Posted by Kostas Pantremenos in Gait Rehabilitation - Foot Drop on May 24, 2022
The likelihood of regaining independent walking after stroke influences rehabilitation and hospital discharge planning.
This study aimed to develop and internally validate a tool to predict whether and when a patient will walk independently in the first 6 months post-stroke.
Adults with stroke were recruited if they had new lower limb weakness and were unable to walk independently. Clinical assessments were completed one week post-stroke. The primary outcome was time post-stroke by which independent walking (Functional Ambulation Category score ≥ 4) was achieved. Cox hazard regression identified predictors for achieving independent walking by 4, 6, 9, 16, or 26 weeks post-stroke. The cut-off and weighting for each predictor was determined using β-coefficients. Predictors were assigned a score and summed for a final TWIST score. The probability of achieving independent walking at each time point for each TWIST score was calculated.
We included 93 participants (36 women, median age 71 years). Age < 80 years, knee extension strength Medical Research Council grade ≥ 3/5, and Berg Balance Test < 6, 6 to 15, or ≥ 16/56, predicted independent walking and were combined to form the TWIST prediction tool. The TWIST prediction tool was at least 83% accurate for all time points.
The TWIST tool combines routine bedside tests at one week post-stroke to accurately predict the probability of an individual patient achieving independent walking by 4, 6, 9, 16, or 26 weeks post-stroke. If externally validated, the TWIST prediction tool may benefit patients and clinicians by informing rehabilitation decisions and discharge planning.
Posted by Kostas Pantremenos in Artificial intelligence, REHABILITATION on April 1, 2022
Background: Machine learning (ML) is an artificial intelligence technique in which a system learns patterns and rules from a given data. Objectives: The objective of the study was to investigate the potential of ML for predicting motor recovery in stroke patients. Methods: We analyzed data from 833 consecutive stroke patients using 3 ML algorithms: deep neural network (DNN), random forest, and logistic regression. We created a practical ML model using the most common data measured in almost all rehabilitation hospitals as input data. Demographic and clinical data, including modified Brunnstrom classification (MBC) and functional ambulation classification (FAC), were collected when patients were transferred to the rehabilitation unit (8–30 days) and 6 months after stroke onset and were used as input data. Motor outcomes at 6 months after stroke onset of the affected upper and lower extremities were classified according to MBC and FAC, respectively. Patients with an MBC of <5 and an FAC of <4 at 6 months after stroke onset were considered to have a “poor” outcome, whereas those with MBC ≥5 and FAC ≥4 were considered to have a “good” outcome. Results: The area under the curve (AUC) for the DNN model for predicting motor function was 0.836 for the upper and lower limb motor functions. For the random forest and logistic regression models, the AUCs were 0.736 and 0.790 for the upper and lower limb motor functions, respectively. The AUCs for the random forest and logistic regression models were 0.741 and 0.795 for the upper and lower limb motor functions, respectively. Conclusion: Although we used simple and common data that can be obtained in clinical practice as variables, our DNN algorithm was useful for predicting motor recovery of the upper and lower extremities in stroke patients during the recovery phase.
Posted by Kostas Pantremenos in Epilepsy, Video on January 31, 2022
For people who live with epilepsy, it can be an agonizing question: when will the next seizure come. But a new treatment is providing a better answer — and relief.
Posted by Kostas Pantremenos in Hemianopsia on June 7, 2021
To determine any factors that predict how an individual will adapt to post-stroke hemianopic visual field loss, with close monitoring of the adaptation process from an early stage.
The Hemianopia Adaptation Study (HAST) is a prospective observational longitudinal cohort clinical study. Adult stroke survivors (n = 144) with new onset homonymous hemianopia were monitored using standardised mobility assessment course (MAC) as the primary outcome measure of adaptation.
Several baseline variables were found to be good predictors of adaptation. Three variables were associated with adaptation status at 12-weeks post-stroke: inferior % visual field, % total MAC omissions, and MAC completion time (seconds). Baseline measurements of these variables can predict the adaptation at 12 weeks with moderate to high accuracy (area under ROC curve, 0.82, 95% CI 0.74–0.90). A cut-off score of ≤25% target omissions is suggested to predict which individuals are likely to adapt by 12-weeks post-stroke following gold standard care.
Adaptation to hemianopia is a personal journey with several factors being important for prediction of its presence, including MAC outcomes and extent of inferior visual field loss. A clinical recommendation is made for inclusion of the MAC as part of a functional assessment for hemianopia.
Posted by Kostas Pantremenos in Epilepsy on December 23, 2020
By Rich Haridy December 20, 2020
An international study is showing, for the first time, that it may be possible to predict the onset of epileptic seizures several days in advance. By analyzing data from a clinically approved brain implant designed to monitor and prevent seizures, the new research hopes to develop a model offering patients with epilepsy a seizure forecasting tool to predict the likelihood of upcoming episodes.
The research looked at data from a responsive brain stimulation implant called NeuroPace. The device was approved for clinical uses back in 2013 and it works to prevent seizures by delivering imperceptible pulses of electrical stimulation to certain parts of the brain upon detecting abnormal brain activity.
Scientists have been working on a variety of seizure prediction tools for decades. But despite some incredible advances, such as the NeuroPace device, no innovation to date has successfully shown it possible to predict seizures more than a few minutes in advance, at best.
The NeuroPace innovation offers researchers the first chance to study the relationship between seizures and brain activity using years of EEG data. The new study initially analyzed long-term data from 18 patients with the brain implant who were closely tracked for several years. From this data the researchers developed predictive algorithms to forecast seizures. These predictive algorithms were then tested on long-term data gathered from the more than 150 people who participated in the decade-long clinical trials testing the brain implant system.
Vikram Rao, co-senior author on the new study, says the data shows seizure risk could be effectively forecasted three days ahead in nearly 40 percent of subjects and one day ahead in 66 percent of subjects.
