Antidepressants and emotions: therapeutics and iatrogenic effects

Hans-Jürgen MÖLLER,MD, PhD

Department of Psychiatry and Psychotherapy, Ludwig-Maximilians-University Munich

Antidepressants and emotions: therapeutics and iatrogenic effects

by H. J . Möller and
F. Seemüller,

Antidepressants can ameliorate depressive symptoms. Apart from the specific pharmacological action of the respective compound, we still have little knowledge about the way antidepressant drugs modulate neural processing of emotional and affective information. One proposed mechanism is an antidepressant-induced increase in processing of positive information in healthy volunteers and acutely depressed patients early in treatment. Such action may help explain the role of monoamines in emotional dysfunction in depression and how antidepressants work. This article will first provide an overview of pathomechanisms of emotional processing in depression and then review data on emotional processing of serotonergic and noradrenergic compounds. It has also been speculated that antidepressants may, in the same manner that they have positive effects on depression, lead to unwanted secondary emotional effects. Some early case reports suggested that selective serotonin reuptake inhibitors, especially, may lead to emotional blunting, a term commonly referred to as a restricted range of emotions. In the years that followed, this side effect was systematically studied. Relevant articles have been critically reviewed and are summarized in the manner of a systematic review. This article will also discuss the neurobiological underpinnings and possible clinical implications of emotional blunting.

Medicographia. 2013;35:304-309 (see French abstract on page 309)

The effects of antidepressant drugs on emotion and on negative biases in depression were usually thought to reflect nonspecific consequences of mood improvement. In recent years, the pharmacological modulation of emotion has become a growing field of research. One of the basic theories on the role of emotional modulation and depression is the so-called “mood congruency hypothesis.” It suggests that mood may enhance the processing of mood-congruent material and impair the processing of mood-incongruent material. This in turn may imply a pronounced tendency to attribute negative emotions to neutral faces in depressed patients.1,2 Recent findings suggest that antidepressants directly affect the way in which emotional information is processed and, therefore, help to reduce the negative bias in perception and memory in depression.2 However, it is noteworthy that even in remission the impairment in processing of neutral faces was evident. Remitted patients attributed not only sadness, but also happiness, to neutral faces. This suggests that this impairment may also be a trait characteristic independent of mood improvement.3 In other words, patients in remission after a major depressive episode may also interpret emotionally neutral social cues as emotionally meaning- ful. This may explain why remitted patients are still vulnerable to development of further depressive episodes.2 Antidepressants seem to affect the way in which emotional information is processed, finally leading to a reduced negative bias in perception and memory that are believed to contribute to the symptoms of depression. Antidepressants may, therefore, work in a manner quite similar to that of psychological treatment that aims to redress negative biases in information processing. Given the possibility that antidepressants—especially selective serotonin reuptake inhibitors (SSRIs)—may neutralize the processing both of negative and positive emotions, one might speculate that such an iatrogenic effect could also blunt subjective response in patients who take them.

In the early nineties, some case reports described a new dimension of side effects which were described as blunted emotional behavior.4,5 The term blunted emotion is commonly referred to as a restricted range of emotions.

Blunted emotions are also commonly thought to be one of the main obstacles in combining pharmacologic SSRI treatment with psychotherapy. For example, some psychotherapists recommend cessation of antidepressant SSRI treatments before exposure therapy in anxiety patients. The rationale behind this is that exposure treatment, in theory, can only be effective if emotions, including anxiety, are not suppressed or blunted by drugs. Consequently, antidepressants are sometimes tapered down or stopped, which is unfortunate, taking into account the significant risk of a recurrence or relapse of the depressive episode.

However, the question remains unanswered whether such side effects really exist or whether they, for example, only represent residual symptoms of the major depressive disorder, such as affective rigidity, or whether they represent personality traits. Recently, with the growing database on side effects of antidepressants, some treatment-emergent emotional side effects for tricyclic antidepressants (TCAs), such as sudden appearance of “anger” and “outburst,” have been described. This article will review the therapeutic effects of serotonergic and adrenergic antidepressants on emotional processing on the one hand, and summarize what is known about the concept of “emotional blunting” on the other, and will, finally, briefly comment on the clinical management of emotional blunting.

