Antidepressant basics

Depressive illness responsive to drugs cannot be reliably distinguished from other depressive states except by its response to anti-depressant drugs. There are no diagnostic tests.

Many anti-depressant drugs that have been approved for use have subsequently failed to prove themselves to possess useful antidepressant efficacy in general clinical use (e.g. mianserin, trazodone, moclobemide). Others are little used and regarded as weak antidepressants by psycho-pharmacologists, such recent examples are reboxetine, nefazodone and mirtazapine.

Then there are other drugs that have efficacy, but clearly have significant problems (e.g. side effects) or rare but severe adverse effects (neutropenia, liver toxicity, auto-immune reactions).

A word of caution to all those doctors who want to look up-to-date by using the latest shiny new drug — it may take some years before the true benefits and risks emerge. Almost all new drugs appear to be better for a while, but by consenting to take them, or by giving them, you are in ‘guinea pig’ territory. This is especially true since drugs are not properly tested or assessed before marketing. If you think that is overstating that case, just read this for a fuller discussion of this important issue, here.

It has taken twenty years for the longer-term side effects of the SSRI to emerge. One is the increased risk of bleeding in various situations: there is an increased risk of death from, inter alia, GI bleeding, post-partum haemorrhage and intracranial Hemorrhage (1-3). The sexual SEs are prominent and have always been under-emphasised; that is where bupropion may be useful as mono-therapy in milder illness.

Agomelatine is already under a shadow re liver toxicity (4, 5), and everyone seems to have forgotten about blood dyscrasias with both mianserin and mirtazapine, despite a warning in 2007 noting several deaths (by the Australian authority) stating all patients should be warned (I still recall the first case ‘we’ reported in 1976).

Whilst one does not wish to overemphasise these risks, they do highlight the fact that most of these newer drugs are not quite as innocuous as some people would have one believe, and should be administered after due consideration for the severity of the illness and potential disadvantages of drug treatment.

Remember that it is axiomatic that doctors must keep up-to-date with the benefits and risks of the drugs they prescribe or they may be found liable, or even negligent, when (not if, but when) patients come to harm.

Older drugs

Such knowledge helps us to balance the relative merits of many of the older drugs which it has become fashionable to decry as being less satisfactory and less safe. It is worth remembering that the supposed advantages of the SSRIs over the older tricyclic antidepressants were always agreed to be marginal, and this more recent recognition of a greater level of medium to long-term side-effects from SSRIs, and other newer drugs, suggests that the older drugs have less disadvantages than many have imagined.

For major (‘endogenous’) depressive illness selective serotonin reuptake inhibitors are, it is generally agreed, probably less effective than TCAs like amitriptyline. It is my opinion that the SSRIs are of low effectiveness. The indications are that the more severe the illness the less effectively SSRIs, and many other newer drugs, work. Clinical experience supports that more strongly.

Sertraline is a logical first choice (see here), especially in less severe cases and where first line treatment with a tricyclic antidepressant is not appropriate. Most SSRIs have problematic cytochrome P450 interactions; these are fluoxetine (2D6 and 3A4), paroxetine (2D6) and fluvoxamine (1A2 and 2C9 /19). I consider it is very hard to justify their use as first line in primary care, or even in a specialist setting, especially since many specialists do not understand drug interactions.

If side effects have resulted in an inadequate therapeutic trial then a second serotonin reuptake inhibitor may be tried, citalopram (or if you must its isomer, more expensive, escitalopram*) is the logical next choice. There is no sense or logic in trying a second or third SSRI. Anyone who gives three or more SSRIs should be smacked around the ear. Likewise, the evidence strongly supports the notion that giving doses above the usual recommended dose is usually a waste of money and pointless. Better to try a different strategy, which, especially in primary care, may be non-drug intervention.

*As of mid 2016 I think this is no longer true as escitalopram is almost the same price: however it does not alter the fact that a misplaced perception of the difference from citalopram has cost healthcare budgets a great deal (6).

Nortriptyline (one of the ‘old’ TCAs) is the best ‘non-SSRI’ choice: it has good evidence of efficacy. It is a ‘pharmacologists’ drug because it possesses most of the attributes that pharmacologists like, e.g. a good side effect profile, near-linear pharmacokinetics, fairly selective NRI (see (7)), low toxicity etc. (also blood levels are easily done).

