Download Death in Tecumseh, Michigan

Predisposing Factors in Sudden Cardiac
Death in Tecumseh, Michigan
A Prospective Study
By BENJAMIN N. CHIANG, M.D., LAWRENCE V. PERLMAN, M.D.,
MARY FULTON, M.B., M.R.C.P.E., LEON D. OSTRANDER, JR., M.D.,
AND
FREDERICK H. EPSTEIN, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
SUMMARY
In an epidemiologic study of the total population of Tecumseh, Michigan, 98 deaths
from coronary heart disease were observed between 1959 and 1965. Forty-five of the
fatalities occurred within 1 hour of the onset of symptoms and were classified as
sudden. The proportion of sudden deaths among men was nearly twice that among
women, and the incidence increased progressively with age in both sexes.
Hypertensive heart disease, coronary heart disease, or diabetes mellitus had been
detected on prior examination in 62% of those who died suddenly. Physiologic abnormalities associated with a high risk of coronary heart disease were also found more
frequently than in the total Tecumseh population.
All but seven of the persons who died suddenly had abnormalities including arrhythmias and conduction defects which were detected in the standard 12-lead electrocardiogram.
Although sudden deaths often seem to occur without warning, the victims are
predisposed by conditions which are detectable long before the catastrophic event.
The identification of conditions which are precursors of sudden death from coronary
heart disease permits a rational consideration of possible preventive measures.
Additional Indexing Words:
Epidemiologic study
Hypertension
Coronary heart disease
Risk factors
Electrocardiographic abnormalities
hospital." 2 Potential victims must be identified as the first step in the development of a
preventive program. Although sudden deaths
often occur seemingly without warning, predisposing conditions which were detectable
before the catastrophic event are usually
found.3 4 This is a report of the antemortem
findings among 45 persons who died suddenly
during a 6-year period in Tecumseh, Michigan.
RECENT ADVANCES in coronary care
have reduced the immediate mortality
from myocardial infarction within hospitals,
but a large proportion of stricken individuals
die so suddenly that they do not reach the
From the Department of Epidemiology, University
of Michigan School of Public Health, Ann Arbor,
Michigan.
Study was supported by the Cardiovascular
Research Center, University of Michigan, under
Program Project Grant HE-6378 from the National
Heart Institute, U. S. Public Health Service.
Dr. Chiang is a recipient of an International Postdoctoral Research Fellowship (5-FOJ-TW 1025) and
Dr. Epstein of a Research Career Award (HE-K66748) from the National Institutes of Health, U. S.
Public Health Service.
Circulation, Volume XLI, January 1970
Diabetes mellitus
Address for reprints: Dr. Frederick H. Epstein,
Department of Epidemiology, School of Public
Health, University of Michigan, Ann Arbor, Michigan
48104. Dr. Fulton is a Visiting Scientist from the
University of Edinburgh, Scotland.
Received May 13, 1969; accepted for publication
September 10, 1969.
31
CHIANG ET AL.
32
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
wvere added to the upper quintile, so that the
hyperglycemic segment included 22% of men and
24% of women past 40 years of age.
The data on 3,643 persons 30 years of age or
older included information on all the deaths from
coronary heart disease which occurred during the
first 6 years of surveillance. Sudden coronary
death was defined as any fatality which occurred
outside of the hospital and within 1 hour of the
onset of symptoms. Deaths due to violence or any
suspected cause other than coronary heart disease
were excluded. This information was carefully
collected, documented, and summarized. Death
certificates were reviewed for each participant
who died, and all obtainable information from
physicians' notes, hospital records, statements
from relatives, and coroners' reports was also
examined. Only three of those who died suddenly
had autopsies and in each case the postmortem
examination revealed extensive coronary artery
atherosclerosis and occlusion.
Methods
The Tecumseh study is a prospective epidemiologic investigation of nearly the total population
of the community. The details of the study have
been described previously.5-8 During the initial
examination of the population in 1959 and 1960,
8,641 participants (88% of the total population)
gave a detailed medical histoiy and had a
thorough physical examination, biochemical determinations, anthropometric measurements, chest
roentgenograms, and standard 12-lead electrocardiograms.
