Electrocardiographic Criteria for the

Diagnosis of Right Ventricular Hypertrophy

Verified at Autopsy*
Jari Lehtonen,
Markku

M.D.;

Ikaheimo,

Seppo Sutinen,
M.D.,

FCC,P.;

M.D.;
and Paavo Paakko,

M.D.

Four commonly used sets of electrocardiographic ( E C G )

showing isolated R V H , with a left-to-right ratio of two or

criteria for diagnosing right ventricular hypertrophy ( R V H )

less, had died of respiratory causes. T h r e e sets of E C G

were tested for sensitivity and specificity in 12 m e n with

criteria showed a specificity o f 1 0 0 percent, but sensitivities

isolated R V H , 1 5 with combined right and left ventricular

only from 26 to 44 percent. O n e set was more sensitive (74

hypertrophy and 24 with normal ventricular weights. T h e

percent) but much less specific than the others (79 percent).

cardiac ventricles w e r e weighed separately at autopsy and

A n e w combination of E C G criteria attained a 63 percent

the left-to-right

sensitivity and a 96 percent specificity.

ratio was calculated. A l l the patients

H T h e standard 12-lead electrocardiogram is in general

readily

available,

and

numerous

electrocardio-

graphic ( E C G ) criteria have been proposed for diagnosing

right

ventricular

hypertrophy

(RVH)

pulmonary heart disease (cor p u l m o n a l e ) .

813

1 7

and

T h e sen-

sitivity of the proposed criteria has varied in different

studies from 10 to 100 percent and the specificity from
90 to 100 p e r c e n t .

7 1 1 1 4 1 6

In practice, however, the

sensitivity of the criteria is relatively low, varying from

25 to 40 percent.

17

By definition,

autopsy data are

essential for an accurate diagnosis of R V H or pulmonary heart disease in correlative studies, since the muscular mass of both cardiac ventricles should be measured separately at autopsy, eg, by the method of Fulton
et al.

18

Not all studies on E C G criteria for the diagnosis

of RVH are based on sound anatomic facts, however.

The aim of the present study was to evaluate the
specificity and sensitivity of four commonly used sets
of E C G criteria for diagnosing R V H retrospectively in
a series of male patients whose cardiac ventricles were
weighed

separately

at autopsy

We

also

made

an

attempt to compile a new set of criteria giving improved sensitivity without a decrease in specificity.
PATIENTS A N D METHODS

A total of 51 adult male patients were selected from a total of 186

adult male autopsy cases in which the free wall of the right ventricle
(RV) and the left ventricle with the septum (LV + S) were weighed
separately and the ratio of LV" + S to RV (the left-to-right ratio) was
calculated. Twenty-four patients (group 1) had normal ventricular
weights, 15 (group 2) had combined right and left ventricular
hypertrophy, and 12 (group 3) had isolated right ventricular hyper-

trophy; No patient with valvular abnormalities was included.

The ventricular weights were considered normal
if the total
ventricular weight was less than 250 g, the weight of the RV less
than 65 g, that of the LV + S less than 190 g and the left-to-right
ratio from 2.3 to 3.3 (group 1). Combined right and left ventricular
hypertrophy was considered to be present if the weight of RV was
more than 80 g and the left-to-right ratio more than 2 (group 2), and
isolated right ventricular hypertrophy was considered to be present
if the weight of the RV was more than 80 g and the left to right ratio
was 2 or less (group 3).
1718

The distributions of ventricular weights in groups 1 to 3 are shown

in Figure 1. The mean ages of the patients was 708.6 years in
group 1, 68 0.7 years in group 2, and 68 7.8 years in group 3.
The underlying causes of death were recorded from the patients'
death certificates.
Electrocardiograms recorded not more than one month before
death were analyzed. Patients with a left bundle branch block
(LBBB) were excluded. The following sets of ECG criteriaforRVH
were tested: World Health Organization, Heikkila, Louridas et al
and Butler et al (Table 1). In accordance with Milnor, the criterion
R or R'aS in V, or \' was substituted for the criterion RaS. The
sensitivity and specificity of each individual criterion and each set
of criteria were determined using the following formulas. The
specificity of a test is defined as its ability to exclude a disease in
those who are free of it.
8

