Journal of Clinical Neuroscience 22 (2015) 29–34

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Review

Brain tumour-associated status epilepticus Janindu Goonawardena a,⇑, Laurence A.G. Marshman a,b, Katharine J. Drummond c,d a

Cairns Clinical School, School of Medicine, James Cook University, Cairns Hospital, Cairns, QLD 4870, Australia Department of Neurosurgery, Institute of Surgery, The Townsville Hospital, Douglas, Townsville, QLD, Australia c Department of Neurosurgery, Royal Melbourne Hospital, Parkville, VIC, Australia d Department of Surgery, University of Melbourne, Parkville, VIC, Australia b

a r t i c l e

i n f o

Article history: Received 3 December 2013 Accepted 29 March 2014

Keywords: Brain tumour Status epilepticus Tumour associated epilepsy

a b s t r a c t We have reviewed the scant literature on status epilepticus in patients with brain tumours. Patients with brain tumour-associated epilepsy (TAE) appear less likely to develop status epilepticus (TASE) than patients with epilepsy in the general population (EGP) are to develop status epilepticus (SEGP). TASE is associated with lesions in similar locations as TAE; in particular, the frontal lobes. However, in contrast to TAE, where seizures commence early in the course of the disease or at presentation, TASE is more likely to occur later in the disease course and herald tumour progression. In marked contrast to TAE, where epilepsy risk is inversely proportional to Word Health Organization tumour grade, TASE risk appears to be directly proportional to tumour grade (high grade gliomas appear singularly predisposed). Whilst antiepileptic drug (AED) resistance is more common in TAE than EGP (with resistance directly proportional to tumour grade and frontal location), TASE appears paradoxically more responsive to simple AED regimes than either TAE or SEGP. Although some results suggest that mortality may be higher with TASE than with SEGP, it is likely that (as with SEGP) the major determinant of mortality is the underlying disease process. Because all such data have been derived from retrospective studies, because TASE and SEGP are less common than TAE and EGP, and because TASE and SEGP classification has often been inconsistent, findings can only be considered preliminary: multi-centre, prospective studies are required. Whilst preliminary, our review suggests that TASE has a distinct clinical profile compared to TAE and SEGP. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Epilepsy in the general population (EGP) is characterised by recurrent seizures of spontaneous onset, associated with hypersynchronous discharges (typically emanating from focal regions of the supratentorial cerebral cortex) and by spontaneous seizure termination (usually within 2 minutes) [1]. Following seizure termination, there is typically a refractory period (which can span minutes to several days) during which it is usually difficult to provoke any further seizure activity. Status epilepticus in the general population (SEGP) is a potentially life-threatening medical emergency in which seizure activity continues for a prolonged period of time, or where seizures recur before there has been complete recovery from the consequences of the preceding seizure [2]. SEGP is a dynamic condition that, if inadequately treated, may result in permanent changes in clinical condition, responsiveness to treatment and in patient behaviour [2,3]. SEGP may also lead to permanent changes in cerebral ⇑ Corresponding author. Tel.: +61 7 4226 6830; fax: +61 7 4226 6831. E-mail address: [email protected] (J. Goonawardena). http://dx.doi.org/10.1016/j.jocn.2014.03.038 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.

histopathology, and in the electroencephalogram (EEG) [2,3]. Despite several advances, the fundamental mechanism by which a single seizure fails to be terminated, or by which the post-seizure refractory period fails to become established (or lapses), remains unknown [4]. We have reviewed and synthesised the scant literature on brain tumour-associated status epilepticus (TASE) as a prelude to further studies. Publications were identified by searching the electronic databases of PUBMED, MEDLINE and CINAHL for the period between January 1955 – June 2013. The medical subject heading (MeSH) terms ‘‘brain tumour’’, ‘‘status epilepticus’’, ‘‘non-convulsive status epilepticus’’, and non-MeSH terms ‘‘tumour associated epilepsy’’ and ‘‘antiepileptic drug’’ were used. Results were restricted to the English language. In addition, a manual search of bibliographies was conducted. The ‘‘find similar’’ function in MEDLINE was used to identify any additional studies. If the electronic version was not available, a hard copy of the article was requested through the local hospital library. Unfortunately, the study of SEGP and TASE (associated with primary brain tumours and cerebral metastases) has hitherto been confounded by a number of factors:

30

J. Goonawardena et al. / Journal of Clinical Neuroscience 22 (2015) 29–34

1. The changing definition of status epilepticus [2,5,6]. SEGP was first defined in the 1962 10th European conference on Epileptology and Clinical Neurophysiology as: ‘‘any seizure lasting more than 30 mins, or intermittent seizures over 30 mins from which full consciousness is not regained’’ [7]. The rationale for 30 minutes derived from animal research which had previously identified this period as a threshold for neuronal injury [2]. In 1991, Bleck defined SEGP as continuous or repeated seizures lasting more than 20 minutes [8], whilst in 1999 Lowenstein et al. defined SEGP as continuous seizures lasting for more than 5 minutes (or two or more discrete seizures without regaining consciousness) [6]. Clearly, studies using older definitions will yield significantly different incidences of TASE than subsequent studies. 2. Studies including variable seizure types [9–12]. The mechanism of each seizure type is distinct, and this may have important implications for SEGP and TASE biology. For example, ictal EEG activity stops abruptly, and without post-ictal sequelae, in absence seizures: yet neither may be the case in generalised or complex partial seizures [1]. Moreover, in another example, the mechanism which prevents the spatial cortical propagation of epilepsia partialis continuans is likely to be different from that which terminates typical seizure activity (and which subsequently initiates the onset of the refractory period). In consequence, series with differing proportions of seizure sub-types may therefore have contributed to inconsistent findings. This may be particularly relevant to primary brain tumours, which may possess a greater propensity to elicit partial TASE than generalised convulsive TASE [13]. 3. The retrospective nature of most studies. To our knowledge only four studies have provided significant data on TASE [9,10,14,24]. All of these studies have been retrospective, and suffered from small sample sizes (especially regarding subgroup comparisons and tumour types). 4. The changing incidence of generalised convulsive SEGP. It has been asserted that the incidence of generalised convulsive SEGP has been progressively falling, and that this potentially relates to improvements in emergency management [13,14]. However, with reported incidences for generalised convulsive SEGP as wide as 6–61 per 100,000/year, such assertions might be difficult to substantiate [13,15]. Similar issues beset non-convulsive SEGP, where inconsistencies in definition and classification have been associated with ranges spanning 4–50% [14–16]. 5. Epidemiological factors. Age is an important and overlooked issue. For example, the incidence of generalised convulsive SEGP follows a U-type distribution with age: one peak occurring at 0–4 years, the other at 60–75 years [13,14,17,18]. Furthermore, generalised convulsive SEGP has a slight male preponderance (1.3:1) [13]. 6. Differences in tumour grade and tumour type between studies. As discussed below, tumour grade and the type of primary brain tumour has a distinctive effect on TASE. In consequence, tumour histology should be considered with all statistical analyses.

non-enhancing lesion in the right posterior frontal region with no calcification and minimal mass effect (Fig. 1). She was neurologically intact. An awake craniotomy was performed and cortical mapping revealed both motor and sensory function within the tumour, thus after discussion with the patient, only a generous biopsy was performed. The histopathological examination of the surgical specimen showed a ganglioglioma, therefore observation only of the tumour was recommended. The seizures were initially controlled with phenytoin, however a few months after diagnosis crescendo focal seizures occurred over a period of 1 week culminating in focal status epilepticus of the left upper and lower limb requiring admission to the intensive care unit for a midazolam infusion for 1 week. After recovery from the Todd’s paresis related to this episode a mild hemiparesis, with spasticity most prominent in the lower limb, was noted. Multiple anti-epileptic drugs (AED) have been trialled over subsequent years with variable seizure control. She has had three subsequent episodes of focal status epilepticus requiring admission to the intensive care unit and after each her lower limb function has worsened, such that while she is able to mobilise independently it is with marked spasticity of the left lower limb. Serial imaging has shown slow increase in the tumour extent with no change in signal characteristics over 6 years, however due to ongoing seizures, radiotherapy is now planned.

2. Epidemiology The cumulative incidence of EGP by 74 years is 3.0%, for all unprovoked seizures 4.1%, and for any convulsive disorder 10% [19]. However, epilepsy is vastly more common in patients with

Despite the long list of confounding factors, most are in fact surmountable. The principal appellation from this review, therefore, is for the establishment of multi-centre, prospective studies to validate the trends herein implied. Such collaboration may prove fruitful since, based on the current weak evidence, TASE pathophysiology appears to deviate from brain tumour-associated epilepsy (TAE) in several domains. 1.1. Illustrative patient This 37-year-old woman suffered a single generalised tonicclonic seizure at work. A CT scan and subsequent MRI showed a

Fig. 1. Sagittal T1-weighted (upper) and axial fluid attenuated inversion recovery (lower) MRI showing a non-enhancing lesion in the right posterior frontal region with no calcification and minimal mass effect.

