SIR: Few studies have documented gross structural abnormalities in posttraumatic stress disorder (PTSD). In one magnetic resonance imaging (MRI) study, a developmental variant known as cavum septum pellucidum was found in half of PTSD subjects but in only 14% of control subjects.1 In another study, gross abnormalities (cortical atrophy, focal white matter lesions) were reported in PTSD subjects (10 of 42 subjects), but not in control subjects.2 Of particular interest has been the finding of reduced hippocampal volume in PTSD, and a question that arises is whether hippocampal injury is an outcome of the disorder or is a preexisting vulnerability to its development.
This report briefly highlights MRI findings of mesial temporal sclerosis (MTS) in a 16-year-old female with PTSD.
Ms. A., age 16 years, presented with PTSD after being raped by her grandfather at age 6 years. Since the age of 10, she had made several attempts to injure herself (cutting her wrists and breasts, overdoses of pills). Of note on medical examination was an ability to dissociate readily and a history of several dissociative episodes per day in the past year. During exploration of the rape, she appeared to go into a light trance. She reported vivid flashbacks, nightmares, irritability, and insomnia. There were no impairments in short-term or long-term memory. She had no history of epilepsy or other medical illness and no history of drug use.
Ms. A. did not improve with cognitive-behavioral therapy. There was also little improvement with various medications: (i) citalopram 40 mg/day+risperidone 4 mg/day; (ii) fluoxetine 60 mg/day+quetiapine (Seroquel) 400 mg/day+valproic acid 100 mg/day; (iii) fluoxetine 60 mg/day+lamotrigine 200 mg/day+risperidone 3 mg/day; and (iv) fluoxetine 80 mg/day +quetiapine 600 mg/day +valproic acid 700 mg/day. Side effects included extrapyramidal effects (risperidone 3—4 mg/day), akathisia (fluoxetine 60—80 mg/day), and galactorrhea (quetiapine 300—600 mg/day). On a treatment regimen of fluoxetine 20 mg/day+quetiapine 200 mg/day+valproic acid 700 mg/day, there was some improvement in flashbacks and self-mutilation. In view of her frequent dissociations, electroencephalography (EEG) and MRI were requested. EEG and Holter-EEG were normal. MRI displayed mesial temporal sclerosis: right hippocampal atrophy and amygdala asymmetry.
MRI findings in adult victims of child physical/sexual abuse3,4 and in combat veterans5 suggest that PTSD patients exposed to early life trauma have greater left hippocampal volume reduction, while those exposed to later life trauma (combat veterans) have bilateral or right hippocampal atrophy. In sexually abused women with PTSD, a correlation has been found between the degree of atrophy and dissociative symptoms.
One can hypothesize that in some cases of PTSD, psychological trauma predates neuronal changes but that in other cases, underlying neuronal dysfunction may predispose to its development. These processes are not mutually exclusive, and some combination of the two may be useful in explaining individual variations in trauma response. Although MTS has long been recognized as a common pathologic finding in patients with drug-resistant temporal lobe epilepsy, its etiology remains unknown. Hypotheses have included infectious etiologies and defects in energy metabolism. In a recent MRI study, the lesion was documented in patients with psychogenic nonepileptic seizures.6 Histologically, MTS has been characterized by atrophy and gliosis of the hippocampus. Administration of epileptogenic agents has been shown to cause MTS and recurrent seizures in animals, perhaps on the basis of excessive release of excitatory amino acids such as glutamate. Interestingly, Ms. A. had no response on lamotrigine, a drug with antiglutamatergic properties.
The question of a relationship between MTS and PTSD cannot be answered by the data at hand. Nevertheless, we present this case to underscore the possible link between gross neurological lesions, more subtle impairments, and PTSD. Ultimately a comprehensive understanding of PTSD is likely to require more fine-grained analysis of dysfunction, including work at the molecular level. In the interim, the anomaly documented here is an interesting one to explore in patients with PTSD.