Therapeutic effects of antidepressants on emotional processing

The improvement of symptoms on a depression rating scale provides good evidence for the efficacy of an antidepressant. While the actions of antidepressants are well characterized (eg, serotonin reuptake inhibition) there is only little understanding of how these mechanisms lead to an improvement of symptoms. However, the effects of antidepressants on the modulation of emotional processing might provide a deeper understanding of its psychological mode of action.6 It also helps us as clinicians to explain to our patients what psychological changes can be expected on an antidepressant medication.

There are many studies of acute and long-term administration of antidepressants in animal models, healthy volunteers, and depressed populations. In the following, we will focus on studies in healthy volunteers and in clinical samples.

For example, a single-dose administration of citalopram helped healthy females to correctly process happy facial expressions and to recognize fear. However, citalopram had no effect on any other negative emotions like sadness, anger, and disgust.7

Another electrophysiological study by Kemp investigated the effects of acute serotonergic administration of citalopram on cortical electrophysiological responses to the processing of pleasant and unpleasant visual stimuli. Kemp’s findings suggest that acute serotonergic augmentation with citalopram modulates cortical processing of emotionally valenced stimuli, such that the response to pleasant valence is potentiated, whereas the response to unpleasant valence is suppressed.8

With respect to long-term administration of SSRIs, Harmer9,10 found a reduction in the processing of negative emotional stimuli following a seven-day trial of citalopram at a daily dose of 20 mg. At the end of the seven-day period, healthy subjects showed reduced recognition of negative facial expressions and also an improved memory for positive information. A reversal in fear processing from acute to chronic treatment with SSRIs has also been described in many preclinical anxiety models and clinical anxiety can easily be exacerbated initially with SSRI treatment. It is thus tempting to speculate that the initial increase followed by a final decrease in fear perception may relate to opposing effects on neural substrates involved in fear processing and that these changes may relate to the therapeutic effects of SSRIs in depression and anxiety.11

With respect to the noradrenergic effect of a single dose of the compound reboxetine, Harmer and coworkers have conducted several important studies. She examined the effect of re-boxetine on emotional processing compared with placebo in 24 healthy controls. On three different measures, reboxetine biased perception toward positive rather than negative information in the absence of effects on neurocognitive performance. For example, the facial expression task revealed greater recognition of happy facial emotions after reboxetine compared with placebo, without improvement in cognitive performance.12

The effect of depression on emotional processing and its correction by reboxetine is also nicely illustrated in Figure 1.6,9 Harmer and colleagues demonstrate that healthy controls seem to have a positive recall bias regarding the recall of positive adjectives, which cannot be found in depressed subjects. However, after single-dose administration of 4 mg of reboxetine, as compared with placebo, the patients regain the usual “healthy” positive recall bias (Figure 1).9

Another interesting study by Harmer13 compared reboxetine or citalopram with placebo over seven days in a double-blind manner in 42 healthy volunteers. Both compounds significantly reduced the recognition of fearful and angry facial expression as compared with placebo. Of note, subjects receiving citalopram were more likely to misclassify anger, disgust, and fear than happy emotions, suggesting again a positive bias in facial expression tasks.

These findings suggest that antidepressants can have early effects on emotional processing with some effects being seen with acute administration and independently of changes in mood. Serotonin potentiation might lead to positive biases in emotional processing and memory. This is exactly the opposite effect seen in depressed states. However, it is well established that drug treatments for depression usually require repeated administration before their effects become clinically apparent. Some theories, therefore, assume that the delay in antidepressant action arises because of apparent delay in the initiation or expression of the relevant pharmacological and/or cellular actions. The delay might also be caused by the requirement for psychological consolidation and processing of the effects of these neurochemical actions.

Figure 1
Figure 1. Mean numbers of
positive and negative self-referent
adjectives in a recall task.

