Obsessive-Compulsive Disorder

Kenneth E. Towbin

Mark A. Riddle

Introduction

Obsessions are among the oldest mental symptoms for which there are detailed descriptions. Obsessions and compulsions were depicted as possession by the devil in 1467 in Malleus Malficarum (1) and described in the “Obsessi” of Paracelsus. In the 1600s, pious texts tell of extremes of religious doubting and “scrupulosity,” or excessive devotion (2). Pioneers of psychiatry like Esquirol (3), Maudsley (4), Freud (5), and Janet (6) took up this troubling and fascinating disorder and their writings reflect the prevailing philosophy of thought, motivation, and free will.

Two discoveries promoted OCD research in the last three decades. First, there was the discovery that medications that inhibit serotonin reuptake are effective for many OCD patients. Subsequently, powerful techniques for observing structures and measuring regional brain activity were applied to learn about brain function during performance of tasks tapping specific regions of interest. The effort to understand OCD has deepened our understanding of the prevalence, course, etiology, and pathology of these symptoms and along the way, broadened our knowledge of the neuroanatomy of voluntary cognitive functions and behavior.

Definitions

Obsessions are unwanted thoughts, images, or impulses that are recognized as senseless or unnecessary, intrude into consciousness involuntarily, and cause functional impairment and distress. Despite this lack of control, a person with obsessions is aware that these thoughts originate in his or her own mental activity. Since they arise in the mind, obsessions can take the form of any mental event— simple repetitive words, thoughts, fears, memories, pictures, or elaborate dramatic scenes.

Compulsions are actions that are responses to a perceived internal obligation to follow certain rituals or rules; they too cause functional impairment. Compulsions may arise as direct consequences of obsessions or indirectly through efforts to ward off certain thoughts, impulses, or fears. Children often report that their compulsions do not have a preceding mental component. Like obsessions, compulsions are often viewed as being unnecessary, excessive, senseless, involuntary, or forced. Individuals suffering from compulsions will often elaborate a variety of precise rules for the chronology, rate, order, duration, and number of repetitions of their acts.

These definitions reflect three critical concepts that are relevant to the differential diagnosis. An essential criterion is
functional impairment as a consequence of symptoms. Two others draw on classic definitions (4,7): Individuals feel that they are being forced or controlled by the symptoms, while they possess insight into the senselessness or excessiveness of their thoughts or acts. Although most patients see their compulsions as unnecessary or their thoughts as senseless, some have this only intermittently. Consequently, some investigators have reservations about the criterion that patients possess insight about their illness. Insel and Akiskal (8) and Lelliott and coworkers (9) reported on severely impaired patients who at times doubted the need to perform their rituals or thought their behaviors were senseless and, at others, were convinced of their necessity to the point of near or actual psychotic proportions. DSM-IV criteria for obsessive-compulsive disorder (OCD) (Table 5.5.2.1) have been modified such that awareness of the senselessness or excess of the symptoms only must be present at some phase of the illness. For children, this criterion is set aside altogether.

Few studies differentiate between participants with childhood- and adolescent-onset OCD. Therefore, in this chapter, “child,” “childhood,” or “children” will be used to signify children and adolescents. There are studies that have sampled adolescent subjects only. When this is so, the more exclusive term will be employed.

