The Ultrastructure of the Heart
In Systole and Diastole
CHANGES IN SARCOMERE LENGTH
By Edmund H. Sonnenblick, M.D., John Ross, Jr., M.D.,
James W. Covell, M.D., Henry M. Spotnitz, M.D., and
David Spiro, M.D., Ph.D.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
ABSTRACT
The relations between ultrastructure and function have been examined in
the canine left ventricle under known hemodynamic conditions. Left ventricles
of normal dogs were fixed acutely by rapid perfusions of the coronary arteries
in either diastole or systole, and also following acute ventricular distension, or
potentiation of contraction. Sarcomere lengths at the midwall of the left ventricle averaged 2.07 fi at end diastole and 1.81 \L at end systole, changes that
are adequate to explain the degree of normal ventricular emptying. Following
marked ventricular emptying induced by postextrasystolie potentiation, sarcomeres shortened to an average of 1.60 ft; with acute ventricular distension
resulting from over-transfusion, average sarcomere length in diastole increased
to 2.25 fi. In each instance, sarcomere lengths correlated with the changes in
sarcomere length predicted from changes in the dimensions of a thick-walled
sphere. The importance of dispersion of sarcomere length within a given layer
of the ventricular wall was noted and its potential significance discussed.
ADDITIONAL KEY WORDS
Starling's law of the heart
myocardium
• The sarcomere comprises the basic ultrastructural unit of contraction in both skeletal
and heart muscle (1, 2). Recent studies in the
isolated cat papillary muscle preparation (3)
and in the passively distended cat and dog
left ventricle (4) have suggested that alterations of sarcomere length help to explain the
length-tension relation of heart muscle and, in
turn, form the basis of the Frank-Starling law
of the heart (5). However, no information is
available concerning the changes in sarcoFrom the Cardiology Branch, National Heart Institute, Bethesda, Maryland 20014 and the Department of Pathology, Columbia College of Physicians
and Surgeons, New York, New York 10032.
Dr. Sonnenblick's present address is Cardiovascular
Unit, Peter Bent Brigham Hospital, Boston, Massachusetts.
This paper was presented in part at the Annual
Meeting of the American College of Cardiology,
February 1966 and appeared in abstract form in
Am. J. Cardiol. 17: 139, 1966.
Accepted for publication August 15, 1967.
OrcuUtitm Rti—rcb, Vol. XXI, Octobtr 1967
heart muscle
dog
mere length which occur in the myocardium
in vivo during contraction or following ventricular dilatation.
The recent development of a method for
abruptly arresting the heart in various phases
of the cardiac cycle has now permitted a
direct approach to the problem of ultrastructure in relation to ventricular contraction (6).
It is the purpose of this study to define the
dimensions of the sarcomere in normal diastole and to delineate the changes in sarcomere
length which occur during systole. Further,
the changes which occur in sarcomere length
during either marked potentiation of contraction or following acute ventricular dilatation
have been explored.
Methods
Dog ventricles have been arrested and fixed in
either diastole or during systolic contraction by
the technique detailed in the preceding communication (6). In brief, glutaraldehyde fixative was
introduced into the coronary circulation by a
423
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
424
SONNENBLICK, ROSS, COVELL, SPOTNITZ, SPIRO
power syringe at a predetermined time during
diastole or systole (6). Only those ventricles
that were clearly fixed either at end diastole or
end systole were vised in this analysis. In two instances, ventricles were fixed by chance during
the potentiated beat following extrasystole.
