Clinical manifestations

Presentation and course

The DSM-5 has defined the diagnostic criteria for intermittent explosive disorder (312.34 [F63.81]) as (04):

Recurrent outbursts consequent to inability to control impulsive aggression. This can result in verbal or physical offensive behavior on average twice weekly over a three month period (without damage/destruction to property of more than trivial value or serious physical injury to others) or three outbursts causing such damage/destruction and/or serious physical injury to others over a one-year period.

Additional diagnostic criteria include:

• Disproportionate aggression relative to provocation; • Impulsive, angry act without premeditation; • Outbursts evoking significant consequences; • Patient chronological age of 6 years or older; • No better explanation for the aggressive outbursts.

These diagnostic criteria carry the caveat that the diagnosis may be concurrent with the diagnosis of attention deficit hyperactivity disorder, conduct disorder, oppositional defiant disorder, or autistic spectrum disorder if the aggressive outbursts exceed what is accepted within the criteria for these conditions and merits "independent clinical attention."

The clinical manifestations associated with intermittent explosive disorder include: rapid onset devoid of significant prodrome; episodes usually lasting less than 30 minutes duration without significant provocation; minor aggressive outbursts between more significant episodes; periodicity as defined above (twice weekly for 3 months with less serious outbursts or thrice yearly for more destructive episodes); lack of control over impulsive aggression with disproportionate provocation; impulsive or anger-based outbursts without premeditation; age 6 years or older (either chronologically or developmentally); and exclusion of those diagnosed with disruptive mood dysregulation disorder or in whom an alternative etiology is more persuasive. If impulsive aggression occurs in the context of an adjustment disorder in those aged 6 to 18 years, intermittent explosive disorder should not be diagnosed.

Because impulsive aggressive behavior appears in patients with many diagnoses, many clinicians are reluctant to make a diagnosis of intermittent explosive disorder in the presence of other psychiatric diagnoses. In fact, impulsive aggressive behavior is manifest in all humans early in life and before the onset of other psychiatric disorders. In the vast majority of people impulsive aggressive behaviors diminish over time, frequently well before adolescence (55). In the adolescent supplement to the National Comorbidity Survey Replication, lifetime DSM-4 intermittent explosive disorder by “narrow” criteria was 5.3% (44), similar to what was found in adults (31). Clinical studies suggest significant comorbidity of intermittent explosive disorder with mood disorders, anxiety disorders, and substance use disorders (09). In each case, with the exception of phobic anxiety disorders, the age of onset of intermittent explosive disorder is reported to be earlier than that of the comorbid disorder. This suggests independence of the disorders or that intermittent explosive disorder might be a risk factor for the comorbid disorder. A similar finding was found in a family history study of intermittent explosive disorder (07). Finally, some also claim that the diagnosis of intermittent explosive disorder should not be made in the presence of borderline, or antisocial, personality disorder. However, when examined empirically, levels of lifetime aggressive behavior among “borderline personality disorder/antisocial personality disorder only” individuals are markedly lower than those who also meet criteria for intermittent explosive disorder, indicating that both diagnoses should be made when criteria for both are met (08).

Prognosis and complications

Intermittent explosive disorder is a chronic condition that can begin as early as the first decade (though a diagnosis cannot be made until age 6). It peaks in the second decade and new cases rarely appear after the fourth decade. Severity of intermittent explosive disorder can vary but the average level of lifetime aggression is typically greater than 18 (n.b., maximum score for the Life History of Aggression is 25) on the life history of aggression assessment (10). By way of comparison, nonaggressive individuals without any psychiatric disorder (healthy controls) have average Life History of Aggression scores of about 5 whereas nonaggressive individuals with other mood, anxiety, and/or personality disorders (psychiatric controls) have average life history of aggression scores of about 9. Assessment of current aggressive behavior using the Overt Aggression Scale Modified for Outpatient Use yields even larger differences, with total aggression scores of about 2 for healthy controls, about 12 for nonaggressive psychiatric controls, and about 100 for individuals with intermittent explosive disorder (14).

