Drugs 43 (Suppl, 2): 48-54.

1992
0012-6667/92/0200-0048/$3.50/0
© Adis International Limited. All rights reserved.
DRSUP3404

Review of Fluvoxamine Safety Database

W. Wagner, I B. Plekkenpol, I T.E. Gray, I H. Vlaskamp2 and H. Essers-
I Solvay Pharmaceuticals Inc., Marietta, Georgia, USA
2 Solvay Duphar B.V., Weesp, The Netherlands

Summary A review of the safety and tolerability of fluvoxamine in worldwide marketing studies in-
volving 24 624 patients, predominantly receiving fluvoxamine treatment in uncontrolled studies
in depression, has been conducted. There was a marked preponderance of female patients and
patients aged between 30 and 50 years. The majority of patients were treated for 6 weeks, with
the most frequent modal total daily dose being 100mg. The greatest proportion of adverse ex-
periences occurring, by COSTART body system, affected the digestive system (24.1%), the nerv-
ous system (23.7%), and the body as a whole (15.3%). The only adverse experience with an
incidence greater than 10% was nausea (15.7%), with somnolence (6.9%) and asthenia (6.2%) as
the next most frequent experiences. Notably, the rates of agitation and anxiety were only 1.4 and
1.3%, respectively. The incidences of adverse experiences increased with age, and were slightly
higher in females than males. 15.1% of patients discontinued treatment prematurely as a result
of adverse experiences, principally nausea, dizziness, vomiting, somnolence, abdominal pain, and
headache. The overall incidence of serious adverse events associated with fluvoxamine treatment
was 2.5%, and the incidence of overall suicidality, including suicidal ideation, overdose, and
intentional overdose as well as attempted and completed acts of suicide, was remarkably low at
0.8%.

We have conducted a review of the safety and
tolerability of fluvoxamine in marketing studies
worldwide, the main results of which are here pre-
sented publicly for the first time. The data assem-
bled represent probably the largest aggregate of in-
formation regarding safety ever amassed and
published for any antidepressant drug.
1. Methods
The data were analysed by a multinational task
force assisted by 6 contract research organisations,
with the entire review and data management being
completed in 13 months. In total, 25372 original
case record forms in 7 languages were collected and
subjected to a medical review, to provide the re-
quisite degree of rigour and consistency in the eval-
uation of the safety profile. Subsequently, 4069
narrative patient summaries were written and later
stored on an optical laser disc imaging system; all
safety data were double entered into intermediate
databases that reflected the layout ofthe study case
record form, then transferred into a standard in-
dividual study database format, and finally merged
into one global safety database. The data manage-
ment and the analysis were conducted and moni-
tored according to Good Clinical Practice (GCP)
guidelines, and employed the US Food and Drug
Administration (FDA) criteria for defining 'seri-
ous' adverse events.
Fluvoxamine Safety Database 49

2. Overview of Studies Table III. Demographic characteristics of the 24624 patients

treated with fluvoxamine

As table I indicates, the majority of studies, % of patients

namely 43, were conducted in depression, with 10
trials in a variety of other indications, such as pain Gender
Male 28.4
states, anxiety disorders, alcoholism etc., and 1 trial
Female 69.3
in obsessive-compulsive disorder. In total, 54 stud- No data available 2.3
ies were analysed, with 53 originating in Europe
Age (years)
and I in Pakistan.
< 18 0.1
18·30 14.1
3. Exposure to Fluvoxamine 31-50 44.1
51-64 25.5
The total cohort receiving fluvoxamine for which ~ 65 14.4
data have been reviewed and analysed is 24 624 No data available 1.7

patients. By far the greatest number of patients, ap-

proximately 23 500, received fluvoxamine in un-
controlled studies in depression, with only a very other indications (table II). Dosages employed
small proportion of patients treated with fluvox- ranged from 50 to 300mg per day, while the dur-
amine in controlled studies in depression and in ation ofthe studies varied between 4 and 52 weeks.

