Article Text
PDF
Abstract
Background Atrial fibrillation (AF) is the most common type of cardiac arrhythmia. Future estimations suggest an increase in global burden of AF greater than 60% by 2050. Numerous studies provide growing evidence that AF is not only associated with stroke but also with cognitive impairment and dementia.
Aim The main goal is to assess the impact of the combined use of cardiac rhythm monitoring devices, echocardiography, biomarkers and neuroimaging on the early diagnosis of AF, silent strokes and cognitive decline, in subjects at high risk of AF.
Methods and analysis Two-year follow-up of a cohort of individuals aged 65–85 years at high risk for AF, with no prior diagnosis of either stroke or dementia. The study involves baseline echocardiography, biomarkers, and neuroimaging, yearly cardiac monitoring, and semiannual clinical assessments. Different parameters from these tests will be analysed as independent variables. Throughout the study period, primary outcomes: new diagnoses of AF, stroke and cognitive impairment, along with any clinical and therapeutic changes, will be registered. A first descriptive and bivariate statistical analysis, appropriate to the types of variables, will be done. The information obtained from the data analysis will encompass adjusted risk estimates along with 95% confidence intervals. Event risk predictions will rely on multivariate Cox proportional hazards regression models. The predictive value of the model will be evaluated through the utilisation of receiver operating characteristic curves for area under the curve calculation. Additionally, time-to-event analysis will be performed using Kaplan-Meier curves.
Ethics and dissemination This study protocol has been reviewed and approved by the Independent Ethics Committee of the Foundation University Institute for Primary Health Care Research-IDIAP Jordi Gol (expedient file 22/090-P). The authors plan to disseminate the study results to the general public through various scientific events. Publication in open-access journals and presentations at scientific congresses, seminars and meetings is also foreseen.
Trial registration number NCT05772806.
- Primary Care
- Primary Prevention
- Cardiology
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The study considers the use of a novel algorithm that will enable the selection of a high-risk population, allowing the analysis of sufficient events with a relatively small sample size and follow-up period.
The study innovatively targets individuals without a prior atrial fibrillation (AF) diagnosis, challenging current guidelines that typically focus on those already diagnosed with AF, and often also with stroke or other AF consequences.
The longitudinal prospective nature of the study design will allow time-to-event analysis.
The outpatient setting introduces heterogeneity in monitoring adherence and device proficiency; variability will be factored into the analyses.
The diagnostic approach employed in this study (encompassing heart rate detection devices, echocardiography, neuroimaging and biomarker tests) may pose limitations on its applicability to other primary healthcare settings.
Introduction
In Europe, the number of ischaemic strokes related to atrial fibrillation (AF) among people aged ≥80 years will triple (2010–2060). In addition to this increase in incidence, the socioeconomic gradient in both the age of diagnosis and the burden of comorbidity is widening.1–3 Ischaemic strokes and AF, which are interlinked and have common complex pathways, are becoming a double epidemic, especially among older persons, with significant health implications, increased disability and a great socioeconomic burden for society. Together with the well-established relationship among AF and ischaemic strokes, there is a growing evidence supporting the association between AF and cognitive impairment, and dementia.4–10 Furthermore, in 24% of patients with a stroke episode, a previously unknown AF is detected,11 with a higher risk of new cognitive impairment diagnoses or worsening of pre-existing ones, heart failure, sudden death and cardiovascular morbidity.12–14 Early detection of AF is crucial because it enables healthcare providers to implement appropriate management strategies, such as anticoagulation therapy, to reduce the risk of recurrent strokes and other associated complications. However, despite this evidence, there is very limited research regarding interventions in high-risk individuals without a known diagnosis of AF.
