Vectorcardiographic QRS loop in spontaneous pneumothorax

J. ELECTROCARDIOLOGY 20(5), 1987, 375-382

Vectorcardiographic QRS Loop in Spontaneous Pneumothorax BY TAKESHI TSUTSUMI, M.D.,* TOSHIHIRA KATO, M.D., HIROFUMI OSADA, M.D.,t KEN-ICHI HARUMI, M.D., F.A.C.C.,t KEN INOUE, M.D., HIDEO KOMOTO, M.D., HAZIME SUZUKI, M.D. AND EISEI NOGUCHI, M.D.**

SUMMARY Vectorcardiographic QRS loops were recorded in twenty-nine patients with primary spontaneous pneumothorax (SP), comprised of eighteen of left SP and eleven of right SP. The configurations of QRS loops in acute to recovery phases were compared. The patients were classified into three groups according to the degree of coUapse of the lung (Group A: 25% or less, Group B: 25% to 50%, Group C: 50% or more). The major features of the QRS loop in SP were as follows: Left SPmLeftward QRS force was markedly reduced and the mean QRS axis showed a shift to the inferior and posterior. The greatest changes in the QRS loop appeared in group B. Right SP--The mean QRS axis tended to shift to the posterior and to the right. For clarifying the cause of the changes in the QRS loop, a simulation study was performed with a two-dimensional electrical field model. The results of the simulation study strongly suggested that the alterations of the QRS loop in spontaneous pneumothorax were mainly due to extracardiac reasons. to the change in the extracardiac electrical field. 2 Lepeschkin supported this hypothesis in his textbook. 3 Subsequently several authors reported electrocardiographic changes in pneuomothorax including certain special cases. 4~6However, we could not find any detailed studies on the vectorcardiographic QRS loop changes in spontaneous pneumothorax (SP). In the present study, vectorcardiograms in acute and recovery phases of primary spontaneous pneumothorax were recorded by Frank's corrected orthogonal lead system. A comparative analysis of the findings in both conditions was undertaken. The genesis of the alterations of the QRS loop was evaluated with a two-dimensional model study.

In 1928, Master reported seven cases of pneumothorax and pointed out the characteristic electrocardiographic changes, including a prominent S wave in Lead I and a right axis deviation of the QRS wave. 1 The author also suggested that a rotation of the heart and a displacement of the mediastinum would be the major factors for the electrocardiographic changes in pneumothorax. 1 Later on, electrocardiographic changes in two cases of mediastinal emphysema with left pneumothorax were reported by Littman. 2 He first noted the effect of the body position on the electrocardiographic changes in pneumothorax, which were attributed

* Associate; Division of Cardiology,Showa University Fujigaoka Hospital, Yokohama 227, Japan.

MATERIALS AND METHODS Vectorcardlographic Analyses

t Professor of Medicine; Division of Cardiology, Showa University Fujigaoka Hospital, Yokohama 227, Japan.

Twenty-nine patients consisting of twenty-five men and four women who had primary SP were selected. Vectorcardiograms were recorded in these patients during the acute phase and after complete recovery. The polarity of the Frank lead system was chosen as +X left, +Y foot and - Z back. Eighteen of the patients had left pneumothorax and eleven right pneuomothorax. Ages ranged from 11 to 52 years. In all cases, chest X-ray films (anterior-posterior view) were taken immediately after recording the vector-cardiograms. The degree of collapse

** Professor of Medicine, Pulmonary Division, Showa University Fujigaoka Hospital, Yokohama 227, Japan. From the Division of Cardiology, Showa University Fujigaoka Hospital, Yokohama 227, Japan. Reprint requests to: T. Tsutsumi, M.D., Division of Cardiology, Showa University Fujigaoka Hospital, 1-30 Fujigaoka, MidoriKu, Yokohama 227, Japan.

375

376

TSUTSUMI ET AL

TABLE I Number of Each Group of Patients with Pneumothorax

Group A (25%->)

Group B (25-50%)

Group C (50%_-<)

Total

Left P.

