The serotonin reuptake inhibitor potency of most of the commonly used opioid analgesics is now known from quality in vitro data. This data is entirely consonant with my previous interpretation concerning which of these drugs were, or were not, capable of precipitating serotonin toxicity. The most important substantiation of my predictions made from the spectrum concept of serotonin toxicity concerns the widely used drug fentanyl, and its congeners. Fentanyl is now established to have no significant serotonin reuptake inhibitor potency and it can be concluded that it poses no risk of serotonin toxicity when mixed with MAO inhibitors (or any other ‘serotonergic’ drugs).


Note: Much of the data in the old commentary, which is retained here below this update, was published in my review in the British J. Anaesthesia (1).

The situation with opioid analgesics is now clear and this March 2015 updating of this commentary has been precipitated by recent research (2, 3) which contains useful new data concerning the in vitro SRI potency of opioids, including fentanyl*. This data allows a considerable contraction of some of the detailed interpretation in my previous summary of this subject. I have inserted this new more concise note in front of the old commentary which I have preserved for historical interest, and for the interest of that proportion of researchers who may wish to have access to it.

It is useful to begin by noting that the spectrum concept of serotonin toxicity has yet again yielded reliable predictions of which drugs do and do not have significant serotonin reuptake inhibition actions from their propensity to provoke ST when combined with MAOIs. This is especially relevant and important in view of the epidemic of unnecessary warnings issued by various government agencies in the Western world concerning a whole raft of drugs including triptans, 5-HT3 antagonists and many others. These erroneous warnings are based on case reports and even less satisfactory reports to drug monitoring agencies. My universal experience in assessing such data (in the context of serotonin toxicity) is that it is almost always of negligible value and has repeatedly lead to mistaken conclusions and actions.

It is therefore especially gratifying to report another successful prediction of the spectrum concept of serotonin toxicity concerning fentanyl and its congeners.

The Current Situation

Members of the phenylpiperidine series of opioids have long been known or suspected to posses SRI potency of possibly sufficient degree to give risk of ST (viz. pethidine (meperidine), methadone and fentanyl, and d-propoxyphene) and also tramadol.

My previous commentary and published comments explained that the phenanthrene series (morphine etc.) were known to be safe (i.e. they have no SRI activity) whereas tramadol and pethidine (meperidine) appeared to carry a significant ST risk, but fentanil was probably safe.

The classes of opioids are; phenanthrenes: Morphine, Hydromorphone, Oxymorphone, Levorphanol, Codeine, Hydrocodone, Oxycodone, Buprenorphine, Butorphanol, Nalbuphine, Pentazocine, Dezocine;

phenylheptylamines: Methadone, Propoxyphene, Levomethadyl;

phenylpiperidines: Fentanyl, sufentanyl, Meperidine aka pethidine.

Tramadol is a weak atypical opioid (4-phenyl-piperidine analogue of codeine and structurally very similar to venlafaxine and with a similar toxicity profile).

We now have a good indication from in vitro data produced by Barann, Stamer et al. (2) of the relative SRI potency of tramadol, pethidine (meperidine), fentanyl, alfentanil, morphine and hydromorphone. Tramadol is the most potent SRI which is in keeping with its seemingly greater potential to induce ST, followed by pethidine (meperidine). The others fentanyl, alfentanil, morphine and hydromorphone have no SRI potency.

Until now the main drug of clinical relevance about which there was significant uncertainty was fentanyl. I have previously argued that it is almost certainly not associated with reports of serotonin toxicity and has frequently been used in conjunction with MAOIs without any signs of toxicity (see discussion of this in relation to MB), thereby proving it has negligible SRI potency (4, 5). That view is firmly supported by this new data.

This new data and analysis enables clinicians to be even more confident when making decisions about patient management. The degree of risk with known serotonergic opioids like meperidine is relatively low (see details in old commentary below), but it is not precisely predictable and the potentially fatal outcome of ST suggests that in usual circumstances it would be prudent to avoid serotonergic opioids in patients who have been taking MAOIs (like tranylcypromine and phenelzine etc see link) and to use the drugs known not to be SRIs where possible.