J. Goonawardena et al. / Journal of Clinical Neuroscience 22 (2015) 29–34

brain tumours, with TAE occurring in up to 80% of patients with low grade gliomas (LGG) [20–22]. SEGP constitutes a neurological emergency, and has significant associated morbidity and mortality [13]. The proportion of patients with SEGP attributable to an underlying brain tumour is reported to be between 4–12% [10,13,15,23]. However, the proportion of patients with TAE who develop TASE – although higher at 15– 22% – appears to be significantly less than the proportion of patients with EGP who develop SEGP (30–40%) [9,24]. Unfortunately, only two studies have examined the incidence of TASE in patients with TAE to our knowledge. Janz et al. reported an incidence of 22% in 123 patients: however, no attempt was made to define seizure type in this study [24]. By contrast, Moots et al. reported an incidence of 15% in 65 patients, all displaying simple partial TASE [9]. These estimates are likely to be affected by heterogeneity in seizure and tumour sub-type with TASE versus TAE, it being likely that the incidence of TASE is underestimated. Notwithstanding, if these incidences are representative, they potentially suggest that factors related either to seizure termination, or to the maintenance of the post-seizure refractory period, could be more robust in TAE than in EGP. However, other explanations related to psychosocial, pharmacological and sampling bias are also possible. 3. Tumour location in TASE It is well recognised that TAE is more common with tumours in frontal and temporal locations, but less common in occipital and (in particular) infratentorial lesions [20–22,25]. Frontal and temporal locations render tumours more amenable to resection, with potential improvements in survival. Despite this, TAE is typically more refractory to treatment in frontal lesions than TAE arising elsewhere [26]. The reason for such refractoriness is currently unknown. In a similar fashion, TASE is also more likely in patients with frontal lobe tumours than those in other regions: thus, four of five series have documented a higher incidence of TASE in tumours in frontal locations [23,24,27,28]. The refractoriness of TAE to treatment in frontal lobe tumours, discussed below, could explain the higher incidence of TASE in frontal lobe tumours. 4. Timing of TASE onset in disease course 4.1. TAE coincides with disease onset In most cases, TAE commences early in the course of the disease, or at tumour presentation. Hildebrand et al. reported that in 86% of 158 patients, TAE commenced soon after (or as an initial manifestation of) tumour presentation: only 14% of patients developed de novo TAE later in the course of their disease [29]. Similarly, Van Breemen et al. found that 70% of patients with LGG presented with seizures as the initial manifestation, compared to 52% of those with high grade gliomas (HGG) [30]. 4.2. TASE is delayed from disease onset Only four studies have provided data on the timing of TASE onset after brain tumour presentation [9,10,31,32]; they included patients with gliomas and cerebral metastases. In marked contrast to TAE, TASE was more likely to occur later in disease course, or to herald tumour progression. Indeed, in three out of the four studies, TASE was twice as likely to herald tumour progression than tumour presentation [9,31,32]. These reports potentially suggest that with tumour evolution factors associated either with spontaneous seizure termination,

31

or with the stability of the refractory period, become progressively impaired in patients sufficiently predisposed. However, other explanations related to psychosocial, pharmacological and sampling bias factors are also possible, as discussed below. 5. TASE and World Health Organization tumour grade 5.1. Inverse relationship between TAE and World Health Organization grade It has long been accepted that TAE risk is inversely proportional to World Health Organization (WHO) tumour grade. For example, TAE occurs in 80–90% of gangliogliomas and 75% of LGG but in only 29–49% of HGG and 20–35% of cerebral metastases [20,21,33]. The most commonly cited mechanism to account for this inverse relationship is the longer survival of patients with low grade tumours compared to their counterparts with high grade tumours; this permits sufficient time for the maturation of focal and remote parenchymal changes required for epileptogenesis [34–36]. For example, low grade tumours slowly deafferentate circumscribed cortical areas [12,33,37], whilst high grade tumours offset this by their more rapid growth and by shorter overall patient survival [12,21]. As stated by Rosati et al., the slow tumoural growth of LGG can ‘‘infiltrate and derange the neural network’’ to somehow organise and stabilise the epileptogenic zone, whilst HGG ‘‘destroy and prevent such stabilization’’ [12]. High grade tumours may thus interfere with the neuronal network required to convert isolated neuronal excitability into regional hyper-synchronous excitability. 5.2. Direct relationship between TASE and WHO grade In marked contrast to TAE, however, TASE risk appears to be directly proportional to tumour grade. This appears to be the case for both intrinsic and metastatic high grade tumours: however, patients with HGG appear singularly predisposed to develop TASE [9,10,24,27,28,31]. Several factors known to be associated with epileptogenicity are more prevalent in higher WHO grade brain tumours, for example, blood–brain barrier breakdown, florid cerebral oedema, necrosis and haemorrhage (with haemosiderin deposition) [38]. Notwithstanding, it seems counterintuitive that factors that correlate positively with TASE simultaneously correlate negatively with TAE. Other explanations, instead, must therefore be sought. Whilst the vast majority of patients with high grade tumours undergo surgery at presentation with subsequent adjuvant radiotherapy and/or chemotherapy, a smaller proportion of patients with low grade tumours undergo any form of surgery at presentation; this was especially so in former decades. In a similar fashion, few patients with low grade tumours undergo any form of adjuvant therapy (at least until disease progression). Such factors may bias the de novo development of TASE toward high-grade tumours, or the progression of TAE to TASE. For example, Foy et al. found a 17% risk of EGP at 5 years after craniotomy for any condition (although this ranged from 3–92% dependent upon the underlying condition) [39]. Chemotherapeutic drugs can adversely affect AED levels (and vice versa): indeed, the most common cause for SEGP is sub-therapeutic AED levels [16]. Because patients with HGG are more likely to receive chemotherapy, TAE progression to TASE may therefore be related. Finally, most patients with high grade tumours receive corticosteroids; at least in utero, excess corticosteroids can adversely alter post-natal seizure susceptibility [40]. The most common identifiable causes for SEGP are AED withdrawal, AED non-compliance, and alcohol-related factors [13,41–43].

