- Email: [email protected]
Reduction in ventricular endocardial and epicardial potentials during acute increments in left ventricular dimensions
Reduction in ventricular endocardial and epicardial potentials during acute increments in left ventricular dimensions
Jon Lekven, M.D.* Kanu Chatterjee, M.B., M.R.C.P. (Lond. John V. Tyberg, M.D., Ph.D., F.A.C.C.** William W. Parmley, M.D., F.A.C.C.
and Edin.),
F.A.C.C.
San Francisco, Calif.
The observation that a reduction in right ventricular endocardial potentials occur in patients with acute myocardial infarction or massive pulmonary embolism, clinical situations likely to be associated with acute ventricular dilatation, indicates that a relationship might exist between the magnitude of ventricular potentials and ventricular volumes.‘, ? Recently it was demonstrated in an experimental study that potentials recorded from the endocardial surface of the left and right ventricles decreased markedly when left ventricular volume was increased by blood transfusion.” Preliminary data suggested that potentials recorded from the epicardial surface also decreased during volume expansion. However, the influence of changes in ventricular volumes on endo and epicardial potentials, recorded simultaneously from the same site, has not been analyzed.
The ventricular excitation propagates from the endocardial side with predominantly radial components, whereas the propagation is more tangentially oriented in outer epicardial layers.“, ‘, Analysis of spatial surface electrocardiograms have indicated opposite effects of changes in the intracavitary blood mass and volume on radial and tangential components of the QRS complex.“, i Our preliminary observations, however, indicated directionally similar changes in endocardial and epicardial potentials during acute changes in ventricular volumes. Thus, the relationship between changes in endocardial and epicardial potentials during acute changes in ventricular volume is not totally clear. The present study was, therefore, undertaken to clarify the relationship between epicardial and endocardial potentials during acute variations in left ventricular diameter and, therefore, left ventricular volume. Methods
From the Cardiovascular Division of the Department the Cardiovascular Research Institute, University Francisco, Calif. This work was supported Institute Program Project
in part Grant
by National HL 06285.
Received
for publication
June
12, 1978.
Accepted
for publication
July
17, 1978.
Reprint requests, Room 1186, Moffitt Calif. 94143.
Dr. Kanu Hospital,
Chatterjee, University
of Medicine of California,
Heart,
Lung
Cardiovascular of California, San
*Dr. Lekven is the recipient of a United States Public Health Fogarty International Fellowship (FO 5 TW 2371). His present is: Cardiovascular Research Laboratory, University of Bergen, land Hospital Surgical Department, N-5016 Bergen, Norway. **Dr. Tyberg is the recipient of a United States Public Research Career Development Award (HL ooO16) and from the American Heart Association (AHA-76788).
200
August,
1979, Vol. 98, No. 2
and
and San Blood
Division, Francisco, Service address Hauke-
Health Service a Grant-in-Aid
Experimental preparation. Twelve mongrel dogs weighing 16 to 25 kilograms were anesthetized initially with sodium thiopental(25 mg./Kg. intravenously). Morphine sulfate (45 mg. intravenously) was then given and anesthesia was maintained throughout the experiments with regular injections of 15 mg./hr. For muscle relaxation, a continuous infusion of succinyl choline chloride (20 mg./Kg./hr.) was given. The dogs were intubated and ventilated with a positive pressure respirator (Harvard Apparatus, Millis, Mass.); respiration was adjusted under the guidance of frequent analyses of arterial blood pH, p0, and $0, (BMS MK2, Radiometer, Copenhagen).
0002-8703/79/080200
+ 07500.70/O
3 1979
The
C. V. Mosby
Co.
Endocardial The chest was opened by mid-sternotomy and the pericardium was opened and retracted. The right atrium was paced in nine of the 12 dogs with an electrode attached to the atria1 appendage (Grass Instruments, Quincy, Mass.). Electrical measurements. Potentials from the endocardial surface of the left and right ventricles (Endo-Pot) were recorded with thin wire hook electrodes with a diameter of 0.13 mm. (Elgiloy, Elgiloy Corp., Elgin, Ill.) on an electrocardiograph (Honeywell, San Jose, Ca.). The entire length of the wire, except the distal 4 mm. was insulated. The electrodes were mounted in 23 gauge needles and were introduced into the ventricular cavities; upon withdrawal of the needle, the electrode hook sank approximately 2 mm. into the endocardial tissue layer. Initially, a contact pattern of ST segment elevation was regularly observed, but after one hour the ST segment elevation was markedly reduced. Electrodes showing persistent ST segment elevation greater than 2 mV. after one hour were discarded from subsequent analysis. Potentials from the epicardial surface (EpiPot) were recorded with a cotton wick electrode with an area of 5 mm.’ The epicardial sites were defined according to anatomical landmarks, so that they corresponded to the underlying endocardial wire electrodes. A total of 32 pairs of endocardial and epicardial electrodes were used for recording left ventricular endo and epicardial potentials, respectively. In five dogs one electrode was placed through the right ventricular cavity into an endocardial position on the left side of the interventricular septum. In four dogs, right ventricular endocardial potentials were also recorded from the free wall (endocardial surface) of the right ventricle. In three of the dogs, wire electrodes were also inserted in an epicardial position, and the Epi-Pot recorded with wire electrodes could be compared with the potentials recorded with wick electrodes at the same epicardial sites. The correlation of 158 paired observations at eight different sites with the two methods was: Wick-Pot = 0.93 Wire-Pot + 0.47 The regression coefficient was r = 0.972, and the regression line was not statistically different from the line of identity. The Wilson central terminal of extremity leads was used as reference for all electrograms. Signals below 0.5 Hz and above 100 Hz were excluded by
American Heart Journal
and epicardial potentials
