Background

Obsessive-compulsive disorder (OCD) is among the most disabling neuropsychiatric conditions characterized by the presence of repetitive intrusive thoughts, impulses, or images (obsessions) and/or ritualized mental or physical acts (compulsions) [1]. OCD affects approximately 1–2% of adult general population worldwide [2]. It is associated with significant functional impairment, both due to the primary illness, as well high comorbidity with other psychiatric disorders. Abnormalities in serotonin and/or dopamine neurotransmission have been suggested to underlie the development of OCD [3, 4].

Recommended first-line pharmacotherapies for OCD are serotonergic antidepressants, such as selective serotonin reuptake inhibitors (SSRIs) and clomipramine [5, 6]. However, estimates suggest that around 30–60% of patients do not improve or show a partial response to adequate serotonergic antidepressant treatment, implying that serotonergic dysregulation may not be the one but rather one of many important mechanisms that are involved in the pathophysiology of OCD.

Recently, researchers have proposed that glutamatergic dysfunction, especially in the cortico-striato-thalamo-cortical (CSTC) circuitry may play a key role in the pathophysiology of OCD [7, 8]. Glutamate is the principal excitatory neurotransmitter in the central nervous system. It is also a precursor for gamma-amino butyric acid (GABA), the main inhibitory neurotransmitter in the brain, as well as for the amino acid glutamine and the antioxidant molecule glutathione [7]. Glutamate plays a vital role in various physiological processes including neuronal migration and cell maturation particularly by acting on N-methyl-D-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors [9]. Abnormally elevated or reduced glutamate is shown to have adverse effect on cortical migration. The striatum, one of the major components of CSTC circuitry, is the largest group of receptive neurons in the basal ganglia, receiving a large glutamatergic excitatory input from the cortex [10]. Evidently, the striatum is responsible for planning cognitive and motor actions [7]. Aberrant glutamatergic signaling between orbitofrontal cortex (OFC), anterior cingulate cortex (ACC) and striatum have been widely recognized to be associated with the development of OCD [11]. Interestingly, recent evidence has also shown in vivo evidence for glutamatergic control of presynaptic serotonin release in the striatum. It is well known that dysregulation of the striatal serotonergic system is a primary pathology in OCD [12].

The most direct evidence suggesting altered glutamate homeostasis in OCD derived from cerebrospinal fluid (CSF) studies. These early studies demonstrated that glutamate is excessive in the CSF of a subset of untreated patients with OCD [13, 14]. Additional studies using magnetic resonance spectroscopy (MRS) indicated that glutamate and related compounds are elevated in the basal ganglia and reduced in the anterior cingulate cortex in patients with OCD [7, 15]. There is also some evidence to suggest that polymorphisms in glutamate-associated genes may contribute to OCD risk [16]. Among the implicated glutamate-associated genes in OCD, the most consistent candidates are the SLC1A1 which encodes the neuronal glutamate transporter excitatory amino acid transporter 3 (EAAT3), and The SAPAPs (synapse associated protein 90/postsynaptic density-95-associated proteins)/ DLGAPs (disks large-associated proteins) which are key components of the postsynaptic complex that anchors and spatially organizes glutamate receptors [17, 18].

Further to the aforementioned evidence on glutamatergic dysfunction in OCD, the potential benefits of some glutamate-modulating agents such as riluzole, memantine, N-acetylcysteine (NAC), D-cycloserine, and ketamine have been demonstrated in the treatment of OCD [10, 19]. However, few writers have been able to draw on any systematic research into the potential utility of these agents. Hence, this paper will systematically review the research conducted on the clinical efficacy of glutamate-modulating agents in the treatment of patients with OCD, aiming to serve as a base for future studies in this area.

Methods

We conducted this systematic review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Protocols guidelines [20].

Search strategy and selection criteria

In this systematic review and meta-analysis, controlled clinical trials (irrespective of blinding and randomization) investigating clinical efficacy of glutamatergic drugs (irrespective of modes of administration, dosage, frequency and duration) in patients with OCD (irrespective of age, gender or race) were included (Appendix 1).

