QRS Waves of the Spatial Velocity Electrocardiogram

in Atrial Septal Defect

Hiroyoshi MORI, M.D., Kouichi MIKAWA,M.D., Toshiharu NIKI, M.D.,

Takashi NAGAO, M.D., Satoru MATSUMO,M.D., Tomoo NII, M.D.,
Teiichi ODA, M.D.,* and Hideto MASAKI,M.D.**

SUMMARY
The most characteristic findings of QRS waves of the spatial velocity
ECG in atrial septal defect are that the velocity increased markedly after
the initiation of the terminal delay. The peak velocity in the terminal
delay and the velocities at the beginning portions of the terminal delay
showed positive correlations with pulmonary artery and right ventricular
pressures. Mean values of the pulmonary artery and the right ventricular
pressures were significantly higher in the group with higher peak velocity
in the terminal delay than in the group with lower peak velocity in the
terminal delay. The durations of the terminal delays showed significant
negative correlations with the pulmonary artery and the right ventricular
pressures. Using only 2 parameters concerning the terminal delay of
QRS waves of the spatial velocity ECG, namely the peak velocity in the
terminal delay and the duration of terminal delay, 93% of the normal sub-
jects, 100% of the usual type of the complete right bundle branch block and
95% of atrial septal defect were differentiated each other.
Additional Indexing Words:
VCG Analog computer analysis Terminal delay of ASD
Right ventricular hypertrophy Right bundle branch block

NCOMPLETE right bundle branch block, which is the most characteristic

electrocardiographic and vectorcardiographic finding in atrial septal de-
fect, is attributed to the hypertrophy of the outflow tract of the right ventricle
and is considered to have different clinical significance with the usual type of
the complete right bundle branch block. Intraventricular conduction dis-
turbances such as the bundle branch block are usually expressed as slow in-
scription of QRS loops in vectorcardiography. Such expressions, however,
are only approximations, and the velocity of inscription of spatial vector loop
can be expressed most exactly by the spatial velocity electrocardiography .
This study was intended to clarify the characteristics of the velocity of

From the Second Department of Internal Medicine, Faculty of Medicine, University of Tokushima ,
Tokushima; the Department of Pediatrics* and the First Department of Surgery ,** Faculty of Medi-
cine, University of Kyushu, Fukuoka, Japan.
Received for publication June 29, 1972.
407
 Jap. HeartJ. S
408 MORI, ET AL. eptember, 1972

inscription of spatial QRS loop in atrial septal defect by means of the spatial

velocity electrocardiogram, comparing with those of the normal subjects and the

usual type of the complete right bundle branch block. Correlations with

hemodynamic data were also investigated.

METHODS

a. Materials: Conventional 14 leads electrocardiograms (ECG), spatial

vectorcardiogram (VCG) by Frank system, and the spatial velocity ECG were

recorded in all 56 cases of secundum type of atrial septal defect (SAD), 20 cases of
the complete right bundle branch block and 56 cases of the normal subjects.

Atrial septal defect: Mean age of the cases with atrial septal defect was

16.0•}7.1, ranging from 5 to 32 years old. There were 24 males and 32 females.
Cardiac catheterizations were performed in 51 cases, and the diagnosis was ascer-

tained in 46 cases by cardiac surgery. Routine examinations, including phono-

cardiograms and chest X-ray examinations, were performed in all cases. Standard

ECG showed incomplete right bundle branch block in all cases.

Complete right bundle branch block: 20 cases of the complete right bundle

branch block without demonstrable heart disease, such as hypertensive, rheumatic,

congenital, or advanced coronary heart diseases, were examined. These cases

showed complete right bundle branch block as the sole abnormality. Mean age was

60.5•}15.3, ranging from 28 to 86 years old. There were 13 males and 7 females.
Normal subjects: This group consisted of 56 healthy men whose mean age was

24.5•}6.6, ranging from 19 to 40 years old. The criteria of normality were as follows:

(1) No history, no complaint and no physical sign of cardiovascular disease, (2)

Systolic blood pressure less than 140mm.Hg and diastolic blood pressure less than