4

8

6

18

Right P.

5

3

3

11

severity

Left P; left pneumothorax; Right P: right pneumothorax

of the lung was calculated from the chest X-ray film in each case by Kircher's method. 7 The patients were classified into three groups according to the degree of collapse as shown in Table 1. Group A included the patients whose degree of collapse was 25 % or less, Group B those with 25% to 50% collapse and Group C those with 50% or more collapse. The vectorcardiograms were analyzed for the orientation of the maximum QRS vector (max QRS vector), R and S wave amplitudes in scalar vectorcardiograms and the configuration of the QRS loop in acute and recovery phases.

Simulation Study The model of the horizontal section of the torso was constructed from electrically conductive papers made for mapping out electromagnetic fields (Tomy Anacon Paper: Tomoegawa Paper-manufacturing Company, Ltd., 1-5-15 Kyobashi, Tokyo 104, Japan). The paper is about 0.24 mm thick and processed so as to possess uniformly 160 ohm resistance along the surface. The variations of the values in electric resistance of the papers are less than 5 %. The configuration of the paper model shown in Fig. 1 corresponds to the cross section of the thorax viewed from the rostral side. In Fig. 1 the X axis from right to left and the Z axis from back to front are shown. The lengths of X and Z axes of the model are 37, 28 cm respectively. The insulations simulating the cross section of SP were made up first by cutting out the area a (area A), and then the area a plus b (area B) and then area a plus b plus c (area C). In the recumbent patients air accumulates first in the anterior cost-mediastinal angle; increase in the degree of pneumothorax causes air to extend further posteriorly without much anterior change. This justifies the model configuration. Locations indicated by the numbers i to 40 on the circumference of the model are defined by the section with radial lines drawn from the center at equal angles. The bipolar copper electrodes with an interelectrode distance of 1.5 cm were placed at the site 7 cm posterior from the site 1 and the midpoint of the dipole was located 2 cm from the Z axis. A sine wave signal with 1 KHz, 6.0 Vp-p was applied to the bipolar electrodes from the generator (Model SY-118; N F Circuit Design Block Co., Ltd., Japan). The reference electrode was constructed by connecting three points,

marked by triangles in Fig. 1, on the edge of the model through 100 Kohm resistors. One of the component electrodes was placed at site 1, and the other two at sites to make 30 degrees with the X axis (Fig. 1), corresponding to the horizontal counterpart of Wilson's Central Terminal. But the reference potential used in this study is not identical with Wilson's Central Terminal. Any reference electrode is not exactly with zero potential as the electrodes at infinity in an infinite medium. The selection of the reference bears no critical importance for the relative voltage manifested in the vectorcardiogram. But it is desirable to select a neutral point as the reference in drawing isopotential lines. In the case of a Z dipole, the front electrode is located in the direction of the dipole which may cause a deviation of the reference potential. For instance, however, other reference electrodes constructed from mirror points with respect to the X axis are estimated as less than 0.1 V. The effect of the dipole direction on the reference potential appears practically not significant. The potential distribution on the edge of the model was measured at forty points as indicated in Fig. 1. The isopotential lines within the model were drawn by the following method. One of the pair of search probes was fixed somewhere on the model and another probe was moved arbitrarily on the model to find the point where the potential difference between these points was zero.

RESULTS Left P n e u m o t h o r a x T a b l e II shows the changes in the a m p l i t u d e s of the R a n d S waves in the scalar leads. In the acute phase, R waves of the X axis decreased and those of the Z axis increased. T h e g r e a t e s t changes app e a r e d in group B. T h e S waves of the Z axis were slightly increased in groups B and C. T a b l e I I I shows the changes in the o r i e n t a t i o n of the m a x Q R S vector in each of the planes. In the acute phase, the m a x QRS vector showed an inferior a n d posterior shift a n d a p p r o a c h e d the Y a n d Z axes. T h e greatest shift was observed in group B. T h e m e a n QRS loops in the acute p h a s e s u p e r i m p o s e d with those in the

