Serotonergic Responsivity Obsessive-Compulsive: Disorder
Serotonergic Responsivity Obsessive-Compulsive: Disorder
Serotonergic Responsivity Obsessive-Compulsive: Disorder
Obsessive-Compulsive Disorder
Effects of Chronic Clomipramine Treatment
Joseph Zohar, MD; Thomas R. Insel, MD; Rachel C. Zohar-Kadouch, MD; James L. Hill, PhD; Dennis L. Murphy, MD
\s=b\ Clomipramine is a potent serotonin reuptake blocker that tients." This apparent paradox might be resolved if chronic
decreases the symptoms of obsessive-compulsive disorder treatment with serotonin reuptake blockers induces adapt¬
(OCD). To investigate whether clomipramine treatment in OCD ive down-regulation of serotonergic responsiveness in the
affects brain serotonergic responsiveness, metachloro- patients with OCD who benefit from this treatment. This
phenylpiperazine (mCPP), a selective serotonin agonist, and study was designed to investigate this possibility.
placebo were given under double-blind conditions to nine The tricyclic antidepressant clomipramine has been
patients with OCD before and after treatment with clomipra- shown to be a potent inhibitor of serotonin reuptake into
mine. Unlike our previous observations of a marked transient synaptosomes in vitro.18 In preclinical in vivo studies,
increase in obsessional symptoms and anxiety following clomipramine is significantly more potent in inducing the
0.5 mg/kg of mCPP, readministration of mCPP after four serotonin behavioral syndrome in rodents than is imipra¬
months of treatment with clomipramine did not significantly mine, amitriptyline, or desipramine.4 Although several
increase obsessional symptoms and anxiety. Similarly, the other tricyclic antidepressants are potent inhibitors of
hyperthermic effect of mCPP observed before treatment was serotonin reuptake, these generally are metabolized to
eliminated after treatment with clomipramine. These findings desmethyl compounds, with a loss of serotonergic potency.
are consistent with the development of adaptive subsensitivity Clomipramine is unique in that its major metabolite, des-
to the serotonergic agonist mCPP during clomipramine treat- methylclomipramine, retains a high potency for serotonin
ment. A similar alteration in the response to endogenous reuptake inhibition while also acting as a norepinephrine
serotonin may mediate clomipramine's antiobsessional ef- uptake inhibitor.19
fects. To investigate possible changes in brain serotonergic
(Arch Gen Psychiatry 1988;45:167-172) responsivity during clomipramine treatment, the behav¬
ioral, thermal, and endocrine responses to mCPP before
tricyclic antidepressant clomipramine hydro¬ and during clomipramine treatment were compared. We
Thechloride, potent
shown to reduce
a serotonin
reuptake blocker,
obsessive-compulsive (OC) symptoms
has been
in
chose mCPP as the serotonergic probe based in part on
studies indicating that mCPP readily displaces tritiated
serotonin from brain membrane homogenates and is
patients with obsessive-compulsive disorder (OCD).18
Other nontricyclic selective serotonin reuptake blockers, thought to act selectively on serotonin receptor sites.20*21 In
such as zimelidine,910 fluoxetine,1112 and fluvoxamine,13 have preclinical in vivo studies, mCPP leads to biochemical
also been reported to be effective treatments of OCD, a (decreased central serotonin synthesis and turnover20"22),
disorder otherwise refractory to psychopharmacologic in¬ behavioral (decreased food consumption15·23 and decreased
tervention." One might therefore hypothesize that admin¬ locomotion24), hormonal (increased prolactin, cortisol, and
istration of a serotonin agonist to patients with OCD would corticotropin concentrations25'28), and thermal (hyperther-
reduce OC symptoms, at least briefly. However, single-dose mia25) effects, all consistent with serotonin agonist activ¬
treatment with metachlorophenylpiperazine (mCPP), a ity.21 These effects of mCPP are blocked by serotonin
serotonin receptor agonist,1516 was recently observed to antagonists and are responsive to presumed alterations in
increase rather than decrease OC symptoms in these pa- serotonin receptor sensitivity15·25·26,28 but not to alterations
in adrenergic or dopaminergic systems.15*21 Recent investi¬
Accepted for publication July 2, 1987.
From the Laboratory of Clinical Science, National Institute of Mental
gations in human subjects indicate that mCPP produces
hormonal (increased prolactin, cortisol, and corticotropin)
Health, National Institutes of Health Clinical Center, Bethesda, Md.