Not surprisingly, individuals with intermittent explosive disorder experience significant psychosocial consequences as a result of recurrent, impulsive, aggressive behavior. First, global psychosocial function is reduced in intermittent explosive disorder compared with both healthy and psychiatric controls (33), as is quality of life experience even when differences in global psychosocial function are accounted for (49). Examined more closely, individuals with intermittent explosive disorder report greater dysfunction while at work and with family and friends (33). In addition, there are the primary effects of aggressive outbursts, the damage done, and the potential for litigation or criminal charges.

Clinical vignette

AB, a 31-year-old, single, Caucasian male came for evaluation because of history of several, recent, aggressive outbursts that led to his dismissal from his last 2 jobs in information technology. AB admitted to angry outbursts in response to frustration with “defects” in computer programs and with criticism from his supervisors advising him to be a “bit less detail-oriented” because his insistence on “getting it right” was making him less productive. Two weeks before evaluation AB had a substantial aggressive outburst with his boss that included screaming and threats to harm his supervisor. AB was removed by security and he was served with a temporary restraining order to stay away from his supervisor.

On exam, AB was well groomed and appeared very concerned about his current situation. He admitted to having a “temper” but said he never hurt anyone before and that he never threatened anyone before. The recent episode made him anxious about his future because, perhaps, he was getting worse. AB reported that he was always a bit anxious and that things had to be “just so” or he would become upset. He knew that his “quick temper” had led him to have difficulty retaining friendships so he sought out a career where he could work alone most of the time. He admitted to having previous periods where he was somewhat depressed but further interview revealed that he never had a full blown episode of any specific psychiatric disorder. His medical evaluation was unremarkable and there was no medical or pharmacological reason for him to be at risk for impulsive aggressive outbursts. AB admitted to more than 3 aggressive outbursts a week, though mostly not while at work. The outbursts were characterized by screaming and throwing things around. He denied ever hitting another person but did admit to breaking things worth more than $50 during his tantrums several times a year.

AB was started on a selective serotonin reuptake inhibitor and increased to maximal dose within 2 weeks. His aggressive outbursts started to decrease to about half of their original frequency in 2 months, at which point he began treatment with a clinical psychologist to do cognitive behavioral therapy focusing on his impulsive aggressive behavior. After 12 sessions he reported feeling less angry and less likely to be irritated by the things that used to “set him off”. He continued on the selective serotonin reuptake inhibitor and continued with cognitive behavioral therapy having “booster sessions” every month or 2. He was able to find work again and continues to be much improved, reporting that he only has an occasional outburst outside of work 1 year after treatment began.

Biological basis

Etiology and pathogenesis

The exact etiology of impulsive aggression in the context of intermittent explosive disorder remains to be defined, but clearly is multidetermined in nature with contributions from genetics, neurochemistry, and cortico-limbic function.

Aggression is under both genetic and environmental influence, with up to 50% of the variability in measures of aggression accounted for by genetic factors (45). Our own twin studies using aggression support this, showing genetic influence of the aggression increasing with the severity of the aggressive acts from 28% for verbal aggression to 45% for aggression against others. In addition, impulsivity is under similar genetic influence in these types of studies and the genetic correlation of impulsivity and aggression is correspondingly substantial, supporting the conceptualization of aggression as a form of dysregulation (51). Following from this, it is not surprising that intermittent explosive disorder runs in families of first-degree relatives of those with intermittent explosive disorder by about 3-fold (07).

From a psychobiological standpoint, aggression, particularly impulsive aggression, is associated with a reduction in central 5-HT function that is also observable in more distal indices of 5-HT function. For example, inverse correlations have been reported between measures of aggression and CSF 5-hydroxyindoleacetic acid (38), hormonal responses to 5-HT pharmaco-challenge (18), and platelet 5-HT transporter (5-HTT) binding (19). These findings are highly consistent with animal studies reporting inverse relationships between 5-HT and aggressive responding first reported in the literature in the 1960s. Other neurochemical systems may also be relevant in human aggression. At this time, there is preliminary data supporting positive relationships between aggression and catecholamines: norepinephrine (21), dopamine (37; 07), and other amines such as glutamate (23); peptides like vasopressin (15), substance P (22), and neuropeptide Y (20); and circulating cytokines such as interleukin-6 (54; 40; 17). There is similar data supporting an inverse relationship between aggression and oxytocin (35). However, with the exception of manipulating 5-HT as a strategy to reduce (27) or increase (06) impulsive aggression, few studies have been conducted to explore how manipulating non-5-HT systems reduces, or has no effect on, impulsive aggressive behavior in humans.

