When Discontinuing SSRI Antidepressants Is a Challenge: Management Tips

Selective serotonin reuptake inhibitors (SSRIs) are the most commonly prescribed class of antidepressant (1) and are recommended as one of the first-line treatments (2–4) for major depressive disorder, an often chronic or recurrent disorder that affects one in six adults during their lifetime (5). Since less than one-third of patients with major depression attain remission with an initial antidepressant trial (6), multiple sequential acute-phase (at least 6–8 weeks) trials are often needed for attainment of symptomatic remission (3, 7). Beyond the acute phase, changes to the patient’s antidepressant medication regimen may be warranted to address residual symptoms, persistent functional impairments, loss of initial therapeutic effect (antidepressant tachyphylaxis), occurrence or prevention of relapse or recurrence, side effects, affordability, or anticipated drug-drug interactions with medications for other medical conditions (8–10). Additionally, patients may discontinue their treatment because of concern about adverse effects (11), which are common with antidepressant medications (12). Clinicians may also recommend discontinuation of treatment for patients with no or few risk factors for relapse or recurrence or upon patient request, consistent with treatment guidelines of 6 to 9 months of continuation-phase treatment after symptomatic remission (2–4).

Treatment guidelines and the U.S. Food and Drug Administration prescribing labels (2–4, 13–17) caution that new-onset symptoms may occur on discontinuation of antidepressant medications and recommend gradual tapering to prevent such symptoms. Here, we discuss the varied symptomatic presentations associated with discontinuation of antidepressant medications and some potential mitigation strategies and tactics (Figure 1). We restricted the scope of this report to the use of SSRIs approved for treatment of major depressive disorder. To provide a fuller picture of the range of discontinuation symptoms, we have also included reports in which SSRIs were prescribed for indications other than major depressive disorder.

We have organized the discussion of discontinuation symptoms to answer the following clinically relevant questions:

1. What are discontinuation symptoms?

   1. How common are discontinuation symptoms?

   2. How do discontinuation symptoms present?

   3. What is the differential diagnosis for discontinuation symptoms?

   4. What is the pathophysiology of discontinuation symptoms?

2. When do discontinuation symptoms occur and how long do they last?

3. Who is at risk of discontinuation symptoms?

4. How are discontinuation symptoms distinct from relapse or recurrence?

5. What are the best strategies to manage discontinuation symptoms?

What are Discontinuation Symptoms?

Discontinuation symptoms are a constellation of symptoms that start soon after the discontinuation of SSRIs (symptomatic discontinuation), similar to other psychoactive medications and older antidepressants (18–20). In one of the first reports of symptomatic discontinuation involving SSRIs, Stoukides and Stoukides (21) reported development of acute dystonic reaction, hypertension, tachycardia, diaphoresis, and increased muscle tone in the extremities in a man who suddenly discontinued fluoxetine after a 6-month course. More cases of new-onset physical symptoms following SSRI discontinuation were reported in subsequent years, especially after SSRIs with shorter half-lives became available, such as sertraline and paroxetine (22–37). While early reports focused on physical symptoms associated with discontinuation, Bloch et al. (38) reported new-onset hypomanic symptoms following paroxetine discontinuation in two healthy men who were being treated for stuttering and had no history of major psychiatric disorders. Emergence of manic symptoms has also been reported with discontinuation of escitalopram (39). These reports resulted in recommendations that patients be assessed for SSRI discontinuation syndrome, which was defined as emergence of physical and neuropsychiatric symptoms within days to weeks after discontinuation of an SSRI that resulted in significant distress and were not accounted for by other causes (such as a general medical condition or concomitant medications) (40–42). While outside the scope of this review, one in eight cases of discontinuation symptoms with tramadol, a painkiller with serotonergic activity, resemble symptomatic SSRI discontinuation (with symptoms such as psychosis, confusion, anxiety/panic, and numbness and tingling in the extremities) and are distinct from classical opioid withdrawal (43). Similar symptomatic discontinuations occur with serotonin-norepinephrine reuptake inhibitors (SNRIs) (44–47).

