You are here

Do SSRI Antidepressants Increase The Risk of Extrapyramidal Side Effects In Patients Taking Antipsychotics?

Matthew Allsbrook PharmD, MS
Brant E. Fries PhD
Kristina L. Szafara PhD
Randolph E. Regal PharmD


Depression, a common psychiatric disorder, is the leading cause of disability in the United States.1 Due to SSRIs’ apparent efficacy and lack of major side effects,2 the prescribing of SSRIs is heavily favored over the older tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs). Indeed, current guidelines suggest the use of SSRIs and/or serotonin–norepinephrine reuptake inhibitors (SNRIs) as first-line therapy for the treatment of major depression.3

The commonly known side effects of SSRIs include transient gastrointestinal effects, weight gain and/or weight loss, sexual dysfunction, sleep disturbances, hypomania, the syndrome of inappropriate antidiuretic hormone secretion (SIADH), and movement disorders.4 New cases of movement disorders associated with SSRIs are difficult to identify due to the low numbers of patients involved in trials. Further, a review of the literature for antidepressant-induced EPS indicates that EPS can occur with various classes of antidepressants, are not dose-related, and can occur with short-term and long-term use.5 Prescribers are advised to monitor patients who might be at a higher risk for developing EPS while taking SSRIs.6 However, some studies support the notion of an elevated risk level when SSRIs are used concurrently with antipsychotics.7

The rationale for the development of SSRIs arose from the observation that decreased brain levels of the neurotransmitter serotonin may cause depressive symptoms, and that blockade of a specific transporter for serotonin might therefore exert an antidepressant effect.8 SSRIs’ pharmacological effect results from inhibition of the presynaptic serotonin transporter at the terminal ends of neurons, thereby increasing the concentration of serotonin. It is theorized that serotonin may inhibit the brain’s dopaminergic system, thus causing a reduction in dopamine activity. Reduced dopamine activity, which occurs as a result of the antipsychotics’ dopamine antagonism properties, is postulated to cause EPS-like effects such as parkin-sonian disorders, postural instability, and akathisia. Tardive dyskinesia is believed to be associated with hypersensitivity of post-synaptic dopaminergic receptors that may arise following the chronic use of medications that decrease dopaminergic transmission.9

An understanding of the mechanism of dopamine antagonism in regard to EPS demonstrates how SSRI use may result in extrapyramidal side effects. With the use of conventional antipsychotics, blockade of dopamine transmission in the mesolimbic dopaminergic pathway provides the treatment of the positive symptoms associated with schizophrenia. However, this indiscriminate blockade also results in the inhibition of dopamine transmission in the mesocortical pathway, which is responsible for the negative symptoms and cognitive effects associated with schizophrenia. A decrease in dopamine in the nigrostriatal and tuberoinfundibular pathways gives rise to the side effects associated with antipsychotic use. Dopamine normally inhibits acetylcholine, but with dopamine blocked in this pathway, acetylcholine becomes hyperactive and can lead to effects such as EPS.9

Although all SSRIs have the class effect of potentiating serotonin release, and thus indirectly antagonizing dopamine, individual SSRIs may exert profiles that vary somewhat. For example, when compared with other SSRIs, sertraline causes potent inhibition of the dopamine reuptake transporter as well as the serotonin reuptake transporter. Therefore, sertraline may theoretically be associated with a lower risk of movement disorders than other SSRIs.2

In a 1997 study of depressed geriatric patients, 6% of those receiving an SSRI for depression experienced EPS, such as akathisia, resting tremor, cogwheel rigidity, and bradykinesia.10 In a 1995 study of 5,555 patients taking fluoxetine, 15 (0.3%) reported the emergence of EPS; of these 15, 12 improved either partially or completely after discontinuation of fluoxetine.11 The review paper by Lane showed that the administration of a specific SSRI alone was associated with EPS; however, these reports were also associated with known predisposing factors to parkinsonism, such as Parkinson’s disease, brain damage, and previous use of antidopaminergic therapy.8 The frequency of EPS with use of SSRI antidepressants is estimated to be one in 1,000 or less.11 A review of the literature on EPS incidence with SSRI use found that parkinsonism accounted for the greatest percentage of SSRI-associated EPS (49%), followed by dystonia (27%).5 It is interesting to note that this study put akathisia frequency associated with SSRIs at 2%, far lower than previous reports on extrapyramidal side effects.12

