Therapeutic doses of Moclobemide combined with any (S)SRI carries a significant risk of potentially fatal serotonin toxicity. This is especially so when multiple pharmaco-kinetic interactions via CYP-450 come into play.


Moclobemide definitely confers a risk of severe and fatal serotonin toxicity if combined with (S)SRIs although the history and evidence relating to its interactions is clouded with misleading information, some of it put out by the drug company. This is now largely past history because few people use this drug any more: there seems to be widespread appreciation of the view I promoted decades ago which is that moclobemide is ineffective as an anti-depressant drug. Indeed, I have stated on more than one occasion that those of my colleagues who thought it was an effective antidepressant neither understood depression nor how to treat it. Ridiculous as that might seem more psychiatrists fit into that category than one might imagine. History has proved me to be correct because there is no other reason not to use moclobemide other than it’s ineffectiveness, because it has no significant side effects.

I will re-post my original detailed analysis about moclobemide & ST separately at some later date, the original section in the serotonin toxicity document posted years ago needs tidying up and re-referencing and because it is of ‘historical’ interest to some researchers.

Suffice it to say that even at therapeutic doses, and most definitely following overdoses, one can get quite severe serotonin toxicity reactions and deaths have occurred. I summed it up more than ten years ago in this brief comment (1), see below.

This update and addition to this commentary (March 2015) has been prompted by a discussion that was brought to my attention recently from an Internet site. It may be educative to pick a few themes from this thread to illustrate the perspicacity of the old adage ‘a little knowledge is a dangerous thing’.

An opening comment is ‘This forum is interesting with the amount of information & knowledgeable people here.’ However, remember what Shaw said (2) ‘Beware of false knowledge; it is more dangerous than ignorance.’ The problem for many readers of such forums is the difficulty of distinguishing between real and false knowledge.

From the thread: 'My personal experience with this was very positive with 20mg Fluoxetine and 150mg Moclobemide. Fluoxetine increases blood levels of Moclobemide so low doses can be used, even though the studies use dosages up to 600mg.'

I will not beat about the bush. The doctor supervising this treatment (if there is one) is putting himself in significant danger of a) being successfully sued by the family, should the patient be unfortunate enough to die, or worse, suffer severe permanent morbidity, b) as well as being in danger of being struck off the medical register.

I am prepared to wager a considerable sum that most psychiatrists would not be able to give an accurate précis of the data I am about to summarise here. This combination of Fluoxetine/Moclobemide is a perfect storm looming on the horizon.

First, there is no doubt that a therapeutic level of any SRI combined with a high therapeutic level, or small overdose, of moclobemide is capable of inducing fatal serotonin toxicity.

Second, moclobemide levels are increased by inhibition of CYP450 2C19 and Moclobemide itself inhibits P450 2D6 (3-8), this can elevate levels to an extent comparable to the levels expected with a small overdose (9).

It is therefore inevitable that in some individuals small changes of blood levels, or dose, of a drug that inhibits P450 could elevate blood levels enough to precipitate ST in someone who previously appeared well and without problems on their previous dosage. These references to cases of ST (10-14) all involved ‘therapeutic’ doses, all are pre 1998. I have not even bothered to retrieve further examples after 1998! Published examples are only the tip of the iceberg, this and personal experience (journal refereeing, medico-legal cases and discussions with colleagues who are toxicologists) indicate that quite a lot of people have been killed by such combinations.

This highlights the importance of understanding that the milligram dose ingested does not always equate consistently and predictably to the concentration of the drug at its site of action. There are many intervening variables which mean that although the same dose is being ingested the concentration of the drug at its site of action can vary by as much as tenfold or even more. There are many doctors who fail to appreciate this and/or forget it, never mind non-medical people searching the Internet. Such people are like innocent children walking through a minefield.

The scenario is an example of what I said above which is ‘a little knowledge is a dangerous thing’ and unfortunately this doctor and his patient are in the category of those who have insufficient knowledge.

Fluoxetine is the worst possible choice for an SSRI to combine with moclobemide (fluvoxamine comes a close second) because it is a potent inhibitor of several cytochrome P-450 enzymes including 2C19, which is responsible for metabolizing moclobemide. In such situations potently inhibiting these enzymes can increase blood levels of drug to the equivalent of those in persons who have taken a small overdose (and thus produce a risk of severe ST). The situation may be further complicated by other drugs in the patient's regime which affect blood levels of either fluoxetine or moclobemide, because in this particular instance not only will fluoxetine inhibit moclobemide metabolism but also moclobemide will inhibit the metabolism of fluoxetine making it a two-way street. So any other drug which raises blood levels of either fluoxetine or moclobemide (of which there are plenty) can potentially start the snowball rolling.

The chess-like complexity of this situation is yet further magnified by the fact that fluoxetine inhibits its own metabolism and has an unusually long elimination half-life of up to 10 days, thus blood levels can unexpectedly elevate via various mechanisms. All of which justifies my opinion that fluoxetine should have been taken of the market years ago and that doctors who still use it are pharmacological ignorami of an inexcusable sort (see my editorial ‘Drug interactions and fluoxetine: a commentary from a clinician’s perspective (15)). A double dose taken by mistake and various other circumstances might easily lead to a fluoxetine or moclobemide level that unexpectedly becomes considerably higher than usual.

So the ‘personal experience’ mentioned above of one individual (probably over a short period of time) is unhelpful and seriously misleading for the general ‘unscientific’ reader. One hardly needs to point that we are not going to hear the opinions of people who have taken such combinations and died.

It is also stated:

‘You will find multiple papers endorsing this combination, the mainstream view is its dangerous, which is misinformed.’