“For forty years, efforts to predict seizures have focused on developing early warning systems, which at best could give patients warnings just a few seconds or minutes in advance of a seizure,” says Rao. “This is the first time anyone has been able to forecast seizures reliably several days in advance, which could really allow people to start planning their lives around when they’re at high or low risk.”
Rao does stress the current algorithm can only predict when one is at higher risk of seizure, and not specifically when a seizure will take place. A number of other unaccounted environmental triggers, from stress to erratic sleep, can play a role in the onset of a seizure. So the system currently developed is more like a weather forecast, offering probabilities designed to help guide a person’s future activities.
“I don’t think I’m ever going to be able to tell a patient that she is going to have a seizure at precisely 3:17 pm tomorrow—that’s like predicting when lightning will strike,” explains Rao. “But our findings in this study give me hope that I may someday be able to tell her that, based on her brain activity, she has a 90 percent chance of a seizure tomorrow, so she should consider avoiding triggers like alcohol and refrain from high-risk activities like driving.”
Much more work is needed before the system is ready for clinical use. This preliminary study uncovered a significant amount of variability from person to person. It is unclear why reliable forecasting could not be generated from some patient’s brain activity data. Future investigations to optimize the algorithm and perhaps incorporate multimodal physiological data may enhance the algorithm’s predictive capacity.
Plus, currently the system requires data gathered from a device requiring surgical implantation. This would limit the use of the device to only those with the most severe forms of epilepsy. More superficial subscalp EEG devices could offer a less invasive way of capturing this brain activity data over long periods of time.
“It is worth remembering that, currently, patients have absolutely no information about the future—which is like having no idea what the weather tomorrow might be—and we think our results could help significantly reduce that uncertainty for many people,” adds Rao. “Truly determining the utility of these forecasts, and which patients will benefit most, will require a prospective trial, which is the next step.”
The new study was published in the journal The Lancet Neurology.
Source: UCSF
Posted by Kostas Pantremenos in Paretic Hand, REHABILITATION, tDCS/rTMS on October 30, 2020
Recovery from motor paralysis is facilitated by affected patients’ recognition of the need for and practice of their own exercise goals. Neurorehabilitation has been proposed and used for the treatment of motor paralysis in stroke, and its effect has been verified. If an expected score for the neurorehabilitation effect can be calculated using the Fugl-Meyer Motor Assessment (FMA), a global assessment index, before neurorehabilitation, such a score will be useful for optimizing the treatment application criteria and for setting a goal to enhance the treatment effect. Therefore, this study verified whether the responsiveness to a treatment method, the NovEl intervention using repetitive transcranial magnetic stimulation and occupational therapy (NEURO), in patients with post-stroke upper extremity (UE) motor paralysis could be predicted by the pretreatment FMA score. No control group was established in this study for NEURO treatment. To analyze the recovery of the motor function in the UE, delta-FMA was calculated from the pre- and post-FMA scores obtained during NEURO treatment. The probability of three levels of treatment responsiveness was evaluated in association with delta-FMA score (<5, 5 ≤ delta-FMA <10, and ≥10 as non-responders; responders; and hyper-responders, respectively) according to the reported minimal clinically important difference (MCID). The association of the initial FMA scores with post-FMA scores, from the status of the treatment responsiveness, was determined by multinomial logistic regression analysis. Finally, 1,254 patients with stroke, stratified by FMA scores were analyzed. About 45% of the patients who had FMA scores ranging from 30 to 40 before treatment showed improvement over the MCID by NEURO treatment (odds ratio = 0.93, 95% CI = 0.92–0.95). Furthermore, more than 25% of the patients with more severe initial values, ranging from 26 to 30, improved beyond the MCID calculated in the acute phase (odds ratio = 0.87, 95% CI = 0.85–0.89). These results suggest that the evaluated motor function score of the UE before NEURO treatment can be used to estimate the possibility of a patient recovering beyond MCID in the chronic phase. This study provided clinical data to estimate the effect of NEURO treatment by the pretreatment FMA-UE score.
Motor paralysis due to the aftereffects of stroke impairs the activities of daily living (ADL) and quality of life (QOL) of patients; it also affects their individual or social activities (1, 2). In particular, motor paralysis of the upper extremity has a large impact on ADL (3). Recovery from motor paralysis is facilitated by patients recognizing the need for and practicing their own exercise goals (4). The type of goals that patients set are related to their goal satisfaction scores, with impairment-based goals being rated significantly higher than activity-based and participation-based goals (5). It is known that patients’ level of knowledge of their rehabilitation goals leads to effective treatment results (6). Thus, clinicians and patients are active partners in setting goals within stroke rehabilitation (5). In previous studies, some prognosis prediction systems were developed for motor paralysis (7–9), and they have been used to set goals for rehabilitation in patients with stroke.
Neurorehabilitation has been proposed and used for the treatment of motor paralysis in stroke, and its effect has been verified (10–14). One of the treatment methods, the NovEl intervention Using Repetitive transcranial magnetic stimulation and Occupational therapy (NEURO), facilitates peripheral muscle movement by controlling the excitability of the motor cortices by repetitive transcranial magnetic stimulation (rTMS). It also promotes peripheral muscle exercise and practice, for the active use of the paralyzed upper extremity (15, 16). NEURO’s efficacy has been proved in a randomized controlled study (17). To date, many patients have been treated by using NEURO; however, the prediction regarding whether patients’ recovery from motor paralysis after treatments can be predicted before treatment, has not been verified. If the Fugl-Meyer Motor Assessment (FMA) score before treatment can be used to predict NEURO treatment response, the score can be used as an effective goal for rehabilitation, by patients and therapists.