The University of Michigan project using the InterRAI MH Assessment resulted in comprehensive surveillance of the adult inpatient psychiatric population, providing data to examine associations between SSRI use and EPS. The InterRAI MH includes a broad assessment of the characteristics of inpatient psychiatric hospital patients. In the University of Michigan study, a list of currently prescribed drugs was also recorded. From this drug list, it was possible to identify patients taking a variety of antidepressants (Table 1). These included tricyclic antidepressants (imipramine, clomipramine, amitriptyline, doxepin), SSRIs (fluoxetine, citalopram, paroxetine, sertraline, fluvoxamine, escitalopram), SNRIs (venlafaxine, duloxetine, desvenlafaxine), and others (trazodone, nefazodone, mirtazap-ine, bupropion). Each drug class has different characteristics regarding affinity for certain neurotransmitter transporters. A breakdown of the incidence of EPS with respect to the use of these agents should provide perspective regarding the relative association of SSRIs with movement disorders. This information could suggest pharmacotherapeutic interventions concerning antidepressant options to maximize therapeutic effect and minimize adverse effects. While some articles have analyzed the incidence of EPS with antidepressant use, the body of evidence is lacking in understanding whether concomitant use with antipsychotics exacerbates such incidence.


This is a retrospective cohort study of cross-sectional data collected from the State of Michigan’s four adult mental health hospitals. The data were collected between May 2010 and October 2010 from Caro Center, the Center for Forensic Psychiatry, Walter Reuther Psychiatric Hospital, and Kalamazoo Psychiatric Hospital. Every patient in each hospital on the target dates was assessed; if the patient was discharged or died before the assessment could be performed, the next patient placed in the same bed was used in the study if he or she met inclusion criteria. The Michigan Public Health Institute collected the data using the InterRAI MH.13 This comprehensive tool provides reliable measurement of patient factors in several areas of functional and clinical significance. Assessors use all available sources of data (the patient, the medical record, direct observation, staff reports, etc.) to determine the appropriate information for each item on the instrument.

To understand the connection (or lack thereof) between SSRIs and the incidence of EPS, we examined the data set from three groups, a total of 693 patients. The first cohort consisted of all patients on antipsychotic drugs who were also taking an SSRI antidepressant; the second cohort consisted of all patients on antipsychotic drugs who are not taking any antidepressant; and the third cohort consisted of all patients on antipsychotic drugs who were taking a non-SSRI antidepressant.

We analyzed the scoring for the seven EPS included on the InterRAI MH for each patient in one of the three cohorts. Each extrapyramidal side effect was scored as a 0 (no) or 1 (yes). To compare the overall responses between the cohorts, a mean EPS sum score was calculated as the total number of EPS symptoms reported in the group divided by the number of patients in that group. Thus, if every patient in a group had exactly one of the seven symptoms, the score would be 1.0.

Inclusion and Exclusion Criteria

Any adult inpatient of the four state mental health institutions named previously who was taking an antipsychotic medication on a daily or regularly scheduled basis was eligible for inclusion. Patients were excluded if there were no data for them regarding antipsychotic use. Patients who were listed as taking antipsychotics as needed were also excluded.


Since we compared more than two cohorts to identify a statistical difference in EPS frequency, we performed an analysis of variance (ANOVA) statistical test with data presented in tabular format showing the number of measurements and the response recorded for each symptom. The F critical value was obtained from the F Distribution Table of the University of California at Los Angeles Department of Statistics.14 The P value was calculated using GraphPad Software QuickCalcs.15 To calculate the statistical difference in categorical variables (i.e., gender), the chi-square test was used.


Table 2 compares the three cohorts. The mean weights and ages within each group were similar, but Cohort 1 had a higher percentage of females than the other two cohorts. The differences in the mean antipsychotic defined daily dose (DDD) in Cohorts 1, 2, and 3 were not statistically significant (P = 0.837). All three cohorts had similar mean EPS sum scores of 0.22, 0.23, and 0.21, respectively.