How Common are Discontinuation Symptoms?

The true incidence of discontinuation symptoms is difficult to ascertain, because they are not systematically assessed in clinical practice and are typically not assessed in patients who drop out of care. On the basis of spontaneous adverse event reports for SSRIs, Price et al. (48) estimated discontinuation rates ranging from 0.002 (for fluoxetine) to 0.3 (for paroxetine) reports per 1,000 prescriptions. In a recent systematic review of clinical trials, Fava et al. (49) found that up to 40% of patients reported new-onset symptoms after abruptly discontinuing SSRIs. Paroxetine has the highest rates of symptomatic discontinuation (50, 51) as compared with other SSRIs, including sertraline (52, 53), citalopram (54), escitalopram (55), and fluoxetine (52, 53, 56–59). Although abrupt discontinuation of fluoxetine is generally well tolerated, case reports have documented the development of delirium after discontinuation (52, 60–63). Table 1 summarizes the findings of discontinuation symptoms reported for each SSRI that is approved for treatment of major depressive disorder.

How Do Discontinuation Symptoms Present?

Berber (64) proposed the mnemonic “FINISH” for discontinuation symptoms: “flu-like symptoms, insomnia, nausea, imbalance, sensory disturbances, and hyperarousal (anxiety/agitation).” For more than 60% of patients, discontinuation of SSRIs will not be associated with any significant symptoms. When present, discontinuation symptoms may be mild, although they can also be moderate to severe in intensity, and in some cases they affect multiple organ systems. To aid clinicians, we have devised a focused review of systems that includes the discontinuation signs and symptoms reported in literature (Table 2).

What is the Differential Diagnosis for Discontinuation Symptoms?

When considering symptoms due to SSRI discontinuation, it is prudent to rule out other causes for the symptoms. Causal general medical conditions include viral or bacterial infections of the respiratory or gastrointestinal tracts, foodborne illness from contaminated food or drinks, seasonal or other allergies, and stroke or other neurological conditions (65). Additionally, changes in concomitant medications, including discontinuation of other psychoactive drugs, may account for the presence of discontinuation symptoms. In cases of switching or cross-tapering of medications, where one medication is discontinued while another is initiated, discontinuation symptoms may be confused with treatment-emergent symptoms associated with depression, such as irritability, anxiety, panic, and mania or hypomania (66). Typically, a role for SSRI discontinuation is suspected on the basis of the timing (within days to a week after the medication is discontinued) and the types of symptoms (see Table 2).

What is the Pathophysiology of Discontinuation Symptoms?

A greater prevalence of symptoms after discontinuation of SSRIs with shorter half-lives suggests a role for reduced availability of serotonin in the central nervous system with decreasing levels of SSRIs (67). Across multiple SSRIs, Michelson et al. (53) found that greater reduction in serum levels of SSRIs and their active metabolites on discontinuation was associated with higher rates of discontinuation symptoms. Henry et al. (68) found that higher levels of paroxetine in the brain prior to discontinuation (despite the same oral dosage of 20 mg/day; N=4) were associated with higher levels of symptoms after discontinuation; no such association was seen for fluoxetine. In preclinical studies, chronic but not acute use of SSRIs has been associated with down-regulation of autoreceptors and serotonin transporter that may be sustained for weeks after discontinuation (69, 70). The modulation of serotonin receptors and transporters over several weeks of SSRI treatment may account for why symptoms do not emerge after discontinuing SSRIs taken for just 1–2 weeks (71).

Arguably, the strongest evidence for a pharmacological mechanism is inferred by rapid resolution of discontinuation symptoms by reinitiation of SSRI medication or, when symptoms emerge with a decrease in the dosage, by an increase in dosage. This is supported by animal studies in which discontinuation of chronic SSRI treatment has been associated with increased turnover of serotonin and reduced extracellular serotonin levels, along with an increased acoustic startle reflex, suggesting that reduced serotonin levels may account for anxiety-like discontinuation symptoms (72).