The minimal clinically important difference (MCID) of motor paralysis in the upper extremity has been investigated (18–20). If the expected value of an effect exceeding MCID can be calculated using FMA score measured before NEURO treatment, such a value will be useful for optimizing the treatment application criteria and setting a goal to enhance the treatment effect. For that purpose, it is sufficient to retroactively analyze the band of the FMA score before NEURO for a patient who is significantly improved. Therefore, this study verified whether the responsiveness of NEURO treatment for patients with post-stroke upper extremity motor paralysis could be predicted by the pre-treatment FMA score.[…]
Posted by Kostas Pantremenos in Educational, Epilepsy on July 6, 2020
Stroke is one of the commonest causes of seizures and epilepsy, mainly among the elderly and adults. This seminar paper aims to provide an updated overview of post-stroke seizures and post-stroke epilepsy (PSE) and offers clinical guidance to anyone involved in the treatment of patients with seizures and stroke. The distinction between acute symptomatic seizures occurring within seven days from stroke (early seizures) and unprovoked seizures occurring afterwards (late seizures) is crucial regarding their different risks of recurrence. A single late post-stroke seizure carries a risk of recurrence as high as 71.5% (95% confidence interval: 59.7-81.9) at ten years and is diagnostic of PSE. Several clinical and stroke characteristics are associated with increased risk of post-stroke seizures and PSE. So far, there is no evidence supporting the administration of antiepileptic drugs as primary prevention, and evidence regarding their use in PSE is scarce.
Neurologists frequently encounter seizures related to stroke. Given that post-stroke epilepsy (PSE) is the most common form of acquired epilepsy, it is quite surprising that it has attracted little academic interest and that, so far, only scarce evidence is available to guide clinical practice. The scenario is, however, changing and both basic science researchers and clinical investigators have started to address highly relevant issues, including pathophysiology, prevalence and incidence, diagnosis, prevention, treatment, and prognosis.
When should PSE be diagnosed? When should treatment start? Which is the most effective treatment? Which antiepileptic drug (AED) should be preferred? Not all seizures occurring after stroke are necessarily stroke-related. So, when should PSE be diagnosed? Only an adequate knowledge and correct diagnosis may spare patients the anxiety, stigma, and side effects of unnecessary treatments.
In recent years, interesting developments in the field of epileptogenesis also suggests that the risk of PSE may be modified through pharmacological intervention. In addition, the association found between PSE and risk of vascular events highlights the importance of secondary stroke prophylaxis.
This seminar paper aims to provide an updated overview of PSE and offers clinical guidance to professionals involved in the treatment of patients with post-stroke seizures.
A distinction between early and late post-stroke seizures is mandatory in the field of seizures and stroke since it underscores different pathophysiological mechanisms. Early post-stroke seizures reflect an acute, and perhaps reversible cerebral injury (i.e. acute symptomatic, provoked) whereas late seizures arise from long-lasting changes in the post-stroke brain (i.e. remote symptomatic, unprovoked). In this article, we will use the terms “early and late post-stroke seizures” throughout. The distinction between early and late seizures is closely linked to the theoretical concept of epileptogenesis -the hereto incompletely characterized process by which the brain acquires an enduring predisposition to seizures. Epileptogenesis does not simply represent a process that starts at stroke onset and manifests with seizures at a later stage, but should be considered within the frame of a threshold model in which individual predisposition, stroke characteristics, and subsequent reactions to the primary injury converge with PSE.
The temporal limit to consider a seizure as a “late-seizure” ranges mostly between one and two weeks after stroke, in analogy with the concept of early and late post-traumatic seizures. A pivotal study in Rochester, Minnesota demonstrated that a seizure within seven days of a stroke carries a ten-year risk of a subsequent unprovoked seizure of 33% (95% confidence interval [CI]: 20.7-49.9), whereas a seizure occurring seven days after a stroke carries a ten-year risk of 71.5% (95% CI: 59.7-81.9) (Hesdorffer et al., 2009). Based on such differences in patient prognosis, seven days is currently the recommended cut-off for considering a post-stroke seizure as early or late (Beghi et al., 2010). According to the most recent diagnostic criteria, epilepsy can be diagnosed after a single seizure and with a recurrence risk >60% within the next ten years (Fisher et al., 2014). As shown by the Rochester study, patients presenting with a single late seizure after a stroke carry such a risk. Therefore, one late unprovoked post-stroke seizure can be diagnosed as PSE.
However, it is important to recognize the pitfalls of the seven-day cut-off in order to distinguish between early and late seizures (see below under: “When and how to clinically diagnose post-stroke epilepsy”). The risk of PSE is substantially higher in patients who have presented with an early post-stroke seizure than in patients who have not had any seizure. The occurrence of early seizures is, indeed, an independent risk factor for PSE and weighs heavily in the SeLECT score. The SeLECT score is a recently developed and validated clinical tool to predict late seizures/epilepsy after ischaemic stroke. In addition to the occurrence of an early seizure, it takes into account the severity of stroke, aetiology of stroke, and cortical and arterial territory involved (Ferlazzo et al., 2016; Galovic et al., 2018). An early seizure, not considered epilepsy, should therefore not convey the message that the patient is at no risk of PSE. Clinical risk models and biomarkers must be incorporated in the future to help identify the mechanisms of PSE and refine the diagnosis of PSE in some patients with early seizures and reassure those at low risk of recurrence.
Much of the pathophysiology underlying seizures after stroke remains elusive. Experimental stroke studies have mapped a number of reactions following brain injury, which are common to other models of acquired epilepsy and include inflammatory response, changes in the expression of proteins involved in neuronal signalling, and remodelling of cytoskeleton, but causal links have not been clearly established. Increased blood-brain barrier permeability could also play a pathogenic role (Pitkänen et al., 2016).
