Prepared by Michael B. First, M.D., DSM Consultant to the American Psychiatric Institute for Research and Education (APIRE), a subsidiary of the American Psychiatric Association
The sixth diagnosis-related research planning conference in the "Future of Psychiatric Diagnosis: Refining the Research Agenda" series focused on Obsessive-Compulsive Disorder and other disorders that may be part of an Obsessive-Compulsive Behavior Spectrum. Conference co-chairs were Eric Hollander, MD, Mount Sinai School of Medicine in New York City, and Joseph Zohar, MD, Chaim Sheba Medical Center in Tel-Hashomer, Israel. Twenty-three invited scientists from around the world participated in the APA/WHO/NIH-sponsored conference, which was held at APA headquarters in Arlington, Virginia on June 21 - 22, 2006.
In opening the conference, Drs. Hollander and Zohar noted that distributed throughout the various sections of the DSM are a number of disorders which, like Obsessive-Compulsive Disorder, are characterized by repetitive thoughts or behaviors. These include obsessive-compulsive personality disorder (OCPD), hoarding (not currently in DSM-IV as a separate disorder but one of the criteria for OCPD), Tourette's and other tic disorders, Sydenham's and other PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections), trichotillomania, body dysmorphic disorder, hypochondriasis, autism, eating disorders, Huntington's and Parkinson's diseases, impulse control disorder (e.g., pathological gambling), and substance addictions. Given potential commonalities in terms of brain circuitry, familial/genetic factors, neurotransmitter/peptide systems, phenotypes, and treatment response, it has been proposed that these disorders be grouped together under the rubric of Obsessive-Compulsive Spectrum Disorders (OCSD). Accordingly, the primary aim of the conference was to review the evidence that these disorders may be related to obsessive-compulsive disorder based on an examination of similarities and differences in terms of phenomenology, comorbidity, course of illness, and the additional domains listed above.. Other questions considered at the conference included: 1) Are there underlying endophenotypes that cut across obsessive-compulsive spectrum disorders and clarify their relationship with anxiety disorders? 2) What is the best way to diagnose OCD and the other spectrum disorders? (e.g., are symptom dimensions helpful in the diagnosis of OCD?) 3) What are the strengths and weaknesses of the current criteria? and 4) How should OCD that occurs in other psychiatric disorders be diagnosed, (e.g., is there a case to be made for including a "schizo-obsessive" subtype for schizophrenia?)
Naomi Fineberg MRCPsych (Hertfordshire, UK), focused on the definition of obsessive-compulsive disorder and on whether obsessive-compulsive personality disorder (OCPD) should be included within the OC spectrum disorders. The current definition of obsessive-compulsive disorder requires obsessions or compulsions. Obsessions are defined as repetitive thoughts that are non-rewarding, ego-alien, resisted by the person, recognized, at least initially, as excessive or unreasonable, and not due to a medical condition (ruling out PANDAS and other medical conditions that can cause OC symptoms). Factor analysis of OC symptoms indicates that they can be subtyped according to their content into contamination/cleaning, sexual obsessions, doubting/checking, need for symmetry/ordering, and hoarding. SRI-resistant OCD may constitute a subtype: in adults it is associated with longer duration, earlier onset, and co-morbid tics; in children it is characterized by comorbid tics, attention-deficit/hyperactivity disorder, and oppositional-defiant disorder. When individuals with OCD are compared to individuals with other anxiety disorders using cognitive testing, OCD patients tend to do worse on tests of executive functioning than do patients with panic and generalized anxiety disorder. In contrast, cognitive functioning in social phobia looks more like OCD. Dr. Fineberg called for collaborative expertise integrating diverse fields to agree on a battery of tests and to systematically compare homogeneous OCD subtypes with each other and with related disorders. She then examined the relationship of OCPD to OCD. OCPD accompanies a range of Axis I disorders, including OCD, eating, mood, and anxiety disorders. Although it occurs frequently with OCD (i.e., 25-30% of those with OCD also have OCPD) and in families of OCD probands, OCPD is more commonly associated with depression (i.e., 70% of persons with OCPD have depression), weakening the argument for it being a specific vulnerability factor for OCD. However, since OCPD and OCD are similar in terms of phenomenology, age of onset, course, heritability, comorbidity patterns and possibly serotonin-selectivity for treatment response, a shared etiology is plausible.
Sanjaya Saxena, MD (San Diego, CA) reviewed the relationship between compulsive hoarding and OCD. Compulsive hoarding is defined as the acquisition of, and failure to discard, a large number of possessions that appear to be useless or of limited value, leading to severe clutter that causes impairment or distress. On the basis of phenemonology, comorbidity patterns, family and genetic studies, and neuroimaging data, compulsive hoarding syndrome appears to be a discrete entity with a unique and characteristic pattern of core symptoms, associated features, and neurobiological abnormalities that differ from OCD. Dr. Saxena proposed that compulsive hoarding syndrome be listed as an OCSD in DSM-V with its own diagnostic criteria. Future research needs include epidemiological studies to establish prevalence and gender ratio, family and genetic studies of hoarders that do not exclude hoarders without other OCD symptoms, pathophysiological studies, and treatment studies of non-SRI medications.
