JACC Vol. 27, No. 5
April 1996:1071-8
1071
Fractionated Electrograms in Dilated Cardiomyopathy: Origin and
Relation to Abnormal Conduction
J A C Q U E S M. T. DE B A K K E R , PHD, F R A N S J. L. VAN C A P E L L E , PHD,* M I C H I E L J. J A N S E , MD,*
S A R A T A S S E R O N , RT,* J E S S I C A T. V E R M E U L E N , MD,* N I C O L A A S DE J O N G E , M D , t
J A A P R. L A H P O R , M D t
Utrecht and Amsterdam, The Netherlands
Objectives. We sought to investigate the origin of the fractionated electrogram and its relation to abnormal conduction in
cardiomyopathic myocardium.
Background. Patients with dilated cardiomyopathy have a high
incidence of ventricular tachycardias. Electrograms recorded in
these patients are often fractionated.
Methods. High resolution mapping (200-/xm interelectrode
distance) of the electrical activity was carried out in 11 superfused
papillary muscles and 6 trabeculae from 7 patients who underwent heart transplantation because of dilated cardiomyopathy.
Similar measurements were taken in four papillary muscles from
dog hearts in which electrical barriers had been artificially made.
Ten human preparations were studied histologically.
Results. All preparations revealed sites with fractionated electrograms. In three human preparations, activation patterns
showed a discernible line of activation block running parallel to
Premature beats and nonsustained tachycardias are the most
prominent ventricular arrhythmias in patients with dilated
cardiomyopathy (1-3). Nonsustained ventricular tachycardias
occur in 20% to 60% of these patients (1-3), whereas monomorphic, sustained ventricular tachycardias (4,5) and ventricular fibrillation (2,6) are observed in 1% to 2%.
In various experimental models of dilated cardiomyopathy
electrophysiologic abnormalities have been observed (7-12).
In a number of these studies increased interstitial fibrosis was
also found (7,12). Analysis of human hearts with dilated
eardiomyopathy revealed increased fibrosis, often accompanied by ultrastructural abnormalities (13-19). Regularly arranged myocardial fibers have been reported (20), as well as
myofiber disarray (16,17). In trabeculae and isolated myocytes
from human hearts with dilated cardiomyopathy, membrane
abnormalities have been observed (21,22). Anderson et al. (23)
Fromthe InteruniversityCardiologyInstituteof The Netherlands,Utrecht;
*Departmentof ExperimentalCardiology,AcademicMedicalCenter,Amsterdam; and tHeart LungInstitute, Universityof Utrecht, Utrecht, The Netherlands.
ManuscriptreceivedJune 29, 1995;revisedmanuscriptreceivedNovember
14, 1995,acceptedDecember13, 1995.
Address for corresoondence:Dr. JacquesM. T. de Bakker,Departmentof
ExperimentalCardiology,Meibergdreef9, 1105AZ Amsterdam,The Netherlands.
©1996by the AmericanCollegeof Cardiology
Publishedby ElsevierScienceinc.
the fiber direction. Fractionated electrograms were recorded at
sites contiguous to the line of block. In five preparations, fractionated electrograms were recorded at sites where lines of block
were not identified. In these preparations, electrical barriers
consisted of short stretches of fibrous tissue. In the remaining
nine preparations, fractionated electrograms were recorded, both
from sites contiguous to distinct obstacles and sites without
evidence of a barrier.
Conclusions. Our observations showed that fractionated electrograms recorded in myocardium damaged by cardiomyopathy
were due to both distinct, long strands and short stretches of
fibrous tissue. Delayed conduction was caused by curvation of
activation around the distinct lines of block and by the wavy
course of activation between the short barriers. The latter reflects
extreme nonuniform anisotropy.
(J Am CoU Cardiol 1996;27:1071-8)
have shown that the severity of abnormal propagation correlated with the amount of myocardial fibrosis in patients with
dilated cardiomyopathy who underwent heart transplantation.
Although the mechanisms of fractionation are various
(24-32), the increase in interstitial fibrosis is likely the mechanism of fractionated electrograms in dilated cardiomyopathy.
To investigate the characteristics of fractionated electrograms
and their relation to abnormal conduction, we carried out high
resolution mapping in papillary muscles and trabeculae from
the hearts of patients who underwent heart transplantation in
the end stage of heart failure due to dilated cardiomyopathy.
