The selective serotonin reuptake inhibitors are employed to treat an array of disorders, including depression, social phobia, panic disorder, posttraumatic stress disorder and obsessive-compulsive disorder. With generally fewer side effects and better tolerability than tricyclic antidepressants and monoamine oxidase inhibitors, the SSRIs have become the most widely prescribed antidepressants in the United States.
Owing to this popularity, adverse drug events previously unappreciated in premarketing clinical trials have gained increasing attention. The most notable of these has been the effect on sexual functioning. Less well-known, yet clinically significant, are movement disturbances that can occur with SSRI use. This should be of concern to clinicians, as movement disorders are uncomfortable, can adversely impact compliance, and can undermine the alliance between clinician and patient.
In a review of the literature, 71 cases of SSRI-associated extrapyramidal symptoms (EPS) were found (Leo, 1996) (Table). Akathisia was most common, followed by dystonia, parkinsonism and tardive dyskinesia-like states. In addition, 16 cases of worsening parkinsonism were found in patients with pre-existing Parkinson's disease.
Any review of case reports is subject to inherent limitations. First, some case reports are limited by ambiguous descriptions that make movement disorders difficult to distinguish from other psychiatric disorders or other potential SSRI side effects. This is most notable in case reports of akathisia attributed to SSRI use. At times, it becomes difficult to differentiate akathisia from anxiety or jitteriness (Amsterdam et al., 1994; Maany and Dhopesh, 1990). Mistaking comorbid anxiety for SSRI-associated akathisia may delay or interfere with the appropriate treatment of the patient's anxiety disorder.
Second, in only limited numbers of case reports was the SSRI the sole agent administered. Often, the coadministered medications were also capable of producing EPS. Since the movement disturbances did not appear to occur until after addition of the SSRI, it is possible that pharmacokinetic interactions may have occurred -- leading to increased bioavailability of the SSRI, the concurrently administered drug or both -- thus leading to the emergence of the dyskinesia (Leo, 1996). For example, when coadministered with paroxetine (Paxil), serum perphenazine (Trilafon) levels were significantly increased and accompanied by increased rates of akathisia and parkinsonism (Ozdemir et al., 1997). In addition, medications that normally do not produce EPS may, when combined with an SSRI, predispose patients to dyskinesia (Leo et al., 1995).
Third, in several reports, the presence of pre-existing neurologic disease was evident. Conditions such as head trauma (Coulter and Pillans, 1995) or Parkinson's disease (Jimenez-Jimenez et al., 1994; Steur, 1993) can account for the development or emergence of movement disorders. It is also unclear how many of the patients reported in the literature had undiagnosed or overlooked neurologic conditions coincidentally manifesting around the time of SSRI exposure.
Fourth, reports of SSRI rechallenges are rare (Coulter and Pillans, 1995; Reccoppa et al., 1990). Despite the presence of confounding variables, re-exposure to the SSRI would establish a stronger causal relationship between SSRI treatment and the emergence of movement disorders. These factors limit the ability to draw firm conclusions about a causal relationship between SSRI use and the emergence and/or exacerbation of movement disorders.
A simple schematic model of movement disorders is provided in the Figure. Normally, voluntary movements arise when corticospinal tracts generate impulses to anterior horn cells in the spinal cord, modulated by basal ganglia output. Gamma-aminobutyric acid (GABA) from the basal ganglia is inhibitory, refining the corticospinal tract activation. Deficiency of GABA outflow, e.g., Huntington's disease, is characterized by herky-jerky and extraneous movements.
The GABA outflow from the basal ganglia is, in turn, controlled by the balance between two neurotransmitter systems, i.e., dopamine (DA) arising from the substantia nigra and acetylcholine (ACh). The latter have opposing influences on the activity and, therefore, the outflow from the basal ganglia. Disturbance in the balance between ACh and DA alters the net outflow from the basal ganglia, producing movement disturbances (Figure). Thus, idiopathic parkinsonism arises from cell loss in the substantia nigra, reducing the amount of inhibitory DA input to the basal ganglia. The ACh, now relatively less opposed, stimulates the basal ganglia, increasing the inhibitory output to the corticospinal tracts, producing the bradykinesia, rigidity, mask-like facies, shuffling gait and other symptoms characteristic of Parkinson's disease.