*Reminder: quite a lot of drugs have alternative spellings, usually for no good reason. In this particular instance fentanyl is usually spelt with a ‘y’ whereas sufentanyl is usually spelt with an ‘i’, but one encounters variations of both.


1.Gillman, PK, Monoamine oxidase inhibitors, opioid analgesics and serotonin toxicity. Br. J. Anaesth., 2005. 95: p. 434-441.

2.Barann, M, Stamer, UM, Lyutenska, M, Stuber, F, et al., Effects of opioids on human serotonin transporters. Naunyn. Schmiedebergs Arch. Pharmacol., 2015. 388(1): p. 43-9.

3.Barann, M, Urban, B, Stamer, U, Dorner, Z, et al., Effects of tramadol and O-demethyl-tramadol on human 5-HT reuptake carriers and human 5-HT3A receptors: a possible mechanism for tramadol-induced early emesis. Eur J Pharmacol, 2006. 531(1-3): p. 54-8.

4.Gillman, PK, Fentanyl and serotonin toxicity. Br. J. Anaesth., 2009: p.

5.Gillman, PK, Meperidine (pethidine) and toxicity. Br. J. Anaesth., 2010: p.


Previous Commentary

Note: Much of the data below has now been published in my review in the BJA (1).

The very first cases of serotonin toxicity occurred when patients were given the early anti-tuberculous drugs. Mitchell’s report, which is the first ever case in a human, was an interaction between iproniazid and pethidine. Some of the other early cases referenced below were precipitated by the use of pethidine in patients being treated with iproniazid for angina. The physicians looking after these patients were aware of, and cited, the earlier reports, documented herein, where chlorpromazine had been successfully employed to lessen such reactions, and the earliest reports of successful use of chlorpromazine in humans were in these patients. Some of these were reported by anaesthetists and occurred in patients having operative procedures like leucotomies. As a result of the uncertainty in this area Churchill-Davidson (2) developed a testing procedure for checking whether patients on MAOIs were likely to exhibit a reaction to analgesics. Evans-Prosser (3) modified this procedure in 1968 and described an experiment in which they gave 15 patients injections of pethidine or morphine or water, in graduated doses, under controlled conditions in hospital. The total dose of pethidine was 75 mg and of morphine 7.5 mg. Each of the 15 patients received in blind order each of these three treatments. No reactions suggestive of serotonin toxicity were observed. It may be noted that although the patients were observed carefully these experimenters clearly did not know what symptoms of serotonin toxicity to look for and only measured the blood pressure and heart rate. No significant changes were noted. However, they were carefully observed, and the expected reaction of shivering, agitation, clonus and hyperreflexia would have been very obvious, and may reliably be presumed not to have occurred. This group of patients represents the only case series of this sort, and even now it is valuable data. It demonstrates clearly what the spectrum concept of serotonin toxicity predicts, which is that weak serotonergic opioids like pethidine are capable of precipitating serotonin toxicity, but this is only likely with susceptible individuals, or with particularly large doses.

All the references I have located, through a search of the listed papers and standard databases, are listed under the relevant headings below. Pethidine has undoubtedly caused a number of reactions which have been severe, with some fatalities; the other serotonergic opioids should be expected to precipitate fatalities in similar circumstances. The more recent fatalities are: (4-8). See also Gillman (9) for further comment on older cases.

Especially when taken in conjunction with the Evans Prosser paper above, this is a clear illustration of how a small case series using limited doses may lead to a false sense of security, because the dose-effect relationship is not sufficiently appreciated. This is especially important when phamacokinetic and pharmacodynamic interactions coincide (cf moclobemide and Dingemanse’s papers).

The limited available clinical and experimental data agree that morphine analogues are not serotonin reuptake inhibitors (or releasers), nor do any of them precipitate serotonin toxicity with monoamine oxidase inhibitors (viz.: Morphine, Codeine, oxycodone, buprenorphine). There are no reliable reports of serotonin toxicity with those drugs (see table 7).

