There is substantial doubt that most new so-called dual action drugs (SNRIs, e.g. venlafaxine, duloxetine etc.) actually have useful, never mind optimal, dual action properties

·      The old TCA, clomipramine, remains the benchmark dual action drug

·      Safe and flexible dual action combinations (of an SSRI with an NRI, e.g. sertraline + nortriptyline) probably allow better efficacy and less side effects

·      The widespread reluctance of psychiatrists to mix different antidepressant drugs reflects their lack of confidence in their pharmacological knowledge: there are various perfectly safe options

·      Mixing several drugs with different actions is routine practice in the treatment of hypertension, diabetes, heart disease and other conditions. The objection that ‘poly-pharmacy’ is, prima facie, bad is both illogical and mistaken.

General Summary

Claims concerning dual serotonin and noradrenaline reuptake inhibitor (SNRI) action for antidepressant drugs have been around since the 1990s. However, guess what, the dual action SNRI tag has had more to do with marketing than with pharmacological reality [just like “atypical anti-psychotics”].

There are two separate and good reasons for thinking that dual action is a good thing. Firstly, some of the best evidence that exists for superior efficacy of one drug over others is for clomipramine. Not only does some controlled trial data support that assertion, but also many experienced psycho-pharmacologist find that clomipramine is more effective than any of the other tricyclics (1). Clomipramine is the drug with the best evidence for clinically meaningful dual action. This is an important point and is investigated in detail in the paper I published in the British Journal of Pharmacology (2). The full text of this paper is available free (it has good tables of receptor affinity data)

I am using the term dual action in relation to re-uptake inhibition [therefore I am not discussing drugs like mirtazapine, which are neither reuptake inhibitors, nor IMO effective ADs]; but it is worth noting here that triple reuptake inhibitors (SDNRIs, that is dopamine, serotonin and noradrenaline reuptake inhibitors) are under development (3-7). It is also worth noting, at this point, that dopamine has not received as much attention as it might have over this last couple of decades. I have commented on this several times previously (8, 9), especially in the context of tranylcypromine [Parnate] being one of the only currently available drugs that does substantially increase dopamine levels.

Dual Action and Receptor Affinity Data

One of the things that my research has clarified is the relationship between the receptor affinities measured in experimental test-tube (in vitro) assays and the actual effect of drugs in humans. This is a vital consideration for two particular reasons:

1) Because the potency of many of these drugs varies by several orders of magnitude (see table Receptor Affinity data below)

 2) There is not a simple or direct relationship between receptor affinity and effect

So, one key question is how potent do these drugs have to be before they have significant effects?

I have been using treatments that produce dual SNRI actions for many years and have no doubt that they are of superior effectiveness compared to other currently available “SNRIs”. The evidence and opinion supporting that position is getting better and stronger gradually, see other PUN notes and references (13-17).

Dual action strategies possible in the past have been Clomipramine (the superiority of which was strongly supported by the DUAG studies (18-20)), and since about 1990, combinations like sertraline + nortriptyline.

Fixed Relative Potency

There is one major disadvantage of using a single drug that supposedly has dual action on both serotonin and noradrenaline reuptake, i.e. a so-called SNRI: it is that the ratio of the potency for effecting these two pathways is fixed, and not ideal because most such drugs affect one pathway much more than the other. A useful way of picturing this is to think of a four-wheel drive vehicle that puts 90% of the power to the ground through one set of wheels, and only 10% through the other; that is fixed and cannot be varied to suit different conditions. That is clearly a dumb idea and a sub-standard system. No-one would buy a four-wheel-drive vehicle that did not have intelligent control systems that varied the power to each wheel, depending on the traction available. This analogy makes it clear that using two different drugs (to achieve SNRI) allows the dose of each one to be varied according to the characteristics and severity of the illness, in a way which is definitely more flexible, and probably more effective, than a single drug with a fixed ratio of effects.

It is logical to assume that the potencies for each pathway need to be similar, or else there will be an excessive effect on one pathway and an inadequate effect on the other. How similar is similar? Until evidence to the contrary is available it is logical to assume that a ratio close to 1:1 is optimal. Reference to the table indicates that the new drugs do not come near this figure. But, on paper, some of the old TCAs do. But that is only part of the story.