Euripedes Miguel, MD, PhD (Sao Paulo, Brazil) examined the relationship between OCD and Tourette's syndrome (TS). Among patients with Tourette's syndrome, 60% have obsessions and/or compulsions and 30% have full-blown OCD. Among patients with OCD, 40% have had a history of a tic disorder. Patients with OCD and tics as opposed to patients with OCD without tics have several distinguishing features. For example, patients with both OCD and tics are more likely to have an earlier age of onset, hoarding behavior, repeating rituals, tic-like compulsions, symmetry/ordering, and aggressive obsessions than OCD counterparts without tics. Overall there is a phenomenological continuum of repetitive behaviors along the TS-OCD Spectrum, ranging from tics to compulsions. The subjective experiences that precede these behaviors help in understanding the continuum; while typical compulsions are preceded by obsessions, tic-like compulsions are performed to relieve sensory phenomena (e.g., bodily sensations, general feelings, urges, or a need to repeat the behavior until reaching an specific sensation or feeling "just-right"). Evidence supporting a close association between OCD and tics include similar clinical presentations; comorbidity patterns; course (e.g., tic in childhood predicts obsessive-compulsive symptoms in late adolescence and early adulthood); genetic data (e.g., rates of OCD are higher in first-degree relatives of probands with Tourette's and vice-versa when compared to first-degree relatives of control probands); and certain neuroimaging findings (e.g., involving cortico-striatum-thalamic-cortical circuits in both disorders and particularly the caudate nucleus). Evidence that argues against an association with OCD include neuropsychological data (e.g.,, impairment in executive functioning in OCD but not in tics) and pharmacological treatment involving different classes of drugs (e.g., SRI's for OCD and dopamine blockers and alpha-2 agonists for tics).
Dan J. Stein, MD, PhD (Cape Town, South Africa) presented on trichotillomania (TTM) and OCD. Although TTM is classified as an impulse control disorder, Dr. Stein argued that it may be better conceptualized as a stereotypic disorder. OCD and stereotypic disorders have some overlap (i.e., both have repetitive and ritualistic symptoms) but also have some differences (stereotypies are often driven by affect rather than by obsessions). From a neurobiological perspective, while both OCD and stereotypic disorders may involve striatal dopamine systems, there are likely differences in neurocircuitry, neurochemistry, and neuropsychology. Similarly, while dopamine blockers may play a role in the treatment of both conditions, SSRI's as well as Exposure and Response Prevention (ERP) are effective in OCD, while TTM responds to habit reversal. Stressors and negative affect appear to create a vulnerability for stereotypies, leading to a pattern of behavioral addiction in which the individual gives into reward-based drives without regard to the consequences. Dr. Stein concluded that although stereotypic disorders are not simply a subtype of OCD, there is some clinical utility to putting them in the same section as OCD.
Laura Marsh, MD (Baltimore, MD) reviewed the evidence for obsessive-compulsive spectrum behaviors in Parkinson's disease, explaining that Parkinson's disease provides a model for psychiatric symptoms that occur in the context of neurological conditions. An estimated 80- 90% of Parkinson's patients have a psychiatric disturbance over the lifetime of the disease, some of which are prodromal to the onset of motor symptoms. There is no consistent evidence for increased rates of OCD or OC symptoms in Parkinson's disease, which is in contrast to the higher rates of co-morbid anxiety disorders that occur in patients with Parkinson's. There is, however, the concept of a "Parkinsonian personality" which pre-dates the development of Parkinson's and is characterized by traits of industriousness, orderliness, inflexibility, cautiousness, and low novelty seeking, which bear some resemblance to OCPD. "Punding," characterized by intense engagement in purposeless repetitive activities like assembling/dissembling equipment, is associated with high-dose treatment of Parkinson's patients with dopamine agonists (14% prevalence). Unlike OCD where repetitive behavior is intrusive, resisting, or rule-bound, punding is often calming to the patient. Finally, dopaminergic medications and deep brain stimulation used to treat Parkinson's disease are associated with the development of impulse control disorders such as pathological gambling, hypersexuality, excessive spending, overeating, and abuse of dopaminergic medications.