Cardiac monitoring for early detection of AF
Various risk scales have been proposed to identify patients at high risk, for whom monitoring would facilitate early detection of AF.15–21 The electrocardiogram (ECG) and machine learning classifiers provide valuable insights that can significantly enhance the identification of AF in patients with embolic stroke of undetermined source.22 23 Prolonged monitoring using devices such as Holter monitors, implantable loop recorders or mobile cardiac telemetry can capture intermittent AF episodes that might go unnoticed during a brief in-office ECG.24 Emerging screening tools, including smartphone apps and smartwatches, are rapidly advancing for the detection of AF, particularly in high-risk patients. These innovations hold the potential to enhance AF prediction and improve stroke prevention significantly.23–25
Brain conditions and AF
Silent strokes are associated with the presence of AF irrespective of its subtype26; they are asymptomatic in the majority of cases but have been detected in up to 22% of studies employing computerized tomography (CT) scans and as high as 44% using cranial magnetic resonance imaging (MRI).27 It is possible to identify patterns in brain lesions among patients who have had an ischaemic stroke and then use these patterns to predict the probability of experiencing a first episode of AF.28
At the same time, silent brain infarcts have been associated with progressive cognitive dysfunction.29 Cognitive impairment is notably higher in women,20 and the risk of Alzheimer’s dementia is already elevated by 30% in individuals with AF.30
The role of biomarkers
Several biomarkers have been associated with brain lesions.31 In the case of dementia and cognitive impairment, various investigations are underway with the goal of identifying potential biomarker profiles for both early diagnosis and disease progression monitoring.32 However, challenges persist in determining their cost-effectiveness and applicability in clinical practice.33 A recent publication has demonstrated the cost-effectiveness of integrating biomarkers into stroke risk stratification scales for patients with AF.34 Furthermore, combining parameters derived from images and biomarkers could improve the predictive value for identifying patients at risk of suffering a stroke or developing dementia. However, the current understanding of the predictive value and applicability, particularly in primary care, of combining plasma biomarkers and radio imaging patterns with clinical variables in the stratification of AF, stroke or dementia remains unknown.
Echocardiography, AF, cerebrovascular disease and cognitive impairment
The echocardiographic features of AF have predictive value for cognitive assessment outcomes. Left atrial dysfunction, rather than its size, has shown correlations with the risk of dementia35 and cerebrovascular ischaemic events.34–39
Although its suitability for clinical practice remains unproven, primarily due to the absence of prospective randomised studies, certain research indicates that incorporating parameters of atrial cardiopathy alongside conventional algorithms may potentially improve risk stratification for both ischaemic stroke and AF.40
The need for an early diagnosis of AF
Currently, there is a lack of evidence regarding the outcomes associated with early diagnosis of AF and potential vascular comorbidities before its formal diagnosis. This entails using a combination of high-risk clinical criteria, echocardiography, MRI and monitoring in patients without a prior AF diagnosis.
Thus, the main objective of present study is to determine the impact of the combined use of cardiac rhythm monitoring devices, biomarkers, echocardiography and MRI on the early detection of AF, silent strokes and cognitive decline in subjects at high risk of developing AF.
Methods and analysis
Study design and setting
This is a multicentre, prospective study, which will follow a cohort of patients between 65 and 85 years old at high risk of AF for 2 years.
Participants will be recruited from the usual consultations in six primary healthcare (PHC) centres managed by the Catalan Health Institute and located in Tarragona region (South Catalonia, Spain).
Study population
Persons with an active electronic clinical history at any of the study PHC centres (in our territory they represent 97.6% of the entire population) who meet the following conditions will be invited to participate (eligibility criteria):
High risk of AF, according to the risk score validated in the AFRICAT (Atrial Fibrilation Research in CATalonia) study. This scale considers the following variables for risk calculation: sex, age, weight, cardiac rate and CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 (doubled), diabetes mellitus, prior stroke or transient ischemic attack (doubled), vascular disease, age 65–74, female) score.15
CHA2DS2-VASc score≥2.
Ability to use a smart phone (or at least the caregiver).
Patients with the following conditions will be excluded (exclusion criteria):
Previous diagnosis of AF.
Previous diagnosis of stroke.
Severe cognitive impairment, with a score on the Global Deterioration Scale (GDS)≥3.
Severe functional impairment, with a Barthel score ≤60, or modified Rankin score≥4.
Active anticoagulant treatment at the inclusion.
Vital prognosis less than 1 year.
Pacemaker carriers.
Recruitment evaluation
A list of potential study candidates (individuals aged 65–85 years at high risk of AF) will be provided to researchers at PHC centres by a member of the institutional Information and Communication Technology Unit. Candidates will be randomly selected for a recruitment evaluation until the desired sample size is attained.
The specified eligibility criteria will be evaluated during the recruitment visit by a member of the research team. The investigator will verify whether the scales’ scores are up to date in the electronic clinical records of candidates, or will administer scales, if they are not. All recruiters are well versed in using the proposed scales as they are routinely administered in accordance with territorial protocols for elderly individuals.
The presence of a previous diagnosis of AF or ischaemic stroke will be considered if it is registered in the patient’s clinical electronic record under the International Classification of Diseases, 10th revision, with codes I48 or I63. Furthermore, the absence of these antecedents will be confirmed by directly querying the patient. Individuals meeting the criteria will be invited to participate in the study, and on signing the informed consent, they will be scheduled for a baseline assessment.
Follow-up
A 2-year follow-up after study starting (2023) is planned, and abandonment of the follow-up will occur in cases of death, transfer to another PHC centre or on the patient’s expressed will.
