ORIGINAL ARTICLES
Placebo-Controlled Comparison of the Selective
Serotonin Reuptake Inhibitors Citalopram and
Sertraline*
Stephen M. Stahl
Background: Previous comparative studies of the selective serotonin reuptake inhibitors (SSRIs) have rarely
included a placebo control group and have rarely demonstrated significant between-group differences. The study
reported on here was a placebo-controlled comparison of
the antidepressant effects of two SSRIs, citalopram and
sertraline.
Methods: Three hundred twenty-three patients with DSMIV– defined major depressive disorder were randomized to
24 weeks of double-blind treatment with citalopram
(20 – 60 mg/day), sertraline (50 –150 mg/day), or a placebo. The primary efficacy measure was the Hamilton
Depression Rating Scale (HAMD) and the primary statistical analysis was an analysis of variance comparing the
change from baseline to the last observation carried
forward in each treatment group.
Results: Both citalopram and sertraline produced significantly greater improvement than placebo on the HAMD,
the Montgomery–Asberg Depression Rating Scale, and the
Clinical Global Impression Scale. Significant improvement was observed at earlier timepoints in the citalopram
group than the sertraline group; however, sertraline
treatment was associated with increased gastrointestinal
side effects and a tendency toward early discontinuation,
and analyses that excluded early dropouts revealed similar acute efficacy for the two active treatments. The
Hamilton Anxiety Scale demonstrated a significant anxiolytic effect of citalopram, but not sertraline, relative to
placebo.
Conclusions: This study confirms the antidepressant efficacy of two SSRIs, citalopram and sertraline. It is hypothesized that the more consistent evidence of antidepressant
activity that was observed early in treatment in the
citalopram group was related to more pronounced antianxiety effects and better tolerability upon initiation of
therapy. Biol Psychiatry 2000;48:894 –901 © 2000 Society of Biological Psychiatry
Key Words: Selective serotonin reuptake inhibitors, citalopram, sertraline, depression, antidepressants, anxiety
From the Department of Psychiatry, School of Medicine, University of California,
San Diego.
Address reprint requests to Stephen M. Stahl, M.D., Ph.D., University of California,
San Diego, Department of Psychiatry, School of Medicine, 8899 University
Center Lane, Suite 130, San Diego CA 92122.
Received October 25, 1999; revised May 2, 2000; accepted June 7, 2000.
© 2000 Society of Biological Psychiatry
*See accompanying Editorial, in this issue.
Introduction
S
ince the mid-1980s, new classes of antidepressant
drugs have emerged that offer improved safety and
tolerability profiles compared with previous classes of
antidepressant drugs. There are at least seven distinct
pharmacologic mechanisms of action for the two dozen or
so known antidepressants (Stahl 1998a). Most comparative studies of antidepressants, however, have shown
similar levels of efficacy (Jonghe and Swinkels 1997).
Such findings from large, multicenter trials are at odds
with the common clinical experience that some patients
respond well to one antidepressant but not to another
(Stahl 1998b), even from the same class (Brown and
Harrison 1995; Joffe et al 1996; Thase et al 1997, 1999;
Zarate et al 1996). Although it is not yet possible to predict
which patients will respond to which antidepressants,
some recent trials have identified differences in efficacy
between antidepressants of differing pharmacologic mechanisms (e.g., Wheatley et al 1998).
Fluoxetine, sertraline, paroxetine, citalopram, and fluvoxamine are the five currently available selective serotonin reuptake inhibitors (SSRIs) constituting the most
widely prescribed class of antidepressants. The lack of
demonstrated robust differences in tolerability or efficacy
between the SSRIs in head-to-head comparisons is leading
formularies to treat these agents as interchangeable, resulting in denied access and substitutions, a trend that is likely
to accelerate as the first SSRIs become generic. On the
other hand, the five SSRIs are not members of the same
chemical classes and do not share the same secondary
binding characteristics (e.g., Stahl 1998b). Failure to
demonstrate differences between drugs may be due in part
to lack of large-scale, multicenter, placebo-controlled,
head-to-head comparisons of two SSRIs.
Citalopram was first marketed in Europe in 1989, is
now available in 68 countries worldwide, and is the most
recent SSRI to become available in the United States.