90mm.Hg, (3) Normal resting standard ECG. Mean value of QRS interval of
this group was 88.6•}8.2msec. The incidences of the various patterns of QRS

waves in V4R were as follows: 2 cases of rSR' (3.6%), 10 cases of rSr' (17.6%),
1 case of notching of R wave (1.8%), 14 cases of notching of S waves (25.0%), and

16 cases of slurrs of R or S waves (28.6%). The incidences of the various patterns

of QRS waves in V1 were as follows: 1 case of rSr' (1.8%), 1 case of notching of R

wave (1.8%), 9 cases of notching of S waves (16.1%), and 16 cases of slurrs of R or S

waves (28.6%).
b. Recording of ECG: Conventional 14 leads ECG, including V4R and V7,

were recorded in all cases by means of 4 channel heat-writing electrocardiograph.

c. Recording of VCG: Three planar projections of spatial VCG were photo-

graphed on 60mm. X-ray films simultaneously by 3 channel vectorcardiograph

using Frank system. Vectorcardiographic loops were interrupted by the saw-toothed
waves of 800Hz.

d. Spatial velocity EGG:

1) Recordings: Spatial velocity ECG were recorded by means of the spatial

velocity electrocardiograph constructed by Mori et al.1),2) (1967). Fig. 1 showed
the block diagram of the spatial velocity electrocardiograph. The computations

shown in formula (1) were performed automatically by leading each scalar ECG of

Frank system to the differentiating, squaring, adding and square root circuits in or-

ders. Spatial velocity ECG were recorded by means of 4 channel heat-writing

Vol.13
No.5
SPATIAL VELOCITY ELECTROCARDIOGRAM 409

Fig. 1. Block diagram of the spatial velocity electrocardiograph.

X, Y, Z: scalar ECG of VCG.

Fig. 2. Schema of QRS waves of the spatial velocity EGG of atrial septal
defect and correlations with the hemodynamic data.
**positive correlations (p<0 .01)
*positive correlations (p<0 .05)
negative correlations (p<0.05)

oscillograph simultaneously with 3 scalar ECG of Frank leads. The recording speed
was 100mm./sec. Time constant of 1msec. was used for the differentiation of QRS
waves.

(1)

where SV: Spatial velocity ECG; X, Y and Z: Scalar ECG of Frank system VCG.
The electric characteristics of these circuits and the method of calibration were
described previously (Mori et al.1) 1968).
 Jap. HeartJ. S
410 MORI, ET AL. eptember, 1972

2) Measurements: Fig. 2 is the schematic presentation of the typical pattern

of QRS waves of the spatial velocity ECG in atrial septal defect. The peaks and

nadirs of QRS waves of the spatial velocity ECG were named as A, B, C, D, E, TD, F,
S and S'. A is the initial notch of QRS wave. C and E are the 2 peaks of the main

M-shaped complex of QRS wave. S is the peak in the terminal delay. F is the
nadir between E and S. TD is the point of the beginning of the terminal delay, and

usually appears slightly before F. S' is the second peak following S in the terminal

delay.
Velocities (amplitudes) and time intervals were measured as shown in Fig. 2.

Velocities and time intervals were expressed as capital and small letters respectively.

Velocities of these points and peak velocity were measured, and were expressed by

mV./sec. Peak-D, E/C, peak/S, and peak/(peak-D) were calculated. Time in-

tervals from the beginning of QRS to the various points were measured as well as

QRS interval, and were expressed by msec. QRS interval-d, QRS interval-e, QRS
interval-f, QRS interval-td, f-d, (QRS interval-d)/d, QRS interval/(QRS interval-d),

QRS interval/(QRS interval-td), and peak interval were also calculated.

RESULTS

a. Configurations of QRS waves of the spatial velocity ECG in atrial

septal defect and their relationships with the hemodynamic data.

QRS waves of the spatial velocity ECG in atrial septal defect were clas-

sified in 5 groups based on the following criteria as shown in Fig. 3.

