CHICAGO— The concept of a disorder spectrum may seem stretched beyond the bounds of clinical utility when it encompasses such seemingly disparate disorders as autism, body dysmorphic disorder, and pathological gambling. But many experts believe that each of these diagnoses merits a place in the obsessive-compulsive spectrum, not only because all of the disorders involve repetitive thoughts and behaviors but also because recent evidence suggests that they share neurobiologic origins.

"The idea is that the same genetic mechanisms or circuits that drive the repetitive behaviors in one disorder also play a role in the other disorders," explained Eric Hollander, MD, at the annual meeting of the American Psychiatric Association. "We can think about these disorders as being different phenotypic expressions of the same underlying genotypic difficulties." The result, he noted, is that "this symptom dimension cuts across traditional diagnostic boundaries."

Reviewing the clinical features and neurobiology of the obsessive-compulsive spectrum, Dr. Hollander, who is Professor of Psychiatry at Mount Sinai School of Medicine in New York City, classified the disorders into three subgroups:
• Those involving excessive preoccupation with body appearance or sensation, such as body dysmorphic disorder or anorexia nervosa.
• Neurologic disorders that present with repetitive behaviors, such as Tourette syndrome, Sydenham's chorea, and autism.
• Repetitive behaviors driven by pleasure or arousal, such as sexual compulsions, kleptomania, trichotillomania, and pathological gambling.

BODY OBSESSIONS

Individuals with a disorder from the first subgroup "are obsessed and preoccupied with imagined defects in their appearance," Dr. Hollander said. "They often have cosmetic surgery or perform other rituals to try to change their appearance." The nature of the obsession varies; a patient may be obsessed with some aspect of his face or head, be convinced that he has a foul body odor, or be overly concerned with the size of his body or of individual parts. The result may be any of a variety of repetitive behaviors, including constant mirror checking, the avoidance of social situations, frequent doctor visits and surgeries, and efforts to conceal the perceived defect. About half of patients have one or more delusional convictions; such delusions are rare, however, in obsessive-compulsive disorder (OCD).

Recent neuroimaging studies have revealed striking similarities—but also some important differences—between body dysmorphic disorder and OCD, Dr. Hollander reported. Positron emission tomography (PET) indicates that patients with body dysmorphic disorder, like OCD patients, tend to have increased metabolic activity in the anterior cingulate region; this overactivity may trigger patients' "heightened sense that something doesn't look right or that something bad is going to happen."

And as in OCD, individuals who have increased metabolic activity in the anterior cingulate region are the ones who have the most robust response to selective serotonin reuptake inhibitors (SSRIs). "This sort of finding—that anterior cingulate activity is a predictor of SSRI response—seems to cut across different disorders and occurs not only in OCD and body dysmorphic disorder but also in autism and depression," Dr. Hollander noted.

However, unlike OCD patients, individuals with body dysmorphic disorder have decreased metabolic activity in the orbitofrontal cortex. This finding is consistent with the relatively poor insight and high rates of comorbid depression found in these patients.

SEROTONERGIC SHOWDOWN

In a controlled crossover trial published last year in the Archives of General Psychiatry,Dr. Hollander and colleagues compared the effects of a serotonergic tricyclic antidepressant (clomipramine) with a noradrenergic tricyclic (desipramine). "On all of the primary outcome measures, there was a significant reduction in the obsessive preoccupations with the imagined ugliness—and fewer urges to do the rituals to conceal the defect—with the serotonin reuptake inhibitor," according to Dr. Hollander.

Patients who initially received desipramine generally responded only minimally but had significant improvement when crossed over to clomipramine; conversely, those switched from clomipramine to desipramine tended to relapse. "This is the same pattern we see in OCD," Dr. Hollander noted.

Interestingly, "those patients with the greatest delusional conviction are the ones who had the best response to serotonin reuptake inhibitor," he added. Moreover, successful treatment of OC symptoms also reduced delusional convictions. "This is the first time this has been reported—that psychotic or delusional symptoms respond to an SRI," Dr. Hollander said. "This makes us think that the delusional conviction is really secondary to the obsessive preoccupation in this illness, and not really a separate disorder."

