9 / 74
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 5, September/October 2018
AFRICA
271
with rate control and there was a marked improvement in left
ventricular function. The remaining three started with good left
ventricular function that deteriorated due to poor rate control,
thereby necessitating more drastic measures to achieve rate
control. The mean age of our cohort of patients with heart failure
was 45 years (range 28–73 years), which is less than the mean age
of 74 years that was reported in the Jahingir
et al
. study.
4
Ten percent of patients (
n
=
4) developed thromboembolic
complications. One patient was a male who had developed
hypertension over time. He consequently developed an ischaemic
stroke that necessitated anticoagulation therapy with warfarin.
This highlights the importance of constant vigilance for
thromboembolic risk factors in lone AF patients during follow up.
The remaining three patients with thromboembolic
complications were females over the age of 60 years who
were part of the subset of patients with AF and SSS or the
bradycardia–tachycardia syndrome. They were not on any
anticoagulation therapy prior to the stroke. Intriguingly, in
the study by Rubenstein
et al
.,
27
there was a high incidence
of systemic embolisation in the SSS/AF patients, and they
recommended anticoagulation in this group regardless of age or
the CHADS
2
score. They hypothesised that the fibrillating atria
predispose to thrombus formation, and the sudden cessation of
fibrillation predisposes to a thromboembolic event. This leaves
scope for a study to assess whether young patients with frequent,
recurrent episodes of paroxysmal lone AF are at higher risk of
thromboembolism compared to the young general population
and young patients with persistent/chronic lone AF.
There were no deaths recorded in our series. This could
have been due to the relatively short follow-up period of each
individual patient (mean follow up of 7.7 years). The study by
Kopecky
et al.
2
recorded a death rate of 20% over a 30-year
follow-up period. Seventy per cent of the deaths were due to
cardiovascular causes (coronary artery disease, heart failure,
stroke and aortic aneurysms). Sixteen per cent of these deaths
resulted directly from embolic events.
2
At the GSH Cardiac Clinic, to improve symptoms, therapy is
aimed at achieving rate or rhythm control, which can significantly
improve quality of life and prevent complications such as
tachycardia-related cardiomyopathy. Timely anticoagulation
with aspirin or warfarin is instituted (if bleeding risk is low) to
prevent thromboembolic complications.
Cardioversion was attempted at least once in 29% (
n
=
12) of
patients in an attempt to achieve rhythm control. Two patients
who could afford the drugs were placed on class IC anti-
arrhythmics (flecainide/propafenone) in an effort to maintain
sinus rhythm because of severe symptoms during AF. These
agents may be used in young patients with structurally normal
hearts, however due to various constraints in the public health
sector, the use of class IC anti-arrhythmic drugs was limited.
For rate control, 60% (
n
=
25) of patients were on beta-
blockers, 29% (
n
=
12) were on digoxin and 14% (
n
=
6) on
verapamil. Often beta-blockers were used in combination with
digoxin. Rate control was considered adequate if the resting
ventricular rate was
<
80 beats/min.
If patients failed to respond to rate-controlling agents or could
not tolerate the side-effect profile of these atrioventricular nodal-
blocking agents, they were put on a class III anti-arrhythmic drug,
such as amiodarone, in an attempt to maintain sinus rhythm.
Twenty-two per cent (
n
=
9) progressed on to amiodarone. Of
these patients, 67% (
n
=
6) developed complications related to
the long-term use of amiodarone, including hypothyroidism (
n
=
2), hyperthyroidism (
n
=
2), interstitial pneumonitis (
n
=
1) and
chronic inflammatory demyelinating polyneuropathy (
n
=
1).
This emphasises the non-benign nature of this agent, and that
the decision to place patients on long-term amiodarone should
not be taken lightly. Sotalol was avoided due to its potential
QT-prolonging effect and subsequent risk of torsades des
pointes, which has been documented in patients with structurally
normal hearts.
28,29
Twelve per cent of patients (
n
=
5) had radiofrequency
ablation of the concomitant atrial flutter and one patient
had radiofrequency ablation of the atrial ectopic foci in the
pulmonary veins. This form of AF ablation became available
only late in the course of the study period.
Ten percent of patients (
n
=
4) did not respond or developed
an adverse side-effect profile on medical therapy and went
on to have an atrioventricular nodal ablation and permanent
pacemaker insertion. In patients who are refractory to medical
management, options include atrioventricular nodal (AVN)
ablation and pacemaker insertion, aso known as the ‘ablate and
pace’ strategy. It is considered an extreme form of rate control.
There is lifelong pacemaker dependency after this procedure.
A recent meta-analysis by Chatterjee
et al
.
30
compared AVN
ablation with pharmacotherapy in patients with drug-refractory
AF. Of the five studies included in the efficacy analysis, four were
randomised trials, comprising 314 subjects in total. The efficacy
analysis demonstrated that AVN ablation improved symptoms
and quality of life significantly in patients with medically
refractory disease, compared with pharmacotherapy alone.
30
TheGSHCardiacClinic policy is toassess the thromboembolic
risk using the CHADS
2
scoring system in patients who have AF
without underlying structural heart disease, and balance this
against the bleeding risk so that an anticoagulation strategy can
be tailored for each patient. Patients considered to be at high risk
for thromboembolism are those with congestive heart failure,
hypertension, age above 75 years, diabetes mellitus and previous
stroke. A CHADS
2
score of 1 is at intermediate thromboembolic
risk and aspirin is recommended. A CHADS
2
score
≥
2 is a high
thromboembolic risk and warfarin is recommended.
Limitations
Although the subjects in our study were identified retrospectively,
the history of events was analysed prospectively from the time
of diagnosis. The importance of these observational studies
is determined by the paucity of patients with lone AF in
randomised studies and their relatively short follow-up time.
The sample size was relatively small but this was a consequence
of strictly adhering to a clinical definition of lone AF in a
geographically defined population. Our follow-up time of each
individual patient was relatively short and this may have resulted
in underestimation of certain events.
Groote Schuur Hospital is a tertiary referral centre in the
public sector and as a result, there may have been a referral
bias in our cohort of patients. The number of subjects in some
subsets was small and it may be premature to extrapolate the
associations observed in this study to the general population
with AF and no underlying organic heart disease. Additionally,
these subjects were largely symptomatic, and there may be