University of Ghana http://ugspace.ug.edu.gh TITLE: UNIVERSITY OF GHANA SCHOOL OF BIOMEDICAL AND ALLIED HEALTH SCIENCES ELECTROENCEPHALOGRAPHY IN SEIZURE DIAGNOSIS AND THE PREDICTION OF FUNCTIONAL OUTCOMES OF STROKE PATIENTS AT KORLE BU TEACHING HOSPITAL BY RUTH YEMORKOR LARYEA (10202726) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL PHYSIOLOGY DEGREE DEPARTMENT OF PHYSIOLOGY JULY, 2018 i University of Ghana http://ugspace.ug.edu.gh DECLARATION I Ruth Yemorkor Laryea, do hereby declare that apart from literature cited and acknowledged, this thesis is my own work produced from research under the supervision of Rev. Dr Charles Antwi Boasiako of the Department of Physiology and Dr Albert Akpalu of the Department of Medicine and Therapeutics, College of Health Sciences, University of Ghana. Signature…………………….. Date…………………….. Ruth Yemorkor Laryea (Student) Signature…………………….. Date…………………….. Rev Dr Charles Antwi-Boasiako (Principal supervisor) Signature…………………….. Date…………………….. Dr Albert Akpalu (Co-supervisor) ii University of Ghana http://ugspace.ug.edu.gh DEDICATION I dedicate my thesis to God almighty- my sun and shield, to my family and to my mentors. iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT I am very grateful for the unwavering support of my supervisors (Rev. Dr Antwi-Boasiako and Dr Albert Akpalu) who continually encouraged and guided me with patience. Thank you Rev. for teaching me how to write a good thesis and Dr Akpalu for guiding me on what is relevant and spending time to read the EEGs for me. I appreciate Mr Philip Amoako, who dedicated his EEG machine, expertise and his time to help me acquire the data I needed for this work, Mr Ebenezer Laryea, for the guidance in code writing in MATLAB, Mr Jude Kankam for research support and the entire stroke unit for their encouragement and support in making this study possible. My gratitude also goes to Mr Stephen Laryea Odoi for the sponsorship and to all colleagues and seniors who made the path to success smooth. iv University of Ghana http://ugspace.ug.edu.gh CONTENTS TITLE .............................................................................................................................................. i DECLARATION ............................................................................................................................ ii DEDICATION ............................................................................................................................... iii ACKNOWLEDGEMENT ............................................................................................................. iv CONTENTS .................................................................................................................................... v LIST OF FIGURES ..................................................................................................................... viii LIST OF TABLES ......................................................................................................................... ix LIST OF ABBREVIATIONS ......................................................................................................... x ABSTRACT .................................................................................................................................. xii 1.0 BACKGROUND ...................................................................................................................... 1 1.1 Introduction ........................................................................................................................... 1 1.2 Problem statement: ................................................................................................................ 3 1.3 Justification/Relevance: ........................................................................................................ 3 1.4 Study Hypothesis: ................................................................................................................. 4 1.5 Aim ........................................................................................................................................ 5 1.6 Specific objectives................................................................................................................. 5 2.0 LITERATURE REVIEW ......................................................................................................... 6 2.1. The homeostatic brain .......................................................................................................... 6 2.2. Pathophysiology of stroke .................................................................................................... 7 2.3. Synaptic plasticity after stroke ............................................................................................. 8 2.4. Seizures in stroke ................................................................................................................. 9 2.4.1. Early seizures ............................................................................................................. 10 2.4.2. Late seizures ............................................................................................................... 10 2.4.3. Subclinical seizures- EEG as a diagnostic tool ........................................................ 11 2.5. Relationship between EEG and action potential ....................................................... 11 2.6. Measures of functional outcome in stroke patients ............................................................ 12 2.7. Quantitative EEG (QEEG)- another measure of functional outcome ..................... 13 3.0 METHODOLOGY ................................................................................................................. 15 3.1 Study Design: ...................................................................................................................... 15 3.2 Study Site: ........................................................................................................................... 15 v University of Ghana http://ugspace.ug.edu.gh 3.3 Study population: ................................................................................................................ 16 3.4 Inclusion Criteria ................................................................................................................. 16 3.5 Exclusion Criteria ................................................................................................................ 16 3.6 Sample size:......................................................................................................................... 17 3.7.0 Procedures ........................................................................................................................ 17 3.8 Data handling ...................................................................................................................... 22 3.9 Statistical analysis ............................................................................................................... 23 3.10 Ethical issues ..................................................................................................................... 23 4.0 RESULTS ............................................................................................................................... 24 4.1 Patient demographics: Age and gender characteristics of 30 stroke patients ..................... 24 4.2. Clinical characteristics of patients...................................................................................... 25 4.3.0 Occurrence of seizures ..................................................................................................... 27 4.4.0: Visual characterization of EEG recordings ..................................................................... 30 4.5.0 Comparison of functional outcome at recruitment to one-month measures .................... 32 4.6. Correlation between recruitment measures of functional status and functional outcome values at one-month .................................................................................................................. 32 4.7.0 Prediction of functional outcome of patients at one-month post-stroke .......................... 35 5.0 DISCUSSION ......................................................................................................................... 38 5.1. Patient demographics and clinical characteristics .............................................................. 38 5.2. Seizure occurrence in stroke patients ................................................................................. 40 5.3. Predictors of post stroke functional outcome at one-month in 30 stroke patients ............. 42 6.0 CONCLUSION ....................................................................................................................... 44 STUDY LIMITATIONS .............................................................................................................. 44 RECOMMENDATIONS .............................................................................................................. 44 REFERENCES ............................................................................................................................. 46 APPENDIX ................................................................................................................................... 59 Appendix 1: Information and Consent form ............................................................................. 59 Appendix 2: Study Questionnaire ............................................................................................. 62 Appendix 3: Modified Rankin Scale (MRS) ............................................................................. 64 Appendix 4: Modified National Institute of Health Stroke Scale (NIHSS) .............................. 65 Appendix 5: Barthel Index ........................................................................................................ 73 vi University of Ghana http://ugspace.ug.edu.gh Appendix 6: Copy of Ethical clearance..................................................................................... 75 vii University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 1: Mechanisms of seizure generation after stroke (adapted from Tanaka & Ihara, 2017) 8 Figure 2: Neuroimaging findings of study participants 25 Figure 3: Electroencephalographic and clinical details on seizures 28 Figure 4: Categories of epileptiform activity seen on EEG 30 Figure 5: Distribution of dysfunctional brain activity on EEG 31 viii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1: Categorization of outcome measures 19 Table 2: EEG frequency bands 20 Table 3: Patients’ demographic characteristics 24 Table 4: Functional status of 30 stroke cases at recruitment and follow-up 26 Table 5: EEG findings in 30 stroke patients 27 Table 6: Early seizures, late seizures and the use of anti-epileptic drugs in stroke patients 29 Table 7: Mean functional status measures at recruitment and one-month post stroke 32 Table 8: Correlations between measures of functional status 34 Table 9: Regression analysis between functional outcomes and QEEG indices of all patients 35 Table 10: Regression analysis between outcome measures by stroke type 37 ix University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS ADL Activities of daily living AED Anti-epileptic drug BI Barthel Index BSI Brain symmetry index Ca2+ Calcium CEEG Continuous electroencephalography CNS Central nervous system CT Computed tomography DAR Delta/ Alpha ratio EEG Electroencephalography fMRI Functional magnetic resonance imaging GABA Gamma amino-butyric acid KBTH Korle Bu Teaching Hospital MDT Multidisciplinary team mNIHSS Modified National Institute of Health stroke scale MRS Modified Rankin Scale Na+ Sodium NIHSS National Institute of Health Stroke Scale pdBSI Pairwise derived brain symmetry index PNS Peripheral nervous system PSE Post stroke epilepsy QEEG Quantitative electroencephalography x University of Ghana http://ugspace.ug.edu.gh RAP Relative alpha power SME Surgical and Medical Emergency unit SPSS Statistical Product for Service Solution SSA Sub-Saharan Africa TIA Transient ischaemic attack xi University of Ghana http://ugspace.ug.edu.gh ABSTRACT Background: As compared to developed economies, stroke is a rising epidemic affecting mostly the younger workforce in Sub-Saharan Africa (SSA), exerting a severe toll on the physical, psychoemotional, cognitive and social lives of its victims with a 3-year mortality rate of 84%. The occurrence of seizures lead to poor prognosis and increased mortality in stroke patients. Early screening for seizures and effective prognostication of the functional outcome of stroke may improve the outcomes of patients. General aim: This study aimed to determine the incidence of seizures in 30 stroke patients and how quantitative electroencephalography (QEEG) indices prognosticate their one-month functional outcome. Methodology: Electroencephalography (EEG) is a non-invasive analysis of brain function available at Korle Bu Teaching Hospital. Stroke outcome was measured using the modified National Institute of Health Stroke Scale (mNIHSS), modified Rankin Scale (MRS) and Barthel Index (BI). In this study, routine EEG of thirty (30) consenting acute stroke patients was recorded using the 10/20 standardized format within ten days of stroke onset. The EEG patterns were characterized by a neurologist and then delta/alpha ratio (DAR), relative alpha power (RAP) and the pairwise derived brain symmetry index (pdBSI) were calculated using the EEGLAB software. On the day of recording EEG, the mNIHSS, MRS and Barthel Index scores of the patients were measured and again a month later. Clinical and EEG findings were displayed on bar charts and statistical tables while Spearman’s rank correlation and linear regression was used to determine predictors of one-month outcome using the Statistical Product for Service Solution (SPSS) version 20. xii University of Ghana http://ugspace.ug.edu.gh Results: The 30 participants of this study consisted 60% aged 60years or below, 66.7% male and 33.3% haemorrhagic stroke. Clinical seizures occurred in 13.3% of study participants, electrographic seizure occurred in 46.7% (with 92.9% being generalized seizures) and cerebral dysfunction was diagnosed in 56.7% of the participants. The MRS, mNIHSS and BI measured at recruitment significantly correlated to their measures at one-month (p < 0.01); the least being between MRS and mNIHSS scores at recruitment (rho = 0.753) and the highest was