32

J. Goonawardena et al. / Journal of Clinical Neuroscience 22 (2015) 29–34

It is unknown whether these factors are more prevalent in patients with high grade tumours than those with low grade tumours. Conceivably, cognitive dysfunction (incurred by diffusely destructive cerebral disease) or systemic effects related to adjuvant therapy (or to the effects of metastatic disease) may reduce compliance in patients with high grade tumours. However, a lower incidence of TASE in patients with TAE than SEGP with EGP would seem to count against this theory, especially given the dominance of high grade tumours amongst any brain tumour population. An alternative explanation relates to the development of sub-therapeutic AED levels within the epileptogenic lesion due to the expression of multi-drug resistance (MDR) proteins [44] (section 6.1), or to the development of AED toxicity – both appear greater in patients with high grade tumours. Ultimately, however, the explanation is likely related to the inexorably destructive nature of high grade tumours, and to the molecular mechanisms involved with tumour growth and progression [45], which progressively impair mechanisms associated either with spontaneous seizure termination, or with the stability of the post-seizure refractory period. The following is a brief discussion of molecular mechanisms currently understood with TAE, and with potential application to understanding TASE. 5.2.1. Glutamate receptors Glutamate is the principal excitatory neurotransmitter in the central nervous system and has been strongly implicated in the pathophysiology of TAE. All known glutamate receptor subtypes (a-amino-3-hydroxy-5-methyl-4-isoaxazolepropionate acid [AMPA-R], N-methyl-D-aspartate [nMDA-R], kainite and metabotropic) are expressed by glioma cells [20,21,45,46,47]. Importantly, Aronica et al. demonstrated increased NR2A and NR2B (stimulatory NMDA receptor subunits) expression in 41 patients with LGG compared to controls [46]. Further, van Vuurden et al. showed that oligodendrogliomas and astrocytomas exhibited greater AMPA-R expression than glioblastoma multiforme (GBM): that is, AMPA receptor expression appeared negatively correlated with WHO tumour grade [48]. In similar fashion, van Vuurden et al. demonstrated significant down-regulation of NMDA (NR1, NR2A– C, NR3A), AMPA (GluR1–4) and kainite (GluR5,7; KA2) receptors (ionotropic receptors) in GBM compared to controls [48]. Downregulation of AMPA-R, for example, enables GBM cells to survive a high glutamate environment without suffering excitotoxic cell death [48]. Increased expression of stimulatory receptors in LGG, with decreased expression in HGG, could therefore explain why TAE risk is inversely proportional to tumour grade. However, it is difficult to immediately reconcile this with simultaneously defective spontaneous seizure termination, or with a defective post-seizure refractory period, as is required to explain TASE. 5.2.2. G-amino-butyric acid receptors G-amino-butyric acid (GABA) is the principal inhibitory neurotransmitter in the central nervous system. GABA-A receptor (GABA-A-R) expression is increased in LGG compared to HGG [21]. Whilst GABA-A-R has been shown to be responsive to GABA in most LGG (71% oligodendrogliomas and 62% astrocytomas) in one study, they did not prove responsive in GBM [49], indicating that functional GABA-A-R expression appeared inversely proportional to WHO tumour grade [49]. Unfortunately, however, this result was not validated in another study, where GABA-A-R expression was increased in only 20% of LGG, and where GABA-A-R expression was actually decreased in a further 20% [37]. Down-regulation of GABA-A-R expression or assembly of nonfunctional GABA-A-R in HGG could therefore be a factor in explaining why TASE is more likely in HGG [49]. Unfortunately, however, the GABA agonist diazepam appears unusually efficacious in

