DISSERTATIONS | JUHA ONATSU | INFLAMMATORY, NEUROAXONAL AND IMAGING BIOMARKERS... | No 640
JUHA ONATSU
Inflammatory, Neuroaxonal and Imaging Biomarkers
in Acute Ischemic Stroke
Dissertations in Health Sciences
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND
INFLAMMATORY, NEUROAXONAL AND IMAGING BIOMARKERS
IN ACUTE ISCHEMIC STROKE
Juha Onatsu
INFLAMMATORY, NEUROAXONAL AND IMAGING BIOMARKERS
IN ACUTE ISCHEMIC STROKE
To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in Medistudia auditorium
MS301 of University of Eastern Finland, Kuopio on Friday, October 1st 2021, at 12 o’clock noon
Publications of the University of Eastern Finland Dissertations in Health Sciences
Number 640
Department of Neurology, Institute of Clinical Medicine, School of Medicine Faculty of Health Sciences, University of Eastern Finland
Kuopio 2021
Series Editors
Professor Tomi Laitinen, M.D., Ph.D.
Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences
Professor Tarja Kvist, Ph.D.
Department of Nursing Science Faculty of Health Sciences
Professor Ville Leinonen, M.D., Ph.D.
Institute of Clinical Medicine, Neurosurgery Faculty of Health Sciences
Professor Tarja Malm, Ph.D.
A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences
Lecturer Veli-Pekka Ranta, Ph.D.
School of Pharmacy Faculty of Health Sciences
Distributor:
University of Eastern Finland Kuopio Campus Library
P.O. Box 1627 FI-70211 Kuopio, Finland
www.uef.fi/kirjasto
PunaMusta Oy Joensuu, 2021
ISBN (print): 978-952-61-4293-7 ISBN (PDF): 978-952-61-4294-4
ISSNL: 1798-5706 ISSN (print): 1798-5706
ISSN(PDF): 1798-5714
Author’s address: Department of Clinical Neurology Institute of Clinical Medicine School of Medicine
University of Eastern Finland KUOPIO
FINLAND
Doctoral programme: Doctoral Programme of Clinical Research
Supervisors: Professor Pekka Jäkälä, MD., Ph.D.
Department of Clinical Neurology Institute of Clinical Medicine School of Medicine
University of Eastern Finland KUOPIO
FINLAND
Professor Ritva Vanninen, MD., Ph.D.
Department of Clinical Radiology Institute of Clinical Medicine University of Eastern Finland KUOPIO
FINLAND
Professor Pirjo Mustonen, MD., PhD.
Department of Internal Medicine Middle Finland Central Hospital JYVÄSKYLÄ
FINLAND
Reviewers: Assistant Professor Susanna Melkas, MD., Ph.D.
Department of Neurology, Helsinki University Hospital Clinical Neurosciences, University of Helsinki
HELSINKI FINLAND
Docent Jussi P Posti, MD., Ph.D.
Department of Neurosurgery University of Turku
TURKU FINLAND
Opponent: Docent Tiina Sairanen MD., Ph.D.
Department of Neurology, Helsinki University Hospital Clinical Neurosciences, University of Helsinki
HELSINKI FINLAND
Onatsu, Juha
Inflammatory, Neuroaxonal and Imaging Biomarkers in Acute Ischemic Stroke Kuopio: University of Eastern Finland, Faculty of Health Sciences
Publications of the University of Eastern Finland Dissertations in Health Sciences 640. 2021, 164 p.
ISBN (print): 978-952-61-4293-7 ISSNL: 1798-5706
ISSN: 1798-5706
ISBN (PDF): 978-952-61-4294-4 ISSN (PDF): 1798-5714
ABSTRACT
Stroke is responsible for 10% of all global deaths and is the second leading cause of death and the third leading cause of disability. Ischemic stroke accounts about 80% of all events and an embolism of cardiac origins is the cause of about every fifth of these occurences. Atrial fibrillation is the most common cause of brain embolism, but there are many other structural changes in the heart, that can predispose to the formation of thrombi. Despite a thorough etiological work-up, the reason behind the stroke remains unknown in 25% to 40% of cases - these are called cryptogenic strokes.
Non-contrast computed tomography (NCCT) is the basic imaging method applied in patients with acute stroke since it can rule out or confirm the presence of an intracranial hemorrhage and furthermore, the early signs of ischemia or hints of large vessel occlusion may be seen.
Contrast enhanced CT can be used to diagnose carotid artery stenosis and dissection and recently also cardiac sources of emboli. Until now,
echocardiography has been the golden standard to permit the identification of a possible cardiac source of the embolism. The prediction of outcome after ischemic stroke is mainly based on the patient’s age, stroke severity and premorbid
disability. However, blood biomarkers reflecting different pathophysiological processes in acute stroke may supplement the performance of these models.
We investigated if a composite examination of the heart, aorta and
cervicocranial arteries with computed tomography (CACC-CT) could improve the etiologic evaluation in patients with a suspected stroke of cardiogenic origin. We also evaluated if inflammatory mechanisms were involved in stroke etiology and
outcome. Additionally we studied if neuro-axonal proteins in blood are related to the outcome and diagnosis of stroke or TIA.
We observed that CACC-CT and transthoracic or transesophageal
echocardiography (TTE/TEE) together were more sensitive than TTE/TEE alone for detecting cardiac or aortic high-risk findings for embolism. TTE/TEE was inferior to CACC-CT in visualizing myocardial infarction with left ventricular aneurysm. In contrast, CACC-CT was not suitable for detecting small left atrial thrombi, a patent foramen ovale or for measuring the ejection fraction of the left ventricle.
Soluble urokinase-type plasminogen activator receptor (suPAR) concentrations were found to be higher in those patients who suffered a stroke/transient
ischemic attack due to large artery atherosclerosis as compared to small vessel disease. Additionally, an elevated plasma suPAR concentration was associated with all-cause mortality during the five year follow-up.
Serum levels of neurofilament light chain and serum Tau concentrations were higher in patients with acute ischemic stroke (AIS) than in those with TIA and the serum levels correlated with the stroke volume and functional outcome after three months.
Our results suggest that, CACC-CT can add value to TTE/TEE in the etiological diagnostics of cryptogenic stroke and furthermore that neuroaxonal biomarkers may assist in both stroke diagnostics and in the prognostification of the outcome after a stroke.
National Library of Medicine Classification: : WG 141.5.E2, WL 356, WN 206 Medical Subject Headings: Stroke; Brain Ischemia; Biomarkers; Tomography, X- Ray Computed/methods; Echocardiography, Transesophageal; Receptors, Urokinase Plasminogen Activator; Prognosis; Risk Factors
Onatsu, Juha
Kirjan nimi. Tulehdukselliset, neuroaksonaaliset ja kuvantamisen biomarkkerit akuutissa aivoverenkiertohäiriössä
Kuopio: Itä-Suomen yliopisto, Terveystieteiden tiedekunta Publications of the University of Eastern Finland
Dissertations in Health Sciences 640. 2021, 164 s.
ISBN (print): 978-952-61-4293-7 ISSNL: 1798-5706
ISSN: 1798-5706
ISBN (PDF) : 978-952-61-4294-4 ISSN (PDF): 1798-5714
TIIVISTELMÄ
Aivohaveri aiheuttaa maailmanlaajuisesti noin 10% kaikista kuolemista, ollen toiseksi suurin kuolinsyy ja kolmanneksi suurin rajoittuneen toimintakyvyn aiheuttaja. Iskeemiset aivoverenkiertohäiriöt (=aivoinfarkti) kattavat noin 80%
kaikista aivohavereista ja noin joka viides tapauksista on sydänperäisisen
embolian aiheuttama. Eteisvärinä on yleisin syy aivoemboliaan, mutta sydämessä on monia muita rakenteellisia muutoksia, jotka voivat myös altistaa trombien muodostumiselle. Kattavista etiologisista tutkimuksista huolimatta aivoinfarktin syy jää tuntemattomaksi 30-40% tapauksista, ja näitä kutsutaan kryptogeenisiksi eli salasyntyisiksi aivoinfarkteiksi.
Varjoaineeton pään tietokonetomografia (NCCT) on aivohaveripotilaan
peruskuvantamismenetelmä ja sillä voidaan poisulkea tai vahvistaa kallonsisäinen verenvuoto. Toistaalta myös varhaiset aivoiskemian merkit ja tiivis keskimmäinen aivovaltimo voivat ovat nähtävissä myös CT-kuvantamisella.
Varjoainetehosteista CT-kuvausta voidaan käyttää myös kaulavaltimon
ahtauman ja dissekaation sekä embolialle altistavien sydämen rakennemuutosten diagnosointiin. Kuitenkin ekkokardiografiaa on tähän asti pidetty sydämen
rakennemuutosten tutkimisen kultaisena standardina. Iskeemisen
aivoverenkiertohäiriön jälkeinen toipumisennuste perustuu pääasiassa potilaiden ikään, aivoinfarktin vaikeusasteeseen ja sairastumista edeltäneeseen
toimintakykyyn. Verestä mitattavat eriaiset biomarkkerit, jotka heijastavat akuutin
aivoinfarktin erilaisia patofysiologisia prosesseja, voivat lisätä näiden ennustemallien osuvuutta.
Tutkimme, parantaako sydämen, aortan- ja kaulavaltimoiden yhdistetty CT- tutkimus (CACC-CT) aivoinfarktin etiologian diagnostikkaa potilailla, joilla infarktin epäillään olevan kardiogeenista alkuperää. Arvioimme myös aivoinfarktin
etiologiaan ja toipumiseen liittyviä tulehdusmekanismeja. Lisäksi tutkimme, kykenevätkö veren neuroaksonaaliset proteiinit erottelemaan aivoinfarktin ja TIA:
n ja voiko niiden avulla ennustaa aivoinfarktista toipumista.
Huomasimme, että CACC-CT: n ja transthorakaalisen/ transesophageaalisen ekkokardiografian (TTE / TEE) yhdistetty käyttö oli herkempi osoittamaan sydän- tai aortaperäisiä korkeanriskin rakennemuutoksia kuin pelkästään TTE / TEE.
Sydämen ultraäänitutkimus oli huonompi kuin CACC-CT osoittamaan mm.
sydäninfarktiarpea, vasemman kammion aneurysmaa tai aortasta lähtöisin olevia embolioita. Toisaalta CACC-CT oli heikompi havaitsemaan pieniä vasemman eteisen trombeja, piilevää eteisväliseunän aukkoa eikä soveltunut vasemman kammion ejektiofraktion mittaamiseen.
Liukoisen urokinaasityyppisen plasminogeenia aktivoivan reseptorin (suPAR) pitoisuuksien havaittiin olevan korkeammat potilailla, joiden aivoinfarkti oli suurten suonten valtimokovettuman aiheuttama verrattuna pienten suonten taudista kärsiviin potilaisiin. Lisäksi kohonnut suPAR-pitoisuus ennusti kuolleisuutta viiden vuoden seurantajakson aikana. Seerumin
neurofilamenttiproteiini- ja seerumin Tau-proteiinipitoisuudet olivat korkeammat aivoinfarkti- kuin TIA-potilailla, ja niiden seerumitasot korreloivat aivoinfarktin aiheuttamaan aivoiskemian tilavuuteen ja toiminnalliseen tulokseen kolmen kuukauden kuluttua.
Tuloksemme viittaavat siihen, että CACC-CT lisää diagnostista tarkkuutta TTE / TEE: n lisäksi salasyntyisissä aivoinfarkteissa ja tulehdukselliset sekä
neuroaksonaaliset biomarkkerit voivat auttaa aivoinfarktin diagnosoimisessa ja ennustearvioinnissa.
Yleinen suomalainen ontologia: : aivoverenkiertohäiriöt; markkerit; riskitekijät;
tietokonetomografia
Labor Improbus Omnia Vincit
-The motto above the door of my high school.
ACKNOWLEDGEMENTS
This study is a part of the EMBODECT project and it was carried out in the Department of Clinical Neurology in Kuopio University Hospital during the years 2014-2021 in collaboration with Departments of Clinical Radiology, Clinical
Cardiology and Clinical Chemistry. I have been privileged to be a part of this study group.
I want to thank my principal supervisor, colleague and a friend, Professor Pekka Jäkälä, for introducing me to the project and providing me the opportunity to perform this thesis. He also guided and mentored me a lot through the entire study project. Your continuous support has encouraged me to finish this thesis.
Professor Ritva Vanninen was another supervisor of this thesis. She has an extensive and deep knowledge in the field of science and radiology. I´m
impressed by her brilliant vision on how best to report the study results and her clinical and statistical skills. She provided me a wise guidance along this journey.
I want to thank my third supervisor, Professor Pirjo Mustonen. Pirjo is a clever, enthusiastic scientist. Her guidance is warm, friendly and cheerful. She has always new ideas, how to further analyze our study cohort.
I want to express my deepest gratitude to my `fourth, informal` supervisor MD, PhD Mikko Taina for all your help and many, many efforts during this study period.
I must admit, that without you and your inspiration, I have never been started this project or finished it. I have got all kind of help and assistance from you whenever needed. I am so greatfull from that.
I want to thank Professor Marja Hedman, Adjunct Professor Petri Sipola being key players in starting and analyzing this EMBODETECT study cohort. With our
assistance and efforts our knowledge about cardioembolic stroke has raised markedly.
I thank my all fellow-colleagues and coworkers for their cooperation and help during this study process. Miikus Korhonen, Antti Muuronen, Johannes Parkkonen, Riikka Tyrväinen and Otso Arponen assisted me a lot during this study.
My special thanks goes to my laboratory colleagues in Sweden: Professor Henrik Zettenberg, Proferssor Kaj Blennow and MD, PhD Kina Höglund and in Finland Professor Kari Pulkki and MD, PhD Sanna-Kaisa Herukka for their help in analyzing blood samples and reporting the study results.
I want to express my gratitude to my pre-examiners, Assistant Professor Susanna Melkas and Docent Jussi P Posti for their incisive comments about the thesis. Your hints and proposals have improved the final version of the thesis markedly.
I want to thank biostatistician Tuomas Selander from the University of Eastern Finland for his great help in resolving the statistical puzzles and Ewen MacDonald for the linguistic revision of this thesis.
I want to express my heartful thanks to my late parents Pirkko and Lauri for proving me a good and solid ground to life and education. I owe a debt gratitude to my three loved sisters; Saila, Elena and Päivi. Yours and yours spouses: Jude´s ,Jussi´s and Antti´s support has been important during these years. Additionally, I want to send my warmest thanks to my mother-in-law Marja Liisa Rahikainen and sister-in law Tarja Rahikainen for all kind of help during these busy years.
I am so blessed to have so many good friends and colleagues. I dedicate warm thanks to my colleagues from Department of Neurology, Neurosurgery and Radiology for been so supportive during these years. `Los Angeles Brains`and
`Honolulu Eddies`,our learning trips abroad have been memorable. I want to thank all my long time friends for being there around with me despite of the lack of time to share with you,; Jarppa ( the rubber), Jukka, Pete, Jussi, Jore, Jussi N, Ari, Jari K, Eske, Anne, Jontte, Timo,Maarit, Ipa, Maarit P, Tapsa, Juse,Vesku,
Johanna,Hilve and many, many others. I hope that we´ll have more time to spent together in the near future.
The uttermost gratitude belongs to my wonderful family for encouraging me through the years and being patient when I have been stucked up with this project. My children Ina, Kalle and Vera; I`m so proud to have such a lovely kids.
Lastly, my wife Sari, I do nothave enough words to describe the indebtedness I owe to you. You know that for sure.
This work was financially supported by grants from the Finnish Foundation Cultural Foundation, State Research Funding from the Hospital District of Kuopio University Hospital and Central Finland Central Hospital (VTR) and the Finnish Medical Foundation.
Kuopio on 5th of Septemper 2021
Juha Onatsu
LIST OF ORIGINAL PUBLICATIONS
This dissertation is based on the following original publications:
I Sipola P, Hedman M, Onatsu J, Turpeinen A, Halinen M, Jäkälä P, Vanninen R.
Computed tomography and echocardiography together reveal more high-risk findings than echocardiography alone in the diagnostics of stroke etiology.
Cerebrovasc Dis. 2013; 35 (6): 521-30
II Onatsu J, Taina M, Mustonen P, Hedman M, Muuronen A, Arponen O, Korhonen M, Jäkälä P, Vanninen R, Pulkki K. Soluble urokinase-type plasminogen activator receptor predicts all-cause 5-year mortality in ischaemic stroke and TIA. In Vivo. 2017 May-Jun; 31(3): 381-386
III Onatsu J, Vanninen R, Jäkälä P, Mustonen P, Pulkki K, Korhonen M, Hedman M, Zetterberg H, Blennow K, Höglund K, Herukka S-K, Taina M. Serum
neurofilament light chain concentration correlates with infarct volume but not prognosis in acute ischaemic stroke. J Stroke Cerebrovasc Dis. 2019 Aug; 28 (8): 2242-2249
IV Onatsu J, Vanninen R, Jäkälä P, Mustonen P, Pulkki K, Korhonen M, Hedman M, Zetterberg H, Blennow K, Höglund K, Herukka S-K, Taina M. Tau, S100B and NSE as blood biomarkers in acute cerebrovascular events. In Vivo. 2020 Sep- Oct; 34 (5): 2577-2586
The publications were adapted with the permission of the copyright owners.
CONTENTS
ABSTRACT ... 7
TIIVISTELMÄ ... 9
ACKNOWLEDGEMENTS ... 13
1 INTRODUCTION ... 27
2 REVIEW OF THE LITERATURE ... 29
2.1 Burden of the disease and epidemiology ... 29
2.2 Definition of ischemic stroke and tia ... 30
2.2.1 Risk factors for stroke ... 30
2.2.2 Genetic factors for stroke ... 31
2.3 Diagnosis and evaluation of stroke ... 32
2.4 Classification of ischemic stroke ... 34
2.4.1 Large vessel disease ... 36
2.4.2 Cardioembolism ... 37
2.4.3 Small vessel disease ... 38
2.4.4 Ischemic stroke due to other determined causes ... 39
2.4.5 Stroke of undetermined etiology ... 39
2.4.6 Embolic stroke of undetermined cause ... 39
2.5 Cardioembolic stroke ... 40
2.5.1 Pathophysiological mechanisms of cardioembolic stroke ... 40
2.5.2 Risk factors for cardioembolic stroke ... 41
2.5.2.1Atrial fibrillation ... 42
2.5.2.2Systolic heart failure ... 43
2.5.2.3Recent myocardial infarction ... 43
2.5.2.4Patent foramen ovale (PFO) ... 43
2.5.2.5Prosthetic heart valves ... 44
2.5.2.6Aortic arch atheroma ... 44
2.5.2.7Infective endocarditis... 45
2.5.2.8Other causes ... 45
2.5.3 Cryptogenic stroke and ESUS concept ... 45
2.5.3.1Potential mechanisms behind ESUS ... 47
2.5.3.2Paroxysmal atrial fibrillation ... 48
2.5.3.3Left atrial cardiopathy ... 48
2.5.3.4Occult structural lesion ... 49
2.6 Imaging biomarkers ... 49
2.6.1 Brain and vessel imaging ... 49 2.6.1.1Non-contrast head computed tomography ... 49 2.6.1.2Computed tomography angiography ... 51 2.6.1.3Computed tomography perfusion ... 51 2.6.1.4Magnetic resonance imaging ... 53 2.6.1.5MR-angiography and perfusion MRI ... 54 2.6.2 Cardiac imaging ... 55 2.6.2.1Cardiac ultrasound ... 57 2.6.2.2Cardiac computed tomography ... 57 2.6.2.3Cardiac magnetic resonance imaging ... 59 2.7 Inflammatory and neuroaxonal biomarkers ... 60 2.7.1 General considerations... 60 2.7.2 Inflammatory blood biomarkers in ischemic stroke ... 60 2.7.3 Blood biomarkers for the diagnosis of ischemic stroke ... 62 2.7.4 Etiological blood biomarkers of stroke ... 64 2.7.5 Prognostical blood biomarkers of stroke ... 65 3 AIMS OF THE STUDY ... 67 4 COMPUTED TOMOGRAPHY AND ECHOCARDIOGRAPHY TOGETHER REVEAL
MORE HIGH-RISK FINDINGS THAN ECHOCARDIOGRAPHY ALONE IN THE STROKE ETIOLOGY... 69 4.1
A
bstract ... 69 4.1.1 Backround ... 69 4.1.2 Methods ... 69 4.1.3 Results ... 69 4.1.4 Conclusion ... 70 4.2 Introduction ... 70 4.3 Materials and methods ... 71 4.3.1 Study design and population ... 71 4.3.2 Imaging protocol and image analysis ... 73 4.3.3 Statistical analysis ... 77 4.4 Results ... 77 4.4.1 Patients characteristics ... 77 4.4.2 High-risk findings ... 78 4.4.3 Minor risk findings ... 82 4.4.4 Carotid artery findings ... 82 4.5 Discussion ... 82 5 SOLUBLE UROKINASE-TYPE PLASMINOGEN ACTIVATOR RECEPTOR PREDICTSALL-CAUSE 5-YEAR MORTALITY IN ISCHEMIC STROKE AND TIA ... 87
5.1 Abstract... 87 5.1.1 Aim ... 87 5.1.2 Methods ... 87 5.1.3 Results ... 87 5.1.4 Conclusion ... 87 5.2 Introduction ... 88 5.3 Materials and methods ... 88 5.3.1 Statistical evaluation ... 89 5.4 Results ... 91 5.5 Discussion ... 94 6 SERUM NEUROFILAMENT LIGHT CHAIN CONCENTRATION CORRELATES
WITH INFARCT VOLUME BUT NOT PROGNOSIS IN ACUTE STROKE ... 97 6.1 Abstract... 97 6.1.1 Background and purpose ... 97 6.1.2 Methods ... 97 6.1.3 Results ... 97 6.1.4 Conclusions ... 98 6.2 Introduction ... 98 6.3
M
aterial and methods ... 99 6.3.1 Measurement of neurofilament in serum ... 99 6.3.2 Statistical analysis ... 100 6.4 Results ... 101 6.5 Discussion ... 107 7 TAU, S100B AND NSE AS BLOOD BIOMARKERS IN ACUTECEREBROVASCULAR EVENTS ... 111 7.1 Abstract... 111 7.1.1 Backgound ... 111 7.1.2 Methods ... 111 7.1.3 Results ... 111 7.1.4 Conclusion ... 112 7.2
I
ntroduction ... 112 7.3 Patients and methods ... 113 7.3.1 Study design and patients ... 113 7.3.2 Measurement of biomarkers ... 113 7.3.3 Statistical analysis ... 114 7.4 Results ... 115 7.5 Discussion ... 1238 GENERAL DISCUSSION ... 127 8.1 Study limitations ... 128 8.2 Future aspects ... 129 9 CONCLUSIONS ... 131 10REFERENCES ... 133
ABBREVIATIONS
ADC Apparent diffusion coefficient AIS Acute ischemic stroke AF Atrial fibrillation
ApoC-I Apoprotein C-1
ASA Atrial septal aneuyrysm ASCOD Atherosclerosis Small Vessel
Disease Cardiac Source Other Cause Dissection classification BNP Brain natriuretic peptide BBB Blood brain barrier CBF Cerebral blood flow CBV Cerebral blood volume
CCS Causative Classification System CE Cardioembolic
CEA Carotid endarterectomy CE-MRA Contrast-enhanced MRA CISS Chinese ischemic stroke
classification CMR Cardiac MRI
CNS Central nervous system CPP Cerebral perfusion pressure CRP C- reactive protein
CS Cryptogenic stroke CT Computed tomography CACC-CT Combined examination of the
heart, aorta and cervicocranial arteries with computed tomography
CTA Computed tomography angiography
CTP Computed tomography perfusion
DALY Disability-adjusted life-years DSA Digital subtraction angiography DWI-MRI Diffusion weighted imaging MRI EEG Electroencephalogram
ECG Electrocardiogram ECHO Echocardiography ESUS Embolic strokes of
undetermined cause FLAIR Fluid-attenuated inversion
recovery
GFAP Glial fibrillary acidic protein HF Heart failure
HFABP Heart-type fatty acid binding protein
hsCRP high sensitive CRP
IL-6 Interleukin 6
ILR Implantable loop recorder ICH Intracerebral hemorrhage ICAM-1 Intercellular adhesion molecule
1
INR International normalized ratio LA Left atrium
LAA Left atrium appendage LDL Low density lipoprotein LMWH Low molecular weight heparin Lp-PLA2 Lipoprotein-associated
phospholipase A2 LVO Large vessel occlusions MCA Middle cerebral artery MCP-1 Monocyte chemoattractant
protein 1
MET Mechanical thrombectomy MI Myocardial infarction MMP2 Matrix metalloproteinase 2 MRA Magnetic resonance angiography MRI Magnetic resonance imaging mRNA messenger RNA
MTT Mean transit time
NCCT Non-contrast computed tomography
NDKA Nucleoside diphosphate kinase A
NIHSS National Institute of Health Stroke Scale
NINDS National Institute of Neurological Disorders NMDA N-Methyl-d-Aspartate NO Nitric oxygen
NOAC Novel oral anticoaculant NSE Neuron specific enolase OAC Oral anticoagulation treatment OEF Oxygen extraction fraction PAF Paroxysmal atrial fibrillation PARK7 Parkinson disease protein 7 PFO Patent foramen ovale proBNP Pro-brain natriuretic peptide PWI Perfusion-weighted imaging PWTFV1 P wave terminal force velocity in
lead V1
ROPE Risk of paradoxal embolism SVD Small vessel disease
SAH Subarachnoid hemorrhage SEC Spontaneuous echo contrast
suPAR Soluble urokinase-type
plasminogen activator receptor TIA Transient ischemic attack TEE Transesophageal
echocardiography TNF Tumor necrosis factor α TOF Time-of-flight
TF Tissue factor
t-PA Tissue-type plasminogen activator
TOAST Trial of ORG 10172 in Acute Stroke Treatment Subtype Classification
TTE Transthoracic echocardiography TTP Time to peak
USA United States of America VCAM-1 Vascular adhesion molecule 1 vWF Von Willebrandt factor WHO World Health Organisation WUS Wake-up stroke
1 INTRODUCTION
The term “stroke” was used for the first time in the year 1599 and it referred to the abrupt onset of clinical symptoms (1, 2). However, in the literature, the first
descriptions of stroke were made by Hippocrates over 2,400 years ago. At that time, stroke was called apoplexy, which is a Greek word (3). In the mid-1600s, the Swiss physician Johan Jacob Wepfer (1620-1695) discovered that patients who died from apoplexy, had bleeding in the brain (3). He also found that a vessel occlusion in the brain could cause apoplexy (4).
In 1679, Theophile Bonet published his work in a book named `Sepulchretum sive Anatomia Practica´. His opus described various causes of apoplexy and he found that hemorrhages, tumors and even brain abscesses could be a cause of apoplexy (3). Later Morgani (1682-1771) recognized the basic differences behind strokes. He divided the apoplexy into two groups: ´sanguineous´and ´serous ´.
The first term is related to an intracranial hemorrhage while the latter refers to increased fluid content in the brain (5).
Gradually, the discoveries made by John Abercrombie (in the 19th century) and Rudolf Virchow (in the 20th century) established the vascular origin behind the apoplexy. Virchow divided the causes of apoplexy as hemorrhagic and ischemic (6, 7). Thomas Willis noted the role of the cerebral arteries and stroke became
recognized as a cerebrovascular disease. Nowadays, stroke is sometimes called a
"brain attack" to describe the sudden onset of neurological symptoms.
Today acute ischemic stroke (AIS) is a major global health problem. According to the World Health Organization (WHO), there were 16.9 million cases of incidents of stroke worldwide in the year 2010 and almost 6 million people died because of this disease and in addition, 5 million individuals were permanently disabled. In Europe, there are approximately 650,000 stroke deaths each year.
Currently the diagnosis of ischemic stroke relies on the clinical assessment in combination with neuroimaging. Non-contrast computed tomography (NCCT) of the head has excellent sensitivity in detecting a fresh intracranial hemorrhage and is therefore the first-line imaging performed in emergency rooms. One
disadvantage is its poor sensitivity in the diagnosis of ischemic stroke if the ischemia is very recent, small, or in the posterior fossa.
Cardiac sources account for about 20 % of all ischemic strokes and more than 20 specific cardiac disorders have been implicated in leading to brain embolism.
Atrial fibrillation (AF) is the most common source of a brain embolism but many
other structural changes in the heart can predispose to thrombus formation. Even after a thorough etiological work-up, the reason of stroke remains unknown in 25 to 40% of cases - these are called cryptogenic strokes.
This doctorial thesis investigates whether the combined examination of the heart, aorta and cervicocranial arteries with computed tomography (CACC-CT) would improve the diagnostic yield in addition to transthoracic or transesophageal echocardiography (TTE/TEE) in the diagnosis of suspected embolic stroke. We also assessed the benefits of an inflammatory marker, suPar, in the evaluation of the etiology and the outcome of AIS. In addition, we examined whether an assessment of the serum levels of neurofilament light chain and plasma T-tau concentrations could improve the performance of validated clinical models in the prediction of patient outcome after AIS or TIA.
2 REVIEW OF THE LITERATURE
2.1 BURDEN OF THE DISEASE AND EPIDEMIOLOGY
Stroke was responsible for 11.8% of all deaths in 2016 and thus was the second most common cause of deaths worldwide after ischemic heart disease (14.8% of all deaths). Globally, 5.5 million people died because of stroke and 116.4 million disability-adjusted life-years (DALY´s) were lost due to stroke (8). Thus, it ranks as the second most common cause of disability (4.5% of DALYs from all cause) after ischemic heart disease (6.1%) (9). The economic consequences from this disease are huge; for example, 34 billion dollars were spent on its different aspects in the USA (10). A total of 13.7 million new stroke cases occurred in 2016 worldwide and there were over 80 million people who had suffered from a stroke (8).