The (Figure) also depicts, therefore, the effect of conventional high-potency antipsychotics in producing dyskinesia. These agents, such as haloperidol (Haldol), bind DA receptors in the basal ganglia, thereby preventing access to DA arising from the substantia nigra. The net effect, like that in Parkinson's disease, is unopposed excitatory input from ACh-containing neurons. Consequently, treatment consists of the addition of an anticholinergic agent like benztropine (Cogentin) restoring balance between DA and ACh and re-establishing the normal inhibitory outflow from the basal ganglia.
Serotonin (5-HT)-containing raphe nuclei extend diffuse interconnections to the DA-rich substantia nigra (Dray, 1981). Neurophysiologic and electric stimulation studies demonstrated that the 5-HT released by the raphe nuclei inhibit striatal neurons, an effect which is reversed by 5-HT antagonists (Davies and Tongroach, 1978). Thus, it is plausible that inhibitors of neuronal 5-HT reuptake, by increasing the availability of 5-HT, might be expected to produce an effect similar to that of DA-blocking agents (Figure). In fact, high doses of fluoxetine (Prozac) have been shown to inhibit DA synthesis in the forebrain, hippocampus and portions of the basal ganglia, specifically the caudate-putamen (Baldessarini and Marsh, 1990). Hence, it can be expected that movement disturbances might arise from SSRI use.
The physiologic processes underlying the development of akathisia may involve the interaction of serotonergic and DA pathways innervating mesolimbic systems. While not depicted here, it is suggested that the inhibitory input to these DA pathways produces the overt and covert restlessness characteristic of akathisia. Noradrenergic mechanisms may also be involved.
The mechanisms underlying SSRI-induced movement disorders are likely to be more complex than has been suggested above. A few case reports suggest improvement of parkinsonism and dystonia with the addition of SSRIs (Durif et al., 1995; Keppel Hesselink, 1993; Meerwaldt, 1986). It is possible that other interconnections between 5-HT containing innervations with those of GABA and ACh may contribute to the development of movement disturbances (Fibiger and Lloyd, 1984; Schreiber and Pick, 1995). However, these interconnections have yet to be clarified.
Just how SSRIs induce EPS and other movement disorders in some patients, but potentially improve parkinsonism and dystonia in others, remains unclear. Furthermore, were the mechanisms underlying SSRI-induced movement disorders as simple as illustrated here, the expectation would be that SSRI-induced movement disorders would be common. In fact, the rates of such movement disturbances remain quite low.
Who Is at Risk?
It is possible that some patients are more vulnerable to SSRI-induced movement disorders than are others. Included in the higher-risk category are a) the elderly; b) those exposed to high levels of SSRIs (due to high doses or altered metabolism due to drug interactions); c) patients with concurrent neuroleptic exposure; and d) patients with compromised nigro-striatal functioning. Elder patients may be susceptible to neuronal loss, rendering them vulnerable to the effects of enhanced 5-HT input to nigro-striatal pathways. In addition, due to decreased hepatic functioning, they may be vulnerable by virtue of increased levels of exposure to administered SSRIs. Clearly, neuroleptic exposure increases one's risk of movement disorders, and such exposure may increase the vulnerability of patients who are simultaneously administered SSRIs.
One can only speculate on the influence of gender. While reports featuring female patients who developed dyskinesia associated with SSRI use outnumber those featuring males (Table), one cannot assume that females are more vulnerable. On the one hand, the gender differences observed may merely reflect another trend, i.e., the prevalence of depression is greater among females and more females than males seek treatment for depression (Weissman and Klerman, 1977). In fact, it may be possible that males may be more susceptible than females to dyskinesia associated with SSRI use. Among SSRI-treated patients in a New Zealand medication monitoring program, females (n=3,539) exceeded males (n=1,917) (Coulter and Pillans, 1995). Nevertheless, the proportion of males developing movement disorders (n=8, 0.42%) exceeded the proportion of females (n=7, 0.2%) who did so.
While a majority of cases of SSRI-induced movement disorders involved fluoxetine, at the time of my initial review, fluoxetine had exceeded the other SSRIs in sales and had been available longer than the others. Consequently, the numbers of reports involving fluoxetine may simply have been an artifact of these trends.
On the other hand, there are differences among the effects of SSRIs on DA-reuptake inhibition. For example, sertraline (Zoloft) exhibits a direct augmenting effect on DA-reuptake inhibition (Koe et al., 1983); inhibitory serotonergic input to dopaminergic systems would be mitigated by such direct augmentation. Paroxetine and fluoxetine have lower potencies than sertraline for DA-reuptake inhibition in vitro (Richelson, 1994). Paroxetine also has anticholinergic properties in vitro, which may contribute to reducing the likelihood of EPS as compared to some of the other SSRIs.
