There is broad agreement that pharmacotherapy in schizophrenia
should be evidence-based. With this goal in mind, a number
of groups -- including the Schizophrenia Patient Outcome
Research Team (PORT) (Lehman et al., 1998), the Department
of Veterans Affairs (Mental Health Strategic Health Care
Group and The Psychosis Working Group, 1997), the American
Psychiatric Association (APA, 1997), and the Texas Medication
Algorithm Project (Miller et al., 1999) -- have reviewed
the literature and formulated recommendations for the prescribing
of antipsychotic medications for the treatment of schizophrenia.
However, the recommendations of these groups differ on several
important and sometimes controversial points.
This report summarizes the proceedings of The Mount Sinai
Conference -- a gathering of investigators who had done
the background research for and developed the various clinical
guidelines mentioned above. It was hoped that the group
would reach a consensus about whether the evidence base
concerning antipsychotic prescribing for the treatment of
schizophrenia was clear enough to support establishing a
particular clinical practice as a treatment standard. The
conference focused on treatment issues that arise frequently
in routine practice where there is practitioner disagreement,
as indicated by broad variations in prescribing practice.
It is important to emphasize that the Mount Sinai Conference
focused solely on questions where recent research may have
resolved a controversial issue. Important issues in pharmacotherapy
were not discussed when the group believed the issue had
been previously resolved or when there was insufficient
evidence to make a recommendation. As a result, the recommendations
of the group cannot be viewed as a comprehensive guide for
treating schizophrenia. Rather, the answers given to each
question should be considered the consensus opinion.
Question 1. Should Conventional Agents
Still Be Considered First-line Agents?
We use the term "first-line" to include agents that can
be prescribed for a broad range of individuals with schizophrenia,
including those experiencing a first episode of schizophrenia
(see Question 2 for an exception with respect to the first-line
agent ziprasidone), patients who have a medication-responsive
illness but whose obtainable drug history is inadequate,
and patients who have responded to other antipsychotics
but have been switched to new agents.
A second-line agent is an agent with demonstrated efficacy
but with a risk or use profile that necessitates that it
be prescribed when there is clinical evidence that first-line
agents are ineffective or poorly tolerated. By this definition,
in the United States today, clozapine, mesoridazine, and
thioridazine are second-line antipsychotic agents.
First generation antipsychotics (FGAs) are those conventional
antipsychotic medications that preceded clozapine's entry
into the antipsychotic armamentarium. Second-generation
antipsychotics (SGAs) include clozapine and those agents
brought to market following clozapine: risperidone, olanzapine,
quetiapine, and ziprasidone. Although the SGAs are discussed
as a group, they are a heterogeneous group of medications
with different side effect profiles. All of the SGAs are
associated with less extrapyramidal side effects (EPS) than
FGAs when prescribed at effective doses.
SGAs -- other than clozapine and ziprasidone -- should
be selected before FGAs for patients experiencing a first
episode of schizophrenia or for those patients where there
is no available history concerning response to antipsychotics.
FGAs may be appropriate selections for the following groups
of patients: (1) patients who have a history of responding
well to conventional antipsychotics without experiencing
EPS, (2) patients who have responded better to FGAs than
to SGAs, and (3) patients who have responded better to long-acting
depot medications when compared to oral antipsychotics.
Background
There was a consensus that SGAs are associated with a reduced
risk of tardive dyskinesia (TD). There is also convincing
evidence that SGAs are associated with reduced EPS compared
to conventional drugs and that the effect sizes for these
differences are medium to large. Because EPS are one of
the important components of non-adherence to medication
and one of the drivers of the subjective response to an
antipsychotic, this potential adverse reaction should receive
the close attention of antipsychotic prescribers. One of
the important difficulties in managing EPS is that many
patients will experience akathisia and other forms of EPS
at the antipsychotic doses that are needed for managing
their illness.
A meta-analysis by Geddes and coworkers (2000) took the
position that there is insufficient evidence to conclude
that there is a safety advantage for SGAs. These authors
agreed that there was an EPS advantage for SGAs, but they
suggested that this should be measured against the other
side effects of these agents, such as weight gain. However,
it was the consensus of the participants of the Mount Sinai
Conference that the high prevalence of EPS on conventional
agents and the higher risk of TD outweighed the differences
in liability for weight gain. Moreover, if patients demonstrate
early evidence of weight gain, there are alternative medications
that can be prescribed -- e.g., ziprasidone.
The evidence supporting superior efficacy and effectiveness
of SGAs remains controversial. The previously cited meta-analysis
by Geddes found that efficacy advantages for SGAs over conventional
drugs tended to occur when patients received relatively
high doses of haloperidol as the comparison medication.