The multitude of symptom domains (Table 2) suggests that neurotransmission systems affected by SSRI discontinuation extend beyond serotonin. In fact, discontinuation symptoms have been partly attributed to increased glutamatergic neurotransmission, which is corrected by administration of N-methyl-d-aspartate receptor antagonist (73). Discontinuation of SSRIs has also been postulated to affect dopamine neurotransmission as well as the hypothalamic-pituitary-adrenal axis (67, 73).

When Do Discontinuation Symptoms Occur and How Long Do They Last?

Discontinuation symptoms typically start 1–10 days after abrupt discontinuation or marked dosage reduction of an SSRI (22, 27, 28, 56). With paroxetine, discontinuation symptoms may also occur if doses are skipped for a few days (29, 55). Discontinuation symptoms typically resolve spontaneously over 2–3 weeks (27, 32, 61). Reintroduction of another SSRI or the original SSRI results in improvement of symptoms within 2–3 days (28, 29, 34, 49). However, some reports have documented persistence of symptoms for up to 1 year (61).

Who is At Risk Of Discontinuation Symptoms?

Any patient who either discontinues or markedly reduces the antidepressant medication dosage is at risk of discontinuation symptoms. Antidepressant medications may need to be taken for at least 4–6 weeks before the development of discontinuation symptoms is of concern. In patients with premenstrual dysphoric disorder, discontinuation of SSRIs after up to 2 weeks of daily dosing was not associated with any increased symptoms compared with placebo (71). Patients who are taking the SSRI with the shortest half-life—paroxetine—are at the highest risk of discontinuation symptoms (52, 57). Factors associated with higher likelihood of discontinuation symptoms include earlier age at onset of depression (57), comorbid conditions such as panic disorder (25), and a past history of discontinuation symptoms. Factors not shown to predict development of discontinuation symptoms include duration of SSRI therapy beyond 4–6 weeks prior to discontinuation (52) and length of medication taper (49, 52).

How are Discontinuation Symptoms Distinct From Relapse or Recurrence?

The timing and types of symptoms distinguish discontinuation symptoms from relapse or recurrence. For most patients, the course of major depression is either chronic or recurrent. As expected with other chronic disorders, discontinuation of effective antidepressant treatment is often associated with either a return of the original depressive symptoms or further worsening of symptoms. Return of the original depressive symptoms after SSRI discontinuation can be rapid (hours to days), resembling a rebound phenomenon, which improves rapidly with SSRI reinitiation (74). This symptom trajectory differs from the more insidious return of depressive symptoms over weeks to months following a period of symptomatic remission (relapse) or recovery (recurrence) (75). Relapse or recurrence does occur in patients who had no or minimal discontinuation symptoms immediately after discontinuation of SSRIs. Thus, onset of symptoms affecting multiple organ systems immediately after discontinuation (1–10 days) of an SSRI is suggestive of discontinuation symptoms. Relapse or recurrence can be ascertained by comparing the discontinuation symptoms (affecting multiple systems) with the original depressive symptoms (which may include sad or depressed mood, anhedonia, impaired sleep, changes in weight or appetite, poor concentration or impaired decision making, fatigue or low energy, pessimism or excessive guilt, psychomotor agitation or retardation, and suicidal ideation). Additionally, while discontinuation symptoms typically resolve on reinitiation of SSRIs, attainment of remission after relapse or recurrence may take several weeks or may not occur at all.

What are the Best Strategies to Manage Discontinuation Symptoms?

