When the First SSRI Fails

A review of switching strategies in patients resistant to an initial SSRI.
Charles DeBattista, DMH, MD, and Hugh Brent Solvason, PhD, MD

Dr. DeBattista is Associate Professor, Department of Psychiatry & Behavioral Sciences,
Stanford University School of Medicine, Stanford, CA.

Dr. Solvason is Assistant Professor, Department of Psychiatry & Behavioral Sciences,
Stanford University School of Medicine, Stanford, CA.

Editor’s Note
As primary care practitioners have become more adept at treating depression with
antidepressants, psychiatrists are seeing more and more patients who have failed
to respond to the usual first line of treatment. Sometimes this is due to a low dose,
poor compliance, or both. In other cases, there seems to be a genuine resistance
to the antidepressants used.

The authors review the relatively scant literature dealing with strategies for
changing medication when the first selective serotonin reuptake inhibitor fails.
From this, suggestions are made for the practical management of these difficult
and frustrating patients.

An incomplete response to antidepressant therapy is common. At least 40% of pa-
tients fail to achieve a 50% reduction in depressive symptoms in the average 6- to
8-week trial of selective serotonin reuptake inhibitors (SSRIs). Perhaps more dis-
couraging, only 25% to 45% of patients achieve remission or become free of
symptoms in any given acute trial.1 SSRIs represent at least 80% of new antide-
pressant prescriptions. Thus, the issue facing most clinicians is what to do if the
first SSRI trial proves unsuccessful.

Failure to respond, as opposed to failure to tolerate, may be related to a num-
ber of factors. The most common of these are inadequate dosing and duration. In
primary care settings, prescriptions are rarely refilled and insufficient dosing is common.2
Other factors include noncompliance, inaccurate diagnosis (with atypical, bipolar, or psy-
chotic depression), and comorbid medical or psychiatric factors.

The most common strategies when patients do not respond to treatment are
switching to another antidepressant or attempting to augment a response by
adding an additional agent. Switching to another antidepressant offers a number
of advantages, the first of which is avoiding polypharmacy.The vast majority of ef-
ficacy and maintenance studies have been completed with single agents. We know
relatively little about the efficacy of augmentation strategies in the treatment of
acute episodes and virtually nothing about the role of augmentation strategies in
preventing subsequent episodes. Switching also has the advantage of avoiding
compounding side effects with multiple drugs. It is often difficult to determine
which drug is contributing to which side effect. Likewise, there is less of a chance
of a drug interaction by switching than by augmentation. In some cases, eg, when
switching from an SSRI to a monoamine oxidase inhibitor (MAOI), a proper washout pe-
riod is necessary to avoid potentially lethal drug interactions. Finally, the cost of using
one agent tends to be more economical than the cost of adding agents.

That being said, there are certain disadvantages to switching. It may take
more time because of time lost in tapering one drug and increasing the dose of
another. In addition, switching strategies are poorly studied in the management of
resistant depression. Nonetheless, most psychiatrists are likely to switch to a dif-
ferent class of agents in nonresponding patients and augment in partially re-
sponding patients. In clinical practice, however, many clinicians will switch to an-
other SSRI before moving on to a different class of agents.

Switching to Another SSRI
Very few studies have addressed the efficacy of switching from one SSRI to another
when the first SSRI fails. Although they share a similar mechanism of action, SSRIs
are a chemically heterogeneous class of agents. This is not true for the tricyclic an-
tidepressants (TCAs), which differ only modestly in their chemistry; switching
among the TCAs has not been shown to be particularly effective.3 Clearly, safety
profiles differ to some extent among the SSRIs. Thus, a number of studies have
shown that a patient’s failure to tolerate one SSRI does not necessarily indicate in-
tolerance of another SSRI.

At least 2 lines of evidence suggest that switching to another SSRI might be
useful if the first one fails. First, there is not necessarily congruence in response to
an SSRI: response to one SSRI does not guarantee response to another.In a study
of recurrent major depression, Sachetti and colleagues found that responders to fluoxetine
(Prozac)in one episode had a high response rate when randomized to fluoxetine in subse-
quent episodes, but much less robust responses when randomized to fluvoxamine (Luvox).

Likewise, patients who initially responded to fluvoxamine were much less likely to
respond when randomized to fluoxetine, although they were robust responders in
a subsequent randomization to fluvoxamine. Because responses to SSRIs are not
congruent, it may be reasonable to switch to another SSRI if the first one fails.
Switching from one SSRI to another has been examined in a number of open
trials, and these results have supported the practice of switching within the class.