Early post-stroke seizures should be regarded as a reaction of the neuronal cells to the acute cerebrovascular injury. They reflect transient cellular biochemical dysfunctions, including -among others- the increased release of excitatory neurotransmitter glutamate, ionic imbalance, breakdown of membrane phospholipids, and release of free fatty acids with oxidative stress (Tanaka and Ihara, 2017). Homeostatic or systemic disturbances, such as electrolyte imbalance, acid-base disturbances and hyperglycaemia, may also play a role in the development of early post-stroke seizures (Tanaka and Ihara, 2017). Conversely, late post-stroke seizures reflect a structural change of neuronal networks following the cerebrovascular injury to the brain (Trinka and Brigo, 2014). They are usually attributed to epileptogenic gliotic scarring with changes in membrane properties, neuronal deafferentation, selective neuronal loss or collateral sprouting (Tanaka and Ihara, 2017). Late post-stroke seizures after a primary cerebral haemorrhage or secondary haemorrhagic transformation of an ischemic stroke are thought to be the consequence of haemosiderin deposits leading to increased neuronal excitability. During post-stroke epileptogenesis, the brain undergoes molecular and cellular alterations, which increase its excitability and eventually lead to the occurrence of recurrent spontaneous seizures. These progressive neuronal changes include selective neuronal cell death and apoptosis, changes in membrane properties, mitochondrial and receptor changes (e.g. loss of GABAergic receptors), deafferentation, and collateral sprouting (Pitkänen et al., 2016). Disruption of the brain-blood barrier following endothelial damage causes extravasation of albumin which in turns activates astrocytes and microglial cells; this leads to changes in the extracellular milieu with increased glutamate levels, release of inflammatory cytokines, and further increase in brain-blood barrier permeability (Tanaka and Ihara, 2017). Thrombin, a major component of the coagulation cascade, and its protease-activated receptor 1 (PAR1), may further contribute to maladaptive plasticity leading to permanent structural changes in the brain with altered neuronal firing and circuit dysfunctions (Altman et al., 2019). This complex cascade of events directly enhances neuronal excitability and could explain epileptogenesis after a stroke. Alterations in gene expression after a stroke can also play a role in epileptogenesis, as they can be associated with impaired neuroprotection, aberrant synaptic plasticity, upregulation of neuronal excitability, and enhanced gliotic scarring formation (Pitkänen et al., 2016). Of note, these pathophysiological mechanisms interact with each other and eventually lead to structural and functional alterations of neuronal networks, leading to recurrent spontaneous seizures (Tanaka and Ihara, 2017).
Remarkably, the current pathophysiological perspective of acquired epilepsy favours a threshold model, which also involves individual predisposition. For instance, individuals with a first-degree relative suffering with epilepsy are at higher risk of developing PSE (hazard ratio: 1.18; 95% CI: 1.09-1.28) although this was associated with a small effect size (Eriksson et al., 2019). Lesion characteristics may be more important in most cases such as size of the lesion, cortical involvement and presence of intralesional blood products (see below under “Neuroimaging of post-stroke seizures: pitfalls and differential diagnosis”).
Until now, the early treatment of stroke patients with AEDs during the acute phase has not been effective in reducing the risk of developing PSE (Gilad et al., 2011; Sheth et al., 2015). On the other hand, statins appear to be the only medication to decrease the risk of PSE (Etminan et al., 2010), and to a greater extent in patients who present with early seizures and are considered a high-risk group (Guo et al., 2015). However, causality and mechanisms of the effect of statins are not yet well-established.
The rates of early post-stroke seizures and PSE vary across stroke populations. For ischaemic stroke, the prevalence of early seizures is generally 3-6% (Beghi et al., 2011; Labovitz et al., 2001; Guo et al., 2015; Serafini et al., 2015) but can be up to 15% in selected cohorts (Labovitz et al., 2001; Lamy et al., 2003; Bentes et al., 2017). There is no converging evidence about the risk of early seizures in patients treated with reperfusion therapies, either intravenous thrombolysis or endovascular thrombectomy (Belcastro et al., 2020; Brigo et al., 2020a; Feher et al., 2019). The risk of intracerebral haemorrhage is somewhat higher (Qian et al., 2014), with early seizures occurring in approximately 10-16% of patients (Naess et al., 2004; Beghi et al., 2011; Procaccianti et al., 2012). However, the methodology adopted to ascertain and diagnose early post-stroke seizures can greatly affect the results. For instance, a study using video-EEG recording performed in the first 72 hours following an acute anterior circulation ischaemic stroke revealed early seizures in 14.6% and non-convulsive status epilepticus (SE) in 2.6% of patients; of note, almost a quarter (22.7%) of early seizures were exclusively electrographic (Bentes et al., 2017).
Data on PSE prevalence also depend on the study population and methodology used to collect data. Based on nationwide registers in Sweden, the cumulative incidence of PSE was 6.4% following ischaemic stroke and 12.4% following haemorrhagic stroke after a follow-up of almost five years (Zelano et al., 2016); the latter finding has been replicated in a population-based investigation in a Finnish region (Lahti et al., 2017). In a video-EEG study, 15.2% of patients suffering with an anterior ischemic stroke met the diagnostic criteria for epilepsy at 12 months (Bentes et al., 2017).
A diagnosis of PSE (after ischaemic and haemorrhagic stroke) increases the risk of mortality after adjusting for stroke severity (Zelano et al., 2016) and, unsurprisingly, vascular disease is the major cause of death. These findings call for concerted efforts to prioritise and optimize secondary vascular prophylaxis (Hansen et al., 2017), and AEDs that do not interfere with concomitant medications, such as anti-hypertensives and anticoagulants, should be preferentially chosen.
The main risk factors for early post-ischaemic stroke seizures are cortical involvement, severe stroke, haemorrhagic transformation, age younger than 65 years, a large lesion and atrial fibrillation (Feher et al., 2019). The main risk factors for PSE following ischaemic stroke are cortical involvement, haemorrhage, and early seizures (Ferlazzo et al., 2016).
The concept of early and late seizures and PSE is straightforward to apply in clinical practice in most cases. If a patient has a seizure within a week of stroke, it is an early seizure and considered acute symptomatic. Although such a seizure carries a risk of subsequent epilepsy, this risk does not warrant the diagnosis of PSE. In contrast, a seizure occurring more than one week after stroke is considered an unprovoked late seizure. This infers a >60% risk of seizure recurrence and the patient meets the diagnostic criteria for epilepsy.
