Isolated Systolic and Diastolic Ventricular Interactions in Pacing-Induced Dilated Cardiomyopathy and Effects of Volume Loading and Pericardium
Background Interactions between the closely coupled right and left ventricles are known to play important roles as determinants of ventricular function, and the purpose of this study was to evaluate their effects in a model of heart failure.
Methods and Results A dilated cardiomyopathy resulting in congestive heart failure (CHF) was produced in pigs by rapid ventricular pacing at 230 beats per minute for 1 week. Blood was rapidly withdrawn from the left ventricular (LV) apex into a prosthetic ventricle, and the instantaneous effects on the right ventricle were studied during volume loading and before and after pericardiectomy. The systolic interaction gain between the right and left ventricles (Gs) was calculated as the ratio of changes in mean systolic pressure during isolated systolic LV unloading. Diastolic ventricular interaction gain (Gd) was calculated as the ratio of changes in mean diastolic pressures during LV unloading in the last 150 ms of diastole. With the pericardium closed, all interaction gains were significantly increased during volume loading from a right ventricular end-diastolic pressure of 3 to 9 mm Hg: Gs from 0.045±0.014 to 0.063±0.020 mm Hg/mm Hg (normal pigs) and from 0.077±0.040 to 0.103±0.019 (CHF pigs) and Gd from 0.196±0.116 to 0.493±0.117 mm Hg/mm Hg (normal pigs) and from 0.174±0.101 to 0.341±0.165 (CHF pigs). When the pericardium was opened, Gd was significantly reduced to 0.145±0.071 and 0.129±0.026 mm Hg/mm Hg (normal and CHF pigs, respectively), but Gs showed no significant change (0.067±0.027 and 0.109±0.012 mm Hg/mm Hg for normal and CHF pigs, respectively), and both were also significantly increased during volume loading. Gs was significantly greater in CHF versus normal pigs under all conditions, but there were no differences in Gd between CHF and normal pigs.
Conclusions These results suggest that dilated cardiomyopathy increases systolic but not diastolic interactions, that the pericardium increases diastolic but not systolic ventricular interactions, and that volume loading with and without the pericardium opened increases both systolic and diastolic interactions.
Systolic and diastolic interactions between the closely coupled right and left ventricles are known to play important roles as determinants of ventricular function. Ventricular interactions during passive diastole are especially important in the presence of an intact pericardium, which increases the coupling between the ventricles and enhances the effect that the volume of one ventricle has on the diastolic filling of the contralateral ventricle.1 2 3 4 5 6 7 8 9 10 11 There is also evidence in the left12 and right13 ventricles that anatomic ventricular interactions have effects on diastolic relaxation. Diastolic interactions have been shown to be less important in conditions where the interventricular septum is less compliant, as reported during right ventricular (RV) pressure (RVP)–overload hypertrophy14 and left ventricular (LV) hypertrophy from renovascular hypertension.15
During systole, ventricular interactions are responsible for the transmission of systolic forces between the ventricles, which can be very important to the low-pressure right ventricle that receives part of its pumping force from the left ventricle.16 17 18 19 We previously showed increased systolic ventricular interactions during pacing-induced dilated cardiomyopathy, which suggests that direct anatomic systolic ventricular interactions become more important when the systolic elastance of the ventricular septum is reduced (more compliant).19 However, the effects of this model on diastolic ventricular interactions have not been evaluated. In addition, the effects of the pericardium are not as well understood for systolic ventricular interactions as for diastolic interactions. Some studies indicated that there is little effect of the pericardium on systolic ventricular interactions,4 12 20 21 whereas other reports concluded that ventricular interaction in the intact heart is responsible for an upward shift in the relations between LV pressure (LVP) and area22 and between pressure and segment length23 after pericardiectomy or reported enhanced systolic pressure interaction during isovolumic beats when the pericardium was sutured closed.24 The purpose of the present study was to directly determine LV-to-RV isolated systolic and diastolic ventricular interactions in pigs with pacing-induced dilated cardiomyopathy. This was accomplished by determining the instantaneous response of the right ventricle to transient changes in LVP with the pericardium closed and opened.
