Standards of Care For Non-Convulsive Status Epilepticus: Belgian Consensus Recommendations
Standards of Care For Non-Convulsive Status Epilepticus: Belgian Consensus Recommendations
————
confusional states are common and may lead, in the subjects with NCSE had a prior history of epilep-
absence of EEG evaluation, to misdiagnosis of sy (2). It is probable that poor initial control of
NCSE as a psychiatric condition (4). The more seizures favours conversion to NCSE, as is well
subtle clinical presentations may not be recognised documented for CSE. A retrospective study (13) of
as NCSE and go untreated if EEG is not performed absence status found that the mean age of onset of
or patients may be inappropriately treated for NCSE was twenty years later than the original
another condition such as transitory ischaemic diagnosis of absence epilepsy, and followed the
attacks. Moreover, patients in coma may be admit- development of generalised tonic-clonic seizures.
ted to intensive care units (ICU) without the asso- The aetiology of focal NCSE may involve
ciated EEG anomalies being detected. underlying brain lesions caused by tumours or trau-
The ERF Workshop has attempted to provide a ma, as well as cerebrovascular disorders. These can
classification scheme for NCSE based on age of be generally identified by magnetic resonance
onset and underlying epilepsy syndromes (2). imaging (MRI). In patients with NCSE associated
Whilst this will be invaluable in improving with coma, the epilepsy may be the cause of the
research in NCSE by identifying homogenous coma in some cases, whereas in others both coma
patient groups for evaluating pathophysiology, and NCSE can be attributed to another cause, such
diagnostic procedures and treatments, it is perhaps as hypoxia (14). In some severe epilepsy syn-
less useful in routine clinical practice where the dromes such as Lennox Gastaut syndrome, tonic
underlying epilepsy syndrome may be inadequate- status may appears as coma with subtle clinical
ly characterised. Three phenotypes are commonly signs and fast activities on EEGs. This condition
seen in practice. These are absence status, complex may be precipitated by the introduction of benzodi-
focal status and NCSE associated with coma. azepines given for another seizure type. In addi-
Absence status and focal NCSE can be distin- tion, certain rare genetic conditions or chromoso-
guished easily by EEG, the former being charac- mal anomalies can present as iterative NCSE. An
terised by generalised 2 to 4 Hz spike and wave example is ring chromosome 20 syndrome (15).
activity and the latter by more or less focalised dis- Apparently successful treatment of CSE (ie
charges generally associated with the temporal or absence of convulsive activity) may in a minority
frontal lobe. Absence status is more benign in its of cases actually reflect transformation into
clinical presentation and in its prognosis than focal NCSE (16). For this reason, outcome in CSE needs
NCSE and is more common in younger patients. to be monitored carefully by EEG in patients stay-
ing in an abnormal consciousness state.
EPIDEMIOLOGY In some cases, antiepileptic drug treatment itself
may induce NCSE. This has been described most
It is believed that around one-third of all cases of convincingly for tiagabine (17-21) and has also
status epilepticus correspond to NCSE (5). The ERF been documented for carbamazepine (22). For
workshop (2) identified and reviewed six epidemio- other antiepileptic drugs, the evidence is anecdotal
logical studies of the incidence of NCSE in a hospi- and the causal relationship between treatment and
tal setting (6-11). These have provided incidence emergence of NCSE unclear. NCSE may also
rates of between 10 and 40 cases per 100,000 sub- develop following discontinuation of antiepileptic
jects per year, corresponding to between 1000 and drugs in general or as part of a benzodiazepine
4000 subjects in Belgium per year. However, it is withdrawal syndrome. The latter is particularly fre-
clear that NCSE is under-diagnosed (12) and that a quent in the elderly.
significant proportion of patients with more subtle
clinical presentations are not seen systematically in PROGNOSIS
hospitals. NCSE can arise throughout the lifespan,
with absence status being particularly frequent in The prognosis of NCSE is traditionally consid-
children and teenagers with idiopathic epilepsy ered to be better (23) than that of CSE. However,
syndromes or in young patients with epileptic prognosis varies according to the type of NCSE
encephalopathies such as Lennox-Gastaut syn- and the underlying aetiology (24). In general,
drome, and NCSE associated with coma in the absence status appears to be relatively benign.
elderly. Complex focal NCSE, on the other hand, may
result in permanent brain damage or in long-term
AETIOLOGY cognitive impairment, particularly if uncon-
trolled (5, 25). There have been reports, using seri-
The aetiology of NCSE is poorly understood and al MRI scans, of localised cerebral oedema associ-
apparently very heterogeneous. In many cases, ated with focal NCSE which evolves into tissue
NCSE appears to correspond to a conversion of an atrophy (26). However, in such patients, it may not
existing epilepsy syndrome presenting as self-lim- always be clear to what extent NCSE itself, rather
iting seizures. In the epidemiological studies of than the underlying focal lesion, contributes to the
NCSE referred to above, between 30% and 50% of residual neurological deficit (27). Particularly in
CONSENSUS GUIDELINES ON NON-CONVULSIVE STATUS EPILEPTICUS 119
patients are presented in Figures 2 and 3. If the FIG. 2. — Treatment algorithm for non-convulsive absence
status epilepticus in non-comatose patients. Black boxes cor-
epilepsy syndrome underlying the NCSE is respond to treatment actions and grey boxes to decision points.