Stroke incidence rates remained stable between 1990 and 2010, but the number of first-ever strokes in total increased by 68 % worldwide (11). The prevalence increased slightly, but the total number of stroke survivors was increased by 84%. Mortality rates decreased in high-income countries 37% and 20% in low-income and middle-income countries accordingly. However, the total number of deaths increased by 26% (11).
These favorable trends are probably related to improvements in both the prevention and acute management of stroke mainly in the developed countries.
Due to demographical changes, the aging population and increasing life expectancy, the total numbers and burden of the disease are still increasing.
According to the Finnish National Institute for Health and Welfare`s registry, 25000 new strokes occurred in 2014 with about 15000 of these being ischemic (12, 13). Although the incidence of the ischemic stroke has remained rather stable during the 21st century, the number of recurrent strokes has decreased, perhaps due to better secondary prophylaxis (12). Unfortunately, the incidence of early- onset (< 50 years) stroke has increased since the 1980s (14). Case-fatalities and 1- year-mortality rates of all strokes have also slightly declined during the past two decades in Finland. The case-fatality was 8.6% in 2010 and the 1-year-mortality was 19.2% with ischemic stroke patients (15) reflecting the hazardous nature of the disease.
2.2 DEFINITION OF ISCHEMIC STROKE AND TIA
According to WHO, stroke is characterized as a neurological deficit due to an acute focal injury of the central nervous system (CNS) by no other cause than a vascular origin (16). AIS was distinguished from transient ischemic attack (TIA) if the symptoms persisted by over 24 h or led to earlier death. The latest definition of stroke defines it as an acute episode of a focal dysfunction of brain, retina or spinal cord persisting for 24 h. The symptoms may last also for a shorter period, if CT or MRI or autopsy show focal infarction, hemorrhage or subarachnoid
hemorrhage (SAH), relevant to the symptoms (17). TIA has been defined as focal symptoms of less than 24 h duration and with no evidence of infarction on imaging (17). Usually TIA symptoms persist for less than 1 hour, typically 2-15 minutes (18).
Ischemic stroke is the most common cause of stroke (87%). The reason for AIS is most often thrombosis in a vessel or an arterial embolism reaching the brain.
Cerebral hemorrhage accounts for 10% of cases and it is caused by the leakage of blood into the parenchyma. If blood bursts out from an aneurysm in the brain, it is called SAH and accounts for 3% of cases (19). Acute stroke and TIA symptoms require an urgent evaluation and risk stratification in a hospital with the risk of recurrence of symptoms being highest, almost 5%, during the next 48 hours; the 3 month risk of recurrent stroke can be diminished by about 80% with appropriate secondary prophylaxis (20).
2.2.1 Risk factors for stroke
There are several well-known treatable risk factors for AIS such as hypertension, hypercholesterolemia, carotid artery stenosis and atrial fibrillation (AF). Many clinical studies have shown that the appropriate treatment of these conditions markedly reduces the stroke incidence (21-24). Hypertension is the commonest treatable risk factor for stroke (12,25). If systolic blood pressure is reduced by 10 mmHg, the risk of stroke decreases by approximately 35 % and it continues at least to a level of 115/75 mmHg (26). AF increases the risk for cardioembolic stroke by about 5 to 17 fold as compared to non-AF patients (27). However, targeted oral anticoagulation treatment (OAC) has been shown to reduce the risk of stroke by approximately 70% as compared to placebo (24,28). It has been shown that
lowering of the LDL cholesterol concentration by 1 mmol/L with statins can reduce the risk of stroke recurrence by about 12% (29). Other noted treatable causal risk
factors are diabetes mellitus, migraine with aura, sleep apnea, prothrombotic conditions and periodontal diseases (12,30-32).
About 75 % of ischemic strokes are related to behavioural and social factors such as smoking, excessive alcohol use, narcotic abuse, obesity, high salt intake, long-term hormonal therapy, low physical activity, psychosocial stress and low socio-economic status (33,34). Risk factors together have an additive effect on the person’s stroke risk and risk scores are helpful when evaluating the total stroke risk. It has been demonstrated that the ten most common risk factors are
associated with 90% of the risk of stroke (35). Age is an important non-modifiable risk factor for stroke i.e. the incidence of stroke increases markedly after 40 years (36).
2.2.2 Genetic factors for stroke
Stroke is a disease which can happen due to more than 150 known different pathologies rather than a single reason. Many stroke risk factors are influenced by multiple genes, making it difficult to analyse the genetics behind each individual stroke. Stroke genetic studies are complicated by interactions between different risk factors that modulate their effects. Recent progress in gene studies has highlighted the number of known single-gene disorders associated with stroke.
However, monogenic disorders are rather exceptional and many single allele changes exert only a small impact on the risk of ischemic stroke (37, 38). Especially, this evidence has emerged from studies in twins and from families with a history of stroke (39).
Some monogenic syndromes, their inheritance and the underlining genes are presented in Table 1.
Table 1.Monogenic causes of stroke. Modified from Markus et al. (40)
Stroke subtype Monogenic
condition Inheritance Underlying gene Cardioembolic Familial
cardiomyopathies Various
Familial
dysrhythmias Various
Large vessel disease Familial dyslipidemias Moya-moya
Various Uncertain Small vessel disease CADASIL
CARASIL COL4A1 Autosominal dominant retinal vasculopathy
Dominant Recessive Dominant Dominant
NOTCH3 HTRA1 COL4A1 TREX1
Large and small vessel
disease Homocystinemia
Sickle cell disease Recessive Recessive
Cystathione β synthase MTHFR Β-Globin
Mitochondrial disease MELAS Maternal
Abbreviations: CADASIL, cerebral autosominal dominant arteriopathy with subcortical infarcts and leukoencephalopathy; CARASIL, cerebral autosominal recessive arteriopathy with subcortical infarcts and leukoencephalopathy; COL4A1, a gene coding for the type IV collagen alpha 1 chain; MELAS, mitochondrial encephalomyopathy, lactic acidosis and stroke-like symptoms.
2.3 DIAGNOSIS AND EVALUATION OF STROKE
One characteristic sign used in the diagnosis of ischemic stroke is the abrupt onset of focal neurologic deficits. The symptoms experienced by the patient depend on the area of the brain affected. The typical symptoms encountered after an anterior circulation stroke are sudden unilateral blindness (amaurosis fugax), unilateral weakness or numbness (sensomotoric hemiparesis), partial visual loss, altered speech, hemi-inattention, neglect or other visuospatial promlems (41). Headache as a first symptom is uncommon in AIS and usually is related to ICH, sinus thrombosis or SAH. The symptoms of stroke related to the posterior circulation
stroke include isolated visual loss, binocular blindness, diplopia, dysarthia, dysphagia, drop attack, imbalance, coordination difficulties and altered consciousness (41).
The patient´s medical history in conjunction with a clinical and neurological examination are the first key elements in the evaluation of an individual with stroke symptoms. Non-contrast computed tomography (NCCT) of the head is a good method for revealing a fresh intracranial haemorrhage and is the first-line imaging to be performed in emergency rooms. One disadvantage is that it not sensitive for diagnosing a recent ischemia or if the ischemia is small or located in the posterior circulation. Diffusion weighted MRI (DWI-MRI) is superior to NCCT for the diagnosis of AIS in patients presenting within 12 hours of symptom onset (42).
About 20–25% of patients presenting with a stroke syndrome have a stroke mimic;
most commonly seizures, migraine, syncope, hypoglycemia, tumor, encephalitis, abscess, peripheral vestibulopathy, and toxic or metabolic encephalopathy (43) and DWI-MRI can be used to rule out the stroke mimic. The diagnosis of stroke is difficult soon after the onset of symptoms, particularly if the time of onset is uncertain, the symptoms are atypical or changing, the patient is unwell or agitated, access to imaging is delayed, or brain imaging is normal. Table 2 presents the suggested diagnostical work-up in the case of ischemic stroke.