bandpass filters. Potentials were accepted for analysis provided correct endocardial electrode positions were verified by postmortem examination of the heart, and only the potentials recorded from such satisfactory electrode positioning were analyzed. The amplitude of QRS complexes from four consecutive beats were averaged and regarded as endo or epicardial potentials. Ectopic beats or QRS complexes revealing conduction disturbances were excluded. The potentials were integrated over time by analog computer (Electronics Assoc., Inc., West Long Branch, New Jersey), giving the area of the QRS complex (mV. * msec.) above the isoelectric T-P segment for each beat. The integrator was triggered and reset by the left ventricular pressure signals and appropriate delay controls. Hemodynamic measurements. Left ventricular diameter was continuously recorded by the ultrasound time between two piezoelectric crystals inserted into the anterior and posterior walls of the left ventricle.” Left ventricular pressure was measured with a solid state transducer (Konigsberg Instruments, Pasadena, Ca.) inserted from the left atria1 appendage. Aortic pressure was measured through a femoral catheter connected to a Statham P23Db transducer, and aortic flow was measured with an electromagnetic flowmeter (Carolina Medical Electronics, King, No. Carolina) on the ascending aorta. Experimental procedure. Control measurements of potentials and hemodynamic parameters were performed when the initial electrocardiogram contact pattern had disappeared. Homologous and prewarmed blood was then infused into the jugular vein through a wide-bore catheter for stepwise increase in left ventricular enddiastolic diameter (LVEDD) and pressure (LVEDP). For each step, the infusion rate was slowed down to maintain stable hemodynamics, and measurement of potentials and hemodynamic parameters were repeated. Infusion was continued until LVEDP had reached 15 to 25 mm. Hg; infused blood volume was 550 to 2,500 ml. Thereafter, blood was withdrawn stepwise to allow LVEDD and LVEDP to return to control. The above procedure allowed us to treat the recorded potentials as a regression problem for each electrode separately with respect to the changes in LVEDD during infusion and withdrawal of blood. Table I gives the control values and the values obtained at the highest LVEDD
201
et
Lekven
al.
Highest Load
Recovery
LV END0 r
RV END0
LVEDD
--r-
---?Fc
--f--
-
mm
---y--
-
----.-
-,,----
-
-
Fig. 1. Potentials recorded from the left (LV) and right (RV) ventricles with the interventricular stepwise infusion and withdrawal of blood. Potentials were recorded at corresponding endocardial epicardial @PI) sites of the ventricular walls. LVEDD = left ventricular end-diastolic diameter.
achieved, where each dog and electrode served as its own control. Student’s t test for paired data was applied in calculating statistical probability.” A p value less than 0.05 was regarded as statistically significant. As similar relationships between Endo-Pot and Epi-Pot and LVEDD were observed in paced and non-paced hearts, the data were, therefore, pooled for final analysis. Results Fig. 1 shows changes in potentials recorded from the left and right ventricular endo and epicardium and from the inter-ventricular septum in one experiment. The potentials from both endo and epicardial,surfaces decreased as LVEDD was increased stepwise during blood transfusion. When LVEDD returned to control after withdrawal of blood, the potentials increased and also returned to control values. The absolute magnitude of the potentials recorded either from endocardium or epicardium of the left or right ventricle varied considerably according to electrode site and also position in different dogs, ranging from 17 to 51 mV. (see Fig. 2). However, irrespective of the recording site and the initial magnitude,
202
septum during (ENDO) and
linear reduction in potential was observed for each individual electrode as LVEDD was increased. Both endocardial and epicardial potentials recorded from either ventricle and also potentials from the interventricular septum behaved similarly. Table I summarizes changes in potentials and hemodynamic parameters during maximum volume load achieved. On the average, LVEDD increased by 11 per cent, from 39.41 -t 1.72 to 43.60 f 1.89 mm. Similarly, LVEDP increased from 4.9 + 1.0 to 17.3 + 2.1 mm. Hg. Blood transfusion caused an increase, as expected, in cardiac output and the systolic blood pressures. To assessthe magnitude of changes in endo and epicardial potentials recorded from the same area of the ventricular wall, the ratio of Endo/Epi potentials was calculated for each pair of endocardial and epicardial electrodes separately. At control LVEDD of 100 per cent, the ratio of Endo/Epi potentials recorded from the left ventricle averaged 1.04 + 0.04, and at LVEDD of 111 per cent this ratio decreased to 0.72 & 0.05, suggesting a relatively greater reduction in left ventricular endocardial potential than in its counterpart, epicardial potential. Statistical
August,
1979, Vol. 98, No. 2
Endocardial
and epicardial potentials Endo RV
mV 20 i 1
40
mV 20
i, , , , 100 108
, Ept
1
0L-LL-L 100
108
RV
-\
IF?
1 100 P
I16
LVEDD, Fig. 2. Relationship between each electrode. Abbreviations
% of
control
cardiac potentials and left ventricular = the same as those in Fig, 1
diameter
analysis of the individual regression lines of the ratios of left ventricular Endo/Epi potentials revealed that the ratio was significantly reduced by volume loading (Slope = 0.016 -+ 0.002, p < 0.001). Such changes, however, were not observed in the ratios of right ventricular endo and epicardial potentials. Endo-pot recorded from the left ventricular septum behaved similarly to Endo-Pot recorded from the endocardium of the left ventricular free wall (Table I). The area of the QRS complexes, similar to its absolute magnitude, showed a linear reduction with increasing diameter of the left ventricle. Duration of the QRS complex did not change, suggesting lack of conduction disturbances (average 42.9 -+ 0.3 msec.). An apparently greater reduction in epicardial QRS area compared to the reduction in Epi-Pot could be accounted for by the fact that the integrator substracted small negative areas below the isoelectric base line (Fig. 1). Discussion
The present study was designated to investigate the influence of acute changes in ventricular
American Heart Journal
, II6
during
blood
infusion.
