Multimodality prediction of success of atrial fibrillation rhythm control strategy.

Atrial fibrillation (AF) affects 1-2% of the population, and its prevalence is
estimated to double in the next 50 years. AF is related with increased
mortality and morbidity.
There are different types of AF based on presentation and duration of the
arrhythmia: paroxysmal, persistent, long-standing persistent and permanent AF.
Paroxysmal AF is self-terminating within 7 days. Persistent AF lasts longer
than 7 days or requires termination by cardioversion. Long-lasting persistent
AF is present when AF lasts longer than 1 year and a rhythm-control strategy is
adopted. Permanent AF is present when the arrhythmia is accepted by patient and
physician.1
Long-term management of the arrhythmia can consist of a rate-control or
rhythm-control strategy. In a rate-control strategy the goal is to control the
ventricular rate adequately but not to restore sinus rhythm (SR). In a
rhythm-control strategy the goal is to restore SR. Several large trials
compared rhythm and rate-control strategy 2-3. The AFFIRM trial found no
difference in all-cause mortality between the different strategies.2 The RACE
trial found rate control not inferior to rhythm control for prevention of
cardiovascular mortality and morbidity.3 Therefor the ESC guidelines advocate a
patient tailored therapy.1 The decision to choose between a rhythm or
rate-control strategy is made by the patient and his cardiologist. Usually
symptoms are the main reason to choose a rhythm-control strategy. A
rhythm-control strategy can consist of anti-arrhythmic drug and direct current
cardioversion (DCC). Anti-arrhythmic drugs can have potentially harmful
side-effects and a DCC is a potentially dangerous procedure due to
complications during anaesthesia. Therefore symptoms should not be the only
reason to choose a rhythm-control strategy but also predictors for maintenance
of SR after successful DCC should be taken into account. In patients unlikely
to maintain SR a rate-control strategy might be a better solution.
Several predictors for maintenance of SR after DCC or ablation have been
identified. Clinical predictors such as younger age and a shorter duration of
AF predict maintenance of SR.4-7 Several parameters on a 12 lead ECG have been
identified that predict maintenance of SR.8-15 Atrial dimension and function
measured by echocardiography can predict maintenance of SR. Biomarkers of
inflammation and cardiac specific markers have been investigated and show to be
predictive in some studies.16-25 Recently common variants have been identified
in patients with AF 26-28, presence of some of these variants on chromosome
4q25 predict recurrence after catheter ablation for AF.29-30 Recently a study
showed these single-nucleotide polymorphisms (SNPs) to be predictive for
recurrence after DCC in patients with persistent AF.31
To find patients likely to respond to rhythm control strategy we want to build
a multimodality prediction model in patients with AF scheduled for DCC in order
to restore SR. This model will include several known predictors stated
previously and some additional diagnostic tools. An additional diagnostic tool
included in this study is a transoesophageal electrocardiogram (TE-ECG).
Preliminary results using ECG analysis software on a standard 12 lead ECG show
a possible important role for left atrial complexity. TE-ECG recordings
correspond to recordings within the left atrium.32 TE-ECG has been used in this
hospital before, preliminary results confirm its correlation with left atrial
complexity.
We will also record a 5 minute ECG with 5 additional surface leads. A previous
study advocated at least 4 additional leads which contain the most independent
information about the atrial electrocardiogram.33 In a previous study conducted
in this hospital we performed body surface potential mapping (BSPM) in this
patient population. Due to the promissing results (unpublished) we will also
use the BSPM in this study.
In this study we aim to build a model to predict the maintenance of SR after
successful DCC incorporating standard predictors such as echocardiographic
parameters, clinical parameter and 12-lead ECG parameters and new parameters,
including BSPM, TE-ECG and SNPs. Using this novel integrated prediction model
will help in the identification of patients who will maintain SR after
cardioversion. Another innovative aspect of our study is the use of daily
rhythm monitoring using a versatile device (MyDiagnostick®). The clinical
impact this study will have is that it helps the cardiologist to make a patient
tailored strategy in the treatment of AF as advocated by the guidelines and to
prevent exposing patients not likely to maintain SR to the potentially harmful
side-effect of a rhythm-control strategy.

Primary objectives

To show that a model incorporating the novel predictors BSPM, TE-ECG and SNPs
is superior to existing models in predicting maintenance of sinus rhythm after
successful DCC in patients with persistent AF.

Secondary objectives

To assess the predictive value of new predictors (the TE-ECG, BSPM and SNPs)
for maintenance of SR after successful DCC separately.
To assess the predictive value of new predictors (the TE-ECG, BSPM and SNPs)
for time to recurrence of AF.
To assess the predictive value of predictors with respect to complete failure
of cardioversion.
To assess the predictive value of predictors with respect to cardiovascular
morbidity.

Patients with persistent AF scheduled for DCC will be included in this study.
At inclusion clinical characteristics of the patients will be recorded.
Clinical characteristics include medical history as well as physical
examination before DCC. Physical examination will focus on vital signs and
signs of heart failure.
Patients will receive an echocardiography, if not performed within the last six
months. Echocardiography is standard care in patients with AF. Echocardiography
will be performed to evaluate left ventricular systolic and diastolic function,
valve function as well as left and right atrial dimensions.
At the day of cardioversion 24.5 ml additional blood will be drawn for
biomarkers and DNA-analysis (if additional consent is obtained). This will be
divided in 10 ml EDTA, 10 ml serum and 4,5 ml citrate. This will be
incorporated in the regular blood examination before cardioversion, no
additional venous puncture is needed. The routine blood examination focusses on
potassium, renal function and coagulation. Additional blood examination
focusses on inflammation (i.e. serum high sensitivity C-reactive protein,
interleukin-6, tumor necrosis factor alpha, transforming growth factor beta-1),
cardiac ischemia (Hs-TnT), cardiac specific biomarkers (ANP, NT-proBNP) and
fibrosis. These blood samples will be processed and stored at -80 *C. The
additional analysis will be done at a later stage, this means that these
parameters won*t be used for patient treatment. These parameters are described
in detail in table 3.