There is substantive evidence that clinical effectiveness, at least for serotonin pathways, requires a drug with low single figure SRI potency, i.e. a Ki of less than 10 nM (21). If the same applies to noradrenaline reuptake, which on present evidence would have to be considered a reasonable proposition, then it is evident that the newer drugs may not be sufficiently potent to produce optimal effects (see table).

 If it is not possible to obtain therapeutic levels of clomipramine without unacceptable side-effects, then this suggests an alternative strategy for achieving effective dual action is use two different drugs at the same time (e.g. sertraline + nortriptyline). This strategy is simple and safe (I have used it in many hundreds of cases) providing a few basics about drug interactions are taken into account [see CYP450 notes].

Tyramine Pressor Response 

In the past various methodologies have been applied to attempt to find an answer to this question of what is the minimum required potency for optimum efficacy. It is possible to test noradrenaline reuptake inhibitors [NRIs] for their ability to modify the tyramine pressor response. The amine, tyramine, when ingested in large quantities, raises the blood pressure. The usual methodology is to give increasing doses, either by mouth, or intravenously, until the systolic blood pressure increases by 30 mm of mercury. The dose required to achieve this is referred to as the TYR30. The mechanism of this reaction is via tyramine releasing the body's own noradrenaline from the sympathetic nerve endings. In order for it to do this it must first be taken up into the pre-synaptic nerve ending. This occurs via the noradrenaline transporter [reuptake inhibition]. Because tricyclic antidepressants inhibit the noradrenaline transporter they prevent this process and thereby prevent tyramine from precipitating noradrenaline release and blood pressure increase. There is a great deal of work, over several decades, relating the capacity of various drugs to do this. It is now possible to correlate that with the latest human cloned receptor data. 

Note that the table below contains a range of values for the receptor affinity measurement. This is because it is not a precise process and the in vitro (test tube) assays that produce these measurements are subject to various experimental variations. Therefore when values are compared between different laboratories employing essentially the same technique there is an approximately tenfold variation in the affinity data produced. This needs to be taken account of when comparisons are made and it is frequently not valid to try to make precise comparisons. The data has to be interpreted in the light of other data and the physico-chemical properties of the drug so that a best guess can be made as to what the data indicates. When reading different papers care needs to be taken to note whether the data relating to human cloned receptor assays is being presented, or some other technique. HCR data is probably the benchmark and all data here is HCR data unless otherwise indicated.

Decades of experience with tricyclic antidepressants indicates that those with weak reuptake capacity are ineffective as antidepressants, viz. trimipramine, doxepin etc. However, it may be noted that this hypothesis has never been rigorously tested using up-to-date methodology and rating scales. I think that should be done. Assuming that resulted in a demonstration that weaker noradrenaline reuptake potency was less effective than strong reuptake potency that would in itself be strong evidence for a role for noradrenaline in promoting antidepressant efficacy. I emphasize this point particularly because the selective noradrenaline reuptake inhibitors (NRIs), like reboxetine, do not appear to be effective antidepressants (10). That definitely accords with my experience. But see also (11). I never used reboxetine by itself is an antidepressant because I regarded it as ineffective. I think it probably did augment the effect of sertraline in combination, but would note that my much preferred drug was nortriptyline.

Note also that atomoxetine (a selective NRI like reboxetine) was strategically marketed by Lilly as an ADH drug because it “failed” in AD trials.

Indeed it seems that development of NRIs has ceased, partly because of the unpopularity and perceived low efficacy of reboxetine (despite the usual meta-analyses that find apparent equal effectiveness with SSRIs(11)). Various methodologies have been applied to attempt to answer the question of how potent do these drugs have to be (in terms of receptor affinities) before they have significant antidepressant actions.