Reprint requests to Beer-Sheva Mental Health Center, POB 4600, Beer- and temperature (hyperthermia) changes that are con¬
Sheva 84170, Israel (Dr Zohar). sistent with the effects of other serotonin agonists.29·30
imum change following mCPP or placebo administration from the P<.005) and anxiety (NIMH Global anxiety score, F[l,8] 16.46,
=
corresponding baseline measure, providing a measure of the P<.004). On one measure, the NIMH Global anxiety score, the
maximum change corrected for the baseline of the day on which the increase following mCPP administration during clomipramine
<&&
C _1
c
EÎ 20
¿F g-
<fi¿r Ö ·—·
<r Ç _i
re
fore (hatched bars) and during (dotted bars) clomipramine treat¬ ets
et¡D (
<o
10
ment. NIMH indicates National Institute of Mental Health; anxiety: _
re
treatment.
The time course of the behavioral symptoms was not different
150, 180, and 210 minutes following mCPP. Treatment with clo¬
mipramine was associated with a marked and significant increase
before and during clomipramine treatment, and the behavioral in baseline prolactin concentrations (mean±SD, 5.6±3.9 µg/L
changes roughly paralleled the time of the peak plasma mCPP [5.6±3.9 ng/mL] before treatment, 14.0±7.8 µg/L
concentration and the peak endocrine response. [14.0±7.8 ng/mL] during treatment, paired i9 5.54, P<.001). =
Thermal Responses However, the ability of mCPP to raise the plasma level of prolactin
was not significantly affected by clomipramine treatment as mea¬
During clomipramine treatment, mCPP administration was not sured by maximal increment over baseline (mean±SD, 12.7±7.6
associated with significant temperature changes CF[1,7] 0.35, =
µg/L [12.7±7.6 ng/mL] before treatment, 14.9±3.5 µg/L
P<.6), while mCPP induced hyperthermia when administered [14.9 ±3.5 ng/mL] after treatment, paired i8 0.81, not significant
=
mCPP only one (11%) of the nine patients in the clomipramine- [10.4±3.89 ng/mL], 4) patients with OCD (¿, 0.07, NS) that
= =
treated group reported chills or hot/flushed sensations compared did not change during clomipramine treatment (16.2 ±2.17 µg/L
with six patients (67%) who reported these symptoms before [16.2±2.17 ng/mL] for women, 13.3±4.17 µg/L [13.3±4.17 ng/mL]
treatment. for men).
A
Before During
o
E
I I 280 (10) Scale
Clomipramine
Treatment
Clomipramine
Treatment
Observer
CPRS-OC-5 4.8 + 0.8 1.9 + 0.7f
« S2 Global OC 5.9 ±0.7 4.0 + 0.4-J:
ra
o E m
ü Global Anxiety 4.5 + 0.3 3.6 + 0.6
re cd
Global Depression 4.4 + 0.5 2.7 ± 0.6t
<e. ~0·
Self-rating
Anxiety 14.4+1.8 10.8±3.5
Time, min Depression 12.9 + 2.8 5.3 ±2.5}.
Fig 4.—Mean SEM cortisol time-response
± (left) and distri¬
curve Dysphoria 5.0 + 1.4 3.0±2.4
bution of maximal cortisol increment over baseline (right) after
Euphoria 2.6 + 0.9 1.8 + 0.7
placebo and metachlorophenylpiperazine (mCPP) before and dur¬ Altered self-reality
ing clomipramine treatment (n 9). Pre-Rx indicates before treat¬
= 0.4 + 0.3 1.2 ±0.8
ment; Rx, during treatment. In figure at right, squares indicate men; Functional deficit 12.0±2.7 7.2 ±1.9
circles, women. There were no significant differences in maximal *CPRS-OC-5 indicates Comprehensive Psychiatric Rating Scale-Ob¬
cortisol increment over baseline. Analysis of variance revealed no
sessive-Compulsive five-item subscale; OC, obsessive-compulsive. Each
significant drug effect (mCPP vs placebo) either before or during scorerepresents the mean ± SEM of two baseline measurements taken in the
clomipramine treatment. morning, one on the metachlorophenylpiperazine study day and one on the
placebo study day.
tP<001, paired ( test of patients with obsessive-compulsive disorder
during clomipramine treatment.
tP<.05, paired f test of patients with obsessive-compulsive disorder
Plasma Cortisol Levels during clomipramine treatment.
Figure 4 summarizes the mean cortisol response after adminis¬
tration of mCPP before and during treatment with clomipramine.
