Article Text

Download PDFPDF

The sEROtonergic pathway: a hERO for the brain, a zERO for the valves?
  1. Tine L M De Backer1,2,
  2. Frank Timmermans1
  1. 1 Cardiovascular Diseases, University Hospital Ghent, Ghent, Belgium
  2. 2 Clinical Pharmacology, University Ghent, Ghent, Belgium
  1. Correspondence to Professor Tine L M De Backer, Cardiovascular Diseases, University Hospital Ghent, Gent 9000, Belgium; tine.debacker{at}

Statistics from

In medical school and ever since, we have been taught with the wisdom that pharmacological agents can heal, but also can harm. Medications or their metabolites that target the serotonergic pathways do not escape form this mantra. The last decades, several reports have suggested an association between serotonergic agents and valvular heart disease that eventually led to their withdrawal from the market. This was the case for the anorectic drugs fenfluramine and dexfenfluramine in the 1990s, and for the dopamine receptor agonist pergolide, a second-line Parkinson disease therapy, withdrawn in 2007 in the USA and in 2011 in Europe.1 2 As a large number of commonly prescribed medications such as antipsychotics, antidepressants, anxiolytics and antimigraine medications act on serotonergic mechanisms, concerns were and are raised that these approved agents might also cause drug-induced valvular heart disease (DIVHD).

Most reports that have studied an association between serotonergic medications and valvular disease show an inherent risk of bias and confounding due to methodological issues such as study design (cross-sectional, case–control, …), low number of patients included, variable dose and duration of medication intake, retrospective nature of the studies, short follow-up time, unblinded echocardiographic assessments and variable criteria used to define DIVHD. Yet, most of the Hill’s criteria for causation between serotonergic medications and valvular disease have been considered and fulfilled to a certain extent, and we therefore may conclude that these medications may be harmful to cardiac valves and may cause cardiac morbidity.3 Because DIVHD is reminiscent to carcinoid heart disease, it was subsequently discovered that DIVHD is mediated through 5-hydroxytryptamine receptor (5-HT) 2B signalling, and agents or their metabolites that specifically interact with this receptor are therefore most likely to be at highest risk for DIVHD. Signalling through the 5-HT 2B receptor was shown to elicit mitogenic and secretory responses in ventricular and heart valve interstitial cells.4 Moreover, 5-HT 2B signalling has been demonstrated in human mitral valve prolapse and might contribute to pathological remodelling of mitral valve prolapse.5 Whether serotonergic drugs/pathways may affect the natural history of other degenerative valve diseases such as aortic stenosis remains unknown. Drugs or their metabolites that do not significantly interact with the 5-HT 2B do not appear to induce DIVHD, but vigilance with respect to their metabolites or other 5-HT pathways remains warranted.4

In their  Heart paper, Fortier et al 6 performed a systematic review and meta-analysis on the association between valvular heart disease and drugs with known serotonergic or dopaminergic effects, synthesising the evidence according to the Cochrane methodology. In concert with earlier literature, the authors observed a significant association between valvular heart disease and intake of both serotonergic (OR=3.30) and dopaminergic (OR=2.56) drugs, indicating that overall serotonergic and dopaminergic drugs might increase the risk of valvular heart disease. As already noted above, the current meta-analysis also deserves some notes of caution. First, most of the studies selected are ‘outdated’, and the illustrative funnel plot demonstrates high risk of selection and publication bias in this field. Also, most reports included in the analysis considered drugs that have been withdrawn from the market.

However, the study of Fortier et al is a reminder that DIVHD is clinical reality, and it should be considered in patients with no clear explanation for having valves with leaflet thickening and restrictive, non-calcified features progressively resulting in valvular regurgitation and/or valvular stenosis. The study also reinforces the importance of pharmacovigilance during drug development (especially 5-HT 2B signalling agents) and post market surveillance. As a large number of individuals are at risk for depression, anxiety and other mood disorders, and almost 40% of US citizens have obesity,7 this enormous population considered for treatment implicates that serotonergic antidepressants and novel anorectic drugs that act via serotonergic pathways may potentially cause a considerable burden of DIVHD, even if the relative risk would be (very) low. However, even large randomised controlled trials (RCT) may be too small and of too short duration to detect rare but serious DIVHD. In fact, RCTs are mainly designed and powered to show benefit of an agent, and not necessarily for assessing its adverse effects. Consequently, postmarketing monitoring of any agent at risk remains important. In this regard, as scant data are available on the possible relation between 3, 4-methylenedioxymethamphetamine (‘Ecstasy’; a widely used recreational drug), methylphenidate (a norepinephrine-dopamine reuptake inhibitor for the treatment of attention deficit disorder), selective serotonin reuptake inhibitors and serotonin-norepinephrine reuptake inhibitors (reported in this review) and DIVHD, these agents deserve further attention, given their widespread use.

As for now, a risk–benefit balance should be made for each patient and regular follow-up, appropriate treatment dosing and duration could be one of the measures to minimise risk.



  • Contributors FT and TLMdB contributed equally to this editorial.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

  • Patient consent for publication Not required.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles