We discuss two patients who were admitted to acute psychiatric inpatient units for new-onset manic behavior without any physical findings. Each received a comprehensive workup that included a psychiatric assessment, brain imaging, laboratory studies, lumbar punctures, and neuropsychological testing. In both cases, the final diagnosis was multiple sclerosis.
Ms. A was a 54-year-old woman who first came to an outpatient clinic because she had stopped taking her thyroxine. She was crying, appeared depressed, and reported that God had "told her" to stop taking her medications. She reported no history of psychiatric illness, substance abuse, seizures, traumatic brain injury, or exposure to poisons or toxins. Eight days later, at the psychiatry emergency care center, she was seen with pressured speech, tangentiality, disorganization, grandiose delusions (reporting receiving messages from God), labile affect, and difficulty paying attention. Collateral sources indicated that she had been hearing "God’s voice" for the last year. The voices had instructed her to leave her successful 20-year career to become an evangelist. No clinically obvious cognitive deficits were observed. Ms. A was admitted to the inpatient psychiatry unit with a preliminary diagnosis of bipolar disorder (manic episode) and was given divalproex sodium and risperidone.
Initially, Ms. A appeared to respond to the mood stabilizer and antipsychotic medication. However, over the ensuing hospitalization, she had significant vacillations in affect (crying, screaming, and euphoria) and thought processes (incoherence, tangentiality, and loosening of associations) despite reported medication adherence. Of particular concern were her poor cognitive function and judgment. Although she had a high premorbid level of functioning, at the hospital, she was unable to understand medication regimens and showed little understanding of her medical condition.
A concurrent medical workup was initiated to explain her symptoms. Neurology consultants performed complete physical and neurological examinations. The only abnormal finding was uniformly reduced reflexes bilaterally in the upper and lower extremities. Measures of Ms. A’s electrolyte levels and blood counts were within normal limits, as were results of a urinalysis and measures of B12, folic acid, protime and international normalized ratio, partial thromboplastine time, and liver function. The results of the following tests were negative: rapid plasma reagin, microhemagglutination-Treponema pallidum, hepatitis B and C, HIV, antinuclear antibody, rheumatoid factor, erythrocyte sedimentation rate, a urine drug screen, and a pregnancy test. Ms. A’s initial thyroid function tests revealed a slightly elevated level of thyroid stimulating hormone (6.98 μU/ml) but normal levels of free T3 and T4.
A computed tomography imaging of Ms. A’s head was remarkable for reduced white matter in the parietal lobes. Subsequent magnetic resonance imaging (MRI) of the brain revealed diffuse and multifocal high signal intensity lesions in the brain parenchyma, mainly in the deep white matter of the cerebral hemispheres. These were best seen on fluid-attenuated inversion recovery images. White matter abnormalities, atrophy, and focal lesions in the remaining white matter of the cerebral hemispheres also were present (F1). The lesions were suggestive of demyelinating disease, although none was enhanced with gadolinium contrast; i.e., the lesions were unlikely to be active. A single photon emission computed tomography scan identified mildly reduced perfusion to the frontal and parietal lobes bilaterally, with normal blood flow in the remainder of the brain (F1).
The MRI findings suggested the possibility of multiple sclerosis, prompting a lumbar puncture. Analysis of CSF confirmed the presence of immunoglobulin oligoclonal bands. The remaining CSF indices were found to be within normal limits. Ms. A’s EEG disclosed a focus of slow-wave activity in the left temporal region, suggesting a focal abnormality. The neurological consultants diagnosed her with chronic progressive (primary-progressive) multiple sclerosis. They recommended a course of cyclophosphamide, 750 mg/m2 intravenous pulse once a month for 4–6 months, along with interferon (INF)-β1b, 0.3 mg in a subcutaneous injection every other day, for maintenance treatment of Ms. A’s multiple sclerosis. Levothyroxine, 0.075 mg/day, was reinitiated for hypothyroidism.
Neuropsychological testing was administered to study Ms. A’s deficits in cognition and judgment. Results of the WAIS-III placed her in the borderline range of intellectual functioning, although her premorbid level of functioning was within the high range. She also demonstrated slowed information processing, concrete thinking, impaired memory retrieval, perseveration, and difficulties in constructional skills.
Although Ms. A’s mood fluctuations decreased, she continued to have poor insight and judgment, believing that she could return to work and live independently. After administration of the Kohlman Evaluation of Living Skills (1) and a trial of living with her family with 24-hour supervision failed, Ms. A was discharged to a locked nursing home unit.
Ms. B was a 50-year-old white woman who at the time of her initial psychiatric consultation had been diagnosed with complex partial seizure disorder, which was poorly controlled because of poor medication adherence. At her evaluation, she had pressured speech, flight of ideas, and hyperreligiosity and exhibited sexually inappropriate behavior (exposing herself to staff). Ms. B expressed some grandiose beliefs and was at times guilt ridden about perceived transgressions. She was admitted to the inpatient psychiatry unit and given olanzapine, while maintaining her oral anticonvulsant regimen of 450 mg every 12 hours of lamotrigine and 600 mg in the morning and 900 mg at bedtime of oxcarbazepine. Ms. B was taking no other medications and had no history of psychiatric diagnoses, substance abuse, traumatic brain injury, or exposure to poisons or toxins.
During her admission to the inpatient psychiatry unit, Ms. B refused to take her medications based largely on persecutory delusions. She was noted to have poor insight into her situation, as well as deficits in memory (an inability to provide information chronologically or accurately) and language (word approximation and neologisms). A complete physical examination did not reveal peripheral neurological deficits. Ms. B underwent a complete medical workup to better elucidate her symptoms.
Her laboratory data included normal levels of electrolytes, serum protein electrophoresis, blood counts, prolactin, antinuclear antibody, and Reiter protein reagin. An EEG revealed diffuse disturbances in the brain and focal slow and sharp activity bilaterally in the frontotemporal regions as well as a frontotemporal seizure focus. An MRI examination revealed multiple high signal intensity punctate and patchy periventricular white matter lesions, including a distinctive Dawson’s finger (F2). This suggested a demyelinating process, and accordingly, a lumbar puncture was performed. The CSF revealed oligoclonal bands, which supported the diagnosis of a demyelinating condition. The CSF analysis showed no other abnormalities.
Ms. B also underwent neuropsychological testing. Administration of the WAIS-III revealed a full-scale IQ of 67. This was far below expectations based on her past educational (a master’s degree) and vocational background. Ms. B exhibited word-finding and retrieval difficulties, despite her intact conversational speech. Her recall of auditory verbal or contextual information was severely impaired and characterized by confabulation. She was unable to grasp the concept of alternating sequences on the Trail Making test (2). Perseveration and significant distractibility were also noted. Ms. B’s MMPI results showed an elevation of scores on the hypomania scale. Her score on the Kohlman Evaluation of Living Skills showed deficits in the awareness of dangerous household situations, an inability to identify appropriate actions for sickness and accidents, and an inability to budget monthly income or to use banking forms.
During her stay, Ms. B refused to take multiple antipsychotic medications, including olanzapine, ziprasidone, and risperidone. However, she did continue to take her seizure medications. After refusal of multiple medication interventions, the neurology consultants simply requested that Ms. B be followed as an outpatient. She was discharged with the recommendation of 24-hour supervision and was deemed unfit to return to work.
Multiple sclerosis is an inflammatory demyelinating illness that can affect any aspect of the CNS, including the cranial nerves. The cause of the illness is unknown, although most hypotheses point toward a viral or autoimmune etiology (3). The disorder is generally viewed as a solely neurological condition with clinical manifestations that include motor impairment, sensory deficits, bowel and bladder dysfunction, and optic neuritis. Multiple sclerosis tends to follow one of four clinical courses, each of which may be of varying severity. The most common (in 66% of patients) is a relapsing-remitting course in which there is total recovery after an exacerbation. "Secondary-progressive" refers to a relapsing-remitting course that becomes progressively worse such that there is no complete remission after an attack. Sixteen percent of those diagnosed with multiple sclerosis have a secondary-progressive course. About 15% of the patients diagnosed with multiple sclerosis have a progressive course from the onset that is referred to as "primary-progressive." In primary-progressive multiple sclerosis, the symptoms do not remit, and acute attacks are not observed. Autopsy results indicate a fourth type of course—benign multiple sclerosis. These patients do not have a second relapse and generally do not seek medical attention (3).
The study of multiple sclerosis from a psychiatric perspective was present as early as the turn of the last century. Cottrell and Wilson in 1926 (4) divided the "mental symptomatology" of multiple sclerosis into intellectual and emotional disorders, with infrequent psychotic disorders observed.
Diaz-Olavarrieta et al. (5) confirmed the findings of Cottrell and Wilson when they administered the Neuropsychiatric Inventory (6) to 44 patients with multiple sclerosis who were not undergoing steroid treatment or were not experiencing an acute period of exacerbation of their illness (5). They found that 95% of the patients were experiencing neuropsychiatric symptoms. Their evaluation revealed that dysphoria or depressive symptoms were most common (79%), followed by agitation (40%), anxiety (37%), irritability (35%), apathy (20%), euphoria (13%), disinhibition (13%), hallucinations (10%), and delusions (7%).
Over the years, the high preponderance of psychiatric symptoms in patients with multiple sclerosis has led to the suggestion that this disease should be routinely included in the differential diagnosis of patients being seen for psychiatric complaints (7–9). Lyoo et al. (7) performed brain MRI on 2,783 consecutive inpatients (ages 15–59) who were referred as part of their psychiatric evaluation. Initially, 53 patients (1.9%) were noted to have a pattern of white matter hyperintensity (WMH) consistent with multiple sclerosis. Subsequently, two neuroradiologists who were blinded to the reason for the referral and diagnosis reviewed the MRI scans and categorized them according to the criteria of Paty et al. (10). Their findings indicated that 0.83% of the patients had T2-weighted WMH that was consistent with multiple sclerosis. Although a clear causal relationship between the neurological findings and clinical presentation cannot be drawn, the percentage of psychiatric patients with WMH consistent with multiple sclerosis was almost 15 times the reported prevalence of multiple sclerosis in the United States (0.058%). Again, affective disorders (depressive and bipolar) were found to be the most common psychiatric findings in these patients.
The patients’ lack of a previous psychiatric history, substance abuse, or brain injury, and late onset of psychiatric sequelae, or neuroimaging findings and a limited response to conventional psychotropic medications led us to bipolar disorder due to multiple sclerosis (without the peripheral findings generally associated with multiple sclerosis) as the most likely diagnosis. An alternate hypothesis is that the patients were experiencing late-onset or atypical mania and that the imaging and laboratory abnormalities were incidental findings. However, while a clear causative link may not be made, the data are highly suggestive of a diagnosis of an affective illness due to multiple sclerosis.
An interesting component of both cases involves the initial presenting symptoms in relation to lesion location. Both patients had lesions in their frontal lobes. The frontal lobe, and more specifically, the prefrontal cortex, is responsible for synthesizing information from different regions of the brain and placing it into deliberate and socially appropriate contexts. The well-recognized functions of the prefrontal cortex include planning, organization, inhibition, empathy, and motivation. These are believed to be executed by three distinct brain circuits that are contained in the dorsolateral prefrontal, orbitofrontal, and anterior cingulate portions of the frontal lobe (11, 12). The axonal projections from these cortical areas course through the white matter to subcortical structures. Disruption of these circuits can occur at any of these locations. More than one circuit can be damaged, especially in the white matter (which is affected by demyelinating diseases), where the pathways travel exceedingly close.
The dorsolateral prefrontal cortex is considered to be responsible for executive functioning, i.e., for tasks such as planning, organization, and attention. When patients have damage to this circuit, they are found to perseverate, exhibit concrete thinking, have difficulty with set shifting and screening out environmental distractions, and have impairments in constructional skills and sequential motor tasks. The two patients clearly demonstrated these symptoms during their assessments. The orbitofrontal cortex is believed to govern socially appropriate behavior and empathy. Lesions can cause impulsivity, lability, personality changes, and lack of humanistic sensitivity. This constellation of symptoms is consistent with mania, which was observed in both patients. The third region is the anterior cingulate cortex, which is thought to have at least two further subdivisions, the affect subdivision and the cognition subdivision (13). The affect portion has connections with the limbic and paralimbic regions, including the orbitofrontal cortex. The cognition subdivision connects with the parietal cortex, the spinal cord, and the dorsolateral prefrontal cortex. The connections of the anterior cingulate cortex highlight the linkage and interdependence of the frontal circuits.
Disruption of the aforementioned brain circuitry is thought to explain the presentation of our patients. Similar findings have been implicated in late-life depression. Taylor et al. (14) used statistical parametric mapping to identify frontal white matter lesions in elderly depressed patients. They found an association with lesions connecting white matter tracts extending from inferior frontal regions toward the basal ganglia and depression in patients. Previous evidence also associated late-life depression, the basal ganglia, and WMH lesions (15, 16). Alexopoulos and coauthors (17, 18) have defined the "depression-executive dysfunction syndrome" in the elderly, which is thought to be a distinct depressive disorder marked by executive dysfunction as a result of lesions in the basal ganglia and left frontal regions. They also reported that patients with depression-executive dysfunction syndrome are less responsive to antidepressant treatment and have fewer vegetative symptoms. The description of depression-executive dysfunction syndrome parallels our case presentations in terms of the cause of executive dysfunction and response to psychotropic medications.
A second issue is imaging techniques and lesion detection. Lyoo et al. (7), using T2-weighted images, found that 0.83% of their patients had WMH consistent with multiple sclerosis. We postulate that this could be higher if newer, more sensitive methods of MRI had been available. Improvements to T2-weighted images include the addition of fast spin echo and fluid-attenuated inversion recovery. Fast spin echo allows thinner contiguous sections of T2-weighted images, which allows for better detection. Also, the addition of fluid-attenuated inversion recovery, as in Ms. B, to the T2-weighted MRI provided an image in which the strong signal from the CSF had been removed, thus allowing finer delineation of multiple sclerosis lesions that may be present adjacent to CSF-laden spaces (19). Both of these techniques improve the sensitivity of MRI and consequently the ability to locate lesions.
Two new MRI techniques currently being used for research purposes are diffusion tensor imaging and magnetization transfer. Diffusion tensor imaging uses the speed at which water diffuses in white matter to assess the direction and integrity of the white matter tracts (20). Most recently, Rovaris et al. (21) implemented the technique to differentiate secondary-progressive and primary-progressive multiple sclerosis. Magnetization transfer depends on the exchange of magnetization between less mobile protons (bound to cellular macromolecules) and more mobile protons (present in cellular water). A low magnetization transfer ratio indicates a reduced ability of the macromolecules in the CNS to exchange magnetization with the surrounding water molecules, representing damage to myelin or to the axonal membrane (22). Filippi et al. (22) investigated magnetization transfer as a diagnostic tool in patients with multiple sclerosis. While magnetization transfer has the advantage of increased sensitivity in locating the pathology of multiple sclerosis, the lesions were not found to be specific to multiple sclerosis.
A third point of interest in these cases is the medical treatment of patients with the neuropsychiatric symptoms of multiple sclerosis. A review of the literature did not yield any specific treatment modalities for multiple sclerosis with only neuropsychiatric symptoms. Recent guidelines for the treatment of multiple sclerosis (23) divide treatments into those for acute exacerbations and those that affect the long-term course of the illness (disease-modifying therapy).
The mainstay for treatment of acute exacerbations remains corticosteroids. Polman et al. (23) recommend intravenous methylprednisolone as "the intervention of choice in patients with an acute exacerbation that warrants treatment" (pp. 9–10). The side effects of intravenous methylprednisolone include psychosis, peptic ulceration, and infections. Additionally, intramuscular ACTH, although proven efficacious, is no longer the preferred treatment for the exacerbations of multiple sclerosis because intravenous methylprednisolone may work more quickly and effectively. Also, plasmapheresis should be reserved only for patients with persistent acute and fulminant symptoms after a trial of intravenous steroids.
The review of disease-modifying therapies by Polman et al. (23) defined two approved treatments: IFN-β and glatiramer acetate. There are two types of IFN-β approved for the treatment of multiple sclerosis. IFN-β1a is administered through an intramuscular injection, which may be given once a month or once a week based on the preparation. IFN-β1b is administered by subcutaneous injection every other day. The side effects include flu-like symptoms (such as fever), myalgia, elevations in measures of liver function, or anemia. No one preparation of IFN-β is preferred over the other, although a skin reaction to subcutaneous administration may indicate use of IFN-β1a. An increasing concern about the use of IFN treatments is their depressive effects. A review of the literature revealed two randomized controlled trials that found that depression is not an effect of treatment with IFN-β1a (24) and that there was no evidence of increased depressive symptoms with IFN-β1a (24). No studies were found assessing the possible mood side effects of IFN-β1b; recall that Ms. A received IFN-β1b. Nonetheless, based on preliminary data and clinical practice, it is recommended that patients should be educated about depression and complete a depression inventory at regular intervals during the course of IFN treatment (25). Furthermore, patients with a history of depression who are initiating IFN treatment should be given a serotonergic antidepressant and monitored closely for the symptoms of depression (25). A final consideration of IFN-β1b is the cost of therapy, which is approximately $8,000 per year for the medication alone.