Group A1: D•…1/2 peak value, and S•†1/3 peak value (32 cases, 57%)

Group A2: D•…1/2 peak value, and S<1/3 peak value (10 cases, 18%)

Group B1: D>1/2 peak value, and S•†1/3 peak value (8 cases, 14%)

Group B2: D>1/2 peak value, and S<1/3 peak value (1 case, 2%)

Fig. 3. Various patterns of QRS waves of the spatial velocity ECG of

atrial septal defect.
Vol.13
No.5 SPATIAL VELOCITY ELECTROCARDIOGRAM 411

Table I. Means and Standard Deviations of the Various Hemodynamic

Parameters in Each Group of Atrial Septal Defect
and Their Statistical Comparisons

PA: pulmonary artery pressures, RV: right ventricular pressures, RV-PA: pressure gradient
between RV and PA; Unit: mm.Hg ; *p<0.05

Group C: One-peaked pattern of main QRS wave, or slurring of the peak

of main QRS wave (5 cases, 9%)
A pattern of group A1 was most frequently observed, and was considered
as the basic pattern in atrial septal defect. Mean values and standard devia-
tions of the pulmonary artery and the right ventricular pressures, pressure
gradients between right ventricle and the pulmonary artery (systolic and
mean), and the areas of atrial septal defect determined during surgery in these
5 groups were shown in the Table I. Statistical comparisons of these mean
values in each of 5 groups were also shown. Mean value of the systolic pres-
sure of the right ventricle was significantly higher in A1 than in A2 group
(p<0.05). Mean values of the pulmonary artery and right ventricular pres-
sures (both systolic and mean) were slightly higher in B1 than in A1 group, but
these differences were not significant. Mean values of the pulmonary artery
(mean pressure) and the right ventricle (systolic and mean pressures), and
pressure gradient (systolic) were slightly higher in A1 than in C group, but these
differences were not statistically significant. Mean value of the pulmonary
artery mean pressure was significantly higher in A1 than in A2 group (p<0.05).
There was no significant difference in hemodynamic data between A2 and C,
and also between B1 and C groups. It was assumed from these results that the
grades of the hemodynamic loadings for the right heart were higher in A1 and
B1, in which the velocities of S were higher than in the other groups.
412 MORI, ET AL. Jap. HeartJ. S
eptember,1972

Table II. Means and Standard Deviations of Velocities and Times

of QRS Waves of the Spatial Velocity ECG in Normals and
Atrial Septal Defect, and Their Statistical Comparisons

Vet.:mV./sec.,Time: msec.; *p<0.05, **p<0.02, ***p<0.01

b. Correlations between QRS waves of the spatial velocity ECG and the
various hemodynamic parameters.
Thirty-three parameters of QRS waves of the spatial velocity ECG de-
scribed in the section of measurements were measured or calculated. Statis-
tical comparisons of the mean values of the main items of these parameters
between the normal group and atrial septal defect were shown in the Table II.
These values of QRS waves of the spatial velocity ECG were correlated with
the various hemodynamic parameters such as pulmonary artery, right ventri-
Vol.13
No.5 SPATIAL VELOCITY ELECTROCARDIOGRAM 413

cular pressures and pressure gradients between the right ventricle and the
pulmonary artery (both systolic and mean), and thus 210 of coefficients of
correlations were obtained by means of the electronic computer. These re-
sults were summarized in Fig. 2.
The velocities of D, Td, F, and S showed significant positive correlations
with the mean pressures of the pulmonary artery and the systolic and mean
pressures of the right ventricle (D showed the positive correlation only with
the mean pulmonary artery pressure). QRS-d time showed negative cor-
relations with the mean pulmonary artery and right ventricular pressures.
QRS-e time and (QRS-d)/d ratio showed negative correlations with the mean
right ventricular pressure. QRS-f time showed a negative correlation with
the mean pulmonary artery pressure.
Thus the velocities at the beginning portions of the terminal delay (TD,
F) and the peak velocity in the terminal delay (S) showed positive correlations,
and the duration of the terminal delay showed negative correlation with the
various hemodynamic parameters indicating right ventricular overloadings in
atrial septal defect.
c. Differentiations of atrial septal defect, complete right bundle branch
block and normal subjects by means of the spatial velocity ECG.
Fig. 4 showed the distributions of the peak velocity in the terminal delay

Fig. 4. Distributions of the peak velocities in the terminal delay (S) and
the durations of terminal delay (QRS-td, QRS-f) in normals, usual type of com-
plete right bundle branch block and atrial septal defect. NORM: Normal
(56 cases), RBBB: Usual type of complete right bundle branch block (20
cases), ASD: Secundum type of atrial septal defect (56 cases) .
 Jap. Heart J.
414 MORI, ET AL. September,1972

Fig. 5. Scattergram of the peak velocities in the terminal delay (S) and
the durations of the terminal delay. NORM: Normals (56 cases), RBBB:
Usual type of complete right bundle branch block (20 cases), ASD: Secundum
type of atrial septal defect (56 cases).