Initial clinical diagnoses of diseases were made
by the examing physicians and then reviewed and
classified according to strict criteria by senior staff
physicians. Electrocardiograms were recorded on
a direct-writing instrument at a paper speed of 50
mm/sec and coded according to the system of
Blackburn and associates.9 Serum cholesterol was
determined by the Abell-Kendall method'0 and
blood glucose by the Somogyi-Nelson method."
Relative weight was computed from the ratio of
actual weight to predicted weight, derived from a
regression equation involving height, bi-acromial
diameter, bicristal diameter, and sex.'2 Elevated
values for blood pressure, serum cholesterol, and
relative weight were defined according to the age
and sex-specific upper quintiles of the distributions of the variables in the population. Except
for the first 22% of persons studied in the 19591960 series of examinations, blood samples were
taken 1 hour after a 100-g oral glucose challenge.
Casual specimens were drawn from all unchallenged adults including known diabetics.
Eighty-five adults were classified as diabetics
because of earlier blood glucose elevations
detected and treated by their personal physicians.
Examined persons were grouped according to
age, sex, and type of test, either casual or
postchallenge, and ranked within their group in
order of blood glucose level. Hyperglycemia was
arbitrarily defined as any value above the 80th
percentile for each group. The known diabetics
Results
Incidence of Sudden Death
During the period from 1959 to 1965, 98
died of coronary heart disease and 45
of the deaths were sudden. Most of the
victims (29 men and 16 women) died at
home, but some were at work, traveling, or on
the way to the hospital. The incidence of this
event increased progressively with age from 2
per 1,000 at ages 30 to 39 to 52 per 1,000
among those past 70 (table 1).
persons
Antecedent Illness
The
presence
of
one
of three clinical
diseases-diabetes mellitus, hypertensive heart
disease, or coronary heart disease-at initial
examination of persons past age 30 was
associated with an excessive incidence
Table 1
Sudden Deaths by Age at Initial Examination in Tecumseh, Michigan (1959-1965)*
Number
Age
examined
No.
30 - 39
40 - 49
50- 59
60 - 69
1381
941
658
374
289
3
6
8
13
1a
3643
43
70+
Total
Six-year incidence
of sudden deaths
Sudden deaths
pei- 1,000 popllation
6.7
13.3
17.8
28.9
33.3
100
2
6
12
35
52
12
*Sudden death defined as death ocurring within 1 hour of onset of symtoms outside of hospital.
Circulation, Volume XLI, January
1970
33
SUDDEN CARDIAC DEATH
Table 2
Incidence of Sudden Death Among Persons 30 Years of Age or Older with Cardiovascular
Disease or Diabetes in the Tecumseh Population (1959-1965)
Clinical diagnosis
Coronary heart disease (CHD)t
Diabetes mellitus (DM)
Hypertensive heart disease (HHD)t
None of the above conditions
6-year incidence
per 1,000
population
No. observed
at initial exam
No. sudden deaths*
166.7
94.1
89.5
5.0
108
85
67
3,387
18
8
6
17
*Four persons had multiple conditions: HHD + CHD (1), HHD + DM (1), CHD + DM (2).
tProbable diagnosis of angina pectoris or myocardial infarction.
$Systolic BP 5 160 mm Hg or diastolic BP 5 95 mm Hg and left ventricular hypertrophy in
ECG (Minnesota Code III,).
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
rate is 15-fold or greater among persons with
at least one of the three antecedent illnesses
Table 3
Expected and Observed Numbers of Five Physiologic Variables in the Highest Quintile (Specific
for Age and Sex) Among Persons Who Died
Suddenly in Tecumseh, Michigan (1959-1965)
than
2).
19t
17t
15t
10
16t
unaffected individuals (table
Risk Factors
Sudden death N = 45
(highest quintile*)
Observed
Expected
Systolic blood pressure
Diastolic blood pressure
Relative weight
Serum cholesterol
Blood glucose§
among
A higher proportion of the persons with
sudden fatalities had systolic or diastolic
hypertension, high relative weight, or hyperglycemia than the Tecumseh population as a
whole, but there was no excess of hypercholesterolemia (table 3). The smoking habits of
those who died suddenly were similar to those
of the general Tecumseh population. The
proportion of multiple risk factors among the
sudden fatalities was significantly higher than
among the total adult population (table 4).