Leaving aside the hubris of this contributor and the ultracrepidarian tenor of the statement, I trust readers with a more sophisticated understanding of scientific research and literature will appreciate that this is a dangerous and unhelpful statement. It is incorrect. There are only a very small number of published reports of poorly controlled non-blinded ‘studies’ (16-19) involving only small numbers of patients which claimed, naïvely and overoptimistically, that this treatment was helpful and safe. In Hawley’s larger study many of the side effects encountered were rated as ‘severe’ (20) which influenced them to cease further investigation. See below for detailed references which I have discussed in detail elsewhere.

Another commentator hit the nail on the head when he pointed out that these papers were published 20 years ago and have never subsequently been followed up. A cautious enquirer would want to know why these researchers failed to capitalise on this ‘successful’ breakthrough that would have enhanced their reputations and careers. In most instances I can inform readers from first-hand knowledge that this was because the doctors concerned encountered serious problems which they did not report but which discouraged their efforts and because ethical committees (rightly) vetoed further efforts.

Since those reports I have corresponded with all of the relevant researchers one or two of whom failed to respond to, or declined to answer, enquiries about outcome and difficulties, others revealed they had encountered difficulties. Indeed, only a few months ago I again corresponded with Dr Prof Udo Bonnet in Germany whose paper concerning this was widely cited (21) and commented on at the time by me (1). My comment concluded:

"The above data indicates clearly that moclobemide/SRI combinations represent a predictably risky strategy and constitute a balancing act, between efficacy and fatality, of such delicacy as to be unattainable in clinical practice. Moclobemide, if combined with any SRI, produces a significant risk of severe ST and also the possibility of fatalities even with ‘therapeutic’ doses. There is no substantive evidence of significant therapeutic benefit. The current medico-legal climate in most western countries means that an informed Doctor, or ethics committee, would be courageous to sanction any such trials except in very special circumstances."

In Nov 2014 I asked Prof Bonnet if he had any further views or experience in this field. He did reply and reported that he has not used that treatment since writing that article in 2004. He states he only ever treated half a dozen patients in this way, encountered problems, and confirms that the treatment is officially contraindicated in the German guidelines.

I know of no research centre, reputable or otherwise, that has done work with this combination in the last 15 years.

Next,this thread also states:

He [Gillman] states the same conclusion, this should only be done in low doses. If you go beyond 20mg Fluoxetine for example, you increase the chance of SS quite drastically.

Wrong. This discussion may possibly have life or death consequences for people making decisions as a result of what they have read. This correspondent seems to read the source material carelessly and seriously misrepresents my position. I have never made the statement attributed to me above.

It is not possible to state that some such combination might not be useful, it might be, but it is clear that most of the practitioners who have dabbled in this area do not know enough about pharmaco-kinetics, pharmaco-dynamics or serotonin toxicity to experiment safely. I suspect a lot of ethics committees have agreed with that.

Last, I note one particular curious comment about avoiding this combination in patients who are suicidal. Patients who are potentially going to become suicidal would be the only candidates for this sort of treatment because serious endogenous (drug-responsive) depression, for which such combinations would be the only conceivable indication, may lead periodically to worsening which will possibly entail suicidal thoughts and intent. That fact makes the argument illogical, especially because suicidal thoughts and intent are difficult both to ascertain and to predict.

Caveat lector. Enough said.


1.            Gillman, PK, Moclobemide and the risk of serotonin toxicity (or serotonin syndrome). CNS Drug Rev, 2004. 10: p. 83-85.

2.            Shaw, GB, Maxims for Revolutionists. 1903.

3.            Gram, LF, Brosen, K, and Group, atDUA, Moclobemide treatment causes a substantial rise in the sparteine metabolic ratio. Brit J of Clin Pharmacol, 1993. 35(6): p. 649-52.

4.            Gram, LF, Guentert, TW, Grange, S, Vistisen, K, et al., Moclobemide, a substrate of CYP2C19 and an inhibitor of CYP2C19, CYP2D6, and CYP1A2: a panel study. Clin Pharmacol Ther, 1995. 57(6): p. 670-7.

5.            Yu, KS, Yim, DS, Cho, JY, Park, SS, et al., Effect of omeprazole on the pharmacokinetics of moclobemide according to the genetic polymorphism of CYP2C19. Clin Pharmacol Ther, 2001. 69(4): p. 266-73.

6.            Schoerlin, MP and Da Prada, M, Species-specific biotransformation of moclobemide: a comparative study in rats and humans. Acta Psychiatr Scand Suppl, 1990. 360: p. 108-10.

7.            Jauch, R, Griesser, E, Oesterhelt, G, Arnold, W, et al., Biotransformation of moclobemide in humans. Acta Psychiatr Scand Suppl, 1990. 360: p. 87-90.

8.            Schoerlin, MP, Mayersohn, M, Hoevels, B, Eggers, H, et al., Cimetidine alters the disposition kinetics of the monoamine oxidase-A inhibitor moclobemide. Clin Pharmacol Ther, 1991. 49(1): p. 32-8.

9.            Isbister, GK, Hackett, LP, Dawson, AH, Whyte, IM, et al., Moclobemide poisoning: toxicokinetics and occurrence of serotonin toxicity. Brit J of Clin Pharmacol, 2003. 56: p. 441-450.

10.            Benazzi, F, Serotonin syndrome with moclobemide-fluoxetine combination. Pharmacopsychiatry, 1996. 29([letter]): p. 162.

11.            Liebenberg, R, Berk, M, and Winkler, G, Serotonergic syndrome after concomitant use of moclobemide and fluoxetine. Human Psychopharmacology, 1996. 11([letter]): p. 146-147.

12.            Gillman, PK, Serotonin syndrome: clomipramine too soon after moclobemide. Int Clin Psychopharmacolog, 1997. 12: p. 339-342.

13.            Dardennes, RM, Even, C, Ballon, N, and Bange, F, Serotonin syndrome caused by a clomipramine-moclobemide interaction. J Clin Psychiatry, 1998. 59(7): p. 382-383.

