Introduction
Obsessive-compulsive disorder (OCD) is the fourth common neuropsychiatric disorder
worldwide and is a major risk for serious social disability and morbidity [1]
[2]. Serotonin reuptake inhibitors (SRIs) and cognitive behavioral therapy (CBT) are
recommended as safe and effective treatments for OCD. Despite being an effective non-pharmacological
treatment, CBT has several drawbacks such as limited access to CBT therapists, high
cost, and delayed clinical response [3]. SRIs are first-line pharmacologic treatment of OCD; however, only about half of
the patients respond to SRI therapy [4]
[5]
[6], and a considerable percentage show significant residual symptoms even after multiple
trials of SRIs [1]
[7]
[8]. The practice guidelines for SRI-refractory patients generally include augmentation
with antipsychotics [4]; however, patients often experience intolerable side effects [9]. This shows the importance of alternative approaches to OCD pharmacotherapy.
A variety of underlying mechanisms have been proposed in the development of OCD, such
as disruption of serotonergic function and dopamine system [10], and more recently there is evidence on the role of glutamatergic system in OCD
pathology [11]. Increased glutamate levels in cerebrospinal fluid, glutamatergic over-activity,
and polymorphism of gene coding N-methyl-D-aspartate (NMDA) receptor are shown to
contribute to OCD occurrence [11]
[12]
[13]
[14]. In addition, there is a large body of evidence on the clinical benefits of augmentation
therapy with glutamate-modulating agents such as riluzole, N-acetylcysteine, ketamine,
and amantadine in reducing symptoms in SRI-refractory OCD [15]
[16]
[17]
[18]
[19].
Memantine is a non-competitive antagonist of NMDA receptor and is approved by the
U.S. Food and Drug Administration for the treatment of Alzheimer’s disease [12]. It is reported to be safe and well tolerated at doses as high as twice the dosage
commonly prescribed for patients with Alzheimer’s disease [20]
[21]. Its most commonly reported side effects include dizziness, somnolence, confusion,
and headache that are usually mild and transient [22]
[23]. Although limited data on the off-label uses of memantine is available for various
psychiatric disorders, reported results have mostly been promising [12]. There are a number of controlled trials that showed 5–20 mg/day memantine augmentation
in moderate to severe OCD patients was significantly effective in reducing symptom
severity [24]
[25]
[26]
[27]. In a recent double-blind placebo-controlled trial by Ghaleiha et al., 89% remission
was observed after 8 weeks of memantine augmentation against 32% from placebo [25]. Another double-blind placebo-controlled study by Haghighi et al. showed treatment
response in 93% of patients receiving memantine augmentation against only 27% in the
placebo group [24].
Only a few studies including 2 case reports, and an open-label trial assessed memantine
augmentation in SRI-refractory OCD patients [28]
[29]
[30]. Aboujaoude et al. studied 15 OCD patients who had failed an average of 2.8 SRI
trials; 40% had failed augmentation with atypical antipsychotics and more than half
suffered from psychiatric comorbidities. Over a 12-week trial with 20 mg/day memantine
augmentation, 43% of patients achieved treatment response [30].
This study is the first double-blind randomized placebo-controlled trial that aims
to assess the efficacy of memantine augmentation in SRI-refractory OCD patients with
no psychiatric comorbidity. Given the safety and tolerability of memantine, the dose
of 20 mg/day was considered to achieve rapid relief of symptoms in refractory OCD
patients.
Methods
Study design
A prospective double-blind randomized placebo-controlled trial was conducted on SRI-refractory
OCD patients in Imam Khomeini Hospital in Ahvaz, Iran, from December 2015 to November
2016. Taking into account the trial’s inclusion and exclusion criteria, recruited
patients were randomized to 2 groups receiving either memantine in 10 mg tablets (Alzantin,
Sobhan Darou, Iran) or placebo twice a day for 12 weeks in addition to their existing
SRI therapy. The placebo tablets were made in the School of Pharmacy, Ahvaz Jundishapur
University of Medical Sciences, with the same appearance and packaging to those of
memantine.
All patients were provided with written informed consent as well as full explanation
of the protocol design. The consent form described the study, outlined the possible
risks, and indicated that an experimental medication or placebo would be consumed
daily. The protocol was approved by the ethics committee of Ahvaz Jundishapur University
of Medical Sciences (identifier code: IR.AJUMS.REC.1394.314) and was conducted according
to the Declaration of Helsinki. The trial was registered in the Iranian Registry of
Clinical Trials (www.irct.ir, registration number: IRCT2015110324853N1).