Discontinuation symptoms can be managed by both prevention and treatment strategies. Preventive strategies include steps to be taken prior to onset of discontinuation symptoms. These include patient education regarding the need to continue antidepressant treatment and maintain adherence. Additionally, patients should be advised that inadvertently missing SSRI doses for a few days, especially with paroxetine, can result in discontinuation symptoms. As changes in ongoing SSRI treatment may be necessitated by partial response, nonresponse, tachyphylaxis, or intolerable side effects, the steps to minimize the risk of discontinuation symptoms include 1) increasing the dosage of the current SSRI to the maximal tolerated dosage; 2) switching to another SSRI or an SNRI; 3) using fluoxetine as a bridge to a nonserotonergic antidepressant, such as bupropion; 4) augmenting the SSRI with an atypical antipsychotic, lithium, buspirone, or liothyronine; 5) adding psychotherapy, light therapy, and/or exercise; and/or 6) consideration of treatments that are specific to treatment-resistant depression—repetitive transcranial magnetic stimulation, vagus nerve stimulation, electroconvulsive therapy, and intravenous ketamine. However, as depression is a chronic disorder, we recommend continued, potentially indefinite, treatment to reduce the risk of relapse or recurrence in patients whose depression is highly recurrent or chronic, is difficult to treat, and is comorbid with other psychiatric and medical conditions (2, 4). Adverse events may also be reduced or mitigated by use of alternative treatments (bupropion for sexual dysfunction, sedatives or cognitive-behavioral therapy for insomnia, exercise and diet modification for weight gain). Use of pretreatment biomarkers such as C-reactive protein may enable personalized selection of an initial antidepressant medication (76, 77). Further studies are needed to demonstrate the utility of biomarkers in reducing the need to switch treatments, which would diminish the risk of discontinuation symptoms. Combinatorial pharmacogenomic kits that test for genetic variants of several pharmacokinetic genes are available and have shown some promise in predicting treatment-emergent side effects (78). However, their utility in predicting risk of discontinuation symptoms remains unclear.

Treatment strategies to manage discontinuation symptoms depend on the consequences of SSRI discontinuation (Figure 1). Active surveillance with a focused review of systems (Table 2) should be undertaken whenever SSRIs are discontinued. Even if discontinuation symptoms develop, the active surveillance can be continued for milder symptoms. Using the measurement-based care paradigm, systematic assessment of frequency, intensity, and burden (79) of discontinuation symptoms can guide next-step treatment strategies. If discontinuation symptoms are burdensome, resumption of the original SSRI medication will typically resolve the symptoms quickly. If the reason for the SSRI discontinuation is to initiate a different treatment, cross-titration with another SSRI or an SNRI can reduce the severity of discontinuation symptoms. A brief trial of fluoxetine can be used if the alternative treatment planned is a nonserotonergic antidepressant. If the goal of SSRI discontinuation is to stop all antidepressant treatment, gradual taper may reduce the severity of discontinuation symptoms as compared with abrupt discontinuation. Smaller doses of medications can be dispensed with the use of liquid formulations. Liquid formulations are available for citalopram (10 mg/5 mL [13]), escitalopram (5 mg/5 mL [14]), paroxetine (10 mg/5 mL [15]), fluoxetine (20 mg/5 mL [16]) and sertraline (100 mg/5 mL [17]). Compounding pharmacies may be used to prepare liquid formulations at very low doses too (80). Symptomatic management, such as use of antihistaminic drugs for acute dystonic reaction, benzodiazepines for anxiety or agitation, and antipsychotics for hallucinations, can be used briefly if discontinuation symptoms are severe or resumption of an SSRI is not prudent.

Recommendations

While the discussion about discontinuation symptoms and the conflation of the presence of discontinuation symptoms with addictive potential of SSRIs is not new, it has gained recent media attention (81). As SSRIs are psychotropic medications, development of new symptoms or worsening of previous symptoms after discontinuation is not surprising. Major depressive disorder is a chronic disorder that necessitates treatment plans that effectively attain optimal symptom control, minimize functional impairments, and reduce the burden associated with treatment modalities. In this context, development of discontinuation symptoms should be considered in the same fashion as treatment-emergent adverse events. Development of rebound or new symptoms and worsening of clinical course on discontinuation is not a trait unique to SSRIs; it is shared by treatments for other chronic conditions, such as beta-blockers for hypertension and hypoglycemic agents for diabetes. Hence, as in these other chronic conditions, clinicians treating patients with major depression need to enhance shared decision-making by providing patient education as well as to proactively consider and discuss the potential risk of discontinuation symptoms when making treatment decisions. Finally, it is essential to ensure that the presenting symptoms are not related to reemergence of the depressive episode, which would require a different management strategy.