Nonconvulsive Status Epilepticus in Adults PDF

31 Nonconvulsive Status Epilepticus in Adults Sara Hocker and Peter W. Kaplan IN THIS CHAPTER Approach to the diagnosis of nonconvulsive status epilepticus (NCSE) including controversial EEG patterns Assessing the neurological impact of...

31 Nonconvulsive Status Epilepticus in Adults Sara Hocker and Peter W. Kaplan IN THIS CHAPTER Approach to the diagnosis of nonconvulsive status epilepticus (NCSE) including controversial EEG patterns Assessing the neurological impact of NCSE: surrogate markers of brain dam- age Determining how aggressively to treat NCSE KEY POINTS Nonconvulsive status epilepticus (NCSE) is not one entity with a fixed prog- nosis and management paradigm, but rather a heterogeneous group of dis- orders with variable outcomes and treatment strategies. Paramount to decision making is differentiation of rhythmic or evolving activ- ity that should be identified as seizures or status epilepticus from periodic, rhythmic, or interictal epileptiform activity that may not be responsible for the clinical state and, therefore, may not warrant treatment. Management decisions should include careful consideration of the ultimate prognosis of the individual patient (e.g., postcardiorespiratory arrest; or mul- tiple organ failure in a patient with serious comorbidities vs. subtherapeutic antiseizure drug levels in a patient with epilepsy). Treatment should include utilization of nonanesthetic anti-seizure drugs with careful and expeditious titration of medication to clinical and electrographic response while monitoring for adverse effects. Anesthetic drugs may be nec- essary and appropriate in refractory cases thought to have a favorable prog- nosis. Treatment is optimally conducted in an ICU with continuous EEG monitoring, in addition to particular attention to cardiorespiratory complications. 298 Copyright Springer Publishing Company. All Rights Reserved. From: Handbook of ICU EEG Monitoring, Second Edition DOI: 10.1891/9780826168627.0031  Chapter 31 Nonconvulsive Status Epilepticus in Adults 299 I. BACKGROUND A. Nonconvulsive status epilepticus Advances in neurointensive care EEG monitoring have uncovered many patients with nonconvulsive status epilepticus (NCSE) who were previously undiagnosed. NCSE is not a single entity with diagnosis and treatment options that can be gen- eralized to all patients. Rather, it is a group of heterogeneous disorders that are diagnosed and managed in a number of different clinical settings. The prognosis of NCSE is dependent on the type and etiology of NCSE and patient comorbidities. Some types are benign (i.e., genetic absence epilepsy [typical absence status epilepticus] which does not result in lasting morbidity). At the opposite extreme, electrographic status epilepticus (SE) with coma after cardiorespiratory arrest has an almost 100% mortality or nonreturn to con- sciousness (particularly if not managed with hypothermia). Diagnosis of NCSE is now being approached along three axes (1) Ictal semiology Etiology EEG patterns II. BASICS A. Animal models of NCSE Animal models have not provided conclusive evidence of the potential for brain damage as a result of NCSE. No satisfactory model exists for typical absence status or for NCSE triggered by low antiseizure drug (ASD) levels in the setting of chronic epilepsy. Many animal models initially used for studying the effects of SE are created by inducing significant brain insults that themselves cause chronic neuronal damage. Many models were initially used to investigate excitotoxicity, but were later also used “to replicate” complex partial status epilepticus (CPSE) in humans (2). In these models, SE was induced in nonepileptic animals using powerful chemoconvulsants or prolonged high-frequency repetitive stimulation (2). B. Studies in humans with NCSE Studies evaluating the effects of NCSE in humans are difficult to interpret. NCSE in humans is very different, usually involving lower frequency discharges. Anti-seizure drugs and prior epilepsy may confer some neuroprotection. It can be very difficult to distinguish damage resulting from the cause of SE from damage resulting from the seizures themselves (2,3). Many case reports and small series in the literature include patients with NCSE from a wide variety of causes including preceding convulsive SE, encephalitis, stroke, or traumatic brain injury. Few studies have directly measured possible cognitive consequences of NCSE. Two studies of epilepsy patients who had cognitive evaluations after SE found no permanent decline in cognitive measures (4,5). 