Specialty knowledge and competency standards for pharmacotherapy for adult obsessive-compulsive disorder

Abstract

Obsessive-compulsive disorder (OCD) affects approximately one person in 40 and causes substantial suffering. Evidence-based treatments can benefit many; however, optimal treatment can be difficult to access. Diagnosis is frequently delayed, and pharmacological and psychotherapeutic interventions often fail to follow evidence-based guidelines. To ameliorate this distressing situation, the International OCD Accreditation Task Force of the Canadian Institute for Obsessive-Compulsive Disorders has developed knowledge and competency standards for specialized treatments for OCD through the lifespan. These are foundational to evidence-based practice and will form the basis for upcoming ATF development of certification/accreditation programs. Here, we present specialty standards for the pharmacological treatment of adult OCD. We emphasize the importance of integrating pharmacotherapy with clear diagnosis, appreciation of complicating factors, and evidence-based cognitive behavioral therapy. Clear evidence exists to inform first- and second-line pharmacological treatments. In disease refractory to these initial efforts, multiple strategies have been investigated, but the evidence is more equivocal. These standards summarize this limited evidence to give the specialist practitioner a solid basis on which to make difficult decisions in complex cases. It is hoped that further research will lead to development of a clear, multi-step treatment algorithm to support each step in clinical decision-making.

1. INTRODUCTION

Obsessive-compulsive disorder (OCD) affects approximately one person in 40 worldwide and is a source of enormous suffering (Ruscio et al. 2010; Kessler et al. 2012). Evidence-based treatment strategies, both pharmacotherapeutic and psychotherapeutic, have been developed over the past several decades and can provide substantial relief to a majority of patients (Koran et al. 2007; NICE, 2006; Koran and Simpson, 2013; Phillips and Stein, 2015; Fineberg et al. 2016; Skapinakis et al. 2016). Nevertheless, diagnosis is frequently delayed, and specialist treatment is often difficult to access. Delayed, suboptimal, and even counterproductive treatment can compound the suffering of affected individuals (Jenike, 2004). There is thus an urgent need to disseminate and standardize quality care for individuals with OCD.

The International OCD Accreditation Task Force.

The OCD Accreditation Task Force (ATF) is an international group of experts in the diagnosis, pathophysiology, and treatment of OCD who have banded together to formalize core knowledge domains and competencies that constitute specialized expertise in the assessment and treatment of obsessive-compulsive disorder (OCD). This effort has been spearheaded by the Canadian Institute for Obsessive-Compulsive Disorders (CIOCD, www.ciocd.ca) and has grown to include members from 14 nations (Sookman et al. 2015).

The Task Force initiative consists of four phases. The first phase consisted of the preparation of a series of review articles that summarize the literature on the assessment and treatment of OCD (Fineberg et al. 2015; Franklin et al. 2015; Ivarsson et al. 2015; McKay et al. 2015; Sookman and Fineberg, 2015). The second phase, of which the current manuscript is a component, seeks to formalize core knowledge domains and competencies for the treatment of OCD. The third phase, which has not yet been initiated, will establish procedures whereby these standards can be taught and assessed, with the ultimate goal of creating a mechanism to provide certification/accreditation for qualified practitioners.

Scope and structure of the paper.

This paper is divided into broad sections on core psychopharmacology-related knowledge domains that the specialist practitioner should know, and core assessment and treatment competencies that the practitioner should master. Each of these broad areas is further subdivided into topical subsections. Within each subsection, after orienting introductory material, core knowledge domains/competencies are presented in tabular form, with citations where further information may be sought. Principles undergirding psychotherapy are reviewed briefly, as are targeted somatic treatments such as transcranial magnetic stimulation (TMS), deep brain stimulation (DBS), and targeted lesions.

Existing treatment guidelines and review articles are cited heavily, often in preference to all but the most seminal primary literature, as they are likely to be of most relevance to the specialist practitioner. Most tabular entries include an assessment of the type of evidence upon which a recommendation is based.

2. METHODS

The International Obsessive Compulsive Disorders Accreditation Task Force (ATF) has been created by the Canadian Institute for Obsessive Compulsive Disorders. This second phase of the ATF initiative involves establishment of core knowledge and competency standards for specialty treatment of OCD through the lifespan (ATF Chair, Debbie Sookman PhD and Co-Chair David Veale, MD). This paper focuses on pharmacotherapy and other somatic treatments in adults.

The ATF process seeks to integrate the best available evidence with the expert opinion of senior specialist clinicians and researchers. The first author and the leadership of the ATF assembled an international group of specialist clinicians with deep expertise in the diagnosis and pharmacological treatment of OCD. This group generated a list of general areas of knowledge and competence, with reference to published guidelines (National Institute of Clinical Excellence, 2006; Koran et al. 2007; Koran and Simpson, 2013; Skapinakis, 2016a), other summaries of the relevant literature (e.g. Pittenger, 2017; Phillips and Stein, 2015), and their own clinical expertise. This list was refined by the group through several iterations; topics were then distributed to group members for further elaboration.

Each section was developed by one or more members of the ATF Adult Pharmacology group, with reference to published guidelines, other literature (as cited throughout the document), and their own clinical experience. These sections were assembled by the first author then distributed through the full group for edits and comments. Comments were incorporated by the first author, and sections were distributed for a second round of commentary and edits by the full group. The full document was then formatted following ATF protocols and submitted to the central ATF leadership for final review and edits.

Tabulated knowledge and competency standards are operationalized as clinician abilities with specification of evidence for each standard. The supporting literature is coded in the tables that follow, according to the following scheme developed for standardization across papers in the ATF phase two series:

We note that this is a different categorization of the literature than the evidence grading systems that are sometimes used in treatment guidelines. This is in keeping with our goal, which is not to provide another treatment guideline, but rather to enumerate the domains of knowledge and competencies that a specialist practitioner should master. Please see the Introductory paper of this series (Sookman et al, 2021a) for elaboration of the distinction between current practice guidelines and ATF standards.

3. SPECIALIZED KNOWLEDGE DOMAINS

3.a. Presentation, diagnosis, and clinical characteristics

3.a.i. Basic phenomenology, diagnosis, and symptom heterogeneity

Overview

OCD is characterized by obsessions and compulsions; these are defined below. While either is sufficient to support the diagnosis, most patients have both (Shavitt et al. 2014). There is typically a functional relationship between the two; compulsions are often experienced as neutralizing the anxiety or discomfort caused by obsessions.

This introductory section provides an overview of the phenomenology of OCD, with particular focus on aspects that may inform pharmacotherapeutic treatment.

Definitions of key terms

Obsessions:

Intrusive, repetitive, distressing thoughts, images or urges that are usually recognized as unrealistic or excessive, that cause distress or impairment or are time consuming, and that the sufferer typically makes some effort to resist or control.

Compulsions:

Repeated, often ritualized thoughts or actions that an individual feels driven to perform in response to an obsession, or according to rules that must be applied rigidly, in an effort to mitigate anxiety or distress. Importantly, compulsions need not be observable physical acts but can be patterns of thought, such as counting, mental prayer or other recitations, or mental review in response to obsessions. Such mental compulsions are not infrequently miscategorized by patients as “pure obsessions”.

Insight:

The degree to which an individual with OCD recognizes that her or his obsessions and compulsions are irrational or inaccurate. Insight can vary among patients, or in the same patient over time, from excellent insight to absent insight (i.e. delusional beliefs); the majority of patients have good or fair insight. Poor insight is associated with a poorer prognosis. Because of its clinical importance, insight was added as a dimensional specifier to the diagnosis of OCD. DSM-5 allows three levels of insight: good/fair, poor, and absent insight/delusional beliefs. ICD-11 reduces this to 2 levels of insight: good/fair and poor/absent.

Tics:

Non-rhythmic, repeated, semi-voluntary movements that are often purposeless or fragmentary, often undertaken in response to a sense of tension or discomfort. Tics are commonly seen in OCD and may be associated with differential response to pharmacotherapy. Because of their association with treatment response, the presence of tics was added as a specifier to the diagnosis of OCD in DSM-5.

Avoidance:

Modification of behavior to escape an actual or feared negative outcome. A distinction may be drawn between passive avoidance and active avoidance; both are prominent in OCD. Active avoidance denotes the execution of a behavior in response to, and to escape, a negative affective state (e.g. anxiety, disgust) or a cue that may signal such a state (e.g. the presence of a contaminant). Most compulsions can be considered a form of active avoidance. Passive avoidance denotes the constriction of behavior in order to avoid exposure to a negative affective state, or cues associated with it – for example, refusal to leave the house in order to avoid potentially threatening situations, or refusal to touch anything in order to avoid the possibility of contamination. Some patients can largely avoid obsessions and feelings of anxiety through extensive passive avoidance; this avoidance may itself cause profound disability by constricting patients’ behavioral repertoire.

Incompleteness:

The uncomfortable sense that an action is not right or has not been finished properly, or that a situation is lacking some critical element. In OCD, this discomfort can be the primary affective driver of symptoms for some patients, independent of anxiety or disgust (two other common affective drivers), especially in conjunction with ordering compulsions, repeated actions, magic numbers, and comorbid tics. A similar concept is that of ‘not-just-right’ sensations, which may drive actions to be repeated.

Contamination:

In the context of OCD, this refers to excessive or irrational fear that one is or might be contaminated. Contamination obsessions may focus on infectious agents in general, by a specific agent (e.g. HIV, Ebola), by a disturbing or disgusting agent (e.g. dirt; bodily fluids; sticky substances), or by a non-physical contaminant (e.g. evil). Contamination obsessions are typically associated with washing compulsions and/or with avoidance. Contamination obsessions may be associated with anxiety or with disgust.

Fear-of-harm:

In the context of OCD, this refers to excessive or intrusive fears that a dreaded outcome, either specific or inchoate, may arise from one’s action or inaction. Fear-of-harm is typically associated with checking compulsions, which can be conceptualized as a form of active avoidance. The association of a particular fear with a particular action may be logical (e.g. checking the stove to make sure it is off) or irrational (e.g. reciting numbers in a ritualized way to avoid the death of a loved one).

Symmetry:

In the context of OCD, this refers to a problematic preoccupation with the idea that one’s surroundings or actions must be symmetrical, balanced, or ‘just right’. A failure to achieve symmetry or balance is often associated with a sense of incompleteness rather than with anxiety or disgust; the discomfort may be inchoate, rather than associated with a particular fear of negative consequences.

Disgust sensitivity:

Some OCD symptoms, especially in the domain of contamination/cleaning, are associated with disgust, rather than fear-of-harm or symmetry, as a motivating emotion. Elevated sensitivity to disgust has been hypothesized to be a core abnormality in some cases.

Sensory sensitivity/sensory phenomena:

Some individuals with OCD have a sensitivity to environmental and bodily sensations that leads to great discomfort; compulsions may be associated with an effort to respond to or eliminate these uncomfortable sensations.

Hoarding.

The excessive accumulation of belongings and/or the inability to discard belongings to the extent that accumulated clutter impairs the normal use of living spaces. Hoarding symptoms were traditionally considered a symptom of OCD and are treated as such in older literature, but over the first decade of the 21^st century hoarding was recognized to differ in important ways from other OCD symptoms. For example, individuals with primary hoarding typically have low subjective distress despite clear dysfunction, poor insight, and poor response to standard OCD treatments. These differences were recognized and formalized in 2013 with the adoption of hoarding disorder as a distinct nosological entity in DSM-5 and ICD-11. Because of this distinction, the treatment of hoarding disorder is not extensively discussed in this document. However, excessive accumulation may in some cases still be a symptom of OCD, when it is clearly driven by irrational or excessive anxiety-associated cognitions that meet the definition of an obsession – for example, excessive accumulation of trash due to contamination obsessions (Pertusa et al. 2010).

Level of evidence

An extensive phenomenological literature describes the presentation of OCD and the different forms that obsessions and compulsions may take. Empirical work has examined distinctions between different types of obsessions and compulsions, their dimensional organization, and their evolution over time. Epidemiological studies have characterized the prevalence of these symptoms in the population and the associated morbidity and comorbidity. Descriptive studies, and some theoretical and empirical work, have described the functional association of obsessions and compulsions with each other. However, expert opinions differ as to whether obsessions drive compulsions or vice versa in the ontogeny of OCD, and as to whether the negative affective states drive or result from the cognitive and behavioral symptomatology. These issues are largely moot by the time of clinical presentation, as all components typically co-occur and reinforce one another.

Recommendations for future research

The heterogeneity of OCD symptoms remains both fascinating and vexing. Research is needed to identify the best ways to characterize this heterogeneity and to identify characteristics – phenomenological, psychological, biological, or other – that can meaningfully guide prognosis and treatment.

The ability to:

1. Define obsessions and compulsions

Evidence: TG: Koran et al 2007, Bandelow et al 2012, American Psychiatric Association 2013b, Baldwin et al 2014, Katzman et al 2014

2. Characterize functional relationships between obsessions and compulsions, where present, and the role of negative affective states (anxiety, disgust, and incompleteness) in their maintenance

Evidence: TG: Koran et al 2007 • OR: Phillips and Stein 2015 • EO: Penzel 2017

3. Describe primary subtypes of obsessions and compulsions, and the proposed dimensional structure of this symptom heterogeneity in adults

Evidence: TG: Koran et al. 2007 • MA: Bloch et al. 2010 • EO: Penzell 2017

4. Describe the role of active and passive avoidance in OCD, and attendant disability

Evidence: TG: Koran et al 2007 • PS: Abramowitz et al 2009, McGuire et al 2012

5. Describe the association between tics/Tourette disorder and OCD, and the relevance of the tic specifier to the diagnosis in DSM-5

Evidence: TG: Koran et al. 2007, American Psychiatric Association 2013b • OR: Kirkanski et al 2013

6. Describe variable insight in OCD, and the relevance of the dimensional insight specifier in DSM-5 and ICD-11

Evidence: TG: Koran et al 2007, American Psychiatric Association 2013b, World Health Organization 2018 • OR: Hamblin et al 2017 • CSS: Phillips et al 2012

7. Describe the typical natural history of OCD symptoms, and the concept of clinical staging, which describes how early subclinical symptoms and risk factors may lead to later development of more significant illness and implies that early intervention may be of particular value

Evidence: ThP: Fontenelle and Yücel 2019

8. Understand and explain that childhood-onset and adult-onset OCD may represent distinct clinical entities; childhood-onset OCD has greater familiality and heritability, is more often comorbid with Tourette syndrome, and is more commonly seen in males.

Evidence: OR: Bloch 2017

3.a.ii. Comorbidity

Overview

Lifetime comorbidity is common in OCD. In some clinical settings, up to 90% of patients present with at least one comorbid diagnosis. Comorbidity rates are somewhat lower, but still substantial, in epidemiological studies. Common comorbidities include mood and anxiety disorders, obsessive-compulsive personality disorder (OCPD), Tourette disorder and other tic disorders, and the disorders identified as ‘Obsessive-Compulsive and Related Disorders’ (OCRDs) in DSM-5: hoarding disorder, body dysmorphic disorder (BDD), excoriation (skin-picking) disorder, and trichotillomania (hair-pulling disorder). Comorbidity is also seen, but is challenging to assess, with conditions that overlap phenomenologically with OCD, such as illness anxiety disorder, somatic symptom disorder, and certain eating disorders. Comorbidity with psychotic disorders and autism is less common but can produce clinical and diagnostic challenges. Comorbidity can complicate clinical presentation, diagnosis and assessment, patient engagement and motivation for treatment, treatment selection, and treatment course.

Definitions of key terms

Comorbidity –

the presence of more than one diagnosis, typically (in this context) more than one DSM-5 or ICDM-11 psychiatric diagnosis.

Obsessive-compulsive and related disorders (OCRDs) –

The cluster of disorders that DSM-5 and ICD-11 define as taxonomically related to and classified in the same chapter as OCD. In addition to OCD itself, the OCRDs in DSM-5 are: body dysmorphic disorder (BDD), hoarding disorder, skin-picking (excoriation) disorder, and trichotillomania (hair-pulling disorder). ICD-11 includes these disorders as well as illness anxiety disorder (hypochondriasis) and olfactory reference disorder. Of note, tic disorders could also have been included in this cluster, given their many similarities to OCD. Although there is utility in thinking of the OCRDs as a category, the cluster should be considered provisional; future research will further elucidate the similarities and differences among these disorders and clarify how they should best be categorized.

Strength of evidence

Epidemiological studies have established rates of comorbidity for OCD in the general population. The largest studies are from the United States (e.g. the Epidemiological Catchment Area study and its replication; Ruscio et al 2010); comparative data is more limited. Many reports have described comorbidity in various clinical settings; while these studies are typically smaller and their generalizability may be questioned, they provide a useful sense of comorbidity rates in treatment-seeking populations.

Recommendations for future research

Most epidemiological studies report comorbidity in terms of co-existing categorical diagnoses. Since symptomatology that does not reach the threshold for a categorical diagnosis can nevertheless be a source of morbidity, subsyndromal comorbid symptomatology can affect both suffering and prognosis. Large studies using dimensional measures of both OCD and non-OCD symptomatology, rather than DSM/ICD diagnostic entities, would be valuable and might give a different sense of the patterns of comorbidity in the population. Large epidemiological studies in different countries and subpopulations are needed to better establish commonalities and differences in the patterns of comorbidity across geography and cultures.

Anxiety disorders. The ability to understand and explain that:

1. All anxiety disorders are more prevalent among patients with OCD than in the general population, with lifetime comorbidity of over 75%

Evidence: OR: Calamari et al 2012 • CSS: Ruscio et al 2010

2. Generalized anxiety disorder (GAD) is more common in the relatives of cases than in controls’ relatives

Evidence: CSS: Nestadt et al 2001

Obsessive-compulsive related disorders. The ability to understand and explain that:

1. Lifetime prevalence of BDD in patients with OCD is 12–16%, much higher than the population prevalence

Evidence: CSS: Simeon et al 1995, Bienvenu et al 2000, 2012

2. Insight is more impaired for BDD than OCD in patients with both disorders

Evidence: OR: Eisen et al 2004 • CC: Phillips et al 2012

3. BDD is significantly more common in the first-degree relatives of OCD cases compared to first-degree relatives of controls

Evidence: CSS: Bienvenu et al 2000

4. Lifetime prevalence of trichotillomania in patients with OCD ranges from 4% to 11%

Evidence: CSS: Bienvenu et al 2000, Lovato et al 2012

5. Skin-picking (excoriation) disorder is very common in patients with OCD, with lifetime prevalence rates as high as 31%

Evidence: CSS: Bienvenu et al 2000, Lovato et al 2012

6. Lifetime prevalence of trichotillomania in patients with OCD ranges from 4% to 11%

Evidence: CSS: Bienvenu et al 2000, Lovato et al 2012

7. Hoarding symptoms (but not necessarily hoarding disorder) occur in up to 30% of OCD cases and are associated with earlier age of onset of OCD, more severe OCD symptoms, and poorer response to standard OCD treatments

Evidence: MA: Bloch et al 2014 • OR: Frost et al 1996 • CCS: Samuels et al 2002

Tic disorders. The ability to understand and explain that:

1. Tic disorders are common in patients with OCD   Evidence: CSS: Rosario-Campos et al 2005, Bienvenu et al 2012

2. There is a higher incidence of tic disorders in relatives of OCD probands than in the general population   Evidence: CSS: Grados et al 2001, Browne et al 2015

3. Specific OCD symptoms, namely symmetry, ordering and arranging, and counting, are most frequently, though not exclusively, encountered in patients with tic disorders   Evidence: CS: De Vries et al 2016

4. Patients with comorbid tic disorders have an earlier age-at-onset of OCD, and a male predominance   Evidence: CSS: Rosario-Campos et al 2005

Mood disorders. The ability to understand and explain that:

1. >60% of OCD patients will have a depressive disorder episode during their course of illness   Evidence: SR: Amerio et al 2015 • CS: Nestadt et al 2001 • CSS: Pinto et al 2006, Ruscio et al 2010, Quarantini et al 2010

2. Mood disorders most frequently result in an exacerbation of OCD symptoms (although there may also be a symptom reduction or no effect on OCD symptoms)   Evidence: SR: Amerio et al 2016

3. Patients often report that the mood disorder is secondary to their OCD symptoms. However, this can be difficult to confirm and may not always be accurate. Treatment should ideally target both disorders (e.g. SSRI antidepressants; CBT with behavioral activation). Treatment targeting depression specifically may be of benefit in some cases if depression is severe or impedes the progress of OCD treatment, but it does not take the place of appropriate evidence-based treatment for OCD   Evidence: EO

4. Circadian dysregulation, which often accompanies depression, may contribute to OCD   Evidence: SR: Paterson et al 2013 • CSS: Coles and Stewart 2019

5. The ability to understand and explain that bipolar mood disorders can co-occur with OCD and can greatly complicate treatment.   Evidence: CSS: Dell’Osso et al 2020

Post-traumatic stress disorder (PTSD). The ability to understand and explain:

1. Symptom overlap between OCD, PTSD, and depression   Evidence: CSS: Huppert et al 2005

2. That treatment-resistant OCD patients report a high rate of prior trauma   Evidence: OR: Adams et al 2018 • CSS: Gershuny et al 2008

Autism spectrum disorder. The ability to understand and explain that:

1. Traits of autism spectrum disorder (ASD) are often seen in outpatients with OCD, and that a comorbid diagnosis of ASD may be appropriate in up to a quarter of individuals with OCD   Evidence: OR: Accordino et al 2017 • CSS: Bejerot et al. (2001), Micali et al. (2004), Wikramanayake et al (2017)

2. The ability to understand and explain that OCD patients with high levels of ASD symptoms typically have reduced cognitive flexibility, reduced insight, and impaired social skills and social communication, relative to those with OCD without ASD   Evidence: OR: Accordino et al 2017 • CSS: Wikramanayake et al 2017

Schizophrenia and psychosis. The ability to:

1. Understand and explain that there is a high frequency of obsessions and compulsions (although not necessarily comorbid OCD) in patients diagnosed with schizophrenia (~15%); the relationship between OCD and schizophrenia is not well understood   Evidence: CSS: Porto et al 1997 • OR: Bottas et al 2005, Schirmbeck et al 2015, Poyurovsky et al 2017

2. Distinguish thought insertion from an obsession   Evidence: OR: Poyurovsky et al 2004 • EO: Oulis et al 2013

3. Understand and explain that most second-generation neuroleptic medications, particularly clozapine, can induce obsessions and compulsions, particularly when used without an SRI   Evidence: OR: Poyurovsky and Koran 2005, Lykouras et al 2003 • CR: Ghaemi et al 1995

Substance use disorders. The ability to understand and explain that:

1. The rate of comorbidity of substance use disorders (SUDs) varies, with some (such as nicotine use disorder) being lower in individuals with OCD than in many other psychiatric disorders   Evidence: CSS: Ruscio et al 2010, Abramovitch et al 2015, Toftdahl et al 2016

2. Substance use in individuals with OCD may emerge as a result of “self-medication” by patients   Evidence: CSS: Mancebo et al 2009

Suicidality.

