemere9 reported a high level of overstimulation and poor therapeutic effects. Wilson10 described a lack of psychiatric utility. There are several reports of isoniazid-induced psychosis.11–14 Unfortunately, this off-patent medication is still labeled only as a generic tuberculostatic, making it unlikely to be evaluated by modern psychiatric methods (ie, randomized, placebo-controlled studies) as a commercial product.
Nevertheless, the presumption of isoniazid’s mechanism of antidepressant action being its inhibition of MAO led to molecular manipulation to produce patentable MAOIs. The first to market was iproniazid. However, appropriate attribution for first recognizing iproniazid’s importance is obscure. Zeller and Barsky15 reported that iproniazid was a potent MAOI, but also stated that “The parent compound, isoniazid, had a weaker effect”. Sandler16 states that there is no doubt that Kline17 and colleagues, Loomer et al.,18 deserve all the credit for identifying the beneficial action of iproniazid in depressive illness. However, Kety19 indicates that Crane20 may have first suggested that the stimulatory side effects of iproniazid in patients with tuberculosis may be therapeutic for depression.
Whichever attribution is most accurate, Loomer et al.18 inferred the biochemical and pharmacological actions of iproniazid shown in animals might produce their clinical effects in humans and conducted an open clinical trial. Patients with stable depression, institutionalized for an average of 20 years, were treated with iproniazid for several weeks. Seventy percent showed remarkable improvement of their depressed state. Similar observations were made in ambulatory depressed patients. Later, numerous investigations confirmed these findings. Although developed as a tuberculostatic drug, it was marketed in 1958 as an antidepressant under the trade name Marsilid (Roche, Basel, Switzerland). This new approach to treating depression proved very successful, and sales figures of Marsilid grew during the next few years. Soon, successors of Marsilid—MAOIs of both the hydrazine and nonhydrazine type—were developed and introduced by several drug companies. These drugs included phenelzine (Nardil; Pfizer, New York, NY), tranylcypromine (Parnate; Smith Kline, London, UK), and isocarboxazid (Marplan; Roche, Basel, Switzerland), and these early MAOIs are still available in the United States. It is interesting that tranylcypromine, the only remaining nonhydrazine MAOI, was created as an amphetamine analog in 1948.
The initial enthusiasm about the success of iproniazid and other MAOIs was curtailed by the occurrence of undesirable side effects, as sporadic reports of liver failure and paroxysms of blood pressure elevation caused concern about cerebral hemorrhage. As a result of these dangers, at least 10 MAOIs were withdrawn from most markets, particularly the hydrazines, including iproniazid. The article by Asatoor et al.21 on the “cheese effect” demonstrated that tyramine in the peripheral circulation releases norepinephrine, thereby increasing blood pressure. They inferred that in patients taking MAOIs, sufficient unmetabolized tyramine could enter the blood stream, inducing the hypertensive crises that had been clinically observed.22 In 1967, Blackwell23 clarified the cause of these MAOI-induced hypertensive crises as due to the ingestion of tyramine-rich foods. Because MAOIs inactivate intestinal destruction of tyramine, adequate dosing of MAOIs allows ingested tyramine to be absorbed, which may lead to a massive release of norepinephrine. The resulting temporary blood pressure rise became known as the “cheese reaction” because aged cheese is often rich in tyramine.
Enthusiasm for MAOIs was further deflated by a 1965 study that randomly assigned 250 depressed patients to receive electroconvulsive therapy (ECT), imipramine, phenelzine, or placebo.24 ECT and imipramine were reported superior to placebo whereas phenelzine did not differ from placebo. This seemed to indicate that MAOIs were not only dangerous, but also ineffective. However, multiple problems with the study design, implementation, and data analyses raise questions about the validity of its authors’ conclusions. First, the dose of phenelzine was 45–60 mg/day, well below the marketed maximal dose of 90 mg/day. Second, evaluations were made by 55 physicians untrained in the use of the ratings and who also were not made blind to the treatment their patients were receiving. Another problem is the study authors did not report on an intent-to-treat sample (as is the current standard), instead limiting their report on efficacy after 4 weeks on subjects for whom they had 6 months of data.24 In our view, this trial should have received a more skeptical review rather than being the demonstration of MAOIs’ ineffectiveness it was taken to be at the time.
In 1968, Robinson et al.7 determined that MAO exists in two functional forms, and labeled these isoenzymes MAO-A and MAO-B. Inhibition of MAO-A supported antidepressant action. All marketed MAOIs inhibit both enzymes, although at low doses (5–10 mg/day) oral selegiline selectively inhibits MAO-B.
During the past 3 decades, MAOIs have remained at the periphery of psychiatry despite continued use by occasional avid users. Contrary to the negative findings of the 1965 British Medical Research Counsel study by Pickering et al.,24 multiple subsequent studies have demonstrated placebo-controlled and/or presumptive efficacy of MAOIs in a variety of patient populations (see the article by Arden et al. in this issue, Part 2).