In some circumstances, the distinction between early and late seizures may not be unequivocal. The clinical situation may have been unstable, and the exact time of the latest cerebral insult may not be clear. As per the definition of epilepsy recommended by the International League Against Epilepsy, the diagnosis requires a risk of seizure recurrence exceeding 60%, however, the exact risk in each case is hard to estimate with precision. If there is doubt whether a seizure has occurred within the acute symptomatic phase, then there is no clear evidence of a >60% recurrence risk. In this scenario, the diagnosis of PSE should not be made. A similar approach can be suggested if there is doubt whether a paroxysmal post-stroke event is actually a seizure. In the presence of uncertainty, it is probably better not to diagnose a late seizure/PSE, but rather adopt a wait-and-watch approach. It is important to emphasize, however, that whether a patient is or is not diagnosed with PSE, the decision to initiate treatment with AEDs will depend on clinical characteristics of individual patients.
In the early phase following an ischaemic or haemorrhagic stroke, electroencephalogram (EEG) is an essential diagnostic tool that aims to detect purely electrographic seizures. It can also detect specific patterns, such as lateralized periodic discharges (LPDs), that are independently associated with early seizures (Mecarelli et al., 2011).
Interestingly, brain single-photon emission computed tomography (SPECT) imaging can reveal focal hypermetabolism with increased cerebral blood flow in association with LPDs in patients with post-stroke seizures; such findings support the view that – at least in some patients – this EEG pattern may correspond to an ictal phenomenon (Ergün et al., 2006; Hughes, 2010).
The lack of a systematic electrophysiological assessment with video-EEG can lead to an underestimation of seizures, particularly in the case of focal unaware or non-convulsive seizures (Belcastro et al., 2014; Bentes et al., 2017; Brigo et al., 2020a, 2020b). Neurologists and health personnel working in stroke units should promptly request an EEG recording for patients with sudden onset of unexplained behavioural changes or impairment of consciousness. A continuous EEG lasting ≥24 hours should be recorded as soon as possible in patients with acute supratentorial brain injury presenting with altered mental status or with clinical paroxysmal events suspected to be seizures. In addition, in comatose patients, patients with periodic discharges, or patients who are pharmacologically sedated, a more prolonged EEG (≥48 hours) may lead to the detection of non-convulsive seizures (Herman et al., 2015). The main indications for continuous EEG in patients with acute stroke, to identify non-convulsive seizures and non-convulsive status epilepticus, are presented in table 1.
EEG recordings may also have implications in the prediction of functional outcome, mortality and post-stroke cognitive decline, with different levels of evidence (Doerrfuss et al., 2020).
Only few studies have, so far, assessed EEG as a predictive tool for post-stroke seizures and epilepsy. Abnormalities on EEG can predict the development of epilepsy in the first year after stroke, independently of clinical and imaging-based infarct severity. A retrospective study of 110 patients with ischaemic stroke-related seizures found LPDs in 5.8% of patients, whereas the 275 stroke patients who did not suffer an early and/or a late seizure did not present with LPDs (De Reuck et al., 2006). Diffuse EEG slowing and frontal intermittent rhythmic delta activities also occurred more frequently among patients with post-stroke seizures compared to controls (21.7% versus 5.1% and 24.6% versus 1.1%, respectively) (De Reuck et al., 2006). A prospective video-EEG study enrolled 151 patients with anterior circulation ischaemic stroke and no previous seizures. Asymmetric background activity and interictal epileptiform activity detected on EEG performed during the first 72 hours after stroke were independent predictors of PSE during the first year following the index event (Bentes et al., 2018).
These findings suggest how EEG recorded in the acute stroke phase may not only detect subclinical seizures, but also may provide useful information to predict the development of PSE. Further studies are warranted to assess whether the inclusion of EEG findings with existing scores (e.g. SeLECT [Galovic et al., 2018]) could improve their predictive accuracy (Doerrfuss et al., 2020).
Most studies available in the literature refer to the use of EEG in the acute phase as a predictor of unprovoked late post-stroke seizures. In patients with PSE, EEG usually shows multifocal or focal slowing, typically with a normal background alpha rhythm (Mecarelli and Vicenzini, 2019). Epileptiform abnormalities can be detected, usually as sharp waves, sometimes with a quasiperiodic pattern of recurrence, particularly in PSE associated with large cortical cerebrovascular lesions (Brigo and Mecarelli, 2019; Mecarelli and Vicenzini, 2019).
Seizures are an expression of sudden depolarization of neurons that transiently disrupts ionic and metabolic homeostasis. There are different proposed pathophysiological mechanisms for early and late seizures, which include critically reduced local blood flow, abnormal release of neurotransmitters, metabolic dysfunction, presence of gliotic scarring and aberrant synaptic connectivity (Pitkänen et al., 2016). Of note, only a minority of stroke patients will develop seizures and there is still scarce understanding of magnetic resonance imaging (MRI) signatures that can identify the patients at higher risk. Some studies have identified the following MRI predictors: watershed infarctions, middle cerebral arterial territory strokes, cortical involvement, haemorrhagic strokes and haemorrhagic transformation of ischaemic stroke (Ferlazzo et al., 2016; Galovic et al., 2018).
Caution is, however, necessary in considering a post-stroke seizure as stroke related. In such cases, neuroimaging is fundamental in providing a differential diagnosis.
In acute settings, cranial computed tomography (CCT) is the gold standard to rapidly image patients presenting with seizures. It also represents the only available neuroimaging tool in patients who cannot undergo MRI. Besides standard CCT, perfusion CT (PCT) can be helpful in differentiating between stroke, stroke mimics and status epilepticus (Strambo et al., 2018). Ongoing seizure activity or SE are characterized by regions of hyperperfusion that usually involve atypical vascular territories, whereas strokes typically correspond to hypoperfused areas in a precise arterial territory (Payabvash et al., 2015). PCT can also be helpful to differentiate postictal versus stroke related focal neurological deficits: the former are characterized by transient iso- to hyperperfusion and the latter by areas of hypoperfusion in a vascular territory (Brigo and Lattanzi, 2020). Notably, PCT must be performed within a strict interval from seizure onset (< three hours) to improve its sensitivity (Payabvash et al., 2015).
