Diagnosis of Arrhythmias in Syncope by Pocket-ECG III® or Conventional Holter

Study population: adult patients with recurrent and/or with probable cardiac origin syncopes with indication of cardiac rhythm monitoring.

Estimated sample size: COLAPS trial showed that conventional 48-hours Holter and 4-weeks external loop recorder got a diagnosis in 22% and 56%, respectively, in patients with syncope. Based on this data, to compare two matched-proportions with α-risk=0.05 and β-risk=0.2, 34 patients would be needed; to compensate a 15-20% percentage of lost patients during follow-up, it is estimated that 40 patients is the number to include in the trial to demonstrate the effect.

Variables to collect:

Baseline variables: birthdate date (month, year), inclusion date, gender, number and features of syncopes, cardiovascular risk factors, electrocardiogram, type of cardiopathy, left ventricular ejection fraction, comorbidities and treatments.

Monitoring findings: diagnostic event and diagnostic fulfilled criteria, date of diagnostic event, date of last follow-up.

Safety and adverse events: monitoring length, unexpected battery depletion, number of losses of mobile phone signal, cutaneous reactions to electrodes.

Costs: the costs of every diagnostic procedure used from inclusion to the end of follow-up will be added, based on prespecified costs of every item (Decreto 56/2014 of the Consellería de Sanidade de la Xunta de Galicia, which specifies the costs of every diagnostic or therapeutic intervention in centers depending on the Servicio Gallego de Salud).

Intervention description: patients with indication of conventional Holter monitoring will be consecutively included. After signing written informed consent, a Pocket-ECG III® system will be given to the patient. All participants will be monitored by the Pocket-ECG III® system until a diagnosis is achieved or for a maximum of two months (whatever it happens first). Everyday daily reports sent via e-mail by the manufacturer company (MEDICALgorithmics© S.A.) will be reviewed by the investigator team; in case of a diagnosis event (according to prespecified diagnostic criteria from the European Society of Cardiology), appropriate treatment will be applied.

Every patient will be his/her own control. Conventional 24-hours Holter monitoring will be simulated using data obtained by the Pocket-ECG III® system in the first 24 hours, ignoring the registries of the rest of days. Two and Three 24-hours conventional Holter strategies will also be simulated analyzing data from first 24 hours and 1-2 additional days, respectively, chosen by means of a random number creation system which will identify the days of registry to be considered. Diagnostic criteria in this virtual control group will be the same as used in the actual Pocket-ECG III® system monitoring group.

Follow-up: patients will be followed-up by two complementary manners

1. Real-time remote: Pocket-ECG III® system sends continuously the registered electrocardiogram sinal via GSM mobile phone to a central unit sited in Poland, where the signal is processed and interpreted. From this central unit, a daily report and a weekly summary report is sent by e-mail to the investigator team. If a lost in the mobile phone coverage happens, the system stores the registry in its memory until the coverage is recovered, when the whole stored registry is sent.
2. Purpose-created outpatient clinic.

Statistical methods: categorical variables will be expressed as "absolute number (percentage)" and continuous variables as "mean ± standard deviation", adding "median (interquartile range)" when the distribution of values is far from normal distribution assessed by Kolmogorov-Smirnov and Shapiro-Wilk methods. Proportion of diagnosis achieved by each strategy will be compared by McNemar for intrasubject measures or binomial exact methods, whichever applies. Economic burden of each strategy will be compared with the t-Student for intrasubject measures or T de Wilcoxon methods, whichever applies. All tests of significance will be 2-sided, with p<0.05 considered statistically significant. The analysis will be conducted using PASW Statistics version 18.0 (IBM, Armonk, New York) and EPIDAT 3.1 (provided by the Direccion Xeral de Saude Publica de la Consellería de Sanidade de la Xunta de Galicia).