In a study performed in 18 countries from every continent, the 12-month prevalence estimate of major depressive episodes (according to the DSM-IV criteria) ranged from 2.2% to 10.4%. The midpoint across all countries was 5%. A substantial proportion of people who seek treatment for major depression have a chronic-recurrent course of illness, and the lifetime prevalence is 2–3 times that of 12 months prevalence. Although the proportion of patients with depression who receive treatment may be less than one-half even in high-income countries, antidepressants are widely used. Moreover, other conditions than depression are treated with antidepressants, in the psychiatric field (e.g., obsessive-compulsive disorder, generalized anxiety disorder) or in nonpsychiatric conditions (e.g., chronic pain). All available antidepressants act via the monoamine neurotransmitters, serotonin and noradrenaline (Table 13.1). The older medications, i.e., tricyclic antidepressants and monoamine oxidase inhibitors, have been supplanted in most cases by newer agents with fewer side effects, mainly selective serotonin reuptake inhibitors (SSRIs) and serotonin and noradrenaline reuptake inhibitors (SNRIs). Although not the most frequent adverse effects of antidepressants, movement disorders may occur during the course of treatment. The study of this topic is hampered by several limiting factors: a) the rarity of systematic prospective studies properly designed to detect movement disorders; b) the use of ill-defined terms, such as “extrapyramidal symptoms” in the older medical literature; c) finally, the fact that a number of patients receiving antidepressants also receive other psychoactive drugs that may also generate movement disorders (e.g., neuroleptics, lithium), so that the imputability may be difficult to establish.
|Type of drug||Mode of action||Examples|
|Tricyclics||Inhibition of noradrenaline and serotonin reuptake||Imipramine, desipramine, amitriptyline, clomipramine, amoxapinea,…|
|Selective serotonin reuptake inhibitors SSRIs]||Inhibition of serotonin-selective reuptake||Fluoxetine, citalopram, escitalopram, sertraline, paroxetine|
|Noradrenaline reuptake inhibitors||Inhibition of noradrenaline-selective reuptake||Atomoxetine, reboxetineb|
|Serotonin and noradrenaline reuptake inhibitors [SNRIs]||Inhibition of noradrenaline and serotonin reuptake||Venlafaxine, duloxetine, desvenlafaxine|
|Monoamine oxidase inhibitors||Inhibition of monoamine oxidase A||Isocarboxazid, phenelzine|
Partially unknown or complex. Presumed antidepressant actions
– mirtazapine: enhanced release of noradrenaline and 5-HT1A-mediated serotonergic transmission
– bupropion: enhanced release and inhibition of the reuptake of dopamine and noradrenaline
– trazodone: 5HT2 receptor antagonism, weak serotonin reuptake inhibition
|mirtazapine, bupropion, trazodone|
a. Amoxapine, although usually classified as an antidepressant, has clinically relevant antidopaminergic properties with a profile close to that of atypical antipsychotics [Kapur et al. 1999].
b. Atomoxetine is currently approved in the United States to treat attention deficit–hyperactivity disorder. Reboxetine is not approved for any use in the USA and many countries due to a low benefit/risk ratio.
Antidepressants associated with tremor
Tricyclics, including amitriptyline, clomipramine, and imipramine have been identified as a cause of postural tremor in several controlled trials comparing the effects of these drugs with those of antidepressants of other pharmacological classes. In a systematic review and metaanalysis of 39 trials comparing efficacy and safety of amitriptyline to placebo, significantly more participants in the amitriptyline group suffered from tremor (OR 5.68, 95% CI 3.19 to 10.10, P <0.00001, 10 RCTs, 1230 participants). Tremor is also a common side effect of SSRIs, as shown by systematic reviews and metaanalyses.[5,6]
Typical rest tremor of parkinsonism is not addressed in this section, which focuses on postural and action tremor. Drug-induced tremor involves upper limbs, with a frequency of 6 to 12 Hz.[7,8] In patients disclosing a tremor while receiving antidepressant drugs, some suggestive features may help orientate the diagnosis (Table 13.2). However, the responsibility of an antidepressant drug may be difficult to ascertain: a) patients under antidepressant treatment often receive other medications with tremorogenic properties, such as mood stabilizers (valproic acid, lithium), neuroleptics, or other nonpsychoactive drugs; b) many chronic conditions in the field of movement disorders (e.g., Parkinson’s disease, dystonic syndromes) may display postural tremor and are associated with a high risk of depression and therefore of receiving antidepressant therapy.