Table I. Overview of 54 studies 4. Patient Demographic Characteristics

Indication No. of studies
There was a marked predominance of female
Depression 43 patients, approximately 70%, in accordance with
Other indications 10
the general epidemiological distribution of the dis-
Pain order, and the largest proportion of patients were
Anxiety aged between 30 and 50 years (table III). The sex
Emotional instability and age distribution of patients was similar across
Alcoholism
studies.
Bulimia
Cardiovascular disease
Premenstrual syndrome
5. Duration of Treatment
Obsessive-compulsive disorder Table IV indicates that a considerable number
of patients - 11 428 - were treated for 6 weeks,
Table II. Exposure to f1uvoxamine by indication almost all of these in uncontrolled studies in
depression. This reflects the conventional design of
Indication No. of patients
marketing studies with antidepressants. The sex and
Depression age distribution of patients was similar across the
Controlled studies 818 respective duration intervals.
Uncontrolled studies 23395

Other indications 6. Dosage

Controlled studies 181
Uncontrolled studies 190 For all dose parameters - the modal, maximum,
Obsessive-compulsive disorder and mean total daily doses - the distribution of
Controlled studies 40 dosages was similar across studies, and the sex and
Total 24624 age distribution of patients was not significantly
different across dose ranges.
50 Drugs 43 (Suppl. 2) /992

Table IV. Duration of fluvoxamine treatment dose for 8 weeks' treatment or more was 100 to
Treatment duration (weeks) No. of patients
150mg in 71% of patients, and it is noteworthy that
this is a rather low dose range. Finally, the maxi-
<1 1156 mum total daily dose increased with increasing
1 1223
duration of treatment, reflecting the design of most
2 846
3 807
studies, which included a titration phase; a similar
4 1052 pattern was observed for the mean total daily dose.
5 1742
6 11 428a
7. Adverse Drug Experiences
7 2295
8·12 1580
13-51 515 Adverse experiences were analysed employing
52-103 88 the most conservative approach based on FDA reg-
'" 104 3 ulations, i.e. no causality judgements were made
but all recorded events were regarded as advers~
No data available 1889

a 98% in uncontrolled studies in depression.
The modal total daily dose represents the most
common or frequent dose offluvoxamine that was
maintained during the patient's participation in the
study. It reflects the attainment of a balance be-
tween clinical efficacy and the occurrence of un-
desired secondary effects. The majority of patients
received a relatively low modal total daily dose of
100mg (46%) or 150mg (16%).
The maximum total daily dose is defined as the
highest dose ever prescribed for a single patient for
at least I day. The maximum total daily dose was
similarly low in the majority of patients, at 100mg
(45%), 150mg (18%), or 200mg (14%).
The mean total daily dose of fluvoxamine (the
experiences; where some uncertainty existed, the
'worst case scenario' for the drug was applied.
Table V enumerates adverse experiences by CO-
START body system, and indicates that the great-
est proportion of adverse experiences occurring
affected the digestive system (24.1%), the nervous
system (23.7%), and the body as a whole (15.3%).
Table VI lists the overall incidence of adverse
experiences. The only symptom that may be de-
scribed as 'common', i.e. it occurred with an in-
cidence of greater than 10%, was nausea (15.7%).
The incidence of all other adverse experiences was
less than 10%; of these, somnolence (6.9%) and as-
thenia (6.2%) were the next most frequent symp-
Table V. Adverse experiences in the 24624 patients treated
with fluvoxamine, by COSTART body system