To support interpretation of our data, we performed similar
investigations on papillary muscles excised from dog hearts.
Specimens were superfused in a tissue bath. In superfused
preparations, a subendocardial rim that is 300 to 500/zm deep
survives (33). The thickness of the rim is small enough,
especially in relation to the length and width of the preparation, to regard the specimens as essentially two-dimensional.
Methods
Eleven papillary muscles and six trabeculae were resected
from the left or right ventricle, or both, of seven patients. Four
papillary muscles were harvested from two dog hearts.
0735-1097/96/$15.00
SSDI0735-1097(95)00612-5
1072
DE BAKKER ET AL.
F R A C T I O N A T I O N IN C A R D I O M Y O P A T H Y
Human preparations. Specimens were obtained from patients who underwent heart transplantation because of idiopathic dilated cardiomyopathywithout valvular disease. Before
the operation, nonsustained ventricular tachycardias were observed in four patients, whereas ventricular fibrillation was
documented in two others. In the remaining patient, premature beats were the only ventricular arrhythmias observed. The
ejection fraction of the hearts, measured before operation,
ranged from 13% to 20% (mean 16%).
Just before removal, hearts were perfused with a cardioplegic solution (St. Thomas Hospital No. 2). During transportation, the hearts were submerged in a modified Tyrode's
solution at 4°C and gassed with 95% oxygen and 5% carbon
dioxide. The Tyrode's solution contained (in mmol/liters):
156.5 Na +, 4.7 K +, 1.5 Ca2+, 0.5 HzPO4-, 137 C1-, 28 HCO 3and glucose 20. After incising the left and right ventricular
walls, papillary muscles and trabeculae were resected. Specimens that were at least 4-mm long were selected for the study.
Preparations were pinned to the base of a tissue bath
(volume 100 ml) and superfused with oxygenated Tyrode's
solution at 37°C (50 ml/min). They were allowed to recover for
at least 45 min, until spread of activation during basic stimulation was stable. Stimulation was achieved (twice threshold) at
a cycle length of 600 ms through a bipolar electrode (diameter
of each pole 0.2 ram, interelectrode distance 0.5 ram).
Canine preparations. Mongrel dogs weighing 18 to 32 kg
were premedicated with azaperon (12 mg/kg body weight
intramuscularly) and atropine (500 Ixg intramuscularly) then
anesthetized with sodium pentobarbital (35 mg/kg intravenously) and metomidaat (5 mg/kg intravenously). After thoracotomy, the hearts were rapidly removed and submerged in
Tyrode's solution at 4°C. Then the ventricles were incised and
the papillary muscles removed. The preparations were pinned
to the base of the tissue bath and superfused in the same way
as the human preparations. After a recovery period of 45 min,
electrical barriers were made by incising the preparations with
an ultrafine scalpel (Drukker Diamond Surgical Instruments).
Recording techniques. High resolution mapping was carried out with a compound electrode consisting of eight terminals arranged in a row (interelectrode distance 200 Ixm). Each
terminal consisted of two adjacent silver wires (diameter
100 txm), one 2 mm shorter than the other. Both wires were
insulated, except at the tips. The longer end of the pair was
positioned on the tissue, the other about 2 mm above the
tissue. These electrodes suppress remote effects but have
unipolar characteristics (34).
The compound electrode was mounted in a micromanipulator and positioned with its long axis perpendicular to the
fiber direction. The electrode was moved along the fiber
direction in steps of 200 ~m over a distance ranging from 3 to
6 mm. Electrograms from the electrode pairs were fed into
differential amplifiers having a gain setting of 500 or 1,000.
Subsequently, signals were digitized with a sampling rate of
8 kHz and stored on a hard disk.
Electrograms were often fractionated, that is, they consisted
of multiple deflections. Activation times of all deflections
JACC Vol. 27, No. 5
April 1996:1071-8
within the electrograms were determined. The activation time
was defined as the instant with the largest negative derivative
between either the maximum or the point of inflection and the
ensuing minimum. The activation time of the deflection with
the largest negative derivative in a signal was selected as the
local activation time and used to construct isochronal maps. If
activation times of signals recorded at contiguous sites
(200 Ixm apart) differed by >2 ms, thick black lines were
drawn. These lines indicated either conduction block or apparent (pseudo) block (35,36). Activation times were determined
digitally from the original waveforms using a computer algorithm and corrected manually for artifacts.