J. ELECTROCARDIOLOGY 20(5), 1987

QRS LOOP AND PNEUMOTHORAX

377

(Back)

E$ ~

(right)

~

~

/

=X

(left)

'

25

24

23 22 21 20 19

18

17

I6

27

T5

28

14

30 29(

13 12

z (antericr)

31"

]

% oscilloscope

3

,

~ 35

~ ~ -- - ~ 38

39 40 1

. 2

, 3

, 4

pe

~ ;;;"

7

36 37

'

lOOk~

5

i ~ o s c Fig. 1. Two-dimensional electrical field model. Right panel: Two-dimensional electrical field model is illustrated. The numbers (1-40) around the model show the points of measuring the potentials on the edge of the model. The crosses connected to the oscillator show the position of bipolar electrodes. The round dotted line indicates the isopotential line. Areas a, b, c in the left lung model are cut out in twins. Amp: differential amplifier, OSC: sine wave oscillator. Left panel: The coordinate axes of the model are illustrated. The angular notations and the direction of positive polarity, represented by the arrowhead, are indicated. The axes of horizontal are X and Z. Solid triangles show the location of reference electrodes.

TABLE II

R and S Waves Amplitudes of Scalar Vectorcardiogram in Left Pneumothorax

X

Y

Z

R(mv)

S(mv)

R(mv)

S(mv)

R(mv)

S(mv)

Group A

SP

0.37+0.12

0.05+0.04

0.72+0.03

0.40+0.61

0.35+0.08

1.15+0.31

(n=4)

N

0.59+0.03

0.29+0.07

0.75+0.28

0.23+0.32

0.33+0.03

1.27+0.35

D

-0.32

-0,26

-0.04

+0.17

+0,02

-0.12

Group B

SP

0.37+0.08

0.18+0.12

1.10+0.43

0.16+0.17

0.55+0.28

1.28+0.41

(n=8)

N

1.63+0.25

0.16+0.13

0.95+0.80

0.14+0.09

0.46+0.25

0.96+0.54

D

-1.25

+0.02

+0,06

+0,04

+0.11

+0.30

Group C

SP

0.45+0.17

0.24+0.07

1.24+0.65

0.08+_0.05 0.60+0.22

1.12+0.19

(n=6)

N

1.21+0.30

0.30+0.07

1.22+0.33

0.10+0.09

0.58+0.33

0.86+0.31

D

-0.70

-0.06

+0.02

-0.03

+0.02

+0.27

D: mean differences of R, S waves amplitudes in SP from R, S waves amplitudes after recovery. J. ELECTROCARDIOLOGY 20(5), 1987

378

T S U T S U M I ET A L

TABLE III Average Orientation of Maximum QRS Vector in Left Pneumothorax Frontal

Horizontal

Lt, Sagittal

81.7+17.5

269.7--- 7.2

45.0+19.9

N

41.7+13.6

280.7---44.5

38.3-1-24.3

D

+ 40.0

- 11.0

+ 6.7

GroupB

SP

8 0 . 9+- 5.5

269.0 + 8.5

52.7+24.9

(n=8)

N

27.5---19.3

345.5---16.8

42.5+39.3

D

+43.3

- 76.5

+ 10.2

GroupC

SP

70.2-+ 9.1

276.5+ 9.0

57.0---23.8

(n=6)

N

46.8•

326.5+47.9

87.8-+43.7

GroupA SP (n=4)

D

9.0

+23.5

-38.5

-30.7

Unit: degree. Other abbreviations are the same as Table II.

recovery phase are illustrated in Fig. 2. The mean QRS loops in the acute phase (dotted lines) showed an inferior and posterior shift, and the leftward QRS force was reduced markedly in comparison with that in the recovery phase (solid lines). Consequently, the QRS loops in the acute phase were characteristically narrower. The results from Fig. 2 also indicated that the shift of the QRS loop was not always proportional to the degree of collapse of the lung.

