If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Stefan Lorenzl, MD, Interdisciplinary Center for Palliative Medicine, Klinikum Grosshadern, University of Munich, Marchioninistrasse 15, 81377 Munich, Germany.
Interdisciplinary Center for Palliative Medicine, Munich University Hospital, Munich, GermanyDepartment of Neurology, Munich University Hospital, Munich, Germany
Altered mental status and reduced level of consciousness are common among patients admitted to palliative care units. However, nonconvulsive status epilepticus (NCSE) has rarely been considered as a possible cause, and electroencephalographic confirmation of the epileptic status is sparse. The clinical presentation of patients with NCSE varies from altered mental status to coma, with no or only minimal convulsions. We report a prospective evaluation of patients with altered mental status on admission to our palliative care unit in the year 2007. Of 290 patients admitted in 2007, 49 patients showed signs of confusion or delirium and/or a reduction in their level of consciousness. NCSE was suspected clinically in 22 of these patients, and epileptic activity could be confirmed in 15 (5.2%) of 290 patients. Nine of 15 patients could be effectively treated with anticonvulsants and regained communication ability before death. NCSE appears to be an important, often unrecognized, and potentially treatable cause of altered mental status in palliative care patients. Pharmacological treatment might restore communicative abilities even in severely ill patients.
There are many causes for altered mental status in terminally ill patients, but one important and treatable cause might be nonconvulsive status epilepticus (NCSE). Diagnosis of NCSE is more challenging than diagnosis of convulsive seizures because the clinical presentation ranges from minimally altered mental status to coma without tonic-clonic activity.
Although few data are available, NCSE has been reported in 6% of patients with systemic cancer without evidence of central nervous system involvement and in up to 20% of patients with primary brain tumors or metastases.
The most common etiologies of NCSE in palliative care patients treated on a palliative care unit are likely to be primary brain tumors and metastases, especially when edema is present. In a systematic survey in older patients suffering from status epilepticus, ischemic stroke, and intracranial hemorrhage were the underlying cause in 40%; 17% were because of hypoxia; and 14% because of metabolic disturbances; whereas brain tumors only accounted for 10% of cases.
Although epileptic seizures are common in patients with brain tumors or metastases, little is known about the incidence and outcome of status epilepticus in patients on palliative care units, and NCSE remains poorly understood in this population. Its response to treatment and the outcome after treatment have never been investigated prospectively. To determine the prevalence of NCSE in patients with delirium and its course after a standardized treatment, an observational prospective study was conducted in all consecutive patients with unexplained changes in mental or behavioral status who were admitted to our palliative care unit in the year 2007.
Patients and Methods
We identified patients with clinical signs of confusion or delirium, and/or reduced level of consciousness, who were admitted to the Interdisciplinary Center for Palliative Medicine at the Munich University Hospital in the year 2007. After a complete neurological examination, each patient with a clinical presentation suggesting NCSE received a 36-channel electroencephalographic (EEG) evaluation. Clinical signs suggesting NCSE were mild myoclonic jerks of facial muscles (e.g., angle of the mouth) or epileptic nystagmus. Treatment was initiated if epileptic activity was confirmed by the EEG. A second EEG was obtained in eight of our patients who showed prolonged clinical improvement after therapy. In the remaining patients, response to treatment was assessed clinically by daily neurological examination. Effective treatment was defined as clinical resolution of NCSE with regained consciousness and ability to communicate.
Results
A total of 290 patients were admitted to the Acute Palliative Care Unit of the Interdisciplinary Center for Palliative Medicine in 2007. We identified 49 patients with altered mental status on admission. Of this group, 22 patients showed clinical signs and symptoms suggestive of NCSE, such as mild myoclonic jerks of the angle of the mouth and/or nystagmus. In the EEG recording, 15 of these 22 patients displayed epileptiform discharges (e.g., sharp and slow waves, repetitive sharp waves, intermittent sharp waves, intermittent sharp and slow waves, and intermittent spikes) (Fig. 1).
Fig. 1An example of an EEG of a patient with NCSE. It shows anteriorly dominant sharp and slow wave complexes with a very prominent positive component and a longer duration slow wave, which could be atypical triphasic waves. Although encephalopathy might be a differential diagnosis, this patient improved on antiepileptic treatment. The figures show the 1a) baseline and 1b) recordings 10 seconds and 1c) 20 seconds later.
The diagnoses of the patients are shown in Table 1. Nine of the 15 patients had documented primary or metastatic neoplasms in the brain. The duration of NCSE could not be determined retrospectively because only eight patients had had an EEG recording before admission on our unit, which in five cases was suggestive of NCSE. Clinically, it was known that six patients had been comatose for at least one week before admission.
Table 1Data of the 15 Patients with NCSE
Age, years
Gender
Diagnosis
Presentation
Midazolam
Phenytoin
Others
Outcome
79
M
Cerebral ischemia
Somnolence
+
+
Valproic acid, levetiracetam
No effect, died four weeks later
76
M
Glioblastoma
Coma
Levetiracetam
Initial effect for two weeks, died three weeks later (NCSE recurrence)
77
M
Glioblastoma
Awake/somnolence
+
+
Levetiracetam
Initial effect for one week, died six weeks later (NCSE recurrence)
53
M
Astrocytoma II
Awake/confusion
+
+
Levetiracetam
Initial effect for seven weeks, died eight weeks later (NCSE recurrence)
52
M
Melanoma with brain metastases
Coma
+
+
Levetiracetam
Initial effect for five days, died six days later (NCSE recurrence)
42
F
Oligoastrocytoma
Somnolence
+
+
Levetiracetam
No effect, died three weeks later
71
M
Liver cirrhosis
Somnolence
+
Levetiracetam
No effect, died two weeks later
52
M
HIV, PML
Somnolence
+
Valproic acid
No effect, died two weeks later
96
F
Bilateral thalamic ischemia
Coma
+
Valproic acid, levetiracetam
No effect, died one week later
67
F
Melanoma with brain metastases
Coma
+
+
Valproic acid, levetiracetam
Initial effect for six days, died one week later (NCSE recurrence)
81
F
Bronchial carcinoma with brain metastases
Somnolence
+
Valproic acid, levetiracetam
No effect, died one week later
24
F
Astrocytoma III
Somnolence
+
Levetiracetam
Initial effect for seven days, died two weeks later (NCSE recurrence)
54
M
Astrocytoma
Confusion
+
+
Levetiracetam
Effect after one day, lasted until death four weeks later
62
M
Glioblastoma
Confusion, aphasia
+
Valproic acid levetiracetam
Effect after one day, effect lasted for 32 days
67
F
Oligoastrocytoma
Confusion, aphasia, and hemiparesis
+
+
Valproic acid, levetiracetam
Effect after eight days, lasted for 184 days, and died six days later (NCSE recurrence)
We used IV phenytoin as the first-choice drug, but none of our patients improved on it. Seven patients were then switched to valproic acid and eight patients directly to levetiracetam. Of the seven patients treated with valproic acid, only one showed clinical improvement. In the remaining six patients, levetiracetam was subsequently given as an add-on treatment.
Overall, clinical resolution of NCSE was observed in nine of 15 patients and lasted between five and 92 days. Patients with clinical resolution of NCSE fully regained consciousness and were able to communicate. A follow-up EEG was performed in two patients, which showed no EEG pattern suggestive of NCSE. The time lag between initiation of antiepileptic drug treatment, and clinical response ranged from four hours to six days.
All but one patient suffered recurrence of seizures or NCSE, which proved refractory to further treatment, and eventually died. Only one patient with an astrocytoma survived for another 184 days. In seven patients who presented with coma, four patients became awake and responsive after treatment.
Discussion
We report a case series of 15 patients with NCSE who were prospectively investigated and treated at our Interdisciplinary Center for Palliative Medicine. Although there is a limited number of reports on NCSE in terminally ill patients, this is, to our knowledge, the first prospective study in this patient group.
In our population, 19% of patients with brain tumors or brain metastases presented with NCSE, which is consistent with earlier reports.
Astrocytomas accounted for most primary brain tumors in our population (five patients). These low-grade tumors are known to be even more epileptogenic than high-grade gliomas.
These patients had been transferred to the palliative care unit for treatment of delirium, which turned out to be because of potentially treatable NCSE.
Among our patients, NCSE developed in their last days of life. Seizures in patients with primary brain tumors or brain metastases normally develop at time of diagnosis or tumor progression.