between MRS and BI scores at recruitment (rho = -0.932). The only correlation for QEEG indices was between RAP and DAR (rho = -0.988, p < 0.001), the highest in the entire analysis. Overall, mNIHSS score at recruitment was the best predictor of all three functional outcome measures at one-month- MRS1 (beta = 0.471, p = 0.001), mNIHSS1 (beta= 0.753, p = 0.001) and BI1 (beta= -0.556, p = 0.001). Conclusion: There was a high incidence of electrographic seizure in stroke patients of Korle Bu Teaching Hospital than what could be clinically diagnosed. The QEEG indices- DAR, RAP and pdBSI did not significantly predict post-stroke functional outcomes measures (MRS, mNIHSS or BI) at one-month across all stroke types. Relative alpha power (RAP) measured at recruitment was the only QEEG index useful in the prediction of neurological deficit (mNIHSS) in ischaemic stroke at one-month post stroke. The mNIHSS measured at recruitment was the most significant predictor of all functional outcome measures (MRS, mNIHSS and BI) assessed at one-month post stroke. xiii University of Ghana http://ugspace.ug.edu.gh 1.0 BACKGROUND 1.1 Introduction Stroke is the damage of brain cells which occurs when the blood supply to that part of the brain is interrupted. This interruption can be due to blockage of the blood vessels (ischaemic stroke) by local or migrating thrombus or rupture of the blood vessels (haemorrhagic stroke) at the site of an aneurysm or due to high blood pressure (Barrett, Brooks, Boitano, & Barman, 2010; Mehta & Vemuganti, 2014; Sherwood, 2010). Despite advances in medical practice, World Health Organization statistics reports that each year, a third of the over 15 million stroke victims die and another third live with permanent disabilities (Mehta & Vemuganti, 2014; Offner, Ihara, Schäbitz, & Wong, 2017), emphasizing the importance of all measures to prevent stroke or reduce its impact on lives. Common complications of stroke includes seizures (Nkusi et al., 2017; Silverman, Restrepo, & Mathews, 2002) among others. Seizures are transient abnormal synchronous discharges of electrical activity from brain cells which cause involuntary spasms and altered behaviour or sensation in the patient (Barrett et al., 2010). Epilepsy is the term used when a person has two or more seizures due to persisting neurological abnormality (Fisher et al., 2005). Seizures are either generalized or partial/focal and are common in stroke patients (Barrett et al., 2010; Silverman et al., 2002). Within the first week or two after stroke, the consequent metabolic dysfunction in the brain can induce seizures (early seizures) while persisting changes in neuronal excitability could also cause seizures beyond two weeks after stroke (late seizures) (Siddiqui et al., 2008; Silverman et al., 2002). The occurrence of seizures in stroke patients increase disability and mortality after stroke (Koubeissi, 2015); 1 University of Ghana http://ugspace.ug.edu.gh however, these seizures can occur without observable signs and symptoms (subclinical seizures) putting these patients in inconspicuous danger. Till date, the electroencephalogram (EEG) remains the most common and most affordable tool in the diagnosis of seizures, especially subclinical seizures (Feldwisch-Drentrup et al., 2011) be it routine or continuous EEG (Finnigan & van Putten, 2013; Koren et al., 2016). Piotr Olejniczak defined EEG as “a graphic representation of the difference in voltage between two different cerebral locations plotted over time” (Olejniczak, 2006). EEG can be recorded on the scalp, on the brain’s surface or inside the brain tissue (Barrett et al., 2010) and interpreted primarily by visual characterization or analysed computationally (a process known as QEEG) for diagnosis or cognitive assessment (Kaiser, 2007). Recent studies on EEG shows QEEG indices significantly predict the functional outcome of stroke patients at day 7 (Sheorajpanday, Nagels, Weeren, & De Deyn, 2011) and even 6 months after stroke (Sheorajpanday, Nagels, Weeren, van Putten, & De Deyn, 2011)- a tool that could be useful in informing clinical management (Finnigan & van Putten, 2013). According to the World Health organization, changes in the health status, well-being or quality of life of individuals as a result of a process/ intervention is the definition of health outcome (Mukuha, 2017). (Kwakkel et al., 2010) termed health outcomes which measure a person’s ability to perform tasks as functional outcomes. These outcomes can be measured as ability to perform activities of daily living after stroke (Barthel Index) (Kwakkel et al., 2010) and functional deficits using the National Institute of Health Stroke Scale (NIHSS) (Martin-Schild et al., 2011) and the Modified Rankin Scale(MRS) (Cincura et al., 2009). 2 University of Ghana http://ugspace.ug.edu.gh Despite the clinical relevance of seizures to stroke survival and recovery (Tanaka & Ihara, 2017), EEG is not routinely done on stroke patients in Ghana; a factor which might significantly impact functional recovery from stroke. Knowledge on the occurrence of seizures in stroke patients in Ghana could go a long way in deciding the use of antiepileptic drug (AED) prophylaxis (Sheth et al., 2015; Zandieh, Messé, Cucchiara, Mullen, & Kasner, 2016; Zelano, 2016), or increased vigilance in seizure identification or prevention (Sudalaimani et al., 2016). 1.2 Problem statement: It is firmly established that seizures in stroke patients lead to poorer chances of survival and higher morbidity (Koubeissi, 2015). Seizures are more common in haemorrhagic than ischaemic stroke (Nkusi et al., 2017; Tanaka & Ihara, 2017). With the higher incidence of haemorrhagic stroke in sub-Saharan Africans (Owolabi et al., 2017) including Ghanaians, the incidence of seizures are likely to be higher (Nkusi et al., 2017), leading to post stroke epilepsy (PSE) which will further decrease the quality of life of stroke survivors (Lahti et al., 2017) and increase mortality. How common clinical and subclinical seizures are in Ghanaian stroke patients is a question that needs answering. Early seizure diagnosis can direct clinical management during hospitalization to improve outcomes (Koubeissi, 2015), but routine seizure monitoring with EEG is not done in Ghana for stroke patients. Will the EEG be useful in diagnosing seizures in stroke patients and prognosticating their recovery at one-month? 1.3 Justification/Relevance: In the Korle Bu Teaching Hospital, stroke persistently ranks as the leading cause of medical mortalities. Considering the negative impact of seizures on survival, a study that could result in prevention and early management of seizures could improve stroke survival rates. Also, there is scarcity of literature on the occurrence and impact of seizures (both clinical and subclinical) in 3 University of Ghana http://ugspace.ug.edu.gh Ghanaian stroke patients, this study will contribute to knowledge and form the basis for further research. Although several studies have been done on the correlation of electroencephalography (EEG) indices with stroke outcomes , they were done in other populations such as in the Belgium (Sheorajpanday, Nagels, Weeren, van Putten, et al., 2011), China (Xin, Gao, Zhang, Cao, & Shi, 2012), Australia (Finnigan & van Putten, 2013) and the U.S.A (Wu et al., 2016), but not in the Ghanaian population. Furthermore, these studies mentioned above made use of the NIHSS scale to measure the stroke outcome. There is the need for such a study to be carried out in the Ghanaian population, using the more reliable modified NIHSS (mNIHSS) which could even be extracted from medical records to measure the stroke outcome (Meyer & Lyden, 2009).This is the scale this study intends to use. Stroke recovery relies significantly on neuroplasticity. In resource-poor countries like Ghana, QEEG can be a cost-effective tool for giving patients feedback on their neurological recovery during rehabilitation to motivate them to continue therapy (Kanna & Heng, 2009). Although it is not the gold-standard- functional magnetic resonance imaging (fMRI), it is non-invasive, easy to maintain, can be done at the patient’s bedside, does not require a high level of expertise to record and is affordable enough to be made available in many hospitals (Kanna & Heng, 2009; Piersson & Gorleku, 2017). To top it off, it measures brain activity in milliseconds, an ability fMRI is yet to achieve (Wu et al., 2016). 1.4 Study Hypothesis: Seizures in stroke patients on admission may be diagnosed with EEG and QEEG indices may be useful in prognosticating one-month functional outcome of stroke patients of the Korle Bu Teaching Hospital. 4 University of Ghana http://ugspace.ug.edu.gh 1.5 Aim The aim of this study was to determine the incidence of electrographic seizures in stroke patients and how QEEG indices prognosticate one-month functional outcome. 1.6 Specific objectives The specific objectives of this study were as follows:  Determine the occurrence of seizures (clinical and subclinical) in 30 stroke cases within one month of stroke onset using clinical observations and routine EEG recorded during admission  To compare the prognostic correlation of QEEG indices (RAP, DAR, pdBSI) measured at recruitment with post stroke functional outcomes in 30 stroke patients, measured at one- month using the MRS, mNIHSS and Barthel Index scores.  To compare the prognostic correlation of MRS, mNIHSS and Barthel Index scores measured at recruitment with post stroke functional outcomes in 30 stroke patients, measured at one-month using the MRS, mNIHSS and Barthel Index scores. 5 University of Ghana http://ugspace.ug.edu.gh 2.0 LITERATURE REVIEW Stroke leads to the premature death of victims and lost productivity for survivors and their caregivers; this results in a huge financial loss to individuals and states (Offner et al., 2017). With increasing population age projected to increase the financial burden stroke has on the economy (Hsieh, Wu, & Sung, 2017), all efforts must be made to improve stroke outcomes. This chapter reviews the normal function of brain cells, pathophysiology of stroke, stroke-related seizures, functional status of stroke patients and the use of electroencephalography in seizure diagnosis and functional status prediction at one-month post-stroke. 2.1. The homeostatic brain Control and coordination of activities of the body such as tasting, seeing, talking, walking digestion and respiration is the primary role of the nervous system. The brain and spinal cord form the central nervous system (CNS) while nerve fibres and sensory organs throughout the body make up the peripheral nervous system (PNS). At any particular time, many afferent nerves of the PNS send sensory input from all parts of the body to the CNS and the CNS sends responses back to the peripheries through the efferent nerves of the PNS (Sherwood, 2010). The ability to maintain homeostasis through these complex and extensive communication networks depends on the functioning and interactions of the neurons, glial cells and the vascular endothelial cells that make up the CNS. Neurons (e.g. pyramidal cells and retinal bipolar cells) are specially equipped to carry the rapid signals within the nervous system because of their ability to maintain a potential gradient while at rest (resting membrane potential); with a slightly negative intracellular and a slightly positive extracellular environment (Barrett et al., 2010). The release of chemical messengers (neurotransmitters) by the axon of a pre-synaptic neuron into the synaptic cleft causes a change in 6 University of Ghana http://ugspace.ug.edu.gh the membrane potential of the post-synaptic neuron (excitation or inhibition) with a resultant cascade of activities that result in signal transduction to target cells (Costanzo, 2011). Glial cells on the other hand, are specialized to support the functioning of neurons. Some enhance transduction of neuronal signals by formation of myelin sheaths around neuronal axons to enhance conductivity (oligodendrocytes) or take up excess neurotransmitters from the extracellular space (astrocytes) to maintain the membrane potential at the junctions. Others help maintain the blood- brain barrier, participate in inflammatory response and scar formation at sites of injury (astrocytes, microglia) or formation of new synapses and circuits after brain injury (synaptic plasticity) (Barrett et al., 2010). These activities of neurons and glial cells are highly metabolic and require constant supply of huge quantities of oxygen and nutrients as well as clearance of the by-products of those metabolic activities. These are the essential functions of the brain’s endothelial vasculature (Guyton & Hall, 2006). 2.2. Pathophysiology of stroke When a stroke occurs, be it ischaemic or haemorrhagic, brain cells are damaged from the interruption of blood supply to the brain (Mehta & Vemuganti, 2014). Hypoxia resulting from this interruption causes neurons to release large quantities of neurotransmitters like dopamine, serotonin and glutamate extracellularly (Tanaka & Ihara, 2017) and cells at the core of the ischaemia quickly die through necrosis. Though cells near the core injury site (ischaemic penumbra) maintain metabolism and low neuronal damage through blood supply from adjacent arteries, they lose the ability to maintain sodium (Na+) gradient due to ion channel dysfunction; therefore, intracellular Na+ levels become elevated, making astrocytes unable to reuptake glutamate from the extracellular fluid (Barrett et al., 2010). Glutamate continues to stimulate calcium (Ca2+) influx (glutamate excitotoxicity) and cells in the penumbra die (apoptosis) from the 7 University of Ghana http://ugspace.ug.edu.gh activities of calcium-dependent catabolic enzymes like proteases and nucleases (Mehta & Vemuganti, 2014). Also, proteases and oxidants are released from leukocytes activated by the restoration of blood flow to the ischaemic penumbra, causing cytokine-mediated inflammation and cell death, further increasing the area of damage (Pan, Konstas, Bateman, Ortolano, & Pile- Spellman, 2007). Figure 1: Mechanisms of seizure generation after stroke (adapted from Tanaka & Ihara, 2017) 2.3. Synaptic plasticity after stroke Be it ischaemia or haemorrhage, the effect of stroke is infarction of brain tissue and neuronal damage (Mehta & Vemuganti, 2014). These damages include physical, psychological and neurological debilitations such as muscle stiffness, paralysis, language difficulties and cognitive impairment (Tanaka & Ihara, 2017) and is dependent on the location, severity, and duration of loss of blood supply to the brain tissues (Mehta & Vemuganti, 2014). Neural plasticity is the structural and functional changes that occur in the central nervous system (CNS) in response to external or internal stimuli (Cohen, Quarta, Bravi, Granato, & Minciacchi, 2017). These changes can be in molecules, cells or networks such that a single change in either of 8 University of Ghana http://ugspace.ug.edu.gh these can result in changes in the functioning of the CNS. Neural plasticity is how the CNS maintains homeostasis and also allows for developmental changes (such as learning) (Feinberg, Koresko, & Heller, 1967; Sherwood, 2010). Recent studies have established that formation of new neurons (neurogenesis) occurs after birth and throughout life (Barrett et al., 2010; Cohen et al., 2017). These new cells are formed from stem cells in the dentate gyrus (hippocampus) and the subventricular zone of the lateral ventricles (Barrett et al., 2010). Afferent input to the brain maintains homeostasis of this neuronal proliferation; with increasing input reducing proliferation while balanced input increases proliferation- as over-proliferation can lead to disruptions in neuronal networks (Cohen et al., 2017). When harmful stimuli such as stroke strikes and unbalances the homeostatic state of the brain, synaptic plasticity is triggered and the brain attempts to regain its balance through creation of new synapsis (synaptogenesis), spreading of dendrites to sub-layers (dendritic arborisation) and the recruitment of synapses and axons (Sheorajpanday, Nagels, Weeren, & De Deyn, 2011; Sherwood, 2010) to the damaged region for reformation of lost connections needed for body functions. This is the principle behind rehabilitation therapy (Kanna & Heng, 2009). 