Table Human cloned receptor (HCR) data relating to ‘dual action’

Reuptake inhibitor affinities Ki nM (modified PKG March 2010) 







SSRIs (for comparison)




~ 1:1000



19 - 102

2.8 - 36

















20 - 142

1.3 - 20




0.63 – 8.6

22 -180













No HCR data

No HCR data








Table Legend TYR30: tyramine pressor response +++ potent effect (i.e. strong NRI) ++ moderate + weak 0 no significant effect. NA/5-HT: ratio of potency NA vs 5-HT

What affinity at the noradrenaline transporter (NAT) (i.e. potency as an noradrenaline reuptake inhibitor (NRI)) constitutes a clinically useful effect on the NAT? The pressor response to tyramine, the ‘TYR30’ test, provides an in vivo index of peripheral NRI potency. The first direct comparison between the posited SNRI, venlafaxine, and a TCA, desipramine has only recently been done (12). If academics were more on the ball it would have been done long ago.

Noradrenaline reuptake inhibitors (NRIs) inhibit tyramine uptake and thus lessen or abolish the response, potent NRIs (reboxetine, desipramine and nortriptyline block this response completely (13, 14)).

The effect of tricyclic antidepressants (TCAs) on the TYR30 is proportional to their relative NRI potency, only affinities of nortriptyline, or greater, are associated with marked attenuation (see table). Duloxetine and venlafaxine have little or no effect on TYR30 even at supra-therapeutic doses. There are no known data for sibutramine or milnacipran. Duloxetine, at 60 mg daily attenuates TYR30 very slightly (15), even at 240 mg per day it is still weak (16). Chalon et al found no effect at 80 mg (17). Venlafaxine only marginally lessens the pressor response at the usual maximum dose of 375 mg daily (18, 19). A complete set of data comparing all the TCAs etc under the same conditions would be revealing, but the present evidence indicates that drugs with an NRI potency less than nortriptyline are sub-optimally effective. Dothiepin (dosulepine), doxepin, mianserin and mirtazapine (6-aza-mianserin) are weak NRIs and have no effect on TYR30 (20-22). 

There is no special evidence suggesting the cerebral drug levels are much different between drugs (although that is definitely possible, see e.g. (23)), so the provisional supposition that the central effects mirror the TYR30 data seems reasonable.


In my opinion the two telling bits of data are the SRI potency difference between imipramine and clomipramine and the TYR30 difference between nortriptyline and desipramine when compared to duloxetine and venlafaxine. It is well established that clomipramine is the only tricyclic antidepressant (TCA) effective for obsessive compulsive disorder (and cataplexy) and that imipramine and amitriptyline are ineffective. It is unlikely that this is accounted for by anything other than their differing SRI potency, e.g. 5-HT2A affinity (see also (24)). Furthermore, we also know that there is a congruent difference in the ability of amitriptyline / imipramine / clomipramine to precipitate serotonin toxicity. The above receptor affinity data clearly indicates that (at least for TCAs) potencies of less than one (i.e. Ki < 1 nM) are required for clinical effectiveness. It is therefore reasonable to suggest that in designing a dual action strategy it would be ideal to aim for those kinds of potencies in both pathways. Neither venlafaxine nor duloxetine come anywhere near that. Clomipramine definitely does, as do combinations like sertraline + nortriptyline, or sertraline + reboxetine.

But who is going to advertise or promote such a combination?

It is appropriate to continue to regard clomipramine as the gold standard antidepressant reference compound because the newer ‘dual action’ drugs just do not measure up. But combinations may achieve the same result with less side effects. In my opinion, and extensive experience (well in excess of 1,000 patients), sertraline + nortriptyline probably works a bit better than sertraline + reboxetine (a highly selective NRI), so maybe 5-HT2A receptors, or some other property of TCAs, does make a difference.

It is possible that adding an atypical neuroleptic for the treatment of refractory depression may be emulating the 5-HT2A antagonism that is missing with selective serotonin reuptake inhibitors (SSRIs). Which brings us back ‘full circle’ to clomipramine!

As Alphonse Karr neatly reiterated a couple of centuries ago ‘Plus ça change, plus c'est la même chose’ (25).

