Sex Differences in the Electrocardiogram
By ERNST SNIJONSON, M.D., HENRY BLACKBURN, JR., M.D., THOMAS C. PUCHNER, M.D.,
PAULINE EISENBERG, M.D., FERNANDA RIBEIRO, M.D.,
MANUEL MEJA, M.D.
AND
of these data subjects were excluded when there
was evidence of arterial hypertension (160 mm.
Hg systolic or 95 mm. Hg diastolic, or greater),
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
SEX DIFFERENCES are lnot considered
in contemporary elinical electrocardiographic interpretation. The more commonly
used normal standards make no sex differentiation, and some are derived from men alone
whereas others are composite data from men
and women.1-4
There is evidence from earlier studies that
sex differeniees in the electrocardiogram may
ilndeed exist.5 Small numbers of subjects were
analyzed; however, no statistical treatment
was provided, and age and relative body
weight were not considered in the compari-
any clinical diagnosis of heart disease, cardiac
murmurs greater than grade-II intensity, chronic
bronchopulmonary disease, renal disease, diabetes
mellitus, thyroid disease, peptic ulcer, active gallbladder disease, or convalescence from recent infections. The groups thus represent clinically
screened samples of the American working population, in the age range 40 through 59 years. Relative body weight was calculated from height-weight
data for age from the standard Medico-Actuarial
tables.6 This is a rather crude measure but is
sufficient for gross charaeterization of underweight,
normal weight, and overweight.
A supine resting 12-lead electrocardiogram
(leads I, II, III, aVn, aVL, aVF, VY-V6) was
recorded on standard direct-writing instruments,
at 25 mm. per second paper speed, with 1 mv. =
1 cm. calibration. The measurements were made in
duplicate by two investigators. Smoking and hard
physical work were proscribed for at least 30
minutes before the recording but prior meal intake
was not rigidly controlled.
sons.
This study provides data on differences in
electrocardiographic characteristics of men
and women that may contribute to improved
discriminative value of the electrocardiogram
in clinical interpretationi.
Material and Methods
The 424 men of this study derive from a randomly selected sample of railroad employees, ages
40 through 59 years, from 20 railroad companies
operating in midwestern and northwestern United
States.* The 142 women, ages 40 through 59, are
employees of the Mt. Sinai Hospital, Minneapolis,
the Mutual Service Insurance Companies, St. Paul,
Minnesota, the Asbury Hospital, Minneapolis, and
the Provident Mutual Life Insurance Company,
Philadelphia, Pennsylvania. We selected this age
range because it is most important in regard to
electrocardiographic interpretation of coronary
heart disease.
For the experimental groups of "normal healthy"
inen and women, eliminations from a larger sample
were made on the basis of nonelectrocardiographic
criteria after medical history and phvsical examination, urinalysis, and chest x-ray. On the basis
Results
Statistical analysis revealed no significant
differenees in electrocardiographic characteristics between partial age groups of 40 to
49 and 50 to 59 years. Therefore, all subjects from 40 to 59 years were pooled with
resultant total groups of 424 men and 142
women. The average relative body weight of
the meni is 102.9 per eent and that of the
women 97.4 per cent. Both values for relative
body weight are not significantly different
fromi the "normal" (100 per cent). The 5 per
cent lower mean relative body weight in the
women corresponds to a general trend in regard to the reference data of 1912. Both
From the Laboratory of Physiological Hygiene,
groups include underweight, normal weight,
and overweight persons, but these are similarly distributed in the total groups, and the
samples may be considered representative of
the general population in regard to relative
body weight. Differences in electrocardio-
University of Minnesota, Mt. Sinai Hospital, Minneapolis, Minn., and the Medical Department, Mlutual
Service Insurance Companies, St. Paul, Minn.
Supported in part by a grant from the Minnesota
Heart Association.
*The epidemiologic study of this railroad population
is under the direction of Dr. Henry Taylor.