Because it is the most selective SSRI (Hyttel et al 1995)
and has low potential for drug– drug interactions (Green0006-3223/00/$20.00
PII S0006-3223(00)00957-4
Comparison of Citalopram and Sertraline
blatt et al 1998; Jeppesen et al 1996), citalopram theoretically could offer advantages over other drugs of this class.
Sertraline also has a low potential for drug– drug interactions and may have clinically relevant interactions with the
dopamine transporter (Bolden-Watson and Richelson
1993) and the sigma receptor (Schmidt et al 1989),
secondary binding properties that could distinguish it from
other SSRIs and offer different advantages over drugs of
this class. The study reported here investigated the efficacy and safety profiles of citalopram and sertraline in a
24-week, placebo-controlled trial.
Methods and Materials
Eight centers in the United States (see Acknowledgements)
participated in this double-blind, prospectively randomized comparison of citalopram and sertraline with a placebo and each
other. Patients of either gender, aged between 18 and 60 years,
who satisfied DSM-IV (American Psychiatric Association 1994)
criteria for major depressive disorder with a minimum 2 months
duration of illness could be considered for inclusion. Women
were eligible only if not pregnant and if taking adequate
contraceptive measures. All eligible patients had to score at least
22 on the 17-item Hamilton Depression Rating Scale (HAMD;
Hamilton 1960), have a score of ⱖ2 on the depressed mood item
and a minimum score of ⱖ8 on the Raskin Depression Scale
(Raskin et al 1969), together with a lower score on the Covi
Anxiety Scale (Covi et al 1981).
Patients were excluded from the study if they 1) presented
with another DSM-IV Axis I diagnosis, 2) were taking other
psychotropic medication, 3) were at increased risk of suicide
(serious suicide attempt in last 12 months or HAMD suicidality
item score of ⱖ3), 4) were treatment resistant (i.e., had failed to
respond to adequate courses of one SSRI or two other pharmacologically different antidepressants), 5) had a history of sertraline intolerance or SSRI hypersensitivity reactions, or 6) had a
history of alcohol or substance abuse.
Patients who showed a ⱖ20% decrease in HAMD-17 item
total score during the single-blind placebo lead-in were barred
from further participation. The study protocol and consent form
were approved by the appropriate institutional review boards.
The capacity of the patients to speak, read, write and understand
English, to provide written informed consent, and to complete
the patient-rated assessments was confirmed during the course of
the initial telephone screening, the review of the study procedures by the investigator, and the diagnostic interview. In study
records, patients were exclusively identified by their initials,
screening number, and randomization number.
Study Drug Administration
Having given informed consent, eligible patients began a
1-week, single-blind placebo run-in. Patients who continued to
meet entrance criteria at the end of the lead-in period (baseline)
were prospectively randomized to 24-weeks treatment with
citalopram, sertraline, or placebo.
Dosing began with one capsule/day in the morning with food
BIOL PSYCHIATRY
2000;48:894 –901
895
for the first week (20 mg citalopram, 50 mg sertraline, or
matched placebo) and rose by one capsule/week to three capsules/day in the third week (60 mg citalopram, 150 mg sertraline,
or placebo). This dose was to be maintained for the duration of
the 24-week study; however, for patients intolerant of this dose,
a reduction to two capsules/day was allowed. Patients unable to
tolerate at least two capsules/day were discontinued. Patients
who had difficulty sleeping were permitted chloral hydrate (1000
mg) for no more than 3 days/week for the first 4 weeks. Other
psychotropic medication was not allowed.
Assessments
After the screening and baseline assessments, further reviews
were undertaken weekly for the first 4 weeks of double-blind
treatment, then at the end of weeks 6 and 8, and thereafter at
four-week intervals. Depressive symptomatology was assessed at
each visit by the 21-item HAMD, Montgomery–Asberg Depression Rating Scale (MADRS; Montgomery and Asberg 1979), the
Clinical Global Impression (Guy 1976) Improvement (CGI-I)
and Severity (CGI-S). Additional efficacy data collected at some
visits during the study included the Hamilton Anxiety Scale
(HAMA; Hamilton 1959), the Symptom Checklist-56 (SCL-56;
Guy 1976), and the Quality of Life Enjoyment and Satisfaction
Questionnaire (Q-LES-Q; Endicott et al 1993).
