Smartphone Based-PPG Assessment of REconversion in Persistent Atrial Fibrillation

Cancellation of planned admission for electrical cardioversion.

Defined as the rate of patients having at least one ECG measurement in sinus rhythm before planned cardioversion.

Cost analysis of systematic PPG monitoring prior to cardioversion.

AF or Aflutter recurrence rate during 30 days post-cardioversion

AF or Aflutter recurrence rate during 1 year follow-up.

Time to AF or Aflutter recurrence.

Electrical cardioversion rate during 1 year follow-up.

The AI-enabled ECG-derived atrial fibrillation risk score will be calculated from a standard 12-lead ECG recorded in sinus rhythm. The score ranges from 0 to 100, with higher scores indicating a higher estimated risk of atrial fibrillation. Recurrence will be defined as documented atrial fibrillation or atrial flutter occurring after the planned admission date. Cox proportional-hazards regression will be used to estimate the hazard ratio for time to first AF or atrial flutter recurrence per 10-point increase in the AI-derived risk score.

The AI-enabled ECG-derived atrial fibrillation risk score will be calculated from the most recent standard 12-lead ECG recorded in sinus rhythm. The score ranges from 0 to 100, with higher scores indicating a higher estimated risk of atrial fibrillation. Spontaneous reconversion will be defined as documented restoration of sinus rhythm before the planned admission for electrical cardioversion. Logistic regression will be used to estimate the odds ratio for spontaneous reconversion per 10-point increase in the AI-derived risk score.

Prediction analysis for spontaneous conversion to sinus rhythm before admission.

Prediction analysis for AF/Aflutter recurrence during one year follow-up.

Patient-reported AFEQT overall score and domain scores at baseline and 30-day follow-up.

The accuracy of smartphone-based PPG monitoring in detecting spontaneous reconversion of AF in patients planned for electrical cardioversion will be evaluated by calculating key performance metrics such as sensitivity, specificity, positive predictive value, and negative predictive value. For each participant all AF, Aflutter and sinus rhythm episodes identified by the PPG system automatic algorithm, as well as blinded cardiologist's interpretation will be compared against those recorded by a simultaneous single lead or 12 lead ECG. The concordance between these methods, as determined by two independent, blinded cardiologists will assess the PPG system's ability to accurately classify AF, Aflutter and sinus rhythm events.

Defined as the rate of patients having at least one ECG measurement in AF/Aflutter after experiencing spontaneous reconversion during the monitoring period.

This will be reported as the amount of simultaneous AF/AFlutter ECG and sinus rhythm PPG recording pairs divided by the total AF/AFlutter ECG recordings.

The rates of each arrhythmia will be reported.

This will be defined as the rate of patients having at least one ECG measurement showing a different atrial arrhythmia than the atrial arrhythmia at inclusion during the monitoring period.

Heart rate will be measured simultaneously using smartphone photoplethysmography and a single-lead ECG. For each paired measurement, the absolute difference between the heart rate reported by the PPG measurement and the heart rate reported by the ECG measurement will be calculated as:

Absolute heart-rate difference = PPG heart rate - ECG heart rate

The mean absolute difference across all evaluable paired measurements will be reported.

This will be reported as the percentage of measurements where symptoms have been indicated, the percentage of measurements where no symptoms have been indicated, the percentage of AF/Aflutter measurements where

Correlation analysis of clinical parameters and symptomatic AF/Aflutter recording rate as well as the SRC index.

Self-reported alcohol consumption will be assessed using a study-specific questionnaire at baseline, on admission for electrical cardioversion, and at the 30-day follow-up. At each assessment, participants will report the number of standard alcoholic units consumed during a typical week. One standard alcoholic unit will be defined as (one glass of wine, one beer, or one shot/measure of spirits). Alcohol consumption will subsequently be compared between participants with and without documented recurrence of atrial fibrillation or atrial flutter within one year.

The frequency of alcohol consumption will be assessed using a study-specific questionnaire at baseline, on admission for electrical cardioversion, and at the 30-day follow-up. At each assessment, participants will report the number of days during a typical week on which they consume alcohol. Alcohol-consumption frequency will subsequently be compared between participants with and without documented recurrence of atrial fibrillation or atrial flutter within one year.

Self-reported caffeine consumption will be assessed using a study-specific questionnaire at baseline, on admission for electrical cardioversion, and at the 30-day follow-up. At each assessment, participants will report their consumption of caffeinated beverages during a typical week. A caffeinated beverage will be defined as a cup of coffee (one standard cup/mug of caffeinated coffee) or a caffeinated drink other than coffee (one cup/glass/can/bottle of tea, cola, energy drink, etc).

Caffeine consumption will subsequently be compared between participants with and without documented recurrence of atrial fibrillation or atrial flutter within one year.

The frequency of caffeine consumption will be assessed using a study-specific questionnaire at baseline, on admission for electrical cardioversion, and at the 30-day follow-up. At each assessment, participants will report the number of days during a typical week on which they consume caffeinated beverages. Caffeine-consumption frequency will subsequently be compared between participants with and without documented recurrence of atrial fibrillation or atrial flutter within one year.

Alcohol consumption will be assessed at enrollment using a study-specific questionnaire. Participants will report the number of standard alcoholic units consumed during a typical week. One standard alcoholic unit will be defined as (one glass of wine, one beer, or one shot/measure of spirits). Spontaneous reconversion will be defined as documented restoration of sinus rhythm before the planned admission for electrical cardioversion. Logistic regression will be used to estimate the odds ratio for spontaneous reconversion per additional alcoholic unit consumed per week.

Frequency of alcohol consumption will be assessed at enrollment using a study-specific questionnaire. Participants will report the number of days during a typical week on which they consume alcohol. Spontaneous reconversion will be defined as documented restoration of sinus rhythm before the planned admission for electrical cardioversion. Logistic regression will be used to estimate the odds ratio for spontaneous reconversion per additional day of alcohol consumption per week.

Caffeine consumption will be assessed at enrollment using a study-specific questionnaire. Participants will report the number of standard servings of caffeinated beverages consumed during a typical week. A caffeinated beverage will be defined as a cup of coffee (one standard cup/mug of caffeinated coffee) or a caffeinated drink other than coffee (one cup/glass/can/bottle of tea, cola, energy drink, etc). Spontaneous reconversion will be defined as documented restoration of sinus rhythm before the planned admission for electrical cardioversion. Logistic regression will be used to estimate the odds ratio for spontaneous reconversion per additional caffeinated serving consumed per week.

Frequency of caffeine consumption will be assessed at enrollment using a study-specific questionnaire. Participants will report the number of days during a typical week on which they consume caffeinated beverages. Spontaneous reconversion will be defined as documented restoration of sinus rhythm before the planned admission for electrical cardioversion. Logistic regression will be used to estimate the odds ratio for spontaneous reconversion per additional day of caffeinated beverage consumption per week.

Adherence to smartphone-based self-monitoring will be calculated for each participant as the number of completed scheduled recordings divided by the total number of scheduled recordings during the prescription period, multiplied by 100. The mean or median participant-level adherence percentage will be reported.

Motivation will be calculated for each participant as the number of prescribed monitoring days on which at least two scheduled recordings were completed divided by the total number of prescribed monitoring days, multiplied by 100. The mean or median participant-level percentage of adherent monitoring days will be reported.

Rescheduled electrical cardioversion rate after initial cancellation