Fifty-one per cent of the persons who died
suddenly had two or more risk factors as
compared with 27% in the total population
past age 30. Conversely, 41.3% of the total
population had none of the five risk factors as
9
9
9
9
10.4
*Upper quintile, age and sex-specific.
tDifference significant at P < 0.01.
tDifference significant at P < 0.05.
§Including upper quintile and known diabetes.
rate of sudden death as compared with
persons who had none of these conditions
(table 2). Eighteen of 108 persons with
coronary heart disease, eight of 85 with
diabetes, and six of 67 with hypertensive heart
disease died suddenly. Thus, the incidence
Table 4
Multiple Risk Factors* at Initial Examination in Persons Who Died Suddenly over a
6-Year Period in Tecumseh Population
Sudden death
(N = 45)
None
One factor
Two factors
Three or more factors
N
%
6
16
16
7
13.3
35.5
35.5
15.7
Total population (age >30)
(N = 3,643)
N
%
1505
1148
604
386
41.3
31.5
16.6
10.6
*Systolic and diastolic pressures, cholesterol, relative weight, and blood glucose in the upper
quintile, age and sex specific.
Circulation, Volume XLI, January 1970
CHIANG ET AL.
34
Table 5
Antecedent ECG Findings in 45 Persons Who Died Suddenly over Six-Year Period
(1959-1965) in Tecumseh Population
ECG findings
Q pattern
Left axis deviation
High R amplitude
S-T segment depression
II
III
III,
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
5
8
6
6
5
2
3
5
9
3
IV,
IV2
IV3
Vi
V2
V3
VI3
T-wave changes
First degree A-V block
Ventricular conduction defect
Complete left BBB
Complete right BBB
Incomplete right BBB
Arrhythmia
Premature systoles
Atrial fibrillation
Normal
compared to 13.3% in the sudden death
(table 4).
No. of findings
Minnesota code
VIII
VII2
5
3
1
VIL3
Vill,
11
1
7
VIII3
0
group
Electrocardiographic Abnormalities
Thirty-eight of the 45 persons who subsequently died suddenly had codable electrocardiographic findings at entrance into the study
(table 5). Conduction defects, rhythm distur-
bances, ventricular hypertrophy, and old
myocardial infarction were particularly frequent antecedents of sudden death (table 6).
Three of four persons with "bilateral bundlebranch block" (complete right and partial left
bundle-branch block manifest by left axis deviation beyond - 30° in the frontal plane), five
Table 6
Relationship of Selected Antecedent ECG Abnormalities to Incidence of Sudden Death in the Tecumseh
Population (30 Years of Age or Older)
Six-year
ECG findings
Bilateral BBB syndrome
Left BBB
Old myocardial infarction
Ventricular premature beats
Left ventricular hypertrophy
First degree A-V block
Normal ECG (no codable items)
Minnesota
code
III + VII2
VII,
II
VIII,
III,
VI,
None
incidenee
of sudden
death per
No. sudden death
3
5
5
10
6
3
7
(68.6)*
(59.6)
(62.6)
(65.1)
(74.1)
(57.6)
(49.8)
No. observed in total
1,000
population
population
4
18
26
165
124
100
2,700
(69.3)
(63.2)
(60.5)
(57.6)
(58.1)
(53.4)
(45.3)
750
277
192
61
48
30
2.6
*( ) Mean age in years of the group.
Circulation, Volume XLI, January
1970
35
SUDDEN CARDIAC DEATH
of the 26 persons with old myocardial
infarct, five of 18 with left bundle-branch
block, 10 of 165 with ventricular premature
beats, six of 124 with left ventricular hypertrophy, and three of 100 with first degree A-V
block, died suddenly in the 6-year period.
Conversely, only seven of 2,454 persons in the
population who had normal ECGs died suddenly in the same period.
Multiple Conditions
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
Most of the 45 victims had more than one
antecedent abnormality. Only one person had
no evidence of coronary or hypertensive heart
disease, diabetes, coronary risk factors, or
electrocardiographic abnormalities (as defined
in table 5) at the time of the initial
examination.
