Antidepressants are one of the most prescribed classes of medications. A number of case reports have linked these drugs to extrapyramidal symptoms (EPSs), but no large epidemiologic study to date has examined this association. We sought to quantify the association of EPSs with different antidepressants by undertaking a large pharmacoepidemiologic study.
A nested case-control study was conducted using a large health claims database in the United States from June 2006 to December 2015. Subjects with a diagnosis of primary Parkinson disease and those who received prescriptions of levodopa, ropinirole, pramipexole, domperidone, metoclopramide, entacapone, benztropine, selegiline, rasagiline, diphenhydramine, trihexyphenidyl, typical and atypical antipsychotics, and tricyclic antidepressants were excluded. Cases were followed to the first billing code for an extrapyramidal event or last date of enrollment in the cohort. For each case, 10 control subjects were matched by follow-up time, calendar time, and age through density-based sampling. Rate ratios were computed using conditional logistic regression adjusting for other covariates.
We identified 3,838 subjects with EPSs compared with 38,380 age-matched control subjects. Rate ratios with respect to EPSs were as follows: duloxetine, 5.68 (95% confidence interval [CI], 4.29–7.53); mirtazapine, 3.78 (95% CI, 1.71–8.32); citalopram, 3.47 (95% CI, 2.68–4.50); escitalopram, 3.23 (95% CI, 2.44–4.26); paroxetine, 3.07 (95% CI, 2.15–4.40); sertraline, 2.57 (95% CI, 2.02–3.28); venlafaxine, 2.37 (95% CI, 1.71–3.29); bupropion, 2.31 (95% CI, 1.67–3.21); and fluoxetine, 2.03 (95% CI, 1.48–2.78).
This observational study demonstrates a harmful association between the incidence of Parkinson disease or associated EPSs and use of the antidepressants duloxetine, mirtazapine, citalopram, escitalopram, paroxetine, sertraline, venlafaxine, bupropion, and fluoxetine.