Katharine Phillips, MD (Providence, RI) examined the relationship between body dysmorphic disorder (BDD) and OCD. Available data suggest that BDD and OCD have many similarities but also apparent differences. Shared features include many aspects of phenomenology; for example, both are characterized by intrusive, persistent, repetitive, and unwanted thoughts and behaviors. However, individuals with BDD have poorer insight (i.e., approximately 2% of OCD patients are currently delusional vs. 27-39% of BDD patients). Furthermore, compulsions in BDD (e.g., checking mirrors) appear less likely to reduce anxiety or distress. From the perspective of comorbidity, 32-38% of BDD patients have had OCD in their lifetime, whereas 3-37% of patients with OCD have comorbid BDD. BDD and OCD have many similarities in terms of comorbidity with other disorders and again, key differences. They have similar age of onset (in adolescence) and often-chronic course. Some findings from a prospective course study suggest that BDD and OCD are related disorders, although BDD often persisted in patients who remitted from comorbid OCD, suggesting they are not identical. BDD occurs more frequently in the first-degree relatives of OCD probands than in first-degree relatives of control probands, suggesting that BDD may be a member of the familial OCD spectrum. Like OCD, patients with BDD have impaired verbal and non-verbal memory which appears mediated by deficits in organizational encoding strategies, implicating cortico-striatal systems. A preliminary mMRI study found subtle caudate abnormalities in BDD, supporting conceptualization of BDD as an OCD-spectrum disorder, although, unlike OCD, there was a relative leftward shift in caudate nucleus asymmetry and greater total white matter volume in BDD than in normal controls. From a treatment perspective, like OCD, SRIs appear preferentially efficacious, but preliminary data suggest that augmentation strategies with typical antipsychotics that work in OCD may not be effective in treating core BDD symptoms. Dr. Phillips noted that the inconsistent classification of delusional variants of nonpsychotic disorders (e.g., OCD, BDD, mood disorders, hypochondriasis, anorexia) should be addressed in DSM-V, perhaps by adopting a dimensional construct for insight (delusionality).
Walter Kaye, MD (Pittsburgh, PA) presented on the relationship between anorexia nervosa (AN) and bulimia nervosa (BN) and the obsessive-compulsive spectrum disorders. AN and BN are related disorders that tend to occur mainly in females and most commonly have an onset during adolescence. AN has been viewed as a form of obsessive disorder since the 1930's. Furthermore, AN and BN have a number of symptoms that have the qualities of obsessions or compulsions, including ritualized feeding, odd food choices, obsessive interest in diet recipes, window shopping in supermarkets, preoccupation with body image distortions, "compulsive" exercise (e.g., calisthenics) and fidgeting. Other features that are similar to those seen in OCD include the ego-syntonic nature of the symptoms in AN, cognitive distortions in which the person focuses on details without seeing big picture, and impaired set shifting. Comorbidity between OCD and AN and BN is high, with reported rates ranging from 10% to 70%. In addition, there is a higher rate of lifetime eating disorder diagnosis in OCD patients (8-12%) compared to the general population. Rates of OCD are also elevated in first-degree family members of individuals with eating disorder (7-10%). However, two studies did not find an increased rate of eating disorders in families of probands with OCD. AN in particular is also associated with high rates of obsessive compulsive personality disorder (6-60%). Obsessive, anxious, and perfectionistic traits tend to occur in childhood before the onset of AN and BN, and persist after remission of eating and body image symptoms. Controlled trials show BN, and perhaps AN, respond to antidepressant medications but little is known about whether there is preferential response to serotonin-specific drugs. Neuroimaging data suggests some shared brain circuitry between AN/BN and OCD but state and methodological differences confound comparisons.
Eric Hollander, MD (New York, NY) presented evidence suggesting that autism may fit into the obsessive compulsive spectrum. Autism is a neurodevelopmental disorder with strong genetic factors (e.g., 90% concordance with identical twins). One of the core symptom areas of autism is repetitive behaviors and compulsivity and there are elevated rates of OCD-like behavior in individuals with autism (e.g., up to 25% with contamination obsessions and 32% with ordering compulsions). In OCD, there is a lifetime prevalence rate of 2.7% for Asperger's disorder, which is well above population estimates. Among individuals with autism, there is evidence for heritability of repetitive behaviors (e.g. parents of children with high rates of repetitive behaviors are more likely to have OC traits themselves than do parents of children with low rates of repetitive behavior). Genetic factors in autism and the repetitive behavior domain include linkage areas in chromosomes 1, 6, and 19 in an obsessive-compulsive behavior subgroup of individuals with autism. Neuroimaging results include a correlation between increased right caudate volume and repetitive behavior. Oxytocin may play a role in both autism and OCD; SRI's increase oxytocin levels in children with OCD and oxytocin infusion in adult autistic individuals improve verbal recognition of emotional tone. From a pharmacological perspective, SSRI's are effective in decreasing repetitive behaviors and anxiety and improving global functioning in autism, similar to their effect in OCD.
