Liverpool-Glasgow VISTA Collaboration

Proposed authors and collaborators:
Azmil H Abdul-Rahim, LCCS University of Liverpool [lead]
*Frank Huang, LCCS University of Liverpool
*Hironori Ishiguchi, LCCS University of Liverpool
Wahbi K El-Bouri, LCCS University of Liverpool
Jesse Dawson, University of Glasgow
Gregory YH Lip, LCCS University of Liverpool
for the VISTA collaborators
(*post-docs)

1. Work-Package, WP 1: Stroke-Heart syndrome in people with ischaemic and haemorrhagic strokes

1.1. Introduction
Stroke is a major cause or mortality and morbidity. One of the recently recognised consequences of stroke includes ‘Stroke-Heart syndrome’, a new-onset cardiovascular complication, which is significantly associated with worse prognosis in terms of major adverse cardiovascular events.1, 2 In a heart disease-free population-based cohort study that compared people with first-ever acute ischaemic stroke (AIS) and propensity-matched individuals without stroke, the risk of early major cardiovascular events within 30 days of AIS is reported to be 25-fold in female and 23-fold in male.3 The risk is attenuated after 30 days but remains significant compared to the general population.3 Intracerebral haemorrhage (ICH) also increases the risk of cardiovascular complications with a 2-fold increase in the risk of myocardial infarction following ICH.4

Stroke–Heart syndrome encompasses a broad clinical spectrum of cardiovascular changes including
(1) acute myocardial injury (as evidenced by acute elevation of hs-cTn), (2) acute coronary syndromes (ACS; including type 1 and type 2 myocardial infarction [MI]), (3) systolic and diastolic LV dysfunction including Takotsubo syndrome (TTS), (4) cardiac arrhythmias including atrial fibrillation (AF) and relevant ECG changes, and (5) neurogenic sudden cardiac death.

Nonetheless, most of our understanding on Stroke-Heart syndrome derived from retrospective observational cohort studies. VISTA is a collaborative venture that collates data from completed acute stroke trials (from year 1998). The completed trials contain rich data on patient demography, laboratory measurements, cardiac and brain imaging, adjudicated outcome measures, and adverse events. Disregarding the treatment groups, the data are rich in patient’s natural history and can be used for exploratory analyses.

The primary aim of the WP is to investigate the incidence and impact of Stroke-Heart syndrome in people with AIS and acute ICH; and to facilitate the development of prognostic models for people who are at risk of Stroke-Heart syndrome, following their initial stroke.



2. WP 2: ABC-Stroke Pathway

2.1. Introduction
The management of people with stroke is often multidisciplinary, involving various specialties and healthcare professionals, as well as patient caregivers and next-of-kin. There is an immediate need for a co-ordinated action to address the underlying risk factors for cerebrovascular, requiring greater investment in, and implementation of, stroke prevention and health promotion following a stroke. The importance of an ‘integrated care’ approach has been applied to other chronic conditions.5, 6 In people with atrial fibrillation (AF), which is a common cause for ischaemic stroke (IS), the ABC (Atrial fibrillation Better Care) pathway has been proposed as an integrated care approach with three central pillars: ‘A’ Avoid stroke (with appropriate antithrombotic); ‘B’ Better symptom management; and ‘C’ Cardiovascular and Comorbidity risk optimisation.5

Recently, the European Society of Cardiology (ESC) Council on Stroke position paper proposed a holistic integrated care management approach for people with stroke.7 The integrated ABC pathway is highly relevant when applied to people with stroke to improve their outcomes.

Preliminary data from the Athens Stroke Registry suggested that among 2,823 stroke patients, followed-up for up to 10 years, complete adherence to the ABC stroke pathway was associated with a significant reduction in the risk of stroke recurrence, with greatest magnitude observed for full ABC pathway adherence (adjusted Hazard Ratio [aHR] 0.38, 95% Confidence Interval [CI]: 0.22–0.64). Compared to incomplete adherence to ABC pathway, full adherence was associated with reduced risks of stroke recurrence (aHR: 0.61, 95% CI: 0.39–0.94), MACE (aHR: 0.65, 95% CI: 0.46–0.91) or death (aHR: 0.28, 95% CI: 0.17–0.44). Thus, adherence to an integrated care pathway for stroke was associated with a significant reduction in the risk of stroke recurrence, major adverse cardiovascular events and mortality, underlining the importance of ABC pathway implementation.