Prevalence and Epidemiology

The prevalence of OCD in childhood should be understood in the context of the high prevalence of subclinical obsessions or compulsions in the population. Evans and coworkers (10) sent out mailings to parents of children less than 6 years old and found that urges to make things “just right” and preoccupations with symmetry and rules were very common in this unselected population. It was notable that these concerns declined as children entered grade school age. To learn about the prevalence of obsessive and/or compulsive symptoms and compare prevalences across development, Zohar and Bruno (11) used self-report measures for a study of 1,083 children attending grades four, six, and eight. As predicted, a large segment of the pediatric population confirmed experiencing these features. Sixty percent of fourth graders reported preoccupations with guilt about lying and engaging in checking behaviors and 50% reported contamination and germ fears. By eighth grade, rates for these concerns declined to 40%, but 60% of eighth graders reported worries about cleanliness and 50% noted intrusive rude thoughts. Mean scores across the age range on the Maudsley Obsessive Compulsive Inventory were 11–12.5/30 in this population; like Evans and coworkers’ (10) finding from a younger population, rates of behaviors and symptoms declined over time. A subgroup of eighth graders had high symptom scores (greater than 2 standard deviations from the mean) and reported high levels of anxiety. The authors suggest that this small group (4%) represented a clinically at-risk group because the large number of symptoms and elevated state and trait anxiety were such a contrast to the decline in both in their age-mates.

Twenty years after the Epidemiologic Catchment Area (ECA) survey (12), there continue to be important disagreements over the most reliable prevalence rate for OCD. Significant disparities and varied methods have yielded figures between 0.5 and 3%. It is useful to place these figures in the context of adult epidemiological reports.

TABLE 5.5.2.1 DSM-IV CRITERIA FOR OCD

Either obsessions or compulsions:
Obsessions: as defined by 1), 2), 3), and 4):
1. Recurrent and persistent ideas, thoughts, impulses, or images that are experienced at some time during the disturbance as intrusive and inappropriate and cause marked anxiety or distress.
2. The thoughts, impulses, or images are not simply excessive worries about real-life problems.
3. The person attempts to ignore or suppress such thoughts, impulses, or images to neutralize them with some other thought or action.
4. The person recognizes that the obsessional thoughts are the product of his or her own mind (not imposed from without, as in thought insertion).
Compulsions: as defined by 1) and 2):
1. Repetitive behaviors or mental acts that the person feels driven to perform in response to an obsession, or according to rules that must be applied rigidly.
2. The behaviors or mental acts are aimed at preventing or reducing distress or preventing some dreaded event or situation; however, these behaviors or mental acts either are not connected in a realistic way with what they are designed to neutralize or prevent or are clearly excessive.
At some point, the person has recognized that the obsessions or compulsions are excessive or unreasonable. Note: This does not apply to children.
The obsessions or compulsions cause marked distress, are time consuming (take more than 1 hour a day), or significantly interfere with the person’s normal routine, occupational (or academic) functioning, or usual social activities or relationships.
If another Axis I disorder is present, the content of the obsessions or compulsions is not restricted to it (e.g., preoccupation with food in the presence of an eating disorder; hair pulling in trichotillomania; concern with appearance in the presence of body dysmorphic disorder; preoccupations with having a serious illness in the presence of hypochondriasis; or guilty ruminations in the presence of a major depressive disorder).
Not due to the direct effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition.

(Adapted from American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders-TR (4th ed.). Washington, DC, American Psychiatric Association, 2001.)

The early work (13), using weak methods by current standards, reported a prevalence of 0.05% for adult OCD. The scarcity of OCD was predominant until 1984, when the ECA survey reported surprising prevalences of 1.2–3.29% (12). Subsequent examination of these high rates uncovered weak concordances for OCD. Among all the diagnoses in the ECA study, the concordance between diagnoses derived from lay interviewers and trained clinicians for OCD were the poorest (14,15,16). Lay interviews employing the Diagnostic Interview Schedule (DIS) rely on simple yes or no responses to queries about the presence of broadly defined obsessions or compulsions. These are not reliable measures for actual clinical cases of obsessive-compulsive disorder. Helzer and co-workers (15) went on to say that results from community surveys, where many subjects cluster at the “threshold of the diagnostic definition,” will be unreliable because the response to a single probe carries too much weight. Karno and coworkers (17) reanalyzed ECA data, and prevalence rates were sustained despite the flawed methodology. False negatives balanced false positives.

In an effort to learn more about this problem, Nelson and Rice (18) used ECA methods and interviewed a community sample at two intervals separated by 12 months. The 1-year stability of the diagnosis of OCD was “very low.” Only 20% of those responding to “ever having symptoms” at Time 1 reported, “ever having symptoms” at Time 2. They concluded that “the DIS diagnosis of obsessive-compulsive disorder possesses extremely limited validity.”