Yale-Brown Obsessive Compulsive Scale
The Yale-Brown Obsessive Compulsive Scale (Y-BOCS) is commonly used in clinical studies
to assess the severity of OCD and response to treatment [25]
[26]
[31]. It comprises 2 subscales including obsession and compulsion, each with 5 items
rated from 0 (none) to 4 (extreme). The subscales’ score range is hence 0–20, and
the total Y-BOCS range is 0–40. The Y-BOCS is rated by clinicians and has proven validity
and reliability for measuring OCD symptoms [32]
[33]. OCD severity is defined using Y-BOCS total score as extreme (score 32–40), severe
(score 24–31), moderate (16–23), and mild (8–15) [33]. In the present study, the Y-BOCS was used to quantify the severity of OCD and evaluate
treatment response, defined as≥35% reduction in Y-BOCS total score [34].
Inclusion and exclusion criteria
SRI-refractory OCD patients aged 18–40 years with Y-BOCS≥24 (severe to extreme) were
included. The diagnosis of OCD was based on the Diagnostic and Statistical Manual
of Mental Disorders, Fifth Edition (DSM-5). SRI-refractory was defined as failing
to at least 3 adequate trials of SRI therapy, 1 of which was clomipramine [7]. The adequate trial was considered as being stable for a minimum of 3 months at
the maximum tolerated dose of an SRI therapy [35]. In addition, patients had to be stable on their existing SRI therapy for at least
3 months prior to admission to the study.
Patients with the following conditions were excluded: (1) receiving CBT at the time
of enrollment; (2) additional anxiety disorder (such as panic disorder, post-traumatic
stress disorder, general anxiety disorder); (3) mood disorder (major depressive disorder
or bipolar disorder); (4) current drug or alcohol abuse or dependence; (5) any systemic
disorder (such as diabetes, hypertension, hyper or hypothyroidism, severe renal or
hepatic impairment); (6) a history of seizure; (7) pregnancy or breast-feeding; or
(8) a history of memantine use.
Sample size and statistical analysis
Based on data from a previous trial of memantine in severe OCD patients [24], considering a power of 80% and alpha=0.05, and assuming an attrition rate of 20%,
a sample size of 32 (16 in each group) was calculated.
IBM SPSS version 21.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analyses.
A p-value of 0.05 was considered as significant. Categorical variables were reported
as number (%) and continuous variables as mean±standard deviation (SD). Demographic
characteristics and baseline Y-BOCS scores were compared between the memantine and
placebo groups using t-test or chi-square, whichever was appropriate.
A mixed 4 (time)×2 (group) analysis of variance, with time as the within-subjects
factor and group (memantine vs. placebo) as the between-subjects factor, was run for
the Y-BOCS and its subscales. The homogeneity of variance was verified using Levene’s
test. Results of Greenhouse-Geisser correction were reported whenever Mauchly’s sphericity
test was significant. When the interaction between time×group was significant, post
hoc analysis based on pairwise comparisons was conducted.
Allocation, randomization, and blinding
The participants were randomized to receive either memantine or placebo, in addition
to their existing SRI therapy, using a simple computerized randomization program.
Staff responsible for preparation of trial medications and randomization process were
not further involved in the study. Throughout the study, all patients were visited
by the same psychiatrist who rated Y-BOCS scores, prescribed the trial medications,
and assessed the adverse effects. The patients, psychiatrist, and statistician were
blind to allocation.
Data collection
All patients were visited at baseline and then every 4 weeks during the 12-week trial.
At baseline, the following data were recorded for patients: age, sex, duration of
illness, marital status, and past medical and drug history. The Y-BOCS total score
and its 2 subscales (obsession and compulsion) were recorded at baseline and all subsequent
visits. Adverse drug reactions were also assessed and recorded at each time point.
Outcome measures
The primary outcome was the change in the Y-BOCS from baseline to the end of trial
in the 2 groups. The secondary outcome was defined as the number of individuals who
achieved treatment response (Y-BOCS reduction≥35%) at the end of trial.