13

16

RV

160
140
120

rf>

100

80-1

15

*From the Department of Pathology, University of Oulu, and the

Department of Internal Medicine, Oulu University Central Hospital, Oulu, Finland.
This work was supported in part by grants from the Finnish AntiTuberculosis Association, Helsinki, Finland, and the Academy of
Finland.
Manuscript received June 29; revision accepted September 14.
Reprint requests: Dr. Ikaheimo, Department of Medicine, Oulu
University Central Hospital, SF 90220 Oulu, Finland

60

= Group 1
= Group 2
= Group 3

40
20
)

50

100

!
150

200

250

300

CHEST / 93 / 4 / APRIL, 1988

Downloaded From: http://journal.publications.chestnet.org/ on 01/31/2014

350 400
LV+S
F l G l ' H E 1. Distribution of ventricular weights in patients with
normal weights (group 1), combined right and left ventricular
hypertrophy (group 2) and isolated right ventricular hypertrophy
(group 3). RV, free wall of right ventricle; LV+S, left ventricle with
septum.
839

Table 2Underlying Causes of Death*

Table ISets of ECG Criteria for Diagnosing

Anatomic RVH
World Health Organization
1. qR pattern with intrinsicoid deflection more than 0.03 s in

Control Combined Isolated

Group
RVH
RVH
(Group 1) (Group 2) (Group 3) Total

V,.
2. R amplitude<S amplitude in V .
3. R or R' amplitude<S amplitude in 1.
4. Incomplete right bundle branch block with QRS duration
less than 0.12 second.
An ECG is considered diagnostic for RVH if criterion 1 or two or
more of criteria 2-4 are met. Criteria 5-7 reinforce the diagnosis.
5. P amplitude 2:0.25 mm in 2.
6. Right axis deviations 110 degree.
7. T-inversion in leads V, to V or in leads 2 and 3.
Heikkila *
1. Right axis deviations 110, no RBBB.
2. R or R' amplitude^ amplitude in V, or V .t
3. R amplitude's amplitude in V , no anterior myocardial
infarct.
An ECG is considered diagnostic for RVH if one or more of criteria
1-3 is met. Criteria 4-5 reinforce the diagnosis.
4. P amplitude>0.25 mV in leads 1, 2 or aVE
5. P axis + 6 0 - + 90 degree (P in aVF>P in lead 2).
Louridas et al
1. R or R' amplitude>S amplitude in V,.t
2. R amplitude<S amplitude in V or V .
3. ST segment depression and T inversion in V, and V .
4. Intrinsicoid deflection>0.04 s.
An ECG is considered diagnostic for RVH if two or more of the
criteria are met.
Butler et al
1. A + R - P L a 0 . 7 m V
A = maximal R or R' amplitude in V, or V
R = maximal S in lead 1 or V
PL = minimal S in VI or minimal r in lead 1 or V
2. R amplitude0.2 mV in lead 1.
3. P amplitude20.25 mV in leads 2, 3, aVF, V, or V. .
An ECG is considered diagnostic for RVH if two or more of the
criteria are met.
s

Underlying
Causes of Death
Respiratory diseases
Emphysema
Asthma
Neoplasms
Others
Cardiovascular diseases
Ischemic heart diseases
Cerebrovascular disease
Others
Other causes of death
Total

(n)

(n)

(n)

(n)

8
4
1
2
1
7
2
5
0
9
24

1
1
0
0
0
13
9
1
3
1
15

12
9
0
2
1
0
0
0
0
0
12

21

20

10
51

13

*In this Finnish Textbook of ECG the criteria for RVH are modified
from the studies by Sokolow and Lyon and Milnor.
tAccording to Milnor the criterion R or R'aS in V, or V may be
substituted for the criterion R>S.
2

*Respiratory diseases vs others: chi-square = 25.13, p<0.0005.