Histologic methods. Histologic investigations were carried
out on 10 human specimens. Preparations were fixed with
buffered formaldehyde immediately after the experiment. Sections were taken from selected sites at which the activation
pattern revealed the presence of a barrier or where signals
were fractionated. Sections (10/xm thick) were cut perpendicular to the long axis of the specimens. Consecutive sections
were stained alternately with hematoxylin-eosin and picrosirius red.
Results
Human preparations. Fractionated electrograms. The number of electrograms recorded in each preparation ranged from
112 (8 × 14) to 256 (8 × 32). All preparations revealed sites
where electrograms were fractionated. The number of deflections in the fractionated electrograms was between 2 and 16
(mean 5.3) and depended on the site of stimulation. Duration
of the electrograms--delay between the first and last activation time--was between 1 and 15 ms (mean 5.8 ms).
Distinct lines of "block." In three preparations, activation
maps revealed a distinct line of "block." These lines were
between 0.8 and 1.6 mm long and always parallel to the fiber
direction. Electrograms in the three specimens were fractionated at sites adjoining the barrier.
The upper panel of Figure 1 shows the activation map of a
papillary muscle. A discrete line of block runs parallel to the
fiber direction between rows 3 and 4 and from columns 4 to 11.
The preparation was stimulated from the site indicated. Isochronal lines show that activation proceeds predominantly
perpendicular to the fiber orientation in the area proximal
from the line of block. After arriving at the line of block, the
wave front curves around both ends of it. The ensuing wave
fronts collide distally from the barrier along columns 4 and 5.
Tracings in the lower right panel show three electrograms
recorded near the right extremity of the line of block. Electrogram e, which is recorded near the pivoting point, shows the
beginning of fractionation (arrow).
Electrogram b, recorded near the barrier away from the
end, reveals two clearly separated deflections. The first component is caused by the approaching wave front and coincides
with the deflection of tracing c. The second deflection in
tracing b is caused by the front that curves around the line of
block, generating signal a. At the ends of the line of block
JACC Vol. 27, No. 5
April 1996:1071-8
1073
DE BAKKER ET AL.
FRACTIONATION IN C A R D I O M Y O P A T H Y
fiber d i r e c t i o n
rows
c~16~15~___
panel, Activation pattern during basic
stimulation (cycle length 600 ms) of a papillary muscle
from a human heart impaired by cardiomyopathy. Numbers in the isochronal lines indicate activation times (in
ms) with respect to the stimulus. Bold line running from
column 4 to 11 along rows 3 and 4 indicates a line of
activation "block." The preparation was stimulated in
the upper right corner (marker). Lower panel, Tracings
are electrograms recorded near the line of block at the
sites indicated in the upper panel. Tracings d to f are
recorded near the right end of the barrier. Electrograms
e and b are fractionated. S = stimulus artifact.
i
Figure 1. Upper
1
i
i
" ~ i
0.2mm
10
17 ,
,
11
15
Columns
a
d
b
J/
/
e
C
j
S
f
isochronal lines are crowded, indicating delayed conduction
velocity in this area.
Stimulating the same papillary muscle near its base caused
activation to run parallel to the fiber direction with a conduction velocity of -0.7 m/s. The barrier previously seen at rows 3
and 4 was no longer discernible. Signals recorded at sites where
the activation pattern previously revealed a barrier during
propagation perpendicular to the fiber orientation were no
longer fractionated.
Fractionation in absence of discernible lines of block. In
preparations with distinct barriers, fractionated electrograms
were mainly localized to sites contiguous to the barriers.
In five preparations, fractionated electrograms were recorded at sites where activation maps did not reveal lines of
block. The upper panel of Figure 2 shows the activation
pattern of such a specimen. Two short lines of block only are
present along row 1. The preparation was stimulated in the
upper left corner. Activation runs perpendicular to the fiber
direction in the central part of the preparation; the conduction
$
J
/1t
f
f
5 ms
velocity in this area is -0.2 m/s. Isochronal lines are not
smooth in this area but reveal several irregularities with dips
and bulges, as shown, for instance, at sites m and o. Electrograms recorded at these sites show some similarity to those
recorded at sites near distinct barriers.
Fractionation in the apparent absence of electrical barriers
suggests that these barriers are nevertheless present, but too
short to show up in the activation maps. Stimulation causing
activation to propagate parallel to the fiber orientation resulted in fast conduction velocities.