suppressing SEGP and TASE when used with phenytoin (section 6.2.); GABA mechanisms must therefore be largely intact in both SEGP and TASE. 5.2.3. pH Numerous studies have focused on pH changes during both seizure onset and seizure termination in TAE. For example, Shamji et al. found increased glucose metabolism, ischaemia and interstitial acidosis in TAE; indeed, acidosis extended into the peritumoural environment where it was associated with glial swelling and structural damage [21]. Glial uptake of presynaptic glutamate is a major factor preventing synaptic glutamate accumulation [50]: glial cell acidosis could therefore encourage synaptic glutamate accumulation, prolonged excitation and seizure maintenance; and ultimately result in SEGP or TASE [4]. Damaged glial cells, necrosis, acidosis and haemosiderin deposition (especially epileptogenic [51]) are all histological features of GBM; however, why they should correlate positively with TASE, and yet negatively with TAE, is difficult to reconcile. 5.2.4. Other changes in HGG Saadoun et al. showed that aquaporin 1 was expressed in HGG where it was associated with vasogenic brain oedema [52,53]. During hypoxia, mitochondrial adenosine triphosphate synthesis is diminished, meaning Na+/K+ pump action is therefore impaired, and increased extracellular K+ concentration facilitates neuronal excitability. Because spontaneous seizure termination is an active inhibitory process, relative hypoxia within HGG therefore potentially prevents seizure termination [54]. 5.2.5. Myo-inositol levels Myo-inositol (MI) is located within astrocytes and its concentration is altered in many brain disorders [55]. Numerous studies have demonstrated increased MI levels in LGG compared to HGG [55–57]. Importantly, Solomonia et al. showed that MI significantly reduced the intensity and severity of seizures [58]. Importantly, Nozadze et al. also demonstrated that MI delayed the latent period for seizure onset and decreased seizure duration [59]. Thus, unlike all other mechanisms discussed in this section (which emphasize excitatory/inhibitory imbalances), MI levels address the oftenignored area of the refractory period in favouring TAE over TASE. The pathophysiological changes explained above support the finding TASE risk appears directly proportional to tumour grade. 6. TASE responsiveness to AED: Effect of tumour grade 6.1. AED refractoriness in TAE The prevalence of EGP that is refractory to AED ranges from 20– 40% [60]. However, AED refractoriness appears more prevalent in TAE than in EGP; for example, first-line AED fail in approximately 60% of patients with TAE [35,61]. Hildebrand et al. found that only 13% of patients with TAE gained complete seizure control with AED, even when considered in combination with direct antitumour treatment [29]. In another study of 99 patients predominantly with TAE, 63% of patients with HGG required AED polytherapy compared to only 36% with LGG [30]. Similar results have been found in other series: that is, that AED resistance is directly proportional to tumour grade in TAE [12,30,62]. Such a result suggests that mechanisms associated with suppressing spontaneous seizure onset are relatively more deficient in HGG. The implication is that patients with HGG should be treated more aggressively, perhaps with AED polytherapy from the outset, to potentially avoid TAE progression to TASE. As noted previously, TAE refractoriness is

J. Goonawardena et al. / Journal of Clinical Neuroscience 22 (2015) 29–34

more common in frontal locations than TAE arising elsewhere [27] – no explanation adequately accounts for this. Because AED resistance may be a factor in the subsequent development of TASE, several explanations for AED resistance in TAE are listed below: 1. Sub-therapeutic AED levels. Sub-therapeutic AED levels are apparent in 60–70% with TAE [22,61]. These may relate to pharmacodynamic interactions of AED with concurrent medication, or with altered plasma protein levels [22,61,63]. More prosaically, sub-therapeutic AED levels may relate to poor compliance. 2. Reduced AED efficacy. Several chemotherapeutic drugs (including methotrexate and cisplastin) have been shown to decrease the efficacy of both valproate and phenytoin [16]. 3. Tumour progression. Subsequent seizure recurrence, despite full AED compliance after presentation, may suggest tumour progression in patients with TAE (although the two events do not always coincide and may be unrelated). This may reflect changes at the molecular level within the peri-tumoural environment [61,63], including changes in receptors or receptorcoupling mechanisms that mediate refractoriness. 4. Expression of MDR proteins. Glycoprotein-P, an agent potentially involved in the exo-transportation of AED from the brain parenchyma, is increasingly expressed in tumours of higher grade [44]. This could lead to sub-therapeutic AED levels within the epileptic lesion, thereby increasing the risk of TASE. Additionally, long term toxic side effects of AED also appear to correlate directly with tumour grade [9]. The development of such side effects may encourage poor compliance with AED in patients with high grade tumours, and thereby increase the risk of TASE. However, if this is so, then one might expect a greater incidence of TASE in patients with TAE than of SEGP in patients with EGP; however, the converse is true. 6.2. AED responsiveness in TASE In the general population, AED refractoriness in SEGP appears to occur with a similar frequency to that in EGP (20–50%) [60,64,65]. As noted above, the relatively high AED refractoriness in both EGP and SEGP is even higher in TAE (60%). Given these facts, it is somewhat arresting that AED refractoriness appears less common in TASE (14–18%) than in TAE, EGP or SEGP; that is TASE appears curiously more responsive to simple AED regimes [29,35,60,61,64,65]. For example, Lowenstein et al. found that patients with TASE were the largest epilepsy sub-group (82%) to respond to first-line AED treatment (phenytoin and diazepam) [66]. The results of a systematic review by Neligan et al. also concurred [13]. Finally, Cavaliere et al. found that only 14% of patients with TASE could not be managed with benzodiazepines and phenytoin; the vast majority could be managed with these simple agents alone [10]. If representative, such TASE responsiveness to simple regimes is singular, but felicitous. Notwithstanding, apparent TASE responsiveness to simple regimes is difficult to reconcile mechanistically with established AED refractoriness in TAE. 7. Mortality in TASE Even non-convulsive SEGP is associated with a mortality of 18% [67]. In the systematic review of Neligan et al., mortality depended strongly upon the underlying cause of SEGP [13,15,23,68]. Thus, SEGP mortality ranged from 60–100% for patients with cerebral anoxia, from 20–60% for those with cerebrovascular disease (20–57%), from 10–35% for those with metabolic disorders, from