2.4. Seizures in stroke Generally, the intracellular influx of calcium and sodium due to glutamate excitotoxicity, creates a hyperosmolar environment for neurons, depolarizing the cell membrane to a threshold low enough for seizure activation (Mehta & Vemuganti, 2014; Tanaka & Ihara, 2017). This is especially true for the hippocampus (a highly excitable part of the brain) where focal seizures with symptomatic hallucinations occur (Guyton & Hall, 2006). The increase in neuronal excitability 9 University of Ghana http://ugspace.ug.edu.gh could also be enabled by stroke-induced down-regulation of the inhibitory gamma-aminobutyric acid (GABA) receptors (Nudo, 2013; Tanaka & Ihara, 2017). Various mechanisms are thought to be responsible for seizure occurrence in stroke patients. These include a large size of the area of damage, cortical location of the stroke and the type of stroke. Haemorrhagic strokes have been found to be more likely to develop seizures (Tanaka & Ihara, 2017) through mechanism thought to be due to hemosiderin deposition which irritates the site where they are deposited, making the brain tissue more excitable and increasing the likelihood of focal seizures (Silverman et al., 2002; Tanaka & Ihara, 2017). 2.4.1. Early seizures Early seizures are thought to be due to metabolic mechanisms like glutamate excitotoxicity, calcium overload and oxidative stress (Silverman et al., 2002; Tanaka & Ihara, 2017). As the size of damage increases, the levels of excitotoxins released increase with increasing risk of early seizures (Mehta & Vemuganti, 2014; Pan et al., 2007) as in Figure 1 above. In haemorrhagic strokes, these seizures are associated with reduced consciousness or a history of seizures in patients (Naidech, 2011). 2.4.2. Late seizures Beyond two weeks of stroke onset, changes in neuronal networks are now persistent and neuroglia and immune cells have replaced the formerly healthy neurons. This gliotic scarring has been implicated as the main reason for the manifestation of late seizures in stroke patients (Silverman et al., 2002; Tanaka & Ihara, 2017). 10 University of Ghana http://ugspace.ug.edu.gh In addition, mechanisms of synaptic plasticity such as synaptogenesis, dendritic arborisation and the recruitment of synapses and axons (as described in the next section) increase brain excitability and the risk of seizures (Sheorajpanday, Nagels, Weeren, & De Deyn, 2011). 2.4.3. Subclinical seizures- EEG as a diagnostic tool In their study on the impact of seizures on stroke morbidity and mortality in 5076 stroke cases, Burneo, Fang and Saposnik (2010) stated that “electrographic seizures are common after stroke and that most patients with electrographic seizures may not have clinical correlates” (subclinical), a situation that can undermine the true incidence and impact of seizures on stroke and subsequently, clinical management. In such cases, continuous EEG (CEEG) is the recommended diagnostic tool (Naidech, 2011). In monitoring 570 consecutive critically ill patients with CEEG, Claassen et al. (2004) reported 19% subclinical seizures of which 84% was recorded on the first day of recording and another 5% on the second day. A study by Miskin et al. (2015) concurs that though 20 minutes of EEG recording is sufficient for capturing EEG abnormalities, a 40 minutes recording significantly increases the yield by 11%. In extrapolation, it can be said that the odds of capturing electrographic seizures in stroke patients using about 30 minutes of routine EEG is fair (as used for this study). 2.5. Relationship between EEG and action potential Poitre Olejniczak (2006) in his review on the Neurophysiologic basis of EEG said that synaptic activity is the most relevant source of the electrical potential recorded by the EEG; specifically, from apical dendrites of cortical neurons. He said that scalp electrodes record the summation of the excitatory and inhibitory postsynaptic potentials from these neurons. And also that other possible contributors to EEG potentials in synchronous events like sleep transients and epileptiform discharges are calcium mediated action potentials (calcium spikes) and individual 11 University of Ghana http://ugspace.ug.edu.gh high-amplitude, fast action potentials (sodium) which are short-acting (Olejniczak, 2006). The oscillatory waves recorded by the EEG are grouped according to frequency bands (delta, alpha, beta, gamma etc.) (Sheorajpanday, Nagels, Weeren, van Putten, et al., 2011). Depending on the status of the individual at the time of recording, the appearance of these frequency bands can be physiologic or pathologic (Guérit et al., 1999). 2.6. Measures of functional outcome in stroke patients In clinical care and rehabilitation therapy after stroke, assessment of deficit and recovery is essential. The NIHSS is a well-validated tool for assessing neurological deficits (Kwah & Diong, 2014) and a good predictor of the likelihood of recovery from stroke (Saposnik et al., 2011) but has items with imbalances in scoring based on the hemisphere in which stroke occurs. These items have been eliminated in the more reliable mNIHSS which has scoring between 0 for normal neurological function and 31 for no neurological function (Meyer & Lyden, 2009). However, neither the NIHSS nor mNIHSS has the ability to assess difficulties such as bed mobility, sitting, standing, walking nor upper limb function (Kwah & Diong, 2014). The Barthel Index (BI) is an ideal measure of these activities of daily living (ADL) such as eating, bathing, grooming and toilet use, with scores ranging from 0 to 20 (between no ability and normal function). Neither mNIHSS nor BI clearly indicates whether a patient is alive or dead by their scores. The Modified Rankin Scale (MRS) is a descriptive measure of survival with scores from 0 (normal) to 6 (dead) and is primarily used to assess disability status of stroke patients as described in Appendix 3 (Martin- Schild et al., 2011; Veerbeek, Kwakkel, Van Wegen, Ket, & Heymans, 2011). 12 University of Ghana http://ugspace.ug.edu.gh 2.7. Quantitative EEG (QEEG)- another measure of functional outcome Studies carried out proves that EEG recordings (in either acute or sub-acute stage of stroke) generated from a standard number of electrodes (10/20 system), when quantitatively analysed is a clinically relevant measure of post-stroke status and outcome, independent of comorbidities and stroke type (Sheorajpanday, Nagels, Weeren, van Putten, & De Deyn, 2009). In brain ischaemia, delta waves (1-4Hz) with high amplitudes are usually observed at the regions of ischaemia showing slowing of brain activity with a converse reduction in the occurrence of faster waves like alpha (8-12Hz) (Sheorajpanday et al., 2009). The QEEG indices used in monitoring brain activity after stroke is based on mathematical derivatives of these frequency band measures. Examples are the delta/alpha ratio (DAR) and relative alpha power (defined as equations 1 and 2 in the methodology section). Delta activity is not characteristic of a normal awake brain (Table 2) so the higher the ratio of delta to alpha spectral power, the greater the abnormality and vice versa while a low relative alpha power is indicative of abnormality in brain activity (Finnigan & van Putten, 2013; Kanna & Heng, 2009; Sheorajpanday, Nagels, Weeren, van Putten, et al., 2011). The brain symmetry index (BSI) is another QEEG index defined by Van Putten & Tavy, 2004 and modified by Sheorajpanday et al., 2009 as the pairwise derived brain symmetry index (pdBSI) for comparison of spectral power in contralateral brain locations (such as frontal left to frontal right) over a specified frequency range (1-4Hz, 1-7Hz, 1-13Hz, 1-25Hz etc.) to determine the presence of asymmetry- an indicator of abnormal brain activity at the location with higher asymmetry (Sheorajpanday et al., 2009). 13 University of Ghana http://ugspace.ug.edu.gh Significant correlations have been established between QEEG indices and stroke outcomes such as pairwise derived brain symmetry index’s (pdBSI) correlation with NIHSS in discriminating between stroke and transient ischaemic attack (TIA) or controls (Sheorajpanday et al., 2009); pdBSI in predicting early worsening of outcome (Sheorajpanday, Nagels, Weeren, De Surgeloose, & De Deyn, 2010); delta/alpha power ratio (DAR) and relative alpha power (RAP) were also significantly associated with NIHSS score at 30 days post stroke (Finnigan & van Putten, 2013). Even without using the quantitatively obtained EEG indices, the presence of slow wave activity has also been found to correlate with worsening functional outcome in patients with intracranial haemorrhage (Grays et al., 2016). Since increasing population age is projected to increase the financial burden stroke has on the economy (Hsieh et al., 2017), all efforts must be made to improve stroke outcome; more so as elderly people living with seizures have been reported in the United States of America to spend double the amount those without seizures spend on health care (Lekoubou, Bishu, & Ovbiagele, 2018). In this study, the QEEG variables used were relative alpha power (RAP), the delta/alpha ratio (DAP) and the pairwise derived brain symmetry index (pdBSI) (Sheorajpanday et al., 2009) described in the next chapter. Each of these indices have been found to correlate and even predict functional outcome in stroke patients when measured on admission or some time in their recovery in more than one study (Agius Anastasi, Falzon, Camilleri, Vella, & Muscat, 2017; Finnigan & van Putten, 2013; Kanna & Heng, 2009). 14 University of Ghana http://ugspace.ug.edu.gh 3.0 METHODOLOGY 3.1 Study Design: This was a longitudinal study where all patients who met the eligibility criteria and consented were recruited. 3.2 Study Site: The study was conducted at the stroke unit of the Korle Bu Teaching Hospital. The hospital is the first of the three Teaching Hospitals to be established in Ghana. It is a tertiary care facility with a bed capacity of 2000. It receives referrals from other regional and district hospitals especially in the southern sector of the country. Korle Bu Teaching Hospital has a surgical and medical emergency facility run jointly by the Surgical and Medical Departments of the hospital. It is the first point of call for patients coming into the hospital with non-trauma emergency where patients are admitted for a few hours or days till they can be discharged or transferred to other wards for further management. From the SME and other hospital departments, patients diagnosed with stroke confirmed by neuroimaging are admitted to the stroke unit, a 20-bed facility which is part of the medical department of KBTH. The stroke unit is equipped with specially trained multidisciplinary team (MDT) of clinical staff who assess and manage patients admitted to the unit. Patients for this study were recruited from this unit. In addition, the hospital has a Neurophysiology unit where EEGs are recorded and a radiology department where Computed Tomography scans and Magnetic Resonance Imaging for stroke patients are done by competent experts. 15 University of Ghana http://ugspace.ug.edu.gh 3.3 Study population: All patients admitted to the stroke unit of the Korle Bu Teaching Hospital with stroke confirmed by neuroimaging (CT (computed tomography) scan or MRI) within 10 days of symptom onset. 3.4 Inclusion Criteria Subjects were eligible if they satisfied the following criteria:  They meet the definition of stroke  Evidence of infarction or haemorrhage on Head with CT scan or MRI  First ever symptomatic stroke irrespective of level of consciousness  Presentation within 10 days of stroke onset.  Aged 18 or above  Given informed consent or consent by next of kin to participate in study. 3.5 Exclusion Criteria Subjects were excluded if they had any of the following:  Previous symptomatic stroke  History of chronic disability (physical, mental) prior to illness.  No evidence of infarct or haemorrhage on CT scan or MRI.  Neuroimaging shows other stroke mimics such as tumours, cerebral abscess and other mass lesions.  Did not satisfy inclusion criteria. 16 University of Ghana http://ugspace.ug.edu.gh 3.6 Sample size: The sample size was calculated using the software by Sergeant, 2018 from http://epitools.ausvet.com.au/content.php?page=cohortSS&P1=0.05&RR=10&Conf=0.95&Powe r=0.8 with the following parameters:  Expected incidence in unexposed population= 0.05  Assumed relative risk= 10  Confidence level= 0.95  Power= 0.8 This calculation yielded a sample size of twenty four (24), which has been approximated to thirty to cater for potential drop outs. 3.7.0 Procedures 3.7.1 Selecting study participants Doctors of the Korle Bu Teaching Hospital stroke team are routinely called to see stroke cases in the hospital’s wards and emergency unit (SME). The names of patients diagnosed with stroke (by neuroimaging) were obtained from these doctors and their eligibility was assessed using the criteria stated above. Consecutive patients who met the eligibility criteria were selected for inclusion in the study. 