Following fixation, the ventricle remained in
glutaraldehyde1 for 4 hr and was then transferred to isotonic phosphate buffer. A segment
of the anterior wall of the left ventricle was then
removed for electron microscopic study. This
sample was always taken from a site halfway between the apex and base of the heart, 1 cm to
the left of the left anterior descending coronary
artery (Fig. 1). Sections of 1 mm8 were then selected from this tissue midway between the epicardium and endocardium; in this region circumferential fibers tend to be perpendicular to the
long axis of the ventricle (8). The tissue was
trimmed under a dissecting microscope to assure
orientation of fibers within the plane of sectioning, and it was then fixed in 2% osmium with
sucrose for 30 min. Subsequently, the tissue was
dehydrated progressively in acetone and imbedded in epon2 in a manner previously described (3). Shrinkage with similar techniques
has been estimated to be less than 5% (7). Ultrathin sections of the imbedded material were prepared using an LKB microtome and the sections
were stained with lead citrate. Care was taken to section the tissue with the knife edge perpendicular to the long axes of the fibers to avoid
artifacts resulting from tissue compression (3).
The sections were then examined under a calibrated RCA EMU 3 electron microscope. Overall magnification was generally 20,000. A minimum of 25 electron micrographs was obtained
for each heart. Sections of distinct tissue blocks
were compared as a precaution against sampling
errors. Sarcomere lengths (measured from the
center of the Z line to the adjacent Z line) (Fig.
2) were determined on all photomicrographs and
between 100 and 200 measurements were taken
to obtain an average sarcomere length for the
mid-ventricular wall of each heart.
solution in isotonic phosphate buffer, pH 7.6.
Fluka, A. G., Chemische Fabrik, Buchs, S. G.,
Switzerland.
2
ENDOCARDIUM
EPICAIDIUM
FIGURE 1
Diagram of the heart to show the portion of the left
ventricular wall that was analyzed for sarcomere
lengths. A segment was removed from the left ventricular wall as shown and the central cube (dashed lines)
studied.
Results
Of 46 dogs studied, the hearts from 22 were
judged satisfactory for the present analysis,
having been fixed at an appropriate time during contraction. The over-all results are summarized in Table 1.
Six hearts were fixed in diastole with an
average filling pressure of 8 mm Hg (range 2
to 12 mm Hg) and an average intraventricular
volume of 51.6 ml (range 40.0 to 66.5 ml). A
typical electron micrograph from a heart fixed
in diastole is shown in Figure 2. Sarcomere
length is indicated by the distance between
two Z lines; in this instance it was 2.07 fi. In
hearts fixed in diastole, the sarcomere length
from the left ventricular midwall averaged
2.07 ±0.024 (SD) fi (range 2.03 to 2.10 fi)
(Table l a n d Fig. 3).
In the 8 ventricles fixed in systole, the intraventricular volume averaged 21.4 ml
(range 17.0 to 29.9). The end-diastolic pressure prior to fixation averaged 6 mm Hg and
was not significantly different from that of the
ventricles fixed in diastole, which averaged
8 mm Hg. In the ventricles fixed in systole,
left ventricular midwall sarcomere length averaged 1.81 ±0.096 fi (range 1.68 to 1.91 fi).
A typical electron micrograph taken from a
left ventricle fixed in systole is shown in Figure 4. Of note, in ventricles fixed in systole,
sarcomeres less than 1.8 fi in length were commonly observed and in these sarcomeres an
additional band at the center of the A band
termed the A contraction band could usually
be perceived (Fig. 4). In 2 hearts fixed durCinuUuo* Rutarcb, Vol. XXI, Oacbtr 1967
425
ULTRASTRUCTURE OF THE CONTRACTING HEART
TABLE 1
Correlations between the Intact Left Ventricle and Sarcomere Dimensions
LV
S/D
Expt. no.