MRI remains the most sensitive non-invasive diagnostic tool to image the brain, and conventional MR sequences suffice in most cases of suspected post-stroke seizures. The most common conventional sequences are listed in table 2. Diffusion-weighted imaging (DWI) is an informative sequence and is now routinely used in clinical settings. It is fast (acquisition requires less than a minute) and demonstrates high sensitivity for areas of water restriction, making it a very commonly used sequence to differentiate stroke from stroke mimics. Nonetheless, timing of acquisition is extremely important to avoid pitfalls. Lesions can be falsely positive (those containing a very high water content, also known as “T2 shine-through” artefact) or falsely negatively (MRI scans acquired too late after symptom onset) (Agarwal et al., 2017; Shono et al., 2017). Restricted signal on DWI due to seizure activity is mostly reversible (figure 1), whereas it may last three to four days in stroke, unless cerebral tissue has been reperfused earlier. Other lesions that may present with DWI positive signal include subacute haematomas, hypercellular tumours and abscesses (figure 2). Recently, Koksel et al. proposed the acronym “CRUMPLED” as a helpful way to remember the most important DWI-restricted lesions that are cortically based and have an atypical vascular distribution. The acronym stands for C = Creutzfeldt-Jakob disease; R = reversible cerebral vasoconstriction syndrome; U = urea cycle disorders and uraemia; M = mitochondrial disorders; P = prolonged seizures and posterior reversible encephalopathy syndrome (PRES); L = laminar necrosis (hypoxic-ischaemic encephalopathy) and liver-related (acute hepatic or hyperammonaemic encephalopathy); E=encephalitis (infectious meningoencephalitis) and D = diabetes mellitus (hypoglycaemia) (Koksel et al., 2018) (figure 2).
Intravenous gadolinium-based imaging can be particularly useful to: a) identify areas of disrupted blood-brain barrier; b) provide evidence of reperfusion or presence of collateral flow; and c) identify stroke mimics.
Advanced imaging can provide additional and more accurate information for the differential diagnosis.
Perfusion weighted imaging (PWI) can reliably identify tissue at risk of infarct, defined as an area with a blood flow of less than 50 mL per 100 mL of brain tissue per minute (Jahng et al., 2014). Signal changes in PWI are related to electrographic ictal activity: hyperperfusion is likely to be seen in pre-ictal and ictal phases, whereas hypoperfusion is more common in the post-ictal phase (Takahara et al., 2018). Early seizures are more likely to present as areas of hyperperfusion due to the underlying pathophysiological mechanisms, including metabolic dysfunction and abnormal release of neurotransmitters, whereas late seizures are likely to show a mixed pattern of perfusion as they are more related to gliotic changes and loss of neuronal tissue (Yoo et al., 2017). Hyperperfusion may also precede DWI signal changes as a compensatory mechanism to support the abnormally increased depolarization of neurons (Takahara et al., 2018). Susceptibility weighted imaging (SWI) and gradient-echo (GRE) sequences can be highly informative and detect punctuate microbleeds and areas of iron-laden products in patients with subarachnoid haemorrhage, chronic subdural haematoma, cerebral amyloid angiopathy or superficial siderosis. Extracellular haemosiderin is considered epileptogenic and may cause focal cerebral irritation and initiate seizures, even though the mechanisms are not yet well-established (O’Connor et al., 2014).
Due to the rather low risk of early, acute-symptomatic post-stroke seizures, ranging from 3-6% in cases of cerebral ischaemia to 16% in primary cerebral haemorrhage (Labovitz et al., 2001; Naess et al., 2004; Beghi et al., 2011; Procaccianti et al., 2012; Guo et al., 2015; Serafini et al., 2015), primary prophylaxis with an AED is not recommended. This is also true for those patients who have cerebral haemorrhage involving cortical structures and a risk of early post-stroke seizures of around 35%. If physicians decide to introduce primary AED prophylaxis despite the evidence-based recommendations, an AED that can be titrated very quickly, administered intravenously, and which lacks significant drug-drug interactions should be preferred. One of the most commonly prescribed AEDs that meets these characteristics is levetiracetam (LEV). One randomized controlled trial compared valproate to placebo in 36 patients, both of which were administered directly after intracerebral haemorrhage (Gilad et al., 2011). The groups did not differ with respect to prevention of early post-stroke seizures (defined in that study as occurring within the first 14 days), but the trial was underpowered, and prevention of early seizures was not the primary endpoint.
After the occurrence of one early post-stroke seizure, the risk of developing a second acute symptomatic seizure within the acute phase is only 10-20% (De Herdt et al., 2011; Leung et al., 2017). Due to the low risk of recurrence, guidelines generally do not recommend secondary AED prophylaxis after an early post-stroke seizure (Holtkamp et al., 2017). However, many clinicians prefer to administer an AED to reduce the likelihood of clinical worsening in the acute setting. Conceptually, this approach likely relies on pathophysiological considerations including increased neuronal excitotoxicity, peri-infarct depolarisations, and inflammatory responses in the first hours and days after stroke (Dirnagl et al., 1999), all of which can be risk factors for acute recurrence of epileptic seizures. The criteria used to choose the AED for acute secondary prophylaxis are similar to those for primary prophylaxis.
If patients without or after an early post-stroke seizure have been administered an AED, physicians are encouraged to withdraw it after the acute phase – at best, at discharge from the stroke unit – as the vast majority of these patients will not experience any future seizures (Holtkamp et al., 2017). The risk of a first unprovoked post-stroke seizure within eight years (which would define epilepsy) after cerebral infarct is 8% and 15% after cerebral haemorrhage (Merkler et al., 2018), and the risk of an unprovoked seizure after one early post-stroke seizure with 10 years is 30 to 35% (Hesdorffer et al., 2009; Galovic et al., 2018). Two studies developed scores to estimate the long-term risk of unprovoked seizures after acute cerebrovascular events. The CAVE score indicates a five-year seizure risk of 46.2% in patients after intracerebral haemorrhage based on the following four variables: early post-stroke seizure(s), cortical involvement, bleeding volume of more than 10 mL, and age of less than 65 years (Haapaniemi et al., 2014). The SeLECT score indicates a five-year seizure risk of more than 50% in patients after ischaemic stroke based on the following four or five criteria: early post-stroke seizure(s), severe stroke (NIHSS ≥11), cortical involvement, and large-artery atherosclerosis and/or involvement of the middle cerebral artery territory (Galovic et al., 2018). In these individual risk constellations, long-term secondary AED prophylaxis may be indicated.