Using independent trials data within VISTA, this WP aims: i) to evaluate the impact of adherence to different number of ABC stroke pathway criteria on the risk of outcomes; and ii) to explore whether the expansion of adherence to the non-cardiovascular comorbidities pool for the ‘C’ criterion would provide consistent estimates on the impact of the ABC pathway in people with recent stroke and ICH. We will look into ischaemic stroke and ICH cohorts separately.

3. WP 3: Multimorbidity, polypharmacy and frailty in people with stroke

3.1. Introduction
Multimorbidity and frailty are prevalent in people with acute ischaemic stroke (AIS) and intracerebral haemorrhage (ICH).8, 9 Polypharmacy is also highly prevalent in people with high risk of stroke.10, 11 Despite their prevalence, the impact of multimorbidity, polypharmacy and frailty on stroke outcomes have received relatively little research attention to-date. Some data suggest that the conditions are not only associated with poor stroke outcomes but may be associated with differential responses to acute revascularisation treatment.12 Their presence may also interfere with recommended rehabilitative treatments post-stroke, and increase the risks of polypharmacy. Thus, there is immediate need to better understand the relationships between of multimorbidity, polypharmacy and frailty in people with AIS and ICH, including the impact on functional outcomes following revascularisation therapies.

The WP aims to explore the associations between multimorbidity, polypharmacy and frailty with health-related stroke outcomes (mortality, functional outcomes, stroke prevention, quality of life) using available large high-quality trials data.

1.2. Methods
1.2.1. Data Source and Patients
Data will be sought and extracted from VISTA, to form 2 cohorts: (1) people with AIS, and (2) people with acute ICH, who had been randomised to receive placebo or any drug now known to possess no confirmed influence on stroke outcome. We will evaluate the 2 cohorts separately.

We will exclude people for whom we lack relevant baseline demographics and outcome information: baseline National Institutes of Health Stroke Score (NIHSS), age, sex, occurrence of adverse events as well as serious adverse events, modified Rankin Scale (mRS) at discharge and at day 90, and vital status (death will be recorded as mRS grade 6).

1.2.2. Demographics
Certain demographics identified from literature to be related to ischaemic stroke will be investigated as possible risk factors. These include: systolic and diastolic BP at baseline; pre-stroke mRS; history of hypertension; history of diabetes; history of ischaemic heart disease or MI; history of hyperlipidaemia; history of stroke or TIA; history of atrial fibrillation; history of heart failure; history of smoking; previous medication including antiplatelet therapy or anticoagulant drugs; blood parameters such as baseline glucose, creatinine/eGFR and platelets, (lipids and coagulation screen i.e. PT, APTT, INR- if possible), stroke/ ICH location and size.

We will identify the outcome measures from reviewing the trials’ adjudicated outcomes and reported adverse events.

1.2.3. Data Analyses
Standard list of demographics, past history, physiological measures and drug treatment (i.e. baseline demographics and disease risk characteristics) will be reported as descriptive statistics comparing overall and each profile. These will be displayed as mean (standard deviation) or median (inter-quartile range) for continuous variables and count (percentage) for categorical variables. Unadjusted baseline comparisons will be conducted using the 2-sample t test, the Mann-Whitney U test, the 2 proportions test, or the χ2 test depending on the distribution and nature of the data.

We will report the incidence of Stroke-Heart syndrome and breakdown of its components at 30 days and 90 days post-stroke. We will evaluate the modified Rankin scale (mRS) at 90 days, in people who had experienced Stroke-Heart syndrome compared to those who did not. Kaplan–Meier life table analysis, Cox proportional hazards, and logistic regression models adjusted for age, race, sex, and pre-stroke mRS will be used for analysis of the outcomes. We will also incorporate artificial intelligence- machine learning and in-silico modelling to develop prognostic models for people who are at risk of Stroke-Heart syndrome. The prognostic models can be validated in an independent cohort.

Analyses will be undertaken appropriate statistical and AI-modelling software.