Stein and coworkers (19) employed a different measure (the Comprehensive International Diagnosis Interview) and DSM-IV criteria. Again, rates derived from lay interviews revealed 22–25% of adults expressed having obsessions or compulsions. However, when clinicians reviewed lay interviews, the rates dropped seven-fold (to 0.7%). They also examined rates of “subclinical OCD” in which criteria were met for symptoms, but not impairment in DSM-IV. Rates for this subclinical syndrome were roughly equivalent to clinical OCD (0.6%). Stein and coworkers (19) concluded, like Karno (17) and Nestadt (16), that lay interviews led to many false-positive diagnoses.

The first prevalence rates reporting childhood data ranged from 0.2% (20,21) to 1.2% (22). Flament and coworkers (23) conducted a rigorous study of children and adolescents to discover a general adolescent population prevalence. Screening employed a modified Leyton Obsessional Inventory (LOI); this was followed by direct clinical interviews of subjects with a high score. This produced a point prevalence of 0.35% and a lifetime prevalence of 0.40%. Weightings, such as those used in the ECA study, gave hypothetical point and lifetime prevalence rates of 1% and 1.9%, respectively (23). Of the 18 adolescents diagnosed with OCD, 12 (67%) reported that symptoms resulted in high subjective interference, yet global assessment (CGAS) scores averaged 67 (“generally functioning pretty well”). To ascertain prevalence in a nonclinical population, Zohar and coworkers (24) performed detailed clinician-rated evaluations of 562 consecutive 16- to 17-year-old male and female inductees to the Israeli army. The OCD prevalence rate was 3.6%. Fifty percent of cases were identified as having “obsessions only.” This latter figure was substantially higher than that reported from clinical populations and casts doubt on the validity of the diagnoses generated by this method. However, rates for compulsions compared favorably to those of Flament and coworkers (23).

Overall, the prevalence of OCD is greater than that reported in the 1950s, but is not as great as reported in the ECA (17). Current valid studies place the prevalence closer to 0.6–1% (19,25,26). Adult epidemiological studies have reported prevalence from 0.5 to 3%. However, these figures may be erroneous because of unreliable diagnostic methods using nonclinician interviews, the DIS, or similar highly structured instruments (18,26,27). Also, it is obvious that rates will change as different diagnostic criteria are employed (19,26,27).

Valid epidemiological studies suggest that OCD shows an equal sex distribution. Males appear to have an earlier age of onset (28,29). Noshirvani and coworkers (30) reported that males and females were equally represented in their sample, but 35% of male subjects had their onset between the ages of 5 and 15, compared to 20% of cohort females. Generally, symptoms exist an average of 7–8 years before reaching clinical attention (23). Noshirvani and coworkers (30) suggested that males tend to have a longer duration of illness prior to seeking treatment, yet the ECA study (17) showed that subjects with OCD had rates of medical and mental health service utilization roughly equal to persons with other disorders.

Clinical Description

The variety of obsessions and compulsions match the unlimited capacity of the human mind and body. Typically, patients experience obsessions and compulsions. A few individuals have one or the other exclusively (28,31) and when this arises, having only obsessions appears to be the more common (24). At any one time, most patients experience multiple obsessions or compulsions. Over time, the objects and contents of symptoms change (32). The content of obsessions can show a wide range, but some themes are more frequent and influenced by the individual’s level of development. Adolescents’ obsessions typically focus on dirt and germs, fears of an ill fate befalling loved ones, exactness or symmetry, and religious scrupulousness (28). Bodily functions, lucky numbers, sexual or aggressive preoccupations, and fear of harm to oneself are less common. In adults, these remain frequent, but aggressive and sexual obsessions are more common (29,33,34).

