Page 13 - Laurens Bosman

Aims and outline of this thesis

Our aim to find ways to improve arrhythmic risk prediction in ARVC started with a systematic
review and meta-analysis of the current literature in Chapter 3. In this study we identified
several risk factors with consistent evidence supporting their association with increased risk
of ventricular arrhythmia. These results led to our hypothesis that we could use these risk
factors combined in a multivariable prediction model to estimate the risk of ventricular
arrhythmia risk of individual patients. In order to conduct such a study, we needed to have a
large longitudinal dataset of ARVC patients, for which we had to redesign the Netherlands
ACM Registry database as described in Chapter 4.

In Chapter 5 we confirmed our hypothesis by successfully developing a risk prediction
model using the eight risk predictors we pre-selected (based on the results in Chapter 3). The
resulting model predicts the risk of a first sustained ventricular arrhythmia in ARVC patients
without a prior sustained event (i.e. primary prevention). In Chapter 6 we study the relation
between physical exercise and arrhythmic risk in ARVC patients, and explore if adding
exercise as a risk factor would improve the risk prediction model we developed. As the
prediction model in Chapter 5 was designed to be used for primary prevention patients only,
we subsequently developed a model for all ARVC patients as described in Chapter 7.
Furthermore, instead of predicting any type of sustained ventricular arrhythmia, this second
model predicts fast (>250 beats per minute) events to closer approximate the risk of SCD.

Alternative to our two prediction models, there are several risk stratification flow-charts
available from published guidelines and consensus documents. In contrast to our models,
6-8
these flow-charts are designed by expert consensus, hence their actual clinical performance
was unknown. The aim of our study in Chapter 8 was to estimate and compare the clinical
performance of these stratification flow-charts.

Moving away from prognosis, in Chapter 9 we focus on the clinical diagnosis of ARVC.
There is no single gold standard test to diagnose ARVC, instead, the diagnosis is determined
by a complex set of different tests and criteria as specified in the Task Force Criteria (TFC).
3
While many studies focus on finding ways to improve the diagnostic performance of the TFC,
no study had validated the clinical performance of the TFC as a whole in a real-life consecutive
diagnostic cohort. Therefore, in Chapter 9 we validate the performance and evaluate
strengths and weaknesses that could guide future studies regarding criteria improvements.

11