Tolerability was assessed at each visit by replies to nonleading
questions. In addition, a sexual function questionnaire (results to
be reported elsewhere) was administered at baseline, week 8, and
week 24. Safety assessments included laboratory tests, electrocardiogram (ECG), vital signs (weight, sitting and standing blood
pressure), and a pregnancy test.
Evaluation of Data
The primary efficacy analyses were based on the intention-totreat (ITT), last-observation-carried-forward (LOCF) population.
This included any patient randomized to double-blind treatment
who took at least one dose of study medication and who had at
least one subsequent assessment of efficacy. A secondary analysis was conducted using the subset of patients who completed
the initial 8 weeks of double-blind treatment. All statistical tests
were two sided and used a 5% significance level. Efficacy
variables were analyzed by a two-way analysis of variance with
treatment and study center as the two factors. If the treatmentby-center interaction was not significant, then the interaction
term was dropped from the statistical analysis. To protect the
experimentwise ␣ (.05) against multiple comparisons, pairwise
comparisons between treatment groups with p ⱕ .05 were treated
as statistically significant only if the overall three-group analysis
had also yielded a significant treatment effect.
The primary outcome variable was the change from baseline in
HAMD total score at week 24. Secondary outcome variables
included the change from baseline in HAMD total score at other
visits as well as for the MADRS, CGI-I and -S, HAMA, SCL-56,
and Q-LES-Q. In addition, the HAMD depressed mood item and
the HAMD subscales (Guy 1976) for sleep disturbance, psychomotor retardation, cognitive disturbance, melancholia (Bech
et al 1981), and anxiety were each analyzed individually. Median
896
S.M. Stahl
BIOL PSYCHIATRY
2000;48:894 –901
Table 1. Patient Characteristics
Demographic parameter
Age (years)
Gender (male/female)
Race (white/other)
Weight (kg)
History (recurrent/first episode)
Citalopram
(n ⫽ 107)
Sertraline
(n ⫽ 108)
Placebo
(n ⫽ 108)
39.1
47%/53%
88%/12%
80.0
50%/50%
39.5
35%/65%
88%/12%
79.8
51%/49%
37.3
37%/63%
92%/8%
79.1
53%/47%
cutoff scores on the HAMD subscales were also used to identify
patient subpopulations with more severe symptoms of sleep
disturbance (ⱖ4), psychomotor retardation (ⱖ8), cognitive disturbance (ⱖ5), and anxiety (ⱖ8).
Safety analyses considered all patients who had received at
least one dose of study medication. Treatment-emergent adverse
events (TEAE) included any event that was not present at
baseline or, if present at baseline, increased in severity during
double-blind treatment. Adverse events were coded by System
Organ Class and Preferred Term using a World Health Organization dictionary. Safety assessments included laboratory tests,
vital signs, and ECG parameters.
Approximately 23% (70 of 323; 20% of citalopram patients, 26% of sertraline patients, and 22% of placebo
patients) discontinued during the first 8 weeks of doubleblind therapy. Forty percent (130 of 323; 44% of citalopram patients, 44% of sertraline patients, and 32% of
placebo patients) completed the full 24 weeks. Lack of
efficacy accounted for 55 discontinuations (8% of citalopram patients, 11% of sertraline patients, and 31% of
placebo patients), whereas 47 patients discontinued because of adverse events (14% of citalopram patients, 19%
of sertraline patients, and 10% of placebo patients).
Results
Efficacy
BASELINE COMPARISON. The rating scale scores obtained at baseline (Table 2) were indicative of a moderately-to-severely depressed patient population and a similar severity of depressive symptomatology in the three
treatment groups.
Patient Demographics
There were no clinically relevant differences in demographic variables (Table 1). As is typically found in
antidepressant trials, the mean age of the patients was
about 40 years, and the majority were female.
ENDPOINT COMPARISON. In the endpoint analysis
(Table 2), both citalopram and sertraline showed a significant improvement in efficacy compared with placebo;
however, the magnitude of the mean improvement in
depressive symptoms was numerically greater in the citalopram group. This was true for all outcome measures
Patient Disposition
Each of the eight study sites randomized between 24 and
48 patients to double-blind treatment, resulting in a total of
323 randomized patients, of whom 316 had post-baseline
efficacy data and were included in the efficacy analyses.