Although any action can become a compulsion, some actions are more common. An adolescent clinical cohort (28) displayed (in descending order of frequency) cleaning rituals, repeating actions (doing and undoing), and checking rituals most commonly. Many fewer subjects reported rituals to protect themselves from illness or injury, ordering maneuvers, and counting behaviors. In adults, the most common compulsions are checking and cleaning (29,34,35). Slowness (34), counting (29), or doing things by numbers (35) each have been reported as third most common.

Several investigators (36,37,38) suggest that obsessions and compulsions should no longer be viewed as separate entities. These investigators used factor analysis to reconfigure Children’s Yale–Brown Obsessive Compulsive Scale (CY–BOCS) symptom categories and found a four-factor model that they believed was more meaningful (39). Rather than just one homogeneous entity or as “two factors” (obsessions or compulsions), there are strong reasons to consider that OCD might be better viewed as composed of four or more subtypes. A number of investigators have now identified four-factor subtypes: 1) aggressive, sexual, religious and somatic obsessions with checking compulsions, 2) symmetry obsessions with counting, arranging, ordering, and repeating compulsions, 3) contamination obsessions with cleaning and washing and 4) hoarding obsessions with hoarding and collection compulsions (36,37,38,40,41,42,43,44). Mataix Cols and coworkers (41) found a five-factor model that has been replicated as well: 1) aggressive-checking, 2) symmetry-ordering, 3) contamination-cleaning, 4) sexual and religious obsessions, and 5) hoarding. These solutions are close to one another and it is not yet clear which will be the most predictive.

Subsequent research, using factor analyses of symptoms without starting with Y-BOCS categories, argues for a yet more
meaningful, “multidimensional” way of grouping symptoms and thinking about OCD symptom patterns and subtypes (45). Using latent class analysis of symptoms, Mataix-Cols and coworkers (45) and Nestadt and co-workers (46,47) identified symptom groupings that were more meaningful when placed in the context of age of onset and comorbid diagnoses. From this work there is evidence for factors 1 and 2 (above) that are associated with early onset OCD, and a factor 4 (hoarding) that does not associate with other disorders. Furthermore, these factors show some association with comorbid disorders. Factor 2, described above, has been associated with earlier age of onset and tic disorders (37,41,48). Hoarding was associated with more Axis II psychopathology [particularly obsessive-compulsive personality disorder (49)], social disability and social anxiety disorder (45). Subsequent work looking more closely at comorbid diagnoses also relying on latent class analysis suggests greater likelihood of comorbid diagnostic groupings, such that OCD occurs with recurrent major depression and generalized anxiety (46,48), and another that associates OCD with agoraphobia, panic disorder, and tic disorders (46).

There is mounting evidence that these factors correlate with treatment response (41), neuroimaging results (45,50), neuropsychological function (51), and genetics (52,53). The evidence suggests that the course, genetic risk, neuropathology, and treatment might be different among these subgroups. For example, Alsobrook and coworkers (54) found that a major gene locus model was more strongly supported in families ascertained where the probands had symmetry/ordering symptoms (factor grouping II), than in probands with other symptom groupings. Alonso and coworkers (49) found that those scoring highly on the religious-sexual obsessions dimension fared more poorly over two years. Nevertheless, a serotonin reuptake inhibitor treatment study failed to support differences in response among three of these five subtypes (41). Several other caveats were suggested by Summerfeldt and coworkers (38), including the possibility that “more-than-four” factor models might also be viable, that a variety of items have been excluded in the analyses performed to date that use categories from the YBOCS, and problems with using “lifetime” ratings as these are subject to recall bias.

Last, longitudinal studies suggest that symptoms are likely to change over time. One study suggested that individuals are not confined to one subgroup exclusively over the course of their illness (32). However, in several longitudinal studies of adults, symptoms change but remain within the same symptom-groupings that have been identified by latent class analysis (55). Thus, adults with OCD show a high two-year symptom-grouping stability (41,55). It remains to be shown whether this is true for children.