Another little diversion is appropriate. Long ago van Praag suggested that depression could be better considered as a group of psycho-pathological changes, each represented by different neuro-pharmacological alterations and to different aspects of the syndrome. For instance, he specifically suggested that psychomotor retardation was related to dopamine (26-28). I must say, it has always seemed to me that his ideas obviously have considerable merit, and I have been surprised how little attention they appear to have attracted. They certainly moulded my thinking on multi-aminergic strategies (i.e. SNRI & DA strategies) decades ago (8).

Indeed, the sensible initial assumption concerning the underlying abnormalities in depressive illness would have to be that it is not a unitary entity, and is not represented by a single discrete abnormality. It is most unlikely to be effectively treated with a drug that has a specific action only one pathway (like SSRIs).

Comparisons and meta-analysis concerning dual action drugs have major problems. There is doubt that most of the drugs considered do in fact have dual action; the severity of illness of the patients considered may not be great enough, e.g. to show differences, either between drugs, or between drugs and placebo (29). Most meta-analysis is flawed anyway (garbage in, garbage out). It is clear from the fuss concerning the ineffectiveness of the SSRIs, that clear differences between different drugs and between drugs and placebo are only evident when one considers severely depressed patients (so that rules out almost all drug company sponsored trials).

A little history is relevant as always. The first dual action antidepressant to be trialed, at least that I have found information about, was in the early 1980s. A drug that is still with us, just, in a different guise. It is sibutramine (Reductil etc, google it), used as a diet drug. The initial depression trials were a failure and were lost. [As of mid 2011 now gone again!). Most psycho-pharmacologists I have spoken to do not know that. It re-emerged years later, after company take-overs etc in its diet guise. It is actually a ‘prodrug’. The effect comes from its metabolites, so if some subjects have different pathways, because of variations in activity of CYP450 [see PUN note on sibutramine].






1.               Fawcett, J, Why aren't MAOIs used more often? J Clin Psychiatry, 2009. 70(1): p. 139-40.

2.               Gillman, PK, Tricyclic antidepressant pharmacology and therapeutic drug interactions updated. Br J Pharmacol, 2007. 151(6): p. 737-48.

3.               Guiard, BP, El Mansari, M, and Blier, P, Prospect of a Dopamine Contribution in the Next Generation of Antidepressant Drugs: The Triple Reuptake Inhibitors. Curr Drug Targets, 2009.

4.               Liang, Y, Shaw, AM, Boules, M, Briody, S, et al., Antidepressant-like pharmacological profile of a novel triple reuptake inhibitor, (1S,2S)-3-(methylamino)-2-(naphthalen-2-yl)-1-phenylpropan-1-ol (PRC200-SS). J Pharmacol Exp Ther, 2008. 327(2): p. 573-83.


5.               Shaw, AM, Boules, M, Zhang, Y, Williams, K, et al., Antidepressant-like effects of novel triple reuptake inhibitors, PRC025 and PRC050. Eur J Pharmacol, 2007. 555(1): p. 30-6.

6.               Golembiowska, K, Kowalska, M, and Bymaster, FP, Effects of the triple reuptake inhibitor amitifadine on extracellular levels of monoamines in rat brain regions and on locomotor activity. Synapse, 2012. 66(5): p. 435-44.

7.               Zheng, YY, Guo, L, Zhen, XC, and Li, JQ, Synthesis and antidepressant activity of arylalkanol-piperidine derivatives as triple reuptake inhibitors. Eur J Med Chem, 2012.

8.               Gillman, PK, Plus ça change, plus c'est la même chose. Pharmabulletin, 1994. 18(1): p. 10-11.

9.               Gillman, PK, Psychotic depression and ‘multi-aminergic’ treatment strategies. Br J Psychiatry, 2006: p.

10.             Eyding, D, Lelgemann, M, Grouven, U, Harter, M, et al., Reboxetine for acute treatment of major depression: systematic review and meta-analysis of published and unpublished placebo and selective serotonin reuptake inhibitor controlled trials. BMJ, 2010. 341: p. c4737.