unknown, the more restricted standard treatment The left side of the diagram refers to the acute treatment strat-
algorithm for absence status should be used as a egy and the right half to maintenance therapy. The text should
conservative measure. be referred to for doses and complete information. AED :
In the case of comatose patients with NCSE, the antiepileptic drug ; RT : rapidly titrating. *Narrow spectrum
AED such as carbamazepine or phenytoin should be avoided.
situation is different and treatment will generally
be administered in the ICU. Two types of patient
should be distinguished. On the one hand, patients
with a history of epilepsy and with no or minimal
perturbation of major organ function can be identi-
fied, for whom NCSE is likely to be responsible for
the coma. In these patients, outcome is likely to be
good if the NCSE is controlled, leading to rapid
resolution of the coma. An intensive treatment reg-
imen is appropriate for these patients, equivalent to
that used for CSE (Fig. 4) (1). This involves an
accelerated transition through the different treat-
ment levels compared to the standard treatment
algorithm for non-comatose NCSE, with recourse
to anaesthesia with drugs such as propofol if need-
ed. On the other hand, other patients have major
organ failure and the NCSE is likely to be sec-
ondary to the events that precipitated the coma. In
such patients, use of antiepileptic drugs can be dan-
gerous and intensive treatment may well do more
harm than good (12). Overall prognosis is likely to
depend more on the resolution of the events that led
to the coma in the first place than to successful
management of NCSE. For these patients, a mini-
mal treatment regimen restricted to intravenous FIG. 3. — Treatment algorithm for non-convulsive focal sta-
benzodiazepine administration is appropriate tus epilepticus in non-comatose patients. Black boxes corre-
(Fig. 5). It is evident that there will be a spectrum spond to treatment actions and grey boxes to decision points.
The left side of the diagram refers to the acute treatment strat-
of patients between these two extreme profiles, and egy and the right half to maintenance therapy. The text should
clinical judgement should be exercised in identify- be referred to for doses and complete information. AED :
ing the degree of intensity of treatment that has the antiepileptic drug ; RT : rapidly titrating.
optimal risk-benefit ratio in individual patients.
The appropriate first-line treatment is intra-
venous benzodiazepine administration. A long-act- given at a maximal infusion rate of 2 mg/min).
ing drug should be chosen to maintain control. The Diazepam should be avoided since breakthrough
most appropriate choice is lorazepam (0.1 mg/kg NCSE may occur as the drug redistributes out of
CONSENSUS GUIDELINES ON NON-CONVULSIVE STATUS EPILEPTICUS 121
36. ADIN J., ARTEAGA R., HERRANZ J. L., ARMIJO J. A. 41. KATRAGADDA S. B., ALURI B. C., BURDETTE D. E.
The use of intravenous valproate. Rev. Neurol., Intravenous administration of valproate for status
1999, 29 : 744-53. epilepticus in 12 patients. Epilepsia, 2000, 41
37. LICHT E. A., FUJIKAWA D. G. Nonconvulsive status (Suppl. 7) Abstract 3.164.
epilepticus with frontal features : quantitating sever- 42. THOMAS P., ZIFKIN B., MIGNECO O. et al. F.
ity of subclinical epileptiform discharges provides a Nonconvulsive status epilepticus of frontal origin.
marker for treatment efficacy, recurrence and out- Neurology, 1999, 52 : 1174-83.
come. Epilepsy Res., 2002, 51 : 13-21. 43. CAMACHO A., PEREZ-MARTINEZ D. A., VILLAREJO A. et
38. JHA S., JOSE M., PATEL R. Intravenous sodium val- al. Nonconvulsive status epilepticus : experience in
proate in status epilepticus. Neurol. India, 2003, 51 : 33 patients. Neurologia, 2001, 16 : 394-8.
421-2.
39. PETERS C. N., POHLMANN-EDEN B. Intravenous val-
proate as an innovative therapy in seizure emer-
gency situations including status epilepticus – expe-
rience in 102 adult patients. Seizure, 2005, 14 : 64- Prof. Dr. K. VAN RIJCKEVORSEL, M.D.,
9. Reference Center for Refractory Epilepsy,
40. WHELESS J. W., VAZQUEZ B. R., KANNER A. M. et al. Cliniques Universitaires St. Luc UCL, Neurologie,
Rapid infusion with valproate sodium is well toler- Avenue Hippocrate, 10,
ated in patients with epilepsy. Neurology, 2004, 63 : B-1200 Brussels (Belgium).
1507-8. E-mail : vanrijckevorsel@nops.ucl.ac.be.