Joseph Zohar, MD (Tel Hashomer, Israel), presented a paper outlining the empirical data supporting the possible inclusion of a schizo-obsessive subtype of schizophrenia. He focused on individuals who have both OCD and schizophrenia and whose obsessions and compulsions are unrelated to the content of their psychosis. Available data suggests little correlation between the severity of OC symptoms (which are typically moderate to severe) and core schizophrenic symptoms. Fifteen out of 18 studies revealed at least a five-fold elevation of OCD prevalence in schizophrenia compared to general population (e.g., ranging from a low of 3% to a high of 64% depending on patient setting and methodology). Higher rates of disorders in the OC spectrum (BDD, chronic tic, etc) also occur in these individuals as compared to schizophrenia patients without OCD. Dr. Zohar noted that the prognosis of this schizo-obsessive subtype is worse than that of pure schizophrenia and that the OCD symptoms precede the first psychotic episode in about half the patients. From a family history perspective, although relatives of both schizophrenia and schizo-obsessive patients have comparable risk for schizophrenia-spectrum, only schizo-obsessive have higher risk of OCD in their relatives. On cognitive testing, patients with the schizo-obsessive subtype do worse on tests of executive dysfunction than do schizophrenic patients without OCD. From a treatment perspective, combination of dopamine-blocker and SRI's are required. Dr. Zohar concluded that although it remains unclear whether the high comorbidity indicates a pathological linkage between schizophrenia and OCD, it suggests that special attention should be given to this comorbidity.
Lorrin Koran, MD (Stanford, CA) and Stefano Pallanti, MD (Florence, Italy) together discussed how well the impulse control disorders fit into a possible Obsessive-Compulsive spectrum. With the exception of intermittent explosive disorder (IED), which is primarily a problem with anger control, the other impulse control disorders can be grouped into self-soothing behaviors (i.e., trichotillomania (hair-pulling) or skin picking) and reward-seeking behaviors (i.e., pathological gambling, kleptomania, pyromania, compulsive sex, compulsive buying). How good a "match" each of these disorders is to OCD varies:
pathological gambling is a fair match given some similarity in phenomenology (begins as ego-alien, and is intrusive and resisted), and course (usually chronic, sometimes episodic, with onset during adolescence), but heterogeneous drug response (responding to SSRIs, mood stabilizers, opioid antagonists and bupropion) as well as different brain circuitry goes against a close match with OCD;
pyromania is a fair-to-poor match given the pleasurable nature of the behavior, differences in phenomenology (commonly occurs with dissociation) and lack of comorbidity with OCD.
kleptomania is a fair-to-poor match given its different patterns of comorbidity, course, brain circuitry (frontal lobe), and treatment response (e.g., response to naltrexone and limited response to SSRI's);
compulsive buying is a poor match given its different phenomenology, lack of OCD family history, and different treatment response (SSRI's are ineffective);
hypersexuality (compulsive sexual behavior)) is a poor match given its focus on pleasurable activities, different brain circuitry (inhibition vs. hyperactivity), drug response (response to opiate and androgen blockers), onset (lack of prepubertal onset); and gender ratio (M/F 5:1);
skin picking is a poor match given the pleasurable nature of the behavior, its association with dissociation; gender ratio (F/M 2:1) and response to a different form of psychotherapy (habit reversal CBT instead of exposure/response prevention);
nail-biting is a poor match given its ego-syntonic nature, lack of comorbidity with OCD, typical onset in childhood; and limited response to SRI's;
intermittent explosive disorder is a poor match given its different phenomenology (often episodic and ego-syntonic), gender ratio (M/F 3:1), different brain circuitry (decreased frontal lobe and serotonergic activity), and its association with dopaminergic genes associated with aggression; and
Dr. Koran recommended that diagnostic criteria be developed for the impulse control disorders not included in DSM-IV (e.g., compulsive buying, skin picking) based on an examination of large numbers of cases or of community subjects to determine essential criteria or to establish criteria that predict treatment response. In the ensuing discussion, it was noted that the association of these disorders with OCD may differ according to subtypes of OCD (e.g., childhood-onset type) and that more thought needs to be given to the naming of these new disorders (e.g., should they be called reward-seeking behavior disorders?).