The primary search process was conducted in Web of Science, PubMed, Scopus, ScienceDirect, Cochrane library and Google Scholar databases based on the search strategies described in the protocol (Appendix 1) to gather the body of evidence available from original articles published up to 2021 in English. The first author conducted an electronic database search. Then, the titles and abstracts of studies initially selected were screened to eliminate duplicate citations and those that were obviously irrelevant. The full texts of the remaining studies were obtained for quality assessment, data collection and analysis.

Data extraction and quality assessment

After the initial screening, the full texts were reviewed by two independent researchers to include eligible articles according to the inclusion criteria.

Detailed data extraction was performed based on the pre-designed data extraction forms. Extracted information included the study design, name and address of the corresponding author, participants’ characteristics, interventions and outcomes. The methodological quality of the included studies was then evaluated according to the Cochrane risk of bias assessment tool [21]. In case of disagreement between authors, opinion was sought from a third author.

Evidence synthesis

We developed an evidence synthesis of the findings of the included studies using systematic approaches such as textual descriptions, tabulation, and transforming data into a common rubric using Review Manager (Version 5.3. Copenhagen: The Nordic Cochrane Centre, the Cochrane Collaboration, 2014). Missing data were handled using sophisticated statistical analysis techniques.

A meta-analysis was performed and the weighted average treatment effect was estimated. Heterogeneity across studies were evaluated by the chi-square statistic and calculation of I2(defined as I2 > 40% and/or chi-square statistic p < 0.1). A random-effects model was used in the case of statistical heterogeneity. Moreover, we applied a subgroup analysis in case of clinical heterogeneity.

Results

Description of included studies

We found 476 studies of interest in the initial electronic searches. We then excluded 111 duplicate citations using Endnote software and 204 articles due to obvious irrelevancy of their topics in primary screening (Fig. 1). In secondary screening of 161 full texts, we excluded 144 articles, and finally included 17 controlled trials with 759 patients with OCD in this systematic review (Table 1).

Fig. 1
figure 1

Flowchart of study selection for systematic review and meta-analysis of Glutamatergic Medications as Adjunctive Therapy for Moderate to Severe Obsessive-Compulsive Disorder in Adults

Full size image
Table 1 Controlled clinical trials reporting Glutamatergic drugs in OCD
Full size table

Our primary outcome measure was the mean difference in Yale-Brown obsessive-compulsive score (Y-BOCS) ratings before and after pharmacological intervention in experimental and control groups. All except one of the included studies reported Y-BOCS final scores of each group. In the study which these data were not reported, we considered the proportion of treatment responders (as defined by a 35% decline in Y-BOCS scores) in the experimental group compared to the placebo group. A treatment response was significantly more likely in the glutamate-mediating-augmentation group than in the placebo-augmentation group (z = − 3.83, P < 0.001). Details of the studies are described in Table 1. Diagram 1 illustrates the forest plot of analyses of the included studies. The results of the assessment of studies for six main biases are shown in Fig. 2. The Funnel plot graph for the studies included is shown in Fig. 3. As can be seen, the overall quality of studies was fair (Table 2).

Fig. 2
figure 2

Comparison of Y-BOCS score between patients with moderate to severe OCD groups and control groups in each of the studies and in overall meta-analysis

Full size image
Fig. 3
figure 3

Funnel Plot graph for the studies included in the Meta-Analysis. The asymmetry in the funnel plot may be due to heterogeneity of samples. It means that a subgroup of OCD patients may benefit more from glutamatergic drugs than other groups

Full size image
Table 2 Table of risk of bias for included studies considering Cochrane ‘Risk of bias’ tool. The overall quality of studies was fair
Full size table

Memantine

Memantine is one of the approved medications for moderate to severe Alzheimer’s disease. It is a non-competitive antagonist of NMDA receptor, one of the main receptors of glutamatergic system [22]. Memantine blocks the effects of sustained, pathologically elevated levels of glutamate that may otherwise lead to neuronal dysfunction [23].

Four of seventeen studies in this review included memantine as an adjuvant therapy. In three of these studies reporting final Y-BOCS scores, the mean difference of Y-BOCS score between two groups with 95% confidence interval was − 5.68[− 6.96, − 4.41] (P-value< 0.0001).