At the 1 month visit an additional 24.5 ml blood will be drawn, the same as at
the day of cardioversion. This additional venous puncture is needed to study
the kinetics of biomarkers in an AF population. Furthermore, we will look more
detailed into the dynamics of proteomics (i.e. change in composition due to the
electrical and possible structural remodelling that occurs due to the
cardioversion). There is more and more evidence for a role of proteomics in the
prognoses of a successful rhythm strategy in patients with AF.43 This
additional blood collection is for both the patients in AF and in SR at that
time.

DNA-analysis will focus on SNPs identified in a meta-analysis of genome-wide
association studies associated with AF.34 These SNPs are located near different
genes involved in pacemaking activity, single transduction and cardiopulmonary
development. Besides theses SNPs 2 other SNPs have been identified that predict
recurrence of AF after ablation or DCC.29-31 The SNPs that will be determined
in the patients and closest genes are represented in table 4. Patients won*t
receive the results of the DNA-analysis about these SNPs, in line with the METC
azM/UM protocol about DNA-research. This DNA-analysis will not have any
consequences for the patients regarding future insurances.

Patients will routinely receive a 10 second 12 lead ECG before DCC. In the
time needed for the regular blood analysis we will perform a 5 minute 17 lead
ECG, TE-ECG and Tissue velocity imaging (TVI). In total this will take about 30
minutes.
A BSPM is a 184-lead surface ECG containing unipolar surface leads on the front
and the back of the thorax. These BSPMs have been used in a previous study to
predict recurrence after DCC and success of pharmacological cardioversion for
AF. This 184-lead ECG will cause no additional discomfort for the patient
compared to a regular 12-lead ECG.
The TE-ECG is a 7 French catheter containing 4 unipolar leads for oesophageal
ECG monitoring. This catheter needs to be swallowed and held into place for 2
minutes. After these 2 minutes the catheter will be removed. The swallowing of
the catheter is discomforting for the patient.
TVI is an echocardiographic technique used for noninvasive quantification and
timing of local myocardial wall motion. TVI can be used to determine AF cycle
length (AFCL) and the velocity of the local atrial fibrillatory wall motion
(AFV).35 There is a strong correlation between the AFCL measured by TVI and the
AFCL determined during electrophysiology studies.36 Unpublished data show that
both AFCL and AFV measured by TVI can predict recurrence of AF after DCC for
persistent AF.
During DCC continuous rhythm monitoring will take place to assess initial
therapy success. After cardioversion another 12-lead ECG will be made to assess
success after one hour, this is part of routine clinical practice.
Patients in whom SR sustains for at least one hour will receive a
MyDiagnostick. The MyDiagnostick has been used previously in the study with
study number 124038. The MyDiagnostick is a stick patients need to hold in
their hands for one minute. After this minute the stick will indicate whether
AF is present. In the PERFORMAF study the MyDiagnostick had a sensitivity of
100% and a specificity of 95.7% in decting AF, this study has been submitted.
Patients are asked to hold the stick during one minute once daily. When the
Mydiagnostick indicates AF recurrence patients will receive a 12 lead ECG to
confirm AF. To minimize problems with compliance and to standardise follow-up,
patients will receive a 12 lead ECG after 1 month (standard care) and 6 months.
When the main study endpoint is met thus AF confirmed by 12 lead ECG or
patients completed the initial 6 months follow-up the daily rhythm follow-up
off the study is completed. If the main study endpoint is not met within the
first 6 months an additional rhythm follow-up will take place 12 months after
ECV. After a maximum duration of 12 months the study is completed. Further
follow-up as indicated by the patients cardiologist. The additional 24,5 ml of
blood drawn at one month is for both the patients in AF and SR at that time.
The study ends after this time point if the patients are in AF. If patients are
still in SR after 12 months the study will end also for those patients.

Daily rhythm monitoring is used to detect recurrence of AF as soon as possible.
Recurrence of AF will be detected at an earlier stage and therefor additional
rhythm control treatment can be initiated earlier, if necessary. This treatment
strategy will be discussed with the referring cardiologist, further follow-up
as indicated by the referring cardiologist.

Patients in whom SR sustains for at least one hour will receive a
MyDiagnostick. The MyDiagnostic is a stick patients need to hold in their hands
for one minute. After this minute the stick will indicate whether AF or SR is
present. Patients are asked to hold the stick during one minute once daily with
a maximum period of 6 months. If the stick indicates AF patients will have to
make 1 extra hospital visit for an ECG. Furthermore, an additional rhythm
follow-up moment at 12 months will be performed.
The investigations at the day of cardioversion will take about 30 minutes.
The recording of a TE-ECG is often being used in other studies without
reporting of complications. We will use a commercial available electrode,
Osypka TO 4-electrode 7F catheter CE1275. As part of another study (METC
03-071) we did this examination in 30 patients without any problems. As an
extra safety measurement we will exclude patients with known oesophageal
disease. In large studies with these catheters no complications occurred.
Patients will receive an additional venous puncture after one month to draw
additional blood.