In 9 of the 17 preparations, areas where fractionated
electrograms were recorded in the vicinity of lines of activation
block varied with areas in which fractionation was present in
the apparent absence of lines of block.
Histologic studies. The location of pronounced lines of
block in the activation patterns could be traced histologically as
prominent strands of fibrous tissue traversing the surviving
subendocardial rim. Areas with fatty tissue were found only
sporadically. Figure 3 show sections of a papillary muscle
1074
DE BAKKER ET AL.
F R A C T I O N A T I O N IN C A R D I O M Y O P A T H Y
JACC Vol. 27, No. 5
April 1996:1071-8
_J-L
fiber direction
b
•h
14__
rows
f
.._____~
.^
- --15
~ - 1 ~
/-t~/~
1
8
..----~1~
"__c
~
b
15
_~___~mk
~
,
~
24
j
d
m
g
0
h
I
y
~
J
columns
b
J
f~
~d
17
0,2 mm
mp~
32
Figure 2. Upper panel, Activation map of a
trabecula from an explanted human heart affected by cardiomyopathy.Bold lines indicate
activationblock or pseudoblock.Numbers in the
isochronal lines are activationtimes in milliseconds measuredwith respect to the stimulus. The
preparation was stimulated from the upper left
corner. Lower panel, Tracings are signals recorded at sites indicated in the upper panel.
Signals c through h and k through p are fractionated, despite the fact that they were recorded at sites where lines of "block" were not
detected. See text for discussion.
p
5 ms
where a distinct line of block was present. Figure 3A shows a
broad strand of fibrous tissue traversing the subendocardial
rim (arrow). Figure 3B is just to the left from the line of block
and shows that the barrier disappears subendocardially. The
remainder of the barrier is present 500/xm below the endoeardium (arrow).
Areas where fractionated electrograms were recorded in
the absence of a zone of activation block also showed the
presence of fibrosis. This either consisted of small subendocardial strands (dark lines in Fig. 3C) or revealed a more patchy
architecture (Fig. 3D).
Canine preparations. A single, artificially made barrier. Figure 4 shows the activation pattern of a trabecula from one of
the dog hearts. An electrical barrier was artificially created by
incising the preparation along the fiber direction in rows 4 and
5, from column 8 to the right end of the preparation. The
preparation was stimulated at the site indicated in the upper
right corner. Electrograms a through c, recorded across the
barrier yet well away from both extremities, revealed two
clearly separated deflections. The first deflection, caused by the
approaching wave front, is predominantly positive and has its
largest amplitude in tracing a. In tracings b and c, the initial
positive deflection is not immediately followed by a negative
one, because the barrier terminates activation. The second
deflection, caused by activation running distally from the
barrier, is largest in tracing c; this component is remote in
tracings a and b.
The two components are separated by a nearly isoelectric
interval that decreases in duration as the recording site is
closer to the left extremity of the barrier. This is shown in
tracings d and e, which have double components separated by
a short isoelectric interval. The initial positive deflection in
both tracings d and e is generated by the approaching wave
front--the departing front induces the ensuing negative deflection. At the far left of the barrier, crowding of isochronal lines
indicates activation delay, which is responsible for the delay
between the positive and negative deflection.
Multiple, short, artificial barriers. In one preparation, electrical barriers with a length of 400 ~m were made at distances
of 400 t~m, as shown in the upper right panel of Figure 5. The
preparation was stimulated from the upper right corner ($1).
Before incising the preparation, recordings were made at
locations along the vertical line (upper left panel). None of the
signals shown in the lower left panel were fractionated.
The lower right panel shows tracings from the same sites
after incisions were made. Now, the signals reveal fractionation
J A C C Vol. 27, No. 5
April 1996:1071-8
A
FRAC*FIONAT1ON
1075
D E B A K K E R E T AL.
IN CARDIOMYOPATHY
G
e
!
t
I
B
!
13
I
!
very similar in configuration to that observed in signals recorded near nonartificial barriers. The activation pattern
showed a line of block only at recording site 6. As in Figure 2,
isochronal lines revealed curvation at sites where lines of block
could not be demonstrated. These observations support the
concept that short electrical barriers, not evidenced in an
isochronal map, can cause fractionation, as shown in Figure 2.