One line for
volume on concurrently recorded right and left ventricular endo and epicardial potentials. The findings indicate that acute changes in left ventricular diameters markedly influence not only the magnitude of the right or left ventricular endocardial potentials, as previously reported,” but also of their epicardial potentials. Thus, an increase in left ventricular diameter during blood transfusion was accompanied by a decrease in both endo and epicardial potentials recorded either from the right or the left ventricle. With an 11 per cent average increase in left ventricular end-diastolic diameter, there was a 15 per cent decrease in left ventricular epicardial and 37 per cent decrease in right ventricular epicardial potentials. As observed in previous investigations, both left ventricular endocardial (-27 per cent) and right ventricular endocardial (-36 per cent) potentials also decreased concurrently. The precise mechanism of such changes in ventricular potentials with changes in ventricular diameter is not clear. That changes in ventricular geometry and wall thickness may contribute needs to be considered. A considerable decrease in thickening of ventricular walls may occur during acute expansion of ventricular volume. During
203
Lekven
et al.
I. Effect of blood infusion on cardiac potentials (Mean t SEM in twelve dogs) Table
N Left ventricle: Endo-Pot, mV. Area, mV. * msec. Epi-Pot, mV. Area, mV. * msec.
Hg
load
P
33.1 k 1.8 622 F 43
23.9 zk 1.6 508 i 42
0.001 0.001
33 33
34.7 -t 1.3 281 + 35
29.6 f 1.2 219 k 32
0.001 0.01
31.8 f 4.4 568 k 124
23.0 i 3.4 431 I 99
0.002 0.05
4 4
22.8 k 2.1 264 + 81
14.5 + 0.9 205 k 61
0.01
4 4
27.0 + 3.5 222 t 28
17.0 -+ 3.2 106 + 34
0.01
Hemodynamic parameters: LVEDD, mm. 12 39.41 -+ 1.72 LVEDP, mm. 12 4.9 5 1.0 Hg LVSP, mm. HR, beats/ min.f CO, ml./min.
Highest
33 33
Interventricular septum: Sept-Pot, mV. 5 Area, 5 mV * msec. Right ventricle: Endo-Pot, mV. Area. mV. * msec. Epi-Pot, mV. Area, mV. * msec.
Control
NS
0.05
43.60 k 1.89 17.3 f 2.1
0.001 0.001
12 12
129 f 6 164 f 6
155 k 7 159k7
0.001 NS
5
3730 Ifr 120
5420 + 800
0.05
Endo-Pot
= endocardial potential; Epi-Pot = epicardial potential: Sept-Pot = interventricular aeptal potential; Area = integrated area of QRS complex above the isoelectric T-P segment; LVEDD = left ventricular end-diastolic diameter; LVEDP = left ventricular enddiastolic pressure; LVSP = left ventricular systolic pressure; HR = heart rate; CO = cardiac output; N = number of observations; P = probability value; NS = not significant. tAtria1 pacing in nine dogs.
such thinning of the ventricular wall, the tissue mass in the immediate vicinity of the recording electrode that delivers the electrical signals is reduced.3 This might account for reduction of both endo and epicardial potentials during volume expansion. In the present study decrease in left ventricular endocardial potentials was of greater magnitude than that of corresponding epicardial potentials. That such a difference in changes in left ventricular endo and epicardial potentials was observed also supports the hypothesis that thinning of the left ventricular free wall may be at least partly responsible for reduction in ventricular potentials during an increase in ventricular volume. During volume expansion subendocardial layers will be expected to be
204
stretched more than the subepicardial layers. The diameter crystals in the experiments in the present study were placed in mid-wall position. Assuming a spherical and symmetrical left ventricle with a wall thickness of 12 mm. before volume expansion, it can be calculated from the changes in left ventricular end-diastolic diameter (Table I) that the inner endocardial diameter increased by 51 per cent during blood transfusion, whereas the outer epicardial surface increased only by 8 per cent. The actual increase in the diameter of subendocardial layers could have been less than expected because of the presence of wrinkling and trabeculations. However, it is most likely that the subendocardial layers will be more stretched and thinned than the subepicardial layers of the left ventricular wall during volume expansion. The absence of a significant difference in the changes of endocardial and epicardial potentials, when recorded from the right ventricular free wall, can also be explained by the fact that the right ventricular free wall is much thinner. Therefore, a much smaller difference in the increased diameter of endocardial and epicardial layers of right ventricular free wall would be expected during volume expansion. Studies of body surface potentials and simulator experiments have suggested that the increase in intracavitary blood volume, a highly conductive mass, augments the radial components and decreasesthe tangential components of ventricular excitation potentials.“. 7. lo. I1 A radial propagation of ventricular excitation is dominant in endocardial layers, whereas the epicardial propagation is much more tangentially oriented.+, A If such mechanisms was operative in the present study, divergent changes between endo and epicardial potentials would be expected during volume expansion by blood transfusion. Such discordant changes, however, were not detected in the present study, as both endo and epicardial potentials decreased during an increase in ventricular diameter. These findings imply that potentials recorded directly from endo and epicardial surfaces cannot be compared to spatial surface vectorcardiograms. Our findings are in agreement with the observations of Angelakos and Gokhan,” who also reported that changes in epicardial and body surface potentials were different during vena caval obstruction, which presumably caused an alteration in ventricular volume. Surface potentials reflect the total cardiac electri-
August,
1979,
Vol.
98, No.