14.            Graudins, A, Stearman, A, and Chan, B, Treatment of the serotonin syndrome with cyproheptadine. J. Emerg. Med., 1998. 16(4): p. 615-619.

15.            Gillman, PK, Drug interactions and fluoxetine: a commentary from a clinician’s perspective. Ex Op Drug Saf, 2005. 4: p. 965-969.

16.            Bakish, D, Hooper, CL, West, DL, Miller, C, et al., Moclobemide and specific serotonin reuptake inhibitor combination treatment of resistant anxiety and depressive disorders. Human Psychopharmacology, 1995. 10: p. 105-109.

17.            Ebert, D, Albert, R, May, A, Stosiek, I, et al., Combined SSRI-RIMA treatment in refractory depression. Safety data and efficacy. Psychopharmacology, 1995. 119: p. 342-344.

18.            Joffe, RT and Bakish, D, Combined SSRI-moclobemide treatment of psychiatric illness. J Clin Psychiatry, 1994. 55: p. 24-25.

19.            Hawley, CJ, Ratnam, S, Quick, S, Echlin, D, et al., Safety and tolerability of combined treatment with moclobemide and SSRIs: A preliminary study. Eur. Neuropsychopharmacol., 1996. 6(3): p. s127.

20.            Hawley, CJ, Quick, SJ, Ratnam, S, Pattinson, HA, et al., Safety and tolerability of combined treatment with moclobemide and SSRIs - a systematic study of 50 patients. Int Clin Psychopharmacolog, 1996. 11: p. 187-191.

21.            Bonnet, U, SSRI Moclobemide-Combination in the Treatment of Resistant Depression. CNS Drug Rev, 2004. 10(1): p. 86-8.


Old Commentary

This essay considers how the pharmaceutical company Roche have represented their drug moclobemide and how this has influenced the perception of doctors, and their ability to understand the limitations and risks associated with moclobemide. This is now largely a lesson in history, because moclobemide is now used so little, presumably because of general recognition of its weak effects, that it is fading into obscurity. However, although it is safe by itself, even after large over-doses, it does contribute to deaths if it is mixed with selective serotonin reuptake inhibitors (SSRIs) or MDMA, ecstasy (3,4-methylenedioxymethamphetamine).

It is educative to understand how medical training and pharmaceutical companies interact to influence this outcome. It is a highly undesireable outcome and one that persuades me that pharmaceutical companies should cease to have any direct involvement in either education or clinical research. Separation, as suggested by myself, Sotelo and Mansfield must be the preferred option (1-4).

The new MAOI drugs were assigned the epithet RIMAs (reversible inhibitors of monoamine oxidase-A). Moclobemide is the only one still widely available. Moclobemide must be treated as a special case in relation to serotonin toxicity. This is because much emphasis has been placed on moclobemide's differences from the old MAOIs, indeed that must be the presumed reason for the marketing decision to re-name them as RIMAs rather than calling them MAOIs. Most doctors were wary of the old MAOI drugs that were frequently (but inappropriately and incorrectly) referred to as'dangerous' (5,6); an image one would not wish to be associated with.

The way that Roche have dealt with the investigation and marketing of moclobemide may be considered as an exemplar of all those factors that make the drug development process what it is today. Also, the slowness which the medical profession have exhibited in understanding the drug and its interactions, and their naivety in accepting partisan data without sufficient critical analysis, is a model for much of what has been amiss in the medical scientific field for decades.

Moclobemide has a usefully reduced tyramine potentiation propensity. That means it does not generally require a tyramine restricted diet which constitutes a significant advantage over the old irreversible MAOIs; or at least it would be an advantage if moclobemide was a good antidepressant, which it is not. Also, the irreversible binding to monoamine oxidase is the reason old irreversible MAOIs take 2-4 weeks to loose their effect (this is the time taken for the body to synthesise new monoamine oxidase from the genome). Moclobemide, being competitive and reversible, has a much shorter duration of action (2-4 hours). This is expected to lessen the severity and duration of any adverse interactions, including serotonin toxicity with SRIs. Lastly moclobemide is selective for monoamine oxidase-A; whereas the old irreversible MAOIs block both monoamine oxidase-A and B (except selegiline and now rasagiline).

Serotonin toxicity with therapeutic doses of moclobemide (not over-doses) + SRIs has been extensively documented. There are about 25 such cases with doses ranging from 150 mg to 750 mg daily (see below).

The position with the old irreversible MAOIs is clear and incontrovertibly high risk. Doctors who combine the old irreversible MAOIs with any serotonin reuptake inhibitor may be, and probably should be, be considered negligent.

Opinion that it is safe to combine moclobemide with SSRIs has been repeatedly expressed, based on preliminary and incomplete evidence. Moclobemide is the most frequently used MAOI drug, and thus the one most likely to be mixed with an SRI, whether by accident or on purpose.

The CNS Drugs review (Bonnet 2003)

The opinion that the risk benefit ratio for patients may make moclobemide combined with (S)SRIs a reasonable prospect for the treatment of refractory cases of depressive illness continues to be put forward. As recently as 2003 Bonnet suggested:--

"Cases of refractory depression might improve with a combination of moclobemide with other antidepressants, such as clomipramine or a SSRI. Since this combination has rarely (KG's italics) been associated with a potentially lethal serotonin syndrome, it requires lower entry doses, a slower dose titration and a more careful monitoring of patients."

Bonnet cited the rather partisan Dingemanse papers (see below) from the makers, Roche, in support of his views. Both Freeman (a former editor of the Brit J Psychiatry who was not a psychopharmacologist at all) and DeBattista have quoted Dingemanse and made similar statements (7,8). Freeman"s view, especially, was expressed at a time when he was clearly unaware of various important facts and it should be disregarded. It shows that even people of Freeman's principles and stature can be flattered into expressing opinions about subjects of which they know little.