patients' certificates are presented in Table 2. All 12

patients with isolated RVH (group 3) but only one
patient with combined RVH had died of respiratory
causes (group 2) (p<0.0005).
The specificities and sensitivities of the individual
criteria and sets of criteria are shown in Table 3. The
criteria of Butler et al

which were originally aimed

at diagnosing RVH due to mitral stenosis, attained the

highest sensitivity (74 percent) but the lowest specificity (79 percent). All the other sets of criteria showed a
specificity of 100 percent but only very low sensitivities (26 to 44 percent). In only one case did one of the
other sets of criteria (Heikkila ) diagnose RVH cor6

rectly when the criteria of Butler et al had failed.

7

The criteria of Butler etal yielded five false-positive

7

RVH diagnoses, cases in which the mean RV weight

was lower than among the others in the control group
(40 vs 53 g; p<0.01), although the left-to-right ratio
was almost the same (3.00 vs 3.06). One of these
patients had an apical myocardial infarct. On the other
hand, among the 19 in the control group without RVH
diagnosis, two had an apical or posterior myocardial

No. of true negative cases

- X 100 percent
No. of cases without disease (RVH)
The sensitivity of a test is defined as its ability to detect a disease
in those who harbor it:
No. of true positive cases
- X100 percent
Sensitivity =
No. of cases with disease (RVH)
For the assessment of emphysema, one lung of each patient was
radiographed, fixed intrabronchially with formalin-polyethylene
glycol-alcoho! solution, dried with air insufflation, sliced sagittally,
and examined on the basis of gross specimens, histologic sections
and radiographs of excised lungs as described previously The
severity of emphysema was evaluated using a set of photographs of
standard grades. Grade 0 represented normal lung; grades 5 to 50,
mild to moderate emphysema; and grade 60 or more, severe
emphysema. In seven cases in the control group (group 1), the
severity of emphysema was evaluated only on the basis of histologic
sections taken from both lungs.
Specificity =

19

20

21

22

infarct. There were five patients in the combined RVH

group and one in the isolated RVH group with an
apical or posterior myocardial infarct. The underlying
causes of death among those five false-positive RVH
cases, according to the criteria of Butler et al, were
7

as follows: Two patients had cerebrovascular disease,

one emphysema, one pulmonary fibrosis (other respiratory diseases) and one other causes of death (Table
2). O f them, three had severe emphysema, while four
out of 19 among those with negative RVH diagnoses
by the criteria of Butler et al had severe emphysema,
7

the difference being not statistically significant.

A new set of criteria, compiled by adding the first
criterion of Butler et aP to the first, second and third
of Heikkila

RESULTS

The underlying causes of death recorded on the

840

Downloaded From: http://journal.publications.chestnet.org/ on 01/31/2014

and considering an E C G diagnostic of

RVH when one or more of the criteria were met,

attained a sensitivity of 63 percent and a specificity of
ECG Criteria for Diagnosis of Right Ventricular Hypertrophy (Lehtonen

et al)

Table 3Specificity and Sensitivity of Individual Criteria

and Sets of Criteria for Diagnosing Anatomical RVH

Speci-
ficity
Groups
(Group 1) Group 2 Group 3 2 & 3
(n = 24) (n == 15) (n == 12) (n == 27)

World Health Organization

qR with intrinsicoid
defleetion>0.03 s
in Vj
R<S in V
R<S in lead 1
Incomplete RBBB
with QRS<0.12s
Pa0.25mVinlead2
QRS axisallO degree
T-inversion in V,-V or
in leads 2, 3
Heikkila
QRS axisadegree, no
RBBB
R o r R ' a S i n V, or V
R s S in V , no anterior
infarct
P>0.25 mV in leads 2,
3, aVF
P in aVF>P in lead 2
Louridas et al
R>S in V,
R<S in V
ST segment
depression and T
inversion in V, and

(100)

No. (%) No. (%) No. (%)

4

(27)

(25)

7 (26)

0 (100)
6
(75)
1
(96)

3 (20)
5 (33)
4 (27)

1
5
4

(8) 4 (15)
(42) 10 (37)
(33) 8 (30)