Surgical and instrumentation protocols were described previously for determining isolated systolic ventricular interaction gains.18 19 Briefly, blood was rapidly withdrawn from the left ventricle into a prosthetic ventricle immediately after the r wave of the ECG, and the instantaneous resultant effects on systolic RVP were determined. Diastolic gains were determined by use of similar techniques but with LV unloading isolated to the last 150 ms of diastole, and the resultant instantaneous effects on diastolic RVP were determined. Seventeen new experiments were performed in 40- to 45-kg farm pigs: 10 normal pigs and 7 pigs in which a dilated cardiomyopathy resulting in congestive heart failure (CHF) was produced by rapid ventricular pacing. Methodology for the CHF model was also presented previously.19 25 26 These protocols are briefly reviewed here.
After anesthesia with ketamine (20 mg/kg IM for preinduction) and intravenous injections of thiamylal sodium (4.5 mg/kg initial bolus and 2.5 mg/kg maintenance doses every 15 to 20 minutes), the 7 pigs from this group were intubated and connected to an artificial ventilator. Respiratory rate, tidal volume, and Fio2 were adjusted as necessary to maintain acid-base equilibrium. A Medtronic unipolar sutureless pacing lead was attached to the LV apex through a small opening in the diaphragm under the xiphoid process and connected to a Medtronic pacemaker (modified model SX-5985) implanted in a subcutaneous abdominal pocket. After recovery from anesthesia, the pigs were returned to a chronic care facility, where they received a standard diet, free access to water, and an antibiotic regimen of Penicillin-G. The pacemaker was then programmed at a rate of 230 beats per minute. The animals were returned to the laboratory for the short-term experiment 7 days later. Before induction of general anesthesia, the pacemaker was turned off. The chest was opened, and the heart was instrumented as described below.
After anesthesia with ketamine (20 mg/kg IM for preinduction) and intravenous administration of thiamylal sodium (4.5 mg/kg initial bolus and 80 μg · kg · −1 · min−1 maintenance infusion), each of the normal and CHF pigs was intubated and connected to a mechanical ventilator. Atropine (0.09 mg/kg SC) was given to control bronchial secretions, and muscle paralysis was induced as needed with d-tubocurarine (0.3 mg/kg IV). Respiratory rate, tidal volume, and Fio2 were adjusted as necessary to maintain acid-base equilibrium. The heart was exposed through a median sternotomy and connected to a Thoratec ventricular assist device (VAD) modified by removal of the usual mechanical inflow and outflow valves, and the outflow port was blocked off. A 2-cm incision was made in the pericardium at the apex, a 12-mm-ID wire-wrapped cannula was inserted into the LV apex through a stab incision, and the cannula was connected to the VAD inflow port. The VAD pneumatic control console was also modified to create single-cycle operation in which there is rapid filling of the VAD and therefore rapid unloading of the left ventricle on command in a single beat at different times as required during systole and diastole.
LVP and RVP were measured with high-fidelity microtipped catheters (Millar Instruments, Inc). These transducers were set to zero in blood at body temperature before insertion and checked at the end of the experiment. Aortic pressure was measured with a standard fluid-filled catheter positioned in the carotid artery and connected to a pressure transducer (Statham). With as little disruption to the pericardium as possible, the pulmonary artery was exposed for placement of an electromagnetic flow probe (Carolina Medical Electronics, Inc) to measure continuous beat-to-beat cardiac output (CO), and the small pericardial slit was closed with clips. All signals were continuously recorded on an eight-channel Gould chart recorder and simultaneously sampled on-line at 200 samples per second with an analog-to-digital convertor (Metrabyte) connected to a personal computer.
All data were collected after a 15- to 30-minute stabilization period after completion of instrumentation and LVAD implantation. The basic protocol consisted of measurements in 8-second-long data groups, with the respiration held at end expiration. In each group, six normal cardiac cycles were followed by one experimental beat, with rapid LV unloading achieved by rapidly reducing the pneumatic pressure of the VAD from 200 to −100 mm Hg, thus allowing the VAD to fill directly from the left ventricle. For isolated systolic interaction studies, LV unloading was achieved after the R wave of the seventh beat. For isolated diastolic studies, unloading began 150 ms before the R wave. The instantaneous changes produced by these perturbations on RVP were then evaluated. Data groups were repeated at least five times, after allowing for stabilization between groups.