(mm Hf)
Fixation
prosure
(mm HgHe)
LV
wt
LV
vol
(g)
(ml)
97
106
107
80
82
101
95.5
5.28
62.1
40.0
68.5
47.8
45.0
48.5
51.6
4.62
2.09
2.07
2.08
2.10
2.03
2.06
2.07
.01
0.059
0.017
0.027
0.037
0.032
0.038
.035
20.8
20.0
29.9
25.0
18.0
17.0
21.5
19.0
21.4
1.6
1.73
1.91
1.68
1.82
1.84
1.74
1.86
1.87
1.81
.03
0.024
0.040
0.036
0.010
0.033
0.038
0.044
0.047
0.034
8.7
6.9
1.62
1.58
0.022
0.048
64.5
54.8
82.0
77.2
71.3
83.0
72.1
4.9
2.29
2.29
2.25
2.32
2.13
2.27
2.25
.03
0.051
0.058
0.078
0.043
0.040
0.036
0.051
Sarcomere length (fl)
SD
Mean
Diastole
17
41
42
43
44
46
Average
112/13
135/4
105/4
95/8
90/7
105/7
107/7
SE
12
2
10
11
11
7
8
1.68
Systole
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
3
6
7
22
23
24
25
27
Average
150/7
100/9
115/8
110/6
112/5
105/1
106/5
105/6
129/6
125
105
102
103
95
108
90
91
102.4
4.26
SE
Potentiated systole
21
26
105/4
115/8
102
86
Dilated distole
5
15
16
18
19
20
Average
75/30
95/25
75/40
110/23
67/33
55/20
80/29
SE
32
32
40
24
37
25
32
2.8
96
120
117
115
106
96
108.3
4.8
Abbreviations: LV = left ventricular; S/D = systolic/end diastolic pressure.
ing postextrasystolic potentiation, intraventricular volume decreased to an average of
7.2 ml while average sarcomere length decreased to 1.60 fi.
By utilizing the calculated changes in the
midwall circumference of the left ventricle
between diastole and systole as outlined previously (6), it was possible to calculate the
changes in sarcomere length which should
have accompanied changes in ventricular volume. Such a line of prediction is shown on
CircnUtion Kuurcb. Vol. XXI, Octobtr 1967
Figure 3. Using the average sarcomere length
at normal diastolic volume as the midpoint for
calculation, the 59% decrease in ventricular
volume which was observed in systole should
have been accompanied by a 17? decrease in
sarcomere length. Experimentally, a 13% average decrease in sarcomere length was observed. In Figure 3, the close approximation
between predicted sarcomere lengths (dashed
line) and those sarcomere lengths observed
experimentally between systole and diastole
426
SONNENBLICK, ROSS, COVELL, SPOTNITZ, SPIRO
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
FIGURE 2
Electron micrograph from a left ventricle fixed in diastole (Expt. 41). The A and I bands of the
sarcomere are noted on the figure. Sarcomere length is represented by the distance between
two Z lines. M = mitochondrion. A 1-p marker is ihown on the figure.
^
*\h
aottolt
is apparent. Those hearts fixed during postextrasystolic potentiation had the smallest intraventricular volumes (average 7.2 ml).
Nevertheless, changes in sarcomere length
followed the line of prediction for sarcomere
length relative to intraventricular volume that
was previously calculated (Fig. 3).
Following acute dilatation of the left ventricle induced by over-transfusion of the dog,
- OToltd
' mttjn * I
f
at
o
'
u
£
/
y J-o
o
3 •
FIGURE 3
PRCOtCTEO
z
LLI
>
t
*>-
1.6
L7
L8
L«
SARCOUERE
2.0
2J
LENGTH
2.2
-
u
2.3
The relation of left ventricular volume to average
midwdIX sarcomere length for diastolic, dilated diastolic, systolic and potentiated systolic left ventricles.
The dashed line represents the predicted changes in
midwaU sarcomere length that should have accompanied these changes in intraventricular volume,
assuming the left ventricle to be a thick-walled sphere
of constant mass.
CircuUiiou Rtsurcb, Vol. XXI, Octobtr 1967
427
ULTRASTRUCTURE OF THE CONTRACTING HEART
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
FIGURE 4
An electron micrograph of a ventricle fixed during systolic contraction (Expt. 3). The I bands
have narrowed considerably from, diastole (see Fig. 2), while the A band has remained unchanged. In these shortened sarcomeres a second dark band, the A contraction band, has
appeared at the center of the A band as denoted by the dark vertical lines with arrows. The
distance from the Z line to the far edge of the A contraction band (dashed arrow) is 1 n.
left ventricular volume increased to an average of 72.1 ml (range 54.8 to 83.0 ml) while
the diastolic filling pressure increased to an
average of 32 mm Hg (range 24 to 40 mm
Hg). Sarcomere length in the mid-ventricular
wall increased to an average of 2.25 ± 0.061
fi. (range 2.13 to 2.32 /x) (Fig. 5). This 8.7$
increase in average sarcomere length correlates well with the 8.6% change in sarcomere
length which would be predicted from the
change in midwall circumference (Fig. 3).