(A) The effects of depression per se.
Comparison for positive stimuli between
depressed patients and comparison
subjects receiving placebo significant
at P<0.05. (B) The acute effect of oral
reboxetine (4 mg) or placebo in depressed
group only. Comparison for
positive stimuli between depressed patients
receiving placebo and those receiving
reboxetine significant at P<0.01. After reference 9: Harmer et al. Am J
Psychiatry. 2009;166(10):1178-1184.
© 2009, American Psychiatric Association.

Generally speaking, the hypothesis is that correcting emotional bias may lead to an improved homeostatic mood response to experience and that recovery in depression is delayed because of the need for relearning of external contingencies and internal states under more positive emotional bias.14 On the other hand, the neutralization of negative emotions might in the long run lead to secondary effects like emotional blunting, which will be critically reviewed in the following paragraph.

Iatrogenic, secondary effects of antidepressants on emotions

In 1990, Hoehn-Saric et al reported apathy, indifference, loss of initiative, and disinhibition in panic disorder and depressed patients on SSRIs.5 A bit later, in 1991, Hoehn-Saric reported apathy, indifference, inattention, and perseveration in an obsessive- compulsive patient taking fluoxetine.4,15 Subsequently, Oleshansky and Labbate (1996) described rapid improvement of excessive or inappropriate crying without apathy or indifference in depressed patients on SSRI treatment.15 Finally, Vinar et al reported that eight depressed women spontaneously cried while watching movie scenes on television over a period of years. This crying disappeared during long-term treatment with SSRIs.

The first systematic case-control study on blunted emotions was a brief report by Opbroek published in 2002.16 The authors studied 15 depressed patients in remission, who took three different SSRIs (fluoxetine, paroxetine, and sertraline). They were all recruited froma study of patients reporting SSRI induced sexual dysfunction. Patients and healthy controls were rated with a newly developed blunted-emotions rating scale (Laukes Emotional Intensity Scale, [LEIS]).16 About 80% of the 15SSRI-treated remitted patients reported emotional blunting. The LEIS consists of 17 items. In only two items—work and energy levels—did the authors find no significant difference as compared with the healthy controls. The frequency of patients experiencing diminished emotional response was 93% for sexual interest, 80% for sexual pleasure, 60% for inability to cry, 53% for erotic dreams, 50% for creativity, and 47% for becoming irritated or upset. There was no difference in the total LEIS score between the different antidepressant classes. The author finally concluded that sexual dysfunction as seen with SSRI treatment may be just one of a larger set of diminished emotional responses as a consequence of increased serotonin neurotransmission.

Despite the numerous methodological limitations of this study, this paper has stimulated a wider debate and enthusiastic reactions from some researchers. Luckily enough, a detailed comment on this study, published in the same issue, summarized the most important limitations17: The applied scale was used for the first time and lacked validity as well as reliability. There was no baseline measurement of the LEIS. The patients were instructed to rate the 18 questions relative to their usual state. This may bias the memory of the respondents toward attribution of the experienced phenomena to side effects of SSRIs. The lacking correlation with depressive symptoms in remitted patients also is not surprising given the minimal variance of depressed symptoms in remitted patients. In addition, the comparison of depressed patients undergoing SSRI treatment with healthy controls not undergoing SSRI treatment, rather than making the comparison with a placebo control group, does not make much sense. Thus, it is impossible to distinguish depressive symptoms from side effects.