Discussion
Increased interstitial fibrosis is a common finding in hearts
from patients with dilated cardiomyopathy (13-19), suggesting
that fibrosis is a likely mechanism for generating fractionated
electrograms in these hearts. Fractionation is associated with
abnormal conduction, which may promote reentry arrhythmias
(24,25). A correlation between the amount of myocardial
fibrosis and the severity of abnormal propagation in patients
with dilated cardiomyopathy has been documented by Anderson et al. (23).
Activation block. Definition of the thick black lines as
markers for activation block or pseudoblock was based on data
from published reports (35,36) and the results of one of the
canine papillary muscles (Fig. 5). In the latter, activation block
was evident, and delay between the two sites at the pivoting
I
I
Figure 3. A, Photomicrograph of a tissue section of a papillary muscle
with a discernible line of"block." The section was taken perpendicular
to this line of block. A wide band of fibrous tissue (arrow) that
traverses the surviving subendocardial rim coincides with the line of
block. B, Photomicrograph from the same tissue preparation as in A
but made just to the left of the block zone. At this level, only a vestige
of the barrier seen in A remains (arrow). C, Photomicrograph of a
section of the trabecula from Figure 2 in which fractionated signals
were recorded in the absence of lines of block. Small strands of fibrous
tissue are present in the subendocardial rim. D, Photomicrograph of
another trabecula in which fractionation occurred in the absence of
lines of block. The subendocardial rim is very patchy and subdivided
into fibrous tissue (dark areas) and surviving myocytes (bright areas).
Calibration lines = 250/zm.
point was 1.9 ms. This corresponds to a conduction velocity of
-0.1 m/s. Data from published reports show that slow conduction in ischemic preparations ranges from 0.21 to 0.13 m/s (35).
During anisotropic conduction, velocities between 0.05 and
0.2 m/s were reported; in these cases lines of block often
indicated apparent block. Activation seemed to pass these lines
of pseudoblock by way of a tortuous route (36).
In a number of hearts, anatomic barriers were not evidenced by lines of block in the activation pattern. This is to be
expected if the barriers are too short to generate an activation
delay of more than 2 ms between activation at either side of the
1076
D E B A K K E R E T AL.
FRACT1ONATION IN CARDIOMYOPATHY
J A C C Vol. 27, No. 5
April 1996:1071-8
fiber direction
I----.
_I-L. $1
. J - l _ $1
IP
8
5
1
-
incision
0.4 m m
0.2 mm
7
9
14
b
5 ms
8
G
Figure 4. Upper panel, Activation pattern of a papilla~ muscle from
a canine heart in which an artificial barrier was made by incising the
preparation parallel to the fiber orientation, from column 8 to its right
end. The preparation was stimulated at the site indicated by the
marker. Isochronal lines are drawn 1 ms apart; numbers in the
isochronal lines indicate activation times in milliseconds (stimulus is
t = 0 ms). Lower panel, Electrograms recorded at sites indicated in the
upper panel. See text for discussion. S = stimulus artifact.
barrier and thus did not manifest themselves in the activation
map.
Influence of Purkinje system. One may argue that the
Purkinje system, which is scattered over the surface of normal
canine papillary muscle, may have produced effects on the
activation sequence and extracellular waveforms. Activation
running through the Purkinje system would speed up the
apparent conduction velocity and create sites with early breakthrough. However, in all cases conduction was essentially
uniform. The conduction velocity in the longitudinal as well as
the transversal direction was equal to or less than that in
normal myocardium, indicating that participation of the Purkinje system is unlikely. It is also unlikely that activation in
Purkinje fibers explains the transfer of activation through the
pseudoline of blocks. Because of the high spatial resolution of
the mapping, one would expect to record Purkinje spikes at
sites near the crossing point. This, however, was not the case.
Origin of fractionated electrograms. Fractionated electrograms originate in infarcted myocardium, where fibrous tissue
10 ms
Figure 5. Electrograms recorded with the compound electrode in one
of the canine preparations. Recording sites are indicated with respect
to the site of stimulation in the upper panel. Tracings in the left panel
were made in the untouched preparation and reveal biphasic deflections only. Tracings in the lower right panel were recorded at the same
sites as those in the left panel, but after incising the preparation, as
shown in the upper right panel. These incisions resulted in fractionation of the electrograms. See text for discussion. $1 = stimulation
sites.
separates myocardial fibers, giving rise to separated conducting
paths (24-26). Multiple deflections have also been recorded in
the atrioventricular junctional area, where superficial and
deeper layers seem to be electrically separated (27-29).