Children with OCD might be described as more selectively impaired than children with other psychiatric disorders. Academic achievement and extracurricular functioning are often preserved, although the quality of peer relationships may be variable (56). Studies with adults point to significant impairment in social and role function (57). In Koran and coworkers’ study (57), moderate to severe illness was correlated in a linear way with social impairment. Clinical studies of children entering treatment programs have consistently demonstrated average intellectual quotients, although the selection bias of these cohorts must be considered in interpreting these data.

It is hard to convey how limiting OCD can be. On the surface children and adolescents appear to function well and seem relatively well adapted to their lives. However, severe symptoms often envelop the patient and his or her family completely (57,58). It is common to learn of washing rituals that consume 4 hours of scrubbing daily, dissolve an entire bar of soap each session, leave the patient’s hands worn raw and macerated, and raise monthly water bills dramatically. Counting or ordering compulsions can waste half a day and lead to complete obstruction. Rituals repetitively executed from night to early morning may curtail sleep to a few hours. Checking or cleansing rituals can produce physical injury such as skin lacerations, ulcers, and chemical burns.

The family’s reaction to the patient’s symptoms is crucial. Several common response patterns may produce delays in evaluation and treatment of childhood OCD. Although patients are embarrassed and secretive about the content of, and the limitations imposed by, their symptoms, serious impairments rarely elude family members. Parents may delay obtaining treatment as a result of a false hope that symptoms will extinguish if everyone acquiesces and aids in performing the activities or out of fear of stigmatization (58,59). This kind of family assistance does not relieve the child’s anxiety. Many times, parents cannot extricate themselves once they involve themselves in the rituals. Their children force assistance from them and implore them to continue. Over time, the child and parent may become pathologically entangled in rituals. When a child has sturdy development in other domains, parents find it hard to believe that he or she suffers from a serious disorder. In addition, the child’s claims that the thoughts or acts are ridiculous or unnecessary can instill false security, leading parents to think that “its just a phase.” Parents with subclinical obsessive or compulsion-like behaviors may be unable to recognize symptoms in their child and unwittingly minimize the child’s impairment. Reassurance from clinicians and pediatricians who are unfamiliar with OCD may lead to mistaking severe symptoms for “normal” reactions. It is a frightening and painful moment when parents recognize that their child is ill and has lost control of his or her thoughts and actions.

Etiology and Pathogenesis

Genetic Studies

There is ever-stronger evidence that genetic transmission confers vulnerability to OCD. Elevated concordance rates are observed among monozygotic compared to dizygotic twins (29,60) and higher rates for OCD are seen among first-degree relatives of clinically ascertained individuals (61,62,63).

The initial family studies did not find elevated rates in first-degree relatives (64,65,66). However, after the ECA study (12), methods changed and rates began to climb. Lenane and coworkers (67) reported that 30% of 46 adolescent probands had a first-degree relative with OCD. OCD was reported in 17% of parents and the age-corrected rate for siblings was 35%. Riddle and co-workers (31) reported that 10% of parents from a cohort of 21 clinic patients with OCD were diagnosed with OCD. Twenty-five percent had subthreshold symptoms of obsessions or compulsions. In contrast to studies that relied on questionnaires or telephone inquiry, Pauls and coworkers (61) directly examined relatives of 100 individuals with OCD and of a control group of 33 psychiatrically unaffected individuals. The rate of DSM-III-R OCD among OCD relatives was 10% versus 2% of the unaffected probands’ relatives. There were also significant differences in rates of OC traits (or subclinical OCD), with 8% of OCD probands’ relatives being affected and 2% of the relatives of those in the control group. Combining the subclinical and clinical threshold groups, 18% of OCD probands relatives were affected, compared to 4% of the control relatives. Like Pauls and coworkers (61), Nestadt and coworkers (62) blindly evaluated 343 relatives of adults with OCD and obtained a control sample of 300 relatives of 73 individuals ascertained by random number dialing. OCD patients were obtained from clinic rosters. Most family members were interviewed directly and all family members
were required to participate for a family to be included. Using a threshold of “definite” DSM-IV criteria, 11.7% of relatives of persons with OCD were affected, compared to 2.7% of control relatives. If “probable and definite” OCD were included, then the rates climbed to 16% versus 6%. An additional finding from this group (62) was that earlier age of onset was associated with greater “familiality” (greater likelihood of OCD among relatives). This finding was later replicated in a separate study by another group (69). Nestadt, Lan, and coworkers (68) performed an elaborate study employing segregation analysis of 80 families and a control group. The model of an autosomal dominant pattern of transmission provided the best fit to the data by both groups (68,69).

