Disinhibition Due to Disruption of the Orbitofrontal Circuit Treated Successfully With Carbamazepine: A Case Series

T he frontal-subcortical circuitry controls human behavior by providing a connection between the frontal lobes and their subcortical counterparts. Injury to any segment of the circuit can disrupt the entire pathway, producing various neuropsychiatric symptoms. Injury to the orbitofrontal circuit may produce behavioral disinhibition. 1 , 2 It is possible that the antikindling properties of carbamazepine may play a role in its ability to treat these symptoms. 3 We present three cases of disinhibition and aggression in patients with neurologic insults which improved after treatment with carbamazepine.

CASE REPORTS

Patient 1

Our patient is a 57-year-old Caucasian man with a history of seizure disorder treated with levetiracetam, 500 mg p.o. b.i.d., and valproic acid, 500 mg p.o. b.i.d. Prior to hospitalization, he was living at a long-term care facility due to poor medication compliance. He was brought to the emergency department after becoming extremely agitated with the facility staff.

Computerized tomography (CT) of the head revealed a lacunar infarct of the left internal capsule which was not seen on a CT performed 2 months prior. His blood alcohol level was 0.0 on admission. A normal awake EEG did not demonstrate slowing or epileptiform activity. Physical examination and laboratory data were unremarkable except for a temperature of 101.8°F and a WBC count of 12.5K/mm ³ , respectively. He was empirically treated with ceftriaxone and his fever and WBC count attenuated, but he continued to be agitated. The agitation was treated by physical restraints and lorazepam, 1 mg i.v. every 4 hours as needed, until we were consulted 5 days later.

The patient also had several instances of sexual disinhibition, in which he attempted to masturbate in the presence of hospital staff. Our service was consulted at this time.

Our initial evaluation was complicated by the patient’s disinhibition, and unfortunately a “baseline” cognitive examination could not be performed. He was oriented only to person. The patient denied hallucinations or delusions.

He was initially treated with aripiprazole, 9.75 mg i.m. b.i.d. for 12 days, which did not stabilize his behavior. Subsequently, aripiprazole was discontinued in lieu of carbamazepine, 200 mg t.i.d., 17 days after he was initially admitted. His carbamazepine levels were 10, 9, and 7 mcg/ml, at weeks 1, 2, and 4, respectively. This was also the maintenance dosage for this patient. Two days after beginning treatment with carbamazepine, the staff was able to successfully remove his restraints. There were no further behavioral problems and his mental status gradually improved. The patient initially scored a 15 on the disinhibition section of the Neuropsychiatric Inventory, 4 but was discharged with a score of 2. By discharge, our patient was oriented times three and was able to recall three out of three words immediately and at 3-minute recall, but his working memory continued to be poor. He was discharged on carbamazepine, and at 2-month follow-up continued to be behaviorally stable.

Patient 2

Our patient is a 42-year-old white male who was admitted to the hospital status-post a closed head injury. A brain CT scan indicated a left frontoparietal subdural hematoma with evidence of subarachnoid hemorrhage and midline shift. After neurosurgical intervention, our patient’s condition stabilized, and a repeat brain CT scan indicated no new hemorrhages.

Behaviorally, our patient’s hospitalization was characterized by disinhibition, starting 2 weeks after admission. He scored a 16 on the disinhibition section of the Neuropsychiatric Inventory, including self-removal of his small bowel feeding tube, requiring intermittent restraints and sedation. Additionally, he was sexually inappropriate and verbally abusive toward staff, which interfered with his rehabilitation. Our team was consulted at this time.

On initial evaluation, our patient denied dysphoria, psychoses, or suicidal/homicidal ideation. Cognitively, our patient was oriented times one, his attention span was grossly impaired and while he could recall three out of three words immediately, on delayed recall he was unable to list any of the three words, even with cues.

Our patient was started on a treatment of carbamazepine, at the time of the initial consultation, two days after admission. His treatment started with carbamazepine, 200 mg p.o. b.i.d., and 5 days later the dosage was increased to maintenance of 400 mg p.o. b.i.d. Blood levels ranged from 6–7 mcg/ml. The result was a rapid decrease in his disinhibition and aggressiveness, such that he was participating appropriately in rehabilitation within 4 days of initiating carbamazepine; however, after 10 days of gradual hyponatremia, carbamazepine was discontinued as his medical team felt that carbamazepine was inducing syndrome of inappropriate anti-diuretic hormone secretion. Three days after discontinuing carbamazepine, aggressive and disinhibited behavior returned. After consultation with his primary care team, carbamazepine was restarted at 400 mg p.o. b.i.d., along with demeclocycline and free water restrictions to maintain sodium levels within normal limits. From a behavioral vantage point, his disinhibited behavior normalized after 2 days of carbamazepine treatment. Interestingly, our patient’s carbamazepine level after 2 days was subtherapeutic at 3 mcg/ml. After an additional 2 days, his carbamazepine dosage was increased to 300 mg p.o. t.i.d., which was his discharge dosage. His discharge carbamazepine serum level was 8 mcg/ml level. Our patient’s final disinhibition section score on the Neuropsychiatric Inventory was 2. Unfortunately, we were unable to follow-up.