11.             Papakostas, GI, Nelson, JC, Kasper, S, and Moller, HJ, A meta-analysis of clinical trials comparing reboxetine, a norepinephrine reuptake inhibitor, with selective serotonin reuptake inhibitors for the treatment of major depressive disorder. Eur. Neuropsychopharmacol., 2008. 18(2): p. 122-7.

12.             Blier, P, Saint-Andre, E, Hebert, C, de Montigny, C, et al., Effects of different doses of venlafaxine on serotonin and norepinephrine reuptake in healthy volunteers. Int J Neuropharmacol, 2007. 10(1): p. 41-50.

13.             Dostert, P, Castelli, MG, Cicioni, P, and Strolin Benedetti, M, Reboxetine prevents the tranylcypromine-induced increase in tyramine levels in rat heart. J Neural Transm, 1994. 41: p. 149-53.

14.             Doggrell, SA and Woodruff, GN, Effects of antidepressant drugs on noradrenaline accumulation and contractile responses in the rat anococcygeus muscle. Br J Pharmacol, 1977. 59(3): p. 403-9.

15.             Turcotte, JE, Debonnel, G, de Montigny, C, Hebert, C, et al., Assessment of the serotonin and norepinephrine reuptake blocking properties of duloxetine in healthy subjects. Neuropsychopharmacology, 2001. 24(5): p. 511-21.

16.             Vincent, S, Bieck, PR, Garland, EM, Loghin, C, et al., Clinical assessment of norepinephrine transporter blockade through biochemical and pharmacological profiles. Circulation, 2004. 109(25): p. 3202-7.

17.             Chalon, SA, Granier, LA, Vandenhende, FR, Bieck, PR, et al., Duloxetine increases serotonin and norepinephrine availability in healthy subjects: a double-blind, controlled study. Neuropsychopharmacology, 2003. 28(9): p. 1685-93.

18.             Harvey, AT, Rudolph, RL, and Preskorn, SH, Evidence of the dual mechanisms of action of venlafaxine. Arch. Gen. Psychiatry, 2000. 57(5): p. 503-9.

19.             Debonnel, G, Saint-Andre, E, Hebert, C, de Montigny, C, et al., Differential physiological effects of a low dose and high doses of venlafaxine in major depression. Int J Neuropharmacol, 2007. 10(1): p. 51-61.

20.             Coppen, A, Ghose, K, Swade, C, and Wood, K, Effect of mianserin hydrochloride on peripheral uptake mechanisms for noradrenaline and 5-hydroxytryptamine in man. Brit J of Clin Pharmacol, 1978. 5 Suppl 1: p. 13S-17S.

21.             Pare, CM, Kline, N, Hallstrom, C, and Cooper, TB, Will amitriptyline prevent the "cheese" reaction of monoamine-oxidase inhibitors? Lancet, 1982. 2(8291): p. 183-6.

22.             Ghose, K, Gifford, LA, Turner, P, and Leighton, M, Studies of the interaction of desmethylimipramine with tyramine in man after a single oral dose, and its correlation with plasma concentration. Br J Clin Pharmacol, 1976. 3(2): p. 334-7.

23.             Schomburg, R, Remane, D, Fassbender, K, Maurer, HH, et al., Doxepin concentrations in plasma and cerebrospinal fluid. J Neural Transm, 2011. 118(4): p. 641-5.

24.             Gillman, PK, A review of serotonin toxicity data: implications for the mechanisms of antidepressant drug action. Biol Psychiatry, 2006. 59(11): p. 1046-51.

25.             Karr, A, 'Les Guelphes'. 1849.

26.             van Praag, H, Asnis, GM, Kahn, RS, Brown, SL, et al., Monoamines and Abnormal Behaviour.  A Multi-Aminergic Perspective. Br J Psychiatry, 1990. 157: p. 723-734.

27.             van Praag, HM, Diagnosis, the rate-limiting factor of biological depression research. Neuropsychobiology, 1993. 28(4): p. 197-206.

28.             van Praag, HM, The diagnosis of depression in disorder. Aust NZ J Psychiatry, 1998. 32(6): p. 767-72; discussion 773-7.

29.             Carroll, BJ, Bringing back melancholia. Bipolar Disord, 2012. 14(1): p. 1-5.