In the dilated left ventricle, H zones were
commonly observed at the center of the A
bands of the sarcomeres as shown in Figure 5,
but only in sarcomeres whose length was in
excess of 2.20 \i. The H zones appeared as
lighter areas in the central portion of the A
band; they were irregular and difficult to deCircmUHon Rmsurcb, Vol. XX1, Octobtr 1967
fine precisely, even in longer sarcomeres (>
2.25 p).
It was consistently found that within the
midwall of the left ventricle there was considerable variation in sarcomere lengths. In Figure 6 the dispersion of sarcomere lengths is
shown for a left ventricle fixed in diastole
(Fig. 6A), systole (Fig. 6B) and in diastole
following acute dilatation (Fig. 6C). The dispersion of sarcomere lengths about the mean
values (vertical arrows in Fig. 6) was similar
in normal systole and diastole. However, with
acute left ventricular dilatation, the dispersion
of sarcomere lengths was somewhat greater,
as is apparent from Figure 6C.
Discussion
In the present study, hearts were fixed in
428
SONNENBLICK, ROSS, COVELL, SPOTNITZ, SPIRO
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
FIGURE 5
Electron micrograph of the left ventricle fixed in diastole following acute dilatation (Expt. 18).
Sarcomeres with lengths in excess of 2.40 y. are shown and the H zone at the center of the A
band is seen.
situ and the changes in sarcomere length
that occur in the normal heart during the cardiac cycle were defined. Further, the changes
in sarcomere length that occur with either
marked potentiation of contraction or with
acute left ventricular dilatation have been
delineated. In normal diastole, with an average left ventricular filling pressure of 6 mm
Hg, sarcomere length averaged 2.07 fi with a
range from 2.03 to 2.10 \i. Previous studies
in vitro have shown that maximum active tension is developed by the isolated cat papillary
muscle preparation with a sarcomere length
of 2.20 /z, and that actively developed tension
decreases along with sarcomere length as initial muscle length is shortened (3). Further,
in the arrested and passively distended dog
and cat left ventricle, it has been demonstrated that sarcomere length is a function of
filling pressure and that a sarcomere length of
2.20 JX corresponds to a ventricular filling
pressure of 12 to 15 mm Hg. Thus, the present
study indicates that in the normal contracting
ventricle, with sarcomeres of 2.03 to 2.10 /JL,
contraction ensues with sarcomeres on the
ascending limb of their length-active tension
curve, somewhat below the apex of this curve.
The increase in sarcomere length between the
observed 2.07 \i, sarcomere length and the
2.20 fi at the apex of the sarcomere lengthtension curve permits increased muscle shortening for a given load and represents reserve
provided by the Frank-Starling mechanism
(9), although it should be pointed out that
more reserve may be available in the intact
animal than is indicated in these experiments
performed in open-chest animals.
During normal systolic contraction, which
CircuUtioo Riiurlb, VoL XXI, Oaabt 1967
ULTRASTRUCTURE OF THE CONTRACTING HEART
60
1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
60
D/LATED
O/ASTOL/C
4 0
I
$
20
1.6 1.7 1.6 1.9 2.0 2.1 2.2 2.3 2.4 2.5
SARCOMERE LENGTH /u
FIGURE 6
Examples of the nomograms of sarcomere frequency
in a diastolic (A), a systolic (B) and a dilated diastolic
(C) left ventricle. The vertical heavy arrows indicate
the average sarcomere length.
was initiated from left ventricular end-diastolic pressures approximately the same as
those of ventricles studied in diastole, average sarcomere length decreased to 1.81 FI.