1. The ability to understand and explain that OCD patients are at 2–5 fold greater risk of attempting suicide and a 10-fold elevated risk of dying by suicide than the general population. The presence of comorbid major depressive disorder, severity of comorbid depression or anxiety, and presence of comorbid substance use disorder increases suicide risk   Evidence: MA: Angelakis et al 2015 • SR: Albert et al 2019 • CSS: Fernandez de la Cruz et al 2016

Personality Disorders. The ability to understand and explain:

1. That personality disorders are common in patients with OCD   Evidence: OR: Starcevic and Brakoulias 2017

2. That OCPD, schizotypal and borderline personality disorder are the most clinically relevant, with OCPD the most frequent   Evidence: OR: Starcevic and Brakoulias 2017, Wheaton and Pinto 2017

3. That the presence of two or more personality disorders carries a poorer prognosis for OCD outcome   Evidence: SR: Thiel et al. (2013)

4. That schizotypal personality disorder is associated with poorer response to treatment but is not particularly common in OCD   Evidence: SR: Thiel et al 2013 • OR: Starcevic and Brakoulias 2017

5. That OCPD may improve as a result of treatment of OCD with medications or CBT   Evidence: OR: Starcevic and Brakoulias 2017 •TM: Gabbard 2005

6. Strategies to increase treatment effectiveness when working with OCD patients with comorbid personality disorders   Evidence: OR: Starcevic and Brakoulias 2017 • TM: Gabbard 2005

7. The ability to understand and explain effective treatments for OCPD and borderline personality disorder, and describe how to integrate them into a comprehensive treatment plan in individuals with OCD   Evidence: MA: Cristea et al 2017 • OR: Diedrich and Voderholzer 2015

3.b. Pathogenesis and pathophysiology

The causes and pathophysiological mechanisms underlying OCD are incompletely understood and are more the province of the researcher than of the clinical practitioner. Nevertheless, patients often have questions about causes and mechanisms, and a particular conception of disease (neurobiological, psychological, or otherwise) can enhance, or problematize, treatment engagement. The specialist clinician should therefore have a working knowledge of current about pathogenesis and pathophysiology; these topics are therefore treated briefly here.

3.b.i. Genetics

Overview.

The pathophysiology of obsessive-compulsive disorder is not well understood; in this, it resembles other complex neuropsychiatric conditions. Both early heritability studies and more recent genetic investigations suggest that genetics contributes 40–50% of the risk of OCD; the genetic contribution appears to be higher in childhood-onset than in adult-onset disease. Although the application of modern genetic techniques has produced new insights into the overall genetic architecture of the condition, individual genetic risk factors have yet to be clearly identified. Genetic testing is of no benefit in establishing diagnosis and of limited benefit in planning treatment. Psychiatric genetics in general is advancing rapidly; it is to be hoped that our understanding of the genetic underpinnings of OCD risk will grow substantially in the coming years.

Definitions of key terms

Heritability –

the extent to which a disease, trait, or feature is determined by genetic factors, as opposed to developmental, environmental, or random ones. Heritability can be determined from careful analysis of inheritance patterns even when the specific genes that contribute to a disease or trait are unknown.

Mendelian trait or disorder –

A genetically determined trait or disorder that is controlled by variation in a single gene, the genetic transmission of which is thus relatively simple to explain.

Polygenic –

determined by the interactive effects of multiple genes. Polygenic traits or diseases (like OCD) are typically more difficult to analyze genetically than simpler monogenic disease, like Huntington’s disease or sickle cell anemia.

Candidate gene studies –

Genetic studies examining one or a few genes, based on an a priori pathophysiological hypothesis. Individual candidate gene studies have been positive, but across all complex disorders these studies have a poor record of reproducibility and have been largely supplanted in recent years by hypothesis-free genome-wide approaches.

Genome-wide association studies (GWAS) –

Genetic studies that examine the contribution of common mutations across the genome to a disease or trait. Because they query the whole genome, rather than just one or a few individual genes, such studies are not constrained by specific a priori pathophysiological hypotheses; however, since they are probing hundreds of thousands of genetic loci in parallel, they require many thousands of individuals to achieve the required statistical power.

Linkage studies –

Genetic studies that examine the transmission of a disease or trait within individual family pedigrees.

Whole exome sequencing –

Examination of the genetic sequence of all genes in an individual with a disease or trait of interest. This has been made possible by the plunging cost of DNA sequencing and allows individual mutations to be probed and novel mutations to be identified; recently, this approach has been enormously fruitful in identifying genes contributing to autism spectrum disorder.

Whole genome sequencing (WGS) –

Examination of the genetic sequence of the entire genome, not just the expressed genes, in an individual with a disease or trait of interest. This allows examination of genetic variation that is not transcribed but may be relevant to gene expression.

Copy number variation –

Genetic variants in which an individual has more or fewer than the usual two copies of a specific sequence of DNA.

Polygenic risk score (PRS) –

An estimate of an individual’s genetic risk for a disease, calculated by considering what alleles they carry at all loci that have been implicated in the disease. Importantly, a PRS can be computed from a group of genetic variants that might be associated with disease, even if the associations are not yet definitive (as in the case of OCD).

Strength of evidence

The approximate heritability of OCD is well established. Specific genetic contributors, on the other hand, have not been definitively identified. Early studies focused on specific pathophysiological hypotheses, such as that OCD arises from dysregulation of serotonin (suggested by the efficacy of the SSRIs) or of dopamine (because of the role of neuroleptics in treating OCD, and their ability to induce OCD symptoms in some cases, especially when used without an SRI). However, such candidate gene studies have a poor record of reproducibility, across all genetically complex disorders, and they have shed little light on the pathophysiology of OCD. More modern methods (e.g. GWAS) probe the entire genome in parallel and thus do not depend on a specific a priori hypothesis; initial GWAS studies in OCD have produced some tentative initial clues as to its genetic causes, but many more subjects (up to tens of thousands) will need to be included before definitive conclusions can be reached, and still more work characterizing the functional consequences of OCD-associated genetic variants will be needed before implications for diagnosis and treatment can be drawn. New techniques such as CNV analysis, whole exome sequencing, and whole genome sequencing are only now being applied to OCD and related disorders.

Recommendations for future research.

GWAS studies described to date include approximately 4,000 individuals with OCD. In other disorders, such as schizophrenia, definitive results began to emerge as studies included 50,000 individuals or more; therefore, progress in OCD is likely to require the inclusion of many more patients in such studies. New techniques such as whole exome sequencing to identify de novo mutations that may contribute to OCD in individual patients are only now being applied in OCD and may produce new insights. The genetic relationship between OCD and other neuropsychiatric disorders is unclear and requires further research; recent results suggest that many genetic risk factors may be shared across psychiatric diagnoses. Finally, genomic information is currently of little use in treatment planning; pharmacogenomic studies to identify genetic predictors of treatment outcome, which are not currently available, are technically challenging, but they will be essential if genetics is to become clinically useful in planning treatment for individuals with OCD.

The ability to:

1. Understand and explain that OCD is moderately heritable (40–50%), and a diagnosis in a first-degree relative is an important risk factor for disease. Early onset and tic-related OCD may have a higher heritability. This heritability influences morbid risk in family members of diagnosed patients.   Evidence: OR: Van Grootheest et al 2005, Browne et al 2014 • CSS: Davis et al 2013

2. Understand and explain that OCD does not fit a Mendelian model of inheritance and is polygenic   Evidence: OR: Stewart et al 2010, Browne et al 2014

3. Understand and explain that genetic studies of OCD are advancing but to date have not definitively identified mutations that substantially contribute to disease; genetic testing is not yet informative for clinical diagnosis or for prediction of treatment outcome   Evidence: MA: Taylor 2013, IOCDF-GC and OCGAS 2018 • OR: Stewart et al 2010 • CSS: Noh et al 2017, Mattheisen et al 2015

4. Understand and explain that genetic testing, especially of liver P450 enzyme variants, may be useful to predict medication metabolism rates and side effects, although clear algorithms have not yet been developed   Evidence: CSS: Qin et al 2016

5. Discuss data implicating variants in serotonin-associated genes in OCD; interest in these genes stems largely from the efficacy of SSRIs in treating OCD, but any contribution to disease risk appears to be very small   Evidence: MA: Bloch et al 2008, Taylor 2013

6. Discuss data implicating variants in dopamine-associated genes in OCD; interest in these genes stems largely from the utility of dopamine antagonists in the treatment of tics and OCD and from the ability of antipsychotics to trigger OCD symptoms, especially when used without an SRI, but any contribution of these alleles to disease risk appears to be very small   Evidence: MA: Bloch et al. 2008, Taylor 2013

7. Discuss genetic evidence implicating variants affecting glutamate homeostasis in OCD risk. Multiple studies have implicated the glutamate transporter Slc1a1 in OCD; animal studies have implicated glutamate synaptic proteins, such as SAPAP3   Evidence: OR: Ting and Feng 2008, Rotge et al 2010, Pittenger et al 2011, Pauls et al 2014 • CSS: Kwon et al 2009, Mattheisen et al 2015

2.b.ii. Neurobiology: The corticostriatal model

Overview.

Abnormalities in affect, cognition, or behavior – that is, psychiatric symptomatology – necessarily correlate with deviations from normal brain function, although these abnormalities may be subtle and difficult to characterize using available technology and conceptual frameworks. OCD was one of the first neuropsychiatric disorders in which regional abnormalities in brain activity were associated with diagnosis and symptomatology. Early positron emission tomography (PET) studies from Baxter and colleagues in the 1980s identified regional hyperactivity in the caudate, thalamus, anterior cingulate cortex (ACC), and orbitofrontal cortex (OFC) in patients compared to controls, and showed that this hyperactivity was mitigated by both pharmacotherapy and cognitive behavioral therapy. PET studies in the 1990s by Rauch and colleagues, and others, identified hyperactivity of a similar circuitry when symptoms were induced in individual patients. These studies led to the development of the cortico-striatal model of OCD, which proposes that obsessive thoughts and compulsive behaviors derive from dysregulated activity in a circuit encompassing the frontal cortex (especially ACC and OFC), the striatum (which consists of the caudate, putamen, and nucleus accumbens), the globus pallidus, and the thalamus.

Subsequent studies using functional magnetic resonance imaging (fMRI) and other techniques have further refined this general model. Functional connectivity studies, which measure the stochastic relationships between activity in different brain regions and thereby infer the degree to which they are engaged in a functional circuit, have emphasized that the OFC-nucleus accumbens and ACC-caudate subcircuits may be differently dysregulated, and that enhanced connectivity between the nucleus accumbens and OFC/medial prefrontal cortex is particularly correlated with symptomatology. Certain domains of symptomatology may be associated with functional abnormalities outside this circuitry; for example, dysregulated anxiety is associated with amygdala abnormalities, and abnormal sensory sensitivity or susceptibility to feelings of disgust may be associated with functional abnormalities in the insular cortex. Therapeutic studies in which this circuitry is directly targeted, as with deep brain stimulation, provide further evidence that functional abnormalities in the corticostriatal circuitry are causally associated with OCD symptomatology.

Definition of key terms

Positron emission tomography (PET) –

a neuroimaging methodology in which low concentrations of radioactive tracer molecules are injected into the bloodstream and their distribution in the brain is measured in three dimensions, permitting quantification of blood flow, metabolism, or the density of specific receptor molecules (depending on the nature of the radiotracer injected). Early PET studies provided the first evidence for the corticostriatal model of OCD pathophysiology.

Magnetic resonance imaging (MRI) –

a neuroimaging methodology wherein strong magnetic fields and radio waves are used to produce a three-dimensional picture of the brain, or other structures within the body.

Functional MRI (fMRI) –

a variant on MRI imaging wherein relative blood flow to different parts of the brain is visualized; typically blood flow is measured in the same subjects under different conditions, and the difference between different conditions is interpreted as a measure of differential brain activation between them.

Functional connectivity –

an analysis of fMRI data wherein the statistical correlations in activity between different brain regions are computed, either at rest or during performance of a task. The strength of correlation is considered to be a measure of the degree to which the two brain regions or structures are engaged in a functional circuit.

Corticostriatal circuitry –

also called the cortico-basal ganglia circuitry or the cortico-striato-thalamo-cortical, or CSTC, circuitry – a circuitry in the brain, connecting the frontal cortex with a set of evolutionarily conserved subcortical structures, including the striatum (caudate, putamen, and globus pallidus), the globus pallidus, and the thalamus. The corticostriatal circuitry is conceptualized, at least to a first approximation, as consisting of parallel feedforward loops from the cortex to the striatum, through the globus pallidus, to the thalamus, and back to the cortex.

Corticostriatal model of OCD –

a qualitative model of OCD pathophysiology wherein obsessive thought patterns and compulsive behaviors are conceptualized as correlating with, and perhaps deriving from, hyperactivity in the corticostriatal circuitry.

Prefrontal cortex (PFC) –

the region of the cerebral cortex lying anterior to the motor cortex in the front/anterior portion of the brain. The PFC is very complex and is enormously expanded in humans relative to other primates. It is thought to be associated with attention, concentration, working memory, abstract thought, and inhibition of automatic, emotional, and habitual behaviors.

Orbitofrontal cortex (OFC) –

a subregion of the ventral PFC that lies above the eyes. Hyperactivity in the OFC is the most robustly replicated finding in neuroimaging studies of OCD. OFC activity has been implicated in a wide range of functions, including value-based decision-making and behavioral and cognitive flexibility.

Anterior cingulate cortex (ACC) –

a subregion of the PFC lying above the corpus callosum on the midline of the brain. ACC hyperactivity has been described in OCD, and therapeutic lesions of the ACC are sometimes used in profoundly refractory disease. Various subregions of the ACC are implicated in myriad cognitive and affective functions, including error detection, pain modulation, and the interface between cognitive control and affective responses.

Basal ganglia –

a network of evolutionarily conserved subcortical structures; connections from the cortex through the basal ganglia to the thalamus and then back to the cortex constitute the corticostriatal or CSTC circuitry.

Striatum –

the largest nucleus of the basal ganglia and their primary input nucleus; in humans and other primates, the striatum consists of the caudate and putamen nuclei (collectively the dorsal striatum) and the nucleus accumbens (the ventral striatum).

Globus pallidus –

a deep nucleus of the basal ganglia, which mediates between the striatum and thalamus.

Subthalamic nucleus (STN) –

a small nucleus of the basal ganglia; therapeutic stimulation of the STN has been used in both refractory OCD and advanced Parkinson’s disease.

Thalamus –

a large and complex subcortical nucleus that serves as a major relay station whereby information is transferred to the cortex, including from the basal ganglia as part of the CSTC circuitry.

Amygdala –

a subcortical structure within the brain’s anterior temporal lobe; dysregulation of the amygdala has been associated with anxious symptomatology.

Insula –

a region of the PFC that is folded underneath other cortical structures and thus not visible from the surface. The insula has been implicated in self-perception, pain, disgust, and taste; abnormalities in insula function have been associated with disgust sensitivity in OCD.

Strength of evidence

Substantial convergent evidence from multiple studies and neuroimaging modalities links OCD with functional abnormalities in the cortico-basal ganglia circuitry. The causal relevance of these abnormalities is supported by the fact that targeted modulation of this circuitry using deep brain stimulation and other modalities appears to have therapeutic benefit in refractory disease. The nature of these abnormalities and their precise functional relationship to OCD symptomatology, on the other hand, remain unclear. Neural abnormalities that are seen with some consistency at the group level do not have enough discriminant power to be of any diagnostic or clinical utility. Different patterns of neural abnormality may be associated with different symptomatic presentations or subtypes of disease, but these more precise relationships between brain and disease symptomatology remain to be clearly elucidated. Finally, why these brain abnormalities emerge, and why they are associated with particular constellations of symptomatology, is largely unknown.

Suggestions for future research

The cortico-striatal model of OCD is widely accepted at a very general level, but it is not well specified. Why corticostriatal functional abnormalities are associated with OCD phenomenology, where they come from, how they differ among individuals with OCD and related disorders, and how they may inform diagnosis and treatment are important unanswered questions. The CSTC model has dominated thinking about OCD until recently; dysregulation in other structures, such as the amygdala and insula, parietal cortex, and cerebellum has been appreciated more recently and merits further study. Future studies should study large numbers of individuals with OCD to better define subtleties and inter-individual variation in CSTC dysregulation and how these may map onto different domains of OCD phenomenology. Such understanding may lead to further advances in personalized treatment, especially using targeted neuromodulatory approaches.

The ability to understand and describe:

1. The basic components and interconnection of the cortico-striato-thalamo-cortical circuitry   Evidence: OR: Haber (2017)

2. Evidence that CSTC hyperactivity is seen in OCD and is associated with symptomatology and treatment   Evidence: MA: van der Straten et al 2017 • OR: Brennan and Rauch 2017, Stein et al 2019, van den Heuvel et al 2015 • CSS: Baxter et al 1987, Rauch et al 1994, Saxena et al 2002

3. Evidence that CSTC hyperactivity is causally associated with OCD and may represent a therapeutic target   Evidence: OR: Maia et al 2008, Greenberg et al 2010

4. Amygdala dysregulation and its association with anxiety in OCD   Evidence: OR: Milad and Rauch 2012

5. Insula dysregulation in OCD and its association with disgust symptoms   Evidence: CSS: Shapira et al 2003

3.b.iii. Neurochemistry

Overview

SSRIs are the first-line pharmacotherapy for OCD. This has led to the idea that serotonin imbalance may be central to pathophysiology. However, evidence in support of this hypothesis from other sources, such as genetics and brain imaging, remains equivocal. More recent work has focused on other neurotransmitters, including dopamine, histamine, GABA, and glutamate. Recent genetic and pharmacological studies have particularly highlighted glutamate dysregulation as a possible contributor to OCD, although this cannot be considered proven. A focus on neurochemistry elides the complexity of the brain at a network and circuit level; serotonin and glutamate, and all other transmitters, play multitudinous roles in the nervous system, and neither OCD nor any other complex neuropsychiatric disorder is likely to be explicable in terms of simple neurochemical imbalance. Nevertheless, as most pharmacological treatments focus on specific neurochemicals and their receptors, understanding how their dysregulation may contribute to disease remains an important area of study, and advancing knowledge may lead to new treatments.

Definition of key terms.

Neurotransmitter.

A chemical that communicates between neurons where they contact one another, leading to electrical activation or other modulation of their activity. Neurotransmitters act by binding to receptors on the surface of neurons. Excitatory and inhibitory neurotransmitters directly electrically excite or inhibit the excitation of neurons. Modulatory neurotransmitters alter the behavior of neurons in more complex and often long-lasting ways. The same transmitter can function as both excitatory/inhibitory and modulatory, depending on which receptors it binds to.