The phenylpiperidine series opioids, pethidine (meperidine), tramadol, methadone and fentanyl, and d-propoxyphene, are all borderline, being very weak serotonin and re-uptake inhibitors (table 7). These drugs have been involved in multiple reports which are almost certainly serotonin toxicity, as judged by a subjective interpretation of the old case reports, informed by more accurate recent data concerning the characteristic features of serotonin toxicity. More precise data from human cloned receptor assays would be valuable because we can reliably predict that only those drugs that are significant serotonin reuptake inhibitors (or releasers) will precipitate serotonin toxicity.

We can reliably predict that only those opiates that are significant serotonin reuptake inhibitors (or releasers) will precipitate serotonin toxicity with MAOIs. Human cloned receptor data 5-HT transporter affinity of narcotic analgesics (antidepressants for comparison) from (10).

DRUG Ki nM ST reports with MAOIs
  Other refs Codd  
All SSRIs 0.13 - 2.2 potent   Definite, and frequent fatalities
Clomipramine 0.14 potent   Definite, and frequent fatalities
Imipramine 1.3   Definite, and some fatalities
Amitriptyline 2.8-36 weak   None, no fatalities
Venlafaxine 7.5-102 anomalous*   Definite, and fatalities (> SSRIs?)
Morphine 500,000 >100,000 None
Codeine - >100,000 None
Buprenorphine - >100,000 None
Oxycodone - >100,000 None
Tramadol 760-1820*** 528 Definite, and possible fatalities
Dextromethorphan - 23 Definite
d-propoxyphene   30,000 ??? see refs
Pethidine 413 weak, anomalous** - Definite, and some fatalities
Pentazocine - - No reports known
Fentanil - - Uncertain. One case of ST, and one possible death reported
Remifentanil - - No case or death reported
Methadone 270 14.1 No reports known, but possible

Table notes:-- In vitro receptor assays give an estimate of the drugs potency as an SRI; a lower Ki indicates higher potency. Until replicated human cloned receptor data is available for all of these drugs no precise comparisons will be possible. It is important to note that this data can only give a guide as to potency. The results from Codd are presented separately because this is the only paper that has screened a group of drugs using the same methodology.

*venlafaxine, may have other serotonergic action as well as serotonin reuptake inhibitor potency: this might be related to the anomalous toxicity, it is weaker as an SRI than at any other antidepressant that is capable of inducing ST. There is also evidence that tramadol, like the close structural homologue venlafaxine, may have some other serotonergic action as well as weak serotonin reuptake inhibitor potency (11-14).

** More recent data (15).

*** More recent SRI data re tramadol (16,17). As a single 100 mg dose of tramadol produces peak plasma concentrations of 300 ng/ml, equivalent to about 1 µM conc. that suggests tramadol might have some effect on serotonin transporter (SERT).

Codd’s paper

Codd’s paper is the only one (that I know of) that examines the serotonin reuptake inhibitor potency of a range of narcotic analgesics. It indicates that Phenanthrene opioids (with an oxygen bridge between C4 and C5), such as morphine (group I, see original paper), did not act as serotonin reuptake inhibitors at all. Phenanthrene and nonphenanthrene opioids such as levorphanol and levomethorphan (group II), and d-propoxyphene and methadone (group III) did act as serotonin reuptake inhibitors. I have made enquiries of many researchers, including Toll and Codd but no-one seems to know of any other data on the serotonin reuptake inhibitor potency of narcotic analgesics (12,18), except recent additions above i.e.(16,17). See also PDSP for possible recent additions of data

References to cases involving tramadol: (19-29)

References to cases involving pethidine: (3, 6, 30-50)

References to cases involving dextromethorphan and propoxyphene: (5, 8, 51, 52)

References to cases involving fentanyl: (53-60)

In my opinion the Noble case report of a fatality was probably serotonin toxicity; so as predicted from Codd’s data there is a question mark over fentanyl: however its potency as an serotonin reuptake inhibitor is low, so as expected (and like pethidine) it has been reported to have been used with impunity. The more recent Roy case is another peculiar example of reactions with venlafaxine; this is a drug that seems to produce unexpected interactions with a variety of other classes of drugs not adequately explained by its serotonin reuptake inhibitor action.

Uncertain: (61).

