How to diagnose atrial fibrillation and set up a screening programm
- 1. Disclosures: none Pasquale Vergara MD, Ph-D ARITMIE: Programma di diagnosi
- 2. AF Challenge Growing number of patients 14–17 million AF patients are anticipated in Europe by 20301 120,000–215,000 newly diagnosed patients per year1 Poor outcomes 1.5 to 2-fold increase of all-cause mortality1 Increased morbidity, including heart failure and stroke1 High cost of care €18 billion overall annual costs to treat AF within UK/DE/FR/SP/IT2 Over 50% of healthcare costs are due to hospitalisations2,3 1. 2016 ESC Guidelines. Eur Heart J. 2016;37(38):2893-2962. 2. Mantovani L, et al. The burden of atrial fibrillation in the more populated European countries: perspectives from the GARFIELD-AF registry. Abstract presented ESC Congress 2017.
- 4. ⇝ Clinical AF AF diagnosis » Symptomatic or asymptomatic AF that is clearly documented by an ECG (12-lead ECG or other ECG devices). » The minimum duration to establish the diagnosis of clinical AF for ambulatory ECG is not clear and depends on the clinical context. Periods of 30 s or more may indicate clinical concern, and trigger further monitoring or risk stratification for thromboembolism. Gelder European Heart Journal 2024; 00, 1–101 ⇝ Device-detected AF » asymptomatic episodes of AF detected on continuous monitoring devices. These include implanted cardiac electronic devices, for which most atrial high-rate episodes may be AF, as well as consumer-based wearable monitors. Confirmation is needed by a competent professional reviewing intracardiac electrograms or an ECG-recorded rhythm. Device-detected subclinical AF is a predictor of future clinical AF
- 5. ⇝ All patients »Medical history • to determine AF pattern, relevant family history, comorbidities, to assess risk factors for thromboembolism and bleeding »12-lead ECG »Assess symptoms and functional impairment »Patient-reported outcomes »Blood tests: full blood count, kidney function, serum electrolytes, liver function, glucose/HbA1c, thyroid function »Transthoracic echocardiography Diagnostic work-up for patients with AF 2024 ESC Guidelines Recommendations for diagnostic evaluation in patients with new atrial fibrillation Level Class Recommendations C I A transthoracic echocardiogram is recommended in patients with an AF diagnosis where this will guide treatment decisions.
- 6. 2024 ESC Guidelines for the management of atrial fibrillation (European Heart Journal; 2024 – doi: 10.1093/eurheartj/ehae176) p for patients with atrial fibrillation Selected patients Ambulatory ECG monitoring for assessing AF burden and ventricular rate control Exercise ECG to evaluate rate control or effects of Class IC antiarrhythmic drugs AF pattern, relevant ties, and to assess risk and bleeding Further blood tests for investigation of cardiovascular disease and refinement of stroke/bleeding risk (e.g. NT-proBNP, troponin) Transoesophageal echocardiography for left atrial thrombus and valvular disease assessment nal impairment Coronary CT, angiography, or ischaemia imaging for suspected CAD patient-reported CMR for evaluation of atrial and ventricular cardiomyopathies, and to plan interventional procedures kidney function, serum cose/HbA1c, and Brain imaging and cognitive function assessment for cerebrovascular disease and dementia risk hy where this will guide ns Diagnostic work-up for patients with AF 2024 ESC Guidelines
- 7. Screening tools for AF
- 8. SmartWatch ECG recording modalities PHOTOPLETHYSMOGRAPHY (PPG) Green LED lights paired with light sensors detect blood volume pulsatility; this enables the creation of tachograms (plot of the time between heartbeats) to assess heart rate variability
- 9. Apple Heart Study Perez NEJM 381;20 8 months period 117 days
- 10. Apple Heart Study Perez NEJM 381;20 The algorithm was designed to minimize false positive findings The algorithm was not designed to detect short episodes of atrial fibrillation, and participants with a low burden of atrial fibrillation could have been missed.