Reports of elevated rates of OCD among patients with Tourette syndrome (70) and of tics and a family history of tics among OCD probands (63,71,72,73) suggest that some OCD may arise from the same genetic etiology as Tourette syndrome (74). Reevaluation of the National Institute of Mental Health (NIMH) cohort and their relatives in a 2–7-year followup lends support to this hypothesis (75). Fifty-seven percent of 54 probands had a lifetime history of tics, 15% met criteria for Tourette syndrome, and 22% had chronic multiple tics. Among their relatives, 14% had lifetime diagnoses of tics. Using a control group, Grados and coworkers (73) and Hanna and coworkers (63) found that tics and OCD were significantly more likely in relatives of probands ascertained for OCD. Even more provocative were findings that “any tic disorder” was equally likely in relatives of probands with OCD, whether the proband had a lifetime history of tics or not. An additional finding from both studies (63,73) was that those relatives with tics plus OCD had an earlier age of onset when compared to those relatives with only OCD.

Neuropsychological Functions

There is good evidence from neuropsychological investigations that OCD is associated with impairment in prefrontal lobe functions. The evidence is strongest for impairments in executive functions of set-shifting and motor inhibition, and in nonverbal memory, visual motor integration, and visual-spatial memory (76,77,78). The most robust findings for pediatric OCD point to deficits in set motor inhibition and response suppression (76). These data point to deficits in dorsolateral prefrontal cortex, orbitofrontal cortex, cingulate, and parietal lobes (77). Also, the data suggest that attention, planning, working memory, and decisionmaking are not particularly weak in those with OCD (77). There appear to be significant problems in frontal lobe functions affecting visual-spatial integration, reasoning, and memory in studies performed on adults with chronic OCD (79). This does not hold true for adolescents or children, however. Flor-Henry and coworkers (80) studied 11 subjects with adult-onset OCD employing age- and sex-matched controls to find frontal deficits, especially in the dominant lobe. They proposed that associated dominant temporal and parietal dysfunction stemmed from the failure of frontal inhibitory responses. Others (81) could not confirm frontal lobe findings. Behar and coworkers (82) found signs of immaturity among 16 adolescent subjects on two measures reflecting frontal lobe function and visual-spatial tasks. The performance of six subjects on another test suggested “neurodevelopmental immaturity.” Neurological soft signs, such as synkinesia, were seen in five of the seven adolescents. Cox and coworkers (83) suggested deficits in regulatory functions localized to the frontal lobe, although they did not confirm dominance of the frontal hemisphere. Few measures supported right–left hemispheric differences. The group underscored that findings were independent of impairment from OCD, implying a stable deficit that was unrelated to severity. Beers and coworkers (84) examined 21 drug-naïve children with new onset OCD using a neuropsychological battery and found no abnormalities.

Imaging Studies

Structural Imaging Studies

Mataix-Cols and coworkers (50,51) employed an entirely novel approach when they posed aversive experiences to patients based on symptom profiles in an fMRI environment. The profiles were based on factor-analytic subgroups as discussed above. The results showed symptom-specific significant activation in brain regions based on clinical phenotype (50,51). Individuals with washing/contamination symptoms showed activation in ventromedial prefrontal cortex, left middle temporal gyrus, and anterior cingulate, while those with checking/aggression symptoms activated basal ganglia structures (globus pallidus, putamen) and dorsolateral prefrontal cortex (50). Individuals with hoarding symptoms showed activation in the left precentral, frontal gyrus, fusiform gyri, and right orbital-frontal regions (50). These studies show that employing undifferentiated groups would have failed to identify differences. The investigators went on to suggest that the findings lend support to a concept of OCD as multiple conditions with separate, overlapping abnormalities in neuropsychological function (45).