sum of each dose level attained by the patient, Body system Incidence (Ofo)
multiplied by the number of days at that level, di-
vided by the total duration of treatment in days) Digestive 24.1
Nervous 23.7
ranged between 100 and 150mg in 48% of patients
Body as a whole 15.3
exposed. In terms ofthis safety review, we consider Cardiovascular 3.5
the mean total daily dose to be a less reliable para- Skin 2.7
meter, since in essentially all studies, an initial ti- Special senses 2.1
tration phase was involved, which would tend to Musculoskeletal 1.7
Urogenital 1.6a
move the mean dose downwards relative to the
Metabolic/nutritional 1.5
maintenance dose at which patients were stabilised Respiratory 1.3
- i.e. this parameter tends to underestimate the true Haematic/lymphatic 0.1
mean dose of maintenance therapy. Endocrine < 0.1
In terms of the relationship between dose and a Denominator adjusted for gender-specific symptoms.
duration of treatment, the maintenance modal daily
Fluvoxamine Safety Database 51

toms. It is noteworthy that agitation and anxiety Table VII. Principal reasons for discontinuation in the 15.1 %
are represented at very low incidences of 1.4 and of patients withdrawing from fluvoxamine treatment as a result
of adverse experiences
1.3%, respectively. This is in marked contrast to
the adverse experience profile of other serotonin Symptom (COSTART term) Patients discontinuing
reuptake inhibitors. treatment (%)
The overall incidence of adverse experiences was Nausea 4.5
observed to increase with age, and the highest rates Dizziness 1.6
were observed in patients aged 65 years or more. Vomiting 1.5
An overall sex gradient was apparent, so that a Somnolence 1.2
Abdominal pain 1.1
slightly higher incidence was found in females.
Headache 1.1
Similarly, relatively more females discontinued
fluvoxamine treatment because of adverse experi-
ences (see below). In addition, some specific se- 8. Premature Discontinuation of
verity gradients were identified, so that in the Treatment
digestive and nervous systems symptoms were mild
or moderate rather than severe, whereas in the body Of the 24 624 patients receiving fluvoxamine,
as a whole symptoms were moderate or severe 17 133 (69.9%) completed their study treatment, as
rather than mild. compared with 545 of 723 (75.4%) on active con-
trol and 15 of 25 (60.0%) on placebo. The reasons
Table VI. Overall incidence of adverse experiences in the 24 624
for premature discontinuation of fluvoxamine
patients treated with fluvoxamine treatment included adverse experiences (15.1%),
administrative reasons (11.0%), lack of efficacy
Symptom(COSTART term) Incidence (%)
(2.1%), clinical improvement (2.1%), and intercur-
Common events (> 10% incidence) rent illness (0.2%). Demographic characteristics of
Nausea 15.7 patients were distributed similarly with regard to
Frequentevents (> 1-10%incidence) reasons for discontinuation, with the exception that
Somnolence 6.9 a relatively higher proportion of female than male
Asthenia 6.2 patients withdrew from treatment prematurely as
Headache 4.7 a result of adverse experiences (15.7% and 13.0%,
Dry mouth 4.6
respectively).
Insomnia 4.5
Abdominal pain 4.0 With respect to the 15.1 % of patients discontin-
Dizziness 3.8 uing prematurely as a result of adverse experi-
Nervousness 3.6 ences, 68.2% (2531 of 3709) withdrew during the
Tremor 3.3 first 2 weeks of treatment, and 72.4%(2684 of 3709)
Vomiting 3.2
withdrew on a low total maximum daily dose
Dyspepsia 3.2
Constipation 2.9 (lOOmg or less). This again reflects the design of
Diarrhoea 2.2 most studies, which included an initial titration
Anorexia 2.1 phase. The most frequent reasons for premature
Vertigo 2.0 discontinuation due to adverse experience were
Sweating 1.7
nausea, dizziness, vomiting, somnolence, abdom-
Palpitation 1.7
Abnormal thinking 1.7 inal pain, and headache (table VII).
Gastrointestinal disorder 1.4
Agitation 1.4
9. Serious Adverse Events
Anxiety 1.3
Serious adverse events were defined according
Malaise 1.3
Amnesia 1.1
to FDA criteria, namely adverse events that are