DeBattista et al state, re combining the SSRIs and moclobemide "... moclobemide may hold promise for less problematic drug interactions. Usual therapeutic doses of the SSRIs have generally been well tolerated in combination with moclobemide". They also quote Dingemanse (9,10) in justification of this, but seem to have been unaware of Hawley's prior work (11,12). They were more cautious in their advocacy of the combination and made appropriate note of the deaths with over-doses.

This is a contentious issue with ethical considerations. One is reminded of the American civil war general who did not think snipers were effective: he raised his head above the parapets and was rudely"cut short" in his ultimate utterance: "Nonsense, they couldn't hit an elephant at this dist...." (13)

I have published a comment on Bonnet"s CNS Drugs review suggesting, inter alia, that these combinations are risky, "The above data indicates clearly that moclobemide/SRI combinations represent a predictably risky strategy"(14), to which Bonnet responded (15) :-- …"Dr. Gillman is correct in emphasizing the danger of SS and the predictive nature of the Hunter criteria, which could help us to identify and treat SS in early stages. It should be realized, however, that the use of a potentially hazardous combination therapy is sometimes unavoidable to overcome the burden of treatment-resistant affective and anxiety disorders. Usually, SS resolves within 24 h after discontinuation of the offending drug(s)."

I agree about the burden of treatment-resistant affective and anxiety disorders. I agree that treatments regarded by mainstream texts as"dangerous", eg MAOIs + TCAs, are valuable and I use them extensively (according to figures available to me, 10 times more frequently than the"average" psychiatrist). Indeed I have published the opinion that the term "<strong>dangerous</strong>" is quite inappropriate when used, as it not infrequently is, in association with MAOIs (5).

When describing moclobemide + SSRIs as a"predictably risky strategy" I am not labelling that as dangerous in an absolute sense; but I do wish to emphasise that those who engage in such attempts will serve their patients best if they are as well informed as possible. The apparent level of knowledge about which drugs can be predicted to cause serious and life-threatening toxicity is well illustrated by numerous examples where many researchers, most of them from reputable hospitals and teaching institutions, have gained ethical approval to perform studies that exposed participants to risks that those researchers appear not to have had the knowledge to analyse or to quantify. It is unlikely such studies would have gained ethical approval (or volunteers to participate in them) had a fully informed estimate of the risks been available. It is evident from the citations in the references quoted below that a proportion of these authors were unaware of relevant published material that would and should have allowed them to develop a more critically informed view (10, 12, 16-31).

I am not so positive about the evidence for effectiveness or safety of moclobemide + SSRIs; but I think more work could be done. I am not sure who will do it, I hope they are experienced and well informed about serotonin toxicity.

In 1995 I published the following warning: "Whether or not moclobemide will exhibit the same propensity as the older MAOIs for causing the SS in man is still uncertain… In relation to the serious interactions of moclobemide with SRIs it is notable that, despite reports from 1962 onwards of "serotonin syndrome" in animals, no cases were noted in humans for twenty years, until Insel"s report of 1982 (they certainly ocurred, but were not recognised). It is important to ensure that this mistake is not repeated with moclobemide, SRIs and opioids like meperidine (pethidine)" (32).

I also criticised Freeman"s Lancet review; I was informed that his review generated some angst in the Lancet editorial office, especially since Freeman was not a psychopharmacologists (33). "There are an increasing number of reports of SS; a computer search located 6 in 1991 and 25 in 1995. Some of them were fatalities after overdose, and near fatalities with therapeutic doses; and so it is salutary to observe that some twenty years elapsed between the original animal work on SS, and its recognition in humans. Indeed it is still under recognised and under reported and the mechanism of the interaction poorly appreciated; so much so that a major journal published a drug profile of moclobemide (a reversible inhibitor of MAO A) which said it had no interactions with SSRIs or TCAs, despite a pre-existing report in the same journal."

I was referring to the report from Neuvonen that Freeman did cite but appears to have discounted, perhaps because they were over-doses? (34). However there were other reports he did not cite and was presumably unaware of (35,36).

In my Biological Psychiatry paper I have summarised my view: "These interactions are complex, especially when phamacokinetic factors influence drug levels, which then have further pharmacodynamic interactions. The misunderstanding of drug interactions played a part in the demise of mixed treatment regimes using the old MAOIs plus TCAs (37,38). Adequate trials of these various combinations have not been undertaken, so valuable data concerning potentially effective treatments may remain undiscovered. Sargent's words from 40 years ago are still worth noting: 'We are never going to learn how to treat depression in an MRC statistician's office' (39). Most current antidepressant agents produce small improvements that require clarification utilizing statistics, but only because the clinical effect is small. Sargent meant that when the treatment effect is large statistics become a refinement, rather than a necessity" (40).

The"Dingemanse" papers

It is important to document this part of the moclobemide story because of the lessons it can teach us about the influence and bias of pharmaceutical companies and how that influences the literature. The view that moclobemide is a safe, and that it may be a useful treatment strategy, when combined with SSRIs, has been promulgated repeatedly in various papers published (mostly) by Roche personnel like Dingemanse. These papers are still being quoted to support this view despite the preliminary nature of the observations, and what has subsequently become a large body of contradictory evidence (as detailed here) (9, 10, 29, 41-43).

The potentially misleading view that there is probably no interaction between SSRIs and moclobemide has been reiterated by Dingemanse most recently in 1998; when the conclusion was offered:--"Combination treatment with fluoxetine and moclobemide did not provide any indication of development of the serotonin syndrome."