Serotonin –

also 5-hyrdoxytryptamine, or 5-HT. An aminergic modulatory neurotransmitter that plays multiple modulatory roles in the brain and is implicated in the regulation of mood, anxiety, impulsivity, and aggression. Most serotonin in the brain comes from the raphe nuclei in the midbrain; these neurons project widely, innervating nearly every structure, including the cortico-striato-thalamo-cortical circuitry that is frequently implicated in OCD. The serotonin reuptake transporter is the primary molecular target of the SSRIs and clomipramine.

Dopamine.

An aminergic neurotransmitter with broad modulatory effects throughout the brain, implicated in reward and reinforcement, arousal, attention, and addiction. Most dopamine in the brain comes from the substantia nigra and the adjacent ventral tegmental area, in the ventral midbrain. Dopaminergic innervation of the striatum is particularly strong, and its dysregulation there has been implicated in Parkinson’s disease and other movement disorders, drug addiction, and Tourette syndrome. The association with OCD is less clear.

Histamine.

A diamine signaling molecule that functions as a modulatory neurotransmitter, in addition to having a prominent role in the regulation of inflammation and allergy. Neurotransmitter histamine derives primarily from the posterior hypothalamus; these neurons project throughout the brain. Interest in histamine dysregulation in OCD derives from the recent finding in a single family that a rare mutation associated with reduced histamine production can cause Tourette syndrome, with comorbid OCD in some cases.

Acetylcholine.

A neurotransmitter used in both the peripheral and central nervous systems. There are multiple sources of acetylcholine in the brain. Interest in cholinergic dysregulation in OCD derives from the observation that cholinergic interneurons in the basal ganglia are reduced in number in Tourette syndrome; the relationship with OCD is less clear. Many commonly used medications, including clomipramine and paroxetine, have prominent anticholinergic effects, which can contribute to side effects.

GABA – also gamma-aminobutyric acid.

The primary inhibitory neurotransmitter in the adult brain; GABA is the primary neurotransmitter in the basal ganglia and is used by interneurons throughout the cortex and elsewhere.

Glutamate.

The primary excitatory neurotransmitter in the brain; glutamate is used by a majority of neurons in the central nervous system. This is in dramatic contrast to the modulatory neurotransmitters (e.g. serotonin, dopamine), which are produced by only small numbers of neurons in discrete places. Genetic, neuroimaging, and neurochemical data suggest that glutamate dysregulation may contribute to OCD in some cases, and several glutamate modulators have shown promise in early studies.

Strength of evidence

The evidence for dysregulation of specific neurotransmitters in OCD is not strong. There have been positive reports of abnormalities in different neurotransmitters, but non-replication across studies is common. The efficacy of SRIs in OCD has been taken to support the idea that serotonin dysregulation is important in OCD pathophysiology, but this need not be the case: modulating serotonin might mitigate symptoms even if serotonin imbalance has nothing to do with the underlying pathophysiology. There has been substantial recent interest in glutamate dysregulation in OCD, supported by preliminary evidence from genetic studies, by some neuroimaging and neurochemical studies, by work in animal models, and by some preliminary treatment studies (further discussed below). However, while this convergent evidence is intriguing, it cannot be considered conclusive.

Suggestions for future research.

Our tools for measuring neurotransmitter dysregulation in vivo in humans remain limited. More research is needed on multiple fronts to better characterize how and to what extent neurotransmitter dysregulation may contribute to OCD. It is important to distinguish between neurotransmitter dysregulation as a cause, as opposed to a correlate or consequence of pathophysiology. Since neurotransmitters are involved in everything that the brain does, any dysregulation of brain function may produce a secondary abnormality in one or another neurotransmitter.

The ability to understand and describe:

1. The limited evidence for mutations in serotonin-related genes in OCD   Evidence: OR: Graat et al 2017, Bandelow et al 2017, Pittenger, 2021

Mutations in the serotonin transporter (SERT) and the 5HT2a receptor have been reported in OCD. However, these findings derive from candidate gene studies, which are no longer considered reliable, and have not been replicated in modern genome-wide studies; most OCD is probably not caused by genetic abnormalities in serotonin-related genes   Evidence: MA: Taylor 2013, 2016 • OR: Graat et al 2017

Serotonin depletion can worsen symptoms of depression and PTSD but not OCD   Evidence: OR: Bandelow et al 2017

2. The limited evidence for primary dysregulation of the dopaminergic system in OCD   Evidence: OR: Graat et al 2017, Bandelow et al 2017, Pittenger, 2021

Mutations in the catechol O-methyltransferase (COMT) gene may contribute to OCD in males, although the effect size is small and again derives from candidate gene association studies, which are considered unreliable   Evidence: MA: Taylor 2013, 2016

Neuroimaging studies have not consistently shown abnormalities in dopamine or its receptors in OCD   Evidence: MA: Graat et al 2017

Dopamine excess is more strongly implicated in tic disorders, which can be comorbid with   OCD Evidence: CCS: Denys et al 2013

3. The limited evidence associating histamine dysregulation with tic disorders, and perhaps OCD. Genetic evidence and studies in mouse models suggest that impaired production of brain histamine may contribute to tic disorders in rare cases. The evidence for a similar effect in OCD without tics is less clear   Evidence: OR: Pittenger 2017a

4. The limited evidence for dysregulation of acetylcholine in OCD

Disruption of cholinergic interneurons has been implicated in Tourette syndrome   Evidence: CSS: Kataoka et al 2010

Evidence for a role for acetylcholine dysregulation in OCD is less clear   Evidence: ThP: Carlsson 2001

Patients with OCD are less likely to smoke than the general population   Evidence: CSS: Abramovitch et al 2015

5. The evidence for glutamate dysregulation (possibly excess) in OCD   Evidence: OR: Pittenger et al 2011, Graat et al 2017, Bandelow et al 2017, Pittenger, 2021 • ThP: Carlsson 2001

Genetic evidence remains inconclusive, but some studies suggest that abnormalities in glutamate-related genes may contribute to OCD   Evidence: MA: International OCD Foundation Genetics Collaborative et al 2018 • OR: Rajendram et al 2017

Some magnetic resonance imaging (MRS) studies have found glutamate dysregulation in OCD, although others have not   Evidence: OR: Brennan et al 2013

Some mouse models with abnormalities in glutamatergic neurotransmission exhibit behaviors that have been interpreted as recapitulating aspects of OCD   Evidence: OR: Ting and Feng 2008, Pittenger et al 2011

6. The ability to understand and explain the limited evidence for dysregulation of GABA in OCD – A single MRS imaging study has suggested elevated GABA in the ACC in OCD   Evidence: OR: Graat et al 2017 • CCS: Simpson et al 2012

3.b.iv. Inflammation/neuroimmune abnormalities

Overview.

A substantial literature, albeit a heterogeneous one, suggests that individuals with OCD may, as a group, have immune abnormalities, as compared to the general population. This may be truer in children than in adults, and more studies have addressed the potential role of immune modulators in treatment of the pediatric population. In particular, the syndrome Pediatric Acute Onset Neuropsychiatric Syndrome (PANS) describes children with the sudden, even overnight, onset of symptoms, and the older entity Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcus (PANDAS) captures the hypothesis that at least some of these cases are attributable to the development of self-reactive antibodies following an infection. These concepts and their therapeutic implications, which have been the focus of some controversy, are discussed in detail in the accompanying paper in this series on pediatric OCD [Piacentini, et al.,);

they are presented briefly here. While treatment algorithms have been developed for pediatric disease in which an autoimmune or neuroinflammatory etiology is suspected, they are not universally endorsed. The clinical implications of neuroinflammatory processes in adults are not established.

Definition of key terms.

Autoimmunity –

The development of antibodies against self-antigens, leading to inappropriate immune activation and damage to cells and tissues. Autoimmunity may arise in a susceptible host idiopathically, in response to an infection, or in response to other pathological processes such as cancer.

PANS, or Pediatric Acute-Onset Neuropsychiatric Syndrome –

A pediatric syndrome in which OCD symptoms and/or extreme food avoidance, together with a constellation of other neuropsychiatric symptoms, develop extremely suddenly, even overnight, in a child. This syndromic diagnosis is etiologically agnostic and does not require a temporal association with infection or evidence of neuroinflammatory dysregulation; but it remains an unanswered empirical question whether it reflects a meaningful diagnostic entity.

PANDAS, or Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcus.

A pediatric syndrome in which OCD symptoms and/or tics and other motor abnormalities develop extremely suddenly, even overnight, in temporal association with an infection with Group A Streptococcus. An autoimmune etiology and a causal role for streptococcal infection is explicitly implied by this diagnosis; however, this has been the cause of some controversy, as the details of this hypothesized pathophysiology remain to be established with certainty.

Strength of evidence.

Enough studies have reported abnormalities in measures of immune function in patients with OCD that it seems likely that there is some immune dysregulation in this population, or at least a subset of it, although the clinical relevance of this dysfunction is not clear. There has been more work in this area in the pediatric population. In adults, a recent meta-analysis showed a higher incidence of autoantibodies in patients, and a recent PET study documented microglial activation in the basal ganglia in patients with OCD. The nature of this immune dysregulation, whether it exists in all patients or in a subset, whether it is causally associated with disease onset or progression, and whether it is a specific cause of disease, a nonspecific exacerbating factor, or an epiphenomenal consequence of other pathophysiological events remain unclear.

Suggestions for future research.

The potential contribution of neuroinflammatory processes to OCD is important to clarify as it may influence treatment in certain cases. This area has been a focus of some controversy, especially in the pediatric literature, with some authors skeptical of the existence of syndromes such as PANDAS and others suggesting that many cases of OCD can be traced to infectious or neuroinflammatory processes. Establishing the extent to which neuroinflammatory processes contribute to OCD, and in which cases (if any), is a critical area for future research. Neuroinflammatory processes could be a primary cause of OCD symptoms in some cases; alternatively, they could represent a less specific triggering or exacerbating event. This needs to be elucidated. Finally, while guidelines have been developed for the treatment of pediatric cases in which a neuroinflammatory contribution is suspected, they remain controversial. The role of immune-modulating therapies in adults, if any, remains to be clarified.

The ability to understand and describe:

1. The diagnosis of Pediatric Acute-Onset Neuropsychiatric Syndrome (PANS)   Evidence: EO: Swedo et al 2012, Chang et al 2015

2. The diagnosis of Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcus, and the hypothesized role of autoimmunity in its etiopathogenesis   Evidence: OR: Swedo & Williams 2016

3. The role of immune-modulating therapies in the treatment of PANDAS and PANS, and the potential drawbacks of such treatments   Evidence: EO: Chang et al 2015

4. The more limited evidence for immune dysregulation in adult OCD

PET imaging with a ligand that recognizes activated microglia shows elevated binding in adults with OCD   Evidence: OR: Attwells et al 2017

A mixed literature suggests that there may be elevated anti-basal ganglia antibodies in OCD   Evidence: MA: Pearlman et al 2014

3.c. Associated cognitive domains

Overview.

Obsessive-compulsive disorder is characterized by unrealistic or excessive preoccupations or fears can dominate conscious experience and irrational or excessive behaviors can occupy many hours of the day. These thoughts and behaviors persist despite the presence, in many cases, of excellent insight into their unrealistic character and harmful effects. This suggests that OCD may be associated with dysregulation of fundamental cognitive processes required to align internal experience and motivation with the facts of the outside world. These matters are highly relevant in the psychotherapy of OCD and are thus dealt with in more detail in the ATF papers in this series on specialized CBT treatment of adults and children (see Sookman et al., 2021b and Piacentini et al., under review); but a general appreciation for this aspect of symptomatology is also relevant for understanding phenomenology in the context of pharmacological treatment, and a brief overview is thus presented here.

Cognitive biases in five major domains have been described in patients with OCD: overimportance/overcontrol of thoughts; overestimation of threat; inflated sense of responsibility; intolerance of uncertainty; and perfectionism. Importantly, these cognitive biases are found in the general population and can also contribute to other psychiatric diagnoses. Classic neuropsychological tests paint a murkier picture; individuals with OCD exhibit small to medium-sized deficits in a number of neuropsychological domains (e.g. executive function, working memory), but no large or consistent deficits. One major recent research focus in the research literature has been on cognitive inflexibility and overreliance on automated patterns of behavior and thought (i.e. habits) in OCD, although how best to operationalize this concept remains to be clearly established.

Definition of key terms.

Overestimation of threat –

A cognitive bias seen in OCD (as well as other anxiety disorders and some other OCRDs, and to some extent in the general population) in which the potential threat associated with real or anticipated events or contexts, or the probability of the occurrence of a negative outcome, is exaggerated, and the corresponding salience and affective response to those events or contexts is thus enhanced.

Inflated responsibility –

A cognitive bias seen in OCD (as well as in the general population) in which an individual’s sense of responsibility for potentially harmful circumstances or outcomes is enhanced beyond what most would consider reasonable; this may lead to an exaggerated need to control those circumstances and to an exaggerated sense of guilt over bad outcomes.

Over-importance of thoughts –

A cognitive bias seen in OCD (as well as in the general population) in which the practical or moral importance of thoughts is exaggerated, as if having a thought of a negative outcome increased the odds of it occurring, or having a thought of an odious action was morally equivalent to performing the action (thought-action fusion). Overestimating the importance of thoughts may lead to efforts to inhibit or control them, which end up being counterproductive.

Intolerance of uncertainty –

A heightened sense of unease or discomfort with uncertainty of outcomes or with the need to make decisions without having all potentially relevant information in hand. This may contribute to indecisiveness.

Perfectionism –

A sense that everyday actions must be performed to an exacting or unrealistic degree of perfection to be acceptable.

Cognitive flexibility –

The ability to change beliefs or patterns of behavior in the face of new or changing information or environmental feedback. This has been operationalized in a variety of ways in experimental studies and is likely to represent a number of distinct interacting cognitive capacities.

Habit –

A learned behavior that becomes automatized and inflexible, which may lead it to be executed even when it is no longer contextually appropriate. An over-reliance on habit has been described in OCD.

Neurocognitive testing –

characterization of cognitive function across a range of domains, such as global intelligence (IQ), attention, working memory, processing speed, and executive function, usually using standardized test batteries. The relationship between cognitive function, as assessed by such testing, and the specific patterns of cognition described above is not straightforward.

Strength of evidence

The documentation of specific cognitive biases that can contribute to symptomatology has been of substantial value in informing strategies and techniques for psychotherapy. More recently there has been a focus on cognitive inflexibility and dysregulated habit learning. The specificity of these abnormalities and how they map onto underlying pathophysiology and treatment outcome is unclear. There is ample evidence for the presence of subtle neurocognitive abnormalities in OCD, but these are not large or specific; this is supported by recent meta-analyses of neuropsychological data. Cognitive abnormalities are likely to be heterogeneous across individuals with OCD; individual variation in cognitive abnormalities has not been well characterized.

Directions for future research

Substantial progress has been made in recent years in fleshing out the cognitive abnormalities that accompany OCD. Whether these abnormalities are uniform across all OCD patients or, as seems more likely, different patterns of cognitive difficulty and bias exist in different individuals with the diagnosis remains unclear. Some studies in this domain have used self-report measures while others have used behavioral tests; whether these two approaches tap into the same underlying constructs remains unclear and requires further study. The relationship between these various cognitive constructs and dimensions of clinical symptomatology remains unclear, as does their specificity to OCD (versus transdiagnostic nature). It is also unclear to what extent the constructs measured in these various studies map onto the organization of the underlying neurobiological and cognitive systems, as opposed to being imprecise probes that are influenced by multiple interacting underlying processes. Clearer insight is likely to derive from behavioral, clinical, or self-report measures that are isomorphic, to whatever extent possible, with underlying processes; identifying and characterizing such constructs and probes is an important goal of ongoing and future research and may lead to new strategies for individualized treatment.

The ability to understand and explain:

1. The domains of cognitive bias identified in OCD by the Obsessive-Compulsive Cognitions Working Group: • overestimation of threat • inflated responsibility • over-importance/overcontrol of thoughts • intolerance of uncertainty, perfectionism   Evidence: PS: Obsessive Compulsive Cognitions Working Group 2003, 2005

2. The evidence for cognitive inflexibility in OCD, and how it may relate to symptomatology   Evidence: OR: Gruner and Pittenger 2016

3. How over-reliance on habit may contribute to compulsions in OCD and related disorders   Evidence: OR: Gillan 2017

4. That individuals with OCD, as a group, have normal intelligence but small- to medium-sized deficits across all major domains. However, no clear or specific patterns of deficit have emerged.   Evidence: MA: Abramovitch et al 2013, Shin et al 2014, Snyder et al 2015

5. The ability to understand and explain evolutionary theories as to why cognitive biases that predispose to OCD have persisted, while recognizing that these theories remain speculative   Evidence: OR: van Schalkwyk and Leckman 2017

3.d. Principles undergirding pharmacological treatment

3.d.i. Choice of treatment modality and sequencing of treatment

Overview.

There are several evidence-supported first- and second-line treatments for OCD. These include pharmacotherapy with the SSRIs, the older agent clomipramine, and various pharmacological augmentation strategies, as well as specialized cognitive-behavioral therapy. These treatment options, and their evidence base, are discussed in more detail in the sections that follow; details on both the theoretical background and the practical execution of specialized CBT are given in the ATF papers in this series focusing on CBT for adult and pediatric OCD (see Sookman et al., 2021b and Piacentini, et al., under review). The availability of several different options for treatment raises the clinical challenge of how best to select an initial treatment, and how to sequence treatments when initial interventions produce an inadequate response. Head-to-head trials suggest that both pharmacotherapy and CBT (especially the strategy of exposure and response prevention, or ERP) are quite effective and that ERP delivered with moderate intensity under optimal conditions may be preferable, at least for patients without complicating comorbidity. ERP can be combined with cognitive therapy approaches and behavioral experiments (Sookman et al., 2021b, this series). CBT reduces relapse, which is common after the discontinuation of pharmacotherapy. Patient preference, the availability of skilled therapists, comorbidity, potential adverse effects, personal and family history, treatment history, illness severity, and other factors may influence the choice of initial treatment modality.

Definition of key terms.

SSRI (selective serotonin reuptake inhibitor) –

A class of antidepressant medications, also effective for OCD, BDD, anxiety, and many other disorders and symptoms. This class is typically defined to include fluoxetine (Prozac; Sarafem), fluvoxamine (Luvox; Faverin; Fevarin; Floxyfral; Dumyrox), sertraline (Zoloft; Lustral), paroxetine (Paxil; Pexeva; Brisdelle; Seroxat), citalopram (Celexa; Cipramil), and escitalopram (Lexapro; Cipralex), all of which are of proven efficacy in the pharmacotherapy of OCD.

SRI (serotonin reuptake inhibitor) –

A broader class of drugs that include both the SSRIs and the older tricyclic medication clomipramine (Anafranil).

Neuroleptic –

A class of drugs, also called antipsychotics, that antagonizes the dopamine D2 receptor and has been shown to be of benefit in augmentation after nonresponse to SSRIs in some patients; neuroleptic use is further discussed below.

CBT (cognitive-behavioral therapy) –

A group of psychotherapies that focus on the systematic examination and alteration of maladaptive patterns of behavior, thought, and emotion, and their meanings. There is less focus than in dynamic/exploratory psychotherapy on details of personal history that may have given rise to current symptomatology, although these may be addressed when they are functioning as a maintaining factor. More detail is provided in the accompanying ATF papers on adult and child CBT in this series.

ERP (exposure and response prevention or, sometimes, exposure and ritual prevention) –

a CBT strategy that focuses on triggering obsessive thoughts and the attendant distress, and then helping patients refrain from engaging in compulsions, avoidance, or other compensatory behaviors. The efficacy of ERP for OCD has been repeatedly proven. More detail is provided in the accompanying ATF papers on adult and child CBT in this series.

Cognitive therapy –

CBT that focuses on clarifying, examining, and modifying maladaptive patterns of thought and emotional responses in OCD, and the information processing biases that may reinforce them. More detail is provided in the accompanying ATF papers on adult and child CBT.

Treatment response –

In the OCD literature, treatment response is mostly often measured in terms of improvement in the Yale-Brown Obsessive Compulsive Scale (Y-BOCS; Goodman et al, 1986a). There has been some variation in how large an improvement constitutes ‘response’; the most widely recognized metrics are 35% improvement for ‘response’ and 25% for ‘partial response’ (Mataix-Cols et al, 2016). It is noteworthy that someone with moderate to severe symptoms can respond to treatment but still have significant symptoms.

Remission –

Extended, marked improvement of symptoms, such that they no longer significantly affect quality of life. In OCD, remission is often defined as reduction of symptoms to the point that the DSM diagnostic criteria are no longer met, and the Y-BOCS score is below 12. Remission of moderate to severe OCD symptoms is not common in response to pharmacotherapy and therefore is not often used as an endpoint in clinical studies.

Relapse –

Recrudescence of symptoms after a period of remission

Strength of evidence.