Marc N. Potenza, MD, PhD (New Haven, CT) presented on OCD and addictive disorders. (A presenter from the Substance Use conference participates in every other conference to underscore the need for appropriate attention to the high comorbidity of substance use and other mental disorders.) The core components of addictive behavior include continuing the behavior despite adverse consequences, diminished control or compulsive engagement in the behavior, and a craving or appetitive urge to engage in the behavior, opening up the construct to behaviors beyond substance use, like gambling, sex, shopping, eating, and computer use. Whether to classify these disorders as addictions, compulsive disorders or impulse control disorders is a matter of ongoing scientific debate. A compulsion is defined as an irresistible impulse to act regardless of the motivation. In OCD, compulsive behaviors are perceived as interfering, distressing, and/or bothersome, and are performed to relieve anxiety. In addiction, there is a progression from impulsivity to compulsivity, so that the behavior becomes less reward-driven over time. Epidemiological data support a greater than expected co-occurrence of OCD and substance dependence (odds ratios from 2 to 4), but a weaker relationship between pathological gambling and OCD (odds ratio 0.6). Regarding clinical course, addiction is characterized by high rates in adolescents and young adults and a decline in prevalence over the age spectrum. In contrast, OCD often has an earlier onset and lacks a peak with a rapid decline. With respect to gender differences, addictions typically have a male to female ratio of 2:1 which differs from a 1:1 ratio for OCD. Familial influences have been observed for a broad range of drug and behavioral addictions. Some genetic factors have been linked to specific drugs and their metabolism (e.g., alcohol and ADH/ALDH alleles; CYP alleles). Common genetic factors also may influence putative endophenotypes relevant across addictions, such as impulsivity, risk-reward decision-making, sensation-seeking, novelty-seeking, and stress-responsiveness. Regarding the neurocircuitry of OCD and addictions, in addiction relatively greater emphasis has been placed on "reward-related" pathways, particularly the mesolimbic dopamine circuitry. In OCD, increased activation of cortico-striato-thalamo-cortical circuitry has been consistently observed. The pharmacological and behavioral therapeutic approaches for OCD and addiction are largely different (e.g., cognitive behavioral therapy for OCD involves exposure/response inhibition whereas that for addiction treatment involves identifying internal/external cues associated with drug taking). In summarizing his presentation, Dr. Potenza noted that OCD and addiction are largely more different than similar, but if divided into less heterogeneous groupings based on biological data, more similarities may be found.
Trevor W Robbins, FRS (Cambridge, UK) reviewed cross-species animal models for OCD. Animal models can be used in two ways, modeling a presumed etiological factor or modeling a particular symptomatic expression. He noted that there can be effective animal models of psychiatric symptoms as long as these models are integrated fully with neurobehavioral phenotypes of disorders and a psychological theory linking it all together Animal models need to be validated by treatment (e.g., clomipramine, fluvoxamine, other SSRIs, cingulotomy, D2/D3 blockers like haloperidol) or genotype (e.g., COMT polymorphism). Mouse genetic models of OCD work superficially at a symptomatic level and include disruptions of Hoxb8 in mutants leading to excessive 'pathological' grooming behavior and 5-HT2C knockout mice with increased screen or clay chewing and perseverative head-dipping. A wide range of repetitive behaviors have been hypothesized to provide a neuropsychology of repetitive behaviors. Typical animal responses used to model compulsive behaviors include canine acral lick dermatitis, schedule-induced polydipsia, excessive wheel-running, and excessive grooming. The presentation also covered cognitive aspects and neural substrates of perseverative behavior, particularly the task of attentional set shifting, as it occurs in both animal models and in humans. Impairment in attentional set shifting is not only relevant in patients with OCD but is also seen in their first degree relatives.
James Leckman, MD (New Haven, CT) focused on Obsessive-Compulsive symptom dimensions. OCD is a consequence of the dysregulation of evolutionary conserved complex, and partially overlapping, neural systems that serve to detect, appraise and respond to potential threats across a number of dimensions. From this perspective, each of the OCD dimensions focuses on some aspect of our lives; for example, had our ancestors not been concerned about contamination, they would not have been successful. Factor analytic studies of OC symptoms from the widely used Y-BOCS Symptom Checklist have identified five dimensions: contamination-washing, harm-checking, hoarding, symmetry-ordering, sex-religion. Preliminary data indicate that a dimensional approach to OC symptoms may be heuristically valuable for future genetic, neurobiological and treatment studies as well as being of practical importance to clinicians. In response to critiques of the factor analytic methodology used to derive these dimensions (e.g., based on the a priori categories of the Y-BOCS), the DYBOCS was developed to measure symptoms within six dimensions: Aggressive; Sexual and Religious; Order and Symmetry; Contamination and Cleaning; Hoarding; and Miscellaneous. The factors emerging from studies of the DYBOCS are largely consistent with earlier studies using the Y-BOCS symptom checklist and emphasize several families of factors: a harm, sexual and religious factor; a Hoarding and ordering and symmetry factor, a Somatic/superstitions factor, and a Pathological grooming factor. Dr. Leckman recommended that symptom dimensions be documented when diagnosing patients with OCD in DSM-V given their implications to prognosis and treatment response.
Scott Rauch, MD (Charlestown, MA) addressed the neuroimaging and neurocircuitry of OCD and related disorders. Brain circuitry in OCD involves pathways from cortex to striatum to thalamus. Two pathways, however, run from striatum to thalamus; one, termed the direct pathway, amplifies activity in the circuit, while the other - the indirect pathway - suppresses or filters transmission through the circuit. A variety of neuroimaging methods have been applied in a complementary fashion to yield convergent findings regarding OCD. Specifically, the nodes of this circuit (i.e., orbitofrontal cortex, anterior cingulate cortex and caudate regions) exhibit hyperactivity at rest, which is accentuated during symptom provocation and attenuated following treatment. There are few candidate findings, however, that offer sufficient specificity and reliability to be useful for differentiating OCD from other disorders. A dimensional approach may be informative in this regard. Dr. Rauch suggested that imaging data could be used to support several of various options for restructuring the DSM.