fatal, life-threatening, permanently disabling, or that
52
Table VIII. Serious adverse events in association with fluvox-
amine treatment. I. Overall incidence
Event Incidence (%)
Hospitalisation 2.0
Suicide attempt < 0.5
Death 0.28
Any other < 0.08
a Includes 18 suicides in studies of depression.
necessitate hospitalisation, as well as congenital
anomaly, cancer, and overdose. Thus, this is an
operational definition, based on outcome rather
than on severity. The overall incidence of any se-
rious adverse event in association with fluvox-
amine treatment was 2.5% as compared with 3.0%
with active control. The most frequent serious ad-
verse event was hospitalisation, occurring in 496
patients (2.0%), followed by suicide attempt (101
patients; < 0.5%), death (47 patients; 0.2%), and
any others « 0.08%) [table VIII). The 47 patients
who died during or after the studies included 18
patients in depression trials who committed sui-
cide. Of the remaining 29 deaths, many occurred
in elderly patients with predisposing medical con-
ditions. To reiterate, no attempt was made to as-
cribe causality, and the most conservative ap-
proach to interpretation was taken.
With respect to the 2.0% of patients hospital-
ised, 43.8% of these were hospitalised because of
adverse experiences, most commonly exacerbation
of depression or emergence of psychotic depression
(table IX). Another 19.0% were hospitalised after
the end of the study, usually for other medical rea-
sons; 14.3% were hospitalised because of ineffec-
tiveness, and 10.3% because of suicide attempt.
The incidence of overall suicidality, which in-
cludes suicidal ideation and tendencies, overdose
and intentional overdose as wen as attempted and
completed acts of suicide, was 0.8% with fluvox-
amine, 0.7% with the active reference compounds,
and 0% with placebo. However, the patient popu-
lations in the 2 latter groups were too small (723
and 25 patients, respectively) to permit direct com-
parison of incidences. Table X provides further de-
Drugs 43 (Suppl. 2) 1992
tail regarding suicidality in association with flu-
voxamine treatment. Suicide attempts were made
by 101 « 0.5%) patients, including 98 patients with
depression and 3 patients with chronic alcoholism.
The incidence of completed suicide was 0.08%,
representing 18 patients being treated for depres-
sion. Thus, the incidence of suicidality with flu-
voxamine was remarkably low, and compares very
favourably with the published literature concern-
ing other antidepressant compounds.
Other serious adverse events recorded in asso-
ciation with fluvoxamine treatment occurred with
an incidence of less than 0.08%. The most clinic-
any significant among these were carcinoma (19
patients), convulsion (n = 11), syncope (n = 10),
myocardial infarction (n = 9), coma (n = 4), hep-
atitis (n = 3), cerebrovascular accident (n = 3), gas-
trointestinal haemorrhage (n = 2) and alcoholic
neuritis (n = 1). In most cases, pre-existing medical
conditions were apparent, and in others there was
insufficient data to determine the patient's history.
Notably, the incidence of convulsions was ex-
tremely low, confirming that fluvoxamine does not
possess proconvulsant activity. One of the II
patients had a prior history of epilepsy, and in 4
patients therapy with fluvoxamine was continued
after a convulsion with no subsequent attacks.
Analysis of laboratory values from 17 studies in
1630 patients, and of vital signs in 37 studies in
Table IX. Serious adverse events in association with fluvox-
amine treatment. II. Hospitalisation (overall incidence 2%)
Incidence (%j
among patients
hospitalised
Reason for hospitalisation
Adverse experience 43.8
(most frequently depression.
psychotic depression, nausea.
agitation, worsening of clinical state,
vomiting, anxiety, malaise)
Ineffectiveness 14.3
Attempted suicide 10.3
Other reasons 11.0
Unknown 1.6
Hospitalisation after completion of stUdy 19.0
Fluvoxamine Safety Database
Table X. Serious adverse events'In association with fluvoxamine treatment. III. Suicidality
Parameter Incidence (%)
fluvoxamine
(n = 24624)
Overall suicidality 0.8