3.7.2 Obtaining consent Eligible patients who were lucid and gave consent were informed about the study and given an opportunity to ask questions. Those who agreed to participate were asked to sign the consent form (Appendix 1) and recruited into the study. For patients who could not comprehend or give consent, the study was explained to their next of kin and if they consented, they signed on behalf of the patient (as clearly shown in Appendix 1). 17 University of Ghana http://ugspace.ug.edu.gh 3.7.3 Demographic data Questionnaires (as in Appendix 2) were used to gather participant demographics including participant ID, date of birth and age. The contact details of participants and their proxy were stored on a separate document for one-month follow-up visit. Details on the incidence of clinical seizures before stroke occurred and within the one month of study were documented on the questionnaire. Patients’ use of antiepileptic drugs within the study period were also documented as they may reduce the occurrence of seizures in stroke patients (Tanaka & Ihara, 2017). 3.7.4 Stroke classification The inclusion criteria for this study required participants’ stroke status to have been confirmed by neuroimaging (CT scan or MRI). Patients neuroimaging reports reviewed and approved by stroke specialists prior to admission were used to fill in details on stroke classification in the questionnaire (Appendix 2) such as stroke type, cortical location, hemispheric location and subcortical structures affected. 3.7.5 Outcome measures Modified Rankin Scale (Appendix 3) was measured for each patient recruited to rank their level of disability after the stroke, both at recruitment (MRS0) and a month (MRS1) after stroke started (Cincura et al., 2009; de Haan, Limburg, Bossuyt, van der Meulen, & Aaronson, 1995). The modified National Institute of Health Stroke Scale (Appendix 4) was filled for each patient at recruitment (mNIHSS0) and a month (mNIHSS1) after stroke started (Lyden et al., 2001; Meyer & Lyden, 2009) as a measure of neurological deficit. The Barthel Index (Appendix 5) was recorded at recruitment (BI0) and at one-month (BI1) as a measure of activities of daily living (Kwakkel et al., 2010; Liu, Unick, Galik, & Resnick, 2015). 18 University of Ghana http://ugspace.ug.edu.gh The three outcome measures were then categorized in the order in Table 1 below into three comparable categories of mild, moderate or severe (Govan, Langhorne, & Weir, 2009). Table 1: Categorization of outcome measures Stroke Severity Categorization Categories Scale Mild (1) Moderate (2) Severe (3) Modified NIHSS 0-5 6-14 15-31 Modified Rankin Scale 0-3 4 5 Barthel Index (20 point scale) 10-20 3-9 0-2 3.7.6 EEG recording The 32 channel portable EEG Maximus 24/32* (“RMS-Maximus Electroencephalograph,” 2013) was used to record the electrical activity of the brain. The scalp of each participant was thoroughly cleaned with abrasive gel to remove all dirt and dead cells to enhance connectivity of electrodes. Electrodes were connected to the scalp with conductive gel according to the international 10/20 system of electrode placement (Kaiser, 2007). A 19 channel awake-EEG recording (with electrode positions Fz, Cz, Pz, Fp1, Fp2, F3, F4, F7, F8, C3, C4, T3, T4, P3, P4, T5, T6, O1 and O2) was done for patients in the ‘eyes-open’ and ‘eyes-closed’ states for 30 minutes including a 3 minutes period of photic stimulation. Patients who were conscious and cooperative did another brain activation test by hyperventilation. The EEG recordings were stored on the recording computer as well as a hard-drive (after appropriate de-identification). 19 University of Ghana http://ugspace.ug.edu.gh 3.7.7 EEG pattern classification and seizure diagnosis Each recording was visually characterized by a neurologist with attention to the general pattern and characteristics of the recording including comments on seizure activity and slow wave activity (characteristics which point to pathophysiology). This data was entered using the questionnaire (Appendix 2). The frequency bands to be used for EEG characterization and quantification are as in Table 2: Table 2: EEG frequency bands EEG frequency Typically observed during Delta (1-4Hz) Dominant during deep sleep. High delta levels show more stable sleep state Theta (4.1-8Hz) Light sleep or extreme relaxation state Alpha (8.1-12.5Hz) Relaxed wakefulness. Beta (12.6-30Hz) Wide awake and focused Gamma (≥30.1Hz) Formation of ideas, memory processing and learning Another characteristic EEG activity is the ‘spike and wave’ activity known as epileptiform discharge (Boro, 2016). Since slow wave activity has also been found to correlate with worsening functional outcome in patients with intracranial haemorrhage, all EEGs recorded were analysed to determine presence of slow wave activity (Grays et al., 2016). The recordings were quantitatively analysed on the EEGlab software in MATLAB for relative alpha power (RAP), delta/ alpha power ratio (DAR) (Finnigan et al., 2007; Kanna & Heng, 2009) and pairwise derived brain symmetry index (pdBSI) which have been found to be significantly correlated with 30-day NIHSS score in ischaemic stroke (Sheorajpanday et al., 2009). 20 University of Ghana http://ugspace.ug.edu.gh 3.7.8 EEG pre-processing EEG recordings were de-identified and imported into the EEGlab software (Delorme & Makeig, 2004; Sheorajpanday, Nagels, Weeren, & De Deyn, 2011) in EDF format, choosing default (BESA) channel locations (Budzynski, Budzynski, Evans, Abarbanel, & Thatcher, 2009; Phillips, Rugg, & Friston, 2002) with exclusion of channels which were not part of the conventional 19- channel EEG recording electrodes. All electrodes were re-referenced to average reference (Sheorajpanday et al., 2009; Sheorajpanday, Nagels, Weeren, van Putten, et al., 2011; Thatcher & Lubar, 2009). The recording was then filtered using a highpass filter of 0.3Hz and a lowpass filter of 30Hz (Sheorajpanday et al., 2009), after which line-noise was removed by means of the CleanLine toolbox (Mullen, 2012) in EEGlab. Sections of the recording which had high artifacts were automatically selected and after visual inspection, they were deleted. Although no channels were deleted, at this stage, channels with high voltage artifacts even after filtering were excluded from spectral analysis. 3.7.9 Quantitative EEG analysis For each patient, 5 minutes of artifact-free EEG data was selected from the recording for spectral analysis according to the QEEG experts recommendation (Budzynski et al., 2009). Delta/ Alpha Ratio (DAR), Relative Alpha Power (RAP) and the pairwise derived Brain Symmetry Index (pdBSI) were computed from EEG recordings using modified algorithms on the EEGlab software according to the formulae below. Equation 1: Equation 2: 𝐾 4𝑁/𝑓𝑠 𝐾 13𝑁/𝑓𝑠∑1 ∑𝑁/𝑓𝑠 𝑃 ∑1 ∑8𝑁/𝑓𝑠 𝑃 𝐃𝐞𝐥𝐭𝐚: 𝐀𝐥𝐩𝐡𝐚 𝐑𝐚𝐭𝐢𝐨 = 13𝑁/𝑓𝑠 𝐑𝐞𝐥𝐚𝐭𝐢𝐯𝐞 𝐀𝐥𝐩𝐡𝐚 𝐏𝐨𝐰𝐞𝐫 = 𝐾 25𝑁/𝑓𝑠 ∑1 ∑ 𝑃 ∑𝐾1 ∑ 𝑃8𝑁/𝑓𝑠 𝑁/𝑓𝑠 21 University of Ghana http://ugspace.ug.edu.gh Where:  P is the absolute power at a given frequency  K is the discrete number of channels  N is the Fast Fourier Transform (NFFT)  fs is the sampling frequency rate in Hertz Equation 3: 1 pdBSI = ∑𝑀 𝑁 𝑅 ∑ 𝑖𝑗 −𝐿𝑖𝑗 𝑗=1 𝑖=1 | | where: 𝑀𝑁 𝑅𝑖𝑗+𝐿𝑖𝑗  Rij = the FFT based power spectral density using Welch’s method of signal obtained from the right channel homologue  Lij = the FFT based power spectral density using Welch’s method of signal obtained from the left channel homologue  i= the pair of homologous channels (1, 2,...,M)  j= the frequency (Fourier coefficient with index j= 1, 2,…,N)  M=8 and  N= 1-4Hz, 1-8Hz, 1-12.5Hz, 1-30Hz for (delta, theta, alpha and beta respectively) 3.8 Data handling All EEG recordings were de-identified and secured on an external hard-drive with password and locked in a steel cabinet in a secure room when not in use. Consent forms and questionnaires generated were also secured in separate locations to maintain patient privacy and confidentiality. 22 University of Ghana http://ugspace.ug.edu.gh 3.9 Statistical analysis Data collected were analysed using the Microsoft Excel (version 2013) and the Statistical Product for Service Solution (SPSS) version 20. Bar charts frequency tables were used to represent categorical data collected and tables of means with standard deviations were used to summarize data on continuous variables, including p-values for proportions at a significance level of 95%. To determine predictors of one-month functional outcomes, Spearman’s correlations and linear regressions were analysed at a significance level of 95% (p < 0.05). 3.10 Ethical issues Ethical clearance was sought from the Ethical and Protocol Review Committee (EPRC) of the College of Health Sciences (Appendix 6). The details of the research were explained to potential participants before enrolment, emphasising that participation is voluntary and their choice will not affect the quality of care the hospital gave them. Data obtained was de-identified and coded to maintain confidentiality, stored in secure rooms with locks and on computers with passwords. 23 University of Ghana http://ugspace.ug.edu.gh 4.0 RESULTS This chapter displays data collected from 30 stroke patients in bar charts and statistical tables to answer the objectives of this research. Data that answers the first objective are in the first four sub- sections and most of the data for the second research objective are in the last three subsections. 4.1 Patient demographics: Age and gender characteristics of 30 stroke patients Of the 30 stroke patients studied, 18 were 60 years old or below (60%) and 12 patients were older than 60 years (40%), but the difference between the two proportions was not significant (p = 0.12). From the student’s t-test for unpaired data, the mean age of patients who had ischaemic stroke (63.1 ± 15.3years) was significantly higher than those who had haemorrhagic stroke (48.1 ± 15.1years), (p = 0.02) as shown in Table 3. There was also a significant difference in the gender ratio of the patients, 10 females (33.3%) and 20 males (66.7%), (p = 0.01). Table 3: Patients’ demographic characteristics Age group ≤ 60 years > 60 years p-value Number (percentage) 18 (60%) 12 (40%) 0.12 Stroke type Ischaemic Haemorrhagic p-value (Mean ± SD) (Mean ± SD) Mean age (years) 63.1 ± 15.3 48.1 ± 15.1 0.02* Gender Female Male p-value Number (percentage) 10 (33.3%) 20 (66.7%) 0.01* *significant at p < 0.05; SD= standard deviation 24 University of Ghana http://ugspace.ug.edu.gh 4.2. Clinical characteristics of patients According to neuroimaging findings (shown in Figure 2), of the 20 ischaemic and 10 haemorrhagic stroke patients studied, 21 had lesions in the cerebral cortex (16 ischaemic and 5 haemorrhagic strokes, p = 0.17) and 17 patients had lesions in subcortical structures (13 ischaemic and 4 haemorrhagic strokes, p = 0.29); including patients who had no cortical lesions. The differences in proportions were not significant (at 95% significance level) after performing the z-test for two proportions. In terms of hemisphere where stroke occurred, 10 patients had left hemisphere lesions (8 ischaemic, 2 haemorrhagic), 13 had right hemisphere lesions (8 ischaemic, 5 haemorrhagic), 4 had bilateral hemispheric lesions (2 ischaemic, 2 haemorrhagic) but 3 patients had no hemispheric categorization of stroke. However, none of the difference in proportions of ischaemic to haemorrhagic were significant. 18 16 16 14 12 13 10 8 8 8 6 4 5 5 4 2 2 2 2 1 2 0 Cerebral cortex Subcortical Left hemisphere Right Bilateral Hemisphere structures hemisphere hemisphere unstated Area of damage in the brain Ischaemia Haemorrhage Figure 2: Neuroimaging findings of study participants 25 Number of patients University of Ghana http://ugspace.ug.edu.gh Table 4 shows the distribution of functional status among patients at recruitment and one-month post stroke. At recruitment, 36.7% had mild disability on MRS, 36.7% had moderate disability and 26.7% had severe disability. These percentages improved to 63.3% mild, 20.0% moderate and 16.7% severe disability for assessment done at one-month post stroke. The distribution of neurological deficit (mNIHSS) in the patients at recruitment was 41.4% mild, 44.8% moderate and 13.8% severe neurological deficit which improved to 71.4% mild, 21.4% moderate and 7.1% severe neurological deficit a month later. Inability to perform activities of daily living at recruitment was in a ratio of 53.3% mild inability, 13.3% moderate inability and 33.3% severe inability at recruitment with increase in the percentage who had mild and moderate inability as 73.3% and 20% respectively, and a decrease in the percentage with severe inability as 6.7% at one- month. Table 4: Functional status of 30 stroke cases at recruitment and follow-up Outcome Severity at recruitment Severity at one-month measure Mild Moderate Severe Mild Moderate Severe MRS 36.67% 36.67% 26.67% 63.33% 20.00% 16.67% mNIHSS 40.00% 46.67% 13.33% 66.67% 26.67% 6.67% BI 53.33% 13.33% 33.33% 73.33% 20.00% 6.67% 26 University of Ghana http://ugspace.ug.edu.gh 4.3.0 Occurrence of seizures 4.3.1 Diagnostic parameters of EEG recordings Based on the neurologist’s report on the 30 EEG recordings in Table 5, a significantly lower number of patients (36.7%) had normal readings (p = 0.04), 14 patients (46.7%) had EEG activity diagnostic of electrographic seizures (13 generalized seizures and 1 focal seizures) and 17 (56.7%) had activity signifying cerebral dysfunction (these patients had abnormal EEG recordings with or without electrographic seizure activity). Further details on the EEG findings and occurrence of clinical seizures in the patients studied are displayed in Figure 3 below. Table 5: EEG findings in 30 stroke patients Normal EEG activity Abnormal EEG Activity p-value Number (percentage) 11 (36.7%) 19 (63.3%) 0.04* Generalized seizures Focal seizures p-value Number (percentage) 13 (92.9%) 1 (7.1%) < 0.001* Cerebral dysfunction No cerebral dysfunction p-value Number (percentage) 17 (56.7%) 13 (43.3%) 0.302 * Significant at p <0.05 27 University of Ghana http://ugspace.ug.edu.gh 4.3.2 Occurrence of seizure and EEG abnormalities in ischaemic and haemorrhagic stroke patients Figure 3 shows that, of the 19 patients with aberrant brain activity on recorded EEGs, 12 were from ischaemic stroke patients and 7 from haemorrhagic strokes. Electrographic seizures were seen in 9 ischaemic stroke patients and 5 haemorrhagic stroke patients. Clinical seizures were observed in 4 patients in a 1:1 ratio for ischaemic and haemorrhagic strokes including 2 patients who also had seizure activity on EEG (1 ischaemic and 1 haemorrhagic stroke); however none of these difference in proportions were significant. 