After data were obtained under baseline conditions, volume loading was begun by infusing combinations of normal saline, low-molecular-weight dextran, or Hetastarch to increase RV end-diastolic pressure (RVEDP) from 3 to 6 mm Hg. Data collection was repeated for both systolic and diastolic unloading and again after volume loading to an RVEDP of 9 mm Hg. The pericardium was then completely opened, and all data measurements were repeated. Partial occlusion of the inferior vena cava was sometimes used to reduce the RVEDP to 3 mm Hg to compare with baseline conditions.
Data were analyzed with a data acquisition and analysis software package developed in our laboratory. Data sets with evidence of arrhythmias or cycle-to-cycle instability were discarded. For systolic studies, mean systolic LVP and RVP were calculated by integration of the LVP and RVP waveforms during systole. Mean systolic ventricular interaction gain (Gs) in millimeters of mercury per millimeter of mercury was defined as Gs=ΔRVMSP/ΔLVMSP, where ΔRVMSP is the change in mean systolic RVP and ΔLVMSP is the change in mean systolic LVP for each of the six normal beats compared with the one unloaded beat in each group. These six values were then averaged to provide one gain for each group. Diastolic ventricular interaction gain (Gd) was measured in the last 150 ms of diastole (through end diastole) asGd=ΔRVMDP/ΔLVMDP, where ΔRVMDP is the change in mean diastolic RVP and ΔLVMDP is the change in mean diastolic LVP in this 150-ms period during isolated diastolic unloading for each of the six normal beats compared to the one unloaded beat.
RV stroke volume was computed by integration of pulmonary artery flow for the whole cardiac cycle; RV stroke work, by the product of developed RVP times stroke volume. Zero pulmonary artery flow baseline was established as the average signal in the last 25% of the cycle before the beginning of systole.
In each group, the data from the six normal cycles (ie, not unloaded) were averaged to provide one value for each parameter. The data from all repeated groups during each experimental condition were subsequently averaged for each animal. Next, the data from all animals in each group were pooled and are presented as mean±SD. Statistical significance of the effects of CHF, the pericardium, and volume loading on interaction gains was determined by an ANOVA with one between-subject factor (CHF versus normal) and two within-subject factors (pericardium closed versus opened and volume loading with repeated measures for three levels of RVEDP). At highest levels of volume loading in these experiments (to an RVEDP of ≈9 mm Hg), statistical comparisons between the two groups of pigs were performed with a nonpaired t test and with a paired t test between control and unloaded beats and between pericardium closed and opened. A value of P<.05 was considered significant.
All animals received humane care in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the NIH (NIH Publication No. 80-23, revised 1978).
Fig 1⇓ shows examples of data recording illustrating the effects of diastolic LV unloading on diastolic RVP with the pericardium closed and opened from a typical normal pig. Pressure waveforms from the acutely unloaded beats are shown in dashed lines superimposed on the corresponding waveforms from the previous cardiac cycle (a control beat). With the pericardium closed in this example, the LVMDP during unloading was reduced by 3.9 mm Hg, which resulted in an instantaneous decrease in RVMDP of 1.3 mm Hg. Thus, the LV-to-RV diastolic interaction gain was Gd=1.3/3.9=0.33 mm Hg/mm Hg. With the pericardium opened, there was only a 0.1 mm Hg decrease in RVMDP resulting from a 5.9 mm Hg decrease in LVMDP; thus, Gd was reduced to 0.02 mm Hg/mm Hg.
Typical waveforms during isolated systolic LV unloading in a normal pig show that changes in systolic pressure were achieved without changes in end-diastolic conditions (Fig 2⇓). With the pericardium closed or opened, there were noticeable instantaneous changes in RVP of similar magnitude. Note that systolic LVP and RVP did not change until after the rapid upswing in LVP caused by pneumatic and mechanical delays of ≈60 ms. Once LVP began falling, there were corresponding instantaneous reductions in RVP. In this example, Gs=0.061 with the pericardium closed and Gs=0.067 with the pericardium opened.