Moreover, the patient sample consisted of a subsample of patients reporting sexual side effects; the LEIS scale is biased toward sexual dysfunction as it (i) contains four sexual items and (ii) highly correlates with a sexual side effect scale. In contrast to the authors’ conclusion that sexual side effects may only be the tip of the iceberg regarding unwanted emotional side effects, it may well be the other way around: severe sexual dysfunction and impairment may cause or at least contribute to emotional blunting. Sexual dysfunction is a known SSRI side effect due to two main mechanisms: (i) elevation of prolactin levels (as has been shown for almost all antidepressants,18 leading to a decrease in libido and (ii) SSRI specific orgasmdelay.19 Thus, it could also be that the decrease in libido secondarily leads to other emotional phenomena, since a healthy sexual life might substantially contribute to a higher quality of life and might thus be a prerequisite for a vital emotional life. Furthermore, it has not been distinguished whether these phenomena impaired patients’ quality of life or whether they might also have been wanted and desired effects of the treatment, as in some of the reports described above. Finally, these questions remain unanswered: Exactly what symptoms constitute the concept of emotional blunting? Is the normalization of pathologic symptoms such as pathologic crying emotional blunting? Do our standardized depression rating scales like the Hamilton Depression Rating Scale (HAM-D) or the Montgomery-Asberg Depression Rating Scale (MADRS) not capture similar symptoms which would be classified as residual symptoms?

Measuring emotional blunting

A different approach has been suggested by Barnhart et al.20 He introduced the term apathy in the context of SSRI-induced emotional blunting and adopted a definition from Marin et al21: a syndrome in which there is a primary absence of motivation that is not attributable to cognitive impairment, emotional distress, or diminished level of consciousness. However, this approach also lacks specificity regarding side effects of antidepressants, as Marin himself suggested the following possible psychiatric differential diagnoses of apathy: delirium, dementia, abulia, akinesia, despair, and depression.21 We would like to suggest including other common psychiatric diagnoses, such as negative symptoms in schizophrenia and chronic residual states as well. However, important longitudinal information, such as the late appearance of apathy during SSRI treatment in the absence of depressive symptoms and diminishing apathy during titration downward may provide more helpful information in distinguishing apathy or emotional blunting as a side effect from depressive residual symptoms.

Rating scales for blunted emotions

Regarding standardized rating scales, there are four instruments for the measurement of SSRI-emergent emotional blunting thus far available: the first is the Marin Apathy Evaluation Scale (AES). However, the scale was developed for an older population and has been used in schizophrenia trials and neurological disorders.22 The second is the above-mentioned LEIS, comprising 18 questions, which thus far lacks validity and specificity, as outlined above.16 The third is the Bell-Shipman Apathy/Emotional Blunting Questionnaire.6,23 In the literature, it is described as “under development,” but up to now there is still no published evidence of its completion. The fourth and perhaps most comprehensive and elaborate questionnaire is the recently published Oxford Questionnaire on the Emotional Side effects of Antidepressants (OQuESA).6,24 It comprises three different sections with 26 self-reported items altogether. It has been tested in a cohort of 207 depressed people: 26% reported that they did not experience side effects; 16% reported insignificant emotional side effects, 30% mild side effects, 23% moderate, and 6% severe. Patients with emotional side effects were significantly younger, had a significantly higher Beck Depression Inventory (BDI) score and shorter treatment duration. The association with depressive symptoms raises the possibility that the OQuESA may not be specific for side effects, but might rather capture depressive symptomatology. Moreover, emotional side effects were more common in patients with a shorter treatment duration, which also suggests greater illness activity and acuity and contrasts with findings from case reports where emotional blunting seems to appear more often in the long run with antidepressant treatment. In fact, the authors themselves state in their discussion that: “the OQuESA measures one or more aspects of depression, rather than necessarily measuring only emotional effects.”24 The authors state that appropriate double-blind studies are under way using the questionnaire, hopefully clarifying the question of overlap between depressive and residual symptomatology.24 However, it would have also been interesting to look for correlations between the BDI and the OQuESA, which unfortunately has not been analyzed in this study.