2.2. Methods
2.2.1. Data Source, Patients and Demographics
Data will be sought and extracted from VISTA, to form 2 cohorts: (1) people with AIS, and (2) people with acute ICH. We will exclude people for whom we lack relevant baseline demographics and outcome information: baseline National Institutes of Health Stroke Score (NIHSS), age, sex, co-morbidities (past medical history), medications, occurrence of adverse events as well as serious adverse events, modified Rankin Scale (mRS) at discharge and at day 90, and vital status (death will be recorded as mRS grade 6). We will include patients with complete data to evaluate retrospectively the adherence to the ABC pathway, and follow-up data on the primary outcome defined for this analysis and composite outcome of all-cause death and major adverse cardiovascular events (MACEs).

We will define two versions of the ABC pathway, which differed according to the ‘C’ criterion definition. For our primary analysis, we will define a ‘standard’ version of the ABC pathway, in which we evaluate adherence to ‘C’ according to the most common comorbidities found in people with stroke; for the exploratory secondary analysis on the ‘expanded’ version of the ABC pathway, we will also evaluate additional non-cardiovascular comorbidities (e.g. hyperthyroidism, hyperlipidaemia, dyspeptic disease).

Adherence to each criterion of the ABC pathway is defined as follows:
‘A’ Criterion: post-stroke patients were considered adherent to this criterion if appropriately prescribed antithrombotic agent (i.e. for AIS – antiplatelet agent or oral anticoagulant for AF).
‘B’ Criterion: post-stroke patients who received stroke unit care and rehabilitation, and those who showed improvement in mRS at 90 days. We will also evaluate whether antidepressant medication or psychology input was considered in post-stroke patients with low mood.
‘C’ criterion: Assessment is made according to the presence and treatment of baseline comorbidities. For the ‘standard’ version of the ‘C’ criterion, we will include comorbidities which are most commonly found in stroke patients and that were previously used to evaluate adherence to the ABC pathway: hypertension, diabetes, coronary artery disease (CAD), congestive heart failure (CHF), history of previous stroke/TIA and peripheral artery disease (PAD). For the exploratory “expanded” version, we additionally will evaluate the presence and treatment of dyslipidaemia, dyspeptic disease (as defined by the presence of gastritis/duodenitis or peptic ulcer) and hyperthyroidism.

2.2.2. Data Analyses
In the primary analysis, the cohort will be divided according to the number of ABC criteria fulfilled, from 0 (none) to 3 (all criteria). For our secondary analyses, we will also consider a) adherence to 0–1 vs. 2–3 ABC criteria; b) full-adherence vs. non-adherence to ABC pathway; and c) pattern of ABC criteria adherence.

We will identify the outcome measures from reviewing the trials’ adjudicated outcomes and reported adverse events. These will include stroke, thrombo-embolism, MI, major bleeding, MACEs, all-cause death and cardiovascular death.

Statistical analysis
Baseline characteristics will be reported as mean and standard deviation (SD) or median and interquartile range [IQR] for normally and non-normally distributed continuous variables and compared with appropriate parametric (including t-test and ANOVA) and non-parametric tests (including Mann–Whitney U and Kruskal–Wallis), respectively. Frequencies and percentages will be reported for categorical variables, and were compared using chi-square test.

Incidence rates and 95% Confidence Intervals (CI) will be calculated according to the number of events and person-years of follow-up, and multivariable Cox-regression analyses to be performed to evaluate the effect of adherence to the ABC pathway on the risk of major outcomes, after adjustment for age (modelled as a linear variable), sex, stroke severity (NIHSS score) and major comorbidities (hypertension, diabetes mellitus, CHF, CAD, PAD and history of stroke/TIA); results will be reported as Hazard Ratio (HR) and 95% CI. For the primary outcome, we will produce Kaplan–Meier curves to represent the cumulative hazard of patients, and survival distributions compared using Log–Rank test. We will also perform secondary analyses: First, we will evaluate the effect of being adherent to 2–3 vs. 0–1 ABC criteria; second, we will explore the effect of full adherence to ABC pathway; finally, we will also assess the contribution of different patterns of adherence to ABC criteria.

All the analyses will be performed using appropriate statistical software.

3.2. Methods
3.2.1. Data Source, Patients and Demographics
Data will be sought and extracted from VISTA, to form 2 cohorts: (1) people with AIS, and (2) people with acute ICH. We will exclude people for whom we lack relevant baseline demographics and outcome information: baseline National Institutes of Health Stroke Score (NIHSS), age, sex, co-morbidities (past medical history), medications, occurrence of adverse events as well as serious adverse events, modified Rankin Scale (mRS) at discharge and at day 90, and vital status (death will be recorded as mRS grade 6).