Table 1⇓ gives the data pooled for all experiments under volume-loaded conditions to an RVEDP of 9 mm Hg. At these filling pressures, CO, stroke volume, and RV stroke work were significantly less in CHF compared with control pigs, but there were no differences in mean arterial pressure or heart rate. When the pericardium was opened, the diastolic ventricular interaction gains were reduced significantly from 0.493 to 0.145 mm Hg/mm Hg in normal pigs and from 0.341 to 0.129 in CHF pigs. These changes were direct results of reduced responses of ΔRVMDP during diastolic unloading: from −2.1 mm Hg (in normal pigs) in response to ΔLVMDP of −4.4 mm Hg with the pericardium closed to −0.4 mm Hg in response to ΔLVMDP of −3.0 mm Hg with the pericardium opened. A similar attenuation in ΔRVMDP during LV diastolic unloading was found in the CHF pigs when the pericardium was opened, and there were no significant differences between normal and CHF pigs.
Systolic interaction gains were significantly greater in the CHF than normal pigs with or without the pericardium opened (Table 1⇑). Opening the pericardium significantly increased RV stroke volume, RV stroke work, and both peak and mean systolic RVPs but did not affect ΔRVMSP or ΔLVMSP. During systolic unloading, the average experimental reduction in ΔLVMSP was −52 to −61 mm Hg with the pericardium closed or opened. The resultant instantaneous ΔRVMSP was −3.5 to −3.9 mm Hg for normal pigs and −5.3 to −5.8 mm Hg for CHF pigs with the pericardium closed and opened, respectively. The mean systolic LV-to-RV pressure gain, which is the ratio of these changes, did not significantly change after the pericardium was opened (from 0.063 to 0.067 mm Hg/mm Hg for normal pigs and from 0.103 to 0.109 mm Hg/mm Hg for CHF pigs, Table 1⇑).
The results of the ANOVA (Table 2⇓) demonstrate that CHF had a statistically significant effect on systolic interaction gains (P=.003) but no significant effect on diastolic interaction gains. The pericardium had a statistically significant effect on diastolic interaction gains (P<.001), whereas there were no statistically significant effects of the pericardium on systolic interaction gains. There were significant effects of volume loading found by ANOVA on both systolic (P<.001) and diastolic (P<.001) interaction gains. The interaction term (between pericardial effects and volume loading effects) was also significant (P=.015), indicating that the relations of diastolic ventricular interaction gains with volume loading (ie, RVEDP) are different with and without the pericardium opened. When the effects of volume loading and pericardium were examined separately in normal and CHF pigs, the same results were found: volume loading affected both systolic and diastolic interactions, but the pericardium had a significant effect only on diastolic ventricular interaction.
These findings are also graphically illustrated under all three volume load conditions in Figs 3⇓ and 4⇓. Fig 3⇓ shows the relation between diastolic ventricular interaction gain and RVEDP during volume loading for all animals for the pericardium closed and opened, and Fig 4⇓ shows a similar relation for systolic interaction gains. Over the range of preload studied, there was a marked reduction in diastolic interaction when the pericardium was opened, with considerable overlap between CHF and normal pigs (Fig 3⇓). On the other hand, opening the pericardium had no significant effect on systolic ventricular interaction gain (Fig 4⇓). However, systolic gains were noticeably higher for CHF pigs at each RVEDP, and they increased during volume loading with or without the pericardium opened.
The principal results from this study indicate that this model of dilated cardiomyopathy has a significant effect on isolated systolic (but not diastolic) ventricular interaction, that the pericardium has significant effects on diastolic (but not systolic) ventricular interaction, and that volume loading with or without the pericardium opened increases both systolic and diastolic interactions.