Clear temporal relationship between tremor onset and drug initiation
Relationship between tremor severity and dose
Improvement or cessation of tremor when antidepressant is suspended
Exclusion of common medical or neurological causes of tremor (e.g., hyperthyroidism, essential tremor)
Relative symmetry of upper limb tremor, and absence of head (i.e., neck) tremor
Absence of worsening over time (if doses of antidepressants are stable)
Little is known about the mechanisms of drug-induced tremors in general, and of antidepressant-induced tremors in particular. The current view is that tremorogenic drugs act through an enhancement of the oscillations of peripheral physiological tremor, through an increase in the gain of the muscle receptors and spinal reflex loops. However, an electrophysiological study of patients with postural hand tremor following amitriptyline administration suggested an enhancement of the centrally driven component of physiological tremor in spinal or supraspinal pathways. It has been suggested that fluoxetine-induced tremors result from an increase of serotoninergic activation of the inferior olivary nucleus and/or the red nucleus, two structures which have been involved in the generation of tremors.
Antidepressants associated with parkinsonism
Although occasionally mentioned, the existence of parkinsonism associated with tricyclic use is questionable. Indeed, in a review of the literature describing the tricyclic-related “extrapyramidal” (i.e., movement disorder) side effects, Vandel et al. identified 30 case reports of such side effects, but none of them was called parkinsonism. Another review on the same topic did not report parkinsonism as a side effect of tricyclic antidepressants. Similarly, in a more recent systematic literature review, movement disorders (“extrapyramidal symptoms”) were identified in 89 patients taking only one antidepressant, but none of them had parkinsonism. However, in a study of 155 cases of drug-induced parkinsonism reported to a French pharmacovigilance center, tricyclics were involved in three (2%). Conversely, imipramine was found to have a beneficial effect on parkinsonian signs, independently of an effect on mood, in two early controlled studies, which raise a number of methodological issues.[17,18] Interestingly, a metaanalysis using a database compiled from six clinical trials suggested that early Parkinson’s disease patients treated with tricyclics, amitriptyline in particular, experienced a significant delay in the initiation of dopaminergic therapy, an effect not attributable to a symptomatic effect. The evidence of a role of newer antidepressants in inducing or aggravating preexisting parkinsonism is stronger. In 1979, Meltzer et al. published the case of a 25-year severely depressed man who, four days after the onset of a treatment with fluoxetine, acutely developed torticollis and parkinsonism. These symptoms cleared rapidly under treatment with trihexyphenidyl, despite continuous administration of fluoxetine. Since that report, many cases of parkinsonism induced by SSRIs have been reported.[16,20–22]
Clinical features and risk factors
Although drug-induced parkinsonism is often said to be relatively symmetrical with no rest tremor, this is far from being the rule, and the clinical presentation may mimic that of Parkinson’s disease. Parkinsonism usually occurs within the first three months of treatment. Subjects aged 65 years and older are more likely to develop parkinsonism under SSRIs. Identified potential risk factors include in one study the presence of the A1 allele of the D2 dopamine receptor gene [DRD2] Taq1A polymorphism. A facilitating role for CYP2D6 polymorphisms has also been postulated.
Do SSRIs worsen motor condition in patients with Parkinson’s disease?