Positron emission tomographic imaging studies have extended these findings in impulsively aggressive personality disordered study participants, most of whom would meet criteria for intermittent explosive disorder. Frankle and colleagues reported reduced availability of the 5-HT transporter (5-HTT) in several cortical and subcortical regions that was significantly lowest in the anterior cingulate cortices of impulsively aggressive personality disordered individuals (28). Although another study did not replicate this finding (56) another positron emission tomography study labeling the 5-HT-2a receptor reported increased binding of 5-HT-2a receptors in the orbitofrontal cortex of impulsively aggressive personality disordered individuals (47).

Similarly, studies of individuals with intermittent explosive disorder note categorical differences in structural and functional imaging studies as would be expected from studies of aggression in general. Individuals with intermittent explosive disorder display a reduction in gray matter volume in fronto-limbic circuits (13), abnormalities in the shape of the amygdala and hippocampus (24), and reduced fractional anisotropy in the superior longitudinal fasciculus (34), all suggesting important structural deficits in critical emotion regulating areas of the brain. Despite the possibility of a smaller, less smooth amygdala, functional fMRI studies note hyperactivity of the amygdala to hostile social threat (ie, anger faces) in intermittent explosive disorder compared with heathy controls, with either unchanged or reduced activation of prefrontal cortical regions (25; 43). We found that although the amygdala exhibits a hyperactive response to anger faces, amygdala responses to induced emotion are not different in intermittent explosive disorder compared with healthy controls (24a). This suggests that heightened amygdala activity to hostile social threat may be an endophenotype for intermittent explosive disorder. Finally, we have found important differences in fMRI responses to videos displaying socially ambiguous situations in which 1 individual experiences a potentially aggressive event. In these studies we have found that individuals with intermittent explosive disorder display multiple deficits in cortico-limbic activation as a function of the different steps in social-emotional information processing (11). Specifically, compared with healthy controls, individuals with intermittent explosive disorder display reduced activation of medial prefrontal and anterior cingulate cortices when watching aggressive versus nonaggressive videos, reduced activation of dorsolateral prefrontal cortex and at the temporoparietal junction while evaluating if the person in the video was hostile in their actions, and reduced activation of superior prefrontal cortex and the periaqueductal gray when estimating how angry they would be if the potentially aggressive event happened to them.

Epidemiology

Before the advent of the DSM-4, intermittent explosive disorder was thought to “rare”. This is because the DSM criteria excluded most individuals with recurrent impulsive aggressive outbursts because intermittent explosive disorder was thought to be a disorder in which someone was otherwise “fine” but that occasionally they had extreme outbursts of aggression (eg, “Bruce Banner” vs. “The Hulk”). With DSM-4, and especially with DSM-5, intermittent explosive disorder appears to be much more common than previously thought. The National Comorbidity Study Replication reported a lifetime prevalence of DSM-4 intermittent explosive disorder in the United States of 7.3% by “broad criteria” and 5.4% by “narrow criteria,” and past year prevalence as 3.9% and 2.7%, respectively (31). Inspection of the data reveals meaningful differences between the 2 intermittent explosive disorder types, with “narrow” intermittent explosive disorder being far more severe than “broad” intermittent explosive disorder (08). “Broad” intermittent explosive disorder stipulates only 3 aggressive outbursts during a lifetime whereas “narrow” intermittent explosive disorder requires at least 3 aggressive outbursts in a single year. This prevalence rate may not be true in non-U.S. countries, however, where the lifetime estimate of DSM-4 intermittent explosive disorder appears to range from only 0.1% (ie, Nigeria) to 1.2% (ie, Columbia, South Africa) averaging 0.6% compared with 2.7% for the United States (50). Reasons for this cross-national difference are not known with any certainty but are likely related to variation in general risk factors, cultural factors that impact on willingness to disclose information about one’s own psychopathology, and other methodological factors. A separate analysis of the U.S. data suggests the lifetime rate of DSM-4 intermittent explosive disorder in the U.S. is 3.6% (12), down from the 5.4% in the original report (31). In addition, a survey study of over 5300 nondeployed army service members (Army STARRS) reported lifetime pre- and post-enlistment diagnoses of DSM-4 intermittent explosive disorder of 3.7% and 3.8%, respectively (46). Despite these data, the community rate of DSM-5 intermittent explosive disorder is not known. This is because DSM-5 criteria for intermittent explosive disorder require either low intensity but high-frequency impulsive aggressive outbursts (A1 criteria) or high intensity but low frequency outbursts (A2 criteria). Because the data instrument used in the National Comorbidity Study Replication(31) and the multinational (50) study only sought information about A2, not A1, type of impulsive aggressive outbursts, the prevalence of those meeting only the A1 criteria in the National Comorbidity Study Replication sample is unknown. New data from our own studies suggest that more than one fifth of those with current intermittent explosive disorder meet the A1 criteria only (had only frequent nondestructive/injurious outbursts) but not the A2 criteria for DSM-5 intermittent explosive disorder (Coccaro et al submitted). If so, one might estimate the lifetime prevalence of DSM-5 intermittent explosive disorder in the United States as 3.3% (50) to 4.4% (12), until another community survey is done assessing the epidemiology of DSM-5 disorders.