Table 2. Baseline and Endpoint Efficacy Variables
Citalopram (n ⫽ 103)
Outcome
measure
HAMD
MADRS
CGI-I
CGI-S
HAMA
SCL-56
Q-LES-Q
Baseline
Change
from
baseline
26.5
32.4
na
4.38
17.0
114
45.1
⫺14.5a
⫺18.0a
2.12a
⫺1.8a
⫺7.5a
⫺24.4a
⫹10.1a
Sertraline (n ⫽ 106)
Placebo (n ⫽ 107)
Baseline
Change
from
baseline
Baseline
Change
from
baseline
26.6
31.2
na
4.34
16.7
116
45.3
⫺13.0b
⫺15.7a
2.27a
⫺1.6b
⫺6.1
⫺24.6a
⫹9.0a
26.4
31.1
na
4.32
16.7
110
45.6
⫺10.0
⫺11.1
2.71
⫺1.2
⫺4.8
⫺14.1
⫹4.0
HAMD, Hamilton Depression Rating Scale; MADRS, Montgomery–Asberg Depression Rating Scale; CGI-I, Clinical Global
Impression Improvement; CGI-S, Clinical Global Impression Severity; HAMA, Hamilton Anxiety Scale; SCL-56, Symptom
Checklist-56; Q-LES-Q, Quality of Life Enjoyment and Satisfaction Questionnaire.
a
Significantly different from placebo, p ⬍ .01.
b
Significantly different from placebo, p ⬍ .05.
Comparison of Citalopram and Sertraline
BIOL PSYCHIATRY
2000;48:894 –901
897
Figure 1. Mean change from baseline on
the Hamilton Depression Rating Scale in
depressed patients treated with sertraline,
citalopram, or a placebo. Error bars represent the SEM. *p ⬍ .05 and **p ⬍ .01, as
compared with the placebo; †p ⬍ .05, as
compared with sertraline.
except the SCL-56 (Table 2). On the HAMA, significant
improvement relative to placebo was observed in the
citalopram group but not the sertraline group.
COMPARISON BY VISIT. For citalopram, mean
changes from baseline on the HAMD were significant
compared with placebo from week 3 until the end of the
study (Figure 1). At week 3, however, the overall threegroup comparison achieved only borderline significance
(p ⫽ .07). For sertraline, significant differences from
placebo on the HAMD occurred at weeks 12, 20, and 24
(Figure 1). Similar sustained advantages for citalopram
over placebo were observed on the MADRS from week 2
(Figure 2). On the CGI-I and CGI-S, significant citalopram–placebo differences first appeared at week 2 and
week 4, respectively. For sertraline, mean changes from
baseline were significant compared with placebo at weeks
12, 20, and 24 for the CGI-S, and at weeks 12, 16, 20, and
24 for the MADRS and CGI-I. A significant treatmentby-center interaction (p ⬍ .05) was observed for the
MADRS at week 12 and 16. This interaction was probably
attributable to lack of consistency of the sertraline effect
across sites, because the sertraline group exhibited numerically greater mean improvement than the placebo group at
only 5 of the 8 centers at these visits. On the basis of a
responder criterion of a 50% decrease from baseline in the
HAMD, the response rate in the citalopram group was
significantly greater than placebo beginning at week 6, and
the response rate in the sertraline group was significantly
greater than placebo from week 12 onward (Figure 3). At
study endpoint, the percentage of patients with a
HAMD-17 total score below 8 was 28% in the placebo
group, 37% in the sertraline group, and 45% in the
citalopram group. Compared to sertraline, citalopram
Figure 2. Mean change from baseline on
the Montgomery–Asberg Depression Rating
Scale in depressed patients treated with sertraline, citalopram, or a placebo. Error bars
represent the SEM. *p ⬍ .05 and **p ⬍ .001,
as compared with the placebo; †p ⬍ .05, as
compared with sertraline.
898
S.M. Stahl
BIOL PSYCHIATRY
2000;48:894 –901
Figure 3. Percentage response rate in depressed patients treated with sertraline, citalopram, or a placebo. The responder criterion
was a ⱖ50% reduction from baseline in the
Hamilton Depression Rating Scale. *p ⬍ .05
and **p ⬍ .01, as compared with the
placebo.
showed significantly greater efficacy in the mean changes
from baseline on the HAMD (Figure 1), the CGI-I and the
CGI-S at week 2, and the mean change in MADRS at
weeks 2, 4, 6, 8, and 16 (Figure 2).