David Mataix-Cols, PhD (London, UK) presented on the neuroimaging and neurocircutry of the OCD symptom dimensions. He described an experimental paradigm in which subjects were exposed to images related to each symptom dimension (e.g., a photograph of a toilet seat for contamination concerns) and were given functional MRI's. Each of the studied symptom dimensions was mediated by distinct but partially overlapping neural systems. While both patients and controls activated similar brain regions in response to symptom provocation, patients showed greater activations in bilateral ventromedial prefrontal regions (washing experiment), putamen/globus pallidus, thalamus and dorsal cortical areas (checking experiment), and left precentral gyrus and right orbitofrontal cortex (hoarding experiment). These results were further supported by correlation analyses within the patient group, which revealed highly specific positive associations between subjective anxiety, questionnaire scores and neural response in each experiment. Based on these preliminary neuroimaging findings, Dr. Mataix-Cols concluded that OCD may not be a unitary entity; consequently,, using OCD as the benchmark to decide if a disorder is part of the OCD spectrum is complicated since OCD itself can be conceptualized as a spectrum of multiple overlapping syndromes. Removing OCD from the Anxiety Disorders umbrella will require careful thought: some OCD dimensions may be more like anxiety disorders (i.e., washing) whereas others may be more like neurological disorders (i.e., symmetry/order) and hoarding itself is heterogeneous (some patients are "compulsive" and others "impulsive"). He also recommended using the same imaging paradigms across patient groups in order to establish the common and distinct neural correlates of the symptom dimensions of OCD and the OCD spectrum disorders.
Susan Swedo, MD (Bethesda, MD) presented on neuroimmunological issues across the Obsessive-Compulsive spectrum, using Sydenham chorea (SC), one of the major manifestations of rheumatic fever, as a model of post-streptococcal OCD/tics. The association between SC and OC symptoms has long been recognized, with up to 75% having obsessive-compulsive symptoms. More recently, it has been recognized that a subset of children with OCD have PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections). The five criteria for PANDAS include: the presence of OCD or tic disorder, prepubertal onset, episodic course of symptom severity; association with neurological abnormalities (e.g., choreiform movements during symptom exacerbation) and a temporal relationship between symptom exacerbation and streptococcal infections. In the PANDAS subgroup of OCD and tic disorder, the diagnosis is made based on historical associations between streptococcal disease (positive streptococcal throat culture or a rise in antistreptococcal titers) and neuropsychiatric symptom exacerbations. A model for the pathogenesis of PANDAS proposes that Group A beta-hemolytic strep infects a susceptible host (only about 1 in 20 families are susceptible), leading to abnormal immune response with cross-reactive antibodies that target the surface of human neuronal cells. The treatment strategy is to obliterate the strep and then prophylax against it. Penicillin and zithromycin in prophylactic doses are effective in reducing number of strep infections and neurological sequalae. Dr. Swedo questioned whether there might be other symptom complexes that represent a post-infectious neuropsychiatric disorder. Furthermore, PANDAS raises questions about how DSM-V should handle etiologic subgroups of disorders. In a complementary presentation, Euripedes Miguel, MD, PhD (Sao Paulo, Brazil) discussed additional research suggesting an association between rheumatic fever (RF) and OCD. More recent studies report high frequencies of OC symptoms, OCD, and tic disorders, not only in patients with SC but also in patients with rheumatic fever without SC. These psychopathological manifestations have been described in both the acute and chronic phases of RF (e.g, patients with valvular heart disease developed as a sequela of carditis). The presence of OC symptoms and tics in non-acute-RF patients suggests several possible mechanisms. First, acute changes related to RF might have persisted or triggered late pathophysiological changes that increase susceptibility to these neuropsychiatric disorders. Second, it is possible that OCD and related disorders such as tic disorders and RF share a common familial/genetic etiology. This possible familial relationship was supported in a case-control family study that found that the rate of OCD/tics and related disorders such as BDD to be significantly higher among first degree relatives of RF probands than among first degree relatives of controls.