Possible mechanisms of emotional blunting

In an excellent review of case reports, Barnhart et al discusses two mechanisms: (i) Frontal lobe activity may be modulated though serotonergic projections, finally leading to emotional blunting.20 (ii) SSRIs may influence serotonergic systems, which in turn might modulate midbrain dopaminergic systems also projecting to the prefrontal cortex and triggering emotional blunting.4,20

Clinical management of emotional blunting

According to Barnhart et al, there are three different ways to manage or respond to emotional blunting.20 Probably, the simplest approach is to reduce the dosage of the antidepressant. In most case reports, particularly in patients taking SSRIs with a shorter half-life, a dose reduction led to complete resolution of emotional blunting. However, in these cases, this side effect typically occurred later, after several months of treatment with an antidepressant. Earlier, it might be even more difficult to distinguish emotional blunting from depressive symptomatology. Thus, one might not want to put the patient at risk of worsening depression through antidepressant dose reduction. A case report described the resolution of emotional blunting when buproprion was added. The addition of buproprion is also a well-established augmentation strategy.25

This strategy might especially be helpful in patients with a partial response to an antidepressant and where it is not clear to what extent the phenomena might be attributable to the initial depressive episode. The third and last management option would be to switch the antidepressant. Based on the above observation that dose titration was a more effective strategy with short–half-life SSRIs, one might think of switching to an SSRI with a shorter half-life. Some case reports also described patients experiencing apathy with an SSRI and who did not experience apathy after switching to monoamine oxidase inhibitors or TCAs.20

An interesting new alternative might be a switch to agomelatine. This is an antidepressant whose properties (MT1/MT2 agonist and serotonin 2C [5-HT2C] antagonist, causing no release of serotonin in the brain) might protect from emotional blunting with a pharmacologic effect that is primarily upon circadian synchronization and enhancement of dopaminergic and adrenergic input to the frontal cortex through the synergy of its receptors.26,27 Agomelatine facilitates positive versus negative affective processing, including emotional memory and fear-potentiated startle.6,28


There is a wide range of evidence available suggesting that the psychological antidepressant effect of pharmacologic treatments can arise from a reduction in negative bias in emotional processing. Apart from positive effects, emotional blunting has been discussed as an unwanted emotional side effect. There are specific rating instruments available and several reports suggest the existence of emotional blunting under SSRI treatment. However, so far there are no convincing criteria that can separate emotional blunting from residual depressive symptoms. Clinical reports suggest that emotional blunting tends to appear rather late with SSRI treatment and also seems to be dose dependent. Thus, a dose reduction or switch to an SSRI with a shorter half-life might be the first choice in the management of emotional blunting. _