We previously showed that this model of dilated cardiomyopathy increases systolic ventricular interaction gains from 0.04 to 0.10 mm Hg/mm Hg, indicating that, relative to the RV free wall, the systolic elastance of the interventricular septum had been reduced, thus increasing the coupling between the ventricles during systole.19 The new experiments reported here indicate that these differences remain with or without the pericardium opened over wide ranges of volume loading. In the current study, the gains from the CHF animals were in the same range as we reported previously at the same filling pressures; however, the gains from the normal pigs appear to be slightly greater (an average of 0.067 mm Hg/mm Hg at an RVEDP of 9 mm Hg with the pericardium opened) but not statistically different from our previously reported values of 0.054 mm Hg/mm Hg at an RVEDP of 3.6 mm Hg18 and 0.040 at an RVEDP of 3.9 mm Hg19 obtained with the same technique. These slightly greater interaction gains are at least partially attributed to another finding of the current study, which is that systolic interaction gains increase with volume loading. Before volume loading with the pericardium open (see Fig 4⇑), the systolic gains in the normal pigs were 0.051±0.022 at an RVEDP of 4 mm Hg, which is consistent with our past reports, and increased about 31% to 0.067 after volume loading. In the CHF pigs, systolic interaction gains also increased with volume loading, but they were consistently higher at each level of RVEDP than in the normal pigs, again indicating that the systolic elastance of the ventricular septum had been reduced compared with normal pigs.
With this model of heart failure, we previously showed significant increases in septolateral LV and RV end-diastolic dimensions of approximately 15%, indicating biventricular dilatation, but with no significant change in wall thickness.25 This biventricular dilatation combined with low CO after 1 week of pacing provides a realistic model of dilated cardiomyopathy for studying ventricular function and interaction. Longer periods of pacing may produce even more severe heart failure with other chronic compensatory changes that would be of interest; however, the results most likely would be similar to those after 1 week but of larger magnitude.
Previous studies showed that systolic ventricular interaction gains during abrupt increases in ventricular pressure (0.12 mm Hg/mm Hg from Yaku et al21 and 0.086 mm Hg/mm Hg calculated from data published by Langille and Jones27 ) are substantially higher than during decreases in ventricular pressure (0.05 mm Hg/mm Hg in normal animals with our techniques from the present or past reports18 19 or 0.04 mm Hg/mm Hg from Yamaguchi et al17 ). As pointed out by Yaku et al,21 these higher systolic gains during abrupt increases in ventricular pressure may be due to higher LV volume and LVP, which increase the coupling between the ventricles because of increased septal tension. This is probably also the mechanism for the increased interaction that we found during volume loading, which concomitantly increased systolic ventricular pressure. However, Schertz and Pinsky24 reported data during increases in systolic LVP from which we calculate a much smaller systolic gain of 0.023 mm Hg/mm Hg with the pericardium opened, increasing to 0.076 mm Hg/mm Hg with a sutured pericardium.
Slinker and Glantz15 22 used a different approach in the intact heart. They separated the direct (anatomic) and indirect (in series through the pulmonary circulation) ventricular interactions using a statistical analysis of transient changes to pulmonary artery pressure and vena caval constrictions. They determined that there were significant effects of the pericardium at end diastole and end systole in normal hearts22 and that the pericardium has less modulating effects on direct ventricular interaction in hypertrophied hearts than in normal hearts.15 Kanazawa et al23 also reported a shift in the end-systolic area relation when the pericardium was removed after short-term volume loading, suggesting an effect of pericardial restraint on systolic ventricular interaction. On the other hand, Slinker et al12 reported no significant effect of the pericardium on RV-to-LV interaction gain, which was calculated to be 0.13 mm Hg/mm Hg when measured directly during sudden pulmonary constriction. In isolated heart experiments, effects of the pericardium on systolic ventricular interaction were reported but are quantitatively small and variable.4 20
Maughan et al28 presented a three-compartment volume elastance model that describes systolic interaction as consisting of the elastances of the RV free wall, interventricular septum, and LV free wall. If the septum is stiff compared with the free walls, then systolic interaction is reduced; if septal elastance is reduced, systolic interaction increases, such as during CHF in the present experiment and in our previous report.19 Santamore and Burkhoff16 incorporated the interaction model of Maughan et al28 into their computer model and used 0.095 mm Hg/mm Hg for LV-to-RV ventricular interaction gain, which is higher than what we found for normal pigs and more in the range of what we found for CHF pigs. These models do not address the effects of the pericardium on systolic interaction, but they are relevant to changes in ventricular interaction resulting from dilated cardiomyopathy.