Table 2. Emergency diagnostic tests in acute stroke patients
In all patients
1. Brain imaging: CT or MRI 2. Electrocardiogram, ECG 3. Laboratory tests
o Complete blood count and platelet count, prothrombin time or INR, partial thromboplastin time
o Serum electrolytes, blood glucose o C-reactive protein or sedimentation rate o Hepatic and renal chemical analysis When indicated
4. Extracranial and transcranial Duplex/Doppler ultrasound 5. MRA or CTA
6. Diffusion and perfusion MR or perfusion CT
7. Echocardiography (transthoracic and/or transoesophageal) 8. Chest X-ray
9. Pulse oximetry and arterial blood gas analysis 10. Lumbar puncture
11. EEG
12. Toxicology screen
Adapted from The European Stroke Organization (ESO) guideline 2008
2.4 CLASSIFICATION OF ISCHEMIC STROKE
The most widely used ischemic stroke classification is the Trial of ORG 10172 in Acute Stroke Treatment Subtype Classification (TOAST). This was basically developed to better characterize the TOAST cohort of patients in order to investigate the benefits of administration of a low molecular weight heparin (LMWH), danaparoid, in different stroke subtypes (44). According to TOAST, ischemic stroke can be divided into five categories based on its etiology; large vessel disease (LVD), cardioembolic (CE), small vessel disease (SVD), stroke of other determined etiology, such as vasculitis, traumatic dissections, and cryptogenic stroke. The estimated proportions in the different stroke categories are presented in Figure 1 (45). However,in the age-group 50-years or younger, the proportion of LVD is decreased and cryptogenic strokes (CS) account for over 30% of strokes (36). Moreover, race and ethnicity have an impact on the proportions of etiological subtypes of stroke.
Figure 1. Etiology of stroke. Modified from Grau et al. 2001 (45) LVD, large vessel disease; CE, cardioembolic; SVD, small vessel disease; CS, cryptogenic stroke.
Other established classifications are: the ASCOD phenotyping system (A:
atherosclerosis; S: small-vessel disease; C: cardiac pathology; O: other cause; D:
dissection (46), the Causative Classification System (CCS) (47) and the Chinese ischemic stroke classification (CISS) (48). Approximately 25%-40% of ischemic stroke cases remain with an undetermined cause i.e. cryptogenic stroke, in which a subgroup is nowadays defined as having embolic strokes of undetermined cause (ESUS) (49). Common causes of ischemic stroke are presented in Figure 2.
21 % LVD
26 % CE 21 % SVD
23 % CS
Other
9 %
Figure 2. Modified from the Source: Fauci AS, Kasper DL, Braunwald E, Hauser SL, Longo DL, Jameson JL, Loscalzo J: Harrisons`s Principles of Internal Medicine. 17th
2.4.1 Large vessel disease
The etiology of stroke is defined as a large vessel disease (LVD), if clinical
symptoms are due to ipsilateral atherosclerotic stenosis or occlusion (> 50%) in the carotid, vertebral or intracranial artery and other possible causes are excluded.
The patient’s clinical symptoms should also associate with the radiological findings both in the brain and vascular imaging. The clinical cortical symptoms include aphasia, visuospatial neglect, sensory or motor dysfunction in the extremities or brain stem and cerebellar symptoms such as dysphagia, diplopia, oculomotor dysfunction and vertigo (50, 51). The cortical or cerebellar infarcts are typically larger than 1.5 cm in CT or over 2 cm diameter in DWI-MRI (44, 52). Notably, high grade carotid stenosis carries as much as a 15-20% annual recurrence risk, which
is the key reason that these patients should be identified and evaluated as quickly as carotid endarterectomy candidates (CEA) (53).
Atherosclerotic stenosis affecting less than 50% of the extra-cranial carotid vessels lumen and fragile substenotic plaques can also act as culprit lesions behind AIS or TIA. These changes most likely cause an in-situ thrombosis or
strokes due to arteriogenic embolization. Intra-cranial atherosclerotic plaques may also cause infarcts; for example, by occluding the origins of perforating vessels or ulcers over those plaques may initiate the formation of a thrombus.
2.4.2 Cardioembolism
In cardioembolism, the source of the ischemic event is an embolic occlusion of a brain vessel with the embolus originating from the vicinity of the heart (27).
However, the occluding thrombus can originate also from proximal cerebral arteries or the aortic arch (arteriogenic embolism) as well as from the veins (paradoxical embolism) (49) and its composition can vary e.g. it may consist of tumor cells, calcific fragments and infective material, but mainly it is a so-called red thrombus, containing fibrin, red blood cells and platelets. Cardioembolic strokes are a subtype of ischemic infarcts with the highest in-hospital mortality during the acute phase of stroke (54) and they are responsible for the most severe disability (55).
About 80% of brain embolisms are found in the anterior circulation and typically they manifest as an acute hemiparesis (56). To confirm that a
cardioembolism is the cause of the stroke, large-artery atherosclerosis should be excluded by performing the appropriate neurovascular imaging and at least one cardiac source for the embolus must be identified (57). These cardiogenic sources are typically divided into high-risk and medium-risk findings based on their propensities to cause an embolic stroke. Table 3. (below) enumerates known high and medium-risk embolic sources according to the Stop Stroke Study TOAST (SSS- TOAST) classification (58).
Table 3. Sources of a cardiogenic embolus. Adapted from Ay el al. (58).
High-risk findings Medium-risk findings
Mechanical or prosthetic valve Mitral annulus calcification Rheumatoid mitral or aortic valve disease Left ventricular aneurysm without
thrombus
AF or PAF Atrial septal aneurysm
Sustained atrial flutter Patent foramen ovale Thrombus in the left atrium or in the left
ventriculum
Isolated left atrial turbulence (`smoke`) without mitral stenosis or AF
Sick sinus syndrome
Recent myocardial infarction (within 4 weeks)
Chronic myocardial infarction with ejection fraction less than 30%
Symptomatic congestive heart failure with ejection fraction less than 30%
Dilated cardiomyopathy Infective endocarditis * Papillary fibroelastoma * Left atrial myxoma *
* Not predominantly of thrombotic origin
2.4.3 Small vessel disease
According to the TOAST criteria, a stroke is due to small vessel disease, if it is one the traditional lacunar syndromes and no evidence of cortical dysfunction and a cardioembolic source and a carotid or vertebral stenosis of > 50% of the vessel diameter has been excluded (259). Infarcts are typically due to the occlusion of penetrating arteries and also termed as ´lacunes´. They are small and located
subcortically in the caudate, putamen, external capsule, internal capsule, corona radiata, pontine tegmentum and thalamus. The size in the CT or MRI examination has to be over 1.5 cm in diameter (44). The most important risk factors for small vessel disease are hypertension, diabetes and current smoking (59).
2.4.4 Ischemic stroke due to other determined causes
There are many more rare causes with different mechanisms which can lead to cerebral ischemia. The TOAST classification defines this group as follows: 1. A clear ischemic lesion should be seen in CT or MRI. 2. A specific cause of the stroke should be revealed by blood examinations or arteriography and cardioembolism and large artery disease should be excluded by diagnostic testing. Some examples of this category of strokes are caused by vasculitis, cervical artery dissections, hypercoaguable states and strokes due to hematological diseases or
thrombophilia.