Discussion
The reported frequency of sudden death as
the initial manifestation of coronary heart
disease varies from 20% to 25%.3' 13 14 Careful
investigation of such cases reveals that few
individuals were free of significant predisposing abnormalities prior to death. In a series of
1,098 sudden deaths reported by Kuller,'3 only
8% of the persons had no history of diabetes,
heart disease, hypertension, or stroke, had not
seen a physician within a month prior to
death, or had not been hospitalized during the
year before death. The reports of several large
series support the impression that most
sudden deaths among white Americans are
due to coronary heart disease. In a study of
1,348 sudden deaths, Myerburg and Davis'5
found that 25.8% had a definite history of
coronary heart disease, and another 33.4% had
suggestive symptoms. In a large autopsy study
from Westchester County, New York, Spain et
al.'6 reported that 90% of sudden fatalities
resulted from coronary artery disease.
All 45 sudden deaths in Tecumseh occurred
outside the hospital and within 1 hour of the
onset of symptoms. All obtainable information
from physicians' notes, prior hospital records,
witnesses to the event,
coroners'
reports, and
reviewed in addition to the
death certificates. Because postmortem examinations of sudden fatalities are infrequent in
autopsy data
were
C-r'culation, Volume XLI,
January
1970
Tecumseh, coronary heart disease is strongly
suspect but usually unproven as the cause of
such deaths among the study population.
Deaths known or suspected to have occurred
from conditions other than coronary heart
disease were excluded from the analysis.
Only a few prospective studies of predisposing factors among sudden fatalities have been
reported. In general, the same nsk factors
(hypertension, hypercholesterolemia, overweight, diabetes, cigarette smoking, and physical inactivity) which predispose to other
manifestations of coronary disease have been
associated with sudden death.3 13 The reported incidence increases with age and shows
a marked male predominance.3 4,13 In the
Framingham study, high serum cholesterol
concentrations and electrocardiographic evidence of left ventricular hypertrophy were
associated with sudden deaths.3 17 Pell and
D'Alonzo,4 in a study of employees of a large
company, found that immediate mortality
from acute myocardial infarction was associated with increasing age, antecedent hypertension, and prior electrocardiographic abnormalities such as conduction defects and
ventricular premature beats. Morris and coworkers'8 reported that physical inactivity was
related to sudden death. Cigarette smoking
has also been implicated in sudden demise
from coronary heart disease.19
In Tecumseh, 45.9% of coronary heart
disease fatalities were sudden. Although intensive coronary care may reduce early mortality
in the hospital, a large proportion of coronary
fatalities occur beyond such assistance. Most
of those who died suddenly in Tecumseh had
hypertension, diabetes, coronary heart disease,
or electrocardiographic abnormalities which
were recognized long before death (tables 2
and 5). Only one of the 45 victims was free of
recognizable antecedent disease, predisposing
risk factors, or abnormalities detected in the
electrocardiogram at the time of the initial
examination.
Risk factors which predispose to other
manifestations of coronary heart disease were
also closely associated with sudden deaths in
Tecumseh (table 3). Those who died sudden-
36
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
ly had the greatest number of risk factors,
with hypertension, hyperglycemia, and overweight predominant. Multiple risk factors
were detected more frequently among the
sudden fatalities than in the total Tecumseh
population (table 4). Hypercholesterolemia
was no more prevalent among those who died
suddenly than in the total population.
The risk factor analyses in Tecumseh, which
are based on age and sex-specific levels rather
than fixed cutting points, may minimize the
biologic importance of conditions whose median and upper quintile values tend to be in a
high range among certain groups. The serum
cholesterol is an example. The mean cholesterol
concentration for the persons who died
suddenly was 235 mg%, and 35% had values of
250 mg% or higher, which are far in excess of
"ideal" concentrations. On the other hand,
values for the total Tecumseh population of
like age and sex are similarly high, so that
discrimination is not improved by the use of
fixed cutting points unless one ignores age and
sex differences. Such a maneuver would,
therefore, not be helpful. Also, the association
between serum cholesterol level and coronary
heart disease incidence diminishes with age,
so that the present findings among persons,
many of whom are older, should not be taken
to apply to sudden deaths in general.