Most data on moclobemide has been derived from studies using doses of 300-450 mg in normal healthy volunteers; its propensity to cause serotonin toxicity at doses of 600 mg or greater has not been extensively or well studied. In the 1998 Dingemanse paper all subjects selected were cytochrome P450 2D6 extensive metabolisers, tending to make them have lower blood levels and be lower risks for interactions. The authors did not explain or discuss the reason for that odd choice. The period of 5 days on the usual minimum dose of fluoxetine 20 mg, combined with moclobemide 600 mg daily in 12 subjects, does not give confidence in the conclusion the paper offers. Dingemanse's statement that moclobemide + fluoxetine…"did not provide any indication of development of the serotonin syndrome" is seriously misleading to all but the most sophisticated of readers.

The usual minimum dose of fluoxetine 20 mg daily is frequently exceeded in primary care practice, and doses of 60 mg or more are common in specialist practice. The use of fluoxetine 20 mg daily, in a sample of extensive metabolisers (EMs) of 2D6 dependant drugs, renders the Dingemanse study of diminished value because it is unrepresentative of the clinical situations that will frequently and inevitably occur. There was limited indication of how these important factors might affect the outcome. Because extrapolation from this data to a clinical population is of doubtful scientific validity, this paper will mislead clinicians who are not well informed (i.e. the great majority of clinicians) and who do not read it very critically and carefully. Fluoxetine pharmacokinetics are non-linear, so the frequently used doses of 40-60 mg and more will produce disproportionately higher blood levels and an elevated risk of inducing serotonin toxicity. Also at those levels it will be even more likely to precipitate a pharmacokinetic interaction, thus also elevating moclobemide levels. Dingemanse, using fluoxetine at a dose of only 20 mg, found 'markedly elevated moclobemide levels'. Unless contrary evidence becomes available it must be assumed that higher fluoxetine doses will be more risky by virtue of this combined pharmacokinetic and pharmacodynamic interaction. Indeed, pharmacokinetic and pharmacodynamic interactions between moclobemide and fluoxetine serve as a teaching model for educating doctors about importance of these aspects of pharmacology.

This interaction problem is further compounded by the dangerous potential of the long elimination half-life for fluoxetine***, of up to 5 days, and for norfluoxetine up to 20 days. The role of 2D6 in fluoxetine's elimination is still uncertain; it seems the R and S enantiomers are converted to nor-fluoxetine by 2C19 and 2D6, but the elimination of norfluoxetine, which is active both as an SRI and a cytochrome P450 enzyme inhibitor, is less well elucidated (44-47).

*** dangerous potential of the long elimination half-life of fluoxetine note, in the first series of deaths (total of 7 deaths) from serotonin toxicity in the fluoxetine trials back in the 1980s some deaths occurred when MAOIs were started more than 30 days after ceasing fluoxetine (17).

Prior to the Dingemanse paper, there had been at least 15 publications, reporting serotonin toxicity with moclobemide and SRIs (up to the end of 1996). Most, but by no means all, were with over-doses (note- Dingemanse paper was accepted October 1997): none of them were cited in the 1998 paper, neither was there any mention or discussion of these extensively documented and serious problems, which included deaths involving moclobemide.

Moclobemide data sheet / PI / MIMS

In my opinion the current warnings in relation to the risk of morbidity and death from serotonin toxicity with moclobemide + (S)SRIs * in the, data sheet / PI / MIMS type texts, is incomplete, potentially misleading and out of date. Firstly, it fails to recognise or explain the link between narcotic analgesics, imipramine / clomipramine, SSRIs and other drugs like sibutramine, being implicated by virtue of their common property of serotonin reuptake inhibition. Secondly, there is insufficient emphasis or explanation of the now established probability of severe side effects blending into serotonin toxicity and fatalities with mixtures of SRIs and MAOIs of all types. Thirdly, there is no adequate description of the now established characteristic features of serotonin toxicity which would facilitate clinicians being able to recognise serotonin toxicity when it occurs. Fourthly, there is inadequate information or guidance about the need for urgent assessment and treatment of symptoms that indicate life-threatening toxicity in relation to the established high risk of severe serotonin toxicity, and the possibility of fatalities, with moclobemide and SRI combinations.

In my opinion this may place companies (Roche with moclobemide, and Pfizer with linezolid, and maybe others) in a weak position to defend themselves if they are sued for damages. The risk of serious adverse effects is now a matter of fact and is beyond any argument; the only question is the extent of that risk. The legal concept of "failure to warn" seems relevant. See (48)

It may be noted that some Roche data sheets (PI) contain misleading and contradictory information that indicates a flawed understanding the problem, especially with clomipramine, TCAs and SSRIs. One Roche PI states: "in a clinical study designed to test the interaction between MANERIX and a tricyclic antidepressant (clomipramine), severe adverse reactions emerged and the study was terminated. Data involving other tricyclic antidepressants are limited. Consequently, the concomitant use of MANERIX and tricyclic antidepressants is contraindicated. Clinical data are not available on the concomitant use of MANERIX and selective serotonin reuptake inhibitors or conventional MAO inhibitors. Therefore, until such data become available, MANERIX should not be administered in combination with these agents."

This information is wrong and not in keeping with current knowledge. Also, it is outrageous, unethical and immoral that the"severe adverse reactions emerged and the study was terminated (clomipramine)", ocurred but was never communicated to doctors in any way, or published.

Another Roche data sheet (PI) statement is wrong: "Clinical data are not available on the concomitant use of MANERIX and selective serotonin reuptake inhibitors or conventional MAO inhibitors". That is incorrect. The data from Hawley, analysed in detail herein, incontrovertibly contradicts that.

Serotonin toxicity reports prior to the 1998 Dingemanse paper

For completeness, the serotonin toxicity cases with moclobemide that might have been noted in the Dingemanse paper, are cited below. References up to the end of 1996 (Dingemanse paper was accepted Oct 1997, so I am being generous) (32, 34-36, 49-59).