598
Circulation, Volume XXII, October 1.(0O
599
SEX DIFFERENCES IN ELECTROCARDIOGRAMS
Table 1
Means (M), Standard Deviations (SD) of Amplitudes, Axis, and Intervals in 424 Healthy Men and 142
Healthy Women, Ages 40 through 50 Years
Standard Leads
M1
SD1
Women
M2
SD2
0.59
1.06
0.23
0.25
0.41
0.23
0.38
0.30
0.35
0.83
0.59
1.02
0.19
0.23
0.38
Men
P1
P2
Qi
Q2
Q3
0.28
0.44
0.31
0.42
0.69
Men
Mi-M2
.01
.04
.04
.01
.03
QRS axis
-QRS
T1
T2
T3
R1
iR
5.97
7.50
3.21
S1
S2
S3
0.70
0.82
1.61
R2
2.69
3.33
3.10
6.16
8.09
3.59
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
2.70
3.37
3.19
0.59
1.07
2.39
-.19
-.60
-.38
T axis
IT
Women
M1
SD1
M2
SD2
37.2
19.8
32.0
5.7
40.1
20.3
28.8
6.4
0.75
1.90
2.33
0.43
1.93
2.25
0.36
0.69
0.79
0.81
- .03
0.91
0.86
37.7
5.01
20.2
1.64
36.9
4.91
18.7
1.43
.8
.10
.35*** Intervals (.01 sec.)
P-R
16.7
1.6
15.5
2.0
.15
9.0
1.3
6.7
QRS
0.9
.21
Statistical significance of the mean differences (M1-M2) is evaluated by means of the t test
(*p = <0.05; **p = <0.01; ***p = <0.0001).
0.95
1.07
2.21
0.34
0.67
1.40
graphic characteristics between the sex groups
are therefore not attributable to a different
distribution of obese and underweight individuals.
Table 1 shows the means (M) and standard
deviations (S. D.) of the amplitudes, axes,
and intervals in the standard limb leads ot
both groups, the mean differences between the
groups, and their statistical significance evaluated by the t test. The most significant difference is the shorter P-R and QRS duration
in women. The differences in amplitudes are
small and statistically not significant except
for the slightly larger S wave in lead I in men.
Table 2 shows the comparison of Q, R, S.
and T amplitudes in augmented unipolar
leads. Since the QRS axis and T axis (table
1) are not significantly different between the
groups, subdivision into partial positional
groups was not necessary for the comparison
of the whole groups. The somewhat higher R
wave in aVF in women corresponds to an average axis deviation to the right, and is statistically significant at the p<0.05 level. The
smaller S wave in aVR and aVL, and the
slightly greater T wave in aVT, were significantly different.
In general, the sex differenees in amplitudes
found in limb leads are small although some
attain statistical significance.
Circulation, Volume XXII, October 1960
MI-M2
-2.9
- .5
.07
.07
1.2***
2.3***
In contrast, figure 1 shows the highly significant differences in precordial leads, with
smaller amplitude of R, 5, and T waves in the
women in all V leads, particularly from V2
to V5. The largest difference in R-wave amplitude occurs in V5 (about 25 per cent larger
in men), and in the S wave and T wave in V3
(about 80 per cent larger in men).
The mean R/S ratio in V1 is identical in
both groups (0.22 in men, 0.21 in women). An
analysis of the frequency distribution (in per
cent) of the transitional zone is shown in table
3. The transitional zone in men occurs most
often in V3 (31.1 per cent), nearly identical
to the frequency for V3 in women (33.3 per
cent). The frequency of the transitional zone
to the left of V3 is significantly higher in men,
and to the right of V3 is significantly higher
in women (tested by the chi-square test). The
transitional zone from V4 to V6 is three times
more frequent in men than in women, and to
the right of V3 iS two times more frequent in
women than in men. Therefore, the transitional zone of women is farther to the right,
or counterclockwise.