2
Endocardial
cal activity as a result of vastly complex propagation patterns of potentials throughout the body,13 whereas a unipolar electrode placed directly in cardiac tissue most likely will reflect predominantly the electrical events in the surrounding myocardial tissue. Changes in hematocrit have been shown to influence surface potentials.“, lZ Marked reduction in hematocrit to below 35% might reduce endocardial potential3 However, only minor changes in hematocrit (38 to 44 per cent) occurred during transfusion of whole blood in these experiments, and reduction in endocardial and epicardial potentials were clearly seen even in those experiments where the hematocrit increased slightly during blood transfusion. Hemolysis was observed in some of the experiments, and it might be argued that increased plasma concentrations of potassium from hemolysis might have increased conductance, accounting for changes in potentials. This is, however, unlikely, because following withdrawal of transfused blood, control values of cardiac potentials were almost always regained as the left ventricular end-diastolic diameter also returned to control. It has recently been suggested that reduction of the R wave amplitude of the epicardial QRS complex may indicate enhancement of myocardial ischemic injury following coronary artery ligation.16 Such a decrease in R wave amplitude, however, may be related to increase in left ventricular volume which is consistently observed during experimental myocardial infarction. The development of Q waves and reduction of R wave magnitude after infarction might, however, also be influenced by the relationship described here between epicardial potentials and left ventricular dimensions. Ventricular dilation is a consequence of acute myocardial ischemia,” and it is possible that this is reflected in reduced QRS voltage in epicardial leads from all parts of the left ventricle. The same argument also applies to the interpretation of changes in QRS voltage to assessthe effect on the ischemic injury of interventions that simultaneously alter left ventricular dimensions. It should be emphasized, however, that our measurements of the relationship between cardiac potentials and ventricular dimensions were performed in acute experiments and that the long-term relationship remains unknown. Although in the present study the absolute magnitude of potentials, endocardial and epicar-
American
Heart
Journal
and epicardial
potentials
dial, varied considerably according to the electrode sites, a strong negative correlation existed between changes in left ventricular diameter and endo or epicardial potentials recorded from each site, whether from the right or the left ventricle. These findings strongly suggest that monitoring of right ventricular endocardial potentials may be useful clinically to detect acute change in ventricular volume. Summary
Unipolar potentials were recorded from the endocardium (Endo-Pot) and the epicardium (Epi-Pot) of the left and right ventricles of anesthetized open-chested dogs during acute changes in left ventricular dimension by blood transfusion. A pair of implanted ultrasonic crystals were used to detail changes in left ventricular (LV) anteroposterior diameter. When the diameter increased by an average of 11 per cent, LV Endo-Pot decreased by 28 per cent and LV Epi-Pot decreased by 15 per cent. Right ventricular Endo-Pot and Epi-Pot concurrently decreased by similar magnitude (-36 per cent). The relationship between potentials and LV diameter showed negative linearity over the ranges examined, and was not influenced by changes in hematocrit. No inverse relation between changes in Endo-Pot and Epi-Pot was observed. It is suggested that potentials when recorded directly from the endocardium or epicardium mainly reflect the electrical activity of the tissues in the immediate vicinity of the electrode. It is postulated that an increase in ventricular volume by producing stretching and thinning of ventricular walls, reduces the effective tissue mass represented in the electrode signal, thereby accounting for a reduction in both endo and epicardial potentials. Although the precise mechanism of changes in ventricular potentials remains unclear, such changes, nevertheless, may indicate, in clinical circumstances, an acute shift in left ventricular volume. REFERENCES
1.
2.
3.
Chatterjee, K., Davies, G., Harris, A., et al.: Fall of endocardial potentials after acute myocardial infarction, Lancet 1:1308, 1970. Chatterjee, K., Sutton, G. C., and Miller, G. A. H.: Right ventricular endocardial potential in acute massive pulmonary embolism, Br. Heart J. 34:271, 1972. Lekven, J., Chatterjee, K., Tyberg, J. V., et al.: Pronounced dependence of ventricular endocardial QRS potentials on ventricular volume, Br. Heart J. 40:891, 1978.
205
Lekven et al. 4.
5.
6.
7.
8.
9. 10.
11.
Sher, A. M., and Young, A. C.: Ventricular depolarization and the genesis of QRS, Ann. N. Y. Acad. Sci. 66:768, 1956. Spach, M. S., and Barr, R. C.: Ventricular intramural and epicardial potential distributions during ventricular activation and repolarization in the intact dog, Circ. Res. 37:243, 1975. Brody, D. A.: A theoretical analysis of intracavitary blood mass on the heart-lead relationship, Circ. Res. 4:731, 1956. Bayley, R. H., and Berry, P. M.: “Body surface” potentials produced by the eccentric dipole in the heart wall of the nonhomogeneous volume conductor, AM. HEART J. 66:200, 1963. Theroux, P., Franklin, D., Ross, J., Jr., et al.: Regional myocardial function during coronary artery occlusion and its modification by pharmacological agents in the dog, Circ. Res. 35896, 1974. Zar, J. H.: Biostatistical analysis, Englewood Cliffs, New Jersey, 1974, Prentice-Hall, Inc. Voukydis, P. C., Angelakos, E. T., and Nelson, C. V.: Electrical effects of a highly conductive mass inside the thorax, AM. HEART J. 86:382, 1973. Sam, Y., and Marcus, M. L.: Computer simulation of the precordial QRS complex: Effects of simulated changes in
Information
12.
13.
14.
15.
16.
17.
ventricular wall thickness and volume, AM. HEART .I. 92:758, 1976. Angelakos, E. T., and Gokhan, N.: Influence of venous inflow volume on the magnitude of the QRS potentials in viva, Cardiologia 42:337, 1963. Abildskov, J. A., Burgess, M. J., Lux, R. L., et al.: Experimental evidence for regional cardiac influence in body surface isopotential maps of dogs, Circ. Res. 38:386, 1976. Rosenthal, A., Restieaux, N. J., and Feig, S. A.: Influence of acute variations in hematocrit on the QRS complex of the Frank electrocardiogram, Circulation 44:456, 1971. Hodkin, B. C., Millard, R. W., and Nelson, C. V.: Effect of hematocrit on electrocardiographic potentials and dipole moment of the pig, Am. J. Physiol. 232:H406, 1977. Hillis, L. D., Askenazi, J., Braunwald, E., et al.: Use of changes in the epicardial QRS complex to assess interventions which modify the extent of myocardial necrosis following coronary artery occlusion, Circulation 54:591, 1976. Lekven, J., Mjes, 0. D., and Kjeksus, J. K.: Compensatory mechanisms during graded myocardial ischemia, Am. J. Cardiol. 31:467. 1973.
for authors
Most of the provisions of the Copyright Act of 1976 became effective on January 1, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: “The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the work is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors.” Authors will be consulted, when possible, regarding republication of their material.