Results
Patients’ baseline characteristics
A total of 58 SRI-refractory OCD patients were screened for eligibility in order to
recruit 32 individuals who met the inclusion criteria. The patients were randomized
into 2 groups of 16, but 1 from each group withdrew from the study before the end
of week 4 ([Fig. 1]). The withdrawal cases did not report side effects or worsening of mental status
as the cause of withdrawal. Among the remaining 30 patients aged between 18 and 40
(mean 30.7±5.7) years, 19 (63.3%) were female and 17 (56.7%) were married. Prior to
enrolment, patients had received between 3 and 4 (mean 3.5±0.5) adequate trials of
SRI therapy including clomipramine and at least 2 selective serotonin reuptake inhibitors
(SSRIs). At baseline, they had all been on a maximum tolerated dose of an SSRI for
at least 12 weeks ([Table 1]) and suffered from severe or extreme OCD with baseline Y-BOCS scores between 25
and 38 (mean 33.7±3.3); 23 (76.7%) patients suffered from extreme and 7 (23.3%) patients
from severe OCD.
Fig. 1 Randomization, treatment, and follow-up procedures.
Table 1 Baseline characteristics of patients.
|
Study groups
|
p-value
|
|
Memantine group (n=15)
|
Placebo group (n=15)
|
t-test
|
chi-square test
|
|
Age (years), mean±SD
|
30.6±6.8
|
30.7±4.7
|
0.97
|
|
|
Sex
|
|
|
|
0.70
|
|
Female
|
9 (60%)
|
10 (67%)
|
|
|
|
Male
|
6 (40%)
|
5 (33%)
|
|
|
|
Marital status
|
|
|
|
0.71
|
|
Single
|
6 (40%)
|
7 (47%)
|
|
|
|
Married
|
9 (60%)
|
8 (53%)
|
|
|
|
Y-BCOS, mean±SD
|
|
|
|
|
|
Obsession subscale
|
17.0±2.8
|
17.4±2.2
|
0.67
|
|
|
Compulsion subscale
|
16.8±1.9
|
16.0±2.2
|
0.26
|
|
|
Total score
|
33.9±3.5
|
33.4±3.2
|
0.59
|
|
|
Duration of illness (months), mean±SD
|
20.5±4.6
|
19.5±5.4
|
0.57
|
|
|
Concurrent SSRIs, dose range
|
|
|
|
0.98
|
|
Sertraline, 150–200 mg/day
|
2 (13.3%)
|
2 (13.3%)
|
|
|
|
Fluvoxamine, 200–300 mg/day
|
5 (33.3%)
|
6 (40%)
|
|
|
|
Citalopram, 60–80 mg/day
|
3 (20%)
|
3 (20%)
|
|
|
|
Fluoxetine, 70–80 mg/day
|
5 (33.3%)
|
4 (26.6%)
|
|
|
The baseline characteristics including demographic information, the type and dose
of the concurrent SSRI regimen, and the mean Y-BOCS scores were not statistically
different between the 2 groups ([Table 1]). During the trial, patients did not participate in any CBT session and there was
no alteration in patients’ treatment regimen including dose change or initiation/discontinuation
of medications.
Y-BOCS total score
[Table 2] shows that there was a significant difference across the 4 time points and between
the 2 groups in the Y-BOCS total score. Most importantly, there was a significant
interaction between time and group; hence, post hoc analyzes were conducted. It indicated
that there was no significant difference in the Y-BOCS total score between groups
at baseline (p=0.71) and at week 4 (p=0.92). However, the difference between the 2
groups became significant at weeks 8 and 12 (both with p<0.001). The means of Y-BOCS
total score reduction in the memantine and placebo groups were respectively 17.2%
versus −0.8%, p<0.001, by week 8 and 40.9% versus −0.3%, p<0.001, by week 12.
Table 2 Mean scores of Y-BOCS total and its obsession and compulsion subscales and results
of statistical analyses.
|
Score (mean±SD)
|
Time point
|
Study groups
|
|
|
Memantine group
|
Placebo group
|
Statistics: mixed 4 (time)×2 (group) ANOVA
|
|
Y-BOCS total
|
Baseline
|
33.93±3.56
|
33.47±3.29
|
Time: F(2.1,57.9)=191.0, p<0.001, η2=0.87 (L)
|
|
Week 4
|
33.8±3.63
|
33.67±3.22
|
Group: F(1,28)=16.5, p<0.001, η2=0.37 (L)
|
|
Week 8
|
28.13±3.93
|
33.67±2.92
|
Time×Group: F(2.1,57.9)=189.8, p<0.001, η2=0.87 (L)
|
|
Week 12
|
20.00±2.56
|
33.53±3.16
|
Greenhouse-Geisser epsilon (ε)=0.69
|
|
Y-BOCS obsession subscale
|
Baseline
|
17.07±2.81
|
17.46±2.26
|
Time: F(2.2,61.8)=142.1, p<0.001, η2=0.84 (L)
|
|
Week 4
|
17.06±2.84
|
17.40±2.26
|
Group: F(1,28)=13.0, p=0.001, η2=0.32 (L)
|
|
Week 8
|
13.73±2.79
|
17.26±2.09
|
Time×Group: F(2.2,61.8)=129.7, p<0.001, η2=0.82 (L)
|
|
Week 12
|
9.33±1.91
|
17.26±2.31
|
Greenhouse-Geisser epsilon (ε)=0.74
|
|
Y-BOCS compulsion subscale
|
Baseline
|
16.87±1.92
|
16.00±2.23
|
Time: F(2.0,55.1)=56.8, p<0.001, η2=0.67 (L)
|
|
Week 4
|
16.60±1.68
|
16.27±2.31
|
Group: F(1,28)=6.9, p=0.014, η2=0.20 (L)
|
|
Week 8
|
14.27±1.94
|
16.40±2.10
|
Time×Group: F(2.0,55.1)=68.2, p<0.001, η2=0.71 (L)
|
|
Week 12
|
10.47±1.55
|
16.40±2.13
|
Greenhouse-Geisser epsilon (ε)=0.66
|
Effect sizes are indicated with the partial eta squared (η2), with 0.01 ≤ η2 ≤ 0.059 indicating small (S), 0.06 ≤ η2 ≤ 0.139 indicating medium (M), and η2 ≥0.14 indicating large (L) effect size.
Considering the groups separately, the mean Y-BOCS total score did not change over
time in the placebo group, but it decreased significantly in the memantine group from
week 4 to week 8 (16.8%) and then from week 8 to week 12 (28.5%). [Fig. 2] shows the change in Y-BOCS total score for both groups across the 4 time points.
Fig. 2 Variation of mean Y-BOCS total score for both study groups across the four time points.
NS: non-significant; **p≤0.001.
Y-BOCS obsession and compulsion subscales
As shown in [Table 2], a similar trend was observed in the Y-BOCS subscales including obsession and compulsion
scores. Having seen significant interaction between time and group in both subscales,
the post hoc analysis showed that the difference in Y-BOCS subscales between the 2
groups was not significant at baseline and week 4 but became significant at weeks
8 and 12 (p-values were p<0.001). The means of Y-BOCS obsession subscale reduction
in the memantine and placebo groups were respectively 19.5% versus 0.88%, p<0.001,
by week 8 and 45.2% versus 1.1%, p<0.001, by week 12. The corresponding results in
the compulsion subscale were respectively 15.2% versus −2.9%, p<0.001, by week 8 and
37.6% versus −3.1%, p<0.001, by week 12.
Treatment response
By week 8, despite the significant reduction of Y-BOCS total scores in the memantine
group (17.2%), no patient in either of the groups met the criteria for treatment response.
At the end of trial (i. e., week 12), however, treatment response was observed in
11 (73.3%) patients. No patient in the placebo group met the criteria for treatment
response by week 12.
Adverse effects
There was no statistically significant difference between the 2 groups with respect
to adverse effects (p=0.87), as shown in [Table 3]. Reported adverse effects were mild and transient and did not result in major discomfort.
The 2 withdrawal cases were not due to adverse drug reaction.
Table 3 Reported adverse effects during the trial.
|
Adverse effect
|
Memantine group (n=15)
|
Placebo group (n=15)
|
|
Headache
|
2 (13.3%)
|
1 (6.6%)
|
|
Constipation
|
1 (6.6%)
|
3 (20%)
|
|
Nausea
|
1 (6.6%)
|
1 (6.6%)
|
|
Dizziness
|
2 (13.3%)
|
2 (13.3%)
|
|
Decreased appetite
|
1 (6.6%)
|
1 (6.6%)
|
Discussion
This study, to the best of the authors’ knowledge, is the first randomized placebo-controlled
trial of adjuvant 20 mg/day memantine in SRI-refractory OCD patients. The key finding
of the study was that adding memantine to the maximum tolerated dose of an SRI significantly
improved OCD symptoms without causing increased side effects.
The assessment of patients at the end of week 4 showed no significant improvement
in the Y-BOCS total score in either of the groups. However, by week 8, a significant
reduction in the mean Y-BOCS score was observed in the memantine group, while the
placebo group showed no improvement. Despite the noticeable Y-BOCS score reduction
in the memantine group, no patient met the criteria for treatment response.
From week 8 to week 12, another significant reduction in the mean Y-BOCS score was
seen in the memantine group that resulted in treatment response being achieved in
73.3% of patients. However, no improvement in the Y-BOCS score was observed in the
placebo group by the end of week 12, which was in fact expected given SRI-refractoriness
of the patients recruited in the trial. The results showed that a time-to-effect of
8 weeks was necessary to observe significant improvement in OCD symptoms, while treatment
response was only seen after 12 weeks of memantine augmentation.
The findings from the present study corroborate the results from previous studies
on memantine augmentation for OCD patients. A double-blind placebo-controlled study
by Haghighi et al. on severe OCD patients showed that low-dose (5–10 mg/day) memantine
added to standard SRI monotherapy not only accelerated the response time, but also
increased the overall response rate [24]. A considerable reduction in the mean Y-BOCS score was observed in the memantine
group at weeks 8 and 12, resulting in a significantly higher response rate versus
the placebo group (93% vs. 27%) at the end of week 12. Despite the title of study
implies that refractory patients were assessed, it was actually stated in the discussion
section that all recruited OCD patients were treated for the first time.
In a double-blind placebo-controlled study by Ghaleiha et al. on severe OCD patients,
concomitant administration of memantine (20 mg/day) and fluvoxamine (200 mg/day) versus
fluvoxamine monotherapy resulted in 100% treatment response in the former group against
32% in the latter group at the end of the 8-week trial [25]. In addition, remission (i. e., Y-BOCS ≤16) rate was 89% in the memantine against
32% in the placebo group. No patient in the present study achieved remission at the
end of trial. The shorter time-to-effect and high remission rate reported by Ghaleiha
et al. compared to the present study may well be due to the fact that the patients
recruited by Ghaleiha et al. were not treatment-refractory.
In the only previous study on the effects of memantine augmentation in SRI-refractory
OCD patients by Aboujaoude et al., treatment response was observed in 43% of patients
receiving 20 mg/day adjuvant memantine over a 12-week open-label trial. The patients
had failed an average of 2.8±1.8 SRI trials prior to admission to the study. The lower
response rate compared to that of the present study may be due to the presence of
patients with psychiatric comorbidities who were not excluded by Aboujaoude et al.
Given the safety and high tolerability of memantine, even at higher doses than those
usually used in clinical practice [20], the starting dose of 20 mg/day was considered for the current trial to achieve
faster onset of action [25]
[27]
[30]. One notable finding from the present study was that using the starting dose of
20 mg/day adjuvant memantine did not result in a higher rate of adverse effects. In
all the previous studies mentioned above, the maintenance dose was reached by upward
titration of 5 mg per week to reduce the potential risk of adverse effects. Nevertheless,
the present study showed that memantine dose titration may not be necessary in SRI-refractory
OCD patients.
A limitation of the current study was that patients suffering from other concomitant
psychiatric disorders were excluded. Although the effect of memantine in such patients
remained unclear, for standard clinical trials it is common to include comorbidity-free
populations. Another limitation was that the Y-BOCS was assessed every 4 weeks. Weekly
assessments can help to determine time-to-effect more precisely. In addition, adverse
effects were recorded at each visit by clinical interview and no formal psychometric
measure was used. Despite the relatively small sample size, statistically significant
results were obtained for both primary and secondary outcomes, which shows that the
results would be applicable to larger population.
In conclusion, the results from the present clinical trial do support the limited
evidence on the effectiveness of memantine as a safe and well-tolerated augmentation
in the treatment of SRI-refractory OCD. Current treatment strategies such as higher
doses of first-line therapies or antipsychotic augmentation need to be weighed against
the potential side effect burden. The promising findings from this study on the efficacy
of memantine augmentation in SRI-refractory OCD patients, in addition to its mild
and transient side effects, opens interesting debate on exploring memantine monotherapy
in OCD patients. Given the proven effectiveness of NMDA antagonists in OCD treatment,
it is recommended that future studies may consider memantine as monotherapy in non-medicated
OCD patients.
Funding: This work was funded by a research grant from Ahvaz Jundishapur University of Medical
Sciences in support of a Doctor of Pharmacy thesis by Setareh Razooghi (Grant number:
GP94097).