The most common extrapyramidal symptom observed was tremor, which was consistently seen in 8.6% to 8.7% of patients in each of the three cohorts. Akathisias were the second most common EPS observed; they were higher in the Cohort 1 antipsychotic-plus-SSRI group compared with the other two groups (7.38% versus 3.4% [Cohort 2] and 1.94% [Cohort 3]). Dyskinesias, the third most common EPS, were highest in Cohort 3 (5.83%) versus Cohort 2 (2.95%) and Cohort 1 (2.01%). This was followed by dystonias and slow shifting gait, which were seen at rates between 0 and about 2%. The final two symptoms, bradykinesia and rigidity, were the least frequent, with rates between 0 and 0.68%. The overall average EPS rate among all three cohorts ranged from 2.64% to 2.87% within the three groups, but the differences were not significant (F2,18 = 0.01; P < 0.9901) (Table 3).


In a cross-sectional, single-point assessment using the InterRAI MH in relatively young, institutionalized psychiatric patients taking antipsychotics, the incidence of EPS in those taking SSRIs did not appear to be greater than in those using antipsychotics alone. Mean EPS sum scores in each of the three cohorts were similar at 0.22, 0.23, and 0.21, respectively. When averaging the incidence of the seven listed EPS, all three groups had an average EPS rate of just under 3% (Table 3).

Akathisias and tremors were the most commonly reported EPS. While the cohort using SSRIs had an appreciably higher rate of akathisias than the other two groups (7.38% versus 3.4% and 1.94%), tremor rates were not higher in the SSRI population; this symptom was reported in just over 8.6% of patients in each group. The tremor rate in this study is very similar to the tremor rate of citalopram, based on package insert data.16 Previous reports have shown that akathisia accounts for 45% of EPS associated with SSRIs, followed by dystonia at 28%.12 However, in a review of the literature, the incidence of akathisia has been portrayed at just 2% of adverse events.5 Some SSRI medications have been associated with akathisia and related symptoms of restlessness.11 Perhaps in this and earlier studies the subjective observation of akathisias was not adequately differentiated from the motor restlessness/agitation sometimes seen with SSRIs. This may be especially true among the most “activating” agents, such as fluoxetine.

Important limitations of this study could have affected the results substantially. Most important, it was a cross-sectional study done at a single point in time during a patient’s stay at an institution. The InterRAI MH instrument has 19 clinical assessment protocols (CAPs) that are “triggered” for a patient based on responses to certain items. The trigger may indicate that further care and assessment are needed in a particular area. For this study population, the most commonly triggered CAP was “medication management and adherence” (67.8% of forensic patients and 82.3% of nonforensic patients). While the adherence rate was not captured for each patient, the need to assess adherence may raise concerns about whether medications were being taken appropriately. The InterRAI MH specifically asked about EPS “within the last three days” from the date of assessment. Also, since EPS are related both to cumulative time on a drug as well as daily dose, we have no way of knowing how long the patient had been using a given medication regimen on the day he or she was assessed. It is well established that EPS are dose-related.

In another confounder inherent with this study, we defined mean antipsychotic DDDs within each group and found those numbers were very similar between Cohorts 1 and 2, but we did not characterize the type of antipsychotic medications used within each cohort, which could have great importance. For instance, if the non-SSRI group happened to have a higher incidence of use of the older, typical antipsychotics that have a higher proclivity for EPS (such as the potent typical anti-psychotic haloperidol), that could blunt the SSRI contribution to EPS. We are also uncertain why the antipsychotic DDDs in Cohort 3 were so much lower than those in Cohorts 1 and 2.

Possibly further obscuring differences between Cohorts 1 and 3, medications in Cohort 3 (non-SSRI antidepressants) included venlafaxine, desvenlafaxine, and duloxetine. These medications are classified as SNRIs. In addition to inhibiting norepinephrine reuptake from the synaptic cleft, these drugs also inhibit the reuptake of serotonin, a pharmacological property shared with the SSRI medications. Although these SNRIs were not included in Cohort 1 of the analysis, it is known that these drugs are 10 to 30 times more selective for serotonin than norepinephrine.17 Further research should be conducted to fully elucidate the differences in efficacy and safety of SNRI versus SSRI therapy when it comes to long-term effects. Differences should be studied between individual drugs rather than using pooled data about SSRIs versus SNRIs.