Y

Y

%

Right Pneumothorax The R waves of the Y axis were slightly reduced, and the S waves of the Z axis were slightly increased in groups B and C. The orientations of the max QRS vector in the frontal and horizontal planes were not shifted significantly, but in the left sagittal plane they tended to shift to the inferior. These changes in the QRS loops were proportional to the degree of collapse of the lung.

Y

Y

L X

'

Z

\ ~ s

Z

I : X

Y

A

i

x

]

X

z

",~

Y

sI Z

,~ ...... >

Y

Y

", ?

~,

~+~

'.

~'~'-..,," Lt. S.P.

A (25% > )

a (25<<50%)

C (50% < )

Z ,~

Fig. 2. Mean QRS loops in left spontaneous pneumothorax and in the recovery phase. Solid lines show QRS loops in the recovery phase. Dashed lines show the QRS loops in spontaneous pneumothorax. In each QRS loop, the mean spatial instantaneous vectors were calculated every twenty milliseconds in each plane and linked with the successive lines. The mean QRS loops in pneumothorax and in the recovery phase in groups A, B and C were superimposed on the same scale. The left panel illustrates a representative case of left pneumothorax. J. ELECTROCARDIOLOGY 20(5), 1987

QRS LOOP AND PNEUMOTHORAX

379

2O 0! ' ~

A

+4-

_

Potential (mV)

!

0 -2 .4 012O

B

+ 4 84 Poterllial (mV)

+2

0 -4 1

Fig. 1 was parallel to the X axis in panel A and that in panel B was parallel to the Z axis. The solid lines show the isopotential lines with values of 0.1 V steps. The right panel of Fig. 3 shows the distribution of the potential along the edge of the model.

Two-Dimenslonal Electrical Field The electrical field in the homogeneous torso model is illustrated in Fig. 3. The direction of a dipole shown by the line connecting the poles in

C

B A

Fig. 3. Isopotential maps and potential distributions around the models in control state. Panel A: The potential distributions when an X dipole was energized. Panel B: The potential distributions when a Z dipole was energized. The potential distributions around the model are shown on the right side of both panels: the transverse axes in the graphs indicate the distance from the anterior center of the model. The numbers on the transverse axes correspond with the numbers around the models.

20

20

20

o~

1

"t"4'

"1-4

~

10

"t"4

20

10

20

--2

' --3

9- - 2

--4

. --4

-4

Fig. 4. Isopotential maps and the potential distributions around the model produced by an X dipole source after areas A, B and C (shaded areas) were cut out. Panels A, B, C show areas A, B, C cut out respectively. The same expressions apply as in Fig. 3. Unit: 100 inv.

J. ELECTROCARDIOLOGY 20(5), 1987

380

TSUTSUMI ET AL C

B

A 20

i

10

30

1

+4

+4

§

+;

_J

+2

\\

"\, 1 10

20

2'1

~'o

10

21

30

t

-

~'o,

Fig. 5. Isopotential maps and potential distributions around the model produced by a Z dipole source after areas A, B, C (shaded areas) were cut out. Panels A, B, C show areas A, B, C cut out respectively. The same expressions were used as in Fig. 3. Unit: 100 mv.

Fig. 4 demonstrates the electrical field produced by X dipole after shaded area A, B and C cut out, respectively. When area A was cut out, the potentials at locations 1-10 (left anterior edge) were slightly reduced. When areas B and C were cut out, the potentials at locations 3-20 (left antero-lateral edge) were reduced. At the same time, the current density in the area between the generator and the lung was increased. In addition, at the area opposite to the insulation corresponding to the right lung at the locations 21-31, the potential was also reduced to a lesser degree than the reduction of the potentials outside the left lung. Fig. 5 demonstrates the electrical field produced by a Z dipole after areas A, B and C were cut out. As the area of insulation became larger, the zero line on the left-side edge was shifted posteriorly, and the potentials at the locations 1-16 (anterolateral edge) augmented. The results from simulation studies of the electrical field showed that the effects on the potential distributions after unilateral insulation depended largely on the relationship between the orientation of the dipole and the insulation. DISCUSSION According to older studies, 1-2,8-1~the QRS wave changes of pneumothorax were chiefly described in