This was confirmed by an evaluation of the
spatial orientation angle in the horizontal
plane. Ijeads are expressed in term-ls of this
angle, with use of the reference system of an
earlier communication,7 in which an angle of
SIMONSON, BLACKBURN, PUCHNER, EISENBERG, RIBEIRO, MEJA
6(00
Table 3
0
Distribution of QRS Transitional Zone Expressed
in Per Cent for 424 Normal Men and 142 Normal
WFomen., Ages 40 through 59
Location
1.2
V -V2
1.2
V2
V2-V3
V3
V3-V4
V4
V4-V5
V-
V6-VG
V'
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
V4
V.
< Dt
V,
V
Figure 1
Mean R,S( and T amplitude in Vi tionV in 424
healthy rnen and 142 healthy wornen f rom 4<0
through 59 years. Statistical s3igniJic(tnce o f the
differences is marked by asterisk8.
Table 2
Means (M) and Standard Deviations (SD) of
Amplitudes in 424 Healthy Men and 142 Healthy
WVomen, Ages 40 through 59 Years
Augmented Unipolar Limb Leads
Women
Men
Ml
SD1
M2
SD-,
0.24
0.23
0.34
0.35
0.23
0.22
0.42
0.36
.01
.01
0.47
3.37
4.71
0.63
2.50
3.26
0.39
3.33
5.33
0.45
2.47
3.21
.08
.04
-.63*
4.25
1.12
1.43
.94*
.39*
.11l
0.64
0.67
0.70
-.07
-.18**
.07
Q
aVF
B
aVR
aVL
aVF
S
3.59
3.91
4.53
0.69
1.51
1.08
aVi,
0.76
1.30
0.87
aVF
T
-2.01
0.72
aVR -2.08
1.06
0.88
0.71
aVL
0.81
1.32
1.40
aVF
Statistical significance of the
(MI1-M2) is evaluated by means of
aVR
<0.05; 4*p
<0.01; ***p
=
mean differences
the t test (*p =
<0.0001).
poinlts left lateral, and an angle of 90°
points directly forward with respect to the
anatomic center of the heart. With this techlnic
zero
7.5
11.8
31.1
23.8
16.0
4.2
2.8
0
0.2
Women
0
4.9
9.2
26.8
33.1
18.3
6.3
0.7
0
0
0.7
i
Eif
aVL
Men
VI
the mnean orientation angle of the transitional
zone is 64.3° in women and 59.4° in men. The
differencee is statistically highly significantt,
and confirms the results of the distribution
analysis (table 3).
However, the tranisitional zone in the majority of men and women is located from the
left of V2 to position V4, and extreme clockwise rotation (transitional zone to the left of
V5) or extreme counterclockwise rotation
(transitional zone to the right of V2) is rare
in both groups. The sex difference in location
of the transitional zone does not account for
the large differelnces in amplitudes of all deflections in chest leads.
Discussion
Major sex differences in electrocardiographic wave amplitudes were found only in
precordial leads (horizontal plane), and this
was unexpected. The reason is not immediately apparenlt, though the smaller heart of
women, both absolute and in relation to the
body weight,8 may be a contributing factor.
In a recent study9 of the influence of chest
configuration in men on the electrocardiogram
and the spatial vectoreardiogram, direction
rather than magnitude was affected. This
imakes it less likely that the different sex characteristics of chest configuration account for
the smaller electrocardiographic amplitudes
found in women. Breast adipose tissue in
women would not likely account for the amCirculation, Volume XXII, October
1960
SEX DIFFERENCES IN ELECTROCARDIOGRAMS
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
plitude differences, since differences are apparent in V2 at the sternum and V5-V6 in the
left chest, although this factor cannot be entirely eliminated.
It can be predicted that the magnitudes of
the mean and maximal spatial vectors for
QRS and T are smaller in women, though no
analysis of vector magnitudes was made.
For clinical application it appears that no
sex differentiation need be made in regard to
limb-lead electrocardiographic amplitudes in
this age group, but the shorter QRS duration
in women, found as well in earlier studies,10' 11
and the shorter P-R interval, must be considered. This might also be explained by
smaller heart size in women. On the other
hand, normal limits for precordial-lead electrocardiographic amplitudes, largely derived
from men, may not be valid for women. Such
limits are of more practical value for clinical
interpretation than the means and S.D. In
these samples, for example, the upper 97.5per cent limit of R-wave amplitude in V5 iS
21 mm. in women, 25 mm. in men.