Determinations (baseline and follow-up assessments)
Once during study period
At first, all included patients will undertake echocardiography, laboratory biomarkers and cranial MRI. These tests will detect the main conditions associated with a major risk of presenting the primary outcomes of the study.
From the echocardiogram, signs of morphological or functional dysfunction of the left atrium will be recorded as predictors of AF, such as size and volume, strain and ejection fraction of the left atrium, as well as valvular dysfunctions.
In the cranial MRI, signs of chronic ischaemia or atrophy, likely related to cognitive impairment, will be examined for, including microinfarts, enlarged ventricles, white matter hyperintensities (along with Fazekas score), microbleeds and perivascular dilations.
Specific serum biomarkers that will be determined are N-terminal pro-brain natriuretic peptide, angiopoietin-2, fibroblast growth factor-23, bone morphogenetic protein-10 and troponin T. All of them are presumptive markers of cardiac or cerebral damage or dysfunction. The collection of serum samples for biomarkers determination will take place at each PHC centre for the convenience of participants. However, both the processing of the samples and the determination of biomarkers will strictly adhere to a standardised protocol meticulously defined and established by the central laboratory responsible for performing these determinations.
Every year
Cardiac monitoring (during 14 days) for active search of AF, with two different electronic devices, such as the FibriCheck App41 and the Fitbit bracelet,42 will be performed at the beginning and 1 year after inclusion.
Every 6 months
ECG and complete clinical assessment will be done to register risk factors and comorbidities, new cardiovascular events, cardiovascular parameters (arterial pressure, heart rate.) and scores at different tests related to functional status (Barthel or Rankin), cognitive function (GDS, Mini Mental State Examination), and AF/stroke risk (CHA2DS2-VASc score). Drug prescriptions (according to electronic registers), at baseline, and changes during the study period will be also recorded biannually.
Figure 1 shows an outline of the most important assessments during the baseline visit and study follow-up.
Main evaluations during study follow-up.
Data collection and management
Study data will be recorded from tests and evaluations performed on participants: anamnesis (including scales), physical examination, laboratory tests and reports provided by specialists (available at computerised clinical records of hospital centres). The echocardiogram and cranial MRI data will be recorded from the reports made by the specialists who will carry out the tests: a cardiologist and a radiologist. Data registered in the electronic PHC records of participants will also be collected. Therapeutic (pharmacological) changes will be registered from the integrated electronic prescription system.
Finally, heart rate monitoring data will be collected through the electronic devices’ registers.
An ad hoc data collection questionnaire will be created to register and store all study variables. The electronic questionnaire will be available at a specific application, and only accessible from the corporate Intranet of the Catalan Health Institute. The data will remain stored for 5 years and can only be recorded and accessed by study’s researchers using personal passwords.
Primary outcomes
Time to detection of AF: To diagnose the AF, confirmation by a 12-lead ECG or Holter will be required.43 44
Time to new stroke diagnosis: Confirmation through a neuroimaging test or neurologist assessment will be required.
Time to new cognitive impairment or dementia diagnosis: Changes in cognitive function will be assessed through scores at the GDS. Dementia diagnosis will be established with scores ≥4. The diagnosis shall also be made if there is a suspicion confirmed by a neurologist’s evaluation.
Time to death (if it occurs).
Sample size calculation
To estimate the prevalence of AF observed with a difference of 0.1 units compared with the reference population (0.07), a sample of 148 individuals will be required.14 An alpha risk of 0.05 and a beta risk of 0.20 are accepted, and a loss rate of 15% has been estimated.
Statistical analysis
A descriptive analysis of the study variables will be carried out, providing absolute and relative frequencies for categorical variables, and mean and standard deviation (or median and interquartile range for variables with non-normal distributions, according to the Shapiro-Wilk test) for continuous variables. For bivariate analysis (which involves comparing people with study outcomes with people without) the χ2 test, or Fisher’s Exact test if applicable, will be used to compare categorical variables. The Student’s t-test or analysis of variance will be applied to compare two or more groups in continuous variables with normal distributions. In instances of non-normal distributions, the non-parametric Mann-Whitney U test will be used for comparing two groups, while the Kruskal-Wallis test will be employed for comparing more than two groups.
The information extracted from data analysis will include adjusted risk estimates and 95% condidence intervals (CIs). The predictions of study outcomes risk will be based on multivariate Cox proportional hazards regression models. Receiver operating characteristic curves and area under the curve will be used to evaluate the predictive value of the models. The analysis of time to event will be conducted using Kaplan-Meier curves.
Multiple imputation will be performed if there are more than 5% missing values in any variable. Follow-up losses, as well as the reasons for each loss, will be reported and analysed.