Next, three presentations looked at family studies and genetics. Humberto Nicolini, MD, PhD (Mexico City, Mexico) presented on family and genetic studies of OC spectrum disorders. Noting that OCD rates are elevated in relatives of probands with OCD, especially when there is comorbidity with tics and an earlier age at onset, he reported that they also are more likely to have the types of obsessions and compulsions displayed by the probands - for example, ordering, checking, and symmetry. Segregation analysis, used to assess modes of inheritance, suggests that there is evidence of a single gene (autosomal dominant) for the following OCD subtypes: symmetry and ordering symptoms, OCD with eating disorders, and early age of onset. Using linkage analysis, which provides etiological evidence for single gene in a disorder, two studies indicate involvement of a gene on 9 p24. Gerald Nestadt, MD (Baltimore, MD) presented on familial approaches to addressing the question of what disorders should be included as OC spectrum disorders. The prevalence of OCD in relatives of probands is clearly elevated; 12% in first-degree relatives, compared to 2% in relatives of normal controls. For the anxiety disorders, there is no elevation in rates for specific and social phobia but there are higher rates of GAD, separation anxiety disorder, panic, and agoraphobia in first degree relatives of probands with OCD. Controlled for the presence of these disorders in the probands, GAD and agoraphobia remain significantly higher in first-degree relatives, suggesting a strong relationship to the OCD phenotype. While major depressive disorder is elevated (in contrast to bipolar and dysthymic disorders), it is no longer significant when adjusting for MDD in the relatives, suggesting that MDD in relatives may be secondary to OCD. For other spectrum disorders, although there are elevated rates of BDD, hypochondriasis, eating disorders, and trichotillomania/nail biting in relatives, controlling for the presence of these in the probands indicates that only the relationship with BDD remains significant. While there were higher rates of tics in relatives or probands with OCD, rates of OCD were not higher in relatives or probands with tics. In his presentation, James Kennedy, MD (Toronto, Canada) discussed candidate genes and subtypes in OCD. The best candidate genes for OCD include 5HTTLPR and 5HT1B (related to the serotonin system) and GRIN2B and Glutamate transporter (related to the glutamate system). Regarding the symmetry/ordering subtype of OCD, candidate genes include 5HTTLPR and the glutamate transporter. Finally, BDD is associated with the GABA-A gene and 5HTTLPR but not 5HT1B. Future research directions, he suggested, include further exploration of the glutamate system that will iteratively refine phenotypes to genotype correlations, and combining genetics and imaging. Once whole genome association studies become available in the next few years, it could conceivably lead to a kind of "fingerprint" for OCD based on dozens to as many as 100 genes.
Next, three presentations considered the nosological implications of different treatment strategies. Herman Westenberg, PhD (Utrecht, Netherlands) proposed a pharmacologic dissection of OC behavior spectrum disorders. Noting that pharmacological treatments have a global effect on the brain neurochemistry and may work differently (e.g.., in different brain regions) in various disorders, he urged caution about making inferences from treatment studies. OCD responds only to potent serotonin reuptake inhibitors (e.g., clomipramine, venlafaxine, and SSRI's); norepinephrine reuptake inhibitors, such as imipramine, and anxiolyics have been shown to be ineffective. The effective dose of SRIs is higher in OCD than for most anxiety disorders and MDD, and the duration of time to onset of action is longer for OCD. With respect to subtypes, symmetry and ordering, in contrast to washing compulsion, may respond more poorly to SRIs . The effectiveness of SRIs in OCD, however, does not imply that there is a serotonergic dysfunction in OCD given that the response rate is less than 50% and that doses of SSRI's are substantially higher than necessary to completely occupy the 5-HT transporter in most brain regions. SSRIs at high doses may affect other neurotransmitter systems such as dopamine. Evidence suggesting an important role for dopamine includes the observation that atypical antipsychotics can augment SSRI's in refractory patients; that neuroimaging studies show abnormalities in brain regions densely endowed with dopaminergic terminals; and that stimulation of dopamine by buproprion exacerbates OC symptoms. Dopamine probably plays a central role in the pathophysiology of OCD, Dr. Westenberg, concluding, noting that certain disorders within the OC spectrum may best be understood within the conceptual framework of behavioral addiction. Benjamin D. Greenberg, MD, PhD (Providence, RI) presented on anatomically focal brain probes and treatments in OCD. The relatively highly focal effects on brain circuitry make these interventions potentially informative for understanding pathophysiology and mechanisms of treatment response, although their effects may be more on traits associated with an illness. One caveat in interpreting these results is that surgical candidates and general patient pools may differ meaningfully. Another issue is that different symptom dimensions or clusters within the categorical construct of OCD may respond differently to focal brain circuit-based interventions within an individual (e.g., for one patient treated with gamma knife capsulotomy, most OCD symptoms decreased, while hoarding symptoms increased). Although Transcranial Magnetic Stimulation (TMS) cannot get deeper than cortex, in some studies it decreased compulsions and/or obsessions and improved mood. Stereotaxic lesion procedures include anterior capsulotomy (for OCD, MDD, non-OCD anxiety), anterior cingulotomy (OCD, MDD, pain), and limbic leucotomy (MDD, OCD, self-mutilation with tic-like compulsions), and each has had some evidence of effectiveness (30%-70%) in open trials. Deep brain stimulation also has been effective in some treatment-resistant cases and has the advantage of reversibility, adjustability, and suitability to controlled trials. Interestingly, in a very small pilot study, a strong predictor of decrease in OCD severity and improvement in function after DBS response was the ratio of extraversion to neuroticism, two dimensions of personality, measured at presurgical baseline. Lee Baer, PhD (Boston, MA) reviewed research on the response of OC spectrum disorders to Cognitive Behavioral Therapy. Since CBT is a catch-all phrase that has little specific meaning, it is important that individual treatment components be specified when reviewing study results. Exposure and Response Prevention (ERP) is most effective for fear (or disgust) and avoidance in OCD-spectrum problems. For OCD, it is most effective for washing and checking compulsions but less effective for ordering, superstitions, and hoarding. Habit Reversal (including awareness training and affect management) is most effective in those stereotyped behaviors of OCD spectrum disorders that do not involve anxiety or cognitions (e.g., tics, skin picking, hair-pulling). Cognitive therapy (CT) is a relatively non-specific technique effective in OCD-spectrum problems only when the patient has irrational beliefs underlying his or her rituals or obsessions (e.g.," if my face is not free of wrinkles I am hideous").
