Risk Factors
037 Predictors for detection of atrial fibrillation in cryptogenic stroke patients: Insights from insertable cardiac monitor data in the CRYSTAL AF study
Thijs V1, Brachmann J2, Morillo C3, Passman R4, Sanna T5, Bernstein R6, Diener H7, Di Lazzaro V8, Rymer M9, Assar M10
1Department of Neurology, University Hospitals Leuven, Leuven, Belgium
2Hospital Klinikum Coburg, Teaching Hospital of the University of Würzburg, Coburg, Germany
3Population Health Research Institute, McMaster University, Hamilton, Canada
4Northwestern University Feinberg School of Medicine, Chicago, USA
5Institute of Cardiology, Catholic University of the Sacred Heart, Rome, Italy
6Davee Department of Neurology, Northwestern University, Chicago, USA
7Department of Neurology and Stroke Center, University Hospital Essen, Essen, Germany
8Institute of Neurology, Campus Bio-Medico University, Rome, Italy
9University of Kansas Medical Center, University of Kansas, Kansas City, USA
10Baylor Heart and Vascular Institute, Dallas, USA
Introduction: Undiagnosed atrial fibrillation (AF) may be responsible for a significant proportion of cryptogenic strokes (CS). However, the factors associated with the occurrence of AF in CS patients are not well understood. We assessed these factors in CS patients who received an insertable cardiac monitor (ICM, Reveal XT) for arrhythmia surveillance in the CRYSTAL AF study.
Method: We assessed whether age, gender, race, body mass index, type of index event, CHADS2 score, PR-interval, and the presence of diabetes, hypertension, congestive heart failure, or PFO predicted AF detection within the initial 12 months of follow-up using Cox proportional hazards models and receiver operator characteristic (ROC) curves.
Results: Among 221 ICM recipients (age 61.6 ± 11.4 years, 64% male), adjudicated episodes of AF were detected in 29 patients within 12 months. In univariate analysis, significant predictors of AF included age >65 years (HR 2.8 [95% confidence interval 1.3–5.8], p < 0.01), CHADS2 score (HR 1.9 per one point [1.3–2.8], p < 0.01), PR interval (HR 1.3 per 10 ms [1.2–1.4], p < 0.0001), and diabetes (HR 2.3 [1.0–5.2], p < 0.05). In multivariable analysis, age >65 (HR 2.5 [1.2–5.2], p < 0.05) and PR interval (HR 1.3 [1.2–1.4], p < 0.0001) remained significant and together yielded an area under the ROC curve of 0.73 (95% confidence interval 0.64–0.83).
Discussion: Age >65 years and a longer PR interval at enrollment were independently associated with an increased propensity for AF in CS patients, however they offered only moderate predictive ability in determining which CS patients had AF detected by the ICM in the CRYSTAL AF study.
038 Diabetes in a stroke population: Incidence and mortality
Turner M1, Macleod M1, Philip S2
1Division of Applied Medicine, University of Aberdeen, Aberdeen, UK
2Grampian Diabetes Research Unit, University of Aberdeen, Aberdeen, UK
Introduction: Diabetic patients have an increased risk of stroke with previous studies suggesting ranges from 1.8 to 6 times that of the non-diabetic population. Mortality is also increased. Enhanced awareness of the impact of diabetes on stroke risk and outcomes might influence both primary and secondary prevention strategies. We investigated the incidence of diabetes and subsequent mortality from stroke in this group of patients.
Methods: Population-based data are available for people with stroke and diabetes in NHS Grampian. Anonymised data were extracted and linked from two datasets (local SSCA dataset and local SCI-DC dataset) for stroke patients admitted to hospital between 2000 and 2011 in NHS Grampian. Descriptive statistics and logistic regression models were generated.
Results: 6747 index stroke events were recorded between 2000 and 2011. Of these 897 (13.3%) had a diagnosis of diabetes before the stroke event, per admission year this ranged from 12.1% to 23.2%. Of the 897 who had diabetes prior to stroke, 803 (89.5%) had type II diabetes. 7.9% had diabetes for less than a year, 55.7% between 1 and 10 years, and 21.5% greater than 15 years, prior to the stroke event. Mortality at 7 days post stroke was 4.9% for stroke patients without diabetes compared to 4.8% for stroke patients with diabetes (p = 0.86). By one year mortality was 19.0% and 22.7%, respectively (p = 0.008).
Discussion: Diabetes is a common and important preventable stroke risk factor in our local population. The causes of increased mortality at one year require further exploration.
039 What is current best Stroke Prevention Therapy (SPT) for carotid stenosis in patients requiring intervention? Short & longer-term SPT results for 1500 patients in the ACST-2 Trial
Halliday A, ACST-2 Collaborators
Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK
Introduction: Asympotomatic Carotid Surgery Trial 2 (ASCT-2) is one of the largest trials currently being conducted comparing carotid artery stenting (CAS) with carotid endarterectomy (CEA). Patients with severe asymptomatic carotid stenosis requiring revascularization are entered when CEA & CAS are both possible, but where there is uncertainty as to which is procedure is more appropriate. Currently, 1500 patients have been recruited from 100 centres.
Methods: Before entering ACST-2, all patients should be on appropriate SPT, have recovered from any necessary coronary procedures, and should have an expected lifespan of 5 or more years. Participants are followed-up via annual questionnaire, and use of SPT at 1 month and yearly thereafter is recorded.