14 12 10 8 6 12 9 4 7 5 2 2 2 1 1 0 Abnormal EEG Electrographic seizures Clinical seizures Clinical & electrographic seizures Patient's diagnosis Ischaemia Haemorrhage Figure 3: Electroencephalographic and clinical details on seizures 28 Number of patients University of Ghana http://ugspace.ug.edu.gh 4.3.3 Details on patients with clinical seizures Further details on the 4 cases who had clinical seizures (in Table 6 below) showed that the two ischaemic stroke patients had only early seizures (1 each), one of the two haemorrhagic stroke patients had early seizures (3) while the other patient had both early and late seizures (2 each). Three of the four patients were given anti-epileptic drugs (Levetiracetam, Phenytoin and both Levetiracetam and Carbamazepine respectively). Of the two who had no seizure activity on EEG, one had a normal EEG recording while the other had evidence of cerebral dysfunction on EEG and one of the two who had seizure activity on EEG also had evidence of cerebral dysfunction. Table 6: Early seizures, late seizures and the use of anti-epileptic drugs in stroke patients ID Stroke type Early Late Total No. of AEDs EEG seizures seizures seizures comments 004 Ischaemia 1 0 1 Levetiracetam Normal EEG 006 Ischaemia 1 0 1 None GS,CD 010 Haemorrhage 2 2 4 Phenytoin CD Levetiracetam, GS 028 Haemorrhage 3 0 3 Carbamazepine GS = generalized seizure activity; CD = cerebral dysfunction activity; AEDs = anti- epileptic drugs 29 University of Ghana http://ugspace.ug.edu.gh 4.4.0: Visual characterization of EEG recordings 4.4.1 Characterization of electrographic seizure activities Observing the types of EEG activity seen in the 14 patients that led to a diagnosis of electrographic seizures (in Figure 4), almost all patients (13) had spike and slow wave activity, 11 had sharp waves on recording, 5 patients also had spike and wave activities but only one patient had diffuse slowing characteristic of seizures. 14 13 12 11 10 8 6 5 4 2 1 0 difuse slowing spike and wave sharp waves spike slow wave activity activity activity activity Type of epileptiform activity Figure 4: Categories of epileptiform activity seen on EEG 30 Number of EEG recordings University of Ghana http://ugspace.ug.edu.gh 4.4.2 Electrographic indications of cerebral dysfunction Of the three types of EEG activities showing cerebral dysfunction in 17 patients, the most prevalent was diffuse slowing (found in 14 patients), followed by focal slowing ( in 10 patients) but only one patient had slow burst activity. 16 14 12 10 8 6 4 2 0 slow burst activity focal slowing diffuse slowing Type of dysfunctional brain activity Figure 5: Distribution of dysfunctional brain activity on EEG 31 Number of EEG recordings University of Ghana http://ugspace.ug.edu.gh 4.5.0 Comparison of functional outcome at recruitment to one-month measures 4.5.1 Relationship between mean functional status at recruitment and one-month after stroke The mean MRS of patients one-month after stroke (2.9 ± 1.4) was significantly lower (p = 0.001) than the mean MRS at recruitment (3.6 ± 1.3) as was the mean mNIHSS a month after stroke (5.2 ± 7.0) to the mean mNIHSS at recruitment (7.4 ± 5.8) with p = 0.007. The mean Barthel Index at one-month post stroke (13.4 ± 7.6) was however significantly higher (p = 0.001) than the mean Barthel Index at recruitment (9.4 ± 7.3) as displayed in Table 7. Table 7: Mean functional status measures at recruitment and one-month post stroke At recruitment One-month after stroke p-value MRS 3.6 ± 1.3 2.9 ± 1.4 0.001* mNIHSS 7.4 ± 5.8 5.2 ± 7.0 0.007* Barthel Index 9.4 ± 7.3 13.4 ± 7.6 0.001* *Significant at p < 0.05 4.6. Correlation between recruitment measures of functional status and functional outcome values at one-month Two-tailed Spearman’s rank correlation analysis (Table 8) with significance level set at 95% showed the following relationships between the variables of interest: At p < 0.001, MRS0 showed a significant positive correlation to mNIHSS0 (rho = 0.753), mNIHSS1 (rho = 0.754) and MRS1 (rho = 0.847) and a significantly negative correlation to BI0 (rho= -0.932) and BI1 (rho = -0.813); with its strongest correlation being to BI0. 32 University of Ghana http://ugspace.ug.edu.gh Apart from its correlation with MRS0, mNIHSS0 showed strong and significant negative correlation to BI0 (rho = -0.800) and BI1 (rho = -0.896) and a significantly strong positive correlation with MRS1 (rho= 0.883) and mNIHSS1 (rho= 0.877). Other significant negative correlations for BI0 were to MRS1 (rho= -0.853) and mNIHSS1 (rho= -0.769), while a significant positive correlation was with BI1 (rho= 0.852). In addition to the correlations with functional status measures at recruitment (MRS0, mNIHSS0 and BI0), there was a significant positive correlation between MRS1 and mNIHSS1 (rh0 = 0.917) and a significant negative correlation between mNIHSS1 and BI1 (rho = -0.895) all at p < 0.001. Of the 3 QEEG indices analysed (DAR, RAP and pdBSI), there was no significant correlation to known functional measures (MRS0, MRS1, mNIHSS0, mNIHSS1, BI0 and BI1) or among the QEEG indices, except for a negative correlation between DAR and RAP (rho= -0.988, p < 0.001), which was the highest correlation coefficient in the entire analysis. 33 University of Ghana http://ugspace.ug.edu.gh Table 8: Correlations between measures of functional status MRS0 mNIHSS0 BI0 MRS1 mNIHSS1 BI1 DAR RAP pdBSI MRS0 rho 1.000 0.753 * -0.932* 0.847* 0.754* -0.813* -0.190 0.189 0.049 p-value 0.000 0.000 0.000 0.000 0.000 0.316 0.316 0.796 mNIHSS rho 0.753* 1.000 -0.800* 0.883*0 0.877 * -0.896* -0.048 0.021 -0.057 p-value 0.000 0.000 0.000 0.000 0.000 0.800 0.912 0.765 BI0 rho -0.932 * -0.800* 1.000 -0.853* -0.769* 0.852* 0.126 -0.125 -0.074 p-value 0.000 0.000 0.000 0.000 0.000 0.508 0.512 0.698 MRS1 rho 0.847 * 0.883* -0.853* 1.000 0.917* -0.959* -0.099 0.089 0.052 p-value 0.000 0.000 0.000 0.000 0.000 0.601 0.638 0.785 mNIHSS1 rho 0.754 * 0.877* -0.769* 0.917* 1.000 -0.895* -0.039 0.004 -0.126 p-value 0.000 0.000 0.000 0.000 0.000 0.839 0.984 0.506 BI1 rho -0.813 * -0.896* 0.852* -0.959* -0.895* 1.000 0.096 -0.077 -0.007 p-value 0.000 0.000 0.000 0.000 0.000 0.614 0.686 0.970 DAR rho -0.190 -0.048 0.126 -0.099 -0.039 0.096 1.000 -0.988* 0.124 p-value 0.316 0.800 0.508 0.601 0.839 0.614 0.000 0.512 RAP rho 0.189 0.021 -0.125 0.089 0.004 -0.077 -0.988* 1.000 -0.070 p-value 0.316 0.912 0.512 0.638 0.984 0.686 0.000 0.714 pdBSI rho 0.049 -0.057 -0.074 0.052 -0.126 -0.007 0.124 -0.070 1.000 p-value 0.796 0.765 0.698 0.785 0.506 0.970 0.512 0.714 *significant at p < 0.01 (2-tailed) 34 University of Ghana http://ugspace.ug.edu.gh 4.7.0 Prediction of functional outcome of patients at one-month post-stroke Regression analysis of all patients (Table 9) showed that none of the three QEEG indices significantly predicted functional outcomes (MRS1, mNIHSS1 or BI1), however mNIHSS0 positively predicted MRS1 (beta = 0.471, p = 0.001) and mNIHSS1 (beta = 0.753, p = 0.001) and negatively predicted BI1 (beta= -0.556, p = 0.001) as shown in Table 9. In addition, BI0 positively predicted BI1 (beta = 0.519, p = 0.001). These predictions were significant at p < 0.05. Table 9: Regression analysis between functional outcomes and QEEG indices of all patients MRS1 mNIHSS1 BI1 Beta p-value Beta p-value Beta p-value DAR -0.051 0.849 -0.051 0.849 0.105 0.701 RAP -0.146 0.594 -0.146 0.594 0.044 0.873 pdBSI -0.146 0.464 -0.146 0.464 0.021 0.917 MRS0 0.257 0.131 0.030 0.905 0.163 0.410 mNIHSS0 0.471* 0.001 0.753* 0.001 -0.556* 0.001 BI0 -0.260 0.176 -0.027 0.925 0.519* 0.026 * Beta (standardized coefficient) is significant at p < 0.05 35 University of Ghana http://ugspace.ug.edu.gh 4.7.1 Regression analysis according to stroke type In a sub-analysis of linear regression between recruitment parameters (MRS0, mNIHSS0, BI0, DAR, RAP and pdBSI) and functional outcome measures in ischaemic strokes (Table 10) significantly, mNIHSS0 positively predicted mNIHSS1 (beta = 1.104, p = 0.003), and negatively predicted BI1 (beta = -0.694, p = 0.012) while RAP negatively predicted mNIHSS1 (beta = -0.668, p = 0.021). For haemorrhagic strokes, MRS0 positively predicted MRS1 (beta = 0.536, p = 0.033) and mNIHSS0 positively predicted MRS1 (beta = 0.961, p = 0.012). There were no other significant predictions. The import of all the results displayed in this section are discussed in the next section. 36 University of Ghana http://ugspace.ug.edu.gh Table 10: Regression analysis between outcome measures by stroke type Ischaemic stroke Haemorrhagic stroke MRS1 mNIHSS1 BI1 MRS1 mNIHSS1 BI1 Beta p-value Beta p-value Beta p-value Beta p-value Beta p-value Beta p-value MRS0 0.116 0.666 -0.024 0.939 0.171 0.509 0.536* 0.033 0.237 0.522 0.052 0.849 mNIHSS0 0.323 0.218 1.104* 0.003 -0.694* 0.012 0.961* 0.012 0.99 0.09 -0.959 0.051 BI0 -0.525 0.105 0.188 0.609 0.42 0.168 0.446 0.072 0.346 0.427 0.062 0.843 DAR -0.031 0.881 -0.458 0.083 0.358 0.091 -0.058 0.642 -0.029 0.917 -0.005 0.983 RAP -0.001 0.996 -0.668* 0.021 0.318 0.144 -0.117 0.335 -0.35 0.235 0.274 0.224 pdBSI 0.093 0.478 -0.109 0.482 -0.034 0.786 0.103 0.387 -0.134 0.609 0.026 0.895 * Beta (standardized coefficient) is significant at p < 0.05 37 University of Ghana http://ugspace.ug.edu.gh 5.0 DISCUSSION The implications of the findings of this research are discussed below 5.1. Patient demographics and clinical characteristics A high number of patients in this study were in the pre-retirement age (≤60years) and according to stroke type, a significant number of them were haemorrhagic strokes. A report on the global burden of stroke by Feigin, Norrving, & Mensah (2017) documented that in low and middle income countries like Ghana, 68% of strokes which occurred in 2010 were in persons younger than 75years. This two-decade (1990-2010) analysis of the global burden of stroke showed a 25% increase in the percentage of young and middle aged adults (20-64years) who suffered stroke, contributing to 31% of all stroke cases (Feigin et al., 2017). These stroke in young adults have been attributed to high body mass index, low vegetable intake and air pollution from household fuels while in middle aged adults systolic blood pressure and other metabolic factors have been attributed to the cause of stroke. The implication of these findings is that low and middle income countries like Ghana may lose a number of their younger work-force to stroke associated morbidity and mortality causing further diminution of the nation’s productivity (Feigin et al., 2016). Haemorrhagic stroke patients formed 33.3% of the stroke cases studied. This finding was similar to findings from the INTERSTROKE study of stroke in 22 countries worldwide, which found haemorrhagic stroke accounted for 34% of stroke in Africa (O’Donnell et al., 2010) and the SIREN study of stroke in Ghana and Nigeria, which found the frequency of haemorrhagic stroke to be 32% (Kengne & Mayosi, 2018). When compared to high income countries such as Canada which reported 18.7% (Burneo et al., 2010) and the report from the INTERSTROKE study which showed strokes caused by haemorrhage to contribute as low as 9% of all strokes (O’Donnell et al., 2010), 38 University of Ghana http://ugspace.ug.edu.gh it becomes imperative that more focus be given to the reduction of the occurrence and impact of haemorrhagic strokes in Ghana and Africa at large (Feigin et al., 2017). This high percentage of haemorrhagic stroke in middle and low income countries has been attributed to changes in life- style and increase in urbanization in Africa (Feigin et al., 2016). The male stroke patients in this study were twice the number of female patients. This finding follows the trend in the review of global stroke statistics given by Thrift et al. in 2017 for 205 countries (including some African countries but not Ghana) in all of whom incidence was more in males than females. This variation has been attributed to difference in risk factors for stroke by gender (Roth et al., 2017). In this study, 70% of the patients had lesions in the cerebral cortex; especially the ischaemic stroke patients (though the difference in proportions between ischaemic and haemorrhagic strokes was not significant). In terms of hemispheric location of brain lesion, this study found no significant difference between ischaemic and haemorrhagic strokes; though the group was a mixture of left, right and bilateral hemisphere strokes. As cortical location of damage is one of the known risk factors of post stroke epilepsy, the occurrence of cortical damage in the study participants may have increased the likelihood of seizure occurrence in them (Packiaseeli et al., 2017; Tanaka & Ihara, 2017). When the connections between the two brain hemispheres are damaged in subcortical lesions, the degeneration of these connections could even provoke post stroke seizures (Klein et al., 2018; Tanaka & Ihara, 2017). Subcortical lesions may also lead to secondary degeneration of the cerebral cortex, therefore the occurrence of purely subcortical lesions in a few of the patients in this study did not exempt them from cortical damage (Tanaka & Ihara, 2017). 39 University of Ghana http://ugspace.ug.edu.gh Majority of the group had mild to moderate stroke severity at recruitment with improvement in one-month variables (MRS, mNIHSS and BI). More severe neurological deficits at the initial stages of stroke (as measured by NIHSS) have been found to be associated with the occurrence of epilepsy after stroke (Klein et al., 2018; Tanaka & Ihara, 2017) and death (Nkusi et al., 2017); though no death was recorded within this study. This is because the larger the cortical area of damage, the greater the severity and hence the greater the risk of post stroke epilepsy. 