33

0–10% for patients with sub-therapeutic AED levels, and from 0–30% for those with acute central nervous system infections [13,15,42,68]. Empirically, TASE might be considered to possess a graver prognosis than SEGP, especially given the pre-eminence of high grade tumours. Indeed, many have reported a higher mortality rate with TASE compared to SEGP [15,23,42,66]: for example, Cavaliere et al. reported a 50% mortality in those with TASE and systemic cancer [10]. However, patients with systemic cancer in the latter study were older than those without cancer, and commonly had major comorbidities. Indeed, such associations frequently confound the study of TASE mortality. Cavaliere et al. found a 30 day TASE mortality of 12% in those with intrinsic brain tumours compared to 50% in those with cerebral metastases associated with systemic cancer, however a similar mortality was found between those with HGG and systemic cancer [10]. Blitshteyn et al. found a 30 day mortality of 50% in patients with central nervous system metastases and non-convulsive TASE, however this study consisted of only four patients [69]. In the systematic review of Neligan et al., TASE mortality at 30 days (23%) was, overall, similar to that of SEGP (20%) [13]. In consequence, most have concluded that (as with SEGP) the major determinant of TASE mortality is actually that of the underlying condition. Neligan et al. also emphasised the confounding factors inherent when analysing heterogeneous TASE sub-type comparison between series. Thus, a lower mortality in patients with primary brain tumours could simply reflect their propensity to elicit partial TASE (simple focal motor epilepsy, epilepsia partialis continua or complex partial seizures) rather than generalised convulsive TASE [13].

8. Conclusion Although preliminary, our review suggests potentially divergent pathophysiological mechanisms between TASE and TAE, with potential implications for the development of novel therapeutic regimes. Future prospective studies are required to validate our findings and to direct future research and management advances.

Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] Huff JS, Fountain NB. Pathophysiology and definitions of seizures and status epilepticus. Emerg Med Clin North Am 2011;29:1–13. [2] Chen JW, Naylor DE, Wasterlain CG. Advances in pathophysiology of status epilepticus. Acta Neurol Scand Suppl 2007;115:7–15. [3] Scott RC, Surtees RA, Neville BG. Status epilepticus: pathophysiology, epidemiology, and outcomes. Arch Dis Child 1998;79:73–7. [4] Lado FA, Moshé SL. How do seizures stop? Epilepsia 2008;49:1651–64. [5] Chen JW, Wasterlain CG. Status epilepticus: pathophysiology and management in adults. Lancet Neurol 2006;5:246–56. [6] Lowenstein DH, Bleck T, Macdonald RL. It’s time to revise the definition of status epilepticus. Epilepsia 1999;40:120–2. [7] Gastaut H. Classification of status epilepticus. Adv Neurol 1983;34:15–25. [8] Bleck TP. Convulsive disorders: status epilepticus. Clin Neuropharmacol 1991;14:191–8. [9] Moots PL, Maciunas RJ, Eisert DR, et al. The course of seizure disorders in patients with malignant gliomas. Arch Neurol 1995;52:717–24. [10] Cavaliere R, Farace E, Schiff D. Clinical implications of status epilepticus in patients with neoplasms. Arch Neurol 2006;63:1746–9. [11] Alajbegovic´ A, Loga N, Alajbegovic´ S, et al. Characteristics of symptomatic epilepsy in patients with brain tumours. Bosn J Basic Med Sci 2009;9:81–4. [12] Rosati A, Tomassini A, Pollo B, et al. Epilepsy in cerebral glioma: timing of appearance and histological correlations. J Neurooncol 2009;93:395–400. [13] Neligan A, Shorvon SD. Frequency and prognosis of convulsive status epilepticus of different causes: a systematic review. Arch Neurol 2010;67:931–40.