Remote activity may also produce electrograms with multiple deflections. Recordings in the atrium are often obscured
by deflections owing to remote ventricular activation, which
may hamper the selection of the local atrial deflection (30).
Separation of conducting paths is not exclusively the result
of anatomic barriers such as fibrous or fatty tissue. Zones of
functional block, too, constitute barriers between asynchronous activation fronts, and electrograms recorded from such
zones are also fractionated (31).
Zones of activation block are not a prerequisite for the
generation of fractionated electrograms, however. Spach et al.
(32) have shown, exclusively on the basis of anisotropy of the
tissue, that double potentials may occur in myocardial tissue.
In our preparations, fractionated electrograms were related
to fibrotic barriers. Occurrence of fractionation on the basis of
remote activation is very unlikely because large muscle masses
were absent. Fractionation on the basis of uniform anisotropy
cannot be ruled out.
The characteristics of fractionation in the presence of short
JACC Vol. 27, No. 5
April 1996:1[}71-8
barriers which we found are similar to those observed by Spach
et al. in the limbus of the fossa ovalis of a dog heart (37) and
human pectinate muscle bundles (38). In these cases, fractionated electrograms and reduced conduction velocity were
caused by nonuniform anisotropy at sites where collagenous
tissue was distributed as distinct collagenous septa.
Slow conduction. Although slow conduction in dilated cardiomyopathy seems to be associated with collagen infiltration,
slow conduction may also occur because of a reduction in
depolarizing currents. It is unlikely, however, that this was the
case in our preparations for the following reasons: 1) electrograms revealed sharp deflections, suggesting close to normal
depolarization; 2) microelectrode recordings that were carried
out in small trabeculae from patients with dilated cardiomyopathy (21) showed action potentials with upstroke velocities between 100 and 150 V/s; and 3) the conduction velocity
parallel to the fiber direction was always close to normal.
Isochronal lines near the edges of the lines of block were
closer together than they were farther away from the barrier,
indicating activation delay in that area. It has been shown by
Cabo et al. (39) that curvature of wave fronts results in slowing
of conduction. The delay is caused by a mismatch between
current supply and current demand. Activation near the edges
of the barriers curved, which explains the delay in that area.
The short stretches of fibrous tissue may create areas where a
transition from wide to narrow zones of myocardium occurs.
Load mismatches are also responsible for activation delay in
such areas (40,41).
Study limitations. Papillary muscles and trabeculae were
chosen because of their well-ordered architecture with regard
to fiber direction. These specimens are not necessarily morphologically comparable to tissue from other parts of the hearts
that were studied.
We used the maximal negative derivative of the electrograms as the instant of activation. By analyzing more than 30
test variables, Anderson et al. (30) have shown that this
derivative is one of three variables having the greatest potential for discriminating local from remote activation. They
showed that in some circumstances, remote activation may
induce electrograms that have a maximal negative derivative
greater than that of the local electrogram. In the study of
Anderson et al. recordings were made in atrial tissue at sites
close to the sulcus, where ventricular components may have
large amplitudes. Occurrence of this artifact in our superfused
preparations was unlikely, because there were no large differences in muscle mass. Nevertheless, we cannot rule out that in
incidental cases a remote component was erroneously selected
for determination of the activation time. Such artificial activation times will, however, only marginally affect the activation
patterns.
Conclusions. Our study shows that fractionated electrograms in dilated cardiomyopathy are caused by electrical
barriers. These barriers, consisting of strands of fibrous tissue
running parallel to the fiber orientation, were sometimes too
short to demonstrate a line of block in the activation pattern.
Fractionation depended on the site of stimulation and often
DE BAKKER ET AL.
FRACTIONATION IN CARDIOMYOPATHY
1077
disappeared or was less pronounced when propagation was
parallel to the barriers.
Activation delay was caused by wave fronts curving around
long barriers in concert with the wavy course of activation
between short barriers. The latter seems to mirror an extreme
increase in nonuniform tissue anisotropy.
References
1. Meinertz T, Hofmann T, Kasper W, et al. Significance of ventricular
arrhythmias in idiopathic dilated cardiomyopathy. Am J Cardiol 1984;53:
902-7.
2. Unverferth DV, Magorien RD, Moeschberger ML, Baker PB, Fetters J,
Leier CV. Factors influencing the one-year mortality of dilated cardiomyopathy. Am J Cardiol 1984;54:147-52.
