Cardiovascular Journal of Africa: Vol 25 No 6 (November/December 2014)

34 / 66

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 25, No 6, November/December 2014 284 AFRICA
hospital and six-month mortality rates in patients in whom AF
occurred.1
The mechanisms involved in the development of AF in
the cardiac surgical patient are incompletely elucidated. It is
probable that pericardial inflammation, catecholamine increase,
autonomic disharmony, atrial stretch, metabolic abnormalities,
transcellular fluid and electrolyte shifts, and neurohormonal
activation act alone or in concert in the production of AF.
These factors shorten the atrial refractory period, slowing atrial
conduction. Resultant re-entry wavelets bombard the atrio-
ventricular node to produce rapid and irregular ventricular
contraction.5
Atrial structural alterations, inherent or iatrogenic, affect
individual susceptibility to AF. It is widely accepted that the
initiation and perpetuation of AF requires a triggering factor
and an electrophysiological substrate within the atria. The
substrate is mandatory and perhaps explains why some post-
operative patients develop AF and some do not.6 Furthermore,
a genetic predisposition has been proposed in patients with the
interleukin-6 promoter gene variant.3
AF post cardiac surgery is postulated to arise in a milieu of
sympathetic hyperactivity in the post-operative period. Several
patient and procedure-related factors have been suggested to
confer increased individual vulnerability. The inflammatory
state induced by cardiopulmonary bypass, atrial incisions and
the relative ischaemia of the atrial septum when cardioplegic
solution is delivered via the coronary circulation contribute to
the complex processes involved in the generation of AF.7
The AF incidence of 5.9% in our cardiac surgical unit is
within the widely varied range of incidence reported in most
international series, between 5.5 and 65%.7,8 A meta-analysis of
24 trials estimated the incidence at 26.7%.9 A comparison of
eight studies with cohorts of 500 or more patients evaluating the
incidence and pre-operative risk factors for atrial arrhythmias
after cardiac surgery confirmed a comparable incidence between
AF post coronary artery bypass surgery (CABG) and post valve
surgery, but an increased incidence after combination surgery.7
Coronary and valve surgery patients were virtually equally
represented in our cohort, however Creswell et al. noted that AF
incidence was in fact increased in valve and combination valve–
coronary surgery.10 In the developing world, it is our experience
that patients with valve disease present later, once AF has
already developed. Since these patients were excluded from our
study, this may have accounted for the unremarkable difference
in incidence between the valve and coronary surgery groups.
The vast heterogeneity between definitions and study group
compositions makes a direct comparison of incidence and
other parameters challenging. Several studies do not allude to a
definition for AF or method of detection, while others considered
only patients requiring intervention for AF. The African incidence
of AF post cardiac surgery remains undocumented.
The highest incidence of AF is seen on post-operative days
two to three, with fewer patients developing AF in the early
post-operative period or beyond four or more days.7 Seventy per
cent of patients develop AF before the end of day four and 94%
before the end of day six.11 Mathew et al. documented the peak
incidence of AF on postoperative days two and three in their
prospective, observational study of 4 657 patients undergoing
CABG in 70 centres. Fifty-seven per cent of patients had only a
single episode of AF during their hospitalisation.8
Our study showed that AF was most prevalent from the third
post-operative day and not in the immediate post-operative
period (< 24 hours). This supports the notion that AF generation
is complex and that it extends beyond the peak adrenergic surge
prevalent on the first post-operative day.
Tsikouris et al. demonstrated P-wave dispersion and atrial
conduction time to be greatest on days two to three, and
day three, respectively.12 It remains unclear whether these
electrophysiological alterations contribute wholly or in part to
the increased development of AF after the first post-operative
day.
A wide array of risk parameters has been evaluated globally,
with conflicting results. Mathew et al. found age to be an
independent risk factor for AF post cardiac surgery and this
finding has consistently been reported by others.7,8 The risk of AF
increases by at least 50% per decade and particularly so over 70
years of age. Inflammatory and degenerative changes associated
with advanced age cause atrial fibrosis and degeneration.7,13 The
resultant alterations in electrophysiological properties may act
as substrates for AF. Most of the parameters considered in this
study, albeit few, revealed results that were within normal limits.
Hypertension has been proposed to predict AF after cardiac
surgery and this may be related to associated fibrosis and
dispersion of atrial refractoriness.1,8,11 However, several well-
conducted trials with large numbers of patients have refuted this
proposition.3,14,15
Men are more likely to develop AF than women. This
disparity may be explained by gender differences in ion-channel
expression, and hormonal effects on autonomic tone.11,15 However,
Mathew et al. and Echahidi et al. expressed a contradictory
viewpoint on male predominance.3,8
Withdrawal of β-blockers pre-operatively in patients on
chronic β-blocker therapy causes withdrawal effects, as described
by Kalman et al.16 The poor oral absorption of β-blockers in
the post-operative period described by Valtola et al. is likely to
exarcebate this withdrawal effect.17
Beating-heart coronary surgery (OPCAB) did not appear to
provide any obvious protective benefit to the development of
AF in this study. Notwithstanding the heterogeneity in OPCAB
trials, off-pump surgery is still believed to be associated with a
significant reduction in post-operative AF.2
In our study, only three patients had redo-operations and
another three β-blocker withdrawal in the pre-operative period.
Therefore no deductions could be made. Twelve patients were
noted to be smoking up to the time of surgery. This information
was confession-based and furthermore, all patients were advised
to cease smoking at the cardiac surgical preview visits.
In their retrospective analysis of 5 058 patients post isolated
CABG in patients older than 50 years, obesity was shown by
Echahidi et al. to be an independent risk factor for AF.3 Higher
cardiac output requirements, left ventricular mass and left atrial
size predisposed to AF in the obese population.18 Filardo et al.
reported a significant relationship between BMI and AF. Their
study of 7 027 consecutive patients post CABG with a mean age
of 64.9 years showed a median BMI of 28 kg/m2.19 In our study,
a mean BMI of 26.4 kg/m2 indicated an overweight population,
as per the World Health Organisation, but no relationship was
readily apparent between BMI and AF in our cohort.
In our study, all patients were found to be AF free at
follow up. The natural history of AF post surgery is that