Imaging studies suggest that the function of the anterior cingulate gyrus (ACC) may have particular bearing on OCD. Van Veen and coworkers (85) and Ursu and coworkers (86) offered new insights when they proposed that the ACC is activated in response to incoming streams of information that are in conflict with other information or in response to errors. In an fMRI study, Ursu and coworkers (86) reported increased activation in ACC regions in persons with OCD compared to controls in response to conflicting information and when given a task that forced increased error rates. Participants with OCD performed as well as controls on the task, but there were substantial between group differences in ACC activation. In addition, levels of ACC activation were correlated with OCD severity (86).

Magnetic Resonance Imagining and Computerized Tomography (CT)

Overall, morphometric findings lend support to the hypothesis of impairments in cortico-striato-thalamo-cortical circuitry, with increased volumes in frontal regions (cingulate gyrus) and decreased volumes in striatal regions (especially globus pallidus) and increased volumes in thalamic regions (87). However, as will be seen below, the implication that OCD can be explained on the basis of a generic impairment in a single network of circuits has been challenged by fMRI studies (88).

Computerized Axial Tomography (CAT)

Views of patients with OCD or OCD whose onset was in adolescence suggested ventricular enlargement independent of sex, age, duration, and types of symptoms (82). Luxenberg and coworkers (89) employed quantitative scanning in males with adolescent-onset OCD and a never-ill control group. In the OCD cohort, decreased mean volumes were seen in caudate nuclei bilaterally. Values of subjects with OCD and healthy controls overlapped considerably, but pooled mean differences were consistent with a hypothesis of basal ganglia changes in OCD.

Structural MRI

Jenike and coworkers (90) studied 10 female patients with OCD and a control group using structural magnetic resonance imaging (MRI). Increased opercular and whole cortex volumes were noted, as well as a decrease of total
white matter. In a parallel study, Alward and coworkers (91) failed to find any differences between 24 adults with OCD when compared to a control group. Garber and coworkers (92) compared MRI scans of adults with OCD and healthy controls and reported abnormalities in frontal cortex, cingulate gyrus, and lenticular nuclei. Past medication treatment or family history did not significantly influence findings. Kellner and coworkers (93) did not discover significant structural differences between OCD subjects and normal controls, but Calabrese and coworkers (94) reported increases in the size of the caudate nuclei. A subsequent study reported left caudate nuclei volumes exceeded those on the right.

Rosenberg and Keshavan (95) proposed a neurodevelopmental view of OCD and performed MRI morphometric studies of drug-naïve children and age-matched healthy volunteers. They observed increased volumes only in the anterior cingulate and not in posterior cingulate, amygdala, dorsolateral prefrontal cortex, hippocampus, or temporal lobe. In addition, unlike healthy volunteers, the anterior cingulate volume did not increase with age (95). The investigators suggested that increased anterior cingulate volume might be an early marker for OCD.

Szeszko and coworkers (96) reported increased gray matter in the left cingulate gyrus and decreased volume in the left globus pallidus when they compared 23 drug-naïve children with OCD to pediatric healthy volunteers. OCD patients and healthy participants showed no between-group differences in caudate nucleus volumes (96). Also, drug naïve children with OCD treated with paroxetine, a highly selective serotonin reuptake inhibitor, showed reductions in caudate nucleus volumes after successful treatment (87). The decrement in caudate nucleus volume was correlated with clinical improvement (87).