206
August,
1979,
Vol.
98, No.
2
Jon Lekven, M.D.* Kanu Chatterjee, M.B., M.R.C.P. (Lond. John V. Tyberg, M.D., Ph.D., F.A.C.C.** William W. Parmley, M.D., F.A.C.C.
and Edin.),
F.A.C.C.
San Francisco, Calif.
The observation that a reduction in right ventricular endocardial potentials occur in patients with acute myocardial infarction or massive pulmonary embolism, clinical situations likely to be associated with acute ventricular dilatation, indicates that a relationship might exist between the magnitude of ventricular potentials and ventricular volumes.‘, ? Recently it was demonstrated in an experimental study that potentials recorded from the endocardial surface of the left and right ventricles decreased markedly when left ventricular volume was increased by blood transfusion.” Preliminary data suggested that potentials recorded from the epicardial surface also decreased during volume expansion. However, the influence of changes in ventricular volumes on endo and epicardial potentials, recorded simultaneously from the same site, has not been analyzed.
The ventricular excitation propagates from the endocardial side with predominantly radial components, whereas the propagation is more tangentially oriented in outer epicardial layers.“, ‘, Analysis of spatial surface electrocardiograms have indicated opposite effects of changes in the intracavitary blood mass and volume on radial and tangential components of the QRS complex.“, i Our preliminary observations, however, indicated directionally similar changes in endocardial and epicardial potentials during acute changes in ventricular volumes. Thus, the relationship between changes in endocardial and epicardial potentials during acute changes in ventricular volume is not totally clear. The present study was, therefore, undertaken to clarify the relationship between epicardial and endocardial potentials during acute variations in left ventricular diameter and, therefore, left ventricular volume. Methods
From the Cardiovascular Division of the Department the Cardiovascular Research Institute, University Francisco, Calif. This work was supported Institute Program Project
in part Grant
by National HL 06285.
Received
for publication
June
12, 1978.
Accepted
for publication
July
17, 1978.
Reprint requests, Room 1186, Moffitt Calif. 94143.
Dr. Kanu Hospital,
Chatterjee, University
of Medicine of California,
Heart,
Lung
Cardiovascular of California, San
*Dr. Lekven is the recipient of a United States Public Health Fogarty International Fellowship (FO 5 TW 2371). His present is: Cardiovascular Research Laboratory, University of Bergen, land Hospital Surgical Department, N-5016 Bergen, Norway. **Dr. Tyberg is the recipient of a United States Public Research Career Development Award (HL ooO16) and from the American Heart Association (AHA-76788).
200
August,
1979, Vol. 98, No. 2
and
and San Blood
Division, Francisco, Service address Hauke-
Health Service a Grant-in-Aid
Experimental preparation. Twelve mongrel dogs weighing 16 to 25 kilograms were anesthetized initially with sodium thiopental(25 mg./Kg. intravenously). Morphine sulfate (45 mg. intravenously) was then given and anesthesia was maintained throughout the experiments with regular injections of 15 mg./hr. For muscle relaxation, a continuous infusion of succinyl choline chloride (20 mg./Kg./hr.) was given. The dogs were intubated and ventilated with a positive pressure respirator (Harvard Apparatus, Millis, Mass.); respiration was adjusted under the guidance of frequent analyses of arterial blood pH, p0, and $0, (BMS MK2, Radiometer, Copenhagen).
0002-8703/79/080200
+ 07500.70/O
3 1979
The
C. V. Mosby
Co.
Endocardial The chest was opened by mid-sternotomy and the pericardium was opened and retracted. The right atrium was paced in nine of the 12 dogs with an electrode attached to the atria1 appendage (Grass Instruments, Quincy, Mass.). Electrical measurements. Potentials from the endocardial surface of the left and right ventricles (Endo-Pot) were recorded with thin wire hook electrodes with a diameter of 0.13 mm. (Elgiloy, Elgiloy Corp., Elgin, Ill.) on an electrocardiograph (Honeywell, San Jose, Ca.). The entire length of the wire, except the distal 4 mm. was insulated. The electrodes were mounted in 23 gauge needles and were introduced into the ventricular cavities; upon withdrawal of the needle, the electrode hook sank approximately 2 mm. into the endocardial tissue layer. Initially, a contact pattern of ST segment elevation was regularly observed, but after one hour the ST segment elevation was markedly reduced. Electrodes showing persistent ST segment elevation greater than 2 mV. after one hour were discarded from subsequent analysis. Potentials from the epicardial surface (EpiPot) were recorded with a cotton wick electrode with an area of 5 mm.’ The epicardial sites were defined according to anatomical landmarks, so that they corresponded to the underlying endocardial wire electrodes. A total of 32 pairs of endocardial and epicardial electrodes were used for recording left ventricular endo and epicardial potentials, respectively. In five dogs one electrode was placed through the right ventricular cavity into an endocardial position on the left side of the interventricular septum. In four dogs, right ventricular endocardial potentials were also recorded from the free wall (endocardial surface) of the right ventricle. In three of the dogs, wire electrodes were also inserted in an epicardial position, and the Epi-Pot recorded with wire electrodes could be compared with the potentials recorded with wick electrodes at the same epicardial sites. The correlation of 158 paired observations at eight different sites with the two methods was: Wick-Pot = 0.93 Wire-Pot + 0.47 The regression coefficient was r = 0.972, and the regression line was not statistically different from the line of identity. The Wilson central terminal of extremity leads was used as reference for all electrograms. Signals below 0.5 Hz and above 100 Hz were excluded by
American Heart Journal
and epicardial potentials
bandpass filters. Potentials were accepted for analysis provided correct endocardial electrode positions were verified by postmortem examination of the heart, and only the potentials recorded from such satisfactory electrode positioning were analyzed. The amplitude of QRS complexes from four consecutive beats were averaged and regarded as endo or epicardial potentials. Ectopic beats or QRS complexes revealing conduction disturbances were excluded. The potentials were integrated over time by analog computer (Electronics Assoc., Inc., West Long Branch, New Jersey), giving the area of the QRS complex (mV. * msec.) above the isoelectric T-P segment for each beat. The integrator was triggered and reset by the left ventricular pressure signals and appropriate delay controls. Hemodynamic measurements. Left ventricular diameter was continuously recorded by the ultrasound time between two piezoelectric crystals inserted into the anterior and posterior walls of the left ventricle.” Left ventricular pressure was measured with a solid state transducer (Konigsberg Instruments, Pasadena, Ca.) inserted from the left atria1 appendage. Aortic pressure was measured through a femoral catheter connected to a Statham P23Db transducer, and aortic flow was measured with an electromagnetic flowmeter (Carolina Medical Electronics, King, No. Carolina) on the ascending aorta. Experimental procedure. Control measurements of potentials and hemodynamic parameters were performed when the initial electrocardiogram contact pattern had disappeared. Homologous and prewarmed blood was then infused into the jugular vein through a wide-bore catheter for stepwise increase in left ventricular enddiastolic diameter (LVEDD) and pressure (LVEDP). For each step, the infusion rate was slowed down to maintain stable hemodynamics, and measurement of potentials and hemodynamic parameters were repeated. Infusion was continued until LVEDP had reached 15 to 25 mm. Hg; infused blood volume was 550 to 2,500 ml. Thereafter, blood was withdrawn stepwise to allow LVEDD and LVEDP to return to control. The above procedure allowed us to treat the recorded potentials as a regression problem for each electrode separately with respect to the changes in LVEDD during infusion and withdrawal of blood. Table I gives the control values and the values obtained at the highest LVEDD
201
et
Lekven
al.