This 13% decrease in sarcomere length is appropriate to explain the observed decreases
in intraventricular volume that occurred during systole. As seen in Figure 3, the close
Circulation Research, Vol. XXI, October 1967
429
approximation of changes in midwall circumference of the left ventricle and in the sarcomere length with changes in ventricular volume
supports the view that changes in sarcomere
length can explain observed stroke volume/
end-diastolic volume relations in the intact
heart (5).
The length of sarcomeres in the ventricle
yields added information when related to the
disposition of myofilaments within the sarcomere. It is now well recognized in both
skeletal (1) and heart (2) muscle that the
sarcomere is composed of two sets of partially
overlapping filaments. The thin actin filaments
which extend from the Z lines, through the I
band and into the A band are 1.0 /JL in length,
while the thicker filaments of myosin, which
are 1.5 fx in length (3), extend the length
of the A band. The length of these filaments
is constant despite changes in sarcomere
length (1, 2, 7), a fact that supports the sliding
filament hypothesis for muscle contraction.
In diastole, with a sarcomere length of 2.07
fx, thin actin filaments lie near the center of
the A band. During systole, these thin filaments move into the center of the sarcomere
and begin to bypass one another as the sarcomere length becomes less than 2.0 ft. In the
presence of a potentiated contraction, the
thin filaments may penetrate the opposite
half of the A band far enough to form a
prominent additional dark band flanking the
center of the A band, termed an A contraction
band. As noted in Figure 4, the distance
from the Z line to the edge of the A contraction band in the opposite half of the sarcomere
is 1.0 ix, which is equal to the length of the
actin filament. Thus, although such extensive
shortening of the sarcomere has been known
to exist in heart muscle in vitro (2, 3), the
documentation of this phenomenon in the
intact heart demonstrates the range of sarcomere shortening that may occur physiologically, i.e., from 2.20 /u. to about 1.60 \x and
shows that a double overlapping of thin filaments at the center of the sarcomere may
occur during potentiated contractions. Further, such a 30% reduction in sarcomere length
theoretically could permit stroke volume to
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
430
SONNENBLICK, ROSS, COVELL, SPOTNITZ, SPIRO
end diastolic volume ratios of 75% or more
during markedly potentiated contractions.
Following ventricular dilatation, thin filaments may be withdrawn from the center of
the A band creating an H zone. The production of H zones, with a decrease in the overlap
of thick and thin filaments, could in itself
result in a decrease in force development,
since force depends on the degree of interdigitation of the two sets of myofilaments
(1, 2, 10). However, such H zones were only
observed with marked acute ventricular distension, at ventricular filling pressures beyond
those usually compatible with life; it would,
therefore, appear unlikely that the production
of H zones plays a primary role in the etiology
of myocardial decompensation.
The demonstration of dispersion of sarcomere length presents certain interesting problems in the relation of ultrastructure to
ventricular performance. It is evident that
effective sarcomere length is a derived value
in the intact heart and that the force generated at any given time will reflect the
average sarcomere length. Further, this finding suggests that tension may not be uniformly
distributed across all sarcomeres in the ventricle in a similar manner, even within the
same layer of the wall, or indeed, even within
a given fiber. The possibility exists that some
of the dispersion of sarcomere lengths during
systole may reflect some delay in fixation,
which could permit some sarcomeres to begin
to relax and lengthen. Although this possibility cannot be excluded, it would not explain
the dispersion of sarcomere length in diastole,
nor the finding of sarcomere length dispersion
in the totally excised ventricle (4, 11).
In the dilated ventricle the interpretation
of sarcomere dispersion is even more complex.
Following acute overdistension of the left
ventricle, average sarcomere length commonly
exceeded 2.20 fi, a value at the apex of the
sarcomere length-active tension curve. However, a considerable portion of the sarcomeres
were shorter than 2.20 p even in the presence
of ventricular dilatation. Thus, with contraction it would be anticipated that the sarcomeres at the apex of the curve (being
strongest) would tend to extend further the
already overstretched, and therefore weakened, sarcomeres. This might tend to increase
the sarcomere dispersion further.