2.4.5 Stroke of undetermined etiology
Quite frequently the cause of a stroke cannot be determined with any degree of confidence and thus patients categorized into this subgroup fail to receive adequate secondary prophylaxis. Despite thorough parallel etiological
investigations, the specific cause of AIS remains unknown and this situation is called cryptogenic stroke (CS). This term was devised for research purposes in the National Institute of Neurological Disorders (NINDS) and Stroke Data Bank (60,61) and subsequently modified in the TOAST trial (44). Patients with two or more possible causes of stroke as well as patients whose etiological evaluation was only cursory belong to this category.
2.4.6 Embolic stroke of undetermined cause
The term embolic stroke of undetermined source (ESUS) was introduced in 2014 to describe patients with a CT/MRI proven non-lacunar ischemic stroke, the absence of at least 50% stenosis in the artery that supplies the ischemic area and no other convincing etiology (49). CS and ESUS are not synonyms, as the latter also includes patients with multiple stroke etiologies or an incomplete diagnostic work-up. ESUS patients are considered as a subset of CS.
2.5 CARDIOEMBOLIC STROKE
Most non-lacunar strokes have an embolic origin and a European study (62) found embolism as the cause of stroke in over 30% of cases. The highest recurrence rate was found also in the CE group (22%), whereas in the LVD group in the same study, the recurrence rate was 10% (62). The number of patients with AF has been
predicted to double in the up-coming years due to demographic changes and the population’s increasing life expectancy, thus in the high-income countries, the numbers of cardioembolic strokes may increase by even several-folds in the next few decades without adequate prevention (63).
There are mainly three mechanisms to account for an embolism originating from the heart:
1. Local blood hemostasis in the cardiac chamber which leads to the formation of a thrombus.
2. Material released from an abnormal valvular surface (e.g., calcific degeneration).
3. Abnormal passage from the venous to the arterial circulation which allows the creation of a paradoxical embolism.
2.5.1 Pathophysiological mechanisms of cardioembolic stroke
Virchow´s triad covers the three basic mechanisms commonly known to lead to blood clotting: a) stasis of blood flow, b) changes in the vessel wall (endothelial injury) and c) hypercoagulability (64, 65). During the clot formation, prothrombin is activated by tissue factor (TF) so that it can be converted into thrombin. Thrombin splits fibrinogen to fibrin and then the fibrin molecules become united together and create a cross-linked fibrin network. This cross-linked fibrin network then comes into contact with thrombocytes, which adhere to the fibrin, building a red thrombus. Any kind of abnormal blood flow or endothelial injury in the heart may predispose to thrombus formation (66).
Flow changes, wall shear stress, viscosity and a high hematocrit are factors prone to form a spontaneous echo contrast (SEC) which is related to increased hemostatic activation (65, 67). Laminar flow and shear stresses are maximal at the vessel wall, and this affects the endothelial cell morphology and function (e.g.
secretion of NO, prostacyclin, t-PA and von Willebrand factor). Platelets near to the vessel wall can become activated by high shear stresses and are able to interact
with vWF and the subendothelium, resulting in platelet adhesion and the initial stages of haemostasis (66, 68). The syntheses of prothrombotic and
proinflammatory endothelial mediators such as TF, von Willebrand factor (vWF), endothelin, ICAM-1 and VCAM1 are also shear-dependent. In addition to changes within the endothelium, some rheological factors can trigger a remodelling of the vessel wall structure e.g. myocytic hypertrophy (65).
2.5.2 Risk factors for cardioembolic stroke
Many cardiac conditions have been identified as potential sources of embolism although AF is recognized as the most common high risk cardioembolic condition.
Other common causes include a recent myocardial infarction, the presence of a mechanical prosthetic valve, dilated myocardiopathy, and mitral rheumatic stenosis. Additional major sources of cardioembolism include infective endocarditis, marantic endocarditis, and atrial myxoma. Minor sources of cardioembolism are patent foramen ovale, atrial septal aneurysm, atrial or ventricular septal defects, calcific aortic stenosis, and mitral annular calcification (69). The distribution of different etiologies is presented in Figure 3.
Figure 3. Sources of cardioembolic stroke. Modified from Wessler et al (70). MI, Myocardial infarction.
2.5.2.1 Atrial fibrillation
In global terms, AF is the commonest sustained arrhythmia (71). The prevalence of AF increases sharply with age, from 0.1 % among adults <55 years to almost 10%
among those >80 years of age (63). The disorder is associated with valvular heart disease, thyroid disorders, hypertension, ischemic heart disease and recent heavy drinking of alcohol. AF predisposes to stroke, because it leads to an irregular contraction of ventricles and stasis, especially in the left atrial appendage (LAA).
Stasis is associated with increased concentrations of fibrinogen, D-dimer, and vWF, which are indicative of a prothrombotic state, which in turn predisposes to
thrombus formation with a consequent increased risk of cerebral embolization (72).
50%
10%
10%
10%
5%
15%
Atrial fibrillation Acute MI
Ventricular thrombus Rheumatic heart disease Prosthetic valves Other, less common sources
AF is classified into first diagnosed, paroxysmal, persistent and permanent AF and recurrent and chronic forms appear to carry a very similar stroke risk (73). It is considered as a high risk for embolic stroke and oral anticoagulation treatment as both primary and secondary prophylaxis are needed. In a meta-analysis with 4792 patients with AF, left atrial (LA) thrombi were detected in about 15% of patients, and almost 90% of these occurred in the LAA (74). An enlargement of the LAA (75) and certain morphological features (76) may predispose the patient to the
formation of the thrombus and a subsequent stroke (77, 78).
2.5.2.2 Systolic heart failure
Approximately 26 million people around the world suffer from heart failure (HF) (79). Poor contractility and low cardiac output, together with likely undiagnosed AF, appear to predispose HF patients to cardiac thrombus (80). As a result, these patients have at least a 3-fold higher risk of stroke than the general population (81).
2.5.2.3 Recent myocardial infarction
Myocardial infarction (MI) is a well-known risk factor for ischemic stroke and 2.5 % of patients suffered a stroke within 4 weeks after an acute MI (82). The MI causes ventricular akinesia, dyskinesia or the formation of a scar on the endothelium resulting in an abnormal blood flow and endothelial denudation with consequent platelet activation (83). The rates of stroke after acute MI are decreasing over time, perhaps due to acute reperfusion therapy, increased use of antithrombotic
medications, and improved secondary prevention therapies (84).
2.5.2.4 Patent foramen ovale (PFO)
PFO is a congenital passage from the right to the left atrium and a potential etiological cause of an embolic stroke via thrombus formation at the foramen or via the paradoxical embolization from the venous system to the cerebral arteries.
It is a common finding in the general population and affects approximately 25% of people (85). Additionally it is observed significantly more frequently in patients with CS than in patients with other stroke etiologies and is associated with an increased risk of stroke recurrence in the presence of an atrial septal aneurysm (ASA) (86,87). Although PFO may be a common stroke risk factor, it is not a major risk factor, except perhaps in very young stroke patients (<50 years).
According to the RoPE-score (Risk of Paradoxical Embolism), the likelihood that a stroke occurred because of paradoxical embolism is higher in younger patients, and also higher with respect to the fewer vascular risk factors (smoking, diabetes, hypertension, prior TIA or stroke) the patient has (88). Previous studies describing the benefit of PFO closure have been equivocal. However, the recent randomized trials of PFO closure yielded impressive results in patients with ESUS aged < 60 years and verified the etiological association between PFO and ESUS (89).
2.5.2.5 Prosthetic heart valves
The prevalence of valvular heart disease is about 2.5% in the general population rising up to 12% in those aged ≥75 years (90). Mechanical or bioprosthetic valves represent a foreign surface in contact with blood and thus pose an elevated risk for thrombus formation. Patients with a mechanical valve have a 4.0% annual risk of stroke, although this can be reduced by the use of oral anticoagulation to 0.8%
for aortic valves and 1.3% for mitral valves (91). Bioprosthetic valves seem to pose a lower risk of stroke than mechanical valves but carry almost the same risk for long-term mortality (92).