Highest Load
Recovery
LV END0 r
RV END0
LVEDD
--r-
---?Fc
--f--
-
mm
---y--
-
----.-
-,,----
-
-
Fig. 1. Potentials recorded from the left (LV) and right (RV) ventricles with the interventricular stepwise infusion and withdrawal of blood. Potentials were recorded at corresponding endocardial epicardial @PI) sites of the ventricular walls. LVEDD = left ventricular end-diastolic diameter.
achieved, where each dog and electrode served as its own control. Student’s t test for paired data was applied in calculating statistical probability.” A p value less than 0.05 was regarded as statistically significant. As similar relationships between Endo-Pot and Epi-Pot and LVEDD were observed in paced and non-paced hearts, the data were, therefore, pooled for final analysis. Results Fig. 1 shows changes in potentials recorded from the left and right ventricular endo and epicardium and from the inter-ventricular septum in one experiment. The potentials from both endo and epicardial,surfaces decreased as LVEDD was increased stepwise during blood transfusion. When LVEDD returned to control after withdrawal of blood, the potentials increased and also returned to control values. The absolute magnitude of the potentials recorded either from endocardium or epicardium of the left or right ventricle varied considerably according to electrode site and also position in different dogs, ranging from 17 to 51 mV. (see Fig. 2). However, irrespective of the recording site and the initial magnitude,
202
septum during (ENDO) and
linear reduction in potential was observed for each individual electrode as LVEDD was increased. Both endocardial and epicardial potentials recorded from either ventricle and also potentials from the interventricular septum behaved similarly. Table I summarizes changes in potentials and hemodynamic parameters during maximum volume load achieved. On the average, LVEDD increased by 11 per cent, from 39.41 -t 1.72 to 43.60 f 1.89 mm. Similarly, LVEDP increased from 4.9 + 1.0 to 17.3 + 2.1 mm. Hg. Blood transfusion caused an increase, as expected, in cardiac output and the systolic blood pressures. To assessthe magnitude of changes in endo and epicardial potentials recorded from the same area of the ventricular wall, the ratio of Endo/Epi potentials was calculated for each pair of endocardial and epicardial electrodes separately. At control LVEDD of 100 per cent, the ratio of Endo/Epi potentials recorded from the left ventricle averaged 1.04 + 0.04, and at LVEDD of 111 per cent this ratio decreased to 0.72 & 0.05, suggesting a relatively greater reduction in left ventricular endocardial potential than in its counterpart, epicardial potential. Statistical
August,
1979, Vol. 98, No. 2
Endocardial
and epicardial potentials Endo RV
mV 20 i 1
40
mV 20
i, , , , 100 108
, Ept
1
0L-LL-L 100
108
RV
-\
IF?
1 100 P
I16
LVEDD, Fig. 2. Relationship between each electrode. Abbreviations
% of
control
cardiac potentials and left ventricular = the same as those in Fig, 1
diameter
analysis of the individual regression lines of the ratios of left ventricular Endo/Epi potentials revealed that the ratio was significantly reduced by volume loading (Slope = 0.016 -+ 0.002, p < 0.001). Such changes, however, were not observed in the ratios of right ventricular endo and epicardial potentials. Endo-pot recorded from the left ventricular septum behaved similarly to Endo-Pot recorded from the endocardium of the left ventricular free wall (Table I). The area of the QRS complexes, similar to its absolute magnitude, showed a linear reduction with increasing diameter of the left ventricle. Duration of the QRS complex did not change, suggesting lack of conduction disturbances (average 42.9 -+ 0.3 msec.). An apparently greater reduction in epicardial QRS area compared to the reduction in Epi-Pot could be accounted for by the fact that the integrator substracted small negative areas below the isoelectric base line (Fig. 1). Discussion
The present study was designated to investigate the influence of acute changes in ventricular
American Heart Journal
, II6
during
blood
infusion.