Summary
In 424 men and 142 women, clinically
"healthy," ages 40 through 59 years, significant sex differences in electrocardiog,raphic
characteristics were found. The differences
were minor in the limb-lead amplitudes (P,
Q, R, S, T), but QRS and P-R duration was
significantly shorter in women. Precordiallead amplitudes (R, S, T) were significantly
smaller in women. Present normal standards
derived from groups of men are not valid for
women.
Acknowledgment
We wish to thank Professor Henry L. Taylor
for his cooperation and permission to use the electrocardiograms of the employees of the railroad companies; Dr. James Dahl for providing the electrocardiograms of female employees of the Ashbury
Hospital, Minneapolis; Dr. Paul Langner for sending
electrocardiograms of female employees of the Provident Mutual Life Insurance Company, Philadelphia,
Pennsylvania; Mrs. Jane Bardon and Mr. Andrew
Circulation, Volume XXII, October
1960
601
Westerhaus for the statistical evaluation; and Mr.
Charles Johnson for participation in the measurements.
Summario in Interlingua
In 424 masculos e 142 femininas omines clinicamente san e de etates de inter 40 e 59 annos-significative differentias sexual esseva constatate in le
characteristicas electrocardiographic. Le differentias
esseva minor in le amplitudes del derivationes extremitatal (P, Q, R, S, T), sed le duration de QRS
e de P-R esseva definitemente reducite in le femiinas.
Le amplitudes del derivationes precordial (R, S, T)
esseva significativemente plus micre in femininas. Le
currentemente usate standards de normalitate derivate ab homines non es valide pro feminas.
References
1. KOSSMANN, C. E.: The normal electrocardiogram.
Circulation 8: 920, 1953.
2. New York Heart Association. Nomenclature and
criteria for diagnosis of diseases of the heart
and blood vessels. Criteria for electrocardiographic diagnosis. New York, New York Heart
Association, 1953, p. 133.
3. LEPESCHKIN, E.: Modern Electrocardiography,
Vol. 1: The P-Q-R-S-T-U Complex. Baltimore,
Williams & Wilkins Co., 1951.
4. WINSOR, T., Editor: Electrocardiographic Test
Book, Vol. 1. New York, American Heart Association, 1956, p. 147.
5. DEEDS, D. E., AND BARNES, A. R.: The characteristics of the chest lead electrocardiogram of
100 normal adults. Am. Heart J. 20: 261, 1940.
6. Association of Life Insurance Medical Directors
and Actuarial Society of America: Medicoactuarial mortality investigation. New York,
Vol. 1, 1912.
7. SIMIONSON, E.: A spatial vector analyzer for the
conventional electrocardiogram. Circulation 7:
403. 1953.
8. SAPHIR, O.: Gross examination of the heart. In
Pathology of the Heart, Edited by S. E. Gould.
Springfield, Ill., Charles C Thomas, Publisher, 1953, p. 948.
9. SIMONSON, E., BROZEK, J., AND SCHMITT, 0. H.:
Unpublished data.
10. AMCGINN, S., AND WHITE, P. D.: Duration of
QRS complex in normal and in abnormal
electrocardiograms; study of 500 cases. Am.
Heart J. 9: 642, 1934.
11. LARSEN, K., AND SKULASON, J.: The normal
electrocardiogram; analysis of extremity derivations from 100 normal persons whose ages
ranged from 30 to 50 years. Am. Heart J.
22: 625, 1941.
Sex Differences in the Electrocardiogram
ERNST SIMONSON, HENRY BLACKBURN, JR., THOMAS C. PUCHNER,
PAULINE EISENBERG, FERNANDA RIBEIRO and MANUEL MEJA
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 1960;22:598-601
doi: 10.1161/01.CIR.22.4.598
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1960 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/22/4/598
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/