WEEK 8 COMPLETER ANALYSIS. A secondary analysis was conducted on the subset of patients who completed the initial 8 weeks of double-blind treatment, at
which point more than 70% of the patients who had
initiated each treatment remained in the study. In this
analysis, both citalopram and sertraline showed significant
efficacy on the HAMD, MADRS, CGI-I, and CGI-S
compared with placebo (Table 3).
HAMD SUBFACTORS. The mean change from baseline to endpoint for the HAMD factors and the depressed
mood item is shown in Table 4. Both citalopram and
sertraline were associated with significantly greater improvement compared with placebo on all of the symptom
clusters except anxiety, where only citalopram showed a
significant effect (Figure 4).
In the subgroups of patients with significant psychomotor retardation or insomnia, both citalopram and sertraline
showed significant improvement compared with placebo
(Table 5). Citalopram, but not sertraline, produced significant improvement in depressive symptomatology among
the anxious depressed patients and patients with cognitive
disturbance.
Safety
During the 24 weeks of the study, 47 (14.5%) patients
discontinued treatment prematurely due to adverse events;
15 (14%) of the citalopram group, 21 (19%) of the
sertraline group, and 11 (10%) of the placebo group. More
patients discontinued for adverse events in the first 8
weeks from the sertraline group (15%) compared with
citalopram and placebo (8% and 7%, respectively). This
difference was most apparent during the first week of the
study when five patients receiving 50mg/day sertraline,
Table 3. Week 8 Completer Analysis
Outcome
measure
HAMD
MADRS
CGI-I
CGI-S
Citalopram
(n ⫽ 83)
change from
baseline
Sertraline
(n ⫽ 78)
change from
baseline
Placebo
(n ⫽ 84)
change from
baseline
⫺15.3b
⫺19.2a
2.0b
⫺1.8b
⫺14.7b
⫺17.4b
2.0b
⫺1.7b
⫺12.2
⫺13.7
2.4
⫺1.4
HAMD, Hamilton Depression Rating Scale; MADRS, Montgomery–Asberg
Depression Rating Scale; CGI-I, Clinical Global Impression Improvement; CGI-S,
Clinical Global Impression Severity.
a
Significantly different from placebo, p ⬍ .01.
b
Significantly different from placebo, p ⬍ .05.
Table 4. Mean Change from Baseline on Hamilton Depression
Rating Scale Subscales
Outcome measure
Anxiety
Cognition
Melancholia
Psychomotor retardation
Sleep disturbance
Depressed mood item
Citalopram
(n ⫽ 103)
Sertraline
(n ⫽ 106)
Placebo
(n ⫽ 107)
⫺4.0a
⫺3.1b
⫺7.4a
⫺4.7a
⫺2.2a
⫺1.8a
⫺3.3
⫺3.0b
⫺6.9a
⫺4.3b
⫺1.9b
⫺1.6a
⫺2.9
⫺2.2
⫺5.1
⫺3.2
⫺1.2
⫺1.2
Significantly different from placebo, p ⬍ .01.
Significantly different from placebo, p ⬍ .05.
a
b
Comparison of Citalopram and Sertraline
BIOL PSYCHIATRY
2000;48:894 –901
899
Figure 4. Mean change from baseline on the
anxiety subscale of the Hamilton Depression
Rating Scale in depressed patients treated with
sertraline, citalopram, or a placebo. Error bars
represent the SEM. *p ⬍ .01, as compared
with placebo; †p ⬍ .05, as compared with
sertraline.
one patient receiving 20mg/day citalopram, and one patient receiving placebo discontinued treatment because of
adverse events. All five of the sertraline patients reported
nausea, and three also complained of insomnia. Overall,
each of the treatments was well tolerated, however, and
the mean dose at the end of the study was 57 mg/day in the
citalopram group and 143 mg/day in the sertraline group.