Depressive disorders as well as depressive symptoms are common in Parkinson’s disease and are important factors affecting quality of life. Although their efficacy has been questioned, antidepressants are therefore widely used in patients with Parkinson’s disease. In these patients, conflicting data as to whether SSRIs worsen parkinsonism have been reported. Although some individual case reports suggested a reversible motor worsening in some patients, several open-label prospective studies on small groups of patients (14 to 33 each at inclusion) found that the following SSRIs did not modify motor function: fluoxetine, sertraline, paroxetine, citalopram, fluoxetine, fluvoxamine, and sertraline. However, in one prospective study of 65 patients with Parkinson’s disease and depression receiving paroxetine during at least three months, 13 patients stopped paroxetine because of adverse events, consisting in two cases in increased “off” time duration and exacerbation of tremor. Motor worsening resolved completely within 48 hours after discontinuation of the drug. In a Dutch retrospective study of levodopa users from a prescription database, it was shown that the start of SSRIs was accompanied by a faster increase of antiparkinsonian drug treatment as compared with starters of tricyclic antidepressants and patients not receiving antidepressants. In a systematic review and metaanalysis of five randomized controlled trials comparing SSRIs to placebos, the dropout rate was not significantly different in both groups of patients. In one of these studies, one patient under citalopram dropped out because of increased bradykinesia, which resolved after treatment discontinuation. Finally, in a recent placebo-controlled trial comparing the efficacy and the safety of paroxetine and of venlafaxine, there was no evidence of treatment-associated worsening of motor function (actually, mean motor scores slightly improved in all groups). In summary, SSRIs seem to very rarely worsen motor function in patients with Parkinson’s disease. If it happens, this effect is rapidly reversible after discontinuation of treatment. Nevertheless, most controlled studies were short (no more than four months), and the risk of treatment-related worsening of parkinsonism cannot be ruled out for long-term use.
Although the exact mechanism or SSRI-induced parkinsonism remains controversial, most authors believe it could involve inhibition of dopaminergic neurons by serotonin. This view derives from early studies showing that experimental manipulations aimed at decreasing central serotonergic function enhanced dopaminergic transmission. Moreover, in one patient who developed parkinsonism on fluoxetine, it was found that serum prolactin levels increased while homovanillic acid levels decreased in the cerebrospinal fluid. However, the picture seems to be much more complex. Midbrain raphe neurons emit 5-HT axons which arborize in virtually all components of basal ganglia, and it has been shown that they exert their influence at both ends of the nigrostriatal dopaminergic projection. The nature of this influence is not simple: several 5-HT receptor subtypes (including 5-HT1A, 5-HT1B, 5-HT2A, 5-HT3, and 5-HT4) act to facilitate dopamine release, while the 5-HT2 C receptor mediates an inhibitory effect of 5-HT on dopamine release. Finally, the relative rarity of parkinsonism on SSRIs, and the identification of risk factors, such as older age, female sex, and drug interaction through cytochrome P450 (CYPD2D6), suggest that the modulatory effect of serotonergic systems on dopaminergic systems is not a factor sufficient for parkinsonism to develop.[15,16]
Akathisia and restless legs syndrome
Sensu stricto, this term refers to the inability to remain seated, but encompasses more generally motor activity as a voluntary effort to relieve continuous uncomfortable sensations of restlessness. Frequently observed in patients treated with antipsychotics, it may also occur with antidepressants, with many observations following that of Zubenko et al. Akathisia has been described following treatment with tricyclic drugs, fluoxetine and other SSRIs,[37,38] and mianserin. Interestingly, trazodone, an atypical antidepressant with serotoninergic antagonistic properties, improved neuroleptic-induced akathisia in a small controlled study.
The antidepressant-induced akathisia has the same characteristics as those observed in patients with antipsychotics. It has been suggested that, in some instances, antidepressants, including the SSRIs, could increase suicidal behavior by inducing akathisia with associated self-destructive impulses. It may also be associated with violent behavior.