Future directions for assessing the correlation of EPS with antipsychotic and SSRI use could include determining the duration of concomitant therapy, the specific SSRIs and anti-psychotics used, and the doses of the medications the patients were taking when EPS were found to occur. A prospective study observing similar parameters could be devised to determine if EPS are occurring more frequently in those taking SSRIs than those who are not.


With new Food and Drug Administration labeling of certain atypical antipsychotics as adjuvants for the treatment of bipolar and major depressive disorders, the future may see an increase in the concomitant use of SSRIs and antipsychotics among younger and more ambulatory populations. This study in mostly nongeriatric psychiatric inpatients taking antipsychotics did not demonstrate an increased risk of extrapyramidal side effects when SSRI antidepressants were used simultaneously.


Psychoactive Drugs Used Within the Three Cohorts



  • ° Chlorpromazine
  • ° Fluphenazine
  • ° Perphenazine
  • ° Trifluoperazine


  • ° Haloperidol
  • Indole derivatives

    • ° Molindone
    • ° Ziprasidone

    Diazepines, oxazepines, thiazepines, oxepines

  • ° Loxapine
  • ° Clozapine
  • ° Olanzapine
  • ° Quetiapine
  • ° Asenapine
  • Benzioxazoles, mood stabilizers, quinolone derivate

  • ° Risperidone
  • ° Paliperidone
  • ° Lithium
  • ° Aripiprazole
  • Antidepressants


    • ° Imipramine
    • ° Clomipramine
    • ° Amitriptyline
    • ° Doxepin

    Selective Serotonin Reuptake Inhibitors

  • ° Fluoxetine
  • ° Citalopram
  • ° Paroxetine
  • ° Sertraline
  • ° Fluvoxamine
  • ° Escitalopram
  • ° Trazodone
  • ° Nefazodone
  • Serotonin–Norepinephrine Reuptake Inhibitors

  • ° Venlafaxine
  • ° Duloxetine
  • ° Desvenlafaxine
  • Aminoketones

  • ° Bupropion
  • Characteristics of the Three Cohorts

    Antipsychotic + SSRI, n = 149 Antipsychotic, n = 441 Antipsychotic + non-SSRI, n = 103 P value
    Mean age in years (SD) 42.6 (12.8) 45.1 (13.5) 44.7 (14.7) P = 0.149
    Male, n (%) 97 (65.3) 333 (75.4) 82 (79.8) P = 0.021
    Mean weight in kilograms (SD) 91.8 (22.0) 90.2 (20.4) 91.7 (27.6) P = 0.667
    Mean antipsychotic DDD in milligrams (SD) 12.5 (17.5) 11.3 (10.8) 8.7 (5.9) P = 0.836
    Mean EPS sum score (SD) 0.22 (0.5) 0.23 (0.52) 0.21 (0.51) P = 0.944

    DDD = defined daily dose; EPS = extrapyramidal symptoms; SD = standard deviation; SSRI = selective serotonin reuptake inhibitor

    Number and Percentage of EPS Reports in Each Cohort

    Antipsychotic + SSRI, n = 149 Antipsychotic, n = 441 Antipsychotic + Non-SSRI Antidepressant, n = 103
    Akathisia, n (%) 11 (7.38) 15 (3.40) 2 (1.94)
    Dyskinesia, n (%) 3 (2.01) 13 (2.95) 6 (5.83)
    Tremor, n (%) 13 (8.72) 38 (8.62) 9 (8.74)
    Bradykinesia, n (%) 1 (0.67) 3 (0.68) 0 (0)
    Rigidity, n (%) 0 (0) 1 (0.23) 0 (0)
    Dystonia, n (%) 0 (0) 7 (1.59) 2 (1.94)
    Slow shift gait, n (%) 2 (1.34) 9 (2.04) 0 (0)
    Sum of percentages 20.12 19.51 18.45
    Mean 2.87 2.79 2.64
    F critical value = 2.62395a
    P < 0.9901; accept the null hypothesisb