In summary: Morphine, codeine, oxycodone, buprenorphine do not precipitate serotonin toxicity with monoamine oxidase inhibitors. Pethidine (meperidine), tramadol, methadone and fentanyl, propoxyphene, dextromethorphan, are borderline. They are weakly serotonergic and some may precipitate serotonin toxicity which must be assumed to be dose dependant.

My conclusion in the BJA paper was: ‘In summary, morphine, codeine, oxycodone and buprenorphine are now known not to be SRIs and they do not precipitate serotonin toxicity with MAOIs. Pethidine, tramadol, dextromethorphan and methadone definitely are weak SRIs (see Table 2), and may infrequently precipitate dose-dependant serotonin toxicity (when administered in conjunction with any type of MAOI), but perhaps only in large doses or susceptible individuals. Our ability to estimate the risk with particular drugs with any precision is compromised by a lack of systematically recorded data on the clinical toxicity of these drugs, and also by the lack of pharmacological data concerning their precise potency, as SRIs or releasers, using the latest assay techniques. For some drugs (propoxyphene, pentazocine, fentanyl, remifentanil and congeners) there is no SRI affinity data at all, and we have to rely on interpolations and the presence or absence of clinical reports (Table 2). It is to be hoped that increasing understanding and awareness of this situation will stimulate further research that will answer these remaining questions.

However, the clinical situation and the risks are now more clearly defined and understood and the information herein will enable many clinicians to be more confident when making decisions about patient management. Choices involving the known serotonergic opioids can now be made in particular clinical situations by balancing the advantages and disadvantages that there may be for individual patients with respect to particular drugs. The level of risk with known serotonergic opioids is probably low, but its unpredictable and serious nature makes it difficult to form judgments. All other factors being equal, it would seem prudent to use the drugs known not to be SRIs where possible. These judgments may be tempered by the knowledge that any reaction is dose-dependent and that we are now confident that reactions can be successfully treated, if severe, with 5-HT2A antagonists.’


1. Gillman, PK, Monoamine oxidase inhibitors, opioid analgesics and serotonin toxicity. British Journal of Anaesthesia, 2005. 95: p. 434-441.

2. Churchill-Davidson, HC, Anesthesia and Monoamine-Oxidase Inhibitors. Br Med J, 1965. 5433: p. 520.

3. Evans-Prosser, A, The use of pethidine and morphine in the presence of monoamine oxidase inhibitors. British Journal of Anaesthesia, 1968. 40: p. 279.

4. Peeters-Asdourian, C, [Fatal drug interactions]. Rev Med Brux, 1986. 7(9): p. 570.

5. Shamsie, SJ and Barriga, C, The hazards of monoamine oxidase inhibitors in disturbed adolescents. Canadian Medical Association Journal, 1971. 104: p. 715.

6. Pollock, RA and Watson, RL, Malignant hyperthermia associated with hypocalcaemia. Anesthesiology, 1971. 34: p. 188.

7. Gong, SN and Rogers, KJ, Role of brain monoamines in the fatal hyperthermia induced by pethidine or imipramine in rabbits pretreated with pargyline. Br J Pharmacol, 1971. 42(4): p. 646P.

8. Rivers, N and Horner, B, Possible lethal interaction between Nardil and dextromethorphan. Canadian Medical Association Journal, 1970. 103: p. 85.

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

10. Gillman, PK, Serotonin toxicity, serotonin syndrome: 2006 update, overview and analysis., 2006: p. Epub 1-125.

11. Giusti, P, et al., Effect of acute and chronic tramadol on [3H]-5-HT uptake in rat cortical synaptosomes. British Journal of Pharmacology, 1997. 122(2): p. 302-6.

12. Codd, EE, et al., Serotonin and norepinephrine uptake inhibiting activity of centrally acting analgesics: structural determinants and role in antinociception. Journal of Pharmacology and Experimental Therapeutics, 1995. 274(3): p. 1263-70.

13. Bamigbade, TA, et al., Actions of tramadol, its enantiomers and principal metabolite, O-desmethyltramadol, on serotonin (5-HT) efflux and uptake in the rat dorsal raphe nucleus. British Journal of Anaesthesia, 1997. 79(3): p. 352-6.