Systematic data on moclobemide/SRI toxicity

The only systematic data concerning moclobemide/SRI toxicity is from the HATS team of an analysis of 106 moclobemide over-doses from their large data base. There were a total of 106 moclobemide ingestions

  • 33-- moclobemide alone, of these 1/33 had ST (52 moclobemide + non-serotonergic drug; no further information given).
  • 21-- moclobemide + SSRI. Of these 11/21 had ST. 6 were 'severe' with temp of > 38.5c needing paralysis and intubation.

The above data has now been fully published in peer reviewed form .

The risk with moclobemide (at therapeutic doses) and (S)SRIs combined is almost certainly much lower than with MAOIs like tranylcypromine and (S)SRIs combined. The studies below demonstrate a significant incidence of serotonergic side effects merging into serotonin toxicity, and Hawley discontinued such work because of toxicity concerns (see below) (10, 12, 22-24, 41, 43).

Moclobemide/SRI toxicity compared with MAOI/SRI toxicity

In contrast, Razani, using the same (low) tranylcypromine dose (20 mg) as Oefele, but combining it with amitriptyline, maximum dose 150 mg (a weak SRI), rather than clomipramine (a potent SRI), reported no serotonin toxicity in a four week trial with 20 patients. This accords with many other early studies (see other references) and years of clinical experience. My own is of the treatment of approximately 1000 patients with tranylcypromine, about half of whom have also been administered concurrently amitriptyline, nortriptyline or doxepin with no serotonin toxicity, nor even problems with significant serotonergic side effects (61-74).

Note: HATS contains some instances of combination over-doses with moclobemide + SRIs (see above; = ~ 30 as of 2003) but few of MAOI + SRI, so these are not (yet) amenable to statistical analysis to elucidate the relative risk of MAOI/SRI vs. Moc/SRI.

It seems that the nearly fatal / fatal cases have been more frequent with MAOI/SRI combinations than the Moc/SRI ones, which, although requiring intubation / paralysis ventilation / 5-HT 2A antagonists, may be somewhat less severe. Nevertheless deaths still occur with Moc / SRI combinations.

The pharmacokinetics of moclobemide

Moclobemide shows low affinity competitive, selective and reversible binding to MAO-A with a Ki in the high micro-molar range (200-400 µ mol). In vitro the half-life for reactivation of MAO-A is 90-120 minutes. In rat liver the ED 50 for MAO A and B inhibition are similar (8.4 vs. 6.6 µmol) but in the brain it appears selective for MAO-A (ED 50 7.6 µmol vs. MAO-B 78 µmol) (75, 76).

The pharmacokinetics of moclobemide itself is linear even in over-dose but levels are raised by interactions, eg with fluoxetine. The daily dose of moclobemide has escalated as the experience (or frustration?) of clinicians grows; although 600 mg is the stated maximum, doses of 2400 mg daily are used by some doctors in practice.

Moclobemide: suggested prescriber information / warning for patients

In my opinion the current warnings, in the moclobemide PI / data sheets (various English language ones vary a little), in relation to the risk of serious adverse effects and death from serotonin toxicity with moclobemide, when combined with all types of SRIs, does not adequately reflect the current knowledge and understanding of serotonin toxicity. I therefore suggest that doctors and pharmacists, who prescribe or supply moclobemide, do all they can to ensure that all non-current supplies of moclobemide and serotonin reuptake inhibitors of any sort are not present in the same household, and also to warn people of the serious risks. A child might only have to take a few (3 or 4?) tablets of each to kill themselves, in a rather horrible way.


1. Gillman, P.K., Disease mongering: one of the hidden consequences. Public Library of Science: Medicine, 2006. 3(7): p. e316. DOI: 10.1371/journal.pmed.0030316.

2. Gillman, P.K., Pharmaceutical company influence. Public Library of Science: Medicine, 2006. 2(12): p. e392

3. Sotelo, J., Regulation of Clinical Research Sponsored by Pharmaceutical Companies: A Proposal. PLoS Medicne, 2006. 3(7): p. e306.

4. Mansfield, P.R., et al., Educating health professionals about drug and device promotion: advocates' recommendations. PLoS Med, 2006. 3(11): p. e451.

5. Gillman, P.K., Safety of MAOIs and side effects clomipramine. Pharmabulletin, 1990. 14: p. 15.

6. Gillman, P.K., Monoamineoxidase inhibitors. Current Therapeutics, 1984. 43: p. 79.

7. Freeman, H., Moclobemide. Lancet, 1993. 342: p. 1528-1532.

8. DeBattista, C., M. Sofuoglu, and A.F. Schatzberg, Serotonergic synergism: the risks and benefits of combining the selective serotonin reuptake inhibitors with other serotonergic drugs. Biological Psychiatry, 1998. 44(5): p. 341-347.

9. Dingemanse, J., An update of recent moclobemide interaction data. International Clinical Psychopharmacology, 1993. 7: p. 167-180.

10. Dingemanse, J., et al., Pharmacokinetic and pharmacodynamic interactions between fluoxetine and moclobemide in the investigation of development of the "serotonin syndrome". Clinical Pharmacology and Therapeutics, 1998. 63(4): p. 403-13.

11. Hawley, C.J., et al., Safety and tolerability of combined treatment with moclobemide and SSRIs: A preliminary study. European Neuropsychopharmacology, 1996. 6(3): p. s127.

12. Hawley, C.J., et al., Safety and tolerability of combined treatment with moclobemide and SSRIs - a systematic study of 50 patients. International Clinical Psychopharmacology, 1996. 11: p. 187-191.

13. Sedgwick, J., Battle of the Wilderness (1864 May 9th) Spottsylvania Court House, Virginia. Bloomsbury Dictionary of Quotations, 1991: p. 337.

14. Gillman, P.K., Moclobemide and the risk of serotonin toxicity (or serotonin syndrome). Central Nervous System Drug Reviews, 2004. 10: p. 83-85.