There is robust evidence, from multiple randomized controlled trials and meta-analyses thereof, for the efficacy of several first- and second-line treatments for OCD; these are treated in more detail in the sections that follow. Studies comparing the efficacy of different treatments are more limited and may not generalize to the complex world of clinical practice. How best to sequence treatments in individual patients has not been examined in well-powered studies; multi-step treatment algorithms are generally guided by expert opinion more than by definitive comparative tests of alternative sequencing strategies.

Suggestions for future research

As is clear in the discussion of strength of evidence above, there is great need for well-powered studies that compare different initial therapies and strategies for sequencing treatment. Comparative treatment studies in real-world settings are also needed, to see whether the randomized clinical trials performed under optimal conditions generalize to other contexts and more heterogeneous populations. Finally, we have limited evidence on the predictors of individual variability in the response to different treatments, and therefore treatment is guided in most cases by generic algorithms, patient and provider preference, and practical considerations, rather than by a well-grounded understanding of which patients are most likely to respond to which treatments. This is an important area for future work.

The ability to understand and explain:

1. Existing data regarding comparative efficacy of SRI vs. ERP vs. their combination for OCD in adults

SRI pharmacotherapy and CBT, including ERP, are both considered first-line treatments for OCD. No studies have examined their comparative effectiveness in a real-world setting in previously untreated patients.   Evidence: TG: NICE 2006, Koran et al 2007, Koran et al 2013, Skapinakis et al 2016a,b, Fineberg et al 2020

In a comparison of clomipramine with intensive ERP and their combination, in a research setting, ERP was found to have the largest effect size; combination treatment gave no additional benefit   Evidence: RCT: Foa et al 2005

In those with clinically significant symptoms despite an adequate SRI trial, addition of twice weekly ERP was found to provide greater benefit than stress management   Evidence: RCT: Simpson et al 2008

2. The principles and basics of practice of specialized CBT for OCD, including ERP and cognitive therapy   Evidence: OR: Foa et al 2012, Abramowitz 2017, Yule and Whittal 2017, Sookman et al. 2021a,b (this series)

3. Real-world limitations that may constrain optimal CBT   Evidence: TG: NICE 2006, Koran et al 2007, Koran et al 2013, Skapinakis et al 2016a,b

Illness severity   Evidence: CT: Mataix-Cols et al 2002 • OCT: Keijsers et al 1994, de Haan et al 1997

Depression and other comorbidities   Evidence: CR: Steketee et al 2001 • CT: Hohagen et al 1998

Individual willingness to engage in CBT, due to a biological conception of illness, a reluctance to engage in exposures to feared stimuli, or the (accurate) perception that it entails extended hard work; this can often be addressed with psychoeducation   Evidence: CS: Simpson et al. 2012 • CR: Reid et al. (2017)

Limited availability or accessibility of specialized CBT/ERP providers   Evidence: OR: Sookman et al 2021a (this series)

4. The importance of psychoeducation in treatment (e.g., review of obsessive thoughts, concept of cognitive distortions, family accommodation, rationale for ERP vs pharmacotherapy) Evidence: OR: Lebowitz et al 2016, Abramowitz 2017

5. Familiarity with key psychoeducation resources, such as patient-targeted books and the IOCDF and CIOCD websites   Web resources: www.iocdf.org; www.ciocd.ca

6. The principles of sequencing in the treatment of OCD.   Evidence: TG: Koran et al 2007 • OR: Simpson 2017

Monotherapy vs combination therapy as initial treatment   Evidence: RCT: Foa et al 2005

Augmentation with psychotherapy when SRI monotherapy is ineffective   Evidence: RCT: Simpson et al 2008

Augmentation with pharmacotherapy when an initial trial of specialized CBT is not optimally effective   Evidence: TG: Koran et al 2007

Augmentation with neuroleptic when SRI monotherapy is ineffective   Evidence: MA: Bloch et al 2006, Veale et al 2014 • RCT: Simpson et al 2013

Relapse risk following the discontinuation of pharmacotherapy; ERP may mitigate relapse risk   Evidence: TG: Koran et al 2007

3.d.ii. Selective serotonin reuptake inhibitors (SSRIs)

Overview.

The selective serotonin reuptake inhibitors (SSRIs) are the first-line pharmacological treatment for OCD, according to all treatment guidelines. Their efficacy has been robustly demonstrated by multiple randomized clinical trials, and meta-analyses thereof. The older agent clomipramine appears to be slightly more efficacious, as further discussed below (section 3.d.iii), but its more problematic side effect profile means that SSRIs are still the preferred first-line agent under most circumstances. All of the SSRIs appear to be of equivalent efficacy. Higher doses and longer durations of treatment are needed for optimal response, relative to treatment guidelines for these agents in major depressive disorder. While the SSRIs are generally well tolerated, potential side effects should not be minimized; some are class effects and typically generalize across agents, while others differ across SSRIs and may contribute to the choice of which agent to use in a particular case. The SSRIs citalopram carries a warning that it may contribute to arrhythmia, especially at high doses; some regulatory agencies apply a similar warning to escitalopram. While the appropriateness and utility of these warnings has been questioned by some authors, they must nevertheless be taken into account, as a practical matter, when prescribing.

Definitions of key terms.

Selective serotonin reuptake inhibitor (or SSRI).

Any of six medications used in the treatment of OCD, depression, anxiety, and other conditions that act primarily by inhibiting the reuptake of serotonin at synapses, thus acutely increasing serotonin’s synaptic/extracellular concentration and extending its half-life in the synapse. The six SSRIs in current use are:

• fluvoxamine (branded as Luvox, Faverin, Fevarin, Floxyfral, Dumyrox)

• fluoxetine (branded as Prozac, Sarafem)

• sertraline (branded as Zoloft, Lustral)

• paroxetine (branded as Paxil, Pexeva, Brisdelle, Seroxat)

• citalopram (branded as Celexa, Cipramil)

• escitalopram (the purified L-enantiomer of citalopram; branded as Lexapro, Cipralex) Four of these – fluvoxamine, fluoxetine, sertraline, and paroxetine – are approved by the US Food and Drug Administration for the treatment of adult OCD. Citalopram and escitalopram have not been approved by the FDA for this indication, but their efficacy has been demonstrated in multiple studies, and they are often used off-label to treat OCD.

Strength of available evidence.

The efficacy of SSRI monotherapy in the treatment of OCD has been robustly demonstrated in multiple high-quality randomized clinical trials, summarized and confirmed by rigorous meta-analyses. As no other class of medications has been shown to work equally well as monotherapy for OCD (with the exception of clomipramine), the SSRIs are considered by all established treatment guidelines and expert consensus to be the optimal first-line pharmacological agent in most cases of OCD. Strong evidence, supported by meta-analysis, also indicates that higher doses and longer duration of treatment than is typically needed for the treatment of depression are needed for optimal SSRI response in OCD, at the group level. There is no evidence to support the use of one SSRI over another, at least on the basis of efficacy; studies done to date suggest that the six agents have equivalent efficacy.

Recommendations for future research

While the SSRIs have been clearly shown to work, there is marked heterogeneity in both the rate and the degree of response; 30–40% of cases of OCD are nonresponsive even to multiple SSRI trials and aggressive dosing. There are few if any clinically useful predictive markers of SSRI response; identification of such markers could substantially improve the efficiency of treatment. We know very little about how SSRIs work to ameliorate OCD symptoms; some studies show normalization of abnormal activity in the CSTC circuitry. Clarification of the relevant therapeutic mechanisms could help guide the development of other treatments that achieve the same benefits. The SSRIs, together with clomipramine, are the only proven first-line pharmacotherapy for OCD; this leaves individuals who do not respond to the SSRIs, or who cannot tolerate them, without other clear pharmacological treatment options. The development of mechanistically distinct alternatives for monotherapy is an important research goal.

The ability to understand and explain:

1. Evidence for the efficacy of the SSRIs in the treatment of OCD. All six SSRIs have been shown to be efficacious as monotherapy in OCD in multiple randomized clinical trials, convincingly summarized by recent meta-analyses.   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Ackerman and Greenland 2002, Soomro et al 2008, Soomro 2012, Skapinakis et al 2016a • SR: Skapinakis et al. (2016b)

2. The equivalent efficacy of the six SSRIs, at least at the population level, and to consider factors that may influence the choice of agent in individual cases   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Soomro et al 2008, Soomro 2012, Skapinakis et al 2016a • SR: Skapinakis et al 2016a

• The SSRIs have different side effect profiles, which may influence the selection of one agent over another • Different SSRIs have different pharmacokinetic properties, which may influence the choice of one agent over another • Individual or family history of response or of side effects to a particular agent may guide choice of agent • Differential SSRI metabolism may help explain why one agent is more effective than another in a particular patient; this is one motivation for trying a second SSRI if a first is ineffective.   Evidence: TG: NICE 2006, Koran et al 2007, Badelow et al 2012, Baldwin et al 2014 • MA: Skapinakis et al 2016a • SR: Skapinakis et al 2016b • OR: Fineberg et al 2016

• Pharmacogenetic testing is currently of limited utility and is not routinely recommended. Pharmacogenetic considerations may help inform choice of agent in the future.   Evidence: OR: Zai et al 2014), Bousman et al 2016, 2017

3. That higher doses of SSRIs, beyond the dose range typically used for depression and anxiety, are often needed for maximum effect in OCD • All treatment guidelines endorse the use of high doses of SSRIs, above the maximum doses approved by regulatory agencies, if the maximum approved dose is not effective but is well tolerated (with the exception of citalopram, which is generally limited by the risk of QTc prolongation and the attendant black box warning it carries – see below • Commonly used doses are 80 mg fluoxetine, 300 mg fluvoxamine, 250 mg sertraline, 40 mg escitalopram • Maximum daily doses in the APA guidelines are 120 mg fluoxetine, 100 mg paroxetine, 450 mg fluvoxamine, 400 mg sertraline, 60 mg escitalopram   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Bloch et al 2010, Issari et al 2016, Skapinakis et al 2016a,b

4. That treatment with SSRIs typically needs to be longer, compared with major depressive disorder and many other conditions, to achieve maximum benefit in treatment of OCD   Evidence: TG: NICE 2006, Koran et al 2007, Badelow et al 2012, Baldwin et al 2014 • MA: Skapinakis et al 2016a • SR: Skapinakis et al 2016b • OR: Fineberg et al 2016

• An ‘adequate treatment trial’ is typically defined as lasting 12 weeks Evidence: MA: Issari et al 2016

• Recent data suggest that symptom improvement begins soon after the initiation of SSRI dosing but accrues slowly, such that it is typically not detectable for many weeks, even in well-powered studies   Evidence: MA: Issari et al 2016

• Higher doses are associated with an increased incidence of side effects, and some individuals achieve an adequate clinical response at a submaximal dosage; the rate of up-titration of dosage must therefore be determined on an individual basis   Evidence: MA: Bloch et al 2010

5. The limitations on SSRI dosage due to cardiac concerns • Citalopram has a black box warning from the FDA for potential QTc prolongation   Evidence: TG: FDA 2012

• QTc prolongation is less pronounced for escitalopram; the US FDA has not issued a warning for escitalopram, although the Medicine and Healthcare Products Regulatory Agency of the United Kingdom has   Evidence: EO: Lam 2013, Hasnain et al 2013

• There is ongoing debate as to whether this QTc prolongation is clinically significant; nevertheless, it must be taken into account when considering high-dose citalopram, even if only for medicolegal reasons   Evidence: EO: Vieweg et al 2012

3.d.iii. Clomipramine

Overview.

Clomipramine, a tricyclic antidepressant drug, was the first medication proven efficacious in the treatment of OCD, in 1984. Its efficacy has been demonstrated in multiple placebo-controlled trials and by meta-analysis thereof. While head-to-head trials have not demonstrated superiority over SSRIs, some meta-analyses suggest that clomipramine has a somewhat larger effect size, and some experts consider it a more efficacious agent. However, it has a greater potential side effect burden than the SSRIs and is thus not considered a first-line agent. Common side effects include weight gain, sedation, and anti-cholinergic effects. At high doses, clomipramine can lower the seizure threshold and produce arrhythmias; monitoring of blood levels is necessary in the high dose range.

Definition of key terms.

tricyclic antidepressant (TCA).

A class of antidepressants, named for their tricyclic chemical structure, that acts on brain monoamines and other neurotransmitter systems.

clomipramine.

A tricyclic antidepressant that was the first drug shown to be efficacious in the treatment of OCD. Clomipramine has the highest serotonin transporter blocking activity of the tricyclics.

desmethylclomipramine.

A major metabolite of clomipramine, which is also measured when serum clomipramine levels are assayed, and which is thought to contribute significantly to side effects and potential toxicity.

Strength of available evidence.

The robust evidence for the efficacy of clomipramine in the treatment of OCD consists of multiple placebo-controlled trials, head-to-head trials, and meta-analyses. The larger effect size of clomipramine relative to SSRIs is well established by meta-analysis, although it has not been demonstrated in head-to-head comparisons.

Suggestions for future research.

Further tests of the efficacy of clomipramine are not needed. The relative efficacy of clomipramine and the SSRIs is not well established, although many experts believe that clomipramine is more efficacious. Early studies suggest a larger effect size, but this may because the populations treated in clomipramine vs SSRI trials differed in important ways, and early clomipramine studies did not use intent-to-treat analysis. Any differential effect size is small – a head-to-head trial would have to be very large to establish superiority, and such a trial is probably not justified. As with the SSRIs, the mechanisms whereby clomipramine reduces OCD symptoms are not well understood; this is an important area of ongoing research. Similarly, no well-established prospective markers are available to identify which patients are likely to respond to clomipramine treatment, in general or among those who have failed to improve after an adequate SSRI trial. The identification of predictive markers of drug response to help guide treatment selection would be a major advance.

The ability to understand and explain:

1. Evidence for the efficacy of clomipramine in the treatment of OCD. Clomipramine monotherapy has been shown to improve OCD symptoms in multiple well-controlled trials   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Skapinakis et al 2016a • SR: Skapinakis et al 2016b • RCT: Insel et al 1984

2. The side effects of clomipramine

• Common side effects include weight gain, sedation, dry mouth, and other anticholinergic effects • At high doses and in susceptible individuals, clomipramine can, rarely, precipitate bundle branch block and cardiac arrhythmia. A baseline EKG is recommended. • Clomipramine can lower the seizure threshold • Because of these side effects, regular checking of blood levels is recommended, and clomipramine dosing is not generally pushed to levels above the FDA-recommended maximum dose, as can be done with the SSRIs   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Skapinakis et al 2016a • SR: Skapinakis et al 2016b

3. Clomipramine metabolism and monitoring of blood levels • Clomipramine is metabolized by CYP2C19, and other liver enzymes, into desmethylclomipramine, which is also psychoactive • Desmethlyclomipramine is thought to be responsible for many of the side effects of clomipramine treatment   Evidence: TG: NICE 2006, Koran et al 2007

• Blood testing to monitor levels is necessary, especially when high doses of clomipramine are used   Evidence: CS: Marazziti et al 2012

• Some experts advocate the use of low doses of an inhibitor of CYP2C19, such as fluvoxamine, can be used to shift the clomipramine/desmethylclomipramine ratio   Evidence: EO: Stewart et al 2017

3.d.iv. Augmentation with dopamine D2 receptor antagonists

Overview.

Antagonists of the D2R dopamine receptor – that is, neuroleptics – can be added to SSRIs or clomipramine as augmentation agents; this is of benefit in about 25% of those who do not respond an SSRI (or clomipramine) alone. Interestingly, neuroleptics are not generally beneficial as monotherapy. Indeed, atypical neuroleptics, especially clozapine, can (rarely) induce OCD symptoms in susceptible individuals, especially when used without an SRI. The evidence for efficacy as augmentation is best for risperidone and aripiprazole. Individuals with tics, or a history of tics, may respond best to augmentation with neuroleptics with higher D2R affinity, such as haloperidol and risperidone.

Definitions of key terms.

neuroleptic.

Any of several drugs that act by antagonizing the D2R dopamine receptor; these drugs were primarily developed for their antipsychotic properties and are thus also called antipsychotics, even though they are frequently used in conditions other than psychosis. While all neuroleptics (except clozapine) are antagonists of the D2R receptor, the specific properties of individual neuroleptics are determined, in large part, by their relative affinities at other receptors.

• typical (first-generation) neuroleptic. An older class of neuroleptic medication, of which haloperidol and chlorpromazine are exemplars. As a class, compared to atypical neuroleptics, these medications are more likely to cause motor effects, such as extrapyramidal motor abnormalities, akathisia, and tardive dyskinesia.

  • haloperidol
• atypical (second generation) neuroleptic. A newer class of neuroleptic drug (also called ‘second-generation neuroleptics’), characterized by affinity for the 5-HT2A and 5-HT1A serotonin receptors as well as the D2R receptor. As a class, these medications have fewer motor side effects and a lower risk of tardive dyskinesia than the first-generation neuroleptics. Some can cause significant weight gain and sedation.

  • risperidone
  • olanzapine
  • quetiapine
  • aripiprazole – an atypical neuroleptic that is a partial agonist at the D2R receptor; it thus acts as an antagonist when dopamine tone is high, but as an agonist when dopamine tone is low.
  • brexpiprazole – also a partial agonist at the D2R receptor

• clozapine. A unique atypical neuroleptic with low affinity for D2R receptors but significant affinity for D4R dopamine receptors and a number of other neurotransmitter receptors

• high-affinity neuroleptic – a neuroleptic medication (either typical or atypical) with a relatively high affinity for the D2R receptor; doses used clinically are correspondingly low. Examples include haloperidol and risperidone.

• low-affinity neuroleptic – a neuroleptic medication (either typical or atypical) with a relatively low affinity for the D2R receptor; doses used clinically are correspondingly high. Examples include chlorpromazine and quetiapine.

extrapyramidal signs and symptoms.

A set of motor phenomena, overlapping with the symptoms of Parkinson’s disease, that derive from dysregulation of the basal ganglia; they are common side effects of the typical neuroleptics. They include tremor, rigidity, masked facies, akathisia, dystonia, and dyskinesia.

akathisia –

a uncomfortable sensation of muscular discomfort that drives incessant restless motion. Akathisia is seen as a side effect of numerous psychoactive medications, including the typical and, sometimes, the atypical neuroleptics.

tardive dyskinesia (TD) –

a persistent motor syndrome that sometimes emerges after longer-term treatment with neuroleptics, especially those with high D2R affinity when used at high doses. Symptoms consist of uncontrolled movements, most often of the lips, tongue, and fingers, and typically (though not always) of low amplitude. TD sometimes resolves gradually after discontinuation of the inciting agent but can be persistent. Clonazepam may be helpful in reducing symptoms. Two medications to treat TD have recently been approved by the FDA: valbenazine and deutetrabenazine.

Strength of evidence.

Evidence from multiple placebo-controlled studies and meta-analyses indicates modest benefit from neuroleptic augmentation in OCD that is unresponsive to SSRI or clomipramine monotherapy. Evidence is strongest for risperidone and aripiprazole but also exists for haloperidol; evidence is mixed for olanzapine, and quetiapine. There is limited evidence to guide decisions as to which of these agents, if any, is most likely to work in a particular patient. One RCT found greater benefit from augmentation with CBT than with risperidone; it is important to consider nonpharmacological options in SRI-unresponsive OCD. There have been no careful studies of neuroleptic augmentation in children. Convincing evidence, in the form of multiple case reports and case series and a few reports from larger studies, indicates that clozapine can induce obsessive-compulsive symptoms, especially when used without an SRI. The evidence for such an effect with other atypical neuroleptics is weaker, and expert opinions differ as to how large a clinical problem this is.

Suggestions for further research.

As with the serotonergic agents reviewed above, no strong clinical predictors can identify which patients are most likely to respond to neuroleptic augmentation. The one exception is that high-potency neuroleptics, such as risperidone and haloperidol, appear to be somewhat more likely to benefit individuals with past or current tics. The identification of predictive markers to help in identifying which patients to attempt neuroleptic augmentation in, and which agent to use, would be a notable advance. The mechanisms whereby neuroleptic augmentation produces benefit, when it does, remain to be elucidated; this is another important area for future research. Advances in both of these areas would be valuable because these agents tend to have more problematic side effects than the SSRIs. Prospectively identifying patients in whom risk/benefit considerations are favorable despite the potential side effects would be a major advance. Identification of the mechanisms of treatment response might lead to the development of novel interventions that produce similar benefits, without the potential side effect burden.