AED treatment is advised based on guidelines when PSE is diagnosed (Holtkamp et al., 2017). As always, there may be individual reasons not to start treatment -for instance, in cases with very mild semiology. Regarding the selection of drugs, two underpowered randomized, open-label studies compared controlled-release carbamazepine (CBZ-CR) to lamotrigine (LTG) (Gilad et al., 2007) and LEV (Consoli et al., 2012). The 12-month seizure freedom rates were 44% and 85% for CBZ-CR and 72% and 94% for LTG and LEV, without significant differences. LTG and LEV were better tolerated than CBZ-CR. A network meta-analysis of these trials showed no difference between LEV and LTG for seizure freedom (OR: 0.86; 95% CI: 0.15-4.89), but demonstrated greater occurrence of adverse events for LEV than LTG (OR: 6.87; 95% CI: 1.15-41.1) (Brigo et al., 2018). A randomized double-blinded trial on AEDs in epileptic patients, aged 60 years and older (two thirds had cerebrovascular aetiology), demonstrated higher one-year retention rates for LEV (62%) compared to CBZ-CR (46%; p=0.02), while LTG (56%) was intermediate (Werhahn et al., 2015). The SANAD trial, a non-blinded randomized controlled study comparing five standard and new AEDs in focal epilepsy, found LTG to have the best retention rate as compared to carbamazepine (CBZ), gabapentin, oxcarbazepine, and topiramate (Marson et al., 2007). Although data were not stratified according to the underlying aetiology, the findings can likely be extrapolated to PSE. The non-blinded, randomized, 52-week KOMET study compared the effectiveness of LEV as monotherapy to extended-release sodium valproate (VPA-ER) or CBZ-CR after the physician had decided which of the two AEDs best suited the individual patient (Trinka et al., 2013). In a post-hoc subgroup analysis of patients aged ≥60 years with newly diagnosed epilepsy (most of which were likely to have cerebrovascular aetiology), the 12-month retention rates in the VPA-ER stratum were 90% in the LEV group and 77% in the VPA-ER group; the corresponding rates in the CBZ-CR stratum were 75% and 53% in the LEV and CBZ-CR treatment arms, respectively (Pohlmann-Eden et al., 2016). In summary, the findings from clinical studies argue in favour of the newer AEDs for PSE due to their better tolerability profiles.
In focal epilepsy, the underlying aetiology does not usually determine the choice of AED. The decision regarding the most suitable compound has to be individualized according to the patient’s age, sex, comorbidities and comedications. Patients with PSE likely carry some burden of cardiovascular risk factors. Accordingly, AEDs such as CBZ, phenytoin, phenobarbital and primidone, which can increase biochemical markers of vascular disease, including total cholesterol, lipoprotein, C-reactive protein and homocysteine (Mintzer et al., 2009), should be avoided. Being strong enzyme-inducers, these AEDs may also increase the metabolism, and thus decrease serum concentrations, of drugs that are concomitantly administered for stroke management, such as warfarin. Post-stroke depression is common, and the detrimental effects of LEV on behaviour (Josephson et al., 2019) may further fuel psychiatric comorbidity, rendering this AED less appropriate in patients with post-stroke depression.
The question to withdrawal the antiepileptic treatment at some time point after the onset of PSE is difficult to address. The overall risk of seizure recurrence within five years after AED tapering is roughly 50%. A meta-analysis on seizure recurrence rate after AED withdrawal, based on 10 retrospective, prospective and randomized-controlled trials involving more than 1,700 patients, allowed the development of a prediction tool for seizure relapse (Lamberink et al., 2017). This tool can be accessed online (Epilepsy Prediction and Tool., 2019) and can assist physicians, but the decision to withdraw the treatment needs to be tailored to each patient individually.
Several issues of PSE remain open to further research and investigation. Studies are warranted to elucidate the mechanisms of epileptogenesis after stroke and identify reliable biomarkers associated with the development of PSE. The role of EEG in predicting the occurrence of post-stroke seizures and epilepsy requires additional evaluation. The duration of EEG recording should be further evaluated in order to establish whether prolonged video-EEG monitoring during the first 72 hours after stroke is cost-effective and can offer advantages over routine, short-lasting EEG to identify post-stroke seizures (Grillo, 2015). The association of systemic thrombolysis and mechanical revascularization procedures with the development of early and late post-stroke seizures is still a matter of debate (Bentes et al., 2020). Similarly, there remain uncertainties about the most efficacious and safe AED to manage PSE.
Long-term, prospective, multicentric, high-quality studies with large cohorts of patients and stroke registries are needed to elaborate a practice guideline on diagnosis and treatment of PSE.
Summary didactic slides are available on the www.epilepticdisorders.com website.
Dr. Brigo received travel support from Eisai; acted as consultant for Eisai, LivaNova, and UCB Pharma; and was one of the organizers of the “Seizures & Stroke” Congress, held in Gothenburg from 20th to 22nd February 2019.
Dr Zelano has received consultancy fees from the Swedish Medial Product agency; speaker honoraria from UCB, was one of the organizers of the “Seizures & Stroke” Congress, held in Gothenburg from 20th to 22nd February 2019; and as an employee of Sahlgrenska university hospital (no personal compensation) is, and has been, an investigator in clinical trials sponsored by GW Pharma, SK life science, UCB, and Bial.
Dr. Holtkamp received speaker’s honoraria and/or consultancy fees from Bial, Desitin, Eisai, GW Pharmaceuticals, LivaNova, Novartis, and UCB (within the last three years).