14. Frink, MC, et al., Influence of tramadol on neurotransmitter systems of the rat brain. Arzneimittelforschung, 1996. 46(11): p. 1029-36.

15. Lomenzo, SA, et al., Synthesis and Biological Evaluation of Meperidine Analogues at Monoamine Transporters. J Med Chem, 2005. 48(5): p. 1336-1343.

16. Barann, M, et al., Effects of tramadol and O-demethyl-tramadol on human 5-HT reuptake carriers and human 5-HT3A receptors: a possible mechanism for tramadol-induced early emesis. Eur J Pharmacol, 2006. 531(1-3): p. 54-8.

17. Rothman, RB and Baumann, MH, Therapeutic Potential of Monoamine Transporter Substrates. Current Topics in Medicinal Chemistry, 2006. 6: p. 1845-1859.

18. Toll, L, et al., Standard binding and functional assays related to medications development division testing for potential cocaine and opiate narcotic treatment medications. NIDA Research Monograph, 1998. 178: p. 440-66.

19. Lange-Asschenfeldt, C, et al., Serotonin syndrome as a result of fluoxetine in a patient with tramadol abuse: plasma level-correlated symptomatology. Journal of Clinical Psychopharmacology, 2002. 22(4): p. 440-1.

20. Reus, VI and Rawitscher, L, Possible interaction of tramadol and antidepressants. American Journal of Psychiatry, 2000. 157(5): p. 839.

21. Kesavan, S and Sobala, GM, Serotonin syndrome with fluoxetine plus tramadol. Journal of the Royal Society of Medicine, 1999. 92(9): p. 474-5.

22. de Larquier, A, et al., [Serotoninergic syndrome after combining tramadol and iproniazid]. Therapie, 1999. 54(6): p. 767-8.

23. Calvisi, V and Ansseau, M, Clinical case of the month. Mental confusion due to the administration of tramadol in a patient treated with MAOI. Rev Med Liege, 1999. 54(12): p. 912-3.

24. Lantz, MS, Buchalter, EN, and Giambanco, V, Serotonin syndrome following the administration of tramadol with paroxetine. International Journal of Geriatric Psychiatry, 1998. 13(5): p. 343-345.

25. Mason, BJ and Blackburn, KH, Possible serotonin syndrome associated with tramadol and sertraline coadministration. Annals of Pharmacotherapy, 1997. 31: p. 175-177.

26. Egberts, ACG, ter Borgh, J, and Brodie-Meijer, CCE, Serotonin syndrome attributed to tramadol addition to paroxetine therapy. International Clinical Psychopharmacology, 1997. 12(3): p. 181-182.

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

28. Raffa, RB, Reimann, W, and Shank, RP, Opiod and nonopiod components independently contribute to the mechanism of action of tramadol, an atypical opiod analgesic. Journal of Pharmacology and Experimental Therapeutics, 1992. 260: p. 275-285.

29. Driessen, B and Reimann, W, Interaction of the central analgesic tramadol, with the uptake and release of 5-hydroxytryptamine in the rat brain in vitro. British Journal of Pharmacology, 1992. 105: p. 147-151.

30. Mitchell, RS, Fatal toxic encephalitis occurring during iproniazid therapy in pulmonary tuberculosis. Annals of Internal Medicine, 1955. 42: p. 417-424.

31. Papp, C and Benaim, S, Toxic effects of iproniazid in a patient with angina. British Medical Journal, 1958. 2: p. 1070-1072.

32. Palmer, H, Potentiation of pethidine. British Medical Journal, 1960. 2: p. 944.

33. Shee, JC, Dangerous potentiation of pethidine by iproniazid and its treatment. British Medical Journal, 1960. 2: p. 507-509.

34. Brownlee, G and Williams, GW, Potentiation of amphetamine and pethidine by monoamine oxidase inhibitors. Lancet, 1961. 1([letter]): p. 669.

35. Clement, AJ and Benazon, D, Reactions to other drugs in patients taking monoamine oxidase inhibitors. Lancet, 1962. 2([letter]): p. 197-198.