300 Part IV Treatment In contrast to this reassuring finding: Two other studies investigated “pure” groups of patients with NCSE from a single etiology along with a control population. Results suggest at least an additive adverse effect of seizures or SE on the ini- tial insult in patients with subarachnoid hemorrhage or head trauma (6). C. Surrogates of brain damage Imaging of patients during and after NCSE: Studies have demonstrated varying results. In patients with known idiopathic generalized epilepsy (childhood absence or juvenile myoclonic epilepsy) there is no evidence of CT or MRI changes following NCSE in patients who have not suffered additional insults from anoxia or trauma. Patients with a history of localization-related epilepsy and NCSE: Most cases show no lasting changes on imaging. During the acute phase, MRI may show transient focal fluid-attenuated inversion recovery (FLAIR) or diffusion-weighted imaging (DWI) changes. There are clear reported cases of temporal lobe and hippocampal atrophy after temporal lobe NCSE, but it is not entirely clear whether these cases were triggered by a temporal lobe insult such as encephalitis. Other case series of NCSE in the ICU setting have shown there is imaging evi- dence of cerebral damage associated with NCSE, but without clear indication that the imaging changes are consequent to the NCSE proper (7). Serum neuron-specific enolase (NSE) level as a surrogate marker of brain damage: NSE is a serum marker of neuronal injury. May rise during NCSE even without a concurrent, identifiable neurologic lesion. NSE levels have been shown to rise following CPSE, partially related to the duration of SE (8). Doubt has been raised regarding the specificity of these changes, as NSE may appear as a result of blood–brain barrier breakdown, and not necessarily reflect acute neuronal damage. The reliability of this test as a reflection of brain damage is debatable: in pub- lished series, some patients with good outcome had significantly increased NSE levels, whereas others with poor outcome had low levels. D. Risks of treatment Clinical experience and the literature have informed us that treatment does not come without risk. Treatment (particularly with benzodiazepines and anesthetics) can cause respi- ratory suppression, hypotension, prolonged ICU stays, and are associated with concurrent infection and death (9). Such treatment cannot and should not be provided with a belief that it is stan- dard therapy, and that there is “known safety and efficacy” uniformly. A “one size fits all” treatment paradigm may tip the risk–benefit ratio toward net harm (3). E. How to approach the diagnosis and treatment of NCSE Several factors may figure in treatment decisions. The degree of certainty regarding the diagnosis of NCSE The perceived prognosis for the particular etiology implicated in the case Whether treatment will improve prognosis or prevent deterioration The refractoriness of the seizures  Chapter 31 Nonconvulsive Status Epilepticus in Adults 301 The patient’s advance directives or the surrogate decision maker’s perception of the patient’s goals of care Clinical presentation of NCSE The diagnosis of NCSE in large part depends on the clinician having a suspi- cion of seizures, triggering a request for an EEG. Following convulsive SE, the patient may enter a state referred to as “subtle SE” in which it is hypothesized that there has been neuronal exhaustion pro- ducing an electroclinical dissociation. An example is a 70-year-old woman who underwent cardiac surgery, and upon holding sedatives had a generalized tonic–clonic seizure. Despite cessation of convulsive activity, she did not awaken. A head CT showed multifocal strokes and an EEG showed frequent multifocal independent seizures representing intermittent focal SE. This pattern gradually subsided over the next 2 days (Figure 31.1). (A) (B) FIGURE 31.1 Focal electrographic seizure arising from the right parietal–occipital region. (A) with sudden offset and independent focal seizure arising from the left parietal–occipital region (B). 302 Part IV Treatment NCSE in comatose patients in the ICU may present with subtle face, eye, head, neck, or limb myoclonus; changes in body or limb tone; or eye deviation and nystagmus. At times, there are few clinical indicators of NCSE, while seizures are clearly seen on EEG. For example, a 46-year-old woman with type 1 diabetes became pro- gressively less responsive in the setting of diabetic ketoacidosis. An EEG showed a generalized spike and wave pattern, frequently faster than 3 Hz (Figure 31.2A), which readily regressed after lorazepam (Figure 31.2B). However, accurate interpretation of the EEG can pose further hurdles in the confirmation of NCSE. EEG diagnosis of NCSE Unfortunately, EEG misdiagnosis of NCSE is common. Inaccuracies range from an incorrect interpretation of EEG artifact (e.g., due to bed vibration, physical therapy tapping, chewing of food) to the overinterpreta- tion of interictal and periodic or rhythmic discharges. Previously known as triphasic waves, generalized periodic discharges (GPDs) with triphasic morphology in the setting of toxic–metabolic encephalopathy may resemble ictal EEG patterns. Furthermore, triphasic morphology is no lon- ger thought to be specific to toxic–metabolic causes of encephalopathy. Much effort has been directed toward identifying EEG patterns believed to be diagnostic for NCSE (1). The clearest EEG evidence of NCSE includes the