Similar proportions of patients in each group reported
TEAEs: 90.7%, 96.3%, and 88% for citalopram, sertraline,
and placebo, respectively. Over the full 24 weeks of the
study, six adverse events occurred in at least 10% of
citalopram or sertraline patients and with a significantly
higher incidence (p ⬍ .05) than placebo: nausea, somnolence, ejaculation disorder (male patients), increased
sweating, anorexia, and decreased libido. All of these
adverse events occurred significantly more frequently in
both active treatment groups versus placebo with the
exception of nausea, which occurred significantly more
frequently in sertraline patients only. The vast majority of
TEAEs were mild or moderate in severity for all treatment
groups.
There were no significant changes in vital signs or ECG
parameters except for a clinically meaningless decrease in
PR interval (2.9 msec) in the sertraline group. In the two
active treatment groups, standing pulse rate was decreased
by 1.4 or 1.5 beats/minute at study endpoint. Neither
citalopram nor sertraline significantly affected body
weight during the 6-month treatment period. Body weight
increases ⱖ7% of baseline occurred in four sertraline
patients (4%), four citalopram patients (4%), and three
placebo patients (3%). Decreases in body weight occurred
in seven sertraline patients (7%), five citalopram patients
(5%), and three placebo patients (3%). The statistically
significant mean changes in laboratory parameters that
were observed—increased cholesterol with citalopram
(⫹1.6%) and sertraline (⫹3.7%), increased alkaline phosphatase (⫹4.5%) with sertraline and decreased uric acid
(⫺6.6%) with sertraline—were small in magnitude and
clinically unimportant.
Discussion
There have been many comparative studies of the SSRIs,
but a recent literature review has identified only one
published placebo-controlled comparison between SSRIs
(Fava et al 1998). Inclusion of a placebo control in studies
of approved antidepressants has been commonly avoided,
in part because of ethical considerations, but also because
of the embarrassingly high rate at which established
antidepressant agents fail to separate from placebo. The
risk of such presumed Type II errors (“false negatives”) is
clearly illustrated by the Fava et al study, in which neither
fluoxetine nor paroxetine significantly separated from
placebo.
The present 24-week, placebo-controlled study provides
clear evidence confirming the antidepressant efficacy of
citalopram and sertraline and demonstrates the sustained
efficacy of these agents for a period of at least 6 months.
The endpoint analysis revealed significant improvement
relative to placebo for both active treatment groups on
each of the depression rating scales administered. Only the
HAMA and the HAMD anxiety subscale failed to detect a
significant effect of sertraline in comparison to placebo.
Significant improvement in the citalopram group was
consistently observed at earlier timepoints than in the
sertraline group. More patients discontinued sertraline
treatment because of adverse events at the beginning of the
study, and such discontinuations before patients had an
opportunity to experience an appreciable therapeutic response probably decreased the mean improvement mea-
900
S.M. Stahl
BIOL PSYCHIATRY
2000;48:894 –901
Table 5. Hamilton Depression Rating Scale Mean Change
from Baseline in Depression Subgroups
Subscale/criterion
Anxiety (ⱖ8)
Cognitive disturbance (ⱖ5)
Psychomotor retardation (ⱖ8)
Sleep disturbance (ⱖ4)
Citalopram
Sertraline
Placebo
⫺14.6a
⫺14.6b
⫺15.0a
⫺15.1a
⫺12.9
⫺14.2
⫺13.7a
⫺13.0b
⫺9.9
⫺11.1
⫺9.8
⫺9.3
Significantly different from placebo, p ⬍ .01.
Significantly different from placebo, p ⬍ .05.
a
included a relatively homogenous group of moderate-tosevere outpatients with uncomplicated first-episode or
recurrent major depressive disorder. Studies conducted in
a broader range of depression subpopulations could be of
use in delineating the clinical factors that determine the
optimal match between an individual patient profile and an
antidepressant’s pharmacologic profile, even within the
SSRI class.
b
sured in the sertraline group. This conclusion is supported
by the results from the week 8 completer analysis, because
patients who discontinued early were excluded from this
analysis, and the analysis revealed significant improvement in sertraline patients as compared to placebo patients.
Alternatively, the maximum allowed dose of 150 mg/day
sertraline may have been insufficient for some patients;
the sertraline maximum recommended dose is 200
mg/day.