(92)
(92)
(100)

3 (20)
3 (20)
1 (7)

3
1
3

(25)

2
0
2
0

(92)
(100)

5 (33)
5 (33)

4
7

(33) 9 (33)
(58) 12 (44)

0
0

(100)
(100)

1 (7)
4 (27)

3
6

(25) 4 (15)
(50) 10 (37)

(100)

(0)

(8)

(4)

3
3
0
0
6

(88)

(0)
(27)
(27)
(33)

2
0
4
4
5

(17)

(75)

3
0
4
4
5

(20)

(88)
(100)
(100)

(0)
(33)
(33)
(42)

5
0
8
8
9

(19)
(0)
(30)
(30)
(33)

(100)

(7)

(0)

(4)

0
5
1
3

(100) 5 (33)
(79) 11 (73)
(96) 5 (33)
(88) 5 (33)

3
9
6
5

(8)
(25)

6 (22)
4 (15)
4 (15)

13

Intrinsicoid deflection
>0.04 s
Butler et al
A + R + PL>0.7 mV
R<0.2 mV in lead 1
Pa0.25 mV in leads
2, 3, aVF, V, or V
New set of criteria
QRS axissdegree, no
RBBB
R o r R ' a S i n V, o r V
RsS in V no anterior
infarct
A + R + PL>0.7mV
7

18

2426

2728

18

15

23

Sensitivity

No. (%)

in some correlative studies of E C G criteria as an

anatomic verification of R V H , although it has been
shown to be unreliable for scientific purposes. The
method of Fulton et al, in which the septum is
weighed with the left ventricle, is similarly not exact,
eg, not taking into consideration the septal hypertrophy observed by echocardiography in primary pulmonary hypertension, for e x a m p l e ,
but it is reprod u c i b l e and g i v e s c o m p a r a b l e results. S o m e
authors
have considered the lower limit for R V H
of 80 g suggested by Fulton et al too high. This was
an advantage in the present study, however, as it
enabled a clear distinction to be made between normal
and hypertrophic hearts. In our material, which included no cases of valvular abnormalities, a reduced

(25) 8 (30)
(75) 20 (74)
(50) 11 (41)
(42) 10 (37)

left-to-right ratio in the sense of Fulton et al ( < 2 )

occurred only in connection with fatal pulmonary
disease (Table 2). Thus, our patients with isolated RVH
very specifically represented pulmonary heart disease.
18

Measurement of pulmonary arterial pressure sometimes is done to search for R V H or cor pulmonale.
The method is reliable, but it would be
reasonable to speak only of increased pulmonaryarterial pressure rather than R V H if the anatomic
degree of hypertrophy is unknown.
1 2 7

2 9

1130

Thallium activity ratio of the left-to-right ventricle

has been shown to correlate closely with the anatomical
left-to-right ratio, and thus with pulmonary heart
disease. The method is inadequate for verifying RVH
combined with left ventricular hypertrophy, however,
for this exists mainly with a normal left-to-right ratio
(group 2 in our study) and is due to cardiovascular
causes. In fact, this may be an advantage.
30

Our results support the common finding of low

sensitivity and high specificity in the case of most
E C G criteria for R V H mentioned in the literature.
Exceptions are the criteria of Butler et al, with a
sensitivity of 74 percent and specificity of 79 percent,
the latter probably being too low for clinical use. The
existence of apical or posterior myocardial infarcts,
which may cause E C G findings similar to RVH, did
not explain the present false-positive findings obtained
using the criteria of Butler et al. Clinical chronic
obstructive pulmonary disease may cause increased P
amplitude, and anatomically verified pulmonary emphysema may reduce the electrical QRS forces, but
in our material any of the criteria for RVH gave no
clear association with the severity of anatomic emphysema. In the original study of Butler et al, the
specificity was 94 percent, but the 500 normal subjects
studied in their series had no ischemic heart disease
or chronic pulmonary disease; half of them were
women and their mean age was 45 years. Our control
group of male patients had a mean age of 70 years and
included subjects with ischemic heart disease and
pulmonary disease, which are common in patients
30

3
1

(88)
(96)

3 (20)
8 (53)

2
9

(17) 5 (19)
(75) 17 (63)

0
0

(100)
(100)

1 (7)
4 (27)

3
6

(25) 4 (15)
(50) 10 (37)

0
1

(100)

0 (0)
5 (33)

1
6

(8) 1 (4)
(50) 11 (41)

6>

(96)

31

96 percent (Table 3).