Functional Studies Using Radioligands

Cerebral Blood Flow, Single Photon Emission Computed Tomography and Positron Emission Tomography (PET)

The general consensus is that exposing children to radioactivity for research that does not benefit them directly is unethical and thus there is only adult research data making use of this technology. These are powerful techniques that allow visualization of highly specific dopamine and serotonin transporter proteins and the specific regions in which they are active. These agents also can be measured rapidly, allowing for close temporal association to functional activity. These techniques have become more powerful when scanning is performed in conjunction with provocation maneuvers in populations that are stratified according to symptom dimensions, noted above. Taken together, these studies support findings of abnormalities in left orbital-frontal cortex and bilateral caudate nuclei. Data from more recent studies suggest different pathways of impairment based on symptom profiles with frontal-striatal regions being involved with ritualistic activity, the insula relating to disgust, and medial temporal regions being involved with anxiety.

Zohar and coworkers (97) employed cerebral blood flow and Single Photon Emission Computed Tomography (SPECT). They observed changes in regional cerebral blood flow (rCBF) with [¹³³Xe] xenon inhalation. Subjects who were stressed specifically to increase anxiety displayed decreased temporal rCBF. This pattern was sustained as the stressful stimuli were changed to induce higher levels of anxiety.

There are no published studies of SPECT in children with OCD. Van der Wee and coworkers (98), using 123-I-labeled 2-β-carbomethoxy-3-β-(4-iodophenyl)-tropane ([123-I]β-CIT), found increased dopamine transporter density in the left basal ganglia of drug-naïve adult patients with OCD compared to healthy volunteers. No differences in serotonin transporter densities were observed (98). Also, Stengler-Wenzke and coworkers (99) measured serotonin transporter density with SPECT using the same agent in adult patients with OCD and healthy volunteers. Patients with OCD had 30% reduction in binding to serotonin transporter in mid-brain and brainstem structures.

Other adult studies with SPECT report increases in medial-frontal rCBF (100,101). Rubin and coworkers (102) examined regional cerebral blood flow with xenon inhalation and technetium uptake. Xenon studies did not reveal differences between adults with OCD and normal controls, but with technetium uptake increased activity in the orbito-frontal cortex was observed, with significantly decreased bilateral activity in the head of the caudate (102), a phenomenon previously noted by others (101). Using SPECT, McGuire and coworkers (103) found increased orbital-frontal, striatum, putamen, globus pallidus, and thalamus rCBF among four patients with OCD when urges to perform rituals were the greatest. They also reported that hippocampus and posterior cingulate showed greater blood flow in response to anxiety surrounding these urges.

Cottraux and coworkers (104) studied patients with predominant checking and washing compulsions using SPECT. Compared to controls, those with OCD showed diminished rCBF at baseline in the putamen and thalamus and greater rCBF in the ACC. In response to provocation, there were substantial increases in rCBF in orbital-frontal and superior temporal regions.

Positron Emission Tomography (PET) has emerged as one of the most powerful tools to learn about OCD. The first wave of studies examined persons with OCD compared to controls (105,106,107,108,109,110). Taken together, these studies replicate findings of increased activity in orbital-frontal cortex (OFC) and caudate nuclei when patients with OCD are compared to healthy volunteers (111). Some, though not all, of these studies also measured increased activity in the anterior cingulate gyrus and thalamus (111).

The technology has expanded to include PET scanning but ascertaining or stratifying by dimensional subtypes and employing tasks that provoke symptoms. Rauch and coworkers (112) scanned individuals with OCD while they were provoked with a feared stimulus and found increased activation in the right caudate, orbital, thalamic and anterior cingulate gyrus. Similarly, Saxena and coworkers (113) compared individuals with symptoms of hoarding to OCD patients without hoarding and to healthy volunteers. Those with hoarding showed a differential pattern of lower glucose metabolism in posterior cingulate and cuneus regions, while individuals with nonhoarding OCD had elevated glucose metabolism in the thalamus and caudate nuclei bilaterally. Unique, large differences between those with OCD plus hoarding and OCD without hoarding were noted in the dorsal anterior cingulate (113).

Only gold members can continue reading. Log In or Register to continue