15. Bonnet, U., SSRI Moclobemide-Combination in the Treatment of Resistant Depression. Central Nervous System Drug Reviews, 2004. 10(1): p. 86-8.

16. Feighner, J.P., et al., Adverse consequences of fluoxetine-MAOI combination therapy. J Clin Psychiatry, 1990. 51(6): p. 222-5.

17. Beasley, C.M., et al., Possible monoamine oxidase inhibitor-serotonin uptake inhibitor interaction: fluoxetine clinical data and preclinical findings. Journal of Clinical Psychopharmacology, 1993. 13: p. 312-320.

18. Insel, T.R., et al., Possible development of the serotonin syndrome in man. American Journal of Psychiatry, 1982. 139: p. 954-955.

19. Oefele, K.V., R. Grohmann, and E. Ruther, Adverse drug reactions in combined tricyclic and MAOI therapy. Pharmacopsychiatry, 1986. 19: p. 243-244.

20. Amsterdam, J.D., F. Garcia-Espana, and M. Rosenzweig, Clomipramine augmentation in treatment-resistant depression. Depression and Anxiety, 1997. 5(2): p. 84-90.

21. Diamond, S., et al., Serotonin syndrome induced by transitioning from phenelzine to venlafaxine: four patient reports. Neurology, 1998. 51(1): p. 274-276.

22. Joffe, R.T. and D. Bakish, Combined SSRI-moclobemide treatment of psychiatric illness. Journal of Clinical Psychiatry, 1994. 55:1: p. 24-25.

23. Bakish, D., et al., Moclobemide and specific serotonin reuptake inhibitor combination treatment of resistant anxiety and depressive disorders. Human Psychopharmacology, 1995. 10: p. 105-109.

24. Ebert, D., et al., Combined SSRI-RIMA treatment in refractory depression. Safety data and efficacy. Psychopharmacology, 1995. 119: p. 342-344.

25. Sovner, R., Amphetamine and tranylcypromine in treatment-resistant depression. Biological Psychiatry, 1990. 28(11): p. 1011-2.

26. Myronuk, L.D., M. Weiss, and L. Cotter, Combined treatment with moclobemide and methylphenidate for comorbid major depression and adult attention-deficit/hyperactivity disorder. J Clin Psychopharmacol, 1996. 16(6): p. 468-9.

27. Hartter, S., et al., Inhibition of dextromethorphan metabolism by moclobemide. Psychopharmacology (Berl), 1998. 135(1): p. 22-6.

28. Fawcett, J., et al., CNS stimulant potentiation of monoamine oxidase inhibitors in treatment-refractory depression. J Clin Psychopharmacol, 1991. 11(2): p. 127-32.

29. Dingemanse, J., et al., Switch in treatment from tricyclic antidepressants to moclobemide: a new generation monoamine oxidase inhibitor. J Clin Psychopharmacol, 1995. 15(1): p. 41-8.

30. Konig, F. and M. Wolfersdorf, Combination therapy using moclobemide with tricyclic and tetracyclic antidepressants to treat therapy-resistant depression. Pharmacopsychiatry, 1997. 30(3): p. 93-6.

31. Ramasubbu, R., Switching to moclobemide to reverse fluoxetine-induced sexual dysfunction in patients with depression. J Psychiatry Neurosci, 1999. 24(1): p. 45-50.

32. Gillman, P.K., Possible serotonin syndrome with moclobemide and pethidine. Medical Journal of Australia, 1995. 162([letter]): p. 554.

33. Gillman, P.K., Serotonin syndrome treated with chlorpromazine. Journal of Clinical Psychopharmacology, 1997. 17: p. 128-129.

34. Neuvonen, P.J., et al., Five fatal cases of serotonin syndrome after moclobemide-citalopram or moclobemide-clomipramine overdoses. Lancet, 1993. 342: p. 1419.

35. Myrenfors, P.G., et al., Moclobemide overdose. Journal of Internal Medicine, 1993. 233: p. 113-115.

36. Spigset, O., T. Mjorndal, and O. Lovheim, Serotonin syndrome caused by a moclobemide-clomipramine interaction. British Medical Journal, 1993. 306(Jan 23): p. 248.

37. Gillman, P.K. and I.M. Whyte, Serotonin syndrome, in Adverse Syndromes and Psychiatric Drugs, P. Haddad, S. Dursun, and B. Deakin, Editors. 2004, Oxford University Press: Oxford. p. 37-49.

38. Gillman, P.K., Serotonin syndrome: history and risk. Fundamental and Clinical Pharmacology, 1998. 12(5): p. 482-491.

39. Sargent, W., Combining the antidepressant drugs. British Medical Journal, 1965. 1: p. 1251.

40. Gillman, P.K., A review of serotonin toxicity data: implications for the mechanisms of antidepressant drug action. Biological Psychiatry, 2006. 59: p. 1046-51.

41. Amrein, R., et al., Interactions of moclobemide with concomitantly administered medication: evidence from pharmacological and clinical studies. Psychopharmacology, 1992. 106(suppl): p. s24-31.

42. Delini-Stula, A. and H. Mikkelsen, Risk of 5-HT syndrome after co-administration of moclobemide (RIMA) and selective 5-HT re-uptake inhibitors (SSRI)- analysis of cases. European Neuropsychopharmacology, 1993. 3(3): p. 337-338 poster.

43. Wallnofer, A., et al., Moclobemide and fluoxetine co-administration: a prospective study in healthy volunteers to investigate the potential developement of serotonin syndrome. Human Psychopharmacologyy, 1995. 10: p. 25-31.

44. Harvey, A.T. and S.H. Preskorn, Fluoxetine pharmacokinetics and effect on CYP2C19 in young and elderly volunteers. J Clin Psychopharmacol, 2001. 21(2): p. 161-6.