Prevention

Factors recognized to be associated with intermittent explosive disorder: a family history of intermittent explosive disorder (07) and presence of certain personality disorders such as paranoid personality, obsessive-compulsive personality, antisocial personality, borderline personality, and narcissistic personality.

Differential diagnosis

Individuals who present with history of recurrent, problematic, impulsive aggressive outbursts should be assessed for any medical or other psychiatric reason for their behavior. This includes, for example, cerebral tumors, dementia, drug effects, and the like.

Those with a history of physical or emotional trauma before 20 years of are at increased risk of intermittent explosive disorder (04). Other factors relevant to such damage must be considered in the differential diagnosis. The DSM-5 stipulates that a diagnosis of intermittent explosive disorder should not be considered if the aggressive outbursts only occur within the context of another psychiatric illness. Similarly, it should not be diagnosed in children or adolescents aged 6 to 18 years if the outbursts occurred during an adjustment disorder. Other confounding diagnoses include: disruptive mood dysregulation disorder; delirium, major neurocognitive disorder, and personality change due to other medical condition (aggressive type); substance intoxication or substance withdrawal; and attention deficit hyperactivity disorder, conduct disorder, oppositional defiant disorder or autism spectrum disorder.

A number of comorbidities, which occur in conjunction with intermittent explosive disorder, are identified in DSM-5 (04) and include: depressive disorders, anxiety disorders, substance use disorders, eating disorders, and individuals with other diagnoses involving disruptive behavior (such as attention deficit hyperactivity disorder, conduct disorder, or oppositional defiant disorder). The critical issues in each of these cases is that individuals with comorbid psychiatric disorders who also meet DSM-5 criteria for intermittent explosive disorder are typically more aggressive than those with these comorbid disorders alone, in which case both disorders can be diagnosed.

Diagnostic workup

There are no diagnostic tests that define intermittent explosive disorder. Medical history and examination should be directed at uncovering the presence of any medical (or pharmacologic) condition that could account for recurrent, problematic, impulsive, aggressive behavior in intermittent explosive disorder.

Management

Various medications can be used to treat impulsive aggression. The best evidence exists for selective serotonin reuptake inhibitors, which have been shown to reduce impulsive aggressive behavior in several, double-blinded, placebo-controlled studies (Coccaro et al 2009; 53; 29). Evidence also exists for the potential efficacy of mood stabilizer/anticonvulsant agents such as lithium (52), carbamazepine (26), phenytoin (05) divalproex (30), and oxcarbazepine (41). Atypical antipsychotic agents, such as quetiapine (32), olanzapine, and risperidone (48) have also been shown to have potential therapeutic benefit.

Nonpharmacological remedies, such as multicomponent, cognitive behavioral therapy, have also been studied for intermittent explosive disorder (42), as has deep brain stimulation of the orbital-frontal projections (39).