This WP will attempt to evaluate the following aspects:
1) Multimorbidity, polypharmacy and frailty prevalence and impact on outcomes in people with AIS and ICH. These would include mortality, functional outcomes and discharged destination.
2) Multimorbidity, polypharmacy and frailty impact on the outcomes in people with AIS who received revascularisation therapy, in particular, mechanical thrombectomy.
3) Multimorbidity, polypharmacy and frailty impact on secondary stroke prevention, including adherence to the ABC-stroke pathway, in people with AIS and ICH.

3.2.2 Data Analyses
Standard list of demographics, past history, physiological measures and drug treatment (i.e. baseline demographics and disease risk characteristics) will be reported as descriptive statistics comparing overall and each profile. These will be displayed as mean (standard deviation) or median (inter-quartile range) for continuous variables and count (percentage) for categorical variables. Unadjusted baseline comparisons will be conducted using the 2-sample t test, the Mann-Whitney U test, the 2 proportions test, or the χ2 test depending on the distribution and nature of the data.

We will report the incidence of mortality, functional outcomes (mRS), stroke prevention, and quality of life for each cohort. Kaplan–Meier life table analysis, Cox proportional hazards, and logistic regression models adjusted for age, race, sex, and pre-stroke mRS will be used for analysis of the outcomes. We will also incorporate artificial intelligence and machine learning to evaluate outcomes of interests.

Analyses will be undertaken appropriate statistical and AI-modelling software.

1. Scheitz JF, Sposato LA, Schulz-Menger J, Nolte CH, Backs J, Endres M. Stroke-heart syndrome: Recent advances and challenges. Journal of the American Heart Association. 2022;11:e026528
2. Buckley BJR, Harrison SL, Hill A, Underhill P, Lane DA, Lip GYH. Stroke-heart syndrome: Incidence and clinical outcomes of cardiac complications following stroke. Stroke. 2022;53:1759-1763
3. Sposato LA, Lam M, Allen B, Shariff SZ, Saposnik G. First-ever ischemic stroke and incident major adverse cardiovascular events in 93 627 older women and men. Stroke. 2020;51:387-394
4. Murthy SB, Zhang C, Diaz I, Levitan EB, Koton S, Bartz TM, et al. Association between intracerebral hemorrhage and subsequent arterial ischemic events in participants from 4 population-based cohort studies. JAMA Neurology. 2021;78:809-816
5. Lip GYH. The abc pathway: An integrated approach to improve af management. Nature Reviews Cardiology. 2017;14:627-628 6. Field M, Kuduvalli M, Torella F, McKay V, Khalatbari A, Lip GYH. Integrated care systems and the aortovascular hub. Thrombosis and haemostasis. 2021;122:177-180
7. Lip GYH, Lane DA, Lenarczyk R, Boriani G, Doehner W, Benjamin LA, et al. Integrated care for optimizing the management of stroke and associated heart disease: A position paper of the european society of cardiology council on stroke. Eur Heart J. 2022;43:2442-2460
8. Gallacher KI, Jani BD, Hanlon P, Nicholl BI, Mair FS. Multimorbidity in stroke. Stroke. 2019;50:1919-1926
9. Evans NR, Todd OM, Minhas JS, Fearon P, Harston GW, Mant J, et al. Frailty and cerebrovascular disease: Concepts and clinical implications for stroke medicine. Int. J. Stroke. 2022;17:251-259
10. Martínez-Montesinos L, Rivera-Caravaca JM, Agewall S, Soler E, Lip GYH, Marín F, et al. Polypharmacy and adverse events in atrial fibrillation: Main cause or reflection of multimorbidity? Biomed Pharmacother. 2023;158:114064
11. Proietti M, Raparelli V, Olshansky B, Lip GY. Polypharmacy and major adverse events in atrial fibrillation: Observations from the affirm trial. Clin Res Cardiol. 2016;105:412-420
12. Joyce N, Atkinson T, Mc Guire K, Wiggam MI, Gordon PL, Kerr EL, et al. Frailty and stroke thrombectomy outcomes-an observational cohort study. Age and ageing. 2022;51

Funding: We are in the process of securing funding for VISTA data access (anticipated costs is £10,000).