In one open trial, 26% to 42% of patients who did not respond to treatment with
fluoxetine responded to sertraline (Zoloft).6 More recently, more than 50% of
patients who did not respond to fluoxetine responded when switched to citalo-
pram (Celexa).8,9 This is consistent with earlier studies suggesting a 51% response
rate when switching from one SSRI to another.10 None of these studies, unfortu-
nately, was a randomized prospective trial. Therefore, the conclusions are less
than definitive. However, relevant studies have been at least consistent in suggest-
ing that it may be worth trying to switch to another SSRI when the first one fails.

Switching Outside the Class of SSRIs to . . .

The rationale for switching a patient who has not responded to an SSRI outside a
drug class is presumably to employ a different mechanism of action. Because de-
pression appears to be biologically heterogeneous, employing an antidepressant
with a different mechanism of action appears sound. Most of what we know about
switching, we know from experience with the TCAs. Studies have shown that
switching within the class of TCAs is largely ineffective, but switching to a differ-
ent class of agents is often effective.

The TCAs are still sometimes regarded as more efficacious than the SSRIs,
particularly in melancholic patients and patients with more serious conditions.
Thus, it might make sense to switch to a TCA in a patient who has not responded
to treatment with an SSRI. A 45% to 60% response rate in patients switched to
TCAs who failed either paroxetine (Paxil)16or sertraline was suggested in a num-
ber of small crossover studies.

The largest prospective trial examining the utility of switching from an SSRI
to a TCA involved a 12-week crossover to imipramine in patients who were initially
randomized to sertraline but failed to respond.17 This study involved more than
635 patients with chronic major depression or double depression. Of the patients
who failed to respond to sertraline, approximately 50% responded when switched
to imipramine. Interestingly, this was approximately the same percentage that re-
sponded when crossed over to sertraline after initially failing imipramine.

Venlafaxine (Effexor):
Venlafaxine (Effexor) resembles the TCAs in that its primary action is to block the
reuptake of serotonin and, to a lesser extent, norepinephrine. Very few studies have evaluated the success of switching from an SSRI to venlafaxine. In one study of 84 highly treatment-resistant depressed patients, all of whom had failed to respond to SSRIs as well as many other interventions, 33% responded to venlafaxine, and most of those maintained the response for 1 year.

Bupropion (Wellbutrin):
Bupropion (Wellbutrin)is primarily a noradrenergic agent that facilitates mild
dopamine reuptake at higher doses. Switching from an SSRI to bupropion is a
common strategy in patients complaining of SSRI-induced sexual dysfunction.
Bupropion is less likely to cause sexual dysfunction than serotonergic drugs. In
one study, 31 patients who discontinued fluoxetine because of sexual dysfunction
tolerated bupropion 250 to 300 mg/day.

Some open-label data suggest that individuals with a lack of response to an
SSRI may benefit from switching to bupropion. For example, 9 patients with
chronic fatigue syndrome who did not respond to fluoxetine, which is commonly
used to treat this condition, responded when switched to bupropion.20 Likewise,
a more recent study found that among fluoxetine nonresponders switched to
bupropion, 28% became full responders and another 24% became partial re-

Nefazodone (Serzone):
Nefazodone (Serzone)is a 5-HT2 antagonist with some mild norepinephrine and
serotonin reuptake blocking properties.22 Its mechanism of action is sufficiently
different from that of SSRIs that switching from an SSRI to nefazodone may ben-
efit some patients. Nefazodone has a low rate of sexual dysfunction associated
with its use. The only study that has examined the efficacy of switching from an
SSRI to nefazodone reported a high rate of success in 115 patients who responded
poorly to treatment with SSRIs. Thase and colleagues reported that 66% of poor
responders to SSRIs responded to nefazodone 300 to 600 mg/day.

Mirtazepine (Remeron):
Mirtazepine (Remeron) is a presynaptic alpha-2 blocker with a variety of effects
on serotonin and histamine receptors. As with bupropion and nefazodone, mir-
tazapine carries a low risk for drug-induced sexual dysfunction. Two recent stud-
ies have suggested that SSRI nonresponders may do well when switched to mir-
tazapine. In one study, 47% of 103 nonresponders to SSRIs responded when
switched to mirtazepine. In another study of 250 SSRI nonresponders, 37% re-
sponded when switched to mirtazapine.

MAOIs are currently indicated for use in treatment-resistant depression. Unfortu-
nately, the concurrent use of MAOIs with SSRIs is contraindicated use because of
the risk for serotonin syndrome. Thus, there are no formal studies of switching
from an SSRI to a MAOI. Fluoxetine requires a 5-week washout period and the
other SSRIs require at least a 2-week washout period before the patient can start
MAOI therapy safely.

