When is a side effect not a side effect? With antidepressants and functional gastrointestinal disorders

ibsNew research has shown that the side effects of tricyclic antidepressants reported by patients with functional gastrointestinal disorders (FGDs) such as irritable bowel syndrome (IBS) often aren’t actually real side effects of the drugs.  Instead, most of the symptoms experienced by the women in the study were present before they started taking the medication, suggesting that the patients receiving antidepressants were simply over-reporting symptoms rather than experiencing side effects.

Interestingly, the authors of this study note that patients with a higher level of psychological distress were more likely to report symptoms than were less anxious patients.  It is possible, therefore, that the ‘side effects’ among such patients were a result of their underlying stress and anxiety and not the antidepressants they were taking.

FGDs comprise gastrointestinal symptoms such as cramps but no clear pathology or anatomical problem.  Instead, the primary abnormality is an altered physiological function (ie a change in the way the body works) rather than an identifiable structural or biochemical cause.  FGDs do have a strong psychological component though, and are often triggered or exacerbated by stress. As such, antidepressants have proved effective in treating patients with FGDs; however, such drugs often have unpleasant side effects.

The authors of this study enrolled 245 women with FGDs: irritable bowel syndrome, painful constipation, functional abdominal pain, or unspecified functional bowel disorder. All patients filled in a questionnaire about their various symptoms at the start of the study, then 57 individuals were randomly assigned to receive the tricyclic antidepressant desipramine for 12 weeks and 36 were assigned a placebo for the same period of time.

At 2 weeks, both groups were reporting the same levels feeling tired in the morning, having trouble sleeping, nausea, blurred vision, headaches, and decreased appetite as they had experienced at the beginning of the study, indicating that these signs were not side effects of the tricyclic antidepressant.  On the other hand, dry mouth, lightheadedness, dizziness, flushing and jitters or tremors at week 2 were more common among the patients on desipramine than among those on placebo, suggesting that these particular symptoms may be related to the known anticholinergic effects of the medication.  By 12 weeks, there was no difference between the two groups in the side effects reported.

Patients with a high score on the SCL-90 GSI test, a measure of psychological distress, reported more symptoms than did patients with lower scores.  The authors thus state that the more symptoms a patient reports, the less likely it is that the symptoms are related to the medication.

I think this study has some interesting implications.  Given that antidepressants by design are often used in anxious individuals, it seems possible that in many cases the side effects reported with this class of drugs might be due to patients reporting every single symptom they experience, however slight.  Given this likelihood of over-reporting, and also the fact that the anxiety itself can cause symptoms, perhaps the side effects of tricyclic depressants are less common than we think, and these drugs might be safe to use in a broader spectrum of patients than is currently treated.


Thiwan S et al. (2009) Not All Side Effects Associated With Tricyclic Antidepressant Therapy Are True Side Effects. Clinical Gastroenterology and Hepatology 7 (4): 446-451 DOI: 10.1016/j.cgh.2008.11.014

You may also like


  1. Thanks for the comments guys.

    You’ve picked up on an interesting point Nestor – the placebo effect can be negative as well as positive so that patients may report unwanted symptoms instead of beneficial effects when taking a placebo. Hopefully doctors will be more thorough when discussing the pros and cons of antidepressants given the results of this study.

    The authors of this study noted that patients with functional gastrointestinal disorders are often reluctant to take antidepressants due to concerns about possible side effects, so your experience is interesting Laura. In fact, they say “the majority of patients in our National Institutes of Health treatment trial who discontinued desipramine reported side effects as the main concern”. Hopefully the knowledge that any symptoms they experience when taking antidepressants are probably not related to the treatment will encourage more patients with functional gastrointestinal disorders to try this therapeutic avenue.

  2. Interesting post. As someone who’s recently started taking tricyclic antidepressants for IBS (uh oh, too much information?), I can say that I was very worried about the potential side effects of psychiatric drugs before I started taking them. It made me really hesitant to begin treatment. Of course, as a neuroscience student I’m really good at over-analyzing these types of things, and was certain that if I messed with my neurotransmitters, all hell would break loose.

    I was actually surprised by how mild the side effects were. The worst is the dry mouth, especially first thing in the morning. It’s easy enough to fix by drinking a glass of water, though. As to whether the drug is helping my IBS — I wouldn’t call myself cured, but I think I’ve noticed some improvement. Whether that’s a real effect or placebo remains to be seen, but I’m thinking of asking my doctor to increase my dose to see if that helps.

  3. Hello Helen, one issue that always strikes me about placebo controlled studies, is how consistent patients are in reporting similar ‘side effects’ to both the placebo and the drug. It is almost like it reflects some sort of mass predisposition or reaction to the general belief that “this drug will probably make me feel bad”. I wonder if this is more common with psychiatric drugs than other drugs.

    The problem is that, placebo effect or not, patients are reporting these symptoms and are often more likely to stop the medication because of these symptoms. So I wonder if health providers should be more proactive in explaining to the patient the limits of the possible side effects ( stating what they are, how unlikely they are, and reaffirming that no other side effects are expected). This may narrow the patients’ expectation of the side effects and possibly prevent them from calling the “this drug will make me feel bad” schema.

    Cheers, Nestor.

Leave a Reply