3. Huang SK, Messer JV, Denes P. Significance of ventricular tachycardia in
idiopathic dilated cardiomyopathy. Am J Cardiol 1983;51:507-12.
4. Poll DS, Marchlinski FE, Buxton AE, Doherty JU, Waxman HL, Josephson
ME. Sustained ventricular tachycardia in patients with idiopathic dilated
cardiomyopathy: electrophysiologic testing and lack of response to antiarrhythmic drug therapy. Circulation 1984;70:451-6.
5. Milner PG, DiMarco JP, Lerman BB. Electrophysiologic evaluation of
sustained ventricular tachyarrhythmias in idiopathic dilated cardiomyopathy.
PACE 1988;11:562-8.
6. Poll DS, Marchlinski FE, Buxton AE, Josephson ME. Usefulness of programmed stimulation in idiopathic dilated cardiomyopathy. Am J Cardiol
1986;58:992-7.
7. Hano O, Mitsuoka T, Matsumoto Y, et al. Arrhythmogenic properties of the
ventricular myocardium in cardiomyopathic Syrian hamster, BIO 14.6 strain.
Cardiovasc Res 1991;25:49-57.
8. Weber KT, Pick R, Silver MA, et al. Fibrillar collagen and remodeling of
dilated canine left ventricle. Circulation 1990;82:1387-1401.
9. Li HG, Jones DL, Yee R, Klein G. Electrophysiologic substrate associated
with pacing-induced heart failure in dogs: potential value of programmed
stimulation in predicting sudden death. J Am Coil Cardiol 1992;19:444-9.
10. Rossner KL, Sachs H. Electrophysiological study of Syrian hamster hereditary cardiomyopathy. Cardiovasc Res 1987;12:436-43.
11. Einzig S, Detloff BLS, Borgwardt BK, Staley NA, Noren GR, Benditt DG.
Cellular electrophysiological changes in "round heart disease" of turkeys: a
potential basis for dysrhythmias in myopathic ventricles. Cardiovasc Res
1981;15:643-51.
12. Fein FS, Capasso JM, Aronson RS, et al. Combined renovascular hypertension and diabetes in rats: a new preparation of congestive cardiomyopathy.
Circulation 1984;70:318-30.
13. Unverferth DV, Baker PB, Swift SE, et al. Extent of myocardial fibrosis and
cellular hypertrophy in dilated cardiomyopathy. Am J Cardiol 1986;57:81620.
14. Sugrue DD, Holmes DR, Gersh BJ, et al. Cardiac histologic findings in
patients with life-threatening ventricular arrhythmias of unknown origin.
J Am Coll Cardiol 1984;4:952-7.
15. Maehashi N, Yokota Y, Takarada A, et al. The role of myocarditis and
myocardial fibrosis in dilated cardiomyopathy. Analysis of 28 necropsy cases.
Jpn Heart J 1991;32:1-15.
16. Roberts WC, Siegel RJ, McManus BM. Idiopathic dilated cardiomyopathy:
analysis of 152 necropsy patients. Am J Cardiol 1987;60:1340-55.
17. Rose AG, Beck W. Dilated (congestive) cardiomyopathy: a syndrome of
severe dysfunction with remarkably few morphological features of myocardial damage. Histopathology 1985;9:367-79.
18. Schaper J, Froede R, Hein ST, et aL Impairment of the myocardial
ultrastructure and changes of the cytoskeleton in dilated cardiomyopathy.
Circulation 1991;83:504-14.
19. Lo YSA, Billingham M, Rowan RA, Li H, Liem LB, Swerdlow CD.
Histopathologic and electrophysiologic correlations in idiopathic dilated
cardiomyopathy and sustained ventricular tachyarrhythmia. Am J Cardiol
1989;64:1063-6.
20. Olsen EGJ, Path FRC. The pathology of cardiomyopathies. A critical
analysis. Am Heart J 1979;98:385-92.
21. Vermeulen JT, McGuire MA, Opthof T, et aL Triggered activity and
1078
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
DE BAKKER ET AL.
FRACTIONATION IN CARDIOMYOPATHY
automaticity in ventricular trabeculae of failing human and rabbit hearts.
Cardiovasc Res 1994;28:1547-54.
Beuckelman D J, N~ibauer M, Erdmann E. Alterations of K + currents in
isolated human ventricular myocytes from patients with terminal heart
failure. Circ Res 1993;73:379-85.