Clinical Considerations
A variety of pharmacokinetic and pharmacodynamic interactions are possible
when a patient switches from an SSRI to a different SSRI or to another class of an-
tidepressants. None of these strategies has been well studied.

The 2 basic strategies are to cross taper or wash out the original drug before
starting the new one. In a cross taper, the SSRI is slowly reduced while the new
agent is titrated up. A cross taper may mitigate discontinuation symptoms and
takes less time; however, cross tapering does increase the risk for drug interactions
and compounding side effects. The washout of the original SSRI and gradual increase
in the dose of the new agent take time, but this process reduces the risk for drug interactions.
For most switches from an SSRI, a cross taper may be advantageous. This is
especially true when drug interactions are unlikely. Thus, for example, switching
to bupropion or mirtazapine may be well tolerated in a cross taper, ie, bupropion
SR 100 to 150 mg/day of can be safely added to the existing SSRI dose. Alterna-
tively, the SSRI dose may be decreased by 25% to 50% with the addition of bupro-
pion. After the first week, the bupropion dose is doubled and the SSRI dose is cut
further. Thus, the entire taper of the SSRI and titration of the bupropion is com-
pleted in 2 to 4 weeks.Mirtazapine 15 mg/day can be added during the first week
and may be increased by 15 mg/week while the SSRI is being tapered. Due to the
potential for additive sedation with an SSRI and mirtazepine, it may be useful to
start with a higher dose of mirtazepine (eg, 30 mg/day), which is actually less se-
dating (a unique property of this drug).

Tapering the SSRIs with short half-lives—such as paroxetine, sertraline, and
fluvoxamine (half lives of 21, 24, and 15 hours, respectively)—will decrease the
risk for symptoms related to discontinuation. Even small doses of paroxetine (20
mg/day) or sertraline (50 mg/day) might still be cut in half to reduce the possi-
bility of symptoms associated with discontinuation of the SSRI. Fluoxetine(half-
life of 7–9 days, including the active metabolite)and citalopram (half-life of 35
hours) may be stopped more abruptly because of their long half-lives.25,26
A cross taper or rapid washout and switch can also be used to switch from one
SSRI to another.A cross taper of SSRIs does increase the risk for adverse seroton-
ergic effects, and many clinicians prefer stopping one SSRI and quickly starting another.

There appears to be a protective effect on discontinuation symptoms
when starting one SSRI rapidly after discontinuing another. Thus, many clinicians
will stop a long-acting SSRI (fluoxetine or citalopram) abruptly and start at a
lower dose of the new SSRI. The metabolism of paroxetine is CYP 2D6 and, thus, flu-
oxetine will artificially raise paroxetine levels for several weeks after it is discontinued.
Therefore, the starting dose of paroxetine may be 10 mg/day instead of 20
mg/day. The risk of a CYP-2D6-mediated drug interaction is small with citalo-
pram; therefore, a different SSRI can be started at the usual dose after discontin-
uing the citalopram.

CYP-2D6 interactions are also important when switching from an SSRI to
either nefazodone, TCAs, or venlafaxine. The metabolism of both TCAs and ven-
lafaxine is mediated by CYP 2D6 oxidative enzyme, as is m-CPP, an anxiogenic
metabolite of nefazadone. Consequently, these drugs compete for the same
degradative mechanisms. All of the SSRIs, except citalopram and fluvoxamine, may
raise TCA serum levels and increase the risk for adverse cardiovascular effects. TCA
blood levels may be monitored during a cross taper while the TCA dose is gradu-
ally increased. Increased venlafaxine levels associated with a 2D6-mediated drug
interaction may result in an elevated blood pressure or nausea. Therefore, the
starting dose of venlafaxine should be low (37.5 mg/day)during a cross taper
when a 2D6-mediated interaction with the other antidepressant is expected.

Switching from an SSRI to nefazodone may be best accomplished with a washout
of the original SSRI. Some patients will be sensitive to the effects of m-CPP
if an SSRI is given simultaneously or if the SSRI washout time is insufficient.
Symptoms can include agitation, dysphoria, anxiety, and visual distortion. Starting
Serzone at a lower dose, such as 100 mg/day the first week, may reduce problems
in sensitive patients. A 1-week washout may be sufficient for short-acting SSRIs and
2 to 4 weeks for longer-acting SSRIs. Citalopram does not require a washout.
Switching from an SSRI to a MAOI is the only switch that mandates a sufficient
washout of the SSRI: 5weeks for fluoxetine, and 2 weeks for the other SSRIs. Sero-
tonin syndrome and fatalities have been reported after an insufficient washout
time for the SSRI.

2003, DeBattista & Solvason

Leave a Reply