AUTISM: THE CINGULATE STRIKES AGAIN

The second subgroup of OC spectrum disorders, those involving neurologic diagnoses, is represented by autism, which is characterized by striking social deficits, speech and language problems, and—of relevance here—repetitive behaviors and a narrow range of interests. "We think of this disorder as having multiple, different genetic components, each coding for a different symptom dimension," Dr. Hollander said.

There does appear to be a substantial genetic component to the OC symptoms in autism. First-degree family members of autistic individuals have high rates of OCD, and they often have subtle social deficits and language problems, Dr. Hollander noted. Rates of OCD and OC symptoms are highest among first-degree relatives of autistic probands with severe repetitive behaviors.

In crossover trials comparing fluoxetine and placebo, the SSRI was effective in reducing global autism severity; the improvements may have been due to reductions in repetitive behavior and perhaps impulsive/aggressive symptoms. Individuals with the greatest metabolic activity in anterior inferior regions of the cingulate had the most severe repetitive behaviors, but also the best response to fluoxetine, "so again anterior cingulate activity may be a predictor of SSRI response across different disorders," Dr. Hollander said.

DIVERSE BIOLOGICAL MARKERS

Several studies suggest that there are deficiencies in serotonin synthesis and release in persons with autism, particularly in autoreceptor-rich cortical and thalamic regions. These studies lead Dr. Hollander and colleagues to hypothesize that 5-HT1D autoreceptors might play a role in symptom genesis. Indeed, subjects with the most sensitive 5-HT1D autoreceptors exhibit the most severe repetitive behaviors, the researchers found.

In addition, Dr. Hollander and colleagues reported last year that the B lymphocyte antigen D8/17—which is associated with an increased risk of developing rheumatic fever and is overexpressed in several disorders in the OCD spectrum—is far more likely to be expressed by the B cells of autistic subjects (78%) than those of medically ill matched controls (21%). Again, autistic persons with the most severe OC symptoms were especially likely to express this marker; this fact suggests that D8/17 is more related to the particular symptom dimension of repetitive behaviors than to autism per se, Dr. Hollander noted. He added that among family members of autistic individuals, those who have this marker have the highest levels of OCD or OC symptoms.

PATHOLOGICAL GAMBLING

The third cluster of disorders in the OCD spectrum includes pathological gambling and other impulsive disorders, which are driven by pleasure, arousal, and gratification. "These patients have a hard time putting the brake on their impulses, and all kinds of functional impairment result," Dr. Hollander explained. Many have bipolar spectrum problems or attention-deficit/hyperactivity disorder and are not getting optimal levels of stimulation; this leads them to pursue risky activities such as gambling.

Studies have implicated the serotonergic, noradrenergic, and dopaminergic systems in pathological gambling. Further clues to the neurobiologic origins of the disorder come from studies using PET to examine brain circuitry in pathological gamblers performing a computerized blackjack task. When the subjects are playing the game but not betting, there is "relatively equal frontal and occipital activation," Dr. Hollander reported. However, when the gambling task involves risk and reward—players can double their money or lose $100—subjects tend to have greater activation of visual cortical regions; similar activation has been described in alcoholics who encounter cues to alcohol.

A HIGH PLACEBO RESPONSE RATE

Although researchers are beginning to explore the effects of 5-HT2 antagonists and opiate antagonists in these patients, other studies have focused on the SSRIs and lithium. In one fluvoxamine study, patients received placebo for eight weeks to stabilize baseline findings and were then switched to the active drug. After fluvoxamine therapy, the subjects "had a significant reduction in the intensity and frequency of thoughts about gambling, and they were better able to delay or inhibit acting on the impulses," Dr. Hollander reported. About 70% became abstinent; the remainder had an early response but were unable to sustain it. All of the latter had mild mood-cycling problems and "probably would have benefited from a mood-stabilizing drug such as lithium," he said.