Results: Baseline data were available for 1500 patients with a median age of 72 years (SD ± 8.1). 31% of patients had diabetes; most (96%) had ipsilateral stenosis of 70–99% (median 80%) with 30% having contralateral stenosis of 50–99% and 8% with contralateral occlusion. Peri- and post-procedural SPT was broadly similar in both treatment groups. The use of dual-antiplatelet therapy was similar in both groups >1 year post procedure. For 1110 patients undergoing intervention with at least 1 month follow-up and Rankin Scoring at 6 months for any stroke, the overall serious cardiovascular event rate of peri-procedural (>30 days) disabling stroke, fatal MI and death was 1.0%. This compares favourably with results from ACST-1, which reported a peri-operative risk of 1.7%.
Discussion: ACST-2 participants are on appropriate SPT therapies both peri-procedurally and in the longer term. The peri-procedural serious complication rate is low.
040 Perceived psychosocial stress: A risk factor for TIA?
Booth J1, Lawrence M1, Connolly L1, Fraser H2, Dunipace L3, Joice S4, McAlpine C3, Chalmers C5
1Institute for Applied Health Research, Glasgow Caledonian University, Glasgow, UK
2NHS Fife, Fife, UK
3NHS Greater Glasgow & Clyde, Glasgow, UK
4Tangata Massey University, Palmerston North, New Zealand
5NHS Lanarkshire, Lanarkshire, UK
Introduction: No research has explored the association between perceived psychosocial stress and transient ischaemic attack (TIA)/minor stroke, yet public perception highlights stress as a key risk factor for stroke. This study examined perceived psychosocial stress in people experiencing transient ischaemic attack/minor stroke.
Method: A mixed–method study involving a cross sectional survey of perceived stress in three groups of adults (n = 1405): those attending TIA clinics diagnosed with TIA (n = 670); TIA clinic attendees receiving a non-TIA diagnosis (n = 341); treatment room (TR) control group (n = 394). Each clinic attendee completed a structured questionnaire. Qualitative interviews to explore perceptions in detail were undertaken with a sub-set of 16 participants who received a diagnosis of TIA.
Results: Survey data showed the TIA group were older and less deprived than the non-TIA and TR control group. They were less likely to believe stress to be a cause of their TIA. Perceived psychosocial stress was common in all groups but the TIA group reported less stress at home, at work and less financial stress than the non-TIA and TR control groups. Qualitative analysis indicated that psychosocial stress was not perceived to be associated with an increased risk of TIA/minor stroke but was reported to increase following a TIA diagnosis.
Discussion: Perceived stress was not a significant risk factor for TIA; however it may increase as a result of a TIA/minor stroke diagnosis. The period following diagnosis may be an optimal time to intervene to reduce perceived stress, contributing to effective secondary prevention of stroke.
041 Clinical relevance of short-runs of atrial tachy-arrhythmia: A UK wide survey of stroke physicians and cardiologists
Tran R1, Rankin A1, Abdul-Rahim A2, Lees K2, Rankin A3
1Western Infirmary, Glasgow, UK
2Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK
3School of Medicine, University of Glasgow, Glasgow, UK
Introduction: Conservative electrocardiography (ECG) definition of atrial fibrillation (AF) with ≥30 seconds recording on ambulatory ECG monitoring is required for the diagnosis of paroxysmal AF. It is unclear whether shorter runs of atrial tachy-arrhythmia detected on ambulatory ECG are relevant in conferring future stroke risk.
Method: A UK wide online survey was distributed to Cardiologists and Stroke Physicians, via their national societies, to assess current diagnosis and management of patients with <30 seconds atrial tachy-arrhythmia detected on ambulatory ECG.
Results: 205 clinicians responded to the survey (130 stroke physicians, 64 cardiologists and 11 other). Regarding the diagnosis of AF, a single 12-lead ECG was accepted by 87% of responders. In contrast, only 45% would accept a single episode lasting <30 seconds detected on ambulatory monitoring. The level of agreement regarding the decision to anticoagulate in hypothetical patients varied depending on CHA2DS2-VASc score, ambulatory ECG findings and past history of stroke/transient ischaemic attack (TIA). The highest level of agreement (94.1%) was observed in the highest risk category, while the lowest level of agreement (56.0%) was observed in the intermediate risk category. There was a trend suggesting that stroke physicians were more likely to anticoagulate a low risk patient (CHA2DS2-VASc = 1) with a single episode of atrial tachy-arrhythmia detected [OR 2.02 (95% CI 0.71–5.76), p = 0.19).
Discussion: There is a lack of consensus on the diagnosis and management of patients with short runs of atrial tachy-arrhythmia detected on ambulatory ECG. More work is needed to clarify the risk of stroke in this unique population of patients.
042 Is anaemia an independent predictor of death and or dependency in acute ischaemic stroke? Retrospective case series
Boovalingam P, Wilson N, Brawn L, Blake M, Ducker D, Srinivasan V, O'Kane D, Day R, Lineham K
Department of Stroke Medicine, Northampton General Hospital, Northampton, UK
Introduction: Strokes are often considered a consequence of hypertension and atherosclerosis. Anaemia as a risk factor for acute ischemic stroke has been reported in only a few case reports.
Methods: We investigated the association between anaemia at admission and clinical outcomes in acute ischaemic stroke in Northamptonshire admitted to Northampton General Hospital with stroke for the period of 7 months from Jan 2013 to July 2013. Retrospective case analysis; All the data including baseline characteristics, factors influencing outcomes, management and outcome were analysed using multivariate logistic regression.
Results: Of the 516 cases, 174 cases had anaemia at admission, which accounted for 33.7%. The independent influencing factors of anaemia were age (OR = 1.09, 95% CI: 0.95–1.30), alcohol excess (more than 21 units per week (OR = 0.99, 95% CI: 0.38–1.42), estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m2 at admission (OR = 1.78, 95% CI: 1.1–2.30).
Discussion: In our case series, about one third of acute stroke patients had anaemia, and anaemia at admission is an independent predictor of death and disability with stroke patients.
