To develop a prediction model for ICD-therapy in a study population of ICD and CRT-D patients using a multimodal dataset containing clinical and historical data from electronic health-records, remote monitoring-data, accelerometer-derived data and…
ID
Source
Brief title
Condition
- Cardiac arrhythmias
Synonym
Research involving
Sponsors and support
Intervention
Outcome measures
Primary outcome
The main study endpoint is the occurrence of inappropriate or appropriate
ICD-therapy between baseline and 12-months follow-up.
Secondary outcome
Time to appropriate/successful ICD-therapy (shock, ATP);
Cumulative incidence of ICD-therapy, both inappropriate and appropriate therapy
Time to inappropriate ICD-therapy (shock);
Incidence of unnecessary ICD-therapy;
Change in accelerometer-derived physical activity from baseline until 12-months
follow-up by activity type data related;
Change in accelerometer-derived sleep behavior from baseline until 12-months
follow-up for different sleep-related metrics;
Number of unique AF episodes from baseline until 12-months follow-up
Cumulative duration of AF during the observation period
Change in device-derived (D-PA) physical activity from baseline until 12-months
follow-up.
Mean of autonomics (night heart rate, heart rate variability) during 12-months
follow-up
Mean ventricular rate during episodes of AF.
Incidence of and time to hospitalization (cardiovascular events, heart
failure), mortality (all-cause, arrhythmic death, cardiovascular death,
unexplained) and Major adverse cardiac events (MACE);
Functional change from baseline onwards per NYHA functional classification;
Incidence of ICD-related complications (device infection, lead dislocation,
generator-related complications);
Comparison of patient-reported health-related quality of life at baseline and
at 12-months follow-up between patients with ICD-therapy and patients free of
ICD-therapy.
Background summary
Previous trials have identified several clinical and demographic predictors
for life-threatening ventricular tachyarrhythmias (VTAs) and appropriate
therapy delivered by an implantable cardioverter-defibrillator (ICD) or ICD
with cardiac resynchronization therapy (CRT-D). The introduction of the concept
of Artificial Intelligence (AI) as a new method of analyzing large datasets
enables the development of new, personalized prediction models for VTA
occurrence aside from the current prediction models. In 2019, Shakibar and
collegues devolped a machine learning (ML) model for the prediction of
electrical storms. The ML random forest performed significantly better than
logistic regression (p<0.01), achieving a test accuracy of 0.96 and an area
under the curve (AUC) of 0.80 (vs. an accuracy of 0.96 and an AUC of 0.75
achieved with logistic regression). The percentage of ventricular pacing and
the daytime activity were the most relevant variables in the prediction model.
Study objective
To develop a prediction model for ICD-therapy in a study population of ICD and
CRT-D patients using a multimodal dataset containing clinical and historical
data from electronic health-records, remote monitoring-data,
accelerometer-derived data and study participant-reported outcomes. The
incidence of ICD-therapy (composite endpoint of appropriate and inappropriate
therapy) is the dependent variable of the prediction model. Specific metrics
and outcomes will be obtained from these data-sources, selected by considering
clinical reasoning (i.e., commonly applied in clinical practice), relevant
literature (e.g., a systematic review of the literature), and/or knowledge of
experts in the field.
Study design
The study involves a cohort of 400 patients who have received an ICD or CRT-D.
Patients will be recruited in both hospitals AUMC and RIGS; 200 study
participants will be recruited in each hospital. Study participants will be
recruited during the same time period from December 2020 onwards. The
enrollment of study participants will be ongoing for a duration of an estimated
2 years and includes 12 months of follow-up (period 2020-2023), depending on
the patient adherence to the wearable. Enrollment in the study does not
interfere with the standard care for ICD or CRT-D study participants. There
will be four main data sources 1) Clinical data will be collected
retrospectively from the electronic health records. Clinical event data will be
collected prospectively (e.g. MACE, mortality, hospitalization), 2) External
accelerometer (wearable) data for the duration of 12 months 3)Remote monitoring
data including device programming settings, appropriate and inappropriate ICD
shocks, anti-tachycardia pacing (ATP) therapy and intracardiac electrograms
will be collected and added to the data set 4) Patient-reported data (e.g.
symptoms such as chest pain, palpitations and dyspnea, weight) will be obtained
using diaries 5) Questionnaires related to the health-related quality of life
and the ICD.
Study burden and risks
We assess the burden of this study for the patients to be low. They will wear
the wearable acclerometer for ultimately 365days, which is comparable to
wearing a watch. Besides, patients are alllowed to remove the wearable in case
of special activities or occasions, for instance in case they are planning to
do a contact sport where the wearable could do harm. Also, patients are allowed
to remove the wearable for a specific period in time, for instance when they
are going on a holiday. Patients will be asked to fill out questionnaires at
three moments, these questionnaires are related to their quality of life and
the ICD-device. Patients are not exposed to intimate or private topics in these
questionnaires.
The risk of this study is very low. Patients are not exposed to risks during
the study. No intervention is performed.
Meibergdreef 9
Amsterdam 1105AZ
NL
Meibergdreef 9
Amsterdam 1105AZ
NL
Listed location countries
Age
Inclusion criteria
- Patients >=18 years old and have undergone ICD or CRT-D implantation for
either primary or secondary prevention less than 5 years prior to enrollment;
- Are participating in a remote monitoring program at AUMC or RIGS;
- Having received appropriate or inappropriate ICD therapy or ventricular
arrhythmias in the last 8 years prior to enrollment.
Exclusion criteria
Unwilling to participate;
Study participants with a life expectancy of less than one year;
Study participants with circumstances that prevent follow-up (emigration,
change of hospital for follow-up);
Study participants who are unable to wear the GENEAactiv wrist-band (e.g.
allergic to the material);
Clinically unstable study participants;
NYHA-class IV;
Study participants unable to complete a questionnaire;
Does not understand the local language (Dutch or Danish);
Serious physical disability (e.g. wheelchair-bound);
A planned ablation for ventricular tachycardia (VT;
Significant movement disorder (i.e. hemiplegia or Parkinsons disease or
similar)
Design
Recruitment
Followed up by the following (possibly more current) registration
No registrations found.
Other (possibly less up-to-date) registrations in this register
No registrations found.
In other registers
Register | ID |
---|---|
CCMO | NL75308.018.20 |