There was no excess of cigarette smokers
among participants who died suddenly. Analysis of the 1959-1960 examination results in
Tecumseh revealed that cigarette smoking was
no more prevalent among persons with
coronary disease than in the general Tecumseh population,6 but smoking was significantly
associated with the incidence of new disease
during the succeeding years, particularly
among persons less than 65 years of age.20 It is
not surprising, then, that the smoking habits
of the sudden death group, who were
generally elderly and had a high prevalence of
known heart disease, were similar to those of
unaffected members of the Tecumseh population of like age and sex.
The immediate cause of sudden death in
coronary heart disease is almost always either
ventricular fibrillation or asystole.2' 22 In the
CHIANG ET AL.
Tecumseh population various abnormal findings in the electrocardiogram including
rhythm disturbances and conduction defects
were more commonly detected among individuals destined to die suddenly than among
other persons. An analysis of the relationship
of ventricular premature beats to coronary
heart disease morbidity and mortality in
Tecumseh has been published earlier.23 Three
of the four participants with "bilateral bundlebranch block syndrome" died suddenly. This
condition has been implicated in the development of complete atrioventricular heart
block.24 Abnormalities detected in the electrocardiogram, particularly arrhythmias and conduction defects, may increase the risk of
sudden death but they were so frequently
associated with other predisposing factors
among the 45 Tecumseh victims that their
relative contribution to the fatal outcome
remains conjectural.
Since myocardial ischemia lowers the
threshold for ventricular arrhythmias and
delays conduction velocities, electrocardiographic evidence of ischemic heart disease
may favor electrical failure, especially when
associated with pre-existent rhythm disturbances or conduction defects.25 While other
risk factors, such as hypertension, hypercholesterolemia, overweight, and hyperglycemia, are probably atherogenic, abnormal
findings detected in electrocardiograms, particularly arrhythmias and conduction defects,
may precipitate sudden death.
Although sudden death often seems to occur
without warning, the victims are almost
always predisposed to their fate by conditions
which are detectable long before the fatal
event. Recognition of these specific abnormalities is a necessary prerequisite to preventive
efforts. Presumably, any regimen which corrects the conditions predisposing to coronary
heart disease in general might also be
beneficial in the prevention of sudden death.
However, the factors which cause electrical
instability and mechanistic myocardial failure
would probably require more specific prophylactic measures.
Circulation, Volume XLI, January 1970
37
SUDDEN CARDIAC DEATH
References
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
1. CROSS EG: Coronary heart disease section. In A
Cooperative Study in Acute Myocardial Infarction, U. S. Public Health Service, edited by
Julian DG, Oliver MF. E. & S. Livingstone,
Ltd., Edinburgh, 1968, p 37
2. JULLAN DG: Coronary care and the community.
Ann Intern Med 69: 607, 1968
3. KANNEL WB, BARRY P, DAWBER TR: Immediate
mortality in coronary heart disease: The
Framingham study. Memorias IV Congress
Mundial de Cardiologia, IVB, 176 Mexico,
1963
4. PELL S, D'ALoNZO CA: Immediate mortality and
five year survival of employed men with a first
myocardial infarction, New Eng J Med 270:
915, 1964
5. NAPIER JA: Field methods and response rate in
the Tecumseh community health study. Amer J
Pub Health 52: 208, 1962
6. EPSTEIN FH, OSTRANDER LD, JoNsoN BC, ET
AL: Epidemiological studies of cardiovascular
disease in a total community-Tecumseh,
Michigan. Ann Int Med 62: 1170, 1965
7. FRANcis T JR, EPSTEIN FH: Epidemiological