Polman et al. (23) recommend that glatiramer acetate should be reserved for those who are resistant to IFN-β or do not tolerate it because of side effects. Their guide to treatment and management also reviews treatments that have not been specifically approved for multiple sclerosis. These treatments may be divided into three groups:
Therapies that are still investigational or experimental, such as antiviral agents, cyclophosphamide, cladribine, and plasmapheresis.
Treatments whose benefits, if any, are outweighed by side effects or risks involved with administration, such as azathioprine, cyclosporin A, IFN-α, methotrexate, and total lymphoid irradiation.
Treatments that are contraindicated, such as corticosteroids (as a disease modifying agent) and IFN-γ.
Ms. A received cyclophosphamide and IFN-β1b. Steroids were not used because of the fear of exacerbation of psychosis. Additionally, as noted, her lesions did not enhance on MRI and therefore were thought not to have been acute. Accordingly, steroids were not indicated in this situation. Cyclophosphamide was preferred because of its availability. Based on the information provided, one may want to explore intramuscular ACTH as an alternative to intravenous methylprednisolone. Also, if the patient is admitted to an inpatient psychiatric facility, intravenous methylprednisolone may be considered with the knowledge that the patient may experience temporary exacerbation of his or her psychiatric symptoms. Ms. B did not receive any acute treatment since she, too, did not have any active lesions. She refused to take medications other than those for seizures and was accordingly asked to follow up with the neurological and psychiatric services for further monitoring and consideration of the use of medications specifically indicated for multiple sclerosis in the future.
The patients’ treatment with psychotropic medications was not as effective as expected. This is consistent with the response to medication in depressed patients with WMH; i.e., the extent of subcortical white matter lesions was negatively correlated with response to the treatment regimen, as reported by Hickie et al. (26) and supported by other investigators (27–30).
These cases highlight the need for psychiatrists to include multiple sclerosis in their differential diagnosis and to initiate the appropriate workup to identify the illness. General medical conditions should always be eliminated before a primary psychiatric condition is diagnosed. Suspicion should be especially heightened when a patient has late-onset or atypical features, peripheral physical findings, a lack of response to standard treatments, and cognitive changes, as were noted for the patients presented. Laboratory tests, brain imaging scans (T2-weighted MRI with fluid-attenuated inversion recovery), and lumbar punctures are recommended to assist in the diagnosis of multiple sclerosis. Neuropsychological testing can better help ascertain the suspected cognitive deficits. If mood and cognitive symptoms are thought to be part of the constellation of symptoms of multiple sclerosis, as suggested in the article, we would recommend treatment of these symptoms with medications aimed at alleviating the demyelinating process as outlined in the discussion. There is a paucity of information about the treatment of neuropsychiatric symptoms of multiple sclerosis with psychotropic medications; therefore, we believe that it is best to individualize treatments and target symptoms with the most appropriate agent(s).
Received Feb. 21, 2003; revision received Sept. 23, 2003; accepted Sept. 29, 2003. From the Menninger Department of Psychiatry and Behavioral Sciences and the Herbert J. Frensley Center for Imaging Research, Department of Radiology, Baylor College of Medicine; and the Mental Health Care Line, Houston Veterans Affairs Medical Center, Houston, Tex. Address reprint requests to Dr. Asghar-Ali, Menninger Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030-3498; firstname.lastname@example.org (e-mail). The authors thank Dr. Glen O. Gabbard for his help with the manuscript.
Magnetic Resonance Imaging (MRI) and Single Photon Emission Computed Tomography (SPECT) Scans of Ms. Aa
aIn image A, multiple confluent lesions are visible in the periventricular and subcortical white matter on this axial T2-weighted fluid-attenuated inversion recovery MRI. In image B, CBF is clearly reduced in the frontal and parietal cortices (arrows) on this axial SPECT image.
Magnetic Resonance Imaging Scan of Ms. Ba
aMultiple lesions are visible in the periventricular and subcortical white matter on this axial T2-weighted fluid-attenuated inversion recovery image. Dawson’s finger (arrow) is a characteristic finding in multiple sclerosis.