34

J. Goonawardena et al. / Journal of Clinical Neuroscience 22 (2015) 29–34

[14] Shah AM, Vashi A, Jagoda A. Review article: convulsive and non-convulsive status epilepticus: an emergency medicine perspective. Emerg Med Australas 2009;21:352–66. [15] DeLorenzo RJ, Hauser WA, Towne AR, et al. A prospective, population-based epidemiologic study of status epilepticus in Richmond, Virginia. Neurology 1996;46:1029–35. [16] Avila EK, Graber J. Seizures and epilepsy in cancer patients. Curr Neurol Neurosci Rep 2010;10:60–7. [17] Chin RF, Neville BG, Scott RC. A systematic review of the epidemiology of status epilepticus. Eur J Neurol 2004;11:800–10. [18] Cocito L, Audenino D, Primavera A. Altered mental state and nonconvulsive status epilepticus in patients with cancer. Arch Neurol 2001;58:1310. [19] Hauser WA, Annegers JF, Rocca WA. Descriptive epidemiology of epilepsy: contributions of population-based studies from Rochester, Minnesota. Mayo Clin Proc 1996;71:576–86. [20] Beaumont A, Whittle IR. The pathogenesis of tumour associated epilepsy. Acta Neurochir 2000;142:1–15. [21] Shamji MF, Fric-Shamji EC, Benoit BG. Brain tumors and epilepsy: pathophysiology and peritumoural changes. Neurosurg Rev 2009;32:275–85 [discussion 284–6]. [22] Brogna C, Gil Robles S, Duffau H. Brain tumors and epilepsy. Expert Rev Neurother 2008;8:941–55. [23] Aminoff MJ, Simon RP. Status epilepticus. Causes, clinical features and consequences in 98 patients. Am J Med 1980;69:657–66. [24] Janz D. Status epilepticus and frontal lobe lesions. J Neurol Sci 1964;1: 446–57. [25] Rajneesh KF, Binder DK. Tumor-associated epilepsy. Neurosurg Focus 2009; 27:E4. [26] Zaatreh MM, Spencer DD, Thompson JL, et al. Frontal lobe tumoral epilepsy: clinical, neurophysiologic features and predictors of surgical outcome. Epilepsia 2002;43:727–33. [27] Oxbury JM, Whitty CW. The syndrome of isolated epileptic status. J Neurol Neurosurg Psychiatry 1971;34:182–4. [28] Oxbury JM, Whitty CW. Causes and consequences of status epilepticus in adults. A study of 86 cases. Brain 1971;94:733–44. [29] Hildebrand J, Lecaille C, Perennes J, et al. Epileptic seizures during follow up of patients treated for primary brain tumours. Neurology 2005;65:212–5. [30] van Breemen MS, Rijsman RM, Taphoorn MJ, et al. Efficacy of anti-epileptic drugs in patients with gliomas and seizures. J Neurol 2009;256:1519–26. [31] Hormigo A, Liberato B, Lis E, et al. Nonconvulsive status epilepticus in patients with cancer: imaging abnormalities. Arch Neurol 2004;61:362–5. [32] Janz D. Conditions and causes of status epilepticus. Epilepsia 1961;2:170–7. [33] van Breemen MS, Wilms EB, Vecht CJ. Epilepsy in patients with brain tumours: epidemiology, mechanisms, and management. Lancet Neurol 2007;6:421–30. [34] Villemure JG, de Tribolet N. Epilepsy in patients with central nervous system tumors. Curr Opin Neurol 1996;9:424–8. [35] Vecht CJ, Van Breemen M. Optimizing therapy of seizures in patients with brain tumors. Neurology 2006;67:S10–3. [36] Smith KC. The management of seizures in brain tumor patients. J Neurosci Nurs 2010;42:28–37. [37] Wolf HK, Roos D, Blümcke I, et al. Perilesional neurochemical changes in focal epilepsies. Acta Neuropathol 1996;91:376–84. [38] Drevelegas A, Karkavelas G. High grade gliomas. In: Develegas A, editor. Imaging of brain tumours with histological correlations. Berlin Heidelberg: Springer; 2011. p. 157–200. [39] Foy PM, Copeland GP, Shaw MD. The incidence of post-operative seizures. Acta Neurochir (Wien) 1981;55:253–64. [40] Velíšek L. Prenatal exposure to excess corticosteroids alters postnatal seizure susceptibility. Epilepsy Res 2011;95:9–19. [41] Sagduyu A, Tarlaci S, Sirin H. Generalized tonic-clonic status epilepticus: causes, treatment, complications and predictors of case fatality. J Neurol 1998;245:640–6.