The ability to understand and explain:

1. Evidence for the efficacy of neuroleptic augmentation in the treatment of OCD refractory to SSRI or clomipramine monotherapy • Augmentation of stable SSRI or clomipramine monotherapy with low doses of neuroleptics has been shown to be of benefit in a minority of OCD patients who are refractory to monotherapy • The strongest evidence for benefit exists for risperidone and aripiprazole; evidence of benefit also exists for haloperidol but is limited for olanzapine and quetiapine   Evidence: MA: Ipser et al 2006, Bloch et al 2006, Veale et al 2014, Dold et al 2015

• Neuroleptic monotherapy does not appear to be of benefit in OCD, although the evidence base is limited   Evidence: TG: Koran et al 2007

• One study showed that adding CBT was more effective than adding risperidone in adults with clinically significant symptoms despite an adequate SRI trial   Evidence: RCT: Simpson et al 2013

2. Principles of dosing and monitoring of response • Typically low doses of neuroleptic are used for augmentation in OCD (e.g. 0.5–2.0 mg risperidone); the evidence for additional benefit from higher doses is very limited   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Ipser et al 2006, Bloch et al 2006

• Efficacy of neuroleptics as augmentation in OCD may correlate with D2R dopamine receptor affinity   Evidence: MA: Ducasse et al. 2014

• Onset of benefit, when it occurs, is more rapid than that of the SSRIs (typically 2–4 weeks), though full response takes longer (6–8 weeks).   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Ipser et al 2006 • EO: Maina et al 2003

3. The side effects of neuroleptics, and how they influence risk/benefit considerations when they are used in OCD • Many neuroleptics can cause sedation and, variably, anticholinergic side effects • First-generation neuroleptics frequently have motor or sensorimotor side effects including tremor and extrapyramidal motor symptoms, akathisia, and occasionally tardive dyskinesia • Some second-generation neuroleptics often lead to weight gain and insulin resistance • These potentially significant side effects dictate that trials of neuroleptic augmentation be of limited duration where possible, and that treatment be continued only when benefit is clear. Side effects that may not quickly resolve with discontinuation of the drug – tardive dyskinesia and weight gain – may be of particular concern   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Ipser et al 2006, Bloch et al 2006

4. The unique characteristics of clozapine and aripiprazole • Clozapine is uniquely efficacious in the treatment of psychosis but not in the treatment of OCD – indeed, clozapine monotherapy can induce OCD symptoms in susceptible individuals. Other atypical neuroleptics may also induce OCD symptoms when used without an SSRI, though the relationship is less clear.   Evidence: OR: Poyurovsky et al 2005 • RCT: McDougle et al 1995 • CR: Ghaemi et al 1995, Lykouras et al 2003

• Aripiprazole augmentation has been shown to be efficacious in the treatment of OCD. Unlike the other neuroleptics it is a D2R partial agonist and thus acts as an antagonist when dopamine tone is high, but as an agonist when dopamine tone is low. It also has serotonergic properties. Brexpiprazole has similar pharmacology but has not been studied in OCD.   Evidence: RCT: Veale et al 2014, Dold et al 2015

3.d.v. Less well proven tertiary pharmacological strategies

Overview.

Unfortunately, only a limited number of treatments that have been unambiguously shown to be efficacious in OCD: specialized cognitive-behavioral psychotherapy, SSRIs, clomipramine, and neuroleptic augmentation of SSRIs or clomipramine. Disease that is refractory to these proven treatments is not uncommon, even when they are optimally delivered. A large number of medications that have regulatory approval for other indications have been investigated off-label for the treatment of refractory OCD, with promising results for several agents. Most of these alternative strategies have focused modulation of the neurotransmitters serotonin or glutamate. None these alternative augmentation agents have been proven efficacious in multiple controlled trials or in meta-analysis. These less well proven strategies should therefore not, in general, be used until the better-proven approaches summarized above, including cognitive-behavioral therapy, have been exhausted.

Definitions of key terms.

buspirone.

A 5-HT1A serotonin partial agonist used for the treatment of anxiety

caffeine.

A stimulant and diuretic alkaloid found in coffee, many teas, and other food and beverage products that releases dopamine, increases cerebral concentrations of tryptophan, and affects serotonin release, uptake, and turnover.

d-amphetamine.

A stimulant drug that releases dopamine and norepinephrine and inhibits dopamine reuptake.

mirtazapine.

An alpha-2 adrenergic antagonist antidepressant; it secondarily modulates serotonin tone.

serotonin-norepinephrine dual reuptake inhibitors (SNRIs).

A class of antidepressants that block the reuptake of both serotonin and norepinephrine; these include venlafaxine, desvenlafaxine, and duloxetine.

psilocybin.

A hallucinogenic agonist of the 5-HT2A and 5-HT2C receptors. Psilocybin is scheduled drug that cannot legally be prescribed; recent very preliminary evidence suggests that it may have some therapeutic benefit in depression and perhaps OCD, but it is available only in specialized research settings.

glutamate.

The primary excitatory neurotransmitter in the brain; there is substantial recent interest in the idea that glutamate dysregulation contributes to OCD and thus that glutamate modulating drugs may be of therapeutic benefit.

• memantine. A noncompetitive antagonist of the NMDA glutamate receptor; memantine has regulatory approval for the treatment of moderate-severity Alzheimer’s disease.

• riluzole. A multifunctional glutamate modulator that reduces glutamate tone; rilzuole has regulatory approval for the treatment of amyotrophic lateral sclerosis.

• topiramate. A calcium channel blocker that modulates glutamate release; topiramate has regulatory approval for the treatment of epilepsy.

• lamotrigine. A sodium channel blocker that modulates glutamate release; lamotrigine has mood stabilizing properties and has regulatory approval for certain epilepsies and bipolar disorder.

• N-acetylcysteine (NAC). An antioxidant amino acid derivative that is widely available without prescription; it has been investigated in OCD, grooming disorders, and a variety of other conditions.

• ketamine. A potent noncompetitive NMDA glutamate antagonist; ketamine has regulatory approval as an anesthetic, but it has been the focus of substantial recent interest as a rapidly-acting antidepressant.

• esketamine. A racemically pure, intranasally delivered form of ketamine that is approved for the treatment of depression. Studies in OCD are limited.

pindolol.

A beta-adrenergic antagonist used to treat of hypertension; it has been studied in the treatment of OCD, including as an augmentation agent for CBT.

tryptophan.

An amino acid, the metabolic precursor of serotonin; it has been studied in a few cases of OCD, based on the theory that it may increase serotonin synthesis. Tryptophan is available as a non-prescription supplement.

morphine.

An agonist of the mu-opioid receptor, used as an analgesic agent; it has been used in the treatment of OCD in a single small study.

tramadol.

A synthetic opioid used as an analgesic agent; tramadol has been used in the treatment of OCD in a few small uncontrolled studies.

Strength of the available evidence.

As noted above in the Overview, all agents discussed in this section have only limited evidence supporting their efficacy. Small controlled trials are available in some cases, but they have not been of sufficient size and quality, or been replicated adequately, to constitute proof of efficacy. These treatment strategies are thus typically considered only after better-proven first and second-line treatments, including CBT, have been attempted.

Suggestions for future research.

Further efficacy studies are needed for all of the agents discussed in this section. Once efficacy is more clearly established, mechanistic studies examining the correlates of therapeutic benefit will be important.

The ability to understand and explain:

1. Evidence for the efficacy of serotonergic agents other than the SSRIs   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Skapinakis et al 2016a • SR: Skapinakis et al. (2016b)

• SSRI augmentation with high doses of buspirone is sometimes used for refractory OCD. Benefit was suggested by case studies and uncontrolled studies but was not confirmed in small double-blind trials.   Evidence: RCT: McDougle et al 1993, Pigott et al 1992 • OCT: Jenike and Baer 1988

• A single small trial suggests benefit from mirtazapine monotherapy of OCD   Evidence: RCT: Koran et al 2005

• Concurrent treatment with mirtazapine and an SSRI may accelerate therapeutic response, though it does not usually change overall improvement   Evidence: RCT: Pallanti et al 2004

• Low-dose ondansetron, a 5-HT3 antagonist used in the treatment of nausea, has been used in OCD; early studies were promising, but a large RCT (never published) was terminated due to lack of efficacy   Evidence: SR: Serata et al 2015 • OR: Andrade 2015 • RCT: Soltani et al 2010, Heidari et al 2014, NCT01275248 • OCT: Hewlett et al 2003, Pallanti et al 2014, 2019

• A single controlled trial has suggested benefit from the 5-HT3 antagonist granisetron   Evidence: RCT: Askari et al 2012

• Several newer serotonergic antidepressants (vilazodone, vortioxetine, levomilnacipran) have recently become available; however, there is no published data on their efficacy in the treatment of OCD, either as monotherapy or as augmentation   Evidence: None

• Case reports suggest benefit from adding L-tryptophan, a serotonin precursor, to SSRI treatment. L-tryptophan was linked to an outbreak of eosinophilia myalgia syndrome, and 36 deaths, in 1989; this appears to have been due to contamination in the supply chain   Evidence: CR: Blier and Bergeron 1996, Yaryura-Tobias and Bhagavan 1977

• Case reports and one small uncontrolled trial suggest rapid benefit from the hallucinogenic serotonin agonist psilocybin in OCD; however, this agent is not legally available outside of specialized research settings   Evidence: OCT: Moreno et al. (2006)

2. That the evidence for benefit from treatment with the SNRI antidepressants in OCD is quite limited   Evidence: TG: NICE 2006, Koran et al 2007 • MA: Skapinakis et al 2016a • SR: Skapinakis et al 2016b

• Small trials suggest some benefit from higher-dose (225–350 mg) venlafaxine monotherapy in the treatment of OCD; it is unclear whether there is any benefit beyond that produced by the SSRIs   Evidence: RCT: Denys et al 2003, 2004 • OCT: Albert et al 2002

• A single trial suggests that, after nonresponse to an SSRI, switching to venlafaxine may be less efficacious than a second SSRI trial   Evidence: RCT: Denys et al 2004

• A large case series suggests benefit from venlafaxine in a minority of patients   Evidence: CR: Balachander et al 2019

• Response to duloxetine has been reported in case reports and one case series, but there have been no controlled trials   Evidence: CR: Dell’osso et al 2008, Juis Blay and Black 2007, Yeh et al 2009

3. The evidence for benefit from augmentation with glutamate-modulating agents in OCD   Evidence: MA: Zhou et al 2019 • OR: Pittenger et al 2011, Pittenger 2015, Pittenger 2017b

• Memantine augmentation has shown promise in case studies and open-label trials   Evidence: CR: Poyurovsky et al 2005, Pasquini et al 2006 • OCT: Aboujaoude et al 2009, Feusner et al 2009, Stewart et al 2010, Bakhla et al 2013

• Two RCTs of memantine showed anomalously high response rates (100% in one study), which is not consistent with the broader literature; further research is needed   Evidence: RCT: Ghaleiha et al 2013, Haghighi et al 2013

• Riluzole augmentation showed promise in a case series and open-label trial. A small controlled study failed to prove efficacy; however, riluzole may have been of benefit to a subset of patients, especially for obsessions. A second small RCT suggested benefit   Evidence: RCT: Pittenger et al 2015, Emamzadehfard et al 2016 • OCT: Coric et al 2005 • OR: Pittenger et al 2008

• Topiramate augmentation showed benefit in case reports, open label trials, and one small RCT. A second RCT showed improvement in compulsions, but not overall OCD. Side effects can be limiting   Evidence: RCT: Mowla et al 2010, Berlin et al 2011 • CR: Rubio et al 2006, Van Ameringen et al 2006, 2015

• Lamotrigine augmentation has been reported to be of benefit in some case reports but not in others. Benefit has been reported in two small RCTs   Evidence: RCT: Bruno et al 2012, Khalkhali et al 2016 • CR: Kumar et al 2000, Uzun et al 2010, Arrojo-Romero et al 2013, Hussain et al 2015

• Limited evidence suggests that N-acetylcysteine (NAC), an antioxidant amino acid derivative with glutamate modulating properties, is of benefit in some cases of refractory OCD. Evidence from four RCTs is mixed • RCTs that showed benefit had shorter duration and lower NAC doses (up to 2400 mg/dy), and included participants taking only an SSRI, not other medications   Evidence: RCT: Afshar et al 2012, Sarris et al 2015, Paydary et al 2016, Costa et al 2017 • CR: Lafleur et al 2006

4. Current research into the use of ketamine infusion in the treatment of OCD

• Intravenous ketamine infusion has been shown to produce a robust, though transient, antidepressant effect   Evidence: OR: Sanacora et al 2017

• Ketamine has been reported to be of benefit in unmedicated OCD patients   Evidence: RCT: Rodriguez et al 2013 • CR: Rodriguez et al 2011, 2016

• In an open label trial of medicated patients with multiple comorbidities, depression improved but benefit for OCD symptoms was minimal, and two patients developed new-onset irritability and suicidal ideation. Caution is warranted.   Evidence: OCT: Bloch et al 2012 • CR: Niciu et al 2013

• Intranasal ketamine is effective for depression; experience in OCD is limited   Evidence: OR: Rodriguez et al 2017 • RCT: Lapidus et al 2014 • CR: Adams et al 2017

5. The mixed evidence for benefit from the beta adrenergic antagonist pindolol in treatment-refractory OCD   Evidence: RCT: Mundo et al 1998, Dannon et al 2000

6. The limited evidence for benefit from the stimulants d-amphetamine and caffeine in the treatment of OCD   Evidence: RCT: Koran et al 2009

7. The limited evidence for benefit from opiate agonists in refractory OCD   Evidence: RCT: Koran et al 2005 • OCT: Shapira et al 1997 • CR: Warneke et al 1997

• Opiate agonist treatment requires careful management; contraindications include history of substance use disorder or prescription medication mis-use, psychosis, mania, antisocial personality disorder, COPD, cardiovascular compromise   Evidence: TG: Koran et al 2007

• Addition of the synthetic opiate tramadol to SSRIs carries a risk of serotonin syndrome   Evidence: TG: Koran et al 2007

8. Pharmacological strategies that good trials indicate do not work as augmentation in OCD

• Lithium augmentation of antidepressant monotherapy is not effective in refractory OCD   Evidence: RCT: Pigott et al 1991, McDougle et al 1991

• Thyroid augmentation of clomipramine does not appear to reduce OCD symptoms   Evidence: RCT: Pigott et al 1991

• Neuroleptic monotherapy (as opposed to SRI augmentation) does not appear to be of benefit in OCD   Evidence: TG: Koran et al 2007

• Tricyclic antidepressants other than clomipramine (which has particularly strong serotonin reuptake blocking effect) are not effective in OCD   Evidence: TG: Koran et al 2007 • RCT: Barr et al 1997, Hoehn-Saric et al 2000

3.e. Treatment options for refractory disease

Overview.

While there are numerous evidence-based treatments for OCD, both pharmacological and psychotherapeutic, a distressingly large minority of patients do not respond to these interventions, even when optimally delivered. How to manage treatment-resistant and treatment-refractory disease is therefore a common problem in specialized clinical practice. A first step in the management of refractory disease is always to confirm that evidence-proven treatments have been optimally delivered; careful review commonly discovers that pharmacological trials have been inadequate (e.g., insufficient SSRI dose or duration, poor treatment adherence), or that CBT has not been expertly delivered. When a case is determined to in fact be refractory to these proven interventions, less well proven pharmacological augmentation strategies, as summarized above, may be tried. Alternatively, more intensive CBT, including multimodal treatment in intensive outpatient and residential settings, may be attempted. These approaches are not covered in detail here, as our focus is on psychopharmacological and other somatic interventions, but considering intensive CBT is essential when treating refractory patients. Finally, a range of brain stimulation manipulations, including transcranial magnetic stimulation (TMS), deep brain stimulation (DBS), and targeted neurosurgical lesions, can be considered. Good evidence exists for the efficacy of all of these strategies, in appropriately selected individuals. These targeted manipulations require particular expertise and are offered only at specialized centers.

Definition of key terms.

Treatment-resistant OCD –

OCD that remains clinically significant despite optimal application of proven treatments

Refractory OCD –

OCD that remains clinically significant despite rigorous application of many evidence-based treatments, including those for which evidence is more limited.

Electro-convulsive therapy (ECT) –

Electrical induction of therapeutic seizure for the purpose of treating psychiatric symptoms. ECT is highly effective for severe depression, but administration for OCD is not evidence based and benefit in OCD is minimal.

Transcranial magnetic stimulation (TMS) –

targeted stimulation of specific brain regions or circuits using rapidly alternating magnetic fields. Mounting evidence suggests that appropriately targeted TMS can benefit some cases of OCD.

Deep TMS –

A newer approach to TMS that uses magnetic coils designed to stimulate deeper structures, such as the cingulate cortex and insula.

Deep brain stimulation (DBS) –

targeted stimulation of deep structures in the brain using surgically implanted electrodes; this is a standard therapy in movement disorders and is increasingly turned to as an option in profoundly refractory OCD.

Cingulotomy –

a neurosurgical lesion of the dorsal cingulate, used as a treatment option in profoundly refractory OCD.

Capsulotomy –

a neurosurgical lesion of the white matter of the anterior internal capsule, used as a treatment option in profoundly refractory OCD.

Strength of evidence.

Studies of refractory OCD tend to be small, and the patients included tend to have many comorbidities and thus to be quite heterogeneous. These characteristics limit the conclusions that can be drawn from this literature. Despite this, there is mounting evidence of benefit from intensive CBT, TMS, neurosurgical lesions, and deep brain stimulation (DBS), in appropriately selected patients treated at specialty centers.

Suggestions for future research.

As for other treatments summarized above, a key gap in the literature is an understanding of which patients are most likely to respond to which treatments. Lacking this knowledge means that treatment is usually based on a statistical understanding of response probability, side effects, patient preference and other incidental factors, and trial-and-error. This lack of knowledge delays effective treatment and can expose patients to side effects unnecessarily (at least in retrospect). These concerns are all the more acute when considering invasive treatments such as DBS that have the potential for severe adverse events and lasting side effects. There, studies to identify prospective markers of response to various treatments are urgently needed. Because studies using invasive tertiary treatment strategies in refractory patients are likely to always be small in size, creative experimental design based on a detailed understanding of pathophysiology is necessary to fill this gap in the treatment research literature. The development of research registries that allow systematic accumulation of data from patients treated at many different sites is needed to overcome the limited numbers typically treated with invasive tertiary approaches at any one site.

The ability to understand and explain:

1. Definitions of treatment-resistant and treatment-refractory OCD   Evidence: ThP: Farris et al 2013 • EO: Albert et al 2013, Mataix-Cols et al 2016

2. Pharmacological options for refractory OCD (as reviewed above)   Evidence: TG: Koran et al 2007 • MA: Skapinakis et al 2016a • OR: Pittenger and Bloch 2014

3. Intensive specialized CBT for refractory OCD   Evidence: MA: Veale et al 2016 • EO: Stewart et al. (2005); Brennan et al. (2014)

4. That electroconvulsive therapy (ECT) is generally not of benefit in OCD • While ECT can be helpful for comorbid depression, it does not appear to be of benefit for OCD symptoms • There have been no RCTs of ECT in OCD to date   Evidence: OR: Fontenelle et al 2015

5. The potential use of TMS for refractory OCD • TMS targeting dorsolateral prefrontal cortex, the primary target used in the treatment of depression, is of little or no benefit in OCD   Evidence: MA: Trevizol et al 2016 • SR: Rapinesi et al 2019

• TMS targeting more medial structures, such as the supplementary motor area (SMA), appears to be of benefit in OCD in sham-controlled trials, although variability in cortical targets and small sample sizes in individual studies limit conclusions that can be drawn at this point   Evidence: MA: Trevizol et al 2016 • SR: Rapinesi et al 2019

• More recently developed deep TMS was shown to be efficacious and has recently received FDA approval for the treatment of refractory OCD   Evidence: RCT: Carmi et al 2019

6. The ability to understand and explain the efficacy of targeted neurosurgical lesions as a treatment option in severe, refractory OCD   Evidence: OR: Greenberg et al 2010) • Keen et al 2017

• Targeted neurosurgical lesions have been used in severely refractory cases of OCD for decades; benefit in many patients is well established, and progressive refinement of technique has reduced adverse effects   Evidence: OR: Greenberg et al 2010, Keen et al 2017

• Neurosurgical anterior cingulotomy shows efficacy, limited side effects, and good long-term outcomes in appropriately selected refractory patients   Evidence: CS: Dougherty et al 2002

• Targeted lesions of the ventral internal capsule, which can be performed without breaching the skull using a gamma knife, show benefit in appropriately selected cases, with good long-term outcomes   Evidence: RCT: Lopes et al 2014 • CS: Sheth et al 2014, Rasmussen et al 2018

• Stereotactic subcaudate tractotomy, which entails disruption of white matter tracts in the substantia innominata and is thought to disrupt connections between OFC and subcortical structures, has shown lasting benefit in a range of severe and refractory psychiatric conditions; the focus has been on psychotic or agitated depression   Evidence: CS: Bridges et al 1994

• Limbic leucotomy, which combines anterior cingulotomy and ventral capsulotomy, has shown benefit in appropriately selected cases of profoundly refractory OCD.   Evidence: CR: Kim et al 2002, Montoya et al 2002

7. The ability to understand and explain the use of deep brain stimulation in the treatment of refractory OCD   Evidence: MA: Martinho et al 2020 • OR: Greenberg et al 2010, Goodman et al 2017

• Deep brain stimulation (DBS), which was developed for the treatment of movement disorders, presents the promise of targeted, adjustable, reversible perturbation of brain circuitry   Evidence: MA: Alonso et al 2015

• DBS targeting the subthalamic nucleus, which is also used in the treatment of refractory Parkinson’s disease, showed efficacy in a blinded, sham-controlled trial   Evidence: RCT: Mallet et al 2008

• DBS targeting the anterior limb of the internal capsule has shown promise in controlled trials; it was approved for clinical use by the US FDA under a humanitarian device exemption in 2008   Evidence: MA: Martinho et al 2020 • RCT: Goodman et al 2010 • OR: Goodman et al 2017

8. To understand and explain the relative advantages and complications/side effects of ablative neurosurgery and DBS in the treatment of severe, refractory OCD.   Evidence: MA: Kumar et al, 2019, Hageman et al 2021 • SR: Pepper et al 2015

4. SPECIFIC COMPETENCIES

4.a. Assessment and diagnostic competencies.