5.2. Seizure occurrence in stroke patients The 13.3% frequency of seizures found in this study was slightly higher than the 11% stroke occurrence in over 530 stroke cases from the Korle Bu Teaching Hospital’s stroke unit (unpublished data) and much higher than the 8.9% found in a multicentre study of over 2000 stroke patients in Canada, Australia, Israel and Italy (Bladin et al., 2000). As discussed above, most patients in this study were young (≤60years), most had lesions in the cerebral cortex and the percentage with haemorrhagic stroke was high. These are three of the most consistent risk factors for post stroke epilepsy (Bladin et al., 2000; Zelano, 2016). Although four patients in this study had early seizures, only one patient developed late seizures- a haemorrhagic stroke patient. This finding concurs with the report from a study in India from which the most common seizure type in stroke patients was early seizures (Packiaseeli et al., 2017). As that one patient had more than one unprovoked seizure in over 24 hour interval, by the definition of the International League Against Epilepsy, this patient has developed post-stroke epilepsy (Fisher et al., 2005). Early seizures is a known risk factor for late seizures and late seizures is a known risk factor for post stroke epilepsy (Klein et al., 2018). Also, stroke has been found to be the main cause of late-onset seizure disorder in African adults (Owolabi, Akinyemi, Owolabi, Sani, 40 University of Ghana http://ugspace.ug.edu.gh & Ogunniyi, 2013); a risk that is increased by the aetiology of haemorrhagic stroke (Zelano, 2016). Seizures are known to increase post stroke morbidity and mortality (Bladin et al., 2000), but no death was recorded within the one-month period of this study and most patients had slight improvement in functional status at one-month post stroke. Apart from the occurrence of clinical seizures, many more patients had electrographic seizures suggestive of generalized seizure disorder. These seizures which are generally as a direct result of the cortical insult to the brain when stroke occurs are more common than clinical seizures and likely lead to worse outcomes (Bleck, 2012). These seizures, which are sometimes inappropriately called subclinical seizures (Fisher et al., 2005) have been found to be a sensitive predictor of clinical seizures (Feldwisch-Drentrup et al., 2011). If the occurrence of these seizures is an indicator of the likelihood of patients to manifest clinical seizures (Feldwisch-Drentrup et al., 2011), then at least 46% of the patients in this study were at risk of post stroke seizures (though only 14% of that number had associated clinical seizures). In addition, with information on the presence of electrographic seizure status of their patients, clinicians could readily provide interventions to reduce the risk of post stroke epilepsy and associated worsening of outcomes. Almost all the electrographic seizures recorded in this study were generalized seizures. Of the numerous aberrant electrographic activity categorized in this study, the most common which was diagnostic of seizures was generalized spike and wave activity and for the diagnosis of cerebral dysfunction, diffuse slowing was the most characteristic. These findings have some similarities to the findings of Siddiqui et al. (2008) in their observational study of EEG recordings in post stroke seizures; though their commonest seizure activity was focal not generalized. Their study also had some similarities to a study of seizure semiotics in Nigeria, which found the most common seizure associated electroencephalographic brain activity to be focal epileptiform discharge followed by 41 University of Ghana http://ugspace.ug.edu.gh focal slowing (L. Owolabi et al., 2013). These findings of aberrant electroencephalography activities are explained by the impairment of neuronal metabolism, neuronal death and mechanisms of plasticity that occur after stroke (Brain Science International [BSI], 2017; Tanaka & Ihara, 2017). 5.3. Predictors of post stroke functional outcome at one-month in 30 stroke patients Despite the good correlations between functional status measures at recruitment and their corresponding one-month measures, only the modified National Institute of Health Stroke Scale turned out to be a good predictor of all the three outcome measures used, and the Barthel Index score at recruitment was able to predict the Barthel Index score at one-month. The ability of NIHSS (or its modified version (mNIHSS) to predict different measures of post stroke functional outcome has been proved over and over by many studies, including a study on post stroke cognitive impairment in a some Ghanaians (Sarfo, Akassi, Adamu, Obese, & Ovbiagele, 2017) and another on measures of activities of daily living (Kwakkel et al., 2010; Veerbeek et al., 2011). There was no significant correlation between any of the QEEG indices studied (DAR, RAP and pdBSI) and other measures of functional status at recruitment or their outcome scores at one-month when measured for all stroke types; none of the indices predicted outcome measures of activities of daily living (BI), death and disability status (MRS) or neurological deficit (mNIHSS). When assessed by stroke type, only RAP significantly predicted mNIHSS score at one-month in ischaemic strokes. Since the mechanisms of damage and repair in the two stroke types differ, the variability in brain activity according to physiological mechanisms in each stroke type may have contributed to the poor correlations of QEEG indices with functional outcomes in this study. Perhaps an analysis with larger samples by stroke type may yield significant QEEG predictors of functional outcome for each stroke type. 42 University of Ghana http://ugspace.ug.edu.gh For this study, EEGs were recorded with scalp-electrode impedance between 20-50kΩ; as opposed to various studies which have found significant associations between the three QEEG indices and functional status measured at admission or at different stages of recovery (Finnigan & van Putten, 2013). The researchers for those studies maintained impedance at levels below 5kΩ or not more than 15kΩ during recording (Finnigan et al., 2007; Sheorajpanday, Nagels, Weeren, & De Deyn, 2011; Sheorajpanday et al., 2009; Sheorajpanday, Nagels, Weeren, van Putten, et al., 2011; Xin, Chang, Gao, & Shi, 2017). Though measures were taken during EEG pre-processing in this study to filter out noise as well as rejecting bad electrodes from the analysis and Ferree, Luu, Russell, & Tucker (2001) concluded that the use of high input impedance-amplifiers and digital-filters for EEG processing eliminates the impact of scalp impedance on EEG recordings, ensuring EEGs are recorded with impedance below 5kΩ may improve study findings. 43 University of Ghana http://ugspace.ug.edu.gh 6.0 CONCLUSION From a discussion above it can be concluded that:  There was a high incidence of electrographic seizure in stroke patients of Korle Bu Teaching Hospital than what could be clinically diagnosed.  The QEEG indices DAR, RAP and pdBSI did not significantly predict post-stroke functional outcomes measures (MRS, mNIHSS or BI) at one-month across all stroke types.  Relative alpha power (RAP) measured at recruitment was the only QEEG index useful in the prediction of neurological deficit (mNIHSS) in ischaemic stroke at one-month post stroke  The mNIHSS measured at recruitment was the most significant predictor of all functional outcome measures (MRS, mNIHSS and BI) assessed at one-month post stroke. STUDY LIMITATIONS A longitudinal study of electroencephalography characteristics of stroke patients during hospitalization and at recovery could not be carried out due to high cost of EEG and time limitation. RECOMMENDATIONS  Electroencephalography should be routinely done for all stroke patients as a surveillance measure for seizure prediction. Perhaps it may help improve the vigilance in clinical seizure detection.  Electroencephalography recordings for studies on QEEG predictors of functional outcome in stroke patients should be recorded at scalp-electrode impedances below 5kΩ.  Prospective studies measuring QEEG indices predictive of stroke at admission and subsequent recovery months should be done to monitor neuronal recovery after stroke. 44 University of Ghana http://ugspace.ug.edu.gh  Further studies should be done to determine respective QEEG indices predictive of functional outcome for ischaemic and haemorrhagic stroke using a large sample size. 45 University of Ghana http://ugspace.ug.edu.gh REFERENCES Agius Anastasi, A., Falzon, O., Camilleri, K., Vella, M., & Muscat, R. (2017). Brain symmetry index in healthy and stroke patients for assessment and prognosis. Stroke Research and Treatment, 2017, 8276136. https://doi.org/10.1155/2017/8276136 Barrett, K. E., Brooks, H. L., Boitano, S., & Barman, S. M. (2010). Ganong’s review of medical physiology (23rd Editi). The McGraw-Hill Companies, Inc. Bladin, C. F., Alexandrov, A. V, Bellavance, A., Bornstein, N., Chambers, B., Coté, R., … Norris, J. W. (2000). Seizures after stroke: a prospective multicenter study. Archives of Neurology, 57(11), 1617–1622. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11074794 Bleck, T. P. (2012). Seven questions about stroke and epilepsy. Epilepsy Currents, 12(6), 225– 228. https://doi.org/10.5698/1535-7511-12.6.225 Boro, A. D. (2016). Focal EEG Waveform Abnormalities. Retrieved from http://emedicine.medscape.com/article/1139025-overview#a2 Brain Science International. (2017). Reading the Brain Science International QEEG Report. Budzynski, T. H., Budzynski, H. K., Evans, J. R., Abarbanel, A., & Thatcher, R. W. (2009). Chapter 11 – EEG Evaluation of traumatic brain injury and EEG biofeedback treatment. In Introduction to Quantitative EEG and Neurofeedback (pp. 269–294). https://doi.org/10.1016/B978-0-12-374534-7.00011-3 Burneo, J. G., Fang, J., & Saposnik, G. (2010). Impact of seizures on morbidity and mortality after stroke: A Canadian multi-centre cohort study. European Journal of Neurology, 17(1), 46 University of Ghana http://ugspace.ug.edu.gh 52–58. https://doi.org/10.1111/j.1468-1331.2009.02739.x Cincura, C., Pontes-Neto, O. M., Neville, I. S., Mendes, H. F., Menezes, D. F., Mariano, D. C., … Oliveira-Filho, J. (2009). Validation of the National Institutes of Health Stroke Scale, modified Rankin Scale and Barthel Index in Brazil: The role of cultural adaptation and structured interviewing. Cerebrovascular Diseases, 27(2), 119–122. https://doi.org/10.1159/000177918 Claassen, J., Mayer, S. A., Kowalski, R. G., Emerson, R. G., & Hirsch, L. J. (2004). Detection of electrographic seizures with continuous EEG monitoring in critically ill patients. Neurology, 62(10), 1743 LP – 1748. Retrieved from http://www.neurology.org/content/62/10/1743.abstract Cohen, E. J., Quarta, E., Bravi, R., Granato, A., & Minciacchi, D. (2017). Neural plasticity and network remodeling: From concepts to pathology. Neuroscience, 344, 326–345. https://doi.org/10.1016/j.neuroscience.2016.12.048 Costanzo, L. S. (2011). Physiology. (C. Taylor, Ed.) (5th Editio). Richmond, Virginiad: Lippincott Williams & Wilkins. de Haan, R., Limburg, M., Bossuyt, P., van der Meulen, J., & Aaronson, N. (1995). The clinical meaning of Rankin “handicap” grades after stroke. Stroke; a Journal of Cerebral Circulation, 26(11), 2027–2030. https://doi.org/10.1161/01.STR.26.11.2027 Delorme, A., & Makeig, S. (2004). EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), 9–21. https://doi.org/10.1016/j.jneumeth.2003.10.009 47 University of Ghana http://ugspace.ug.edu.gh Feigin, V. L., Norrving, B., & Mensah, G. A. (2017). Global Burden of Stroke. Circulation Research (Sixth Edit, Vol. 120). https://doi.org/10.1161/CIRCRESAHA.116.308413 Feigin, V. L., Roth, G. A., Naghavi, M., Parmar, P., Krishnamurthi, R., Chugh, S., … Forouzanfar, M. H. (2016). Global burden of stroke and risk factors in 188 countries, during 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet Neurology, 15(9), 913–924. https://doi.org/10.1016/S1474-4422(16)30073-4 Feinberg, I., Koresko, R. L., & Heller, N. (1967). EEG SLEEP PATTERNS AS A FUNCTION OF NORMAL AND PATHOLOGICAL AGING IN MAN. J. Psychiaf. Res, 5, 107–144. Feldwisch-Drentrup, H., Ihle, M., le van Quyen, M., Teixeira, C., Dourado, A., Timmer, J., … Schelter, B. (2011). Anticipating the unobserved: Prediction of subclinical seizures. Epilepsy and Behavior, 22(SUPPL. 1). https://doi.org/10.1016/j.yebeh.2011.08.023 Ferree, T. C., Luu, P., Russell, G. S., & Tucker, D. M. (2001). Scalp electrode impedance, infection risk, and EEG data quality. Clinical Neurophysiology : Official Journal of the International Federation of Clinical Neurophysiology, 112(3), 536–544. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11222977 Finnigan, S. P., Rose, S. E., & Chalk, J. B. (2006). Rapid EEG changes indicate reperfusion after tissue plasminogen activator injection in acute ischaemic stroke. Clinical Neurophysiology, 117(10), 2338–2339. https://doi.org/10.1016/j.clinph.2006.06.718 Finnigan, S. P., Walsh, M., Rose, S. E., & Chalk, J. B. (2007). Quantitative EEG indices of sub- acute ischaemic stroke correlate with clinical outcomes. Clinical Neurophysiology, 118(11), 2525–2532. https://doi.org/10.1016/j.clinph.2007.07.021 48 University of Ghana http://ugspace.ug.edu.gh Finnigan, S., & van Putten, M. J. A. M. (2013). EEG in ischaemic stroke: Quantitative EEG can uniquely inform (sub-)acute prognoses and clinical management. Clinical Neurophysiology, 124(1), 10–19. https://doi.org/10.1016/j.clinph.2012.07.003 Fisher, R. S., Van Emde Boas, W., Blume, W., Elger, C., Genton, P., Lee, P., & Engel, J. (2005). Epileptic seizures and epilepsy: Definitions proposed by the International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia, 46(4), 470– 472. https://doi.org/10.1111/j.0013-9580.2005.66104.x Govan, L., Langhorne, P., & Weir, C. J. (2009). Categorizing stroke prognosis using different stroke scales. Stroke, 40(10), 3396–3399. https://doi.org/10.1161/STROKEAHA.109.557645 Grays, B., Wettengel, M., Ouyang, B., Bleck, T., Ovalle, A. B., & Cutting, S. (2016). Predictors of functional outcomes and use of antiepileptic drugs in patients with primary acute intracerebral hemorrhage. Neurology. Retrieved from http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L7225231 3%0Ahttp://findit.library.jhu.edu/resolve?sid=EMBASE&issn=00283878&id=doi:&atitle= Predictors+of+functional+outcomes+and+use+of+antiepileptic+drugs+in+patients+with+pr imary+ac Guérit, J. M., Fischer, C., Facco, E., Tinuper, P., Murri, L., Ronne-Engström, E., & Nuwer, M. (1999). Standards of clinical practice of EEG and EPs in comatose and other unresponsive states. The International Federation of Clinical Neurophysiology. Electroencephalography and Clinical Neurophysiology. Supplement, 52, 117–131. Guyton, A. C. (Department of P. and B. of M. M. C., & Hall, J. E. (Department of P. and B. of 49 University of Ghana http://ugspace.ug.edu.gh M. M. C. (2006). Textbook of medical physiology. (W. Schmitt & R. Gruliow, Eds.) (11th Editi). Philadelphia, Pennsylvania: Elsevier Inc. Hsieh, C. Y., Wu, D. P., & Sung, S. F. (2017). Trends in vascular risk factors, stroke performance measures, and outcomes in patients with first-ever ischemic stroke in Taiwan between 2000 and 2012. Journal of the Neurological Sciences. https://doi.org/10.1016/j.jns.2017.05.002 Kaiser, D. (2007). What is quantitative EEG? Journal of Neurotherapy, 10(4), 1–37. https://doi.org/10.1300/J184v10n04 Kanna, S., & Heng, J. (2009). Quantitative EEG parameters for monitoring and biofeedback during rehabilitation after stroke. IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM, (November 2016), 1689–1694. https://doi.org/10.1109/AIM.2009.5229832 Kengne, A. P., & Mayosi, B. M. (2018, April 1). Modifiable stroke risk factors in Africa: lessons from SIREN. The Lancet Global Health. Elsevier. https://doi.org/10.1016/S2214- 109X(18)30030-5 Klein, P., Dingledine, R., Aronica, E., Bernard, C., Blümcke, I., Boison, D., … Löscher, W. (2018, January). Commonalities in epileptogenic processes from different acute brain insults: Do they translate? Epilepsia. https://doi.org/10.1111/epi.13965 Koren, J. P., Herta, J., Pirker, S., Fürbass, F., Hartmann, M., Kluge, T., & Baumgartner, C. (2016). Rhythmic and periodic EEG patterns of “ictal-interictal uncertainty” in critically ill neurological patients. Clinical Neurophysiology. 