(S) and the durations of terminal delays (QRS-f, QRS-td) in atrial septal
defect, complete right bundle branch block without demonstrable heart
disease, and normal subjects. Velocities of S were much higher in atrial septal
defect than in the other 2 groups, permitting relatively good separations for
atrial septal defect from the other 2 groups. The separations between the
normal group and the complete right bundle branch block were very poor by
this parameter.
Durations of terminal delay (QRS-f, QRS-td) were much longer in the
complete right bundle branch block than in the other 2 groups. Normal
subjects and complete right bundle branch block were separated perfectly by
this parameter, but the distributions of right bundle branch block and atrial
septal defect showed some overlappings.
Fig. 5 showed scattergram of the normal subjects, atrial septal defect and
complete right bundle branch block, plotting the duration of the terminal delay
(QRS-td) on the horizontal (X) axis and the velocity of S on the vertical (Y)
axis.
The velocities of S were less than 30mV./sec. and the durations of ter-
minal delay (QRS-td) were more than 55msec. in all cases of the complete
right bundle branch block. Thus the separations of complete right bundle
branch block from the other 2 groups were satisfactory, giving the diagnostic
Vol.13
No.5 SPATIAL VELOCITY ELECTROCARDIOGRAM 415

positivities of 100%. The separations of atrial septal defect from the normal
subjects were obtained in all except 3 of the former and 4 of the latter, using the
criteria of the velocity of S of 50mV./sec. or more and the duration of terminal
delay of 35msec. or more. Thus the diagnostic positivities of 93%, 100%,
and 95% were obtained in normal subjects, in complete right bundle branch
block, and in atrial septal defect respectively, using only 2 parameters, name-
ly peak velocity and duration in the terminal delay of QRS waves of the spatial
velocity ECG.

DISCUSSION
Velocities of inscription of vector loops were usually expressed by marking
the vectorcardiographic loops with saw-toothed waves of the definite frequen-
cies. Such methods, however, offer to us only approximations for the veloc-
ities of inscription of vector loops. The spatial velocity electrocardiography
should be used for the precise investigations.
Shapiro3) (1952), Halmos et al.4) (1961), Angelakos5) (1962), Langner
and Geselowitz6) (1962) reported on the first derivatives of the electrocar-
diograms. The concept of the spatial velocity ECG was first introduced by
Hellerstein and Halmin7) (1960). They obtained the spatial velocity ECG
mainly by calculations and drawings, but also showed a record by means of
the analog computer. Systematic clinical investigations of the spatial velocity
ECG have been investigated by Sano et al.8) (1964) and Mori et al.1) (1967).
Mori et al.2),9)-11)(1967) investigated the spatial velocity ECG as a part of
studies on the analog computer analysis of the electromotive forces of the heart
quantitatively as well as qualitatively, and have been clarifying the clinical
usefulness of this new methods.
The most characteristic ECG findings in atrial septal defect is incomplete
right bundle branch block. This finding is expressed as terminal delay or
closed spacing of time markers at the terminal portion of QRS loop. This
finding is considered as the expression of the hypertrophy of the outflow tract
of the right ventricle. So the terminal delays in atrial septal defect and in the
usual type of complete right bundle branch block have somewhat different
clinical significance. Diastolic overloadings of the right ventricle are present
in atrial septal defect, and also systolic overloadings may be combined in the
advanced cases in which the pulmonary hypertension is present . The
characteristic findings of QRS waves of the spatial velocity ECG in usual type
of complete right bundle branch block were that the grades and the durations
of the delay in the later half of QRS waves were much pronounced although
the initial half of QRS wave was inscribed normally.
416 MORI, ET AL. S
Jap. Heart J.
eptember,1972