4.a.i. Clinician-patient relationship, capacity, and consent; attitudes towards and barriers to care.

Overview.

As in all psychiatric care, a trusting clinician-patient relationship is essential to effective diagnosis, patient engagement, patient acceptance and compliance with prescribed treatment, and accurate assessment of symptom change. The early establishment of a positive clinician-patient relationship is therefore of the utmost importance (Koran et al. 2007).

Patients must provide informed consent to treatment and must have the capacity to do so. These principles are applicable to the treatment of all psychiatric conditions. Patient ambivalence about treatment can be a challenge. Particular sources of ambivalence about pharmacological treatment in individuals with OCD include limited insight, worry that obsessions may reflect reality and thus that reducing compulsions may be immoral or create risk of harm, and heightened concern about side effects or risk to bodily integrity from somatic treatments. Patients with long-standing severe symptoms may find it difficult to imagine what life would be like without them, which can create a paradoxical ambivalence about getting better. A strong clinician-patient relationship is essential to navigating these complexities. Clarifying the patient’s understanding of the nature and meaning of their symptoms, the cause(s) of their distress, and their goals in treatment is essential; mis-alignment between the clinician and the patient on these matters is likely to reduce the efficacy of any treatment.

Cultural factors and language barriers may create challenges to establishing an effective therapeutic alliance. For example, religious obsessions and compulsions may be understood differently by individuals from different religious backgrounds. Attitudes towards contamination and cleanliness, or towards taboo sexual thoughts, may similarly differ across cultures or types of religious beliefs. Educational level and psychological-mindedness may influence which strategies of explanation and communication are most effective. Language barriers must be addressed to ensure that a patient adequately understands therapeutic options and is providing truly informed consent to treatment. Ideally, the patient and clinician are fluent in a shared language; absent this, using a professional translator with experience in psychiatric matters may be necessary.

Strength of evidence

Building of a strong clinician-patient relationship is considered essential to effective treatment throughout medicine. The need for clear communication (with translational where necessary) and informed consent are ethical precepts, rather than empirical considerations that are subject to support or refutation by evidence.

The ability to:

1. Form a therapeutic alliance with a range of patients, with sensitivity to cultural differences, language barriers, religious beliefs, and other complicating factors   Evidence: TG: Koran et al 2007

2. Assess patient capacity to consent for treatment, with particular reference to how OCD symptoms and comorbidities may impair capacity   Evidence: TG: Koran et al 2007 • EO: American Psychiatric Association 2013a

3. Engage patients in informed consent for diagnosis and treatment, with the aim of attaining alignment on goals over the course of treatment   Evidence: TG: Koran et al 2007 • EO: American Psychiatric Association 2013a

4. Assess patient attitudes towards available treatments and to understand their impact on treatment outcome • Biological vs. psychological conceptions of disease   Evidence: EO: Patel and Simpson 2010

5. Assess barriers to care for patients with OCD • Stigma • Lack of knowledge • Poor access to care • Inconvenience of treatment • Resistance to diagnosis or to the idea of psychiatric care, or some of its components   Evidence: CSS: Garcia-Soriano et al 2014

4.a.ii. Assessment of key symptom domains.

As described at length above, OCD is markedly heterogeneous. This raises challenges to differential diagnosis and treatment planning. For example, OCD and obsessive-compulsive personality disorder (OCPD) often co-occur and can sometimes be difficult to distinguish; this can have significant import for treatment selection, as the evidence for efficacy of SSRI pharmacotherapy is far stronger for symptoms of OCD than for OCPD. It is therefore essential that the clinician be experienced in the major symptom domains of OCD, the heterogeneity of their presentation, and common diagnostic dilemmas, especially to the extent that diagnostic subtleties inform treatment decisions.

Various authors, and different strands of research, have emphasized different ways of dividing and describing OCD symptom clusters and domains. For example, Penzel (2016, 2017) presents a useful taxonomy of common symptomatic presentations seen in a psychotherapy practice, based on decades of clinical work. A literature of rigorous factor analytic studies suggests a somewhat different taxonomy, emphasizing the distinction between contamination symptoms, fear-of-harm/checking symptoms, somatic symptoms, intrusive taboo thoughts, and symmetry/ordering symptoms (Bloch et al. 2008b). Others have emphasized the distinction between fear-of-harm and incompleteness as affective drivers of OCD symptomatology (Summerfeldt et al. 2014; Pallanti et al. 2017). For the purpose of building an understanding of the breadth and heterogeneity of OCD symptomatology, any of these taxonomies, be they based on dimensions or exemplars, may be useful to the clinician. To the extent that they reflect differences in underlying pathophysiology and/or that they can usefully inform treatment, the relative merits of different ways of describing OCD symptomatology are still to be determined empirically; this is an area of active ongoing study.

Definition of key terms

Obsessions, compulsions, insight, avoidance, tics, incompleteness.

See above, Section 3.a.1.

OCD subtypes.

Commonly seen clinical presentations of OCD, such as contamination OCD, fear-of-harm OCD, or taboo thoughts OCD. These are useful clinical labels that can adequately describe some cases, but their descriptive utility should not be taken to mean that they represent distinct underlying entities; very few clinical subtypes have been shown to predict treatment response, and the effect sizes are small (see below). Commonly described OCD subtypes are discussed above in Section 3.A.1, Basic phenomenology, diagnosis, and heterogeneity. Of note, these ‘subtypes’ are not mutually exclusive entities; symptoms from multiple subtypes often co-occur in individual patients.

Overvalued idea.

A pervasive, strongly held idea associated with idealized values, which has developed such an over-riding importance that it comes to define the ‘self’ or identity of the individual. Idealized values are characterized by the rigidity with which they are held. Patients with overvalued ideas may be unable to adapt to different circumstances and ignore the consequences of acting on these values. Overvalued ideas are characteristic of OCPD and autism but can also be present in OCD (Veale, 2002).

Rumination.

A pattern of passive and perseverative thoughts, without initiation of active problem solving that might alter the perceived causes of the distress. Ruminations are common in depressive and anxiety disorders; they are classically distinguished from obsessions by their ego-syntonic nature (i.e. they are consistent with patients’ expressed values and sense of self) and their focus on depressive themes of loss, rejection, or worthlessness – typically on questions that cannot be answered (Nolen-Hoeksema et al. 2008; Baer et al. 2017).

Worry.

A pattern of apprehensive expectation characterized by excessive anxious preoccupation with everyday problems or activities. Worries are typically distinguished from obsessions by their realistic nature, lower degree of rigidity, and lack of associated compulsive behaviors (Baer et al. 2017).

Delusion.

A fixed, false belief about the environment. Insight in OCD can range from excellent, in which obsessions are viewed as clearly unrealistic (though still emotionally powerful) to absent, in which obsessions can constitute delusional beliefs. Insight in OCD is usually in the fair to good range; fewer than 5% of OCD patients have delusional OCD beliefs. In these cases, OCD must be distinguished from delusional beliefs accompanying a psychotic disorder; this distinction can sometimes be challenging.

Strength of evidence

Diagnostic characteristics and other features common in OCD are derived from expert clinical observation and clinical phenomenology and have been validated by more rigorous study showing that clinical distinctions map onto other objective measures. For example, insight has been shown to inversely correlate with disease severity and prognosis (Foa et al. 2005; Hamblin et al. 2017); tic comorbidity is associated with genetic signal and with differential response to pharmacotherapy (Leckman et al. 2010). The distinction between obsessions and other distressing thought patterns (worry, rumination) is validated through association with other measures; effective CBT strategies for these different types of clinical phenomenology overlap but have some important differences. Dimensions of OCD symptomatology (contamination, fear-of-harm/checking, symmetry/ordering) have been replicated in multiple factor analytic studies and meta-analysis (Bloch et al. 2008b), but their association with genetics, imaging, treatment outcome, or other objective measures remains unclear.

Suggestions for future research

Further work characterizing OCD subtypes that can reliably predict treatment response or other clinically important variables, based on phenomenology or on other tractable sources of information, is needed, to enable more precise or personalized treatment approaches. The anxiety/incompleteness dichotomy is an interesting area of research and merits further investigation. Characterization of the underpinnings of phenomenological constructs (obsessions; worry; rumination) is needed.

The ability to:

1. Assess obsessions, compulsions, anxiety, incompleteness, disgust, fear-of-harm, major symptom domains.   Evidence: See Section 3.A.1: Basic phenomenology, diagnosis, and heterogeneity

2. Distinguish between obsessions and mental compulsions, and to recognize that patients who self-identify as having only obsessions (sometimes called ‘pure-O) very often have mental compulsions such as counting, mental recitation, or review, or avoidance compulsions.   Evidence: TG: Koran et al 2007 • EO: American Psychiatric Association 2013b

3. Recognize avoidance as an alternative strategy to the management of obsessions, which can independently lead to distress and impairment   Evidence: TG: Koran et al 2007 • EO: American Psychiatric Association 2013b

4. Assess symptoms of depression, and to distinguish between obsessions and depressive ruminations   Evidence: OR: Baer et al 2017, Nolen-Hoeksema et al 2008 • EO: American Psychiatric Association 2013b

5. Assess symptoms of anxiety, and to distinguish between obsessions and worry   Evidence: OR: Baer et al 2017 • EO: American Psychiatric Association 2013b

6. Distinguish between obsessions and overvalued ideas, recognizing that overvalued ideas may be present in individuals with OCD   Evidence: EO: Veale 2002

7. Assess age of onset and historical course of symptoms, including fluctuation over time and relationship to major life events   Evidence: TG: Koran et al 2007 • EO: American Psychiatric Association 2013b

8. Assess insight in OCD, and to recognize how limited insight may complicate treatment; the ability to appropriately assign the DSM-5 ‘insight’ specifier to the diagnosis   Evidence: OR: Hamblin et al 2017 • CSS: Foa et al 1995 • EO: American Psychiatric Association 2013b

9. Distinguish between obsessions with poor (delusional) insight and the delusions seen in psychotic disorders   Evidence: OR: Hamblin et al 2017 • EO: American Psychiatric Association 2013b

10. Distinguish between autistic preoccupations and obsessions, and between stereotypies and compulsions   Evidence: OR: Paula-Pérez 2013, Wu et al 2014 • EO: American Psychiatric Association 2013b

11. Assess level of distress, quality of life, and functional impairment in OCD, and the relationship between symptomatology and quality of life   Evidence: OR: Boisseau et al 2017 • CSS: Ruscio et al 2010

12. Assess suicidality and risk factors for self-harm in patients with OCD, and to recognize the impact of comorbid symptomatology on suicide risk   Evidence: MA: Angelakis et al 2015 • EO: American Psychiatric Association 2013

13. Assess current or past tics in patients with OCD, and to appropriately assign the ‘ticrelated’ specifier to the diagnosis   Evidence: OR: Leckman 2010 • EO: American Psychiatric Association 2013

The ability to distinguish between complex motor tics and compulsions   Evidence: OR: Franklin et al 2012, Kircanski et al 2013

The ability to distinguish between OCD and the behavioral and cognitive rigidity seen in obsessive-compulsive personality disorder   Evidence: TG: Koran et al 2007 • OR: Diedrich and Voderholzer 2015, Starcevic and Brakoulias, 2017, Wheaton and Pinto 2017 • EO: American Psychiatric Association 2013

The ability to assess the relationship between obsessions and culturally appropriate anxieties related to sexual orientation or other aspects of sexuality   Evidence: TG: Koran et al 2007 • OR: Williams et al 2015

The ability to assess the relationship between obsessions, especially scrupulosity and intrusive thoughts about blasphemy, and culturally appropriate religious concerns or preoccupations   Evidence: TG: Koran et al 2007

4.a.iii. Assessment of comorbidity.

Comorbidity is the norm in OCD and can significantly influence the choice and sequence of treatment modalities and can have a marked impact on outcomes. For example, some patients have difficulty engaging in ERP due to comorbid depressive symptoms and may benefit from initial treatment of depression, using pharmacotherapy or targeted psychotherapeutic strategies such as behavioral activation. As another example, the presence of comorbid tic disorders can impact the choice of pharmacotherapy, as OCD with comorbid tics is more likely to respond to SRI augmentation with D2 antagonists. Patterns of comorbidity are discussed above (Section 3.a.ii); the focus here is on how the presence of comorbidity influences treatment decisions.

Definition of key terms

Comorbidity –

See above, section 3.a.ii.

Strength of evidence

Comorbidity in OCD has been described in multiple robust epidemiological and clinical studies, with thousands of individuals. Data to guide treatment in patients with particular patterns of comorbidity are more limited.

Suggestions for future research

Patients with substantial comorbidity, especially with severe depression, autism, or psychosis, are often excluded from research studies; there is, therefore, limited evidence to guide their treatment. This is an important direction for future research. The frequency of comorbidity in OCD suggests overlapping pathophysiology with other conditions, especially mood, anxiety, and tic disorders; ongoing investigations of pathophysiology are needed to better elucidate these relationships.

The ability to:

1. Identify OCRDs (BDD, hoarding disorder, trichotillomania, excoriation [skin-picking] disorder), to distinguish them from OCD, and to diagnose them, either as primary diagnoses or as comorbid with OCD   Evidence: CR: Pertusa et al 2010 • TM: Phillips and Stein 2015 • EO: American Psychiatric Association 2013b

2. Identify when symptoms of OCRDs are the result of some other condition, such as: • stimulant drugs • basal ganglia lesions • Alzheimer’s disease (which may lead to clutter suggestive of hoarding disorder) • scabies or medical condition causing pruritus (which may mimic excoriation disorder) • seborrheic dermatitis or tinea capitis or medical condition causing skin inflammation (which may mimic trichotillomania)   Evidence: TM: Phillips and Stein 2015 • EO: American Psychiatric Association 2013b

3. Identify and diagnose patients with “other specified” OCRDs, such as: • BDD-like disorder in the presence of actual flaws in appearance • body-focused repetitive behavior disorder (e.g., nail biting, lip biting, cheek chewing) • obsessional jealousy • Koro (intense anxiety that the penis or vulva and nipples will recede into the body) • olfactory reference disorder • Olfactory reference disorder (preoccupation with the inaccurate belief that one emits a foul or unpleasant body odor, which is often accompanied by repetitive behaviors, such as checking)   Evidence: TM: Phillips and Stein 2015 • EO: American Psychiatric Association 2013b

4. Identify and diagnose major depressive disorder and bipolar disorder, either as a primary diagnosis or as comorbid with OCD   Evidence: OR: Stefanis et al 2002 • EO: American Psychiatric Association (2013b)

5. Identify and diagnose anxiety disorders – generalized anxiety disorder, social anxiety disorder, specific phobias, panic disorder (with or without agoraphobia), illness anxiety disorder – either as primary diagnoses or as comorbid with OCD   Evidence: EO: American Psychiatric Association 2013b

6. Identify psychosis and diagnose schizophrenia and other primary psychotic disorders, autism spectrum disorder, and post-traumatic stress disorder, either as primary diagnoses or as comorbid with OCD   Evidence: OR: van Os and Kapur 2009 • TM: Nutt et al 2000, Gallo et al 2010, Friedman et al 2014 • EO: American Psychiatric Association 2013b

7. Recognize and characterize the relationship between acute or chronic stress or trauma and the presentation of OCD   Evidence: OR: Adams et al 2018

8. Identify and diagnose substance use disorders, either as primary diagnoses or as comorbid with OCD, including substance abuse as self-medication   Evidence: CSS: Mancebo et al 2009 • EO: American Psychiatric Association 2013b

9. Recognize and diagnosis Tourette syndrome, other tic disorders, and attention deficit disorder, either as primary diagnoses or as comorbid with OCD   Evidence: TG: Koran et al 2007 • TM: Martino et al 2013 • EO: Tourette Syndrome Classification Study Group 1993, American Psychiatric Association 2013b

10. Recognize and diagnosis OCPD and other personality disorders, either as primary diagnoses or as comorbid with OCD   Evidence: TG: Koran et al 2007 • OR: Diedrich and Voderholzer 2015, Starcevick and Brakoulias 2017, Wheaton and Pinto 2017 • TM: Gabbard 2005 • EO: American Psychiatric Association 2013b

4.a.iv. Assessment of relevant clinical history.

The collection of an appropriately detailed clinical history is a cornerstone of diagnosis and treatment planning and is a core competency of good clinical practice for any medical or psychiatric condition. Collecting a good history can be complicated by the vagaries of patient self-report, especially for distant events and for technical details such as the dosage of past medications; by patient reluctance to discuss certain issues, such as suicidality, drug use, or symptoms with taboo content (e.g. sexual obsessions); and by limitations such as time. Careful clinician preparation can mitigate these difficulties.

Definition of key terms

Clinical history –

defined elsewhere in this document

Strength of evidence

Across medicine, clinical history has been found to be essential to accurate diagnosis and treatment selection. This assertion is difficult to assess rigorously, however. Historical details of symptom onset and fluctuation, family history, and comorbidity have been shown in research studies to predict long-term clinical course of OCD, but in adults these effects are by and large small and may be of limited utility in treatment planning for individual cases. Historical details of past medication trials, on the other hand, are particularly essential (e.g. past agents used, maximum dose, trial duration, benefits, side effects, reason for discontinuation). Previous failure to respond to an adequate trial of a specific medication predicts a reduced likelihood of responding to repeated attempts of the same or related medication.

Suggestions for future research

The field badly needs comprehensive longitudinal studies of OCD, to better delineate predictors of clinical course and of the likelihood of response to various interventions. The identification of historical or diagnostic predictors that are sufficiently robust to inform treatment selection in individual patients would be an enormous advance.

The ability to assess:

1. Current and historical pharmacological trials for OCD and comorbid disorders, including dose, duration of treatment, efficacy, side effects, and adherence • Trials of SSRIs or clomipramine • Non-SSRI antidepressant trials • History of use of benzodiazepines and other medications for anxiolysis or sleep, recognizing that sedatives can assist with sleep onset but disrupt the quality of sleep and produce rebound insomnia and other difficulties • Trials of pharmacological augmentation for OCD (evidence-based and other) • Other pharmacotherapy for psychiatric symptoms   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein, (2014)

2. Current and historical psychotherapy for OCD and comorbid symptomatology • ERP treatment, including intensity, duration, fidelity, completion of homework • Other CBT-based treatment, including cognitive therapy targeting OCD symptoms • Mindfulness/ACT treatment • Supportive, dynamic, or other psychotherapy   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

3. Any other past and current somatic treatments for OCD • Transcranial magnetic stimulation • Deep brain stimulation • Ablative neurosurgery • ECT   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

4. Any past or current non-conventional or non-medical treatments for OCD, such as: • Herbal or over-the-counter remedies • Homeopathy • Naturopathy • Chiropractic • Spiritual or religious remedies • Self-medication • Illicit drugs   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

5. Social and occupational functioning, level of impairment, and reduced quality of life due to OCD and any comorbid symptoms • Relationship status and relationship history; interference with social or partner relationships due to symptoms; possible partner accommodation • Educational attainment; interference with educational attainment due to symptoms • Employment or role status; interference with employment status or history due to symptoms   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

6. Sexual history and function as it relates to OCD symptomatology • Interference with sexual function due to symptomatology • Interference with sexual function due to medication side effects • Possible association of symptoms with sexual development, especially in women (menarche, menstrual cycle, pregnancy)   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

7. Relevant medical history • Comorbid medical conditions that may influence treatment or course, including those that may limit treatment adherence (e.g. difficulty with swallowing, mobility) and those that may inform choice of medication (e.g. cardiac history) • Concomitant medications or other medical treatments; possibility of drug-drug interactions   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

8. Relevant family history, including: • OCD • tic disorders • OCRDs • other major psychiatric disorders • suicide   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

9. Family/social context and support • Support for treatment adherence, including both emotional and practical considerations (e.g. transportation, finances, childcare) • Assessment of patient’s role obligations and how they may be influenced by symptoms and by treatment • Presence of family accommodation; need for psychoeducation for primary supporters • Presence of psychiatric symptoms, high expressed emotion, accommodation, or other characteristics that may influence treatment in primary supports   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016 • TM: Phillips and Stein 2014

4.a.v. Use of quantitative measures to inform treatment.