None of the individual criteria nor the sets of criteria
gave any indication of a statistical association with the
severity of emphysema, although isolated RVH was
found more often in patients with severe emphysema
than in those without (p = 0.03, statistical significance
according to Fisher's exact probability test).
DISCUSSION

The time-honored method of measuring the thickness of the right ventricle in millimeters is still used

30

CHEST / 93 / 4 / APRIL, 1988

Downloaded From: http://journal.publications.chestnet.org/ on 01/31/2014

841

with RVH.
Most of the sets of criteria for diagnosing RVH were
slightly more sensitive in detecting isolated rather
than combined RVH, but none of them, not even that
proposed by us, was able to distinguish between these
two types.
We conclude that the presence of anatomic R V H
may be diagnosed from a standard 12-lead electrocardiogram in almost two thirds of men who have this
condition by reference to one or more of the following
E C G criteria: right axis deviation of s i 10 degrees (no
right bundle branch block), R or R ' a m p l i t u d e ' s
amplitude in V, or V , R amplitude^ S amplitude in
V (no anterior myocardial infarct) and A 4- R P L s
0.7 mV Correspondingly, RVH may be ruled out in
practically all of those who do not have it if none of
these criteria is met. This set of criteria also is useful
in RVH patients with E C G changes due to myocardial
infarcts and left ventricular hypertrophy, although the
criteria have not yet been tested on women or young
patients.
2

fi

REFERENCES

13 Louridas G, Galanis N , Patakas D. Vectocardiographic and

hemodynamic correlation in chronic obstructive pulmonary
disease. Chest 1982; 82:593-97
14 Phillips RW, Providence RI. The electrocardiogram in cor
pulmonale secondary to pulmonary emphysema: a study of 18
cases proved by autopsy. Am Heart J 1958; 56:352-71
15 Scott RC. The correlation between the electrocardiographic
patterns of ventricular hypertrophy and the anatomic findings.
Circulation 1960; 21:256-91
16 Weizenblum E, Moyses B, Dickele MC, Methlin G. Detection
of right ventricular pressure overloading by thallium-201 myocardial scintigraphy Chest 1984; 85:164-69
17 Fisch C. Electrocardiography and vectorcardiography. In:
Braunwald E, ed. Heart disease: a textbook of cardiovascular
medicine, vol 2, ed 2. Philadelphia: WB Saunders Company,
1984:210
18 Fulton RM, Hutchinson EC, Jones AM. Ventricular weight in
cardiac hypertrophy Br Heart J 1952; 14:413-20
19 Sutinen S, Paiikko P Lahti R. Post-mortem inflation, radiography, and fixation of human lungs; a method for radiological and
pathological correlations and morphometric studies. Scant! J
Respir Dis J979; 60:29-35
20 Sutinen S, Paakkii P Lohela V, Lahti R. Pattern recognition in
radiographs of excised air-inflated human lungs: IV emphysema
aione and with other common lesions. Eur J Respir Dis 1981;
62:297-314
21 Thurlbeck W M , Dunnill MS, Harking W, Heard BE, Heppleston AG, Ryder RC. A comparison of three methods of measuring
emphysema. Hum Pathol 1970; 1:215-26
22 Sutinen S. Vaajalahti E Paiikko P Prevalence, severity; and types
of pulmonary emphysema in a population oi deaths in a Finnish
city: correlation with age, sex and smoking. Scand J Respir Dis
1978; 59:101-15