Terminal delay (TD) began at 51.4•}8.2msec. (53.9•}8.4msec. for F),

and then the velocity increased again markedly (S) in atrial septal defect. The

peak velocities in the terminal delay (S) and the velocities at the beginning

portions of the terminal delay (TD, F) showed positive correlations with pul-

monary artery and right ventricular pressures. Mean values of the pulmonary

artery and the right ventricular pressures were higher in the groups in which

the velocities of S were larger. These portions of QRS waves of the spatial

velocity EGG corresponded to the instantaneous QRS vectors directing right

wards anteriorly as determined by the simultaneously recorded 3 scalar

ECG, reflecting the right ventricular activations. So it was considered that

the increase of the velocity of these portions might be related with the hyper-

trophy of the right ventricle.

Increased spacings of time markers of QRS loops or low amplitudes of

QRS waves of the spatial velocity ECG lasting for some extent of time indicated

the presence of vectors of similar magnitudes which directed to the similar

spatial directions, reflecting the intraventricular conduction disturbances.

It was considered that both the intraventricular conduction disturbances

and the ventricular hypertrophy were present in atrial septal defect, because of

the wide terminal delay and the partial increase of the velocity in the terminal

delay. Durations of the terminal delay of QRS waves of the spatial velocity

ECG showed significant negative correlations with the pulmonary artery or the

right ventricular pressures. The shortening of the duration of the terminal

delay reflected the increase of the grades of the overloading of the right ven-

tricle, so it was considered that these findings more related to ventricular

hypertrophy than the intraventricular conduction disturbances.

It has well been known that the velocity of inscription of QRS loops were

slower at the initial and the terminal portions of QRS loop than at the middle

portions of the outgoing or the returning limbs of QRS loops in the normal

subjects. Durations of these physiological terminal delay, however, are much

shorter than those in atrial septal defect and complete right bundle branch

block. The velocities of S were less than 30mV./sec., and the durations of the

terminal delay (QRS-td) were more than 55msec. in all cases with usual type

of complete right bundle branch block, and thus the complete right bundle

branch block was easily differentiated from the normal subjects and atrial

septal defect. Atrial septal defect and the normal subjects were differentiated

in all except 3 cases in the former and 4 cases in the latter, using the criteria

that the velocity of S of 50mV./sec. or more and the duration of the terminal

delay (QRS-td) of 35msec. or more. Thus 93% of the normal subjects, 95%

of atrial septal defect, and 100% of the usual type of complete right bundle

branch block were differentiated by 2 parameters concerning the terminal

Vol.13
No.5 SPATIAL VELOCITY ELECTROCARDIOGRAM 417

delay of QRS waves of the spatial velocity ECG.

REFERENCES

1. Mori, H., Nagayama, T., and Nakasone, K.: Jap. Circulat. J. 31: 976, 1967 (abstract) (in
Japanese).
2. Mori, H., Nagayama, T., Oda, T., Oshita, K., Shibata, T., Takeshita, I., Tetsuo, M., Ide, T.,
Wang, J., Nakasone, K., and Hagihara, S.: Jap. Circulat. J. 32: 149, 1968.
3. Shapiro, P.J.: Cardiologia 21: 17, 1952.
4. Halmos, M., Lolossvary, B., Caccamo, L.P., and Saadi, E.: Quart. Bulletin St. Elizabeth
Hosp. 1: 71, 1961.
5. Angelakos, E.T.: J. Appl. Physiol. 17: 1023, 1962.
6. Langner, P.H., Jr. and Geselowitz, D.B.: Circulat. Res. 10: 220, 1962.
7. Hellerstein, H.K. and Hamlin, R.: Am. J. Cardiol. 6: 1049, 1960.
8. Sano, T., Suzuki, F., and Minami, S.: Jap. J. ME. & BE. 2: 277, 1964 (in Japanese with
English abstract).
9. Mori, H.: Jap. Circulat. J. 35: 791, 1971.
10. Mori, H., Nagayama, T., Shibata, T., and Takeshita, I.: Jap. Circulat. J. 33: 931, 1969.
11. Mori, H., Nagayama, T., Takeshita, I., and Hirahashi, T.: Jap. Heart J. 10: 516, 1969.