patients receiving pneumothorax therapy. The authors noted that characteristic QRS changes in pneumothorax consisted of a rightward shift of the frontal QRS axis and a decrease of the R wave amplitude in lead I. In a subsequent study of the electrocardiographic changes in pulmonary collapse therapy, Weinshel et al. 11 used standard 12lead electrocardiograms and noted in the chest leads that the transitional zone showed a tendency to be shifted to the left in right pneumothorax and the amplitude of QRS was diminished in left pneumothorax. Walston et al. 4 studied seven cases of left pneumothorax complicating other diseases and pointed out similar changes as described in older literatures.l-2.S-l~ In the present study, the QRS loop in primary SP was characterized as follows; Left SP: 1) Leftward QRS force was markedly diminished and the shape of the QRS loop became narrower along the Y and Z axes. 2) The max QRS vector was shifted inferiorly, posteriorly and to the right. 3) The greatest changes of QRS loops appeared in group B. Right SP: The mean QRS axis tended to be shifted inferiorly, the R wave of the Y axis slightly reduced and the S wave of the Z axis slightly increased in groups B and C. Compared with other studies, the QRS loop in right SP was changed to a lesser degree because in our series almost all patients with right SP did not

J. ELECTROCARDIOLOGY20(5), 1987

QRS LOOP AND PNEUMOTHORAX

have severe collapse. In addition, changes in the QRS loop in left SP particularly in groups B and C were not proportional to the degree of collapse of the lung. Walston's study also led to the same conclusion. 4 Since Master ascribed electrocardiographic changes in pneumothorax to longitudinal rotation of the heart, displacement of the mediastinum and the insulating effect of intrathoracic air, ~ several clinical studies have referred to it. 2,1~ Todd and Aronson 12 and Treiger and Lundy ~3 stated that although appreciable heart and mediastinal movements were found in chest X-ray films of pneumothorax, the changes in the electrical axis were not parallel with anatomical findings. In 1946, Littman investigated the effects of body position on the electrocardiographic changes induced by pneumothorax. 2 He noted that the electrocardiographic changes in pneumothorax were observed when the patients were in the supine position, but not in the prone or right lateral position. From this fact, it was suspected that the changes of QRS wave in SP are due to the location of the intrathoracic air. Then the author postulated that the alteration of the extra-cardiac electrical field caused by the insulating effect of intrathoracic air might be an important factor in electrocardiographic changes observed in SP. Katz and Korey 14 inserted rubber dams on the heart surface and recorded the changes in the voltage of the QRS waves of the standard limb lead electrocardiograms, and demonstrated electrocardiograms similar to those caused by pneumothorax, but those changes were smaller since the chest was open. In the present study, the effect of left-sided insulators on the field of the X dipole showed marked reduction of left lateral voltage (5-15) as well as a slight reduction of the voltage at the opposite side (20-25) despite the unilateral insulation, and the current density on the inner side of the lung was augmented. The former phenomena may be interpreted as the effect of a conducting plane wall boundary on the dipole oriented tangentially to the interface and on the image dipole oppositely directed. For instance, a planar boundary between conducting and insulating media has an accentuating effect on the tangential dipole and a reducing effect on the normal dipole? 5 Related action of the boundary near the dipole seems to be the reason for the reduction of the voltage on the right edge of the torso. The present results may account for the following vectorcardiographic findings in SP; that is, reduced leftward QRS force, reduced R wave J. ELEGTROCARDIOLOGY 20(5), 1987