The differences between citalopram and sertraline observed in this study were not apparent in a previous
comparative trial with no placebo control group (Ekselius
et al 1997); however, a prior active-controlled study of
citalopram and fluoxetine also demonstrated a significant
advantage for citalopram after 2 weeks of treatment (Patris
et al 1996) and also provided evidence of greater usefulness of citalopram in the treatment of anxiety symptoms
(Bougerol et al 1997). Previous placebo-controlled trials
of sertraline have detected significant antidepressant effects of sertraline at an earlier timepoint than was observed
in the present study (Lydiard et al 1997; Reimherr et al
1990), and an active-controlled sertraline–fluoxetine comparison revealed a trend toward greater antianxiety effects
of sertraline (Bennie et al 1995).
Given the variability of study populations even among
large multicenter trials, the results of the present study
may support the conclusion that some patients respond
differently to one SSRI than to another, and that different
multicenter trials may detect different aspects of this
phenomenon, rather than the conclusion that one SSRI is
clearly superior to another in a manner that is generalizable to the wide population of patients who take antidepressants. These results also support the anecdotal observations of clinicians that different patients have optimal
efficacy on different SSRIs, and argue for keeping the
broadest prescribing options open for clinicians to be able
to match the best agent with the best therapeutic outcome
for each individual patient.
Although all of the SSRIs share a presumed primary
mechanism of action, their individual postsynaptic binding
characteristics, pharmacokinetic profiles, and other secondary pharmacologic properties may be clinically relevant (Stahl 1998b). Participants in the trial discussed here
Coinvestigators for this multicenter trial were John Carman, Lorna
Charles, Eugene Duboff, James Ferguson, James Hartford, Ram Shrivastava, and Charles Wilcox. All coinvestigators, including the author,
received research grant support from Forest Laboratories for this trial.
The author is a consultant, lecturer, and recipient of research grants from
Forest Laboratories and H. Lundbeck A/S, the companies that market
citalopram. The author is also a consultant, lecturer, and recipient of
research grants from Pfizer Inc., the company that markets sertraline. The
author thanks John Rush and Bruce Pollock for their valuable comments
on the draft manuscript.
References
American Psychiatric Association (1994): Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC:
American Psychiatric Press.
Bech P, Allerup P, Gram LF, Reisby N, Rosenberg R, Jacobsen
O, Nagy A (1981): The Hamilton Depression Scale: Evaluation of objectivity using logistic models. Acta Psychiatr
Scand 63:290 –299.
Bennie EH, Mullin JM, Martindale JJ (1995): A double-blind
multicenter trial comparing sertraline and fluoxetine in outpatients with major depression. J Clin Psychiatry 56:229 –
237.
Bolden-Watson C, Richelson E (1993): Blockade by newly
developed antidepressants of biogenic amine uptake into rat
brain synaptosomes. Life Sci 52:1023–1029.
Bougerol T, Scotto J-G, Patris M, Strub N, Lemming O, Petersen
HEH (1997): Citalopram and fluoxetine in major depression:
Comparison of two clinical trials in a psychiatrist setting and
in general practice. Clin Drug Invest 14:77– 89.
Brown WA, Harrison W (1995): Are patients who are intolerant
of one selective serotonin reuptake inhibitor intolerant to
another? J Clin Psychiatry 56:30 –34.
Covi L, Rickels K, Lipman RS, McNair DM, Smith VK,
Downing R, et al (1981): Effects of psychotropic agents on
primary depression. Psychopharmacol Bull 17:100 –101.
Ekselius L, von Knorring L, Eberhard G (1997): A double-blind
multicenter trial comparing sertraline and citalopram in patients with major depression treated in general practice. Int
Clin Psychopharmacol 12:323–331.
Endicott J, Nee J, Harrison W, Blumenthal R (1993): Quality of
life enjoyment and satisfaction questionnaire: A new measure. Psychopharmacol Bull 29:321–326.
Fava M, Amsterdam JD, Deltito JA, Salzman C, Schwaller M,
Dunner DL (1998): A double-blind study of paroxetine,
fluoxetine, and placebo in outpatients with major depression.
Ann Clin Psychiatry 10:145–150.
Comparison of Citalopram and Sertraline
Greenblatt DJ, von Moltke LL, Harmatz JS, Shader RI (1998):
Drug interactions with newer antidepressants: Role of human
cytochromes P450. J Clin Psychiatry 59(suppl 15):19 –27.