[42] Towne AR, Pellock JM, Ko D, et al. Determinants of mortality in status epilepticus. Epilepsia 1994;35:27–34. [43] Varelas PN, Mirski MA. Status epilepticus. Curr Neurol Neurosci Rep 2009;9:469–76. [44] Rogawski MA. Does P-glycoprotein play a role in pharmacoresistance to antiepileptic drugs? Epilepsy Behav 2002;3:493–5. [45] Yuen TI, Morokoff AP, Bjorksten A, et al. Glutamate is associated with a higher risk of seizures in patients with gliomas. Neurology 2012;79:883–9. [46] Aronica E, Yankaya B, Jansen GH, et al. Ionotropic and metabotropic glutamate receptor protein expression in glioneuronal tumours from patients with intractable epilepsy. Neuropathol Appl Neurobiol 2001;27:223–37. [47] de Groot J, Sontheimer H. Glutamate and the biology of gliomas. Glia 2011;59:1181–9. [48] van Vuurden DG, Yazdani M, Bosma I, et al. Attenuated AMPA receptor expression allows glioblastoma cell survival in glutamate-rich environment. PLoS One 2009;4:e5953. [49] Labrakakis C, Patt S, Hartmann J, et al. Functional GABA(A) receptors on human glioma cells. Eur J Neurosci 1998;10:231–8. [50] Benarroch EE. Neuron-astrocyte interactions: partnership for normal function and disease in the central nervous system. Mayo Clin Proc 2005;80:1326–38. [51] Baumann CR, Schuknecht B, Lo Russo G, et al. Seizure outcome after resection of cavernous malformations is better when surrounding hemosiderin-stained brain also is removed. Epilepsia 2006;47:563–6. [52] Saadoun S, Papadopoulos MC, Davies DC, et al. Increased aquaporin 1 water channel expression in human brain tumours. Br J Cancer 2002;87:621–3. [53] Nag S, Manias JL, Stewart DJ. Pathology and new players in the pathogenesis of brain edema. Acta Neuropathol 2009;118:197–217. [54] Holtkamp M. Pathophysiology of status epilepticus. Epileptologie 2009;26:59–65. [55] Castillo M, Smith JK, Kwock L. Correlation of myo-inositol levels and grading of cerebral astrocytomas. AJNR Am J Neuroradiol 2000;21:1645–9. [56] Barton SJ, Howe FA, Tomlins AM, et al. Comparison of in vivo 1H MRS of human brain tumours with 1H HR-MAS spectroscopy of intact biopsy samples in vitro. MAGMA 1999;8:121–8. [57] Cheng LL, Chang IW, Louis DN, et al. Correlation of high-resolution magic angle spinning proton magnetic resonance spectroscopy with histopathology of intact human brain tumor specimens. Cancer Res 1998;58:1825–32. [58] Solomonia R, Nozadze M, Kuchiashvili N, et al. Effects of myo-inositol on convulsions induced by pentylenetetrazole and kainic acid in rats. Bull Exp Biol Med 2007;143:58–60. [59] Nozadze M, Mikautadze E, Lepsveridze E, et al. Anticonvulsant activities of myo-inositol and scyllo-inositol on pentylenetetrazol induced seizures. Seizure 2011;20:173–6. [60] French JA. Refractory epilepsy: clinical overview. Epilepsia 2007;48:3–7. [61] Riva M. Brain tumoral epilepsy: a review. Neurol Sci 2005;26:40–2. [62] Lund M. Epilepsy in association with intracranial tumour. Acta Psychiatr Neurol Scand Suppl 1952;81:1–149. [63] Schaller B, Rüegg SJ. Brain tumor and seizures: pathophysiology and its implications for treatment revisited. Epilepsia 2003;44:1223–32. [64] Mayer SA, Claassen J, Lokin J, et al. Refractory status epilepticus: frequency, risk factors, and impact on outcome. Arch Neurol 2002;59:205–10. [65] Holtkamp M, Othman J, Buchheim K, et al. Predictors and prognosis of refractory status epilepticus treated in a neurosurgical intensive care unit. J Neurol Neurosurg Psychiatry 2005;76:534–9. [66] Lowenstein DH, Alldredge BK. Status epilepticus at an urban public hospital in the 1980s. Neurology 1993;43:483–8. [67] Shneker BF, Fountain NB. Assessment of acute morbidity and mortality in nonconvulsive status epilepticus. Neurology 2003;61:1066–73. [68] Sung CY, Chu NS. Status epilepticus in the elderly: etiology, seizure type and outcome. Acta Neurol Scand 1989;80:51–6. [69] Blitshteyn S, Jaeckle KA. Nonconvulsive status epilepticus in metastatic CNS disease. Neurology 2006;66:1261–3.