50 University of Ghana http://ugspace.ug.edu.gh https://doi.org/10.1016/j.clinph.2015.09.135 Koubeissi, M. (2015, January 4). Seizures worsen stroke outcome: New evidence from a large sample. Epilepsy Currents. https://doi.org/10.5698/1535-7597-15.1.30 Kwah, L. K., & Diong, J. (2014). National Institutes of Health Stroke Scale (NIHSS). Journal of Physiotherapy, 60(1), 61. https://doi.org/10.1016/j.jphys.2013.12.012 Kwakkel, G., Veerbeek, J. M., van Wegen, E. E. H., Nijland, R., Harmeling-van der Wel, B. C., & Dippel, D. W. J. (2010). Predictive value of the NIHSS for ADL outcome after ischemic hemispheric stroke: Does timing of early assessment matter? Journal of the Neurological Sciences, 294(1–2), 57–61. https://doi.org/10.1016/j.jns.2010.04.004 Lahti, A. M., Saloheimo, P., Huhtakangas, J., Salminen, H., Juvela, S., Bode, M. K., … Tetri, S. (2017). Poststroke epilepsy in long-term survivors of primary intracerebral hemorrhage. Neurology, 88(23), 2169–2175. https://doi.org/10.1212/WNL.0000000000004009 Lekoubou, A., Bishu, K. G., & Ovbiagele, B. (2018). Health care expenditures among elderly patients with epilepsy in the United States. Epilepsia, 59(7), 1433–1443. https://doi.org/10.1111/epi.14455 Liu, W., Unick, J., Galik, E., & Resnick, B. (2015). Barthel Index of Activities of Daily Living. Nursing Research, 64(2), 88–99. https://doi.org/10.1097/NNR.0000000000000072 Lyden, P. D., Lu, M., Levine, S. R., Brott, T. G., Broderick, J., & Cote, R. (2001). A Modified National Institutes of Health Stroke Scale for Use in Stroke Clinical Trials : Preliminary Reliability and Validity Editorial Comment : The NIH Stroke Scale: Is Simpler Better? Stroke, 32(6), 1310–1317. https://doi.org/10.1161/01.STR.32.6.1310 51 University of Ghana http://ugspace.ug.edu.gh Martin-Schild, S., Albright, K. C., Tanksley, J., Pandav, V., Jones, E. B., Grotta, J. C., & Savitz, S. I. (2011). Zero on the NIHSS does not equal the absence of stroke. Annals of Emergency Medicine, 57(1), 42–45. https://doi.org/10.1016/j.annemergmed.2010.06.564 Mehta, S. L., & Vemuganti, R. (2014). Mechanisms of stroke induced neuronal death: multiple therapeutic opportunities. Adv. Anim. Vet. Sci., 2(8), 438–446. https://doi.org/http://dx.doi.org/10.14737/journal.aavs/2014/2.8.438.446 Meyer, B., & Lyden, P. (2009). The modified National Institutes of Health Stroke Scale: its time has come. International Journal of Stroke, 4(4), 267–273. https://doi.org/10.1111/j.1747- 4949.2009.00294.x.The Miskin, C., Carvalho, K. S., Valencia, I., Legido, A., & Khurana, D. S. (2015). EEG Duration: The Long and the Short of It. Journal of Child Neurology, 30(13), 1767–1769. https://doi.org/10.1177/0883073815579969 Mukuha, C. W. (2017). The Effect of hospital accreditation on quality of health care services: a case study of Aga Khan University hospital in Kenya. Strathmore University. Retrieved from http://su-plus.strathmore.edu/handle/11071/5498 Mullen, T. (2012). NITRC: CleanLine: Tool/Resource Info. Retrieved July 28, 2018, from https://www.nitrc.org/projects/cleanline/ Naidech, A. M. (2011). Intracranial hemorrhage. American Journal of Respiratory and Critical Care Medicine, 184(9), 998–1006. https://doi.org/10.1164/rccm.201103-0475CI Nkusi, A. E., Muneza, S., Nshuti, S., Hakizimana, D., Munyemana, P., Nkeshimana, M., … Amendezo, E. (2017). Original Article: Stroke Burden in Rwanda: A Multicenter Study of 52 University of Ghana http://ugspace.ug.edu.gh Stroke Management and Outcome. World Neurosurgery. https://doi.org/10.1016/j.wneu.2017.06.163 Nudo, R. J. (2013). Recovery after brain injury: mechanisms and principles. Frontiers in Human Neuroscience, 7. https://doi.org/10.3389/fnhum.2013.00887 O’Donnell, M. J., Xavier, D., Liu, L., Zhang, H., Chin, S. L., Rao-Melacini, P., … INTERSTROKE investigators. (2010). Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. The Lancet, 376(9735), 112–123. https://doi.org/10.1016/S0140-6736(10)60834-3 Offner, H., Ihara, M., Schäbitz, W.-R., & Wong, P. T.-H. (2017). Neurochemistry International Stroke and other cerebrovascular diseases. Neurochemistry International, 107, 3–5. https://doi.org/10.1016/j.neuint.2017.04.003 Olejniczak, P. (2006). Neurophysiologic basis of EEG. Journal of Clinical Neurophysiology, 23(3), 186–189. https://doi.org/10.1097/01.wnp.0000220079.61973.6c Owolabi, L., Akinyemi, R., Owolabi, M., Sani, M., & Ogunniyi, A. (2013). Profile of stroke- related late onset epilepsy among Nigerians. Journal of Medicine in the Tropics, 15(1), 29– 32. Retrieved from https://www.ajol.info/index.php/jmt/article/view/88494 Owolabi, M., Sarfo, F., Howard, V. J., Irvin, M. R., Gebregziabher, M., Akinyemi, R., … Howard, G. (2017). Stroke in Indigenous Africans, African Americans, and European Americans: Interplay of Racial and Geographic Factors. Stroke, 48(5), 1169–1175. https://doi.org/10.1161/STROKEAHA.116.015937 Packiaseeli, C. R., Bobby, E., Radha, M., Saravanan, S., Murugan, P., & Anandan, H. (2017). A 53 University of Ghana http://ugspace.ug.edu.gh Prospective Study of Seizures in Patients with stroke. International Journal of Scientific Study, 5(2), 5–8. https://doi.org/10.17354/ijss/2017/224 Pan, J., Konstas, A.-A., Bateman, B., Ortolano, G. A., & Pile-Spellman, J. (2007). Reperfusion injury following cerebral ischemia: pathophysiology, MR imaging, and potential therapies. Neuroradiology, 49(2), 93–102. https://doi.org/10.1007/s00234-006-0183-z Phillips, C., Rugg, M. D., & Friston, K. J. (2002). Anatomically Informed Basis Functions for EEG Source Localization: Combining Functional and Anatomical Constraints. NeuroImage, 16(3), 678–695. https://doi.org/10.1006/nimg.2002.1143 Piersson, A. D., & Gorleku, P. N. (2017). Assessment of availability, accessibility, and affordability of magnetic resonance imaging services in Ghana. Radiography, 23(4), e75– e79. https://doi.org/10.1016/j.radi.2017.06.002 RMS-Maximus Electroencephalograph. (2013). Retrieved July 29, 2018, from http://www.perfectionnepal.com/products/eeg/eeg-rms-maximus-32.pdf Roth, G. A., Johnson, C., Abajobir, A., Abd-Allah, F., Abera, S. F., Abyu, G., … Murray, C. (2017). Global, Regional, and National Burden of Cardiovascular Diseases for 10 Causes, 1990 to 2015. Journal of the American College of Cardiology, 70(1), 1–25. https://doi.org/10.1016/j.jacc.2017.04.052 Saposnik, G., Kapral, M. K., Liu, Y., Hall, R., O’Donnell, M., Raptis, S., … Group, S. O. R. C. (SORCan) W. (2011). IScore: a risk score to predict death early after hospitalization for an acute ischemic stroke. Circulation, 123(United States eISSN-1524-4539 (Electronic) PT- Journal Article PT-Multicenter Study PT-Research Support, Non-U.S. Gov’t PT-Validation 54 University of Ghana http://ugspace.ug.edu.gh Studies LG-English DC-20110223 OVID MEDLINE UP 20151216), 739–749. https://doi.org/http://dx.doi.org/10.1161/CIRCULATIONAHA.110.983353 Sarfo, F. S., Akassi, J., Adamu, S., Obese, V., & Ovbiagele, B. (2017). Burden and Predictors of Poststroke Cognitive Impairment in a Sample of Ghanaian Stroke Survivors. Journal of Stroke and Cerebrovascular Diseases, 26(11), 2553–2562. https://doi.org/10.1016/j.jstrokecerebrovasdis.2017.05.041 Sergeant, E. (2018). EpiTools. Retrieved January 25, 2018, from http://epitools.ausvet.com.au/content.php?page=cohortSS&P1=0.05&RR=10&Conf=0.95& Power=0.8 Sheorajpanday, R. V. A., Nagels, G., Weeren, A. J. T. M., & De Deyn, P. P. (2011). Quantitative EEG in ischemic stroke: Correlation with infarct volume and functional status in posterior circulation and lacunar syndromes. Clinical Neurophysiology, 122(5), 884–890. https://doi.org/10.1016/j.clinph.2010.08.020 Sheorajpanday, R. V. A., Nagels, G., Weeren, A. J. T. M., De Surgeloose, D., & De Deyn, P. P. (2010). Additional value of quantitative EEG in acute anterior circulation syndrome of presumed ischemic origin. Clinical Neurophysiology, 121(10), 1719–1725. https://doi.org/10.1016/j.clinph.2009.10.037 Sheorajpanday, R. V. A., Nagels, G., Weeren, A. J. T. M., van Putten, M. J. A. M., & De Deyn, P. P. (2009). Reproducibility and clinical relevance of quantitative EEG parameters in cerebral ischemia: A basic approach. Clinical Neurophysiology, 120(5), 845–855. https://doi.org/10.1016/j.clinph.2009.02.171 55 University of Ghana http://ugspace.ug.edu.gh Sheorajpanday, R. V. A., Nagels, G., Weeren, A. J. T. M., van Putten, M. J. A. M., & De Deyn, P. P. (2011). Quantitative EEG in ischemic stroke: Correlation with functional status after 6 months. Clinical Neurophysiology, 122(5), 874–883. https://doi.org/10.1016/j.clinph.2010.07.028 Sherwood, L. (2010). Human Physiology: From Cells to Systems. (M. Arbogast & L. Oliveira, Eds.) (7th Editio). Belmont, CA: Cossio, Yolanda. Sheth, K. N., Martini, S. R., Moomaw, C. J., Koch, S., Elkind, M. S. V., Sung, G., … Woo, D. (2015). Prophylactic antiepileptic drug use and outcome in the ethnic/racial variations of intracerebral hemorrhage study. Stroke, 46(12), 3532–3535. https://doi.org/10.1161/STROKEAHA.115.010875 Siddiqui, M., Yaqoob, U., Bano, A., Malik, A., Khan, F. S., & Siddiqui, K. (2008). EEG findings in post stroke seizures: An observational study. Pakistan Journal of Medical Sciences, 24(3), 386–389. Retrieved from www.pjms.com.pk Silverman, I. E., Restrepo, L., & Mathews, G. C. (2002). Poststroke seizures. Arch Neurol, 59(2), 195–201. https://doi.org/10.1001/archneur.59.2.195 Sudalaimani, C., Asha, S. A., Parvathy, K., Thomas, T. E., Devanand, P., Sasi, P. M., … Thomas, S. V. (2016). Use of electrographic seizures and interictal epileptiform discharges for improving performance in seizure prediction. In 2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015 (pp. 229–234). https://doi.org/10.1109/RAICS.2015.7488419 Tanaka, T., & Ihara, M. (2017). Post-stroke epilepsy. Neurochemistry International, 107, 219– 56 University of Ghana http://ugspace.ug.edu.gh 228. https://doi.org/10.1016/j.neuint.2017.02.002 Thatcher, R. W., & Lubar, J. F. (2009). History of the scientific standards of QEEG normative databases. In Introduction to Quantitative EEG and Neurofeedback (pp. 29–59). https://doi.org/10.1016/B978-0-12-374534-7.00002-2 Thrift, A. G., Thayabaranathan, T., Howard, G., Howard, V. J., Rothwell, P. M., Feigin, V. L., … Cadilhac, D. A. (2017). Global stroke statistics. International Journal of Stroke. https://doi.org/10.1177/1747493016676285 Van Putten, M. J. A. M., & Tavy, D. L. J. (2004). Continuous quantitative EEG monitoring in hemispheric stroke patients using the brain symmetry index. Stroke, 35(11), 2489–2492. https://doi.org/10.1161/01.STR.0000144649.49861.1d Veerbeek, J. M., Kwakkel, G., Van Wegen, E. E. H., Ket, J. C. F., & Heymans, M. W. (2011, May 1). Early prediction of outcome of activities of daily living after stroke: A systematic review. Stroke. https://doi.org/10.1161/STROKEAHA.110.604090 Wu, J., Srinivasan, R., Burke Quinlan, E., Solodkin, A., Small, S. L., & Cramer, S. C. (2016). Utility of EEG measures of brain function in patients with acute stroke. Journal of Neurophysiology, 115(5), 2399–2405. https://doi.org/10.1152/jn.00978.2015 Xin, X., Chang, J., Gao, Y., & Shi, Y. (2017). Correlation between the Revised Brain Symmetry Index, an EEG Feature Index, and Short-term Prognosis in Acute Ischemic Stroke. Journal of Clinical Neurophysiology, 34(2), 162–167. https://doi.org/10.1097/WNP.0000000000000341 Xin, X., Gao, Y., Zhang, H., Cao, K., & Shi, Y. (2012). Correlation of continuous 57 University of Ghana http://ugspace.ug.edu.gh electroencephalogram with clinical assessment scores in acute stroke patients. Neuroscience Bulletin, 28(5), 611–617. https://doi.org/10.1007/s12264-012-1265-z Zandieh, A., Messé, S. R., Cucchiara, B., Mullen, M. T., & Kasner, S. E. (2016). Prophylactic Use of Antiepileptic Drugs in Patients with Spontaneous Intracerebral Hemorrhage. Journal of Stroke and Cerebrovascular Diseases, 25(9), 2159–2166. https://doi.org/10.1016/j.jstrokecerebrovasdis.2016.05.026 Zelano, J. (2016). Poststroke epilepsy: update and future directions. Therapeutic Advances in Neurological Disorders, 9(5), 424–435. https://doi.org/10.1177/1756285616654423 58 University of Ghana http://ugspace.ug.edu.gh APPENDIX Appendix 1: Information and Consent form Title of research: Association between EEG findings and functional outcomes in stroke patients at the Korle Bu Teaching Hospital Investigators: Ruth Y. Laryea MPhil student, University of Ghana- School of Biomedical and Allied Health Sciences (mobile number: +233 243722940) Rev Dr Charles Antwi-Boasiako Senior lecturer, University of Ghana- School of Biomedical and Allied Health Sciences Dr Albert Akpalu Senior Lecturer, University of Ghana School of Medicine and Dentistry You are being invited to participate in a study of the association between EEG findings and functional outcomes of stroke patients at the Korle Bu Teaching Hospital. When someone suffers from a stroke, the brain of that person sustains some injury which can affect the brain’s ability to do its work which includes coordination of body functions. The injury from stroke can also cause the patient to have seizures which could worsen the health status of the patient 59 University of Ghana http://ugspace.ug.edu.gh and slow down recovery. Sometimes these seizures do not show any outward signs for the doctors to see. The EEG (electroencephalography) is a painless and non-invasive diagnostic test (like the ECG test) that records the electrical activity of the brain, which shows how the brain is functioning. The EEG is a unique test that can show if someone’s brain is having a seizure (even when the doctor cannot see it). The purpose of this study is to use the EEG to measure the type of brain activity (seizures, normal etc.) of those who have had stroke while they are on admission and to see if the recorded brain activity can tell us how well the patients will be able to perform their body functions like walking and talking after one month. If you agree to participate in this study, I will record your age, gender, type of stroke you have then a routine EEG test will be done for you. This test will take not more than 30 minutes. In addition, I will ask you to say and do a few things so I can assess how much the stroke has affected your ability to function. This assessment will be done again one month after you do the EEG at which time I will give you a copy of your EEG report. To protect your privacy, I will separate your name from any data I will collect from you. In place of your name, I will give you a code which cannot be traced back to you. All data collected from you will be stored on a password protected computer which will be kept in a locked room. Participation in this research is purely voluntary. If you choose not to participate in this study, it will not affect the care you receive from this facility; you will still receive the quality of care as any other patient here. 60 University of Ghana http://ugspace.ug.edu.gh Consent and Signature The details of the study titled “Association between EEG findings and functional outcomes in stroke patients at the Korle Bu Teaching Hospital” have been explained to me and I have understood it. I agree to participate in this research, knowing that my participation involves a follow-up visit at one month. Name of Participant………………………………………………………………………….. Name of Proxy……………………………………………………………………………….. (if participant cannot sign) Contact details (telephone)…………………………………………………………………… Signature……………………………………………………………………………………… Date Signed (DD/MM/YYYY)………………………………………………………………. Name of Investigator: Ruth Y. Laryea Signature……………………………………………………………………………………… Date Signed (DD/MM/YYYY)………………………………………………………………. 