The present study supports the concept that the pericardium has minimal effects on systolic ventricular interaction, at least in the range of volumes studied, and our results confirm the well-known effects of the pericardium on enhancing diastolic ventricular interaction.1 2 3 4 5 6 7 8 9 10 11 These studies indicate that the pericardium, by defining a rigid finite volume to enclose the four chambers of the heart, enhances the effect that the diastolic ventricular volume of one ventricle has on the pressure-volume relation of the contralateral ventricle. This increased coupling between the ventricles was reported to be due primarily to increased ventricular-pericardial-ventricular coupling rather than to a direct increase in ventricular-ventricular coupling.10
Surprisingly, we found no statistically significant differences with or without the pericardium in LV-to-RV diastolic interaction gains between normal pigs and pigs with pacing-induced CHF. These results suggest that, in this model of dilated cardiomyopathy, elastance of the interventricular septum is reduced only during systole, which is consistent with reduced myocardial contractility of the septum in CHF, and that there are no significant effects on diastolic septal compliance. The diastolic results with the pericardium intact from this study are similar in magnitude to the 0.33 mm Hg/mm Hg reported by Slinker et al29 for RV-to-LV interaction; they also determined gains from sudden hemodynamic transients. Slinker et al found no statistically significant difference in diastolic ventricular interaction during volume loading, which is a rather surprising contrast to our finding of a significant effect of volume loading on diastolic interaction gain with or without the pericardium in both normal and CHF pigs. This suggests that volume loading has a much more significant effect on diastolic LV-to-RV interaction as we determined in the present study compared with RV-to-LV interaction as determined by Slinker et al.
Others also reported changes in end-diastolic pressure in one ventricle in response to changes in the other, from which diastolic interaction gain can be calculated. These studies indicated a wide range of diastolic ventricular interaction gain measurements, probably related to the various techniques used: 0.43 mm Hg/mm Hg with the pericardium removed in normal dogs, which was reduced to 0.21 mm Hg/mm Hg in dogs with RVP-overload hypertrophy14 ; 0.185 before pericardiectomy and 0.0865 mm Hg/mm Hg after pericardiectomy in isolated hearts4 ; 0.86 mm Hg/mm Hg (slopes of linear regression equations between LVEDP and RVEDP) with the pericardium closed and 0.54 with the pericardium opened3 ; 0.45 (also from a slope between LVEDP and RVEDP) with a sutured pericardium30 ; and 0.297 with pericardium intact and 0.140 with pericardium opened in preterm lambs, and 0.650 with pericardium intact and 0.301 mm Hg/mm Hg with pericardium opened in newborn lambs.9 The results from our present experiment have the advantage of providing directly measured diastolic interaction gains over ranges in volume loading in the intact heart. Our results support the concept that the pericardium enhances diastolic interaction but that this interaction exists even without the pericardium, with gains of ≈0.10 to 0.14 mm Hg/mm Hg, which indicate the residual interaction solely through the ventricular septum and free walls independent of pericardial restraint.
One of the limitations of the present study was the need to open sections of the pericardium to insert the LV apex cannula and to connect the electromagnetic flow probe to the pulmonary artery. Although every attempt was made to minimize these effects, these manipulations could have affected the measured interaction gains. The fact that the ventricular interaction gains were substantially higher in studies where ventricular pressures were abruptly increased21 rather than decreased as in the present study deserves further investigation. However, the pericardium was found to have no significant effect on systolic interaction with either method. The present study also determined ventricular interaction at different RVEDPs without knowledge of RV end-diastolic volume. The results may have been different if data were measured at constant volumes, as in isolated heart studies. Also, because of the very small changes in diastolic pressure of fractions of 1 mm Hg, especially with the pericardium closed, there could be considerable error in the diastolic gain calculations. However, we believe that if the integrated changes in diastolic pressure are used and data from replicated data groups are averaged, the confidence in such measurements is acceptable. Finally, our experiments were performed after only 1 week of rapid ventricular pacing to produce CHF. Although a convincing model of heart failure was produced in this short time, results may be different in more long-term