Guy W (1976): ECDEU Assessment for Psychopharmacology,
Revised, DHEW Publication (ADM). Rockville, MD: National Institute of Mental Health.
Hamilton M (1959): The assessment of anxiety states by rating.
Br J Med Psychol 32:50 –55.
Hamilton M (1960): A rating scale for depression. J Neurol
Neurosurg Psychiatry 23:56 – 62.
Hyttel J, Arnt J, Sanchez C (1995): The pharmacology of
citalopram. Rev Contemp Pharmacother 6:271–285.
Jeppesen U, Gram LF, Vistisen K, Loft S, Poulsen HE, Brosen K
(1996): Dose-dependent inhibition of CYP1A2, CYP2C19,
and CYP2D6 by citalopram, fluoxetine, fluvoxamine, and
paroxetine. Eur J Clin Pharmacol 51:73–78.
Joffe RT, Levitt AJ, Sokolov ST, Young LT (1996): Response to
an open trial of a second SSRI in major depression. J Clin
Psychiatry 57:114 –115.
Jonghe F, Swinkels J (1997): Selective serotonin reuptake
inhibitors: Relevance of differences in their pharmacological
and clinical profiles. CNS Drugs 7:452– 467.
Kerr JS, Sherwood NJ, Hindmarch I (1991): The comparative
psychopharmacology of 5HT re-uptake inhibitors. Hum Psychopharmacol 6:313–317.
Lydiard RB, Stahl SM, Hertzman M, Harrison WM (1997): A
double-blind, placebo-controlled study comparing the effects
of sertraline versus amitriptyline in the treatment of major
depression. J Clin Psychiatry 58:484 – 491.
Montgomery SA, Asberg M (1979): A new depression scale
designed to be sensitive to change. Br J Psychiatry 134:382–
389.
Patris M, Bouchard J-M, Bougerol T, Charbonnier J-F, Chevalier
J-F, Clerc G, et al (1996): Citalopram versus fluoxetine: A
double-blind, controlled, multicentre, phase III trial in pa-
BIOL PSYCHIATRY
2000;48:894 –901
901
tients with unipolar major depression treated in general
practice. Int Clin Psychopharmacol 11:129 –136.
Raskin A, Schulterbrandt J, Rectig N, McKeon JJ (1969):
Replication of factors of psychopathology in interview, ward
behaviour, and self-report ratings of hospitalised depressives.
J Nerv Ment Dis 148:87–98.
Reimherr FW, Chouinard G, Cohn CK, Cole JO, Itil TM,
LaPierre YD, et al (1990): Antidepressant efficacy of sertraline: A double-blind, placebo- and amitriptyline-controlled,
multicenter comparison study in outpatients with major depression. J Clin Psychiatry 51(suppl B):18 –27.
Schmidt A, Lebel L, Koe BK, Seeger T, Heym J (1989):
Sertraline potently displaces 3H-(⫹)-3-PPP binding to sigma
sites in rat brain. Eur J Pharmacol 165:335–336.
Stahl SM (1998a): Basic psychopharmacology of antidepressants
(part 1): Antidepressants have seven distinct mechanisms of
action. J Clin Psychiatry 59(suppl 4):5–14.
Stahl SM (1998b): Not so selective serotonin reuptake inhibitors.
J Clin Psychiatry 59:343–344.
Thase ME, Blomgren SL, Birkett MA, Apter JT, Tepner RG
(1997): Fluoxetine treatment of patient with major depressive
disorder who failed initial treatment with sertraline. J Clin
Psychiatry 58:16 –21.
Thase ME, Lydiard B, Feighner J (1999, May): Citalopram
treatment of fluoxetine nonresponders. Presented at the annual meeting of the American Psychiatric Association, Washington, DC.
Wheatley DP, van Moffaert M, Timmerman L, Kremer CME
(1998): Mirtazapine: Efficacy and tolerability in comparison
with fluoxetine in patients with moderate to severe major
depressive disorder. J Clin Psychiatry 59:306 –312.
Zarate CA, Kando JC, Tohen M, Weiss MK, Cole JO (1996):
Does intolerance or lack of response with fluoxetine predict
the same will happen with sertraline? J Clin Psychiatry
57:67–71.