Dr. Trinka received speaker honoraria from Eisai, UCB Pharma, LivaNova, Sandoz, Novartis, Biogen, Everpharma, BIAL-Portela &C, Newbridge, GL Pharma, Boehringer; grants from Biogen, UCB Pharma, Bayer, Novarti, Eisai, Merck, and Red Bull; grants from the European Union, FWF Österreichischer Fond zur Wissenschaftsforderung, Bundesministerium für Wissenschaft und Forschung, and Jubiläumsfond der Österreichischen Nationalbank outside the submitted work; and is a member of the following ILAE Task forces: Medical Therapies, Nosology, Terminology, Congresses, Driving, Regulatory affairs, and Telemedicine.
Dr. Agarwal and Dr. Lattanzi have no conflicts of interest to disclose.
Posted by Kostas Pantremenos in Epilepsy on June 12, 2020
Many seizure-free patients who consider withdrawing from antiepileptic drugs (AEDs) hope to discontinue treatment to avoid adverse effects. However, withdrawal has certain risks that are difficult to predict. In this study, we performed a literature review, summarized the causes of significant variability in the risk of postwithdrawal recurrent seizures, and reviewed study data on the age at onset, cause, types of seizures, epilepsy syndrome, magnetic resonance imaging (MRI) abnormalities, epilepsy surgery, and withdrawal outcomes of patients with epilepsy. Many factors are associated with recurrent seizures after AED withdrawal. For patients who are seizure-free after treatment, the role of an electroencephalogram (EEG) alone in ensuring safe withdrawal is limited. A series of prediction models for the postwithdrawal recurrence risk have incorporated various potentially important factors in a comprehensive analysis. We focused on the populations of studies investigating five risk prediction models and analyzed the predictive variables and recommended applications of each model, aiming to provide a reference for personalized withdrawal for patients with epilepsy in clinical practice.
Posted by Kostas Pantremenos in Epilepsy on January 30, 2018
Imagine going about your daily life, working, shopping, and driving, knowing that you might have a seizure at any moment. But relief is on the horizon, as researchers from the University of Melbourne in Victoria, Australia have developed a potentially life-saving deep learning tool that can predict when an epileptic seizure is about to happen.
Their study was published in the journal eBioMedicine last month. The deep learning-based prediction system “achieved mean sensitivity of 69% and mean time warning of 27%, significantly surpassing an equivalent random predictor for all patients by 42%,” according to the findings.
Dean Freestone, PhD, senior research fellow at the department of medicine at St. Vincent’s Hospital at the University of Melbourne, says the tech could be contained in a chip inside a wearable device such as a wristband or bracelet, “incorporating a person’s behavior, environment, and physiology.” He and fellow co-author Mark J. Cook, MD, chair of medicine at St. Vincent’s Hospital, have launched a company named Seer Medical to pursue this technology. They hope to implant patients with the technology later this year.
“The technology is now proven. We have shown seizure prediction is possible in our previous paper published in Brain and in a Kaggle contest. This new study is just further backup,” Freestone says.
The advance could change the lives of many people with epilepsy, who worry about looming seizures while they are doing everyday activities. Patients who have tested the technology reported that they felt more in control when they used the wearable device and were more confident doing novel activities. They also claimed to have benefited from improved sleep and decision making.
The new forecasting technology would be best suited for someone having seizures once per week, according to the architects. If someone has seizures every hour, or if the seizures are too infrequent, it is difficult to train the algorithms, Freestone notes.
The way the predictive technology works is similar to Facebook’s facial recognition software. Instead of people in photos, the researchers have trained the algorithms to recognize patterns in the electrical activity of the brain that preempt seizures. “It is software that learns from example. The electrical patterns are very subtle and are invisible to the human eye, but the computer algorithms can identify them. The circadian patterns then help to boost the algorithms performance,” Freestone says.
“Patients can take action to actually prevent seizures. This could be in the form of a medication or even just a change in behavioral. We have also learnt a lot about the mechanisms of seizure, such as the strong influence of circadian cycles,” he adds.
Although significant cost and risk comes with new trials of medical devices, researchers are excited about the changes they can make. “We are working toward a system that will constantly provide a person with a risk level of seizure susceptibility,” Freestone says. “It will be a gauge that outputs a probability. We will incorporate as many aspects of a person’s behavior, environment and physiology as we can acquire from wearable technologies and other sensors.”
The findings came about, in part, thanks to the University of Melbourne’s large, long-term data set, which is unique and apt for exploring deep learning for seizure forecasting.
via Deep Learning Device Can Predict Epileptic Seizures | Healthcare Analytics News
Posted by Kostas Pantremenos in Paretic Hand on August 12, 2017
•Three characteristic factors differentiate fine motor control in patients and controls.
•The three factors are grip force scaling, motor coordination and speed of movement.
•These factors are predictors of 69% of variance for the Jebsen Taylor Hand Function Test.
To determine factors characterizing the differences in fine motor performance between stroke patients and controls. To confirm the relevance of the factors by analyzing their predictive power with regard to the Jebsen Taylor Hand Function Test (JTHFT), a common clinical test of fine motor control.
Twenty-two people with slight paresis in an early chronic phase following stroke and twenty-two healthy controls were examined.. Performance on the JTHFT, Nine-Hole Peg Test and 2-point discrimination was evaluated. To analyze object manipulation skills, grip forces and temporal measures were examined during 1) lifting actions with variations of weight and surface 2) cyclic movements 3) predictive/reactive catching tasks. Three other aspects of force control included 4) visuomotor tracking 5) fast force changes and 6) grip strength.
Based on 9 parameters which significantly distinguished fine motor performance in the two groups, we identified three principal components (factors): grip force scaling, motor coordination and speed of movement. The three factors are shown to predict JTHFT scores via linear regression (R2=.687, p<.001).
We revealed a factor structure behind fine motor impairments following stroke and showed that it explains JTHFT results to a large extend.
Significance: This result can serve as a basis for improving diagnostics and enabling more targeted therapy.
TBI Rehabilitation 