The Atrioventricular Nodal Fat Pad in Humans: Fat or Fiction?
SUNEET MITTAL, M.D., and BRUCE B. LERMAN, M.D.
From the Department of Medicine, Division of Cardiology, Cornell University Medical Center, New York, New York
Editorial Comment
The importance of the integrated role of the autonomic
nervous system on coordinating cardiac conduction and
function has long been appreciated. Less well appreciated
are the anatomic correlates associated with this system and
in particular, the parasympathetic nervous system. To date,
most anatomic data have been derived from animal studies.
What can be agreed upon is that the parasympathetic ner-
vous system mediates its effects on supraventricular tissue
via the right and left vagal nerves. Preganglionic neurons
arise in the brainstem and project to cardiac ganglia, located
in discrete epicardial fat pads. Experimental studies in dogs
have shown that parasympathetic nerve bers selectively
innervate the sinoatrial (SA) or atrioventricular (AV)
node.
1,2
Stimulation of a fat pad located at the atrial juncture
of the right pulmonary veins results in a negative chrono-
tropic response, suggesting selective SA node innervation,
whereas stimulation of a fat pad identied at the junction of
the inferior vena cava and right inferior pulmonary vein
results in a negative dromotropic response, consistent with
selective AV node innervation.
1,2
Although the presence of a selective SA node epicardial
fat pad has been shown to be present in humans,
3
there has
been no conclusive evidence for an AV node fat pad. Of
interest, however, endocardial stimulation in the region of
the posteroseptal right atrium and/or coronary sinus ostium
has a negative dromotropic effect in the AV node.
4-6
In this
issue of the Journal , Quan et al.
7
convincingly demonstrate
the existence of an AV node fat pad in humans and conrm
that parasympathetic nerves emanating from this fat pad
selectively innervate the AV node (and a small area of
surrounding atrial myocardium). This epicardial fat pad was
identied near the junction of the left atrium and right
inferior pulmonary vein in seven patients undergoing coro-
nary artery bypass grafting. Complete AV block (without a
change in sinus cycle length) occurred reproducibly with fat
pad stimulation and normal AV node conduction resumed
after stimulation ceased.
The identication of this epicardial fat pad has several
potentially important clinical implications. Endocardial ac-
cess to the AV node fat pad may allow the development and
application of novel techniques for parasympathetic medi-
ated ventricular rate control during atrial brillation.
4-6
Re-
cent preliminary data from the Atrial Fibrillation Follow-Up
Investigation of Rhythm Management (AFFIRM) trial sug-
gest that a strategy of ventricular rate control is as effective
as aggressive attempts to maintain sinus rhythm in patents
with atrial brillation (Wyse DG, et al. Circulation 2002;
105:e9085). As is well appreciated, digoxin, beta-blockers
and/or calcium channel blockers are initially used to achieve
rate control. However, the effect of these drugs is unpre-
dictable and often associated with undesired effects on
cardiac contractility and/or blood pressure. For patients in
whom drug therapy is either ineffective or not tolerated, AV
node modication or AV node ablation has emerged as an
alternative strategy.
8
However, the long-term efcacy of
AV node modication has been disappointing,and AV node
ablation consigns patients to pacemaker dependence and is
hemodynamicallysuboptimalbecause ventricularactivation
originates from the site of pacing rather than via the native
conduction system. Therefore, control of ventricular rate
through manipulation of the AV node fat pad is of potential
interest because it should preserve physiologic conduction.
Furthermore, as demonstrated by Quan et al.,
7
vagal stim-
ulation can be turned on and off virtually instantaneously
and can be repeated as often as required without attenuation
of effect.
Stimulation of the AV node fat pad also allows the
application of closed-loop feedback control algorithms that
may provide physiologic control of AV nodal rate.
9,10
Zhang et al.
10
recently showed that a linear proportional
feedback system using short pulse bursts delivered from the
epicardial AV node fat pad in dogs effectively slowed AV
node conduction during atrial brillation. Because the con-
trol algorithm can be manipulated to decrease AV node
conduction over a wide range of ventricular rates, it is
possible to target an optimal degree of negative dromotropic
effect (as measured by indices of cardiac output).
The ndings in this study may have implications regard-
ing radiofrequency ablation of supraventricular arrhyth-
mias. For example, parasympathetic denervation of the right
atrium during directed radiofrequency catheter ablation can
abolish the induction of vagally mediated atrial brilla-
tion.
11
This observation may provide an explanation for the
reported efcacy (although modest) of right atrial ablation
incorporating linear lesions in the septum and cavotricuspid
isthmus in patients with vagally mediated atrial brilla-
tion.
12
It is worth cautioning, however, that ablation of
parasympathetic nerves from within the atria may have
detrimental effects. Parasympathetic denervation has been
implicated as the mechanism of inappropriate sinus tachy-
cardia, observed after ablation of the slow AV nodal path-
way or posteroseptal accessory pathway.
13
In addition,
given the increasing popularity of pulmonary vein isolation
in patients with atrial brillation, the long-term effect of
ablation in the region of the right inferior pulmonary vein
needs to be considered, given the proximity of this vein to
the AV node fat pad.
The contribution of Quan et al.
7
is that they were able to
demonstrate the anatomic derivative of parasympathetic in-
nervation to the AV node. The challenge ahead will be to
This work was supported in part by grants from the National Institutes of
Health (RO1-HL56139); the American Heart Association, Grant-in-Aid,
New York City Afliate; the Maurice and Corinne Greenberg Arrhythmia
Research Grant; and the Raymond and Beverley Sackler Foundation.
J Cardiovasc Electrophysiol, Vol. 13, pp. 740-741, August 2002.
Address for correspondence: Bruce B. Lerman, M.D., Division of Cardi-
ology, Cornell University Medical Center, 525 East 68th Street, Starr 4,
New York, NY 10021. Fax: 212-746-6951; E-mail: lerman@med.
cornell.edu
740
Reprinted with permission from
JOURNAL OF CARDIOVASCULAR ELECTROPHYSIOLOGY, Volume 13, No. 8, August 2002
Copyright ©2002 by Futura Publishing Company, Inc., Armonk, NY 10504-0418