One line for
volume on concurrently recorded right and left ventricular endo and epicardial potentials. The findings indicate that acute changes in left ventricular diameters markedly influence not only the magnitude of the right or left ventricular endocardial potentials, as previously reported,” but also of their epicardial potentials. Thus, an increase in left ventricular diameter during blood transfusion was accompanied by a decrease in both endo and epicardial potentials recorded either from the right or the left ventricle. With an 11 per cent average increase in left ventricular end-diastolic diameter, there was a 15 per cent decrease in left ventricular epicardial and 37 per cent decrease in right ventricular epicardial potentials. As observed in previous investigations, both left ventricular endocardial (-27 per cent) and right ventricular endocardial (-36 per cent) potentials also decreased concurrently. The precise mechanism of such changes in ventricular potentials with changes in ventricular diameter is not clear. That changes in ventricular geometry and wall thickness may contribute needs to be considered. A considerable decrease in thickening of ventricular walls may occur during acute expansion of ventricular volume. During
203
Lekven
et al.
I. Effect of blood infusion on cardiac potentials (Mean t SEM in twelve dogs) Table
N Left ventricle: Endo-Pot, mV. Area, mV. * msec. Epi-Pot, mV. Area, mV. * msec.
Hg
load
P
33.1 k 1.8 622 F 43
23.9 zk 1.6 508 i 42
0.001 0.001
33 33
34.7 -t 1.3 281 + 35
29.6 f 1.2 219 k 32
0.001 0.01
31.8 f 4.4 568 k 124
23.0 i 3.4 431 I 99
0.002 0.05
4 4
22.8 k 2.1 264 + 81
14.5 + 0.9 205 k 61
0.01
4 4
27.0 + 3.5 222 t 28
17.0 -+ 3.2 106 + 34
0.01
Hemodynamic parameters: LVEDD, mm. 12 39.41 -+ 1.72 LVEDP, mm. 12 4.9 5 1.0 Hg LVSP, mm. HR, beats/ min.f CO, ml./min.
Highest
33 33
Interventricular septum: Sept-Pot, mV. 5 Area, 5 mV * msec. Right ventricle: Endo-Pot, mV. Area. mV. * msec. Epi-Pot, mV. Area, mV. * msec.
Control
NS
0.05
43.60 k 1.89 17.3 f 2.1
0.001 0.001
12 12
129 f 6 164 f 6
155 k 7 159k7
0.001 NS
5
3730 Ifr 120
5420 + 800
0.05
Endo-Pot
= endocardial potential; Epi-Pot = epicardial potential: Sept-Pot = interventricular aeptal potential; Area = integrated area of QRS complex above the isoelectric T-P segment; LVEDD = left ventricular end-diastolic diameter; LVEDP = left ventricular enddiastolic pressure; LVSP = left ventricular systolic pressure; HR = heart rate; CO = cardiac output; N = number of observations; P = probability value; NS = not significant. tAtria1 pacing in nine dogs.
such thinning of the ventricular wall, the tissue mass in the immediate vicinity of the recording electrode that delivers the electrical signals is reduced.3 This might account for reduction of both endo and epicardial potentials during volume expansion. In the present study decrease in left ventricular endocardial potentials was of greater magnitude than that of corresponding epicardial potentials. That such a difference in changes in left ventricular endo and epicardial potentials was observed also supports the hypothesis that thinning of the left ventricular free wall may be at least partly responsible for reduction in ventricular potentials during an increase in ventricular volume. During volume expansion subendocardial layers will be expected to be
204
stretched more than the subepicardial layers. The diameter crystals in the experiments in the present study were placed in mid-wall position. Assuming a spherical and symmetrical left ventricle with a wall thickness of 12 mm. before volume expansion, it can be calculated from the changes in left ventricular end-diastolic diameter (Table I) that the inner endocardial diameter increased by 51 per cent during blood transfusion, whereas the outer epicardial surface increased only by 8 per cent. The actual increase in the diameter of subendocardial layers could have been less than expected because of the presence of wrinkling and trabeculations. However, it is most likely that the subendocardial layers will be more stretched and thinned than the subepicardial layers of the left ventricular wall during volume expansion. The absence of a significant difference in the changes of endocardial and epicardial potentials, when recorded from the right ventricular free wall, can also be explained by the fact that the right ventricular free wall is much thinner. Therefore, a much smaller difference in the increased diameter of endocardial and epicardial layers of right ventricular free wall would be expected during volume expansion. Studies of body surface potentials and simulator experiments have suggested that the increase in intracavitary blood volume, a highly conductive mass, augments the radial components and decreasesthe tangential components of ventricular excitation potentials.“. 7. lo. I1 A radial propagation of ventricular excitation is dominant in endocardial layers, whereas the epicardial propagation is much more tangentially oriented.+, A If such mechanisms was operative in the present study, divergent changes between endo and epicardial potentials would be expected during volume expansion by blood transfusion. Such discordant changes, however, were not detected in the present study, as both endo and epicardial potentials decreased during an increase in ventricular diameter. These findings imply that potentials recorded directly from endo and epicardial surfaces cannot be compared to spatial surface vectorcardiograms. Our findings are in agreement with the observations of Angelakos and Gokhan,” who also reported that changes in epicardial and body surface potentials were different during vena caval obstruction, which presumably caused an alteration in ventricular volume. Surface potentials reflect the total cardiac electri-
August,
1979,
Vol.
98, No.