Soraya Seedat MBChB (Tygerberg, South Africa) and Hisato Matsunaga MD (Osaka, Japan) presented jointly on cross-national and ethnic issues in OC spectrum disorders. Epidemiological and clinical data on OCD and many of the OCSDs across various cultural and geographic settings suggest that this is a group of disorders with a good degree of trans-cultural homogeneity. Relative to the OCSDs, cross-national data are most robust for OCD with respect to consistency in rates, age at onset, and comorbidity. Socio-demographic features and core symptoms are to a large extent independent of cultural, ethnic, religious and geographic differences. Where standard instruments have been used, the symptom profiles of OCD and OCSD have been remarkably similar across countries. However, the themes and course of these disorders may be shaped by cultural, ethnic or religious experience (e.g., religious themes in patients with OCD often prevail in countries where religion plays an important role in society, like the Middle Eastern countries). Little is known about the mechanisms through which culture and ethnicity may affect the nuclear phenotype; this is an area where more work is needed. To ensure better diagnostic reliability and validity cross-culturally, further studies are needed using internationally reliable instruments aimed specifically at comparing trans-cultural aspects (socio-demographic, clinical, and prognostic features) of OCD and the OCSDs. Furthermore, future neurobiological and genetic approaches to studying OCSDs should incorporate an assessment of the interactions between culture and biology to allow for better understanding of its impact on the development and maintenance of these disorders.
Upon conclusion of the presentations, participants reconvened in two breakout groups to formulate recommendations for research and suggestions for future versions of the classifications of mental disorders (e.g. in the DSM-V, ICD-11 and others). The first group, lead by Dr. Hollander, agreed that the data support five dimensions of OCD symptoms (Symmetry, Hoarding, Washing/contamination, Worries about Harm/Checking/Doubt, Taboo thoughts [sex and violence]) but questioned whether these dimensions should be used to guide treatment or are useful only for research purposes. In considering what the criteria should be for deciding whether a disorder falls within the spectrum, the group agreed on the need for strong evidence from at least three out of the five domains (i.e., phenomenology, comorbidity, family history, fronto-striatal brain circuitry [e.g., caudate hyperactivity], or treatment response), with either family history or fronto-striatal brain circuitry required. Based on this criterion, the group concurred that the strongest evidence for a relationship with OCD was for Tourette's, body dysmorphic disorder, and hypochondriasis. There was also support for OCPD, hoarding, Sydenham's/PANDAS, trichotillomania, and eating disorders, and the least support for skin-picking and nail-biting. The group also recommended that the DSM-IV impulse control disorders and the ICD's-NOS (including impulsive-compulsive buying, sex, and internet use) be grouped together into a new category called "Behavioral and Substance Addictions", or alternatively "Impulsive-Compulsive Disorders. Possible subtypes of OCD to be considered include: "tic-related," "childhood-onset," "symmetry type," "hoarding type," "post-partum," "poor insight," "impulsive type," "inattentive typ,e" and PANDAS.
The second group, led by Dr. Zohar, recommended that, based on family studies, response to treatment, course, some biological markers, epidemiological findings, and brain imaging, OCD should be removed from anxiety disorders. Regarding subtypes for OCD, the group agrees that the strongest evidence was for an early-onset type, for which PANDAS could be listed as a specifier (since all PANDAS cases are early-onset but not all early-onset cases are PANDAS). The group concurred that BDD, hypochondriasis, and tic disorders had the strongest evidence arguing for their inclusion into an OC spectrum grouping. The group also proposed that the reward/impulse control related disorders (e.g., pathological gambling, trichotillomania) not be included in the OCD spectrum. The group also suggested that the definition of obsession be revisited since not everyone with OCD experiences them as "intrusive" as is now required in DSM-IV. Finally, although agreeing on the importance of increasing awareness of the comorbidity between schizophrenia and obsessive compulsive disorder, the group was divided on whether a specifier for schizo-obsessive type should be added to schizophrenia.
The complete presentations will be published in a future monograph by American Psychiatric Publishing, Inc.