This point was not investigated in our patients because drug level monitoring was not performed. Other possible factors leading to NCSE in palliative care patients may be metabolic derangement (hypercalcemia, hyponatremia), addition of new drugs (antipsychotics, antidepressants, and diuretics), or progression of the underlying disease.
The time until diagnosis is an important predictor of the outcome and, therefore, of the value when deciding on treatment. Because response to treatment is often delayed, the minimum duration of therapy should be four to five days. Beyond this time, remissions are unlikely. If the treatment requires higher doses of sedative drugs, the potential benefit of the treatment has to be weighted against the side effects. The rationale and consequences of withdrawing the antiepileptic medication should be carefully discussed with the relatives and team members.
Most of our patients were noncommunicative or even comatose on admission. In these patients, the diagnosis of NCSE is challenging because the differential diagnosis is broad and includes drug intoxication and many other causes of encephalopathy. Subtle myoclonic movements or nystagmus may be suggestive of NCSE but are often absent.
might not be investigated when lids are closed. Therefore, a thorough neurological examination is mandatory for all patients with delirium to detect signs of NCSE.
Almost 90% of NCSE of all hospital inpatients is unrecognized at the bedside and is only diagnosed by continuous EEG.
Therefore, the actual number of NCSE in terminally ill patients might be significantly underestimated. In 22 of 49 patients, we suspected NCSE clinically and 15 of them showed EEG patterns consistent with NCSE (Fig. 1). However, because the EEG pattern of NCSE can be highly variable and sometimes difficult to distinguish from encephalopathy (Fig. 2), even clinically suspected NCSE cases without typical EEG patterns might warrant an antiepileptic treatment trial.
Fig. 2The EEG pattern of the 42-year-old female with right frontal oligoastrocytoma. She presented with an NCSE after a right frontotemporal trepanation. The status did not respond to treatment. This bipolar longitudinal recording shows a right frontotemporal status pattern.
Although levetiracetam has a wide spectrum of action and a favorable pharmacokinetic profile, few data exist regarding its use in NCSE. There are only retrospective reports on its use in status epilepticus.
which is a major disadvantage of phenytoin and valproic acid. Our results suggest that treatment should be started with levetiracetam and combined with other drugs when NCSE is not treated effectively. Because of its pharmacological profile, levetiracetam is particularly suited for the treatment of elderly patients,
Of the 15 patients treated in our series, 13 died within the first three months, with survival ranging from five to 184 days. Yet, these were all patients with terminal diseases where death was expected to follow within days to months. As this study was not controlled, we cannot say whether survival time was actually prolonged by NCSE treatment. The most obvious treatment effect, however, was the regained ability to communicate. Retrospective analysis of available data did not identify patient features predictive of treatment outcome. This may be because of the overall small number of patients in this study. Changes in neuropsychiatric performance could not be evaluated because of the lack of baseline data. None of the patients could remember what happened during the time of NCSE.
Two patients with ischemic stroke have been included in this study. One patient had a hemispheric infarction in the territory of the right middle cerebral artery and the other had bilateral thalamic infarctions. In both patients, treatment did not show any effect, and they died soon after. The patient with the thalamic infarctions was the oldest patient in our study (96 years), but effective treatment of NCSE in stroke patients has been reported in this age group.
Convulsive status epilepticus after ischemic stroke and intracerebral hemorrhage: frequency, predictors, and impact on outcome in al large administrative dataset.
NCSE is an important neurological condition in patients on palliative care units. Its incidence may be underestimated because of the lack of diagnostic evaluation. In our observational prospective study, we have shown that treatment of NCSE can restore communicative abilities even in severely ill patients. This might even be important when the remaining life span is very short. Treatment decisions must be made on a case-by-case basis and constantly reevaluated. More research into this syndrome, its treatment, and ethical implications in terminally ill patients are urgently needed.
Acknowledgments
The authors thank Claudia Schneider for secretarial assistance.
References
Grewal J.
Grewal H.K.
Forman A.D.
Seizures and epilepsy in cancer: etiologies, evaluation, and management.
Convulsive status epilepticus after ischemic stroke and intracerebral hemorrhage: frequency, predictors, and impact on outcome in al large administrative dataset.
00388-X&id=gr1.jpg){kind=link}
00388-X&id=gr2.jpg){kind=link}