Following acute ventricular dilatation, there
was no disparity of theoretical and observed
sarcomere lengths, a phenomenon previously
observed in the passively overdistended left
ventricle and ascribed to possible fiber slippage or shear within the wall (4). In the
present study the changes in sarcomere length
follow the line of prediction for changes in
midwall circumference and are entirely adequate to explain observed changes in ventricular volume (Fig. 3). Differences in
methodology may help to explain this discrepancy, since in the studies on the passive
ventricle, multiple pressure-volume curves
were obtained prior to fixation (4). This
procedure could have induced progressive
stress-relaxation and fiber slippage, and the
lack of active contraction in these previous
studies could have prohibited compensation
for excessive distension.
Acknowledgment
The excellent technical assistance of Miss Nancy
Dittemore and Major Reginald Reagan is gratefully
acknowledged.
References
1. HUXLEY, H. E.: Muscle cells. In The Cell,
vol. IV, edited by J. Brachet and A. E. Mirsky.
New York, Academic Press, 1960, pp. 365481.
2. SPIRO, D., AND SONNENBUCK, E. H.: Compari-
son of the ultrastructural basis of the contractile process in heart and skeletal muscle.
Circulation Res. 15 (suppl. II): 11-14, 1964.
3. SONNENBLICK, E. H., SPIRO, D., AND COTTKELX,
T. S.: Fine structural changes in heart muscle in relation to the length-tension curve.
Proc. Natl. Acad. Sci. U.S.A. 49: 193, 1963.
4. SPOTNITZ, H.
M.,
SONNENBLICK, E.
H., AND
SPIRO, D.: Relation of ultrastructure to function in the intact heart: Sarcomere structure
relative to pressure volume curves of intact
left ventricles of dog and cat. Circulation
Res. 18: 49, 1966.
5. SONNENBUCK, E. H., SPIBO, D., AND SPOTNTTZ,
H. M.: Ultrastructural basis of Starling's law
of the heart. The role of the sarcomere in
determining ventricular size and stroke volume. Am. Heart J. 68: 336, 1964.
Cimdttm
Rtsurcb, Vol. XXI, Oaobtr 1967
ULTRASTRUCTURE OF THE CONTRACTING HEART
6. Ross, J., JR., SONNENBLICX, E. H., COVELL,
J. W., KAISER, G. A., AND SPIBO, D.: Archi-
tecture of the heart in systole and diastole:
Technique of rapid fixation and analysis of
left ventricular geometry. Circulation Res.
21: 409, 1967.
7. PAGE, S., AND HUXLEY, H. E.: Filament lengths
in striated muscle. J. Cell. Biol. 19: 369,
1964.
8. STREETEB, D. D., JR., AND BASSETT, D. L.: An
engineering analysis of myocardial fiber orientation in pig's left ventricle in systole. Anat.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
CircmUiio* Rtturcb,
Vol. XXI, October 1967
431
Rec. 155: 503, 1966.
9. STAMJNC, E. H.: Linacre Lecture on the Law
of the Heart, Cambridge, 1915. London,
Longmans Green and Co., Ltd., 1918.
10. GORDON, A. M., HUXLEY, A. F., AND JULIAN,
F. J.: The variation in isometric tension with
sarcomere length in vertebrate muscle fibers.
J. Physiol. 184: 170, 1962.
11. HORT, W.: Makroskopische und mikrometrische
Untersuchungen am Myokard verschieden stark
gefiillter linker Kammem. Virchows Arch. Path.
Anat. Physiol. Klin. Med. 333: 523, 1960.
The Ultrastructure of the Heart in Systole and Diastole: >Changes In Sarcomere Length
EDMUND H. SONNENBLICK, JOHN ROSS, Jr., JAMES W. COVELL, HENRY M.
SPOTNITZ and DAVID SPIRO
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Circ Res. 1967;21:423-431
doi: 10.1161/01.RES.21.4.423
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1967 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/21/4/423
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/