Patient 3

Our patient is a 59-year-old white female who was admitted to the hospital for a subarachnoid hemorrhage involving the left anterior cerebral artery status-post aneurysm rupture. After coiling of the lesion, she had rebleeding. Her past medical history is significant for hypertension. She had no prior psychiatric history, although described as an “introvert” by the family prior to the subarachnoid hemorrhage. Her medications on admission included lisinopril.

Two weeks following rupture, we were consulted due to “bizarre behavior.” She began to develop signs of disinhibition, with inappropriate laughter, distractibility, and “elated” mood. Her physical examination was unremarkable except for right-sided hemiparesis with hyperactive ([3/4]) reflexes and ataxia.

On mental status examination, our patient had a notably elevated mood. She was significantly disinhibited verbally, with inappropriate laughter and sexual innuendos. The patient scored a 12 on the disinhibition section of the Neuropsychiatric Inventory. Cognitive examination was unremarkable except for a poor attention span. She denied psychoses or suicidal/homicidal ideation.

Serologic and urine evaluations were within normal limits. An EEG showed no generalized slowing or seizure activity. A CT scan of the brain demonstrated evidence of subarachnoid hemorrhage over both cerebral convexities with right being greater than the left. Additionally, right frontal subcortical white matter infarcts were noted.

Carbamazepine treatment was started at 200 mg p.o. t.i.d., and after 3 days was increased to 400 mg p.o. b.i.d. Her carbamazepine serum level after 5 days was 7 mcg/ml. After 1 week of carbamazepine therapy, our patient scored a 1 on the disinhibition section of the Neuropsychiatric Inventory. She remained in the hospital for an additional 3 weeks treated with carbamazepine, 400 mg p.o. b.i.d. Her carbamazepine serum level at discharge was 5 mcg/ml, and her carbamazepine dose was 400 mg p.o. b.i.d.

On 3-month follow-up, according to the nursing home staff, the patient’s carbamazepine was discontinued after 2 months at the nursing home. They reported that 1 month after the carbamazepine had been stopped, the patient no longer demonstrated sexual and verbal disinhibition.

DISCUSSION

Disinhibition syndromes, ranging from mildly inappropriate social behavior to mania, may result from lesions to specific brain areas, especially the orbitofrontal cortex (OFC). These patients may show different types of disinhibition: motor (e.g., hyperactivity, pressured speech), instinctive (hypersexuality, aggressive outbursts), emotional (euphoria, irritability), intellectual (grandiose and paranoid delusions), and/or sensory (visual and auditory hallucinations). 5

The orbitofrontal circuit provides a connection between the frontal cortex and subcortical structures and controls socially appropriate behavior. The circuitry includes the orbitofrontal cortex (OFC), striatum (ST), globus pallidus (GP), and thalamus (TH). This circuit, which projects as follows: OFC→ST→GP→TH→OFC, is considered to be an extension of the limbic system. A lesion involving this circuit, either in a specific region or its axonal connections between regions (i.e., white matter), may present with symptoms such as aggression, behavioral disinhibition, and emotional lability. 2

Each of our patients had insults to the proposed orbitofrontal circuit. In our first patient, a lesion to the internal capsule could potentially disrupt this circuit at the level of the orbitofrontal cortex projecting to the striatum. 1 In our second patient, the frontoparietal subdural hematoma/subarachnoid hemorrhage could have potentially resulted in direct injury to the orbitofrontal cortex as it lies directly rostral to the orbital roof making it a common site for injury during a closed head injury. 6 Finally, in our third patient, an extensive subarachnoid hemorrhage involved both cortices and right frontal subcortical white matter.