36. Cocks, DP and Passmore-Rowe, AH, Dangers of monoamine oxidase inhibitors. British Medical Journal, 1962. 2([letter]): p. 1545-1546.

37. Denton, PH, Borrelli, VM, and Edwards, NV, Dangers of monoamine oxidase inhibitors. British Medical Journal, 1962. 2: p. 1752-1753.

38. London, DR and Milne, MD, Dangers of monoamine oxidase inhibitors. British Medical Journal, 1962. 2: p. 1752.

39. Pells Cocks, D and Passmore-Rowe, A, Dangers of monoamine oxidase inhibitors. British Medical Journal, 1962. 2: p. 1545-6.

40. Reid, NCRW and Jones, D, Pethidine and phenelzine. British Medical Journal, 1962. 1: p. 408.

41. Taylor, DC, Alarming reaction to pethidine in patients on phenelzine. Lancet, 1962. 2: p. 401-402.

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

43. Vigran, IM, Dangerous potentiation of meperidine hydrochloride by pargyline hydrochloride. Journal of the American Medical Association, 1964. 187([letter]): p. 953-954.

44. Jounela, AJ, Mattila, MJ, and Knoll, J, Interaction of selective inhibitors of monoamine oxidase with pethidine in the rabbit. Biochemical Pharmacology, 1977. 26: p. 806-808.

45. Leander, JD, Batten, J, and Hargis, GW, Pethidine interaction with clorgyline, pargyline, or 5- hydroxytryptophan: lack of enhanced pethidine lethality or hyperpyrexia in mice. J Pharm Pharmacol, 1978. 30(6): p. 396-8.

46. Rawlins, MD, Drug interactions and anaesthesia. Br J Anaesth, 1978. 50(7): p. 689-93.

47. Meyer, D and Halfin, V, Toxicity secondary to meperidine in patients on monoamine oxidase inhibitors: a case report and critical review. Journal of Clinical Psychopharmacology, 1981. 1: p. 319-321.

48. Starr, C, Interaction between pethidine and selegiline. Lancet, 1991. 337(8740): p. 554.

49. Zornberg, GL, Bodkin, JA, and Cohen, BM, Severe adverse interaction between pethidine and selegiline. Lancet, 1991. 337: p. 246.

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

51. Garbutt, JC, Potentiation of propoxyphene by phenelzine. American Journal of Psychiatry, 1987. 144(2): p. 251-2.

52. Zornberg, GL, Adverse interaction between propoxyphene and phenelzine. American Journal of Psychiatry, 1993. 150(8): p. 1269-1270.

53. Noble, WH and Baker, A, MAO inhibitors and coronary artery surgery: a patient death. Canadian Journal of Anaesthesia, 1992. 39(10): p. 1061-1066.

54. Insler, SR, et al., Cardiac surgery in a patient taking monoamine oxidase inhibitors: an adverse fentanyl reaction. Anesth Analg, 1994. 78(3): p. 593-7.

55. Wells, DG and Bjorksten, AR, Monoamine oxidase inhibitors revisited. Can J Anaesth, 1989. 36(1): p. 64-74.

56. Fobe, F, et al., Heart-transplant and mono-amine oxidase inhibitors. Acta Anaesthesiol Belg, 1989. 40(2): p. 131-8.

57. Stack, CG, Rogers, P, and Linter, SPK, Monoamine oxidase inhibitors and anaesthesia. British Journal of Anaesthesia, 1988. 60: p. 222-227.

58. el-Ganzouri, AR, et al., Monoamine oxidase inhibitors: should they be discontinued preoperatively? Anesth Analg, 1985. 64(6): p. 592-6.

59. Michaels, I, et al., Anesthesia for cardiac surgery in patients receiving monoamine oxidase inhibitors. Anesth Analg, 1984. 63(11): p. 1041-4.

60. Roy, S and Fortier, LP, Fentanyl-induced rigidity during emergence from general anesthesia potentiated by venlafexine. Can J Anaesth, 2003. 50(1): p. 32-5.

61. Brown, DD and Waldron, DH, An unusual reaction to tranylcypromine. Practitioner, 1962. 189: p. 83-86.