Indeed, it is important to consider patient subtypes when treating pathological gamblers. For gamblers with bipolar spectrum problems, "you might want to consider beginning with a mood stabilizer," Dr. Hollander suggested; his group is currently assessing the benefits of this approach. For nonresponders with comorbid ADHD, clinicians should consider adding an agent with dopaminergic/noradrenergic effects. In general, the best approach is to start with an agent with documented benefits, and if that fails, to try an agent appropriate for the patient's comorbidity.

Dr. Hollander cautioned, however, that achieving a sustained biological response may take time. In one study, patients received either an SSRI or placebo and were then crossed over to the other treatment. Both groups initially showed substantial improvements, suggesting that "in highly impulsive populations, there is a relatively high early placebo response rate," Dr. Hollander noted. But in the second phase, subjects crossed over to SSRI maintained their early response and even improved over time; those crossed over to placebo relapsed. "So it took a long time to see this significant difference between the active drug and placebo," he said.

IS THE SPECTRUM TOO WIDE?

Summing up, Dr. Hollander emphasized that the disorders described in his talk not only share important clinical features with OCD but have other important similarities as well, such as a response to SSRIs, comorbidity patterns, and involvement of particular brain regions.

However, a recent report suggests that if one considers a shared genetic component to be a crucial criterion for the inclusion of a disorder in the OC spectrum, several putative members may be distant relatives at best.

In a study published in the August 15 issue of Biological Psychiatry,O. Joseph Bienvenu, MD, and colleagues at The Johns Hopkins School of Medicine examined the frequency of various OC spectrum disorders in 80 individuals with OCD and 343 first-degree relatives, as well as in 73 controls and 300 of their first-degree relatives. The researchers conducted diagnostic interviews with all subjects, generally in person, using a version of the Schedule for Affective Disorders and Schizophrenia that had been modified to include the Yale-Brown Obsessive Compulsive Scale (Y-BOCS) and related material; age-appropriate instruments were used for the 47 child and adolescent participants. Expert psychiatrists reviewed all available material and assigned diagnoses to the subjects.

Not surprisingly, the lifetime prevalence of any "spectrum" disorder was higher in probands with OCD (57%) than in controls (21%). However, only for a few individual disorders—notably hypochondriasis, body dysmorphic disorder, and pathological skin picking—did the difference in rates between probands and controls reach statistical significance. The findings were similar, though even less robust, for first-degree relatives; although family members of OCD probands were more likely than relatives of controls to have any OC spectrum disorder (25% vs 15%), the only individual disorder for which the difference was significant was body dysmorphic disorder.

Combining the OC spectrum diagnoses into subgroups revealed higher rates of somatoform disorders (body dysmorphic disorder, hypochondriasis) and grooming disorders (pathological skin biting or skin picking, trichotillomania) in relatives of OCD cases than in relatives of controls. However, the prevalence of eating disorders (anorexia nervosa, bulimia nervosa) and impulse control disorders (pathological gambling, kleptomania, pyromania) did not significantly differ between the two groups. "This does not rule out the possibility of familial co-transmission of these disorders," Dr. Bienvenu and colleagues wrote, "but it does suggest that these conditions are not prominent parts of the familial OCD spectrum, despite phenomenologic and other similarities." In fact, they noted, generalized anxiety disorder and other psychiatric conditions "appear more substantially related to the familial OCD spectrum than [are] many 'obsessive-compulsive spectrum' conditions."

—Peter Doskoch

Suggested Reading
1. Bienvenu OJ, Samuels JF, Riddle MA, et al. The relationship of obsessive-compulsive disorder to possible spectrum disorders: results from a family study. Biol Psychiatry.2000;48:287-293.
2. Hollander E. Treatment of obsessive-compulsive spectrum disorders with SSRIs. Br J Psychiatry Suppl.1998;35:7-12.
3. Hollander E, Allen A, Kwon J, et al. Clomipramine vs desipramine crossover trial in body dysmorphic disorder: selective efficacy of a serotonin reuptake inhibitor in imagined ugliness. Arch Gen Psychiatry.1999;56:1033-1039.

Return to table of contents