2
Endocardial
cal activity as a result of vastly complex propagation patterns of potentials throughout the body,13 whereas a unipolar electrode placed directly in cardiac tissue most likely will reflect predominantly the electrical events in the surrounding myocardial tissue. Changes in hematocrit have been shown to influence surface potentials.“, lZ Marked reduction in hematocrit to below 35% might reduce endocardial potential3 However, only minor changes in hematocrit (38 to 44 per cent) occurred during transfusion of whole blood in these experiments, and reduction in endocardial and epicardial potentials were clearly seen even in those experiments where the hematocrit increased slightly during blood transfusion. Hemolysis was observed in some of the experiments, and it might be argued that increased plasma concentrations of potassium from hemolysis might have increased conductance, accounting for changes in potentials. This is, however, unlikely, because following withdrawal of transfused blood, control values of cardiac potentials were almost always regained as the left ventricular end-diastolic diameter also returned to control. It has recently been suggested that reduction of the R wave amplitude of the epicardial QRS complex may indicate enhancement of myocardial ischemic injury following coronary artery ligation.16 Such a decrease in R wave amplitude, however, may be related to increase in left ventricular volume which is consistently observed during experimental myocardial infarction. The development of Q waves and reduction of R wave magnitude after infarction might, however, also be influenced by the relationship described here between epicardial potentials and left ventricular dimensions. Ventricular dilation is a consequence of acute myocardial ischemia,” and it is possible that this is reflected in reduced QRS voltage in epicardial leads from all parts of the left ventricle. The same argument also applies to the interpretation of changes in QRS voltage to assessthe effect on the ischemic injury of interventions that simultaneously alter left ventricular dimensions. It should be emphasized, however, that our measurements of the relationship between cardiac potentials and ventricular dimensions were performed in acute experiments and that the long-term relationship remains unknown. Although in the present study the absolute magnitude of potentials, endocardial and epicar-
American
Heart
Journal
and epicardial
potentials
dial, varied considerably according to the electrode sites, a strong negative correlation existed between changes in left ventricular diameter and endo or epicardial potentials recorded from each site, whether from the right or the left ventricle. These findings strongly suggest that monitoring of right ventricular endocardial potentials may be useful clinically to detect acute change in ventricular volume. Summary
Unipolar potentials were recorded from the endocardium (Endo-Pot) and the epicardium (Epi-Pot) of the left and right ventricles of anesthetized open-chested dogs during acute changes in left ventricular dimension by blood transfusion. A pair of implanted ultrasonic crystals were used to detail changes in left ventricular (LV) anteroposterior diameter. When the diameter increased by an average of 11 per cent, LV Endo-Pot decreased by 28 per cent and LV Epi-Pot decreased by 15 per cent. Right ventricular Endo-Pot and Epi-Pot concurrently decreased by similar magnitude (-36 per cent). The relationship between potentials and LV diameter showed negative linearity over the ranges examined, and was not influenced by changes in hematocrit. No inverse relation between changes in Endo-Pot and Epi-Pot was observed. It is suggested that potentials when recorded directly from the endocardium or epicardium mainly reflect the electrical activity of the tissues in the immediate vicinity of the electrode. It is postulated that an increase in ventricular volume by producing stretching and thinning of ventricular walls, reduces the effective tissue mass represented in the electrode signal, thereby accounting for a reduction in both endo and epicardial potentials. Although the precise mechanism of changes in ventricular potentials remains unclear, such changes, nevertheless, may indicate, in clinical circumstances, an acute shift in left ventricular volume. REFERENCES
1.
2.
3.
Chatterjee, K., Davies, G., Harris, A., et al.: Fall of endocardial potentials after acute myocardial infarction, Lancet 1:1308, 1970. Chatterjee, K., Sutton, G. C., and Miller, G. A. H.: Right ventricular endocardial potential in acute massive pulmonary embolism, Br. Heart J. 34:271, 1972. Lekven, J., Chatterjee, K., Tyberg, J. V., et al.: Pronounced dependence of ventricular endocardial QRS potentials on ventricular volume, Br. Heart J. 40:891, 1978.
205
Lekven et al. 4.
5.
6.
7.
8.
9. 10.
11.
Sher, A. M., and Young, A. C.: Ventricular depolarization and the genesis of QRS, Ann. N. Y. Acad. Sci. 66:768, 1956. Spach, M. S., and Barr, R. C.: Ventricular intramural and epicardial potential distributions during ventricular activation and repolarization in the intact dog, Circ. Res. 37:243, 1975. Brody, D. A.: A theoretical analysis of intracavitary blood mass on the heart-lead relationship, Circ. Res. 4:731, 1956. Bayley, R. H., and Berry, P. M.: “Body surface” potentials produced by the eccentric dipole in the heart wall of the nonhomogeneous volume conductor, AM. HEART J. 66:200, 1963. Theroux, P., Franklin, D., Ross, J., Jr., et al.: Regional myocardial function during coronary artery occlusion and its modification by pharmacological agents in the dog, Circ. Res. 35896, 1974. Zar, J. H.: Biostatistical analysis, Englewood Cliffs, New Jersey, 1974, Prentice-Hall, Inc. Voukydis, P. C., Angelakos, E. T., and Nelson, C. V.: Electrical effects of a highly conductive mass inside the thorax, AM. HEART J. 86:382, 1973. Sam, Y., and Marcus, M. L.: Computer simulation of the precordial QRS complex: Effects of simulated changes in
Information
12.
13.
14.
15.
16.
17.
ventricular wall thickness and volume, AM. HEART .I. 92:758, 1976. Angelakos, E. T., and Gokhan, N.: Influence of venous inflow volume on the magnitude of the QRS potentials in viva, Cardiologia 42:337, 1963. Abildskov, J. A., Burgess, M. J., Lux, R. L., et al.: Experimental evidence for regional cardiac influence in body surface isopotential maps of dogs, Circ. Res. 38:386, 1976. Rosenthal, A., Restieaux, N. J., and Feig, S. A.: Influence of acute variations in hematocrit on the QRS complex of the Frank electrocardiogram, Circulation 44:456, 1971. Hodkin, B. C., Millard, R. W., and Nelson, C. V.: Effect of hematocrit on electrocardiographic potentials and dipole moment of the pig, Am. J. Physiol. 232:H406, 1977. Hillis, L. D., Askenazi, J., Braunwald, E., et al.: Use of changes in the epicardial QRS complex to assess interventions which modify the extent of myocardial necrosis following coronary artery occlusion, Circulation 54:591, 1976. Lekven, J., Mjes, 0. D., and Kjeksus, J. K.: Compensatory mechanisms during graded myocardial ischemia, Am. J. Cardiol. 31:467. 1973.
for authors
Most of the provisions of the Copyright Act of 1976 became effective on January 1, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: “The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the work is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors.” Authors will be consulted, when possible, regarding republication of their material.
206
August,
1979,
Vol.
98, No.
2