2.5.2.6 Aortic arch atheroma
About 45% of individuals aged ≥45 years have a concealed atherosclerotic plaque in their aorta (93), and its presence has been associated with stroke (94). Ulcerated and mobile atheromas, which occur in about 8% of the population (93), have been particularly linked with stroke (95). In the routine diagnostical work-up, aortic atheromas may be an under-recognized cause of stroke because the imaging modalities required to detect them, such as TTE/TEE and CTA of aorta, are not routinely performed in all stroke patients (96). The stroke recurrence rates in patients with aortic arch atheromas have been reported to vary from 3% up to 12
% per year in different patient populations (97, 98).
Atherosclerotic thoraco-aortic plaques greater than 4 mm in depth are more common in patients with stroke than in control patients (94). Even proximal plaques in the descending aorta have been speculated to be related to the ischemic stroke risk (99). Patients with mobile plaques have a high risk of subsequent vascular events (100). There is rather little data available on the management of aortic arch atheroma. Surgical techniques, such as stenting or endarterectomy, have a high risk, and have not shown any benefit (101). In
addition, observational studies suggest that statins may reduce the risk of recurrent events.
CT angiography of the aorta is known to be superior to TEE for identifying the presence of aortic arch atheromatous disease (102).
2.5.2.7 Infective endocarditis
The prevalence of infective endocarditis is about 1 per 10 000 individuals in high- income countries (103). This disease predisposes to the formation of a bacterial mass in the heart valve, which may be released into the bloodstream and cause a brain embolism. Approximately 20% of patients with endocarditis are complicated by stroke and multiple studies have found an over 20-fold relative increase in the stroke risk in the month after the diagnosis of a bacteremia (104) or infective endocarditis (105).
2.5.2.8 Other causes
There are several other, but rare, causes of embolism, such as papillary
fibroelastoma, myxoma and mitral calcification with each of them accounting for
<1% of cardioembolic strokes (49).
2.5.3 Cryptogenic stroke and ESUS concept
Cryptogenic stroke describes a subtype of ischemic stroke, in which no cause can be definitively identified. A large population-based study (106) with 2555 AIS or TIA patients found that 32% of strokes were cryptogenic with the 10-year recurrence rate being 32%. This highlights the importance of determining the possible culprit behind the stroke because recurrent strokes are mainly of the same subtype as the index stroke and this could be avoided with the appropriate secondary stroke prevention strategies (107).
According to the TOAST criteria (44), CS includes all strokes without a defined cause, strokes with two or more mechanisms and strokes in which the work-up is incomplete. In order to decrease the portion of CS, the concept of embolic stroke of undetermined source (ESUS) was proposed in 2014 (49,108). This classification sought to identify a subgroup of patients with CS, who have an embolic-appearing stroke but neither a high-risk cardioembolic source, nor a high-grade arterial stenosis in the arterial supply serving the infarcted brain. It was assumed that a covert AF was the most important mechanism in ESUS patients. This idea was also
supported by the fact that in histopathological analysis of specimens extracted during the endovascular treatment of acute stroke in ESUS patients, most thrombi were erythrocyte-rich (13%) or of mixed composition (80%), with platelet-rich thrombi in only 8% (109). One other study provided evidence that thrombi extracted from ESUS patients more closely resembled cardioembolic clots rather than noncardioembolic thrombi (110).
ESUS involves approximately 15% of all AIS patients, who are typically younger patients with milder strokes. The mean age of ESUS patients is 65 years and they have fewer conventional vascular risk factors than non-ESUS patients with ischemic stroke; 42% are women and the average NIHSS is 5 points (108). The annual stroke recurrence rate was 4.5% during a mean follow-up of 2.7 years (108). Currently, the standard care of ESUS patients is anti-platelet therapy and a modification of concurrent vascular risk factors. Many potential cardiac, arterial, paradoxical, and hematological sources have been proposed in ESUS patients e.g.
these might be amenable to treatment with an anticoagulant (49,111).
Based on the above findings, it was hypothesized that novel oral anticoagulants (NOACS) may decrease the risk of stroke recurrence in ESUS patients, and this proposal has been tested in 2 large randomized controlled trials: NAVIGATE-ESUS (rivaroxaban) and RE-SPECT ESUS (dabigatran) (112,113). Unfortunately, both of them failed to show any efficacy of NOACS superior to aspirin in the prevention of secondary stroke and furthermore they demonstrated an increased risk of ICH.
There is a substantial overlap between CS, ESUS and CE which is demonstrated in Figure 4.
Figure 4. Presentation of the ovelap between cardioembolic, cryptogenic and embolic stroke of undetermined source. Adapted from Kamel et al. (114)
2.5.3.1 Potential mechanisms behind ESUS
There are numerous putative thrombosis-mediated and non-thrombosis-mediated mechanisms of ESUS patients which might embolize and cause a stroke. The most likely mechanisms include a covert structural cardiac lesion, paroxysmal atrial fibrillation (PAF), undefined thrombophilia or a hypercoagulable state related to an undiagnosed malignancy. A few of them are discussed in detail bellow.
2.5.3.2 Paroxysmal atrial fibrillation
PAF often presents asymptomatically and can also easily be missed with only short-term cardiac telemetry. Thus, a large proportion of patients with CS may have AF that has not been detected and this is unfortunate as even short periods of AF increase the risk of stroke (115). The ASSERT study examined 2580 patients, age ≥65 years, with hypertension and no history of AF who had the recent
implantation of a pacemaker or defibrillator. Patients with a single episode of AF lasting over 6 minutes during the first 3 months after implantation of the device, experienced a 2.5- fold higher hazard of stroke during the 2.5 year follow-up (116).
The EMBRACE study enrolled subjects who presented with CS or TIA. The patients with no history of AF were randomized to ECG and inpatient cardiac telemetry versus 30-day cardiac event monitoring. Out of the patients who received the usual investigation, 3.2% were found to have AF, while arrhythmias were detected in 16.1% of patients with prolonged cardiac monitoring (117).
A subcutaneous implantable loop recorder (ILR) and inpatient cardiac telemetry was compared in the CRYSTAL AF study. After six months, AF detection was
significantly higher in the monitored group (8.9 percent, versus 1.4 percent in the control group, hazard ratio 6.4, 95% CI 1.9-21.7) (118). Despite those findings, there is still no common consensus on how long an AF episode needs to last if it is to increase the stroke risk, and furthermore the time causality between AF
dignosis and stroke is controversial.
2.5.3.3 Left atrial cardiopathy
Left atrial cardiopathy is defined by a dysfunction of the left atrium as measured by electrocardiographic, serum or imaging markers, in the absence of AF (119) and its markers include LA enlargement, atrial fibrosis, elevated pro-brain natriuretic peptide (proBNP), and P wave terminal force velocity in lead V1 (PWTFV1). Even in the absence of definite AF, biomarkers of atrial dysfunction have been associated with an increased risk of ischemic stroke.
The temporal relationship between subclinical AF and embolic events is still a matter of debate. Kamel et al. examined adults without stroke or AF (120) and found that PWTFV1 and serum NT-proBNP were associated with incident stroke (HR: 1.04, 95% CI:1.001–1.08 and HR:1.09, 95% CI:1.03–1.16 respectively). Atrial fibrosis, verified with cardiac MRI (CMR), is another marker of LA/ LAA dysfunction that increases the risk of embolic stroke. Left atrial fibrosis was found more frequently in patients with CS versus stroke with a determined non-cardioembolic