research in a total natural community, Tecumseh, Michigan. Milbank Memorial Fund
Quarterly Vol. XLIII, No. 2, page 333, April,
1965
8. EPSTEIN FH, FRANcIS T JR, HAYNER NS, ET AL:
Prevalence of chronic diseases and distribution
of selected physiological variables in a total
community, Tecumseh, Michigan. Amer J
Epidemiol 81: 307, 1965
9. BLAcBuRN H, KEYS A, SIMONSON E, ET AL: The
electrocardiogram in population study. Circulation 21: 1160, 1960
10. ABELL LL, LEvy BB, BRODIE BB, ET AL: A
simplified method for the estimation of total
cholesterol in serum and demonstration of its
specificity. J Biol Chem 195: 357, 1952
11. NELSON N: A photometric adaptation of the
Somogyi method for the determination of
glucose. J Biol Chem 153: 375, 1944
12. MONTOYE HJ, EPSTEIN FH, KJELSBERG MO: The
measurement of body fatness. A study in a
total community. Amer J Clin Nutr 16: 417,
1965
13. KULLER L: Sudden and unexpected non-traumatic deaths in adults: A review of epidemiological and clinical studies. J Chron Dis 19: 1165,
1966
Circulatiosu, Volme XU, January 1970
14. KULLER L, LILIENFELD A, FISHER R:
An
epidemiological study of sudden and unexpected deaths in adults. Medicine 46: 341, 1967
15. MYERBURG RJ, DAVIS JH: The medical ecology
of public safety: 1. Sudden death due to
coronary heart disease. Amer Heart J 68: 586,
1964
16. SPAIN DM, BRADESS VA, MOHR C: Coronary
atherosclerosis as a cause of unexpected and
unexplained death: An autopsy study from
1949-1959. JAMA 174: 384, 1960
17. KANNEL WB, KAGAN A, DAWBER TR, ET AL:
Epidemiology of coronary heart disease:
Implications for the practicing physician.
Geriatrics 17: 675, 1962
18. MoRRis JN, HEADY JA, RAFFLE PAB, ET AL:
Coronary heart-disease and physical activity of
work. Lancet 2: 1053, 1953
19. SHAPIRO S, WEINBLATr E, FRANK CW, ET AL:
The H.I.P. study of incidence and prognosis of
coronary heart disease. J Chron Dis 18: 527,
1965
20. EPSTEIN FH: Some uses of prospective observations in the Tecumseh community health study.
Proc Roy Soc Med 60: 56, 1967
21. SuRAwIcz B, PELLEGRINO EE (Eds): In Sudden
Cardiac Death, Chapt. III. Mechanism of
sudden cardiac death, p 16 (Hellerstein HK,
Turell DJ: The mode of death in coronary
artery disease-An electrocardiographic and
clinicopathological correlation) and Chapt. VI.
Ventricular asystole, p 97. (Langendorf R,
Pick A: Causes and mechanisms of ventricular
asystole in advanced A-V block). New York,
Grune & Stratton, Inc., 1964
22. Gu11ERREz MR, CHANGFOOT GH, PERETZ DI:
Significance of T wave interruption by
premature beats as a cause of sudden death.
Canad Med Ass J 98: 1968
23. CHIANG BN, PERLMAN LV, OSTRANDER LD, ET
AL: Relationship of premature systoles to
coronary heart disease and sudden death in
Tecumseh epidemiologic study. Ann Intern
Med 70: 1159, 1969
24. LASSER RP, HAr JI, FRIEDBERG CK: Relationship of right bundle-branch block and marked
left axis deviation (with left parietal or periinfarction block) to complete heart block and
syncope. Circulation 37: 429, 1968
25. WIGGERS CJ, WEGRIA R. PINERA B: The effect
of myocardial ischemia on fibrillation threshold: The mechanism of spontaneous ventricular
fibrillation following coronary occlusion. Amer
J Physiol 131: 309, 1940
Predisposing Factors in Sudden Cardiac Death in Tecumseh, Michigan: A
Prospective Study
BENJAMIN N. CHIANG, LAWRENCE V. PERLMAN, MARY FULTON, LEON D.
OSTRANDER, JR. and FREDERICK H. EPSTEIN
Downloaded from http://circ.ahajournals.org/ by guest on June 12, 2017
Circulation. 1970;41:31-37
doi: 10.1161/01.CIR.41.1.31
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1970 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://circ.ahajournals.org/content/41/1/31
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles
originally published in Circulation 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 is online at:
http://circ.ahajournals.org//subscriptions/