appearance of epileptiform morphologies that evolve in frequency, amplitude, or location, OR have per- sistent frequencies faster than 2.5 Hz (Figure 31.3A, B). (A) FIGURE 31.2 Continuous 2- to 4-Hz generalized atypical spike-waves (A). (continued )  Chapter 31 Nonconvulsive Status Epilepticus in Adults 303 (B) FIGURE 31.2 (continued ) This pattern ceased after 4 mg of lorazepam was given with transition of the background to diffuse slowing in the delta range with moderately low amplitudes (B). (A) FIGURE 31.3 Focal onset electrographic seizure maximal over the right central region which evolves in frequency (A), and then ends abruptly in (B). (continued ) 304 Part IV Treatment (B) FIGURE 31.3 (continued ) Additionally, based on the Salzburg Criteria (10), epileptiform discharges less than or equal to 2.5 Hz would meet criteria for NCSE if they exhibit either: An associated clinical correlate (e.g., limb jerking, facial, axial or limb myoc- lonus, gaze or head deviation, nystagmus) (see ebook for Figure S-31.1) or Clinical and EEG improvement after administering an IV ASD. The need to establish clear EEG evidence of NCSE risks missing an interme- diate segment of patients lying along the ictal–interictal continuum, which typ- ically include periodic or rhythmic patterns in the 1- to 2.5-Hz frequency range that do not meet the previous criteria (Figure 31.4). Although periodic discharges (see ebook for Figures S-31.2 and S-31.3) are known to be associated with seizures and reflect cortical irritability, these patterns are typically seen in temporal proximity to seizures—before or after—and not thought to represent “active” seizures, while others regard these patterns as the end stage of SE. Thus, much controversy surrounds whether periodic discharges should be sup- pressed with parenteral ASDs, especially given the morbidity associated with intensive anesthetic management (3). Various strategies for the management of these controversial patterns have been proposed (11). Prognosis as a factor guiding management of NCSE Individual prognosis largely depends on semiology, level of consciousness, eti- ology of NCSE, and, in part, on EEG patterns (e.g., absence of background activity foreboding more marked underlying brain dysfunction) (Table 31.1).  Chapter 31 Nonconvulsive Status Epilepticus in Adults 305 FIGURE 31.4 Right hemisphere LPDs at a frequency of 1 Hz. LPDs, lateralized periodic discharges. For example, patients whose proximate cause of NCSE is low ASD levels in chronic epi- lepsy have low morbidity and mortality, in the range of 3% (13). Conversely, a typical academic hospital will see a mortality rate of 30% to 50% in critically ill patients with NCSE as a result of an acute, severe brain insult. F. Suggestions on how to approach treatment (Table 31.2) Treatment is optimally conducted in an ICU with continuous EEG monitoring and intensivist support. Clinician must titrate ASD therapy to the patient’s clinical and EEG response while monitoring carefully for adverse effects. Drugs used include: Oral ASDs Benzodiazepines with potential respiratory and blood pressure suppressant effects Parenteral use of phenytoin, valproate, or newer agents such as levetiracetam or lacosamide Anesthetic agents such as propofol, midazolam, barbiturates, or ketamine Tailor treatment intensity to the perceived morbidity of the particular etiology and clinical setting of the case in question. Use level of consciousness, age, patient comorbidities (such as dementia or metastatic cancer), and probable outcome as further guideposts. Avoid lowering the patient’s level of consciousness to a level significantly below that present from the NCSE itself.  306 TABLE 31.1 Prognosis in Nonconvulsive Status Epilepticus TYPES PROGNOSIS DIAGNOSIS RESPONSE TO RECURRENCE OUTCOME ASDs Part IV TAS Excellent Frequently Excellent Frequent No morbidity or mortality missed De novo absence status Excellent Frequently Good, but Occasionally (situation Excellent Treatment in the elderly missed sometimes delayed rated; triggers can be removed) Absence status Guarded to fair Less Variable Frequent Variable, some with with degenerative frequently cognitive decline (difficulty generalized epilepsies missed in determining whether this is due to disease or to episodes of AASE) AASE Fair to poor Less Relatively refractory Frequent Frequent cognitive frequently (when seen in the morbidity, but it is missed setting of epileptic difficult to differentiate encephalopathy/ this from the effects of mental retardation) disease progression and consequences Simple partial Usually good Frequently Excellent Frequent Morbidity and mortality nonconvulsive status to excellent, missed negligible to absent epilepticus occasionally poor (continued )  TABLE 31.1 Prognosis in Nonconvulsive Status Epilepticus (continued ) TYPES PROGNOSIS DIAGNOSIS RESPONSE TO RECURRENCE OUTCOME ASDs Complex partial status When not Less Good to very good, Frequent Only very rare cognitive epilepticus associated with frequently but often delayed sequelae (