    aThe F critical value was obtained from the F Distribution Table from the UCLA Department of Statistics.14

    bThe P value was calculated from GraphPad Software QuickCalcs.15

    EPS = extrapyramidal symptoms; SSRI = selective serotonin reuptake inhibitor

    Author bio: 
    Dr. Allsbrook is a Pharmacy Resident at the University of Virginia Health System in Charlottesville, Virginia. Dr. Fries is a Professor of Health Management and Policy in the School of Public Health Research and a Professor in the Geriatric Center of the School of Medicine at the University of Michigan (UM), and Chief of Health Systems Research at the Ann Arbor Veterans Affairs Healthcare Center, both in Ann Arbor, Michigan. Dr. Szafara is a Research Fellow at the UM Institute of Gerontology. Dr. Regal is a Clinical Associate Professor of Pharmacy in the UM College of Pharmacy and a Clinical Pharmacist in Adult Internal Medicine in the UM Health System Department of Pharmacy Services in Ann Arbor.


    1. Murray CJ, Atkinson C, Bhalla K, et al. The state of U.S. health, 1990–2010: burden of diseases, injuries, and risk factors. JAMA 2013;310;(6):591–608.
    2. Gerber PE, Lynd LD. Selective serotonin-reuptake inhibitor-induced movement disorders. Ann Pharmacother 1998;32;(6):692–698.
    3. Ciechanowski P. Patient information: depression treatment options for adults (beyond the basics). UpToDate. Available at: Accessed October 21, 2015.
    4. Ferguson JM. SSRI antidepressant medications: adverse effects and tolerability. Prim Care Companion J Clin Psychiatry 2001;3;(1):22–27.
    5. Madhusoodanan S, Alexeenko L, Sanders R, et al. Extrapyramidal symptoms associated with antidepressants—a review of the literature and an analysis of spontaneous reports. Ann Clin Psychiatry 2010;22;(3):148–156.
    6. Hawthorne JM, Caley CF. Extrapyramidal reactions associated with serotonergic antidepressants. Ann Pharmacother 2015;49;(10):1136–1152.
    7. Schillevoort I, van Puijenbroek EP, de Boer A, et al. Extrapyra-midal syndromes associated with selective serotonin reuptake inhibitors: A case-control study using spontaneous reports. Int Clin Psychopharmacol 2002;17;(2):75–79.
    8. Lane RM. SSRI-induced extrapyramidal side-effects and akathisia: implications for treatment. J Psychopharmacol 1998;12;(2):192–214.
    9. Stahl S. Stahl’s essential psychopharmacology online. Available at: Accessed October 30, 2015.
    10. Gormley N, Watters L, Lawlor B. Extrapyramidal side effects in elderly patients exposed to selective serotonin reuptake inhibitors. Human Psychopharmacology: Clinical and Experimental 1997;12;(2):139–143.
    11. Coulter DM, Pillans PI. Fluoxetine and extrapyramidal side effects. Am J Psychiatry 1995;152;(1):122–125.
    12. Leo RJ. Movement disorders associated with the serotonin selective reuptake inhibitors. J Clin Psychiatry 1996;57;(10):449–454.
    13. Fries BE, Schmorrow A, D’Souza J. Mental health assessment project Report to the Michigan Department of Community Health University of Michigan. 2011;
    14. Statistics Online Computational Resource, University of California at Los Angeles. F distribution tables. Available at: Accessed November 13, 2013.
    15. QuickCalcs GraphPad Software. P value calculator. Available at: Accessed November 13, 2013.
    16. Citalopram package insert St Louis, Missouri: Forest Pharmaceuticals. 2014;
    17. Bradley AJ, Lenox-Smith AJ. Does adding noradrenaline reuptake inhibition to selective serotonin reuptake inhibition improve efficacy in patients with depression? A systematic review of meta-analyses and large randomised pragmatic trials. J Psychopharmacol 2013;27;(8):740–758.