A number of well-validated instruments are available to assess the severity of OCD and comorbid symptomatology, as well as other aspects of patient functioning (e.g. psychosocial functioning and quality of life). These can be invaluable to the delivery of care, in initial assessment, monitoring treatment response, and deciding when to consider switching treatment modality or augmenting with another approach. Here we summarize standard instruments used to assess OCD, anxiety, and depressive symptomatology. These are selected as being the most relevant to clinical practice. Myriad other instruments to measure different aspects of symptomatology, correlated cognitive constructs, and comorbidity have been described in the research literature and in specialized clinical literatures.

It may also be helpful to use validated quantitative instruments to measure psychosocial functioning, quality of life, and other aspects of the impact of symptomatology on broader functioning. Such measures are not described in detail here but are available from other sources.

Definition of key terms

Clinician-rated instrument.

A semi-structured clinician administered interview to assess the presence and/or severity of symptoms of OCD or other aspects of symptomatology or function

Self-report measures.

Patient reports of varying constructs, usually using a visual analog scale. Self-report measures are generally more convenient than clinician-rated instruments but may be less valid for aspects of symptomatology, like insight, in which patient report may be insensitive.

Patient-proxy report measures.

This approach involves a significant other reporting on aspects of a patient’s symptom presentation or function – for example, OCD symptomatology or family accommodation.

Strength of evidence

Clinician and patient report measures have been validated across numerous studies. In some instances, validation by separate research groups is needed. Translation and validation of such measures is required in many languages, especially for key measures (e.g., Y-BOCS). Convergence of self-report or clinician administered ratings with more objective measures of symptomatology, disability, and function remains an area of research.

Suggestions for future research

Further work is needed examining more objective markers of OCD symptom severity and related distress that move beyond clinician observations and self-report scales. Examples might include harnessing computer vision technology and quantitative assessments of symptom provocation tasks.

The ability to use and interpret:

1. The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) to assess symptom severity in OCD • The ability to interpret Y-BOCS scores as indicative of mild (<16), moderate (16–24), severe (24–32), or extreme (32–40) • The ability to interpret symptom domains that are not well captured by the Y-BOCS, such as avoidance   Evidence: TG: Koran et al 2007 • OR: McGuire et al 2017 • PS: Goodman et al 1989a,b • TM: Phillips and Stein 2014

2. The Y-BOCS-2, which better assesses avoidance and better discriminates severity in the severe and extreme ranges   Evidence: OR: McGuire et al 2017 • PS: Storch et al 2017

3. The Dimensional Obsessive-Compulsive Scale (DOCS), a self-report measure that discriminates validated symptom dimensions of OCD   Evidence: OR: McGuire et al 2017 • PS: Abramowitz et al 2010

4. The Obsessive-Compulsive Inventory-Revised (OCI-R)   Evidence: OR: McGuire et al 2017 • PS: Foa et al 2002

5. Measures of comorbid symptomatology in patients with OCD, such as: • Beck Depression Inventory • Beck Anxiety Inventory • PSD-9   Evidence: TG: Koran et al 2007

4.a.vi. Use of medical and psychological assessments to inform diagnosis and treatment.

Psychiatry in general lacks objective biological and psychological measures that can reliably inform diagnosis or treatment selection. This is true in the case of OCD. While effects of OCD diagnosis on neuroimaging measures, neuropsychological tests, and other measures have been described in the research literature, in general these effect sizes are too small to be useful in the clinical setting. Sensitivity, specificity, and other critical scale metrics have not been established for the use of any imaging or laboratory tests in diagnosis. Laboratory, neuroimaging, neuropsychological, and other tests may be useful to clarify comorbidity and to rule out possible medical or neurological etiologies for observed symptomatology, but not to diagnose OCD itself. Laboratory tests may be helpful to monitor for potential drug side effects (e.g. periodically monitoring clomipramine levels), but they are not generally contributory to the assessment of symptom change or treatment efficacy; interview by a trained clinician and/or use of validated clinician-administered or self-report measures of symptomatology remains the gold standard for these purposes.

Definition of key terms

Laboratory tests –

quantitative or qualitative assessments of the composition of body fluids, typically blood or urine, to assess abnormalities in health.

Urine toxicology –

a laboratory test of the urine to establish the presence of a range of drugs of abuse

Electrocardiogram (EKG) –

a test of electrical events during the beating of the heart, which can be used to identify a range of cardiac pathology

Neuroimaging –

Noninvasive techniques used to examine the structural integrity of the brain and skull. CT scanning and structural MRI are the most commonly used in the assessment of neurological disease (and to rule out neurological disease in psychiatric patients when clinical presentation suggests that this is necessary). Neuroimaging may be used to guide anatomically targeted treatments such as TMS in the future.

Electroencephalogram (EEG) –

a test of electrical events in the brain, recorded from electrodes on the scalp, that can be used to diagnose and characterize seizure disorders. EEG may be used to guide anatomically targeted treatments in the future.

Neuropsychological testing –

A battery of validated quantitative tests of a number of domains of mental function, such as intelligence (IQ) and attention.

Strength of evidence

Many studies in the research literature have sought to characterize diagnostic or prognostic biomarkers in OCD. As in other neuropsychiatric conditions, measures that provide sufficient information to be useful in treatment planning of individual patients with uncomplicated OCD have yet to emerge. The use of diagnostic tests to evaluate neurological comorbidity or other complicating factors is generally dictated by the standards of practice in the potential complicating condition being investigated or ruled out.

Suggestions for future research

The search for diagnostic or prognostic biomarkers is ongoing. The use of multivariate strategies, where many different streams of data are used to identify patterns that may better discriminate diagnoses or correlate with outcome than any one measure does in isolation, holds promise. The search for diagnostic biomarkers is complicated by the phenomenological nature of our diagnoses, which may not correspond to true natural types; the search for clinically useful biomarkers and tests is therefore unavoidably bound up with efforts to better characterize OCD as a diagnosis, and its relationship to other symptom patterns and diagnostic entities.

The ability to:

1. Appropriately order and interpret laboratory tests in the assessment of patients with OCD, understanding that such tests do not inform diagnosis but may help in the evaluation of comorbidity and possible medical factors that may influence treatment • Assessment of general medical health • Assessment of laboratory abnormalities that may explain symptoms of anxiety and depression (e.g. thyroid abnormalities; other hormonal abnormalities; iron deficiency; vitamin D deficiency) • Assessment of laboratory abnormalities that may inform choice of pharmacotherapy (e.g. liver and kidney health) • Use of urine toxicology to inform assessment of substance use disorders   Evidence: TG: Koran et al 2007, American Psychiatric Association 2016

2. Use electrocardiography (EKG) at baseline and at intervals subsequently to guide choice of pharmacotherapy, as appropriate • Clomipramine • Citalopram • Escitalopram • Certain neuroleptics   Evidence: MA: Beach et al. 2014 • TG: Koran et al 2007, Koran and Simpson 2013 • EO: US Food and Drug Administration 2012

3. Use imaging methodologies during diagnosis, recognizing that they do not typically inform the diagnosis of OCD but may occasionally be useful to evaluate or rule out comorbidities or medical causes • Unusually acute onset of symptoms • Presence of focal motor signs • Presence of cognitive decline, disorientation, or other signs or symptoms suggesting diffuse brain dysfunction • Onset after age 40 • Other signs or symptoms or a family history suggestive of an alternative neurological diagnosis   Evidence: TG: American Psychiatric Association 2016

4. Refer for EEG testing if seizure is suspected   Evidence: TG: American Psychiatric Association 2016

5. Refer for neuropsychological testing in assessment as appropriate, recognizing that this is not typically needed for diagnosis of OCD but may be useful for assessment or clarification of comorbidity in occasional cases • Functional disability out of proportion to symptoms • Unclear neurodevelopmental history • Unclear diagnostic picture   Evidence: TG: American Psychiatric Association 2016

6. Seek out appropriate specialty consultation when needed in complex cases or in the presence of medical, neurological, or psychiatric comorbidity requiring specialty assessment or care   Evidence: TG: American Psychiatric Association 2016

4.b. Pharmacological Treatment Capacities

4.b.i. Treatment planning

As further reviewed below, there are a number of evidence-based initial and secondary treatment options for OCD (psychotherapeutic, pharmacotherapeutic, and – increasingly – anatomically targeted). There is also a growing array of tertiary treatments for which the evidence base is weak, but which may be considered after nonresponse to better-established strategies. Establishing a staged treatment plan for an individual patient is a complex challenge. Treatment planning must account for the evidence supporting various treatment options, but also patient-specific factors such as individual preference, availability of various treatments (both geographically and logistically), financial and social support, and comorbidity. Coordination and effective collaboration with other providers and support systems (therapist, social work, family members, educational system) is an essential aspect of treatment planning.

Definition of key terms

Key terms used in this section are defined elsewhere in this document.

Strength of evidence

Most studies of OCD treatments have focused on a single treatment modality, or at best have compared two interventions. Most trials are explanatory rather than pragmatic in nature – that is, they are designed to answer a specific question in isolation, rather than to provide guidance to decision making in the real world. No comprehensive studies provide unambiguous treatment algorithms applicable to most or all patients. Treatment planning is thus generally guided by extrapolation from individual studies, expert consensus, and clinical experience.

Suggestions for future research

Large pragmatic studies examining the differential efficacy of different treatment approaches to treatment and predictors of treatment response that can be used to stratify patients and inform treatment planning (i.e. personalized medicine) are urgently needed.

Ability to:

1. Select initial treatment modality (pharmacotherapy, psychotherapy, or combination)   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • OR: Fineberg et al 2016 • RCT: Foa et al 2005, Simpson et al 2008

Evaluation of patient attitudes towards treatment   Evidence: OCT: Reid et al 2017, Wheaton et al 2016

Evaluation of symptom type and severity   Evidence: OCT: Keijers et al 1994, de Haan 1997, Mataix-Cols et al 2002

Evaluation of complicating clinical factors that may influence treatment selection, such as: • insight • anxiety/distress tolerance • suicidality • comorbidity   Evidence: RCT: Hohagen et al 1998 • OCT: Abramowitz and Foa 2000, Steketee et al 2001

Evaluation of practical considerations that may influence treatment selection, such as: • insurance/financial considerations • availability of skilled therapists   Evidence: RCT: Hohagen et al 1998 • OCT: Abramowitz and Foa 2000, Steketee et al 2001

2. Work collaboratively with other clinicians to pursue a comprehensive treatment strategy   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • OR: Fineberg et al 2016

3. Provide psychoeducation • Symptoms and heterogeneity of OCD • Psychological and neurobiological understandings of illness; cognitive distortions • Treatment options • Comorbidity • Reasonable expectations of improvement (degree and timeframe) • Barriers to treatment • Chronicity of illness; relapse risk • Morbid risk in family members   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • OR: Fineberg et al 2016, Abramowitz 2017, Lebowitz 2017

• Family accommodation   Evidence: OR: Lebowitz 2017

4. Ability to use a range of strategies to engage patients in treatment and overcome barriers • Gauging pace of treatment • Motivational interviewing   Evidence: TG: Koran et al 2007

4.b.ii. Initial pharmacological treatment

The initial pharmacological treatment of OCD is straightforward, as only one class of medications – the selective serotonergic antidepressants – is supported for first-line treatment by robust evidence. The evidence for clomipramine is equally robust, and it may even be slightly more effective than SSRIs, but its higher potential side effect burden means that it is reserved for second-line treatment in all major treatment guidelines.

Definition of key terms

Selective serotonin reuptake inhibitors (SSRIs) are defined and discussed above, in sections 3.d.i and 3.d.ii.

Strength of evidence

The evidence base for SSRI pharmacotherapy in OCD, which is robust, is assessed above in section 3.d.i.

Suggestions for future research

Alternative monotherapies for OCD are greatly needed for patients who do not respond to or cannot tolerate SSRIs.

Ability to:

1. Make an informed choice of SSRI vs clomipramine, recognizing that clomipramine may be more effective but also has a higher potential side effect burden, and that other antidepressants have not been clearly shown to be efficacious as monotherapy in OCD   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • MA: Issari et al 2016, Soomro et al 2008, Soomro 2012, Skapinakis et al 2016a • OR: Fineberg et al 2016

2. Make an informed choice of initial SSRI, recognizing that they are of similar efficacy and that potential side effects, past treatment, patient preference, and similar considerations therefore inform choice • Pharmacokinetic considerations • Drug-drug interactions, including interactions with liver CYP enzymes • Comorbidity (both psychiatric and medical) may inform choice of treatment, though all SSRIs are similarly efficacious for common mood and anxiety comorbidities   Evidence: TG: Koran et al 2007, Koran and Simpson 2013, National Center for Clinical Excellence 2006, Bandelow et al 2012, Baldwin et al 2014 • MA: Issari et al 2016, Soomro et al 2008, Soomro 2012, Skapinakis et al 2016a • OR: Fineberg et al 2015, 2016

• Pharmacogenetic testing is not standard practice and cannot predict whether a particular medication will be effective   Evidence: EO: Zai et al 2014, Bousman et al 2016, 2017

3. Appropriately dose initial SSRI, recognizing that higher doses are more efficacious in OCD, if tolerated • Possible emergence of side effects at higher SSRI doses • Concern of QTc prolongation at high doses of SSRI; FDA black-box warning for citalopram • Titration to reach target dose while minimizing side effects and monitoring for the possibility of early improvement   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • MA: Bloch et al 2010 • Issari et al 2016, Soomro et al 2008, Soomro 2012, Skapinakis et al 2016a • OR: Fineberg et al 2016

4. Complete an adequate clinical trial, recognizing that long duration of treatment is typically needed in OCD and that most clinical trials have been 12 weeks in length   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • MA: Issari et al 2016, Skapinakis et al 2016a • OR: Fineberg et al 2016

5. Appropriately dose and monitor clomipramine monotherapy • Dose and duration of treatment • Monitoring of side effects, including EKG monitoring for QTc prolongation at high doses • Monitoring of trough serum levels of clomipramine and desmethylclomipramine   Evidence: TG: Koran et al 2007 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2016 • CES: Marazziti et al 2012

6. Monitor response and any side effects over the course of initial monotherapy with SSRI or clomipramine • Assessment of any side effects • Monitoring of symptoms, using clinical interview and structured instruments Assessment of adherence; implementation of strategies to enhance adherence • Distinguishing improvement in core OCD symptoms from improvement in comorbid symptoms • Assessment of improvement in quality of life and functional status, which may not always correlate with change in symptoms   Evidence: TG: National Institute for Clinical Excellence 2006, Koran et al 2007, Bandelow et al 2012, Koran and Simpson 2013, Baldwin et al 2014 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2015, 2016 • CES: Marazziti et al 2012

7. Manage medication side effects, using secondary medication, behavioral measures, modulation of medication dose or titration schedule, and simply the passage of time • Management of anxiety, especially around SSRI dose increases • Management of insomnia, gastrointestinal side effects, fatigue, sexual side effects, akathisia, other side effects   Evidence: TG: National Institute for Clinical Excellence 2006, Koran et al 2007, Koran and Simpson 2013 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2015, 2016, Hirschtritt et al 2017

8. Decide when to continue or discontinue pharmacotherapy through consideration of projected risks/benefits   Evidence: OR: Fineberg et al 2016, Hirschtritt et al 2017

9. Manage medication discontinuation through appropriate taper, taking into account pharmacokinetics and side effects, in collaboration with therapist and other treatment clinicians   Evidence: TG: National Institute for Clinical Excellence 2006, Koran et al 2007, Koran and Simpson 2013 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2015, 2016

4.b.iii. Management of initial nonresponse to pharmacotherapy through change of monotherapy or addition of an augmentation strategy

Unfortunately, a substantial fraction of individuals with OCD – up to 40% – do not adequately respond to SSRI monotherapy, even when it is optimized. Similarly, a significant percentage (which varies across studies) similarly do not respond to standard CBT. It is therefore often necessary to explore a range of pharmacological and non-pharmacological augmentation strategies; our focus here is on the former. A few strategies are supported by robust evidence and are reviewed in this section, as well as in sections 3.d.iii and 3.d.iv, above. Other strategies are supported by more limited evidence and remain a focus of active research; these are addressed in a subsequent section.

Definition of key terms

Key terms for this section are defined above in sections 3.d.iii and 3.d.iv.

Strength of evidence

The strength of evidence for these treatment approaches is assessed above, in sections 3.d.iii and 3.d.iv.

Suggestions for future research

The well-supported options for pharmacological augmentation remain very limited. Firmly establishing the efficacy of other strategies in OCD resistant to first-line interventions is an important goal, so that clinicians will have more options when initial treatment does not produce an adequate response.

Ability to:

1. Evaluate when response to initial pharmacotherapy, of sufficient dose and duration, is inadequate, or initial pharmacotherapy is not well tolerated because of side effects, and further increasing dose, changing to an alternate agent or adding an augmentation strategy is appropriate   Evidence: TG: Koran et al 2007

2. Select and implement a second-line treatment strategy after inadequate response to initial pharmacotherapy, taking into account response to initial treatment, available evidence, and patient preferences. • Switch to specialized CBT • Augment with specialized CBT • Increase dose of SSRI • Switch to a second SSRI • Switch from SSRI to clomipramine • Augment with secondary pharmacotherapy (e.g. neuroleptic) • Augment with a second serotonergic agent (e.g. clomipramine + SSRI)   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2016, Simpson 2017 • RCT: Foa et al 2005, Simpson et al 2008, Simpson et al 2013 • OCT: Wheaton et al 2016 Management of SSRI and clomipramine monotherapy is discussed in the section above. Specialized CBT is discussed in companion papers in this series (Sookman et al, 2021b; Piacentini et al, under review).

3. Assess the appropriateness of and implement neuroleptic augmentation of SSRI or clomipramine, recognizing that this is the pharmacological augmentation strategy best supported by available evidence   Evidence: TG: Koran et al 2007, Koran and Simpson 2013 • MA: Skapinakis et al 2016a • OR: Fineberg et al 2016, Simpson 2017

• Efficacy of neuroleptic augmentation, recognizing that data are strongest for risperidone, aripiprazole   Evidence: MA: Bloch et al 2006, Ipser et al 2006, Veale et al 2014, Dold et al 2015

• Choice of agent, recognizing that higher potency D2 antagonists may be indicated in patients with tics or a history of tics   Evidence: MA: Bloch et al 2006, Ducasse et al 2014, Veale et al 2014

• Dosing of neuroleptic augmentation, recognizing that low doses are typically indicated • Assessment of response, recognizing that response is typically more rapid than with the SSRIs (2–4 weeks) • Assessment of any side effects of neuroleptic treatment   Evidence: TG: Koran et al 2007

• Discontinuation of neuroleptic treatment when ineffective or limited by side effects   Evidence: OR: Maina et al 2003

4. Assess the appropriateness and implement the combination of an SSRI with clomipramine, acknowledging that the evidence base behind this strategy is very limited.   Evidence: TG: Koran et al 2007 • OCT: Ravizza et al 1996, Figueroa et al 1998, Pallanti et al 1999, Diniz et al 2010

4.b.iv. Use of agents for which the evidence base is less clear for refractory disease when the better-proven agents have been exhausted

Unfortunately, a substantial minority of patients continue to suffer significant symptomatology even after appropriate trials of available evidence-based treatments (both psychotherapeutic and pharmacological; both first- and second-line). In these cases, medication combinations and augmentation strategies that are supported by only limited evidence must be considered, always carefully balancing risks against the potential benefits, and acknowledging that the potential benefits of strategies supported by limited evidence must be considered provisional. This section, together with section 3.d.v above, reviews strategies described in the literature that are sometimes used in specialty practice, as well as strategies that have been investigated but found to be ineffective.

Definition of key terms

Relevant terms are defined above, in section 3.d.v.

Strength of evidence

The evidence for these strategies is assessed above, in section 3.d.v.

Suggestions for future research

All pharmacological treatment strategies described in this section have been investigated in some published studies but are not sufficiently well supported to be considered standard-of-care first or second-line treatments. Most investigations of these agents and strategies are small, single-site studies from which generalization is difficult. Further research better establishing the efficacy of these various treatment approaches, and identifying those that are lacking efficacy, is needed.