1 Myers GB, Klein HA, Stofer MD. The electrocardiographic

diagnosis of right ventricular hypertrophy Am Heart J 1948;
35-. 1-40
2 Sokolow M, Lyon TE The ventricular complex in right ventricular hypertrophy as obtained by unipolar precordial and limb
leads. Am Heart J 1949; 38:273-94
3 Milnor VVR. Electrocardiogram and vectorcardiogram in right
ventricular hypertrophy and right bundle branch block. Circulation 1957; 16:348-67

23 Lamb D. Heart weight and assessment of ventricular hypertrophy. In: Dyke SD, et al, eds. Recent advances in clinical
pathology, series 6. London: Churchill Livingstone, 1973;133148

4 Goodwin JF, Abdin ZH. The cardiogram of congenital and

acquired right ventricular hypertrophy Br Heart J 1959, 21:
523-44

24 Goodman DJ, Harrison DC, Popp RL. Echocardiographic

features of primary pulmonary hypertension. Am J Cardiol 1974,
33:438-43

5 Roman GT, Walsh TJ, Massie E. Right ventricular hypertrophy;

correlation of electrocardiographic and anatomic findings. Am J
Cardiol 1961; 7:481-87

25 Maron BJ, Clark CE, Henry WL, Fukuda T, Edwards JE,

Mathews EC, et al. Prevalence and characteristics of disproportionate ventricular septal thickening in patients with acquired
or congenital heart diseases; echocardiographic and morphologic
findings. Circulation 1977; 55:489-96

6 Heikkila J, ed: EKG: perusteet ja tulkinta (ECG: the basis and

interpretation), Hiimeenlinna. Orion-yhtyma Oy, 1982: 55-63
7 Butler PM, Leggett SI. Howe CM, Freve CJ, Hindman NB,
Wagner GS. Identification of electrocardiographic criteria for
diagnosis of right ventricular hypertrophy. Am J Cardiol 1986;
57:639-43
8 World Health Organization. Chronic cor pulmonale: report of
an expert committee. Circulation 1953; 27:594-615
9 Selvestcr RH. Rubin HB. Neu criteria for the electrocardiographic diagnosis of emphysema and cor pulmonale. An) Heart
J L965; 69:437-47
10 Millard FJ. The electrocardiogram in chronic lung disease. Br
Heart J 1967; 29:43-50
11 Kilcoyne MM, Davis AL, Ferrer MI. A dynamic electrocardiographic concept useful in diagnosis of cor pulmonale: result of a
survey of 200 patients with chronic obstructive pulmonary
disease. Circulation 1970, 42:903-24
12 Loiitidas G, Anmachalelia N, Patakas D, Stravroponlos C. Right
ventricular echocardiographic, vectocardiographic and electrocardiographic study in cor pulmonale. Acta Cardiol 1980; 35:
429-36

842

Downloaded From: http://journal.publications.chestnet.org/ on 01/31/2014

26 Kaul S. The interventricular septum in health and disease. Am

Heart J 1986; 112:568-81
27 Thurlbeck W M . Chronic airflow obstruction in lung disease,
volume V In: Dennington JL, ed. Major problems in pathology
Philadelphia: WB Saunders Company, 1976
28 Sutinen S, Paiikko P Tienari J. Weights of the body and cardiac
ventricles in pulmonary emphysema. Yirchows Arch Pathol Anat
1985; 407:249-57
29 Cowdery CD, Wagner GS, Starr JW, Rogers G, Greenfield J.
New vectocardiographic criteria for diagnosing right ventricular
hypertrophy in mitral stenosis: comparison with electrocardiographic criteria. Circulation 1980;62:1026-32
30 Kawai S, Tanaka N, Sawada M, Sakaguehi K, Fuhta K, Shiola
K. The usefulness of thallium-201 myocardial perfusion scintigraphy in the diagnosis ol chronic cor pulmonale. Jap Cireul J
1981; 45:1003-13
31 Spodic DH. Electrocardiographic studies in pulmonary disease:
1. electrocardiographic abnormalities in diffuse lung disease.
Circulation 1959; 20:1067-72

ECG Criteria for Diagnosis of Right Ventricular Hypertrophy (Lehtonen

el al)