In addition to these efficacy studies, others sought ways to make MAOIs safer. In 1981, Kline et al.25 demonstrated that concomitant desipramine prevented tyramine’s hypertensive effect in rats treated with tranylcypromine. Pare et al.26 translated Kline et al.’s25 research to patients, showing that amitriptyline augmentation of MAOI greatly diminished the pressor response to intravenous tyramine in patients receiving MAOIs alone. Indeed, the sensitivity to tyramine in patients treated with both MAOI and amitriptyline was reduced to that of healthy controls and untreated depressed patients.26 Although these studies seem convincing, tricyclic antidepressants remain contraindicated with MAOIs in most guidelines. More experimentation would be clarifying. However, neither the pharmaceutical industry nor other funding agencies are interested in doing so.
Another approach was use of MAO-B inhibitors. Because the intestinal MAO that breaks down tyramine is MAO-A, an MAO-A–sparing drug should not cause the “cheese reaction.” Selegiline (also known as l-deprenyl) is a candidate because at low doses it inhibits MAO-B with relatively little inhibition of MAO-A. Unfortunately, oral selegiline appears to be an ineffective antidepressant at MAO-B selective doses (5–10 mg/day) but is effective when the dose is raised to nonselective doses (ie, >20 mg/day).27–29 Furthermore, the “cheese reaction” remains a risk at antidepressant doses of oral selegiline, as at least one food-related hypertensive crisis has been reported.30
Bypassing the intestines via the selegiline transdermal system (STS) spares intestinal MAO-A at the lowest patch strength of STS (6 mg/24 hours); thus, at the lowest STS dose (6 mg/24 hours), tyramine has a negligible effect on blood pressure, so it does not require the patient to adhere to the strict dietary restrictions usually required when taking an MAOI. Unfortunately, higher patch strengths do; thus, the US Food and Drug Administration (FDA)-approved labeling suggests institution of the tyramine-free diet when using the 9-mg/24 hours and 12-mg/24 hours patch strengths.
Reversible inhibitors of monoamine oxidase-A (RIMAs) provide a safety advantage over irreversible MAOI because tyramine can displace RIMAs and still be metabolized. Moclobemide and brofaromine have been shown to be effective in the treatment of depression and appear to have similar efficacy to selective serotonin reuptake inhibitors.31 RIMAs are not commercially available in the United States, although several RIMAS have been tested. However, no company has requested FDA approval for use of a RIMA as an antidepressant. This is unfortunate given the therapeutic benefit they may have for the significant minority of depressive patients for whom MAOIs are the most effective treatment.
In summary, numerous psychotropic agents were identified in the 1950s, mostly serendipitously, including the MAOIs. Initial treatment enthusiasm flagged with the recognition of the risks for hepatotoxicity and hypertensive crisis. Risk of significant liver damage was minimized by the removal of the main culprits from the market, so that today hepatotoxicity induced by MAOIs appears to be rare. Risk for hypertensive crisis remains but can be minimized to an acceptable degree by vigilant use of a tyramine-free diet and avoidance of dangerous concomitant medications, such as decongestants. Despite improved risks, MAOIs are still viewed as dangerous and many psychiatrists do not prescribe them. In our view, MAOIs should not be considered first-line agents, but no depressed or anxious patient should be considered treatment refractory until given an adequate trial of an MAOI.
- Delay J, Laine B, Buisson JF: Note on the action of INH utilized in the treatment of depressive states. Ann Med Psychol. 1952;2:689–692.
- Pletscher A: The Discovery of Antidepressants: A Winding Path. Basel, Switzerland: Swiss Academy of Medical Sciences; 1991.
- Salzer HM, Lurie ML: Anxiety and depressive states treated with isonicotinyl hydrazide (isoniazid). AMA Arch Neurol Psychiatry. 1953;70(3):317–324. doi:10.1001/archneurpsyc.1953.02320330042005 [CrossRef]
- Salzer HM, Lurie ML: Depressive states treated with isonicotinyl hydrazide (isoniazid); a follow-up study. Ohio Med. 1955;51(5):437–441.
- Healy D. Discussion of Lurie M Interview. The Psychopharmacologists II. London, UK: Chapman and Hall; 1998:119–134.
- Zeller EA, Barsky J, Fouts JR, et al. Influence of isonicotinic acid hydrazide (INH) and 1-isonicotinic-2-isopropyl-hydrazide (IIH) on bacterial and mammalian enzymes. Experientia. 1952;8:349–350. doi:10.1007/BF02174413 [CrossRef]
- Robinson DS, Lovenberg W, Keiser H, Sjoerdsma A: Effects of drugs on human blood platelet and plasma amine oxidase activity in vitro and in vivo. Biochem Pharmacol. 1968;17(1):109–119. doi:10.1016/0006-2952(68)90163-9 [CrossRef]
- Smith CK, Durack DT: Isoniazid and reaction to cheese. Ann Intern Med. 1978,88(4):520–521. doi:10.7326/0003-4819-88-4-520 [CrossRef]
- Lemere F: Isoniazid treatment of psychiatric patients. AMA Arch Neurol Psychiatry. 1954;71(5):624–625. doi:10.1001/archneurpsyc.1954.02320410086008 [CrossRef]
- Wilson WP: Isonicotinic acid hydrazide in non-tuberculous mental patients. Dis Nerv Sys. 1953;14(9):278–279.