45. Sallee, F.R., C.L. DeVane, and R.E. Ferrell, Fluoxetine-related death in a child with cytochrome P-450 2D6 genetic deficiency. J Child Adolesc Psychopharmacol, 2000. 10(1): p. 27-34.

46. Margolis, J.M., et al., (R)-, (S)-, and racemic fluoxetine N-demethylation by human cytochrome P450 enzymes. Drug Metab Dispos, 2000. 28(10): p. 1187-91.

47. Sanchez, C., et al., Escitalopram versus citalopram: the surprising role of the R-enantiomer. Psychopharmacology (Berl), 2004. 174(2): p. 163-76.

48. Healy, D., A Failure to Warn. International Journal of Risk and Safety in Medicine, 1999. 12: p. 151-156.

49. Benazzi, F., Serotonin syndrome with moclobemide-fluoxetine combination. Pharmacopsychiatry, 1996. 29: p. 162.

50. Brodribb, T.R., M. Downey, and P.J. Gilbar, Efficacy and adverse effects of moclobemide. Lancet, 1994. 343([letter]): p. 475.

51. Fischer, P., Serotonin syndrome in the elderly after antidepressive monotherapy. Journal of Clinical Psychopharmacology, 1995. 15(6): p. 440-442.

52. Hernandez, A.F., et al., Fatal moclobemide overdose or death caused by serotonin syndrome? Journal of Forensic Sciences, 1995. 40: p. 128-130.

53. Iwersen, S. and A. Schmoldt, Three suicide attempts with moclobemide. Journal of Toxicology Clinical Toxicology, 1996. 34(2): p. 223-5.

54. Kuisma, M.J., Fatal serotonin syndrome with trismus. Annals of Emergency Medicine, 1995. 26([letter]): p. 108.

55. Liebenberg, R., M. Berk, and G. Winkler, Serotonergic syndrome after concomitant use of moclobemide and fluoxetine. Human Psychopharmacology, 1996. 11([letter]): p. 146-147.

56. Power, B.M., et al., Fatal serotonin syndrome following a combined overdose of moclobemide, clomipramine and fluoxetine. Anaesthesia and Intensive Care, 1995. 23: p. 499-502.

57. Reith, D.M., et al., Clinical features of self-poisoning with monoamine oxidase inhibitors and / or serotonin reuptake inhibitors. Proceedings of the Australasian Society of Clinical and Experimental Pharmacologists and Toxicologists, 1996. 3( (Abstract)): p. 29.

58. Robert, P., et al., Syndrome serotoninergique lors d'une intoxication aigue aux antidepresseurs. Annales Francaises d Anesthesie et de Reanimation, 1996. 15(Serotonin syndrome following acute intoxication with antidepressants (Serotonin syndrome in acute poisoning with antidepressive agents)): p. 663-665.

59. Whyte, I.M. and N.A. Buckley, Relative toxicity and resource utilisation in antidepressant self poisoning: Tricyclics vs specific serotonin reuptake inhibitors. Proceedings of the Australasian Society of Clinical and Experimental Pharmacologists and Toxicologists, 1995. 2: p. 103.

60. Isbister, G.K., et al., Moclobemide poisoning: toxicokinetics and occurrence of serotonin toxicity. British Journal of Clinical Pharmacology, 2003. 56: p. 441-450.

61. Sethna, E.R., A study of refractory cases of depressive illness and their response to combined antidepressant treatment. British Journal of Psychiatry, 1974. 124: p. 265-272.

62. Schuckit, M., E. Robins, and J. Feighner, Tricyclic antidepressants and monoamine oxidase inhibitors. Archives of General Psychiatry, 1971. 24: p. 509-514.

63. Sjoqvist, F., Psychotropic drugs (2). Interaction between monoamine oxidase (MAO), inhibitors and other substances. Proceedings of the Royal Society of Medicine, 1965. 58: p. 967-978.

64. Sargent, W., Antidepressant drugs. British Medical Journal, 1965. 1: p. 1495.

65. Randall, J., Combining the antidepressant drugs. British Medical Journal, 1965. 1: p. 521.

66. Gander, D.R., Treatment of depressive illness with combined antidepressants. Lancet, 1965. 1: p. 107-109.

67. Gander, D.R., Combining the antidepressant drugs. British Medical Journal, 1965. 1: p. 521.

68. Nymark, M. and I. Møller Nielsen, Reactions due to the combination of monoamineoxidase inhibitors with thymoleptics, Pethidine or methylamphetamine. Lancet, 1963. 2([letter]): p. 524-525.

69. Davies, E.B., Combining the antidepressants. Lancet, 1963. 2: p. 781-782.

70. White, K., et al., Combined MAOI-tricylic antidepressant treatment: a controlled trial. Psychopharmacology Bulletin, 1982. 18(4): p. 180-1.

71. White, K. and G. Simpson, Combined MAOI-tricyclic antidepressant treatment: a reevaluation. Journal of Clinical Pharmacology, 1981. 1: p. 264-281.

72. Simpson, G.M., Combined antidepressant therapy. British Journal of Psychiatry, 1972. 120: p. 695.

73. Razani, J., et al., The safety and efficacy of combined amitriptyline and tranylcypromine antidepressant treatment. A controlled trial. Archives of General Psychiatry, 1983. 40: p. 657-661.

74. Abdala, E.N., [Combination of tricyclic antidepressants and MAOI in the depressions]. Acta Psiquiatr Psicol Am Lat, 1975. 21(1): p. 52-5.

75. Da Prada, M., et al., Neurochemical profile of moclobemide, a short-acting and reversible inhibitor of monoamine oxidase type A. Journal of Pharmacology and Experimental Therapeutics, 1989. 248(1): p. 400-14.

76. Da Prada, M., et al., The pharmacology of moclobemide. Reviews in Contemporary Pharmacotherapy, 1994. 5: p. 1-18.