They found that when studies used doses of haloperidol below
12mg daily, there were no advantages in efficacy for the
newer agents.
The meta-analysis by Leucht (1999) found that when there
were statistically significant advantages for newer drugs
over older drugs, the effect sizes tended to be very small.
These results differ from a recent meta-analysis by Davis
(personal communication), who compared SGAs and conventional
antipsychotics in controlled trials. Effect sizes for the
advantages in efficacy of clozapine, risperidone, and olanzapine
over FGAs were 0.54, 0.22, and 0.21, respectively. The effect
sizes were 0.00 for quetiapine and -0.08 for ziprasidone,
suggesting that these agents demonstrated similar effectiveness
as FGAs.
There were fewer studies of quetiapine and ziprasidone,
and these agents were studied at later dates than the other
agents. As a result of being studied later, patients may
have entered these trials after already failing to respond
to other SGAs. If such individuals are less likely to respond
to any medication, this would put these agents at a disadvantage
and may explain, at least in part, the smaller effect sizes
observed.
There is also evidence supporting advantages for SGAs for
treating patients who are experiencing their first psychotic
episode. A number of studies -- usually industry-sponsored
-- have randomly assigned first-episode patients to newer
or older drugs and found advantages for the newer agents.
A large international study (Emsley 1999) randomized 183
patients to either risperidone or haloperidol and found
that response rates were high on both drugs. Risperidone
was better tolerated and led to fewer dropouts due to side
effects.
In a large multicenter study (Sanger et al. 1999) that
compared olanzapine and haloperidol, a subpopulation of
83 patients were recovering from their first episode. These
recent-onset patients demonstrated a better response to
olanzapine than haloperidol for both positive and negative
symptoms. Moreover, patients taking olanzapine tended to
demonstrate improvements in EPS, while patients taking haloperidol
worsened. Although we are unaware of controlled studies
of quetiapine and ziprasidone for first-episode patients,
their side effect profiles indicate that they also may offer
safety and tolerability advantages for these individuals.
Question 2. Should Ziprasidone Be
a First Line Agent?
The cardiac safety of ziprasidone will become clearer as
greater numbers of patients are exposed to this agent. Until
sufficient information is available on the rates of arrhythmias
or sudden death on ziprasidone, this antipsychotic should
only be prescribed when patients have received a trial with
another SGA. This is consistent with ziprasidone's package
insert, which states that "other agents should be tried
first."
Background
Ziprasidone is associated with prolongation of the QT interval
of the electrocardiogram. This is a serious concern, since
excessive prolongation can lead to a potentially fatal ventricular
arrhythmia known as torsades de pointes. The number of patients
exposed to ziprasidone has exceeded 150,000 (as of 10/01)
and no cases of sudden death due to drug effects or cardiac
arrhythmia have been reported.
Question 3. What is an Adequate Trial?
Is a 4, 6, 8 or N Week Trial Sufficient? If These Cut-Off
Points Were Used, How Many Individuals Would Be Discontinued
from a Medication That They Would Have Responded to Had The
Trial Duration Been Longer? Are There Individual Characteristics
That Predict a Sl
A. Patients who have received an antipsychotic for
one week at therapeutic doses for an acute exacerbation
and have not demonstrated any evidence of improvement may
be unlikely to respond to that antipsychotic robustly during
that episode. However, it takes up to four weeks on full
therapeutic doses to demonstrate convincingly that a patient
should be considered a nonresponder to that regimen. If
patients demonstrate a partial response, the trial should
be extended to as long as 12 weeks in the face of a continuing
partial response.
Background
There is relatively little information from controlled
clinical trials on the relationships between trial duration,
antipsychotic dose, and treatment response that may be used
to guide decision-making about the trial duration. In one
of the few studies to address this issue, Janicak and colleagues
(1997) examined patients' responses to low, medium, and
high antipsychotic doses. There were no differences in the
rates of treatment response among the three groups. Patients
who failed to respond to either low or high doses tended
to have a better response when they were switched to a medium
dose level.
The results suggest that there is a subgroup of patients
whose treatment response is dose dependent, whereas the
responses of most patients -- as long as they are above
some minimum dose level -- are relatively independent of
dose.
There is evidence indicating that patients who receive
an antipsychotic that is effective for them will continue
to improve over several months, although most improvement
will occur during the first weeks of treatment.
Marder SR, Essock SM, et al.: The Mount Sinai Conference
on the pharmacotherapy of schizophrenia. Schizophrenia Bulletin
2002; 28 (1): 5-16. Reprinted with permission of the author.
For a complete list of references, visit our Web site at
www.manisses.com/sinai.pdf
to view a PDF file.
Volver Atrás