Suicide attempt < O.Sa
Suicide 0.08b
a 98 patients with depression and 3 patients with chronic alcoholism.
b 18 patients with depression.
9689 patients, did not detect any consistent uni-
directional effect of treatment. No cases of Zirnel-
dine syndrome, bleeding syndrome or Guillain-
Barre syndrome were identified in any of the stud-
ies reviewed.
10. Conclusion
Fluvoxamine was safe and well tolerated in the
majority of the 24 624 patients in worldwide mar-
keting studies analysed in this review. No unex-
pected serious adverse events were identified.
Within the adverse experience profile, nausea was
the only commonly occurring symptom, while
stimulating or activating symptoms were less
prominent than sedating symptoms, Finally, the
incidence of suicidal ideation and of attempted or
completed suicide in association with fluvoxamine
treatment appeared to be particularly low.
Correspondence and reprints: Dr med. W. Wagner, Solvay Pharo
maceuticals Inc., 901 Sawyer Road, Marietta,GA 30062, USA.
Discussion
Dr C. V. De Blecourt: I am aware of reports that
concomitant treatment with fluvoxamine may
double or even triple plasma concentrations of
clomipramine, and similarly theophylline. Do you
have any systematic data regarding such interac-
tions?
Dr W. Wagner: Fluvoxamine undergoes exten-
sive hepatic metabolism and therefore an interac-
53
active control placebo
(n = 723) (n = 2S)
0.7 None
0.6 None
0.3 None
tion with other extensively metabolised com-
pounds is possible. We did not identify drug
interactions in this database of marketing studies,
but it must be acknowledged that the designof these
studies in most cases limited or excluded concom-
itant medications other than benzodiazepines and
excluded patients with certain concurrent diseases.
In the worldwide spontaneous reporting system
there are a number of reports of interactions with
propranolol, theophylline, and also tricyclic anti-
depressant drugs.
Dr Y. Lecrubier: What is meant by 'administra-
tive reasons' for withdrawal from study?
Dr Wagner: This mainly refers to patients lost
to follow-up for reasons unrelated to the study
medication,such as moving to another area. Where
any suspicionexisted that the drug could have been
implicated, the 'worst case scenario' was applied
and the patient was recorded as a withdrawal due
to adverse experiences caused by the study drugs,
DrLecrubier: You mentioned that the incidence
of side effectswas higher in females than in males.
Were there side effects specific to female patients?
Dr Wagner: No, the occurrence of particularside
effects was in general distributed similarly across
male and female patients. We cannot account for
the higher overall incidenceamong female patients,
but in my view it is likely that sex differences in
bodily perception are involved.
DrLecrubier: Did you establish any relationship
between the incidence of side effects and efficacy?
Dr Wagner: We did not attempt to appraise ef-
ficacy since most of these studies, being Phase IV
54
trials, were not conducted subject to the guidelines
for Good Clinical Practice (GCP), and could not
have been considered to provide reliable data re-
garding efficacy. However, the fact that the major-
ity of patients received a mean maintenance dose
of 100mg appears to reflect simply the efficacy of
this dose; although the titration schedule in early
marketing studies provided for doses up to 300mg,
in practice physicians seldom titrated the dose be-
yond 150mg, and patients withdrawing from treat-
ment mostly did so on the lower doses.
Drugs 43 (Suppl. 2) 1992
Dr G. Beaumont: Clinically significant sexual
side effects such as delayed ejaculation and
female anorgasmia have been described with some
of the serotonin reuptake inhibitors, for instance
fluoxetine. What is the incidence of sexual side ef-
fects with fluvoxamine?
Dr Wagner: Less than 0.1% in this database.
Moreover, there were no consistent unidirectional
effects in these rare cases, e.g. some patients re-
ported increased, others decreased, libido, so that
it is difficult to attempt any rigorous conclusion.