The effect of mCPP tended to be different from placebo both before Table 2.—Peak Plasma Concentration of mCPP* Before and
and during treatment, although the overall mCPP effect did not
reach statistical significance (F[l,8] 5.24, P>.05). Clomipramine
=
During Clomipramine Treatment
treatment did not affect baseline cortisol levels (mean± SD, Before Clomipramine
nmol/L [9.8 ±3.6 During Clomipramine
270 + 100 µg/dL] before treatment; Treatment Treatment
300±120 nmol/L [10.9±4.2 µg/dL] during treatment), and the
mean ± SD maximal cortisol increments over baseline before Peak Plasma Peak Plasma
(170 + 90 nmol/L [6.2±3.3 µg/dL]) and during (180±140 nmol/L mCPP Time mCPP Time
[6.5 ±5.1 µg/dL]) treatment were similar. There were no gender Concen¬ After Concen¬ After
differences in the cortisol response to mCPP either before tration, mCPP tration, mCPP
(150 ±40 nmol/L [5.6 ±1.36 µg/dL] for men, 180 ±120 nmol/L Patient nmol/L Dose, nmol/L Dose,
[6.7±4.22 µg/dL] for women) or during (200±130 nmol/L
No. (ng/mL) min (ng/mL) min
[7.2±4.79 µg/dL] for men, 160±140 nmol/L [5.9±5.2 µg/dL] 239 (47.0) 150 426 (83.8) 90
for women) treatment. 155 (30.4) 210 551 (108.4) 120
Baseline Behavior 231 (45.4) 150 207 (40.6) 150
Ratings
47 (9.2) 180 157 (30.9) 210
Chronic treatment with clomipramine in patients with OCD was
associated with significant mean decreases in OC symptoms and 77 (15.2) 180 156 (30.7) 150
depression (Table 1). All the patients experienced a reduction in 113 (22.2) 210 23 (44.3) 150
their OC symptoms during clomipramine treatment as indicated by 122 (24.0) 150 156 (30.7) 150
measurements on two rating scales (CPRS-OC-5 and NIMH Global
128 (25.1) 150 283 (55.8) 150
OC rating scales). The mean percentage improvement in OC
192 (37.8) 150 284 (55.9) 210
symptoms was 32% with the NIMH Global OC scale and 60% with
the CPRS-OC-5 scale (mean± SD, 47.6% ±19.4%). However, no Total, 145 ±27.5 170 + 8.66 272 ±50.5 160+14.14
correlation was found between the reduction in OC symptoms as mean (28.5 + 5.41) (53.5 + 9.93)
measured by either scale or their mean and the decreased behav¬ ±SEMf
ioral or thermal responses to mCPP during clomipramine treat¬ *mCPP indicates metachlorophenylpiperazine.
ment. Plasma levels of at least 559 nmol/L (176 ng/mL) for fMean±SEM peak plasma mCPP concentrations were significantly
clomipramine and 1380 nmol/L (414 ng/mL) for desmethylclomipra- higher during clomipramine treatment than before treatment (fB 3.21. =
during clomipramine treatment. By using a peak response reported after mCPP administration17) and to improve if
(ie, subtracting the maximum change following mCPP or serotonin hyporesponsivity develops (as it often does dur¬
placebo administration from the corresponding baseline ing chronic clomipramine treatment40). However, if hyper-
measure), we attempted to adjust at least partially for these responsivity of the serotonergic system is associated with
baseline differences. the psychopathologic characteristics of OCD, we might also
Clomipramine treatment was associated with a very expect to see worsening of OCD symptoms during the initial
significant increase in baseline prolactin concentrations. phase of treatment with serotonin reuptake blockers, as
This effect of chronic clomipramine treatment was reported initially these medications increase serotonin content in the
more than ten years ago34 and does not occur during chronic synaptic cleft. A recent reanalysis of daily ratings of
administration of amitriptyline, although the basis for this patients from an earlier inpatient study8 provides prelimi¬
effect of clomipramine has not been studied. Possible expla¬ nary support for the notion that many patients with OCD
nations included sustained serotonergic stimulation of experience exacerbation of their symptoms during the first
prolactin release by clomipramine, a change in sleep cycle, three to five days of clomipramine treatment (J.Z., T.R.I.,
and partial dopamine receptor blockade; the last possibility D.L.M., unpublished findings, 1987).
is not an expected consequence of clomipramine administra¬ In conclusion, clomipramine treatment was associated
tion. with decreased serotonergic responsivity in patients with
An increased prolactin response to mCPP during clo¬ OCD compared with the same patients' responses to the
mipramine treatment might have been expected on the same serotonergic challenge (0.5 mg/kg of mCPP) before
basis of increased blood mCPP levels. Failure to observe treatment. This finding along with the increased respon¬
this increase was probably not due to a "ceiling effect." sivity of patients with OCD to mCPP17 is consistent with the
Based on prolactin response to neuroleptic agents35 and hypothesis that increased rather than decreased
thyroid releasing hormone,36 it appears that prolactin levels serotonergic responsiveness is associated with the psycho-
can go higher than the peak reached after mCPP adminis¬ pathologic characteristics of OCD.
tration during clomipramine treatment. The apparently
equivalent endocrine responses to mCPP during clomipra¬
mine treatment in the face of significantly higher plasma We thank Terry Tolliver, MS, and P. K. Narang, PhD, for measuring
mCPP concentrations may reflect hyporesponsivity of mCPP concentrations, Francis Szele for measuring prolactin and cortisol
prolactin secretion to serotonergic provocation, since pre- concentrations, and Dottie Drake for secretarial assistance.