The statistical package R (R Foundation for Statistical Computing, Vienna, Austria; version R V.3.4.3 for Windows) will be used for all analyses. Statistical significance shall be established at p<0,05.
Patient and public involvement
The research team has not considered patient or public involvement in the design of the present protocol. However, it is planned to share the main results and, in particular, their possible applicability with patient associations and elderly representatives to define the best strategy for disseminating and transferring to clinical practice these results.
Ethics and dissemination
This study protocol has been reviewed and approved by the Independent Ethics Committee of the Foundation University Institute for Primary Health Care Research-IDIAP Jordi Gol, expedient file 22/090-P.
The authors plan to publish the results from this study in open-access international and national journals as well as to present them at scientific congresses, seminars and meetings. Additionally, it is planned to explain the results and their clinical applicability to the general public in different scientific dissemination events.
Discussion
There is a need for interventions that enable early diagnosis and management of AF to mitigate the social burden resulting from both the condition itself and its associated diseases or complications. These interventions should also take into account and aim to minimise inequities. It has been demonstrated that implementing interventions during the early stages of the condition with effective strategies could yield significant benefits for healthcare systems.3 45–47
Numerous methods have been proposed for detecting undiagnosed AF. However, persistent barriers exist that go beyond individualised person-centred interventions and health promotion efforts.41–48
The experts at the consensus conference recommend systematic screening for all individuals aged ≥75 years and suggest that systematic screening may also be considered for people aged 65–74 years who have additional risk factors, elevated biomarkers or a positive alert from digital devices.43 44 49 Furthermore, the first practical guide on the use of digital devices for monitoring heart rhythm and frequency provides clarification on which technologies to employ and under what circumstances.50 Nevertheless, despite clear recommendations in the European Stroke Organisation guidelines following a stroke episode, significant gaps in primary detection and ambulatory monitoring persist.51
The results of this study could provide new information about the added benefits that the combination of different strategies, each of which has individually shown efficacy, can bring in the early diagnosis of AF. It is also expected that this early detection will be followed by appropriate therapeutic changes, resulting in a reduction in the incidence of both ischaemic stroke and cognitive impairment or dementia. Nevertheless, if this combination of strategies proves to be effective, as anticipated, its cost-effectiveness should be evaluated through appropriate studies before being integrated into standard clinical practice.
Ethics statements
Patient consent for publication
Acknowledgments
We are grateful to the clinical and administrative staff of the participating PHC centres. We thank the Primary Care Research Institute-IDIAP Jordi Gol, the Pere Virgili Institute for Health Research (IISPV) and the biobank of the Vall d’Hebrón Research Institute (VHIR) for collaborate in this project.
References
Footnotes
Collaborators The complete list of PREFA-TE study investigators is: A. Panisello-Tafalla, J. Clua-Queralt (Primary Health Care [PHC} Centre Tortosa est); A. Grau-Roche, T. Forcadell-Arenas, KV. Yañez-Cabrera (PHC Centre Tortosa oest); E. Muria-Subirats, EM. Lloren-Insa; MJ. Forcadell-Peris (PHC Centre Amposta); G. Batlle-Escolies (PHC Centre L’Aldea); S. Sancho-Maigi (PHC Centre Camarles); T. Miralles Saura (PHC Centre L’Ampolla); A. Quilez-Recio, G. Lucas-Noll, C. Calduch-Noll (PHC Centre La Rapita); C. de Diego-Cabanes, A. Diaz-Vilarasau (PHC Centre Salou); D. Ribas-Seguí, C. Molins-Rojas (PHC Centre Sant Pere i Sant Pau); D. Gentille-Lorente, I. Lechuga Duran (Cardiology Department; Verge de la Cinta Hospital, Tortosa); M. Guillen-Marzo, LA. Anmad-Shihadeh, A. Bardají-Ruiz (Cardiology Department, Joan XXIII Hospital. Tarragona); JA. Izuel-Navarro, AM. Viveros-Castaño (Institute of Diagnostic Imaging [IDI}Terres de l’Ebre); M. Rovira-Beltran, E. Salvado-Geli, A. Samitier-Pastor (IDI Camp de Tarragona); P. Delgado-Martínez, A Penalba-Morenilla (VHIR, Barcelona)
Contributors J-LC-E conceived the study and got the funding; J-LC-E, FMM-L, EMS-G and MP-M provided the methodology; AH-P wrote and prepared the original draft; J-LC-E, EMS-G and FMM-L reviewed and edited the writing. All authors have read and approved the final manuscript.
Funding This study has been funded by the Department of Health of the Generalitat of Catalonia in its Strategic Plan in Research and Innovation in Health (PERIS), on the 2021 call (expedient file SLT/21/000027).
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.