1. Gur RC, Erwin RJ, Gur RE, Zwil AS, Heimberg C, Kraemer HC. Facial emotion discrimination: II. Behavioral findings in depression. Psychiatry Res. 1992;42 (3):241-251.
2. Serra M, Salgado-Pineda P, Delaveau P, Fakra E, Gasto C, Blin O. Effects of antidepressant drugs on emotion. Clin Neuropharmacol. 2006;29(3):170-185.
3. Drevets WC. Neuroimaging and neuropathological studies of depression: implications for the cognitive-emotional features of mood disorders. Curr Opin Neurobiol. 2001;11(2):240-249.
4. Hoehn-Saric R, Harris GJ, Pearlson GD, Cox CS, Machlin SR, Camargo EE. A fluoxetine-induced frontal lobe syndrome in an obsessive compulsive patient. J Clin Psychiatry. 1991;52(3):131-133.
5. Hoehn-Saric R, Lipsey JR, McLeod DR. Apathy and indifference in patients on fluvoxamine and fluoxetine. J Clin Psychopharmacol. 1990;10(5):343-345.
6. Goodwin GM. Symptom relief and facilitation of emotional processing. Eur Neuropsychopharmacol. 2011;21(suppl 4):S710-S715.
7. Harmer CJ, Bhagwagar Z, Perrett DI, VollmBA, Cowen PJ, Goodwin GM. Acute SSRI administration affects the processing of social cues in healthy volunteers. Neuropsychopharmacology. 2003;28(1):148-152.
8. Kemp AH, Gray MA, Silberstein RB, Armstrong SM, Nathan PJ. Augmentation of serotonin enhances pleasant and suppresses unpleasant cortical electrophysiological responses to visual emotional stimuli in humans. Neuroimage. 2004;22(3):1084-1096.
9. Harmer CJ, O’Sullivan U, Favaron E, et al. Effect of acute antidepressant administration on negative affective bias in depressed patients. Am J Psychiatry. 2009;166(10):1178-1184.
10. Harmer CJ, Bhagwagar Z, Cowen PJ, Goodwin GM. Acute administration of citalopram facilitates memory consolidation in healthy volunteers. Psychopharmacology (Berl). 2002;163(1):106-110.
11. Sheline YI, Barch DM, Donnelly JM, Ollinger JM, Snyder AZ, Mintun MA. Increased amygdala response to masked emotional faces in depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry. 2001; 50(9):651-658.
12. Harmer CJ, Hill SA, Taylor MJ, Cowen PJ, Goodwin GM. Toward a neuropsychological theory of antidepressant drug action: increase in positive emotional bias after potentiation of norepinephrine activity. Am J Psychiatry. 2003;160(5): 990-992.
13. Harmer CJ, Shelley NC, Cowen PJ, Goodwin GM. Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition. Am J Psychiatry. 2004; 161(7):1256-1263.
14. Harmer CJ, Goodwin GM, Cowen PJ. Why do antidepressants take so long to work? A cognitive neuropsychological model of antidepressant drug action. Br J Psychiatry. 2009;195(2):102-108.
15. Oleshansky MA, Labbate LA. Inability to cry during SRI treatment. J Clin Psychiatry. 1996;57(12):593.
16. Opbroek A, Delgado PL, Laukes C, et al. Emotional blunting associated with SSRI-induced sexual dysfunction. Do SSRIs inhibit emotional responses? Int J Neuropsychopharmacol. 2002;5(2):147-151.
17. Balon R. Emotional blunting, sexual dysfunction and SSRIs. Int J Neuropsychopharmacol. 2002;5(4):415-416; author reply 7.
18. Coker F, Taylor D. Antidepressant-induced hyperprolactinaemia: incidence, mechanisms and management. CNS Drugs. 2010;24(7):563-574.
19. Serretti A, Chiesa A. A meta-analysis of sexual dysfunction in psychiatric patients taking antipsychotics. Int Clin Psychopharmacol. 2011;26(3):130-140.
20. Barnhart WJ, Makela EH, Latocha MJ. SSRI-induced apathy syndrome: a clinical review. J Psychiatr Pract. 2004;10(3):196-199.
21. Marin RS. Differential diagnosis and classification of apathy. Am J Psychiatry. 1990;147(1):22-30.
22. Marin RS, Biedrzycki RC, Firinciogullari S. Reliability and validity of the Apathy Evaluation Scale. Psychiatry Res. 1991;38(2):143-162.
23. Bell S, ShipmanM, Bystritsky A, Haifley T. Fluoxetine treatment and testosterone levels. Ann Clin Psychiatry. 2006;18(1):19-22.
24. Price J, Cole V, Doll H, Goodwin GM. The Oxford Questionnaire on the Emotional Side-effects of Antidepressants (OQuESA): development, validity, reliability and sensitivity to change. J Affect Disord. 2012;140(1):66-74.
25. Bech P, Fava M, Trivedi MH, Wisniewski SR, Rush AJ. Outcomes on the pharmacopsychometric triangle in bupropion-SR vs. buspirone augmentation of citalopram in the STAR*D trial. Acta Psychiatr Scand. 2012;125(4):342-348.
26. de Bodinat C, Guardiola-Lemaitre B, Mocaër E, Renard P, Muñoz C, Millan MJ. Agomelatine, the first melatonergic antidepressant: discovery, characterization and development. Nat Rev Drug Discov. 2010;9(8):628-642.
27. Racagni G, Riva MA, Molteni R, et al. Mode of action of agomelatine: synergy between melatonergic and 5-HT2C receptors. World J Biol Psychiatry. 2010;1 2(8):574-587.
28. Harmer CJ, de Bodinat C, Dawson GR, et al. Agomelatine facilitates positive versus negative affective processing in healthy volunteer models. J Psychopharmacol. 2011;25(9):1159-1167.

Keywords: agomelatine; antidepressants; emotional blunting; iatrogenic effect; SSRI; tricyclic antidepressants