Since interactions between sertraline or citalopram and nortriptyline are minimal and straightforward to allow for and manage I always used to overlap them and start the nortriptyline before attempting to cease the sertraline. This ‘SNRI’ strategy often produces such a good result that many patients do not wish to cease either drug, because, as they tried to reduce either one, they felt worse. Hence continuation of this SNRI or dual action approach was the rule for me.

More effective 1st line antidepressant drugs; various lines of evidence suggest that amitriptyline, clomipramine and sertraline are more effective than the ‘average’, and one can add NTP & IMI.

Less effective and minimally effective drugs include; moclobemide, selegiline, trimipramine, doxepin, dothiepin, trazodone, nefazodone and mianserin, mirtazapine, paroxetine, fluvoxamine.


After cogitating on what to say about these drugs for a while I shall use the phrase that kept echoing in my mind — ‘like tits on a bull’.

What else can one say. As I argue elsewhere there is some doubt about whether most of them do have dual activity, and even if they do it is in a fixed ratio, which despite what some commentators seem to think, is obviously not ideal. The best-selling drug, still, is venlafaxine and there is a more detailed discussion about it here. When you can mix and match to suit yourself with sertraline and nortriptyline (or reboxetine) why would you bother with SNRIs? The main reason so many doctors have difficulty with this is poor pharmacological knowledge.


Although the old monoamine oxidase inhibitors (MAOIs), tranylcypromine and phenelzine, are no longer advertised (how many young doctors do not know they even exist?) it is well to remember that they are not only very effective, but are the treatment of choice for many patients. See info on MAOIs here

Shiny new drug X

Then, it is asked, what about shiny new drug X: agomelatine, vortioxetine etc. Well, that is partly answered above, and there simply is not enough quality replicated unbiased evidence to say anything useful. Remember, most new drugs introduced over the whole history of ADs have been promoted as working faster and better with less SEs. If that was true people would now be getting better so fast they would be well before the doctor had even finished writing the script.

Join the other guinea-pigs if you wish: otherwise, come back in a decade or so.


1.         Renoux, C, Vahey, S, Dell’Aniello, S, and Boivin, JF, Association of Selective Serotonin Reuptake Inhibitors With the Risk for Spontaneous Intracranial Hemorrhage. JAMA Neurol, 2016.

2.         Roose, SP and Rutherford, BR, Selective Serotonin Reuptake Inhibitors and Operative Bleeding Risk: A Review of the Literature. J Clin Psychopharmacol, 2016. 36(6): p. 704-709.

3.         Hanley, GE, Smolina, K, Mintzes, B, Oberlander, TF, et al., Postpartum Hemorrhage and Use of Serotonin Reuptake Inhibitor Antidepressants in Pregnancy. Obstet. Gynecol., 2016. 127(3): p. 553-61.

4.         Gahr, M, Agomelatine in the treatment of major depressive disorder: an assessment of benefits and risks. Curr Neuropharmacol, 2014. 12(5): p. 287-398.

5.         Montastruc, F, Scotto, S, Vaz, IR, Guerra, LN, et al., Hepatotoxicity related to agomelatine and other new antidepressants: a case/noncase approach with information from the Portuguese, French, Spanish, and Italian pharmacovigilance systems. J Clin Psychopharmacol, 2014. 34(3): p. 327-30.

6.         Ng, I, Greenblatt, HK, and Greenblatt, DJ, Stereo‐Psychopharmacology: The Case of Citalopram and Escitalopram. Clinical Pharmacology in Drug Development, 2016. 5(5): p. 331-335.

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

Consider Donating to PsychoTropical

PsychoTropical is funded solely through generous donations, which has enabled extensive development and improvement of all associated activities. Many people who follow the advice on the website will save enormously on doctors, treatment costs, hospitalization, etc. which in some cases will amount to many thousands of dollars, even tens of thousands — never mind all the reduction in suffering and the resultant destruction of family, work, social, and leisure capability. A donation of $100, or $500, is little compared to those savings. Some less-advantaged people feel that the little they can give is so small it won’t make a difference – but five dollars monthly helps: so, do not think that a little donation is not useful.

– Dr Ken Gillman

Dr Ken Gillman