381

on the X axis, increased S wave on the Z axis, and narrower shape of the QRS loop along the Z and Y axes. The present data also showed that the potential outside the insulation (1-15) was increased in proportion to the size of insulation when the Z dipole was energized. This result implies that the voltage outside the insulator near the cardiac generator fluctuated in a different fashion according to the relationship between the direction of the dipole and the insulation. The experimental facts may explain that the development of abnormal vectorcardiograms in SP is not always proportional to the degree of collapse of the lung. Another explanation of why R, S waves in Z lead increased less in group C than in group B is that in group C the insulation extends to the back of the heart and then becomes partly in series with the two dipoles and partly counteracts the augmentation caused by the insulation parallel to them at B. Up to the present, various kinds of model studies for nonhomogeneous media have been reported. TM Two-dimensional models give qualitative insights into the effects of nonhomogeneities. TM A detailed study of the insulating effects of bilateral lungs has been performed by Nelson. 17The results of our study were, in part, similar to Nelson's conclusions, but in contrast to his description on the effect of lungs, our observations are concerned with pathological conditions of unilateral insulating mass. Acknowledgements: The authors express sincere appreciation to S. Mashima, M.D. for his critical review of the manuscript, and to Y. Teramachi, Ph.D. for expert technical assistance. REFERENCES 1. MASTER,A M: The electrocardiographic changes in pneumothorax in which the heart has been rotated. Am Heart J 3:472, 1928 2. LITTMAN,C D: Electrocardiographicphenomena associated with spontaneous pneumothorax and mediastinal emphysema. Am J Med Sci 212:682, 1946 3. LEPESCHKIN, E: Modern Electrocardiography. Williams and Wilkins Co, Baltimore, 1951, p 215 4. WALSTON,A B E, BaEWER, D L, KITCHENS,C S AND KROOK,J E: The electrocardiographic manifestations of spontaneous left pneumothorax. Ann Intern Med 80:375, 1974 5. KURITZKY,P AND GOLDFARB,A L: Unusual electrocardiographic changes in spontaneous pneumothorax. Chest 70:535, 1976 6. FELDMAN, T AND JANUARY, C T: ECG changes in pneumothorax. A unique finding and proposed mechanism. Chest 86:143, 1984

382

TSUTSUMI ET AL

7. KIRCHER,L T: Spontaneous pneumothorax and its treatment. JAMA 155:24, 1954 8. ARMEN,R N ANDFRANK,T V: Electrocardiographic patterns in pneumothorax. Dis of Chest 15:709, 1949 9. FELDMAN,D ANDSILVERBERG,C-"Electrocardiographic changes in pulmonary collapse therapy. Am Heart J 35:800, 1948 10. SILVERBERG, C, KINGSLAND, R AND FELDMAN, D: Electrocardiographic changes in pulmonary collapse: Artificial and spontaneous left sided pneumothorax studied by conventional and unipolar methods. Dis of Chest 17:181, 1950 11. WEINSHEL,M, MACK,I, GORDON,A AND GORDON, S: Electrocardiographic changes accompanying pulmonary collapse therapy and thoracic surgery. Am Rev Tuberc 64:50, 1951 12. TODD,G S ANDARONSON,D M: Effect of altitude on cases of pneumothorax. Lancet 13:597, 1943

13. TREIGER,I ANDLUNDY,C J: The correlation of shifting electrical axis of the heart with X-ray observations in artificial pneumothorax. Am Rev Tuberc 29: 546, 1934 14. KATZ,L N AND KOREY, H: The manner in which the electric currents generated by the heart are conducted away. Am J Physiol 111:83, 1935 15. MCFEE, R ANDRUSH, S: Qualitative effects of thoracic resistivity variations on the interpretation of electrocardiograms: The low resistance surface layer. Am Heart J 76:48, 1968 16. RusH, S AND NELSON, C V: The effects of electrical inhomogeneity and anisotrophy of thoracic tissues on the field of the heart. In Theoretical Basis of Electrocardiology. C V Nelson and D B Geselowitz, Eds, Clarendon Press, Oxford, 1976, p 323 17. NELSON, C V: Human thorax potentials. Ann N Y Acad Sci 65:1014, 1957

J. ELECTROCARDIOLOGY 20(5), 1987