Surprisingly, while disinhibition traditionally has been associated with right hemispheric lesions involving the orbitofrontal circuit, 7 two of our three patients had lesions in the left hemisphere, although the closed head injury in our second patient resulted in a rightward midline shift. Additionally, while right hemispheric lesions increase the risk of disinhibition, this brain asymmetry is not absolute. 5

In addition to the potential pathophysiology of disinhibition that we have reviewed, we would also like to discuss our patients’ responses to carbamazepine. Carbamazepine is approved for the treatment of acute mania and mixed episodes of bipolar disorder. 8 Carbamazepine purportedly acts by blocking sodium channels; 9 however, the molecular mechanism underlying the action of carbamazepine in mood-stabilization is unknown, but three targets have been speculated. 10 Furthermore, carbamazepine is a potent antikindling anticonvulsant. 11

The theory of kindling has long been used to explain the pathophysiology of epilepsy and the efficacy of anticonvulsants. Animal studies have shown that repeated low-level stimulations of the amygdala will eventually evoke seizure activity. Stimuli that previously would not have been able to evoke seizures are now able to do so at the new lowered threshold, with seizures eventually occurring spontaneously in the absence of any stimulus. Anticonvulsants are thought to prevent the seizures by preventing the kindling effect. 3

Kindling refers to a highly persistent modification of brain functioning in response to repeated application of electrical stimulation that results in the development and spread of seizure activity and has been previously described in the visual, somatosensory and motor cortex of the cat. Kindling has been shown to result in long-lasting enhancement in synaptic efficacy and unit activity, and recent evidence suggests that kindling is capable of substantially altering organization in motor cortex of the rat. It was reported that neocortical as well as limbic kindling resulted in reorganization in the caudal forelimb area of the motor cortex with a substantial expansion (doubling) of the cortical area capable of eliciting forelimb movement. 12

This reorganization in rat motor cortex resulted in a change in the balance between excitation and inhibition. That is, kindling “unmasked” previously silent N-methyl-D-aspartate receptors, eventually resulting in an excitatory state. While never being evaluated (as far as we are aware), based on this reorganization theory, kindling may be able to explain potential reorganization of prefrontal cortex, transforming it to an excitatory state, manifested phenomenologically as disinhibition. 12

It has been postulated that this same concept of kindling may be applicable to behavioral changes such as irritability and aggression. Barratt 13 speculated that impulsive anger may be attributed to subseizure neural activity and may share similar neural pathways with seizure disorders. He defines impulsive anger as a defect of adaptive information processing, in which the patient is unable to respond to normal social cues. This type of behavior is often seen in patients with orbitofrontal lesions. 13 Carbamazepine’s antikindling properties make it an interesting treatment option for behavioral changes secondary to orbitofrontal circuitry lesions. 7 , 14 Other studies have also demonstrated carbamazepine’s efficacy in treating behavioral disturbances due to dementia, 15 , 16 frontal brain lesions, 17 and traumatic brain injury. 18

Furthermore, manic syndromes have been reported with bilateral asymmetric temporal lobe lesions. The buildup of subclinical epileptic activity in the anterior temporal lobes and the purported enhancement of mesocorticolimbic dopaminergic transmission may have had pathophysiological importance. By analogy, chronic disinhibition symptoms could develop from subclinical episodic causes which may result in enhancement in key neurotransmitter (i.e., dopamine) transmission, manifesting clinically after a CNS lesion, such as a cerebrovascular accident. 19

Our first patient was already being treated with levetiracetam and valproic acid prior to the development of his lacunar infarct, yet he still developed symptoms of disinhibition. Based on our proposed theory of kindling involvement in the manifestation of disinhibition, it would be tempting to conclude that levetiracetam and valproic acid are weaker antikindling agents than carbamazepine. Nonetheless, all three of these anticonvulsants have antikindling effects, 20 , 21 and we were unable to locate any reference that compared the antikindling effects of these three anticonvulsants.

While our patients’ improvement after being treated with carbamazepine suggests that the kindling model may have some role in the pathophysiology of orbitofrontal syndrome, a significant limitation in our theory is that the relatively short time between our patients’ cerebrovascular insults and symptom onset may indicate another possible neurobiological underpinning for these symptoms. Indeed, the kindling hypothesis warrants repeated low-level stimulation.

Additionally, explanations other than effects of kindling might better account for the greater efficacy of carbamazepine over other anticonvulsants. Foremost is carbamazepine has more purported mechanisms of action than other anticonvulsants, including its effects on multiple neurotransmitter systems (including dopamine) and second messengers systems, thus, potentially offering a variety of “neural signaling” explanations for its enhanced efficacy. 22 Nonetheless, we feel that further research into this theory and its treatment with carbamazepine is warranted.