Antidepressants as a risk factor for RLS, prevalence. The prevalence of RLS in the general population has been found to be around 10%. The prevalence is higher in women than in men and it increases with age. Current evidence is contradictory regarding the question of RLS as a side effect of antidepressants. In a large study of almost 19,000 subjects performed in Europe, the prevalence of RLS was 5.5%. Users of SSRIs but not of other antidepressants were more likely to have RLS. In one prospective study of patients treated for the first time with antidepressants, RLS was recorded as a side effect in 9%. However, another community-based cross-sectional study found a prevalence of RLS of 10.6 and revealed no association between antidepressants and RLS. In a retrospective chart review of 200 consecutive patients referred for sleep initiation insomnia, 45% of the patients had RLS and no significant association was found between the use of antidepressants and RLS. Moreover, one retrospective chart review study found improvement of preexisting RLS by SSRIs in most patients. Another study suggested that the association of RLS and antidepressant use may vary by gender and type of antidepressant: antidepressant was more strongly associated with RLS for men, but not for women, and further analyses revealed an association between RLS and fluoxetine for women, and associations between RLS and citalopram, paroxetine, and amitriptyline for men. The discrepancy between these studies might have various causes: a) in the study by Dimmitt et al., preexisting RLS did deteriorate in 12% of the patients, and RLS developed in 9% of the patients free of RLS prior to treatment, which is in keeping with the results of other studies; b) RLS typically occurs during the initial days of treatment, and cross-sectional studies of patients with long-standing antidepressant treatment may miss the association with RLS; c) finally, the association between the use of antidepressants and RLS should be interpreted cautiously, as several cross-sectional, population-based, and clinic-based studies reported significantly higher rates of depression symptoms in individuals with RLS than in controls. A prospective study in a cohort of women followed up during six years also showed that subjects with preexisting RLS had an increased risk of developing clinical depression and clinically relevant depressive symptoms. In cross-sectional studies investigating the relationship between antidepressants and RLS, analyses should therefore adjust for depression.
Antidepressants associated with restless legs syndrome. Most antidepressants have been reported to provoke or deteriorate RLS: mianserin, mirtazapine, SSRIs, SNRIs, and tricyclic antidepressants.[43,47,50] Conversely, bupropion, an antidepressant devoid of serotonergic effect did not aggravate the symptoms of preexisting RLS in a controlled trial, and there was a trend suggesting a possible beneficial effect, yet to be confirmed.
Pathogenesis. The pathogenesis of RLS is poorly understood. It is generally agreed that dopamine and iron availability in the brain modulate emergence of symptoms, while dopaminergic agonists, iron, but also opioid drugs, alleviate the symptoms. Genetic factors seem critical.[45,52] How SSRIs could aggravate or induce RLS in some subjects is unknown. However, a SPECT study showed that in a group of patients with RLS, the severity of symptoms increased as the availability of serotonin transporter decreased. The authors therefore suggested that an increase in serotonergic transmission might exacerbate RLS, possibly via modulations on striatal dopaminergic transmission and on the spinal motor and sensory neurons.
Periodic limb movements [PLM] are characterized by brief 0.5–10 s stereotyped lower limb movements during sleep, which typically occur at 20–40 s intervals. PLM can result in electroencephalographic arousals or awakenings and contribute to insomnia, or excessive daytime sleepiness. PLM may be associated with restless legs syndrome. An association between the use of fluoxetine and/or myoclonic jerks during non-REM sleep was described in the 1990s. PLM was described with the use of fluoxetine in depressed patients and soon after with venlafaxine in normal volunteers. In a large case-control study, Yang et al. found that in patients referred for polysomnography (mainly to rule out sleep apnea), the use of venlafaxine and SSRI, and the risk of PLM was increased. An increased risk of PLM in subjects receiving SSRI has also been demonstrated in children. The risk of PLM seems to be independent of the type of SSRI, although differences in severity might exist.[57,59] Other antidepressants involved include venlafaxine and mirtazapine.