- Jackson SL: Psychosis due to isoniazid. Br Med J. 1957;2:743–746. doi:10.1136/bmj.2.5047.743 [CrossRef]
- Duggal HS, Nizamine SH: Novel antipsychotic drugs and INH-related psychosis. Aust N Z J Psychiatry. 2000;34(2):343–344. doi:10.1080/j.1440-1614.2000.0719i.x [CrossRef]
- Alao AO, Yolles JC: Isoniazid-induced psychosis. Ann Pharmacother. 1998;32(9):889–891. doi:10.1345/aph.17377 [CrossRef]
- Iannaccone R, Sue YJ, Avener JR: Suicidal psychosis secondary to isoniazid. Pediatr Emerg Care. 2002;18(1):25–27. doi:10.1097/00006565-200202000-00008 [CrossRef]
- Zeller EA, Barsky J: In vivo inhibition of liver and brain monoamine oxidase by 1-isonicotinyl-2-isopropyl hydrazine. Proc Soc Exp Biol Med. 1952;81(2):459–461. doi:10.3181/00379727-81-19910 [CrossRef]
- Sandler M: Monoamine oxidase inhibitors in depression: history and mythology. J Psychopharmacol. 1990;4:136–139. doi:10.1177/026988119000400307 [CrossRef]
- Kline N: Clinical experience with iproniazid (Marsilid), Symposium Marsilid New York 1957. J Clin Exp Psychopathol. 1958;19(Suppl):72–78.
- Loomer HP, Saunders JC, Kline NS: A clinical and pharmacodynamic evaluation of iproniazid as a psychic energizer. Psychiatr Res Rep Am Psychiatr Assoc. 1957;8:129–141.
- Kety S: Introduction, Symposium on Monoamine Oxidase and Its Inhibition. London, UK: CIBA Foundation; 1976.
- Crane G: The psychiatric side-effects of iproniazid. Am J Psychiatry. 1956;112:494–501. doi:10.1176/ajp.112.7.494 [CrossRef]
- Asatoor AM, Levi AJ, Milne MD: Tranylcypromine and cheese. Lancet. 1963;282:733–734. doi:10.1016/S0140-6736(63)90368-4 [CrossRef]
- Grady MM, Stahl SM: Practical guide for perscribing MAOIs: debunking myths and removing barriers. CNS Spectr. 2012;17:2–10. doi:10.1017/S109285291200003X [CrossRef]
- Blackwell B: Hypertensive interactions between monoamine oxidase inhibitors and foodstuffs. Br J Psychiatry. 1967;113:349–365. doi:10.1192/bjp.113.497.349 [CrossRef]
- Pickering G, Bowlby J, Cochrane L, et al. Clinical trial of treatment of depressive illness: report to the Medical Research Council by its Clinical Psychiatry Committee. Br Med J. 1965;1:881–886. doi:10.1136/bmj.1.5439.881 [CrossRef]
- Kline NS, Cooper TB, Suckol RF, Hallstrom C, Bebbria DR: Protection of patients on monoamine oxidase inhibitor against hypertensive crisis. J Clin Psychopharmacol. 1981;1:410–411.
- Pare CM, Kline N, Hallstrom C, Cooper TB: Will amitriptyline prevent the “cheese” reaction of monoamine-oxidase inhibitors?Lancet. 1982;2(8291):183–186. doi:10.1016/S0140-6736(82)91030-3 [CrossRef]
- Mann JJ, Aaron SF, Wilner PJ, et al. A controlled study of the antidepressant efficacy and side effects of l-deprenyl. Arch Gen Psychiatry. 1989;46:45–50. doi:10.1001/archpsyc.1989.01810010047007 [CrossRef]
- McGrath PJ, Stewart JW, Harrison W, Wager S, Nunes EN, Quitkin FM: A placebo controlled study of l-deprenyl in atypical depression. Pharmacol Bull. 1989;25:63–67.
- Sunderland T, Cohen RM, Molchan S, et al. High-dose selegiline in treatment-resistant older depressive patients. Arch Gen Psychiatry. 1994;51:607–614. doi:10.1001/archpsyc.1994.03950080019003 [CrossRef]
- McGrath PJ, Stewart JW, Quitkin FM: A possible l-deprenyl induced hypertensive reaction. J Clin Psychopharmacol. 1989;9(4):310–311.
- Lotufo-Neto F, Trivedi M, Thase ME: Meta-analysis of the reversible inhibitors of monoamine oxidase type A moclobemide and brofaromine for the treatment of depression. Neuropsychopharmacology. 1999;20(3):226–247. doi:10.1016/S0893-133X(98)00075-X [CrossRef]