- 12. ⇝ 2 coprimary outcomes » AF >30’’ on ECG patch monitoring in a participant who received an irregular pulse notification » simultaneous AF on ECG patch during intervals with irregular tachogram. Apple Heart Study Perez NEJM 381;20 Irregular Pulse Notifications, According to Age and Sex
- 13. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Apple Heart Study #detections Perez NEJM 381;20 AF 1489 tg 2089 irregular tachograms other 600 tg PACs 462 tg PVCs 96 tg AT 228 tg 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Sinus arrhy 28 tg The positive predictive value of an individual tachogram was 0.71 PPV of an irregular pulse notification was 0.84 Algorithm: multiple irregular tachograms before triggering a notification
- 14. Apple Heart Study #watched by Apple Campion NEJM 381;20 “Over 400,000... enrolled in the study, not because of any health problem but because they were curious and wanted reassurance”
- 15. Apple Heart Study 76% of notified participants actively sought
- 16. ⇝ AIMS: » Compare waveform similar to a Lead I ECG from a standard 12- Lead ECG » evaluate SR /AF rhythm classification algorithm ⇝ Participants » 301 with known AF » 287 with no known cardiac rhythm abnormalities ⇝ Methods » 3 single-lead ECGs recorded using the ECG app simultaneously with » 3 12-Lead ECGs using a standard FDA-cleared ECG device Clinical evaluation of 1 Lead ECG from Apple Watch Apple “Using Apple Watch® for Arrythmia Detection” Whitepaper, December 2018.
- 17. Clinical evaluation of 1 Lead ECG from Apple Watch Parameter Value Lower Confidence p-value Reference ECG : AF (n) 240 App: AF 236/240 (98.3%) App: SR 4/240 (1.7%) Sensitivity 236/240 (98.3%) 95.8% < 0.0001 Reference ECG : SR (n) 239 App: AF 1/239 (0.4%) App: SR 238/239 (99.6%) Specificity 238/239 (99.6%) 97.7% < 0.0001
- 18. Apple Watch® FDA approval Apple Watch® has been designated a Class II device and has received FDA clearance for its irregular rhythm notification feature and ECG app (Sept. 11, 2018; DEN180042 , DEN180044)2a,b FDA clearance2a,b specifies that the irregular rhythm notification feature and ECG app are “not intended to provide a diagnosis” and “not intended to replace traditional methods of diagnosis or treatment” 2a,b In addition, the irregular rhythm notification feature is “not intended for use in individuals previously diagnosed with AFib” and “not intended to provide a notification on every episode of irregular rhythms suggestive of atrial fibrillation (AFib)”2a
- 19. Who should pay for reviewing the ECGs from the Apple Watch Ellwood British Journal of General Practice 2019 A GP from UK wrote: “The UK National Screening Committee has not recommended systematic population screening for AF. But … ECG screening for AF has already become a reality with the ECG app on the Apple Watch.”
- 20. ⇝ 325 million smart watches sold in 2016 alone ⇝ 13% of the US population owning smartwatches ⇝ 40% of the public showing interest in getting one The SmartWatches invasion ⇝ The use of these wearables needs to be effective in diagnosing subclinical AF, ⇝ with false positives increasing health costs by unwarranted ER visits or anxiety to the patient, ⇝ or with false negatives missing the diagnosis which may lead to morbidity or mortality Total: 53.3 millions of units in 2 years 0.52% received irregular pulse notifications = 277’160 patients with irregular pulse notifications = 230’042 true AF detected + 47’117 “false” AF
- 21. 2024 Guidelines for atrial fibrillation
- 22. AF-CATCH
- 23. LOOP Study Svendsen Lancet 2021; 398: 1507-16 If AF lasting >=6 min was detected, initiation of oral anticoagulation was recommended
- 24. LOOP Study Svendsen Lancet 2021; 398: 1507-16 not all screen-detected atrial fibrillation should merit anticoagulation
- 26. AI ECG algorithm for identification of AF in SR Attia, Lancet 2019; 394: 861–67
- 27. AI ECG BNP CHA2DS2- VASc Papanicolaou cervical cancer -2.77555756156289E-17 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.87 0.65 0.65 0.7 AUC Screening techniques comparison
- 28. Waveforms Kurshid Circulation. 2022;145:122–133 waveform of a random sample of 1000 individuals in the BWH test set with low predicted AF risk (ie, 5-year AF risk <2.5%, green) vs the median waveform of a random sample of 1000 individuals in the BWH test set with high predicted AF risk (ie, 5-year AF risk >5%, red)
- 30. Take home messages ⇝ Episodi asintomatici di FA rilevati da dispositivi con monitoraggio continuo (smartwatches) richiedono la conferma da un professionista. ⇝ Wearables possono essere utili per diagnosticare FA » falsi positivi aumnento costi ed ansia » falsi negativi aumento morbidità or mortalità ⇝ Screening pazienti anziani e con fattori di rischio ⇝ Algoritmi di AI possono identificare pazienti a rischio di FA utilizzando un semplice ECG a 12 derivazioni