Anderson K]?, Walker R, Urie P, Ershier PR, Lux RL, Karwandee SV.
Myocardial electrical propagation in patients with idiopathic dilated cardiomyopathy. J Clin Invest 1993;92:122-40.
Gardner PI, Ursell PC, Fenoglio JJ, Wit AL. Electrophysiologic and
anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation 1985;72:596-611.
de Bakker JMT, van Capelle FJL, Janse MJ, et al. Slow conduction in the
infarcted human heart. 'Zigzag' course of activation. Circulation 1993;88:
915-26,
Josephson ME, Wit AL. Fractionated electrical activity and continuous
electrical activity: fact or artifact? Circulation 1984;70:529-32.
Haissaguerre M, Gaita F, Fischer B, et al. Elimination of atrioventricular
nodal reentrant tachycardia using discrete slow potentials to guide application of radiofrequency energy. Circulation 1992;85:2162-75.
Jackman WM, Beckman KJ, McClelland JH, et al. Treatment of supraventricular tachycardia due to atrioventricular nodal reentry by radiofrequency
ablation of slow-pathway conduction. N Engl J Med 1992;327:313-8.
McGuire MA, de Bakker JMT, Vermeulen JT, Opthof T, Becket AE, Janse
MJ. The origin and significance of double potentials near the atrioventricular
node: correlation of extracellular potentials, intracellular potentials and
histology. Circulation 1994;89:2351-60.
Anderson KP, Walker R, Ershler PR, et al. Determination of local myocardial electrical activation for activation sequence mapping. A statistical
approach. Circ Res 1991;69:898-917.
Shimizu A, Nozaki A, Rudy Y, Waldo AL. Characterization of double
potentials in a functionally determined reentrant circuit. Multiplexing studies during interruption of atrial flutter in the canine pericarditis model. J Am
Coil Cardiol 1993;22:2022-32.
JACC Vol. 27, No. 5
April 1996:1071-8
32. Spach MS, Miller WT, Miller-Jones E, Warren RB, Barr RC. Extracellular
potentials related to intracellular action potentials during impulse conduction in anisotropic canine cardiac muscle. Circ Res 1979;45:188-204.
33. Wilensky RL, Tranum-Jensen J, Coronel R, Wilde AAM, Fiolet WT. The
subendocardial border zone during acute ischemia of the rabbit heart: an
electrophysiologic, metabolic, and morphologic correlative study. Circulation 1986;74:1137-46.
34. Veeustra RD, Joyner RW, Rawling DA. Purkinje and ventricular activation
sequences of canine papillary muscle. Circ Res 1984;54:500-15.
35. Kl6ber AG, Janse MJ, Wilms-Schopman FJG, Wilde AAM, Coronel R.
Changes in conduction velocity during acute ischemia in ventricular myocardium of the isolated porcine heart. Circulation 1986;73:189-98.
36. Dillon S, Allessie MC, Ursell PC, Wit AL. Influence of anisotropic tissue
structure on reentrant circuits in the epicardial border zone of subacute
canine infarcts. Circ Res 1988;63:182-206.
37. Spach MS, Miller WT, Dolber PC, Kootsey JM, Sommer JR, Mosher CE.
The functional role of structural complexities in the propagation of depolarization in the atrium of the dog. Cardiac conduction disturbances due to
discontinuities of effective axial resistivity. Circ Rcs 1982;50:175-91.
38. Spach MS, Dolber PC. Relating extracellular potentials and their derivatives
to anisotropic propagation at a microscopic level in human cardiac muscle.
Evidence for electrical uncoupling of side-to-side fiber connections with
increasing age. Circ Res 1986;58:356-71.
39. Cabo C, Pertsov AM, Baxter WT, Davidenko JM, Gray RA, Jalife J.
Wave-front curvature as a cause of slow conduction and block in isolated
cardiac muscle. Circ Res 1994;75:1014-28.
40. De la Fuente D, Sasyniuk B, Moe GK. Conduction through a narrow isthmus
in isolated canine atrial tissue--a model of the WPW syndrome. Circulation
1971;44:803-9.
4l. Rohr S. Determination of impulse conduction characteristics at a microscopic scale in patterned growth heart cell cultures using multiple site optical
recording of transmembrane voltage. J Cardiovasc Electrophysiol 1995;6:
551-68.