The ability to:

1. Identify refractory disease • Failure to respond to at least two trials of an SSRI, of adequate dose and duration, with an improvement of ≥35% in YBOCS • Failure to respond to clomipramine, neuroleptic augmentation • Failure to respond to expert outpatient CBT   Evidence: OR: Albert et al 2013, Farris et al 2013 • EO: Mataix-Cols et al 2016

2. Evaluate the appropriateness of, and make an appropriate referral for, intensive specialized multimodal treatment, including CBT (intensive outpatient treatment; residential)   Evidence: MA: Veale et al 2016 • CR: Stewart et al 2005, Brennan et al 2014

3. Evaluate the appropriateness of, and implement, pharmacotherapy with serotonergic agents other than the SSRIs, recognizing that the evidence of their benefit is limited   Evidence: TG: Koran et al 2007 • OR: Pittenger and Bloch 2014, Pittenger 2017b • See also section 3.d.v, above

• Controlled trials of augmentation with buspirone have not shown benefit. Some experts nevertheless advocate high-dose buspirone augmentation in refractory patients.   Evidence: TG Koran et al 2007 • RCT: Pigott et al 1992, McDougle et al 1993 • OCT: Jenike and Baer 1988

• Augmentation with mirtazapine has been shown to accelerate SSRI response in a small study, but not to improve ultimate response rate or degree of improvement   Evidence: TG: Koran et al 2007 • RCT: Pallanti et al 2004

• Newer serotonergic agents (vilazodone, vortioxetine, agomelatine) have not been systematically investigated in the treatment of OCD   Evidence: OR: Pizarro et al 2014

• Intravenous administration of high-dose SSRIs (chiefly citalopram) has been investigated in OCD but remains an investigational strategy at specialty centers   Evidence: TG: Koran et al 2007

4. Evaluate the appropriateness of, and implement, monotherapy with serotonin-norepinephrine dual reuptake inhibitors (venlafaxine, duloxetine), recognizing that the evidence for benefit is limited and mixed   Evidence: TG: Koran et al 2007 • OR: Pittenger and Bloch 2014, Pittenger 2017b • See also section 3.d.v, above

• Venlafaxine has been investigated in several controlled trials, with mixed results   Evidence: TG: Koran et al 2007 • RCT: Yaryura-Tobias et al 1996, Denys et al 2003

• Duloxetine has been examined in case reports; there are no controlled studies   Evidence: TG: Koran et al 2007 • CR: Dell’osso et al 2008, Luis Blay and Black 2007, Yeh et al 2009

5. Evaluate the appropriateness of, and to implement, augmentation with glutamate modulators in refractory OCD, recognizing that the evidence for benefit from these agents remains limited   Evidence: OR: Pittenger 2015, 2017b • See also section 3.d.v, above

• Memantine augmentation of SSRI treatment is supported by small controlled studies   Evidence: RCT: Ghaleiha et al 2013, Haghighi et al 2013 • OCT: Aboujaoude et al 2009, Feusner et al 2009, Stewart et al 2010, Bakhla et al 2013 • CR: Poyurovsky et al 2005, Pasquini et al 2006

• Riluzole augmentation of SSRI treatment is supported by small controlled studies in adults but not in children. Monitoring of liver function tests is advised   Evidence: RCT: Pittenger et al 2015, Emamzadehfard et al 2016 • CR: Coric et al 2005, Pittenger et al 2008

• Small studies suggest that single-dose ketamine infusion may be of benefit for comorbid depression, and perhaps in uncomplicated OCD   Evidence: RCT: Rodriguez et al 2013 • OCT: Bloch et al 2012

• Small studies suggest the possibility of benefit from N-acetylcysteine (NAC); evidence for benefit is clearer in other trichotillomania and possibly excoriation (skin-picking) disorder   Evidence: RCT: Afshar et al 2012, Sarris et al 2015, Paydary et al 2016, Costa et al 2017 • CR: Lafleur et al 2006

• Case reports and small studies suggest benefit from the glutamate-modulating anticonvulsants topiramate and lamotrigine and from the amino acid glycine   Evidence: RCT: Mowla et al 2010, Berlin et al 2011, Bruno et al 2012, Khalkhali et al 2016 • OCT: Greenberg et al 2009 • CR: Rubio et al 2006, Van Ameringen et al 2006, 2015

• D-cycloserine, which modulates glutamate neurotransmission, has been investigated as a CBT augmenter, but not for stand-alone pharmacological treatment   Evidence: OR: Pittenger 2015, 2017b • See also Sookman et al 2021b

6. Evaluate the appropriateness of, and to implement, augmentation with other monoaminergic agents in refractory OCD, recognizing that the evidence for benefit from these agents remains limited   Evidence: TG: Koran et al 2007 • OR: Pittenger and Bloch 2014, Pittenger 2017b • See also section 3.d.v, above

• Pindolol augmentation of SSRIs has been tested in two small trials, with conflicting results   Evidence: RCT: Mundo et al 1998, Dannon et al 2000

• Tryptophan augmentation of SSRIs has been tested in open-label studies; serotonin syndrome is a theoretical risk, and impurities in the tryptophan supply led to adverse events in the 1980s and 1990s   Evidence: CR: Blier and Bergeron 1996, Yaryura-Tobis and Bhagavan 1977

• Opiate agonists (morphine, tramadol, buprenorphine) have been investigated in small studies, with promising results; the risk of dependence and addiction complicates their use   Evidence: RCT: Koran et al 2005 • OCT: Shapira et al 1997 • CR: Warneke et al 1997

7. Recognize pharmacological augmentation strategies that good studies suggest do not work • Lithium • Neuroleptic monotherapy • Non-serotonergic tricyclic antidepressants • Bupropion   Evidence: TG: Koran et al 2007 • See also section 3.d.v, above

8. Rationally select from the augmentation strategies for which evidence is limited, once better-proven approaches have been exhausted • Considerations of side effects and tolerability are weighed more heavily when confidence in the efficacy of tertiary strategies is lower; well-tolerated strategies are preferred • Use of less well proven strategies should be stratified on the basis of the strength of what evidence exists, as summarized above • Patient preference, personal or family history of benefit, and other situational factors are weighed more heavily when there is not clear evidence to guide a specific choice   Evidence: TG: Koran et al 2007 • OR: Simpson 2017 • See also section 3.d.v, above

4.b.v. Management of refractory OCD through anatomically targeted treatment approaches

As reviewed above (Section 3.b.2), abnormal activity in a defined neural circuitry is associated with OCD symptoms, and improvement with treatment has been associated with normalization of the elevated activity in this circuitry. OCD is thus ripe for the development of anatomically targeted treatment strategies that seek to disrupt this pathological activity. Indeed, such strategies have been used in the most refractory cases since the 1950s. This section, together with section 3.e above, reviews the evidence for benefit from targeted neurosurgical lesions, deep brain stimulation (DBS), and transcranial magnetic stimulation (TMS). Neurosurgical approaches are in general appropriate only when OCD remains crippling despite robust application of all less invasive interventions. TMS is not invasive, and the selection of appropriate patients can therefore be somewhat more flexible; the evidence base for newer TMS approaches in OCD is limited but promising, and the use of TMS for OCD may grow if supportive evidence continues to accrue. These approaches all require costly hardware and subspecialist expertise and are available only in specialized centers. The specialist OCD clinician in general practice should understand them well enough to evaluate when making a referral to such a center is appropriate.

Definition of key terms

Key terms used in his section are defined above, in section 3.e.

Strength of evidence

The evidence for these treatment approaches is evaluated above, in section 3.e.

Suggestions for future research

For TMS, early evidence for efficacy of ‘deep TMS’ is promising; replication is needed, and other technical approaches to stimulate deep structures using this noninvasive method need to be developed. For DBS and neurosurgical lesion approaches, studies are small and blinding is difficult; while the available evidence is promising, ongoing monitoring of all cases is needed to grow the evidence base and thereby guide future practice.

The ability to evaluate the appropriateness of and make a referral for:

1. ECT treatment, recognizing that available evidence does not support efficacy of ECT for OCD, but it may be appropriate for the treatment of very severe comorbid depression   Evidence: SR: Fontenelle et al 2015

2. Transcranial magnetic stimulation (TMS)   See also section 3.e, above

• TMS targeting the dorsolateral PFC, which is indicated for depression, is of limited benefit for OCD   Evidence: MA: Trevizol et al 2016 • SR: Rapinesi et al 2019

• TMS targeting the dorsomedial PFC, including the supplementary area (SMA), or the orbitofrontal cortex has been shown to be of statistically significant but clinically modest benefit in OCD   Evidence: MA: Trevizol et al 2016 • SR: Rapinesi et al 2019

• TMS targeting the frontal pole may be of benefit   Evidence: OCT: Williams et al 2021

• ‘Deep TMS’, which uses a custom coil to target the medial prefrontal cortex and anterior cingulate cortex, has recently shown benefit in OCD   Evidence: RCT: Carmi et al 2019

3. Deep brain stimulation (DBS) in profoundly refractory OCD   Evidence: SR: Rapinesi et al 2019 • OR: Goodman et al 2017 • See also section 3.e, above

4. Ablative neurosurgery for profoundly refractory OCD   Evidence: SR: Rapinesi et al 2019 • OR: Keen et al 2017 • See also section 3.e, above

4.b.vi. Management of high-risk or complicated patients

As discussed above, comorbidity is the norm in OCD, and each patient is unique. Psychiatric, neurological, medical, and psychosocial complexities can combine to make the management of certain patients particularly challenging. This section reviews some of the more common complexities encountered in clinical practice.

Definition of key terms

Key terms in this section have been described previously in other sections in this document.

Strength of evidence

As patient populations become more and more specific (e.g. characterized by complex patterns of comorbidity or specific psychosocial situations), studies become fewer and smaller, if they are available at all. Treatment in complex cases is typically guided by extrapolation from studies of less complicated populations and/or by clinical experience. In most situations described in this section, therefore, evidence is at the level of case series and expert opinion.

Suggestions for future research

More research is needed regarding the treatment of OCD in complicated clinical contexts. Some of these contexts (e.g. pregnancy, geriatric) are sufficiently common and predictable that careful observational and treatment studies can be designed; more such studies are needed to guide clinical practice. Advancing knowledge of the pathophysiology of OCD will provide new insights into how it interacts with other pathologies, shedding new light on how to rationally manage complex clinical scenarios even when definitive studies are not available.

The ability to:

1. Manage pharmacotherapy for OCD during pregnancy, or to make an appropriate referral for such management

• CBT may be the preferred treatment approach during pregnancy as it carries no risk for the fetus   Evidence: TG: Koran et al 2007 • OR: Guglielmi et al 2017

• Discontinuation of pharmacotherapy carries risk of psychiatric destabilization; risks to the fetus of continuing an SSRI, which are small, must be weighed against risks of discontinuing pharmacotherapy   Evidence: OR: Guglielmi et al 2017

• OCD symptoms can arise or worsen in the peripartum period   Evidence: OR: McGuinness et al 2011, Guglielmi et al 2017

2. Assess and respond clinically to suicidality when it arises in patients with OCD • Suicide risk is elevated by comorbidity, especially with affective and substance use disorders • Suicide risk is elevated several-fold in OCD even after controlling for comorbidity • Importantly, some individuals with OCD have ego-dystonic, intrusive suicidal/self-harm obsessions without suicidal intent. Evaluating risk in this setting is particularly challenging; some patients with suicidal/self-harm obsessions may also have suicidal thoughts associated with depression.   Evidence: TG: Koran et al 2007 • MA: Angelakis et al 2015 • SR: Albert et al 2019 • OR: Veale et al 2009 • CSS: Fernandez de la Cruz et al 2016 • See also section 3.a.ii, above

3. Alter treatment of OCD appropriately in the presence of medical comorbidity and to collaborate with medical providers to provide coordinated care • Medical conditions (e.g. liver and kidney disease) may alter metabolism of OCD medications and thus affect pharmacokinetics • Cardiac conditions may increase the risk of arrhythmia with certain agents, especially citalopram and clomipramine • Neuroleptics can predispose patients to weight gain and the development of dyslipidemia and diabetes • Alterations of hepatic metabolism of other drugs can change their pharmacokinetics; for example, fluoxetine extends the half-life of warfarin and can thus disrupt anticoagulant treatment and predispose to bleeding • SSRIs can interfere with platelet function and thereby predispose to bruising • Cerebrovascular disease can alter the response to any psychoactive drug, change sensitivity to medication effects and side effects, and predispose to delirium • Clomipramine can lower the seizure threshold and should be avoided or used with caution in patients with epilepsy or with other predisposing factors   Evidence: TG: Koran et al 2007

4. Adjust treatment of OCD symptoms in the elderly, and to seek medical or geriatric psychiatric input as appropriate for comprehensive coordination of care, recognizing that the management of OCD symptoms in the elderly has not been well studied   Evidence: TG: Koran et al 2007

• New onset of OCD in the elderly is unusual and should generally trigger workup for a neurological cause. However, worsening of OCD symptoms in the elderly is not uncommon   Evidence: CSS: Ruscio et al 2010

• Drug metabolism is typically slowed in the elderly, so lower doses of medication may be appropriate • Drug-drug interactions and development of delirium are more frequent in the elderly   Evidence: OR: Lotrich and Pollock 2005, Pollock 2005

• The FDA black box warning limits the use of citalopram to a maximum dose of 20 mg/day in the elderly to limit risk of arrhythmia, and so other SSRIs are generally preferred in this age group   Evidence: EO: US Food and Drug Administration 2012

5. Adjust treatment of OCD symptoms in the context of comorbid autism, recognizing that optimal treatment in this population has not been well studied • Obsessions can be difficult to distinguish from restricted interests or overvalued ideas • Compulsions can be difficult to distinguish from stereotypies • Individuals with autism may have difficulty engaging in ERP • The literature on optimal pharmacotherapy of comorbid OCD and autism is sparse   Evidence: TG: Koran et al 2007 • OR: Accordino et al 2017 • CSS: Wikramanayake et al 2017 • See above, sections 3.a.ii and 3.a.iii

6. Take comorbid personality disorder into account in treating OCD • OCPD is common in OCD; OCPD symptoms may respond less well to SRI pharmacotherapy than OCD and may affect motivation for treatment • Schizotypy is uncommon in OCD but can complicate treatment • Borderline personality is moderately common in OCD and can complicate treatment Evidence: TG: Koran et al 2007 • OR: Diedrich and Voderholzer 2015, Starcevick and Brakoulias 2017, Wheaton and Pinto 2017 • See above, sections 3.a.ii and 3.a.iii

7. Treat OCD in the presence of comorbid bipolar disorder   Evidence: TG: Koran et al 2007 • OR: Mucci et al 2018

• SSRIS and clomipramine may be activating in patients with bipolar disorder; the risk of activation may be greater with clomipramine   Evidence: CSS: Perugi et al 2002

• Polypharmacy (e.g. SSRI + lithium or another mood stabilizer) may be necessary in these patients   Evidence: TG: American Psychiatric Association 2002 • OR: Mucci et al 2018

• Cycling or episodic course of OCD symptoms should increase suspicion for comorbid bipolar disorder   Evidence: CSS: Perugi et al 2002

8. Treat OCD symptoms in the presence of comorbid psychosis   Evidence: TG: Koran et al 2007 • OR: Poyurovsky 2017 • See above, sections 3.a.ii and 3.a.iii

• The prevalence of obsessions and compulsions is increased in individuals with schizophrenia   Evidence: OR: Bottas et al 2005, Schirmbeck et al 2015, Poyurovsky et al 2017 • CSS: Porto et al 1997

• Second-generation antipsychotic drugs, especially clozapine, can trigger or exacerbate obsessions and compulsions, especially when used without an SRI   Evidence: OR: Poyurovsky and Koran 2005, Lykouras et al 2003 • CR: Ghaemi et al 1995

• SSRIs are generally well tolerated in schizophrenia   Evidence: OR: Poyurovsky et al 2004

5. Discussion and conclusions.

OCD affects approximately one person in 40 over the course of a lifetime (Ruscio et al. 2010) and causes enormous suffering. Evidence-based treatments, both pharmacological and psychotherapeutic, can benefit many patients and are summarized in a number of treatment guidelines (NICE, 2006; Koran et al. 2007; Koran and Simpson, 2013; Baldelow et al. 2012). Unfortunately, they are often not delivered optimally, and thus millions who might respond to standard-of-care treatment continue to suffer significant symptoms. The promulgation of best practice standards is therefore a high priority. This paper, together with the accompanying papers from the International OCD Accreditation Task Force, seeks to establish standards for specialty care of OCD.

Our focus has been on the pharmacological treatment of adult OCD. However, it should be emphasized that psychotherapy, and in particular the CBT technique of exposure and response prevention (ERP), is also a cornerstone of treatment of both adult and pediatric OCD and must be considered in every treatment plan. Indeed, in uncomplicated OCD, CBT including ERP may be preferable to pharmacotherapy, for at least two reasons. First, while ERP is difficult (for both patient and therapist) to do well, it produces none of the potential side effects or drug interactions that can sometimes complicate pharmacological treatment. Second, because adaptive strategies can be internalized and continue to have benefit beyond the period of active participation in ERP treatment, relapse rates after a successful course of psychotherapy are typically lower than after discontinuation of effective pharmacotherapy. Specialized CBT for OCD typically includes cognitive approaches and behavioral experiments to specifically address complexities in individual presentations of OCD and comorbid symptomatology.

Pharmacotherapy nevertheless has a very important role, both as a first-line treatment and in complicated or refractory disease. High-quality psychotherapy is difficult to access, which is a frequent impediment to optimal care. Some patients may initially feel unwilling or unable to do the work that CBT requires, and some have symptomatology that is difficult to address in ERP; in these circumstances, pharmacotherapy may be a preferable first-line treatment. And in severe OCD, or in the presence of severe comorbidity or other complicating factors such as suicidality, pharmacotherapy is likely to remain an essential treatment component regardless of the availability of high-quality CBT.

Only a few first- and second-line pharmacological strategies that are supported by robust studies and can be considered the unambiguous standard of care. The only agents appropriate for first-line treatment are the SSRI antidepressants (fluoxetine, fluvoxamine, sertraline, paroxetine, citalopram, escitalopram), any of which can benefit 50–60% of patients. The older medication clomipramine is perhaps slightly more effective than the SSRIs, but is generally reserved for second-line treatment due to its more problematic potential side effects. Finally, augmentation of SSRI or clomipramine treatment with low-dose neuroleptic (especially with risperidone, haloperidol, and aripiprazole) can be of clear benefit in some patients. Anatomically targeted methods, especially deep TMS, have also been shown to benefit some patients and are likely to see increasing use.

After this small handful of well-proven pharmacological approaches has been exhausted, and optimal psychotherapy has been attempted or deemed impractical, the evidence to guide third-line interventions is much more equivocal, and specialist practitioners frequently find themselves balancing unclear benefit against side effects and current suffering. Other serotonergic agents (e.g. mirtazapine, buspirone), other monoaminergic agents (e.g. dual reuptake inhibitors), and glutamate modulators (e.g. memantine, riluzole) have all been investigated in SSRI-refractory disease, but none is supported by sufficiently robust evidence to make it the clear next choice in a treatment algorithm. Thus, treatments are typically selected on the basis of the strength of available evidence, patient-specific factors such as previous success (or failure) or comorbidity, and practitioner experience. In summarizing the limited evidence for this range of unproven treatments, we aim to provide a better basis for such decisions. It is to be hoped that ongoing research will convincingly confirm (or refute) the utility of these various strategies, so that a more extensive treatment algorithm, supported at every step by robust evidence, can be developed and promulgated.

The neurobiology of OCD is not well understood, and clear genetic signals as to its pathophysiology have yet to emerge. Nevertheless, we know a significant amount about the neural circuitry that is dysregulated in the disorder, and something about neurochemical and other aspects of its underlying neural and psychological substrates. This growing understanding may be increasingly important in the development and deployment of novel interventions in the future – indeed, transcranial magnetic stimulation, targeted neurosurgical lesions, and deep brain stimulation are all used in refractory disease today, and all seek to target the dysregulated brain circuitry. It is therefore important that specialist practitioners be conversant with the basics of current pathophysiological models, so that they can quickly adapt to new findings and adopt novel interventions as these are developed.

Treatment-resistant and treatment-refractory OCD are distressingly common, and much work is needed to develop new and more efficacious treatment strategies for those who do not achieve adequate benefit from the best evidence-based interventions available today. Nevertheless, currently available treatments do benefit many, and if we were simply able to provide all patients with the current standard of care, much suffering would be alleviated. We hope that the establishment of clear knowledge and competency standards will represent a valuable step towards this goal.

Table 1.

ATF EVIDENCE LEGEND

TG: Treatment Guideline	OCT: Open-Label Clinical Trial	CQS: Clinical Qualitative Study
MA: Meta-Analysis	CC: Case-Control Study	ThP: Theoretical Paper
SR: Systematic Review	CSS: Cross-Sectional Study	TM: Treatment Manual or Book
OR: Other Review	PS: Psychometric Study	TrPN: Non-Data Based Treatment Paper
RCT: Randomized Controlled Trial	CR: Case Report or Series	EO: Expert Opinion
RCS: Randomized Clinical Study	CES: Clinical Experimental Study	AR: Animal Research Study
CS: Cohort Study	AES: Analogue Experimental Study

Highlights.

• Obsessive-compulsive disorder (OCD) affects 1 person in 40 and produces substantial disability

• Evidence-based pharmacotherapy, especially with the SSRI antidepressants, can benefit many

• When SSRI pharmacotherapy is ineffective, pharmacological augmentation strategies are limited

• We present standards for the evidence-based specialty pharmacological treatment of OCD