Treatment Options, Conclusion
The most prudent treatment measures might be dose reduction or discontinuation of the SSRI, switching to an alternate antidepressant, and/or reducing the coadministered medications that may have led to drug interactions and potentiation of the movement disturbances after SSRI administration. Other interventions employed to mitigate dyskinesia associated with SSRI use are summarized in the Table.
Given that patient exposure to the aforementioned SSRIs is currently estimated to exceed 85 million, movement disorders associated with SSRI use are rare. Certain patients may be more vulnerable to the emergence of dyskinesia after SSRI treatment, e.g., the elderly or those with neurological insults. Clinicians may need to pay particular attention to patients treated with SSRIs who require multiple medications for co-existing medical conditions or complicating psychiatric symptoms. Because pharmacokinetic interactions may occur that render patients susceptible to dyskinesia, patients should be examined frequently for signs of an emerging movement disorder.
1.Amsterdam JD, Hornig-Rohan M, Maislin G (1994), Efficacy of alprazolam in reducing fluoxetine-induced jitteriness in patients with major depression. J Clin Psychiatry 55(9):394-400.
Baldessarini RJ, Marsh E (1990), Fluoxetine and side effects. Arch Gen Psychiatry 47(2):191-192 [letter].
Coulter DM, Pillans PI (1995), Fluoxetine and extrapyramidal side effects. Am J Psychiatry 152(1):122-125 [see comment].
Davies J, Tongroach P (1978), Neuropharmacological studies on the nigro-striatal and raphe-striatal system in the rat. Eur J Pharmacol 51(2):91-100.
Dray A (1981), Serotonin in the basal ganglia: Functions and interactions with other neural pathways. J Physiol (Paris) 77:393-403.
Durif F, Vidailhet M, Bonnet AM et al. (1995), Levodopa-induced dyskinesias are improved by fluoxetine. Neurology 45(10):1855-1858.
Fibiger HC, Lloyd KG (1984), Neurobiological substrates of tardive dyskinesia: the GABA hypothesis. Trends Neurosci 7:462-464.
Jimenez-Jimenez FJ, Tejeiro J, Martinez-Junquera G et al. (1994), Parkinsonism exacerbated by paroxetine. Neurology 44(12):2406.
Keppel Hesselink JM (1993), Serotonin, depression, and PD. Neurology 43(8):1624-1625 [letter; see comment].
Koe BK, Weissman A, Welch WM, Browne RG (1983), Sertraline, 1S,4S-N-methyl-4-(3,4-dichlorophenyl)-1,2,3,4-tetrahydro-1-naphthylamine, a new uptake inhibitor with selectivity for serotonin. J Pharmacol Exp Ther 226(3):686-700.
Leo RJ (1996), Movement disorders associated with the serotonin selective reuptake inhibitors. J Clin Psychiatry 57(10):449-454 [see comments].
Leo RJ, Lichter DG, Hershey LA (1995), Parkinsonism associated with fluoxetine and cimetidine: a case report. J Geriatr Psychiatry Neurol 8(4):231-233.
Maany I, Dhopesh V (1990), Akathisia and fluoxetine. J Clin Psychiatry 51(5):210-212 [letter; see comment].
Meerwaldt JD (1986), Treatment of hypokinetic rigid syndrome with fluvoxamine maleate. Lancet 1(8487):977-978 [letter].
Ozdemir V, Naranjo CA, Herrmann N et al. (1997), Paroxetine potentiates the central nervous system side effects of perphenazine: contribution of cytochrome P4502D6 inhibition in vivo. Clin Pharmacol Ther 62(3):334-347.
Reccoppa L, Welch WA, Ware MR (1990), Acute dystonia and fluoxetine. J Clin Psychiatry 51(11):487 [letter; see comment].
Richelson E (1994), The pharmacology of antidepressants at the synapse: focus on newer compounds. J Clin Psychiatry 55(suppl A):34-39 [see discussion pp40-41, 98-100].
Schreiber S, Pick CG (1995), Fluoxetine for blepharospasm: interaction of serotonin and dopamine. J Nerv Ment Dis 183(11):719-721.
Steur EN (1993), Increase of Parkinson disability after fluoxetine medication. Neurology 43(1):211-213.
Weissman MM, Klerman GL (1977), Sex differences and the epidemiology of depression. Arch Gen Psychiatry 34(1):98-111.