61 University of Ghana http://ugspace.ug.edu.gh Appendix 2: Study Questionnaire Date:___/___/2018 Demographics Participant ID:______________________ Age (yrs):_______ Sex: Male/Female Date of Birth:_______________(DD/MM/YYYY) Neuroimaging Report Stroke type: Volume of lesion: Cortical location (one or more): a. Haemorrhage I. 0-3 A. Frontal lobe b. Infarct II. 3.1-5.0 B. Temporal lobe III. 5.1-7.0 C. Parietal lobe IV. 7.1-10.0 D. Occipital lobe V. >10.0 E. Brainstem Hemispheric location of lesion: Subcortical structures affected: a) Left b) Right □ Thalamus □ Pituitary gland □ Basal ganglia Functional Assessment MRS at recruitment:_______________ MRS at one-month:________________ 62 University of Ghana http://ugspace.ug.edu.gh mNIHSS at recruitment:______________ mNIHSS at one-month:______________ Barthel Index at recruitment:_________ Barthel Index at one-month:_________ EEG Pattern Report Type of EEG: a) Awake b) Drowsy c) Asleep d) Sedation e) Sleep deprived Activations done: a) None b) Hyperventilation c) Photo stimulation EEG activity 1. Normal activity 2. Epileptiform activity 3. Cerebral dysfunction i Spike and slow waves i. Focal slowing ii Sharp waves ii. Diffuse slowing iii Spikes iii. Others……………… iv Diffuse slowing Have you ever had a seizure before this stroke? (Yes/ No) If yes, how many________ How many seizures has the patient had since stroke occurred? _________ When did they occur? a)Within two weeks b) After two weeks c) Both Are you currently on any anti-epileptic drugs? (Yes/ No) If yes, name them_______________________________________________________________ EEG Quantification pdBSI__________ RAP___________ DAR___________ 63 University of Ghana http://ugspace.ug.edu.gh Appendix 3: Modified Rankin Scale (MRS) MODIFIED RANKIN SCALE Participant ID: Date assessed: 0 No symptoms at all 1 No significant disability despite symptoms; able to carry out all usual duties and activities 2 Slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance 3 Moderate disability; requiring some help, but able to walk without assistance 4 Moderately severe disability; unable to walk without assistance and unable to attend to own bodily needs without assistance 5 Severe disability; bedridden, incontinent and requiring constant nursing care and attention 6 Dead 64 University of Ghana http://ugspace.ug.edu.gh Appendix 4: Modified National Institute of Health Stroke Scale (NIHSS) Modified National Institute of Health Stroke Scale (mNIHSS) Participant Date assessed:_____________________ ID:________________________ Score Instructions Scale definition 1 2 1b. LOC Questions: The patient is 0 Answers both questions correctly. asked the month and his/her age. 1 Answers one question correctly. The answer must be correct - there is no partial credit for being close. Aphasic and stuporous patients who do not comprehend the questions will score 2. Patients unable to speak because of endotracheal intubation, orotracheal trauma, severe dysarthria from any cause, 2 Answers neither question correctly. language barrier, or any other problem not secondary to aphasia are given a 1. It is important that only the initial answer be graded and that the examiner not "help" the patient with verbal or non- verbal cues. 0 Performs both tasks correctly. 65 University of Ghana http://ugspace.ug.edu.gh 1c. LOC Commands: The patient is 1 Performs one task correctly. asked to open and close the eyes and then to grip and release the non-paretic hand. Substitute another one step command if the hands cannot be used. Credit is given if an unequivocal attempt is made but not completed due to weakness. If the patient does not respond to command, the task should be demonstrated to him or 2 Performs neither task correctly. her (pantomime), and the result scored (i.e., follows none, one or two commands). Patients with trauma, amputation, or other physical impediments should be given suitable one-step commands. Only the first attempt is scored 2. Best Gaze: Only horizontal eye Normal. 0 movements will be tested. Voluntary or reflexive (oculocephalic) eye movements Partial gaze palsy; gaze is abnormal in will be scored, but caloric testing is not 1 one or both eyes, but forced deviation or done. If the patient has a conjugate total gaze paresis is not present. 66 University of Ghana http://ugspace.ug.edu.gh deviation of the eyes that can be overcome by voluntary or reflexive activity, the score will be 1. If a patient has an isolated peripheral nerve paresis (CN III, IV or VI), score a 1. Gaze is testable in all aphasic patients. Patients with ocular trauma, bandages, pre- Forced deviation, or total gaze paresis existing blindness, or other disorder of 2 not overcome by the oculocephalic visual acuity or fields should be tested maneuver. with reflexive movements, and a choice made by the investigator. Establishing eye contact and then moving about the patient from side to side will occasionally clarify the presence of a partial gaze palsy 3. Visual: Visual fields (upper and 0 No visual loss. lower quadrants) are tested by 1 Partial hemianopia. confrontation, using finger counting or 2 Complete hemianopia. visual threat, as appropriate. Patients may be encouraged, but if they look at Bilateral hemianopia (blind including the side of the moving fingers 3 cortical blindness). appropriately, this can be scored as normal. If there is unilateral blindness or 67 University of Ghana http://ugspace.ug.edu.gh enucleation, visual fields in the remaining eye are scored. Score 1 only if a clear- cut asymmetry, including quadrantanopia, is found. If patient is blind from any cause, score 3. Double simultaneous stimulation is performed at this point. If there is extinction, patient receives a 1, and the results are used to respond to item 11. 5. Motor Arm: The limb is placed in No drift; limb holds 90 (or 45) 0 the appropriate position: extend the arms degrees for full 10 seconds. (palms down) 90 degrees (if sitting) or 45 Drift; limb holds 90 (or 45) degrees (if supine). Drift is scored if the degrees, but drifts down before full 1 arm falls before 10 seconds. The aphasic 10 seconds; does not hit bed or patient is encouraged using urgency in other support. the voice and pantomime, but not Some effort against gravity; limb noxious stimulation. Each limb is tested cannot get to or maintain (if cued) in turn, beginning with the non-paretic 2 90 (or 45) degrees, drifts down to arm. Only in the case of amputation or bed, but has some effort against joint fusion at the shoulder, the examiner gravity. should record the score as untestable No effort against gravity; limb 3 (UN), and clearly write the explanation falls for this choice. 4 No movement. 68 Right Left University of Ghana http://ugspace.ug.edu.gh UN= Amputation or joint fusion, explain: ________________ No drift; leg holds 30- 6. Motor Leg: The limb is placed in the 0 degree position for full 5 appropriate position: hold the leg at 30 seconds. degrees (always tested supine). Drift is Drift; leg falls by the end of scored if the leg falls before 5 seconds. 1 the 5-second period but does The aphasic patient is encouraged using not hit bed. urgency in the voice and pantomime, but Some effort against not noxious stimulation. Each limb is gravity; leg falls to bed by 5 2 tested in turn, beginning with the non- seconds, but has some effort paretic leg. Only in the case of against gravity. amputation or joint fusion at the hip, the No effort against gravity; leg 3 examiner should record the score as falls to bed immediately. untestable (UN), and clearly write the 4 No movement. explanation for this choice. UN = Amputation or joint fusion, explain: ________________ 8. Sensory: Sensation or grimace to 0 Normal; no sensory loss. pinprick when tested, or withdrawal from 69 Right Left University of Ghana http://ugspace.ug.edu.gh noxious stimulus in the obtunded or aphasic patient. Only sensory loss attributed to stroke is scored as abnormal and the examiner should test as many body areas (arms [not hands], legs, trunk, face) as needed to accurately check for Abnormal; patient feels pinprick hemisensory loss. A score of 2, “severe is less sharp or is dull on the or total sensory loss,” should only be affected side; or there is a loss of given when a severe or total loss of 1 superficial pain with pinprick, but sensation can be clearly demonstrated. patient is aware of being touched. Stuporous and aphasic patients will, Or patient is not aware of being touched therefore, probably score 1 or 0. The in the face, arm, and leg. patient with brainstem stroke who has bilateral loss of sensation is scored 2. If the patient does not respond and is quadriplegic, score 2. Patients in a coma (item 1a=3) are automatically given a 2 on this item. 9. Best Language: A great deal of 0 No aphasia; normal. information about comprehension will be Mild-to-moderate aphasia; some obtained during the preceding sections of obvious loss of fluency or facility of 1 the examination. For this scale item, the comprehension, without significant patient is asked to describe what is limitation on ideas expressed or form of 70 University of Ghana http://ugspace.ug.edu.gh happening in the attached picture, to expression. Reduction of speech and/or name the items on the attached naming comprehension, however, makes sheet and to read from the attached list of conversation about provided materials sentences. Comprehension is judged difficult or impossible. For example, in from responses here, as well as to all of conversation about provided materials, the commands in the preceding general examiner can identify picture or naming neurological exam. If visual loss card content from patient’s response. interferes with the tests, ask the patient to Severe aphasia; all communication is identify objects placed in the hand, through fragmentary expression; great repeat, and produce speech. The need for inference, questioning, and intubated patient should be asked to guessing by the listener. Range of write. The patient in a coma (item 1a=3) 2 information that can be exchanged is will automatically score 3 on this item. limited; listener carries burden of The examiner must choose a score for the communication. Examiner cannot patient with stupor or limited identify materials provided from patient cooperation, but a score of 3 should be response. used only if the patient is mute and Mute, global aphasia; no usable 3 follows no one-step commands. speech or auditory comprehension. 11. Extinction and Inattention 0 No abnormality (formerly Neglect): Sufficient Visual, tactile, auditory, spatial, or 1 information to identify neglect may be personal inattention or extinction to 71 University of Ghana http://ugspace.ug.edu.gh obtained during the prior testing. If bilateral simultaneous stimulation in the patient has a severe visual loss one of the sensory modalities. preventing visual double simultaneous stimulation, and the cutaneous stimuli are normal, the score is normal. If the patient has aphasia but does appear to Profound hemi-inattention or attend to both sides, the score is extinction to more than one modality; 2 normal. The presence of visual spatial does not recognize own hand or orients neglect or anosagnosia may also be to only one side of space. taken as evidence of abnormality. Since the abnormality is scored only if present, the item is never untestable. TOTAL SCORE AT RECRUITMENT (SCORE 1) TOTAL SCORE AT ONE-MONTH (SCORE 2) 72 University of Ghana http://ugspace.ug.edu.gh Appendix 5: Barthel Index BARTHEL INDEX OF ACTIVITIES OF DAILY LIVING ITEM RANKING SCORE 1 SCORE 2 0 = incontinent (or needs to be given enemata) Bowels 1 = occasional accident (once/week) 2 = continent 0 = incontinent, or catheterized and unable to manage Bladder 1 = occasional accident (max. once per 24 hours) 2 = continent (for over 7 days) 0 = needs help with personal care Grooming 1 = independent face/hair/teeth/shaving (implements provided) 0 = dependent Toilet use 1 = needs some help, but can do something alone 2 = independent (on and off, dressing, wiping) 0 = unable Feeding 1 = needs help cutting, spreading butter, etc. 2 = independent (food provided within reach) 0 = unable – no sitting balance Transfer 1 = major help (one or two people, physical), can sit 2 = minor help (verbal or physical) 3 = independent Mobility 0 = immobile 73 University of Ghana http://ugspace.ug.edu.gh 1 = wheelchair independent, including corners, etc. 2 = walks with help of one person (verbal or physical) 3 = independent (but may use any aid, e.g., stick) 0 = dependent Dressing 1 = needs help, but can do about half unaided 2 = independent (including buttons, zips, laces, etc.) Bathing 0 = dependent 1 = independent (or in shower) 0 = unable Stairs 1 = needs help (verbal, physical, carrying aid) 2 = independent up and down Total Score 74 University of Ghana http://ugspace.ug.edu.gh Appendix 6: Copy of Ethical clearance 75