One of the studies only reported the remission rate of the responders (> 35% decrease in Y-BOCS score). In which, 89% of the patients in the memantine group compared with 32% of the patients in the placebo group achieved remission (χ2(1) = 13.328, P < 0.001) [24].

Minocycline

Minocycline is a known glutamatergic agent with therapeutic effects on neurodegenerative diseases which might be achieved through the blockade of glutamate-mediated excitotoxicity. Moreover, this antibiotic is known for its antioxidant and anti-inflammatory characteristics, which could further explain its neuroprotective effects. The beneficial role of minocycline in the treatment of schizophrenia, depressive and autistic symptoms is reported in previous research [25, 26].

In one of the studies, the efficacy of minocycline was assessed as an augmentative agent to fluvoxamine in the treatment of patients with OCD [27]. Significantly lower Y-BOCS scores were achieved in the minocycline group compared to the placebo group at the end of the study (t-score: − 2.84, P value: 0.0084).

L-carnosine

L-carnosine is a nutritional complementary agent with both antioxidant and glutamatergic properties. Carnosine reduces the glutamate levels in the central nervous system via upregulation of the glutamate transporter 1 [28,29,30]. One of the studies assessed the effect of L-carnosine as adjunct therapy to fluvoxamine in OCD and found significant decrease in final Y-BOCS score (t42:-2.62, P-value:< 0.001) [31].

Riluzole

Riluzole, an anti-glutamatergic agent, is mainly known as a treatment of Amyotrophic Lateral Sclerosis (ALS) [32]. A 10-weeks randomized placebo-controlled trial examined the efficacy of riluzole in the management of OCD. This study showed significant improvement in the patients treated with Riluzole [33]. Two studies were found which used Riluzole for the trial [34]. The reduction of Y-BOCS score at the end of studies were significantly lower than control group. [Z = 4.85 (P < 0.00001)].

N-acetylcysteine (NAC)

NAC is known as a regulator of glutamatergic system and can prevent pre-oxidant effect of glutamate. It has been proposed as a potential therapy for OCD since it can regulate the exchange of glutamate and prevent its pre-oxidant effects [35]. Four studies in this review included this medication. In the analysis of these studies, there was a significant decrease in the Y-BOCS scores in the experimental group [Z = 5.4(P: 0.000139)] [36,37,38,39].

Discussion

Many glutamate-modulating drugs have been reported to have clinical efficacy in the treatment of OCD. As previously mentioned, riluzole which is an anti-glutamate drug, can be effective in the treatment of refractory OCD. This therapeutic effect can be considered an evidence for the abnormal elevation of glutamate in the CNS of patients with OCD [40, 41].

Based on the findings of this review, memantine has strong evidence supporting its clinical efficacy in the treatment of OCD which is in agreement with previous reviews [42]. Additionally, there is promising evidence on the therapeutic effects of riluzole and NAC in other studies [43, 44]. There is also confirmative data on the potential utility of minocycline by Marinova et al. [19]. Other medications with different mechanisms have also been proposed to be effective in the treatment of OCD. For instance, topiramate (through gamma-Aminobutyric acid (GABA) and AMPA/kainite-type glutamate receptors) [45] and lamotrigine (through GABA and reduction of the presynaptic release of glutamate [46], have shown some efficacy in patients with OCD. Moreover, D-cycloserine, a partial agonist of NMDA receptor, has demonstrated some supporting evidence [10, 15, 34, 42, 44, 45], but we did not find any eligible study on these drugs. However, it is unfortunate that this study did not include data reported in peer-reviewed publications other than journal articles. Therefore, it is important to bear in mind the possibility of publication bias.

Conclusions

In summary, there is supporting evidence for glutamate-modulating drugs in treating moderate to severe OCD as an alternative or adjunctive therapy. In the present review we found evidence for several drugs such as memantine, NAC, minocycline, L-carnosine and riluzole. Further research is needed to determine neuroanatomical, neurochemical and basic genetics for the new line of treatments in OCD. The efficacy, effectiveness and risks associated with these glutamate- modulating drugs for the treatment of moderate to severe OCD should be further investigated. In future investigations, it would also be interesting to identify and analyze possible moderator variables.