Chow et al (1) report that significant financial resources are spent on
surgical AF ablation in Canada with unclear long-term benefits and
“scientifically rigorous” data regarding reduction in mortality and
stroke risks. The authors call for large prospective studies examining
clinically important outcomes to justify the routine use of concomitant
surgical AF ablation (SA) and to guide allocation of healthcare funds.
Based on their analysis, they assert that concomitant surgical AF
ablation during cardiac surgery should be reconsidered. To better put
these findings into context, it is important to examine the assumptions
made and data utilized to assess both cost and benefit.
The authors admit that their methods for cost derivation are
significantly limited and do not “constitute the precision of a
detailed prospective cost-utility analysis.” However, even their
“simple methods” are incomplete and miss important alternative costs
if SA is abandoned. They describe the incremental costs associated with
surgical ablation of AF in a specific geographic location - Ontario
(Canada). The authors aggregate costs from a variety of sources but base
much of their calculations on an increased hospital length of stay (LOS)
and increased pacemaker implantation rate from a recent systematic
review and meta-analysis of RCTs of surgical ablation of AF (2).This
systematic review describes an increased hospital LOS of
~ 1.67 days and a trend (not statistically significant)
of increased permanent pacemaker implantation at hospital discharge,
although risk of pacemaker implantation was not increased in the long
term. Using this information plus Ontario-specific information of
institutional costs, physician remuneration and device costs, they
generate Ontario-specific financial costs for the procedure and
calculate the incremental cost associated with this intervention as
$4287 CAD per patient. The authors state that their calculations likely
underestimate the true cost of surgical AF ablation but that at least
over $10 million (CAD) was spent in Ontario province on surgical AF
ablation between 2006-2017. The authors fail to consider, however, the
alternative costs associated with NOT performing SA in patients with AF
undergoing surgery. While surgery for the primary structural abnormality
may alleviate the patients’ symptoms, it is likely that patients with
pre-existing AF will continue to have AF if SA is not performed.
Patients whose AF is left untreated could continue to have symptoms
related to AF necessitating multiple procedures, including
cardioversions, hospitalization for initiation of antiarrhythmic drugs,
and catheter ablation. Even patients with asymptomatic to minimally
symptomatic AF are often treated with this multiplex of therapies, each
of which has more limited efficacy than SA. Given the likelihood of
multiple treatments and procedures if SA is not performed, any
clinically relevant cost analysis must incorporate this real cost.
Consequently, the true incremental cost of SA is likely substantially
less than the numbers generated by Chow et al.
When analyzing the benefits of SA, the conclusions draw heavily from a
recent systematic review and meta-analysis that describes several key
points regarding surgical AF ablation: 1) surgical AF ablation during
cardiac surgery improved freedom from AF at 12 months; 2) there was no
statistically significant evidence of impact on mortality,
thromboembolic or neurovascular events with limited long-term follow-up;
and 3) the evidence for improvement in health-related quality of life
was limited (2). It is important to note that of the 23 studies in this
meta-analysis, only 5 had follow-up >12 months and only 1
> 24 months. This reported time frame may be inadequate to
detect significant differences in clinical outcomes, particularly
long-term outcomes such as incident stroke and mortality.
As a comparison, in the world of catheter-based ablation, we have
recently had the results of the CABANA trial (3). In the
intention-to-treat analysis, this trial did not demonstrate
statistically significant differences between catheter ablation and
anti-arrhythmic drug therapy with regard to the primary composite
endpoint of death, disabling stroke, serious bleeding, or cardiac
arrest. However, there was a substantial and clinically important
benefit of catheter ablation over drug therapy in reducing recurrent
symptomatic and asymptomatic AF over 5 years of follow-up (4). There
were also clinically important and significant improvements in quality
of life (QOL) at 12 months in symptomatic patients (5). There was also a
signal for decreased cardiovascular hospitalizations, although not
centrally adjudicated, and there may be a signal for mortality benefit
of catheter ablation in the subgroup of patients with systolic
dysfunction (6). On-treatment analysis showed even more significant
benefits, even with mortality reduction.
Currently, the American Heart Association and the European Society of
Cardiology provide a Class IIa recommendation for concomitant surgical
AF ablation to maintain sinus rhythm in symptomatic patients. The
guidelines consider surgical AF ablation with cardiac surgery a
“reasonable” treatment, including patients with persistent or
permanent AF (7, 8). The recent 2017 STS guidelines outline that
concomitant surgical ablation to restore normal sinus rhythm during
mitral valve procedures is a Class 1 recommendation, level of evidence
A. Similarly, surgical ablation at the time of isolated AVR and AVR with
CABG is a Class 1 recommendation, level of evidence B-NR (9).
We believe the authors’ suggestion that the incremental cost of surgical
AF ablation, based on this incomplete analysis, is too great to be
shouldered by the taxpayers of Canada is unjustified. This manuscript
touches on some interesting topics regarding the rendering of medical
care, namely – what patient outcomes justify the financial costs of a
procedure and what bar do we set for the level of evidence needed to
justify the performance of a procedure but their cost-benefit analysis
is incomplete requiring careful reevaluation of its conclusions.
The importance of freedom from AF and protection from AF relapses for
patients, in particular those with symptomatic AF, cannot be
underestimated. Both catheter ablation studies such as CABANA and
studies of surgical ablation support a clear decrease in AF burden with
ablation. In CABANA, freedom from AF was long-lasting with the benefit
of ablation sustained, although attenuated, at 5-year follow-up.
Interestingly, the CABANA trial also included patients who had MI, PCI,
or valve/bypass surgery > 3 months prior to enrollment
although we do not have subgroup-specific data. In regard to freedom
from AF, the surgical ablation data from the referenced meta-analysis
only extends on average to 12 months (2), but we cannot exclude the
possibility of a more sustained treatment effect like that seen in
CABANA. In a study of SA, risk-adjusted survival differences were
assessed in 372 propensity matched pairs; at last follow-up, 78% of SA
patients were free of AF, and restoration of sinus rhythm was associated
with improved survival (10). While large-scale data on mortality
following SA remain sparse, the Society of Thoracic Surgeons’ report of
over 28,000 propensity score matched patients with and without SA
support improved survival at 30 days (11).
There is strong evidence for improved symptoms and quality of life in
patients after ablation. In the DISCERN AF study which evaluated
symptomatic and asymptomatic episodes of AF pre and post radiofrequency
ablation using long-term continuous monitoring, rates of symptomatic AF
approached 50% (12). After ablation, arrhythmia events were 3 times
more likely to be asymptomatic and the proportion of asymptomatic
episodes approached 80%. This was also substantiated by CABANA trial
data that showed that when AF recurred after catheter ablation it was
more likely to be asymptomatic – at five years, freedom from recurrence
of symptomatic AF episodes was ~ 80% (4).
Health-related quality of life was also significantly improved at 12
months in the ablation group versus medical therapy (5). In the
systematic review from McClure et al, there was a significant difference
in physical role functioning post-operatively even though there were
only two studies that could be included in the analysis (2). Granted, in
patients undergoing cardiac surgery, health-related QOL outcomes are
confounded by improvements in treating the underlying heart disease in
addition to the restoration of sinus rhythm (13).
Overall, the relatively small cost of $4287 CAD per patient, which
would represent a much smaller incremental cost when accounting for the
subsequent cost for AF treatment if SA is not performed, strongly
supports undertaking this procedure despite the small increased risk of
pacemaker implantation and increased hospital stay given the known and
established benefits of freedom from AF and improvement in symptoms and
quality of life. Additionally, the use of this procedure should not be
halted since there is short-term and mid-term surgical mortality data
that support its use (9-11), and we cannot rule out mortality and stroke
benefit over a period of 5 or more years because of the limited data
available. In this sense, we agree with the authors that updated
long-term outcomes should be pursued to enhance our understanding and to
refine decision making.