Tricyclic antidepressants side effects

by | Last updated Mar 1, 2020 | Published on Mar 1, 2020 | Anti-Depressants, TCAs

The receptor affinities for H1, A1, Muscarinic (m1-6 subtypes) etc determine much of the profile of side effects. This varies a lot between the various available TCAs (see detailed information on receptor affinities in other notes) and see my TCA review paper (free pdf here).

The overall severity of adverse effects (if used at appropriate doses — with blood level monitoring as needed) is probably:

Best: nortriptyline, desipramine* and lofepramine

Intermediate: Amitriptyline, imipramine, clomipramine, Maprotiline

Worst: dothiepin (aka Dosulepin, weak and more toxic in OD), trimipramine

It is difficult to justify the clinical use of Dosulepin for depression, some toxicologists feel it should no longer be on the market.

*Those who are well-read may know that this drug was taken off the market, supposedly because of toxicity. The usual daily dose (referred to in the literature as the DDD, which means the ‘defined daily dose’, was artificially high. That was the result of people trying to get an antidepressant effect by inappropriately using very large doses (recommended doses were frequently as high as 300 mg daily in many sources, which is ridiculous). This meant that a single packet of the drug contained far more milligrams in total than any of the other TCAs — therefore when people took a whole packet as an overdose it was more likely to be harmful. This is another dramatic illustration of the pharmacological incompetence of psychiatrists — it is such a potent NRI that it fully inhibits the uptake pump at a dose of only 10-25 mg; therefore administering much more than that for most people is simply giving an excessive dose.

Doxepin, in doses of 5-25 mg, is a useful sedative and antihistamine (it is the most potent antihistamine currently available); but it should not be regarded as an antidepressant drug (nor should trimipramine).

This enormous differences in the various properties of these drugs illustrates that they should not be considered as a homogenous pharmacological group.

The term TCA is a purely descriptive categorisation: it bears no relationship to pharmacology whatsoever.

This is a consequential error of categorisation and thinking. It has led to much misplaced advice over the decades, concerning the safety, and properties, of these drugs, particularly by labelling all TCAs as dangerous with MAOIs, whereas in fact it is only the ‘SRI’ members of the group, i.e. clomipramine and imipramine, that carry a risk of precipitating serotonin toxicity.

All the other TCAs are perfectly safe with MAOIs.

Effectiveness vs toxicity

The view that one should not use the old TCAs because they are more toxic in overdose, and also more likely to cause motor impairment, and therefore be implicated in causing accidents of various sorts, has long been adduced to strongly promote the successive waves of newer drugs.

This point of view, as an unqualified generalisation, has been overstated. This situation is partly the result of the high advertising profile now needed by most companies to market their new products. This results in the older drugs getting little attention but being much maligned. It is certainly the case that new is not necessarily better.

It is also worth appreciating that there are very considerable differences between the various old tricyclic antidepressants, so much so that lumping them all together (as is usually done) is not scientific or helpful. I have been espousing that view for a long time and I am therefore much gratified to see the traction being gained by the drive to classify drugs by their neuro-pharmacological profile [1-6].

Lofepramine and nortriptyline are probably less toxic in overdose than the ‘average’ TCA — their toxicity is of the same order of magnitude as the SSRIs. The new(er) antidepressant venlafaxine is more toxic (in overdose) than some tricyclic antidepressants, e.g. NTP.

These statements are appropriately couched in cautious terms because the quality ‘toxico-epidemiological’ evidence from which we would be able to make definitive judgements remains incomplete.

There is an interesting debating point around my inclusion of clomipramine as being ‘less toxic’. An explanation is educative. If one looks at the deaths from overdoses per million scripts issued, then clomipramine has a significantly lower death rate than other antidepressants. However, if one looks at the percentage of people who take an overdose of clomipramine who get toxic effects, then that is higher than some comparator drugs, such as SSRIs.

The debating point is this, since less people take an overdose when treated with clomipramine compared to other drugs — this is itself is evidence that clomipramine is a more effective antidepressant.

If you do the calculation of the total number of people treated with the drug, vs the total number who suffer toxic effects from it, then CMI comes comes out on top, despite the fact that for each individual who actually takes an overdose of it there is a slightly increased risk of toxic effects compared to various other drugs.

Which ‘measure’ is more valid as a real-world measure of ‘toxicity’?

This illustrates the point I have discussed in greater or lesser depth in various commentaries I have written: using a less toxic drug which is less effective is generally not a clinically sensible decision.

Bottom line is that overall, you are less likely to come to harm if your depression is treated with clomipramine, even though it is, by a restrictive definition, more toxic than some newer drugs.

Sedative potency (H1 blockade) is important for TCAs and potency for H1 blockade varies enormously between different TCAs, some being virtually devoid of this effect [see ‘receptor affinities’]. Of the newer drugs mirtazapine is equipotent to doxepin.

Falls in old people have not been shown to be less with SSRIs. Indeed, and perhaps a little counter intuitively, TCAs may in fact carry a lower risk of hip fractures [7, 8].

This suggests that if one uses an appropriate TCA (a less sedative and less hypotensive one) then even in special at-risk groups TCAs may be as good as, or better than, SSRIs (or other new drugs).

Even if some TCAs are significantly more toxic in over-dose than alternatives, this is a specific risk that may be satisfactorily managed in most patients by the simple expedient of giving out smaller numbers of tablets at a time; after all, if such patients are not in hospital then they are likely to require frequent visits. Suicide risk is certainly not, of itself, a reason to switch to less ‘toxic’ drugs, which have modest evidence of even equal efficacy (compared to amitriptyline, or clomipramine) and certainly no significant evidence of superior efficacy.

Suicide risk may in fact be a positive and strong indicator that:

A selective serotonin reuptake inhibitor should be avoided

A more potent antidepressant is required

The evidence suggests this means a tricyclic antidepressant, probably amitriptyline or clomipramine, an SNRI, or an MAOI.

For references see my review paper [9] about TCAs which is accessible as fulltext here

References

1. Zohar, J., et al., A review of the current nomenclature for psychotropic agents and an introduction to the Neuroscience-based Nomenclature. Eur Neuropsychopharmacol, 2015. 25(12): p. 2318-25.

2. Nutt, D.J. and P. Blier, Neuroscience-based Nomenclature (NbN) for Journal of Psychopharmacology. J Psychopharmacol, 2016. 30(5): p. 413-5.

3. Blier, P., M.A. Oquendo, and D.J. Kupfer, Progress on the Neuroscience-Based Nomenclature (NbN) for Psychotropic Medications. Neuropsychopharmacology, 2017. 42(10): p. 1927-1928.

4. Gorwood, P., et al., Editorial: Neuroscience-based Nomenclature (NbN) replaces the current label of psychotropic medications in European Psychiatry. Eur Psychiatry, 2017. 40: p. 123.

5. Worley, L., Neuroscience-based nomenclature (NbN). Lancet Psychiatry, 2017. 4(4): p. 272-273.

6. Seifert, R., Rethinking Pharmacological Nomenclature. TiPS, 2018: p. https://www.sciencedirect.com/science/article/pii/S0165614718301056.

7. Souverein, P.C., et al., Understanding inconsistency in the results from observational pharmacoepidemiological studies: the case of antidepressant use and risk of hip/femur fractures. pharmacoepidemiology and drug safety, 2016. 25: p. 88-102.

8. Hung, S.-C., et al., Use of selective serotonin reuptake inhibitors and risk of hip fracture in the elderly: a case-control study in Taiwan. Journal of the American Medical Directors Association, 2017. 18(4): p. 350-354.

9. Gillman, P.K., Tricyclic antidepressant pharmacology and therapeutic drug interactions updated. British Journal of Pharmacology 2007. 151(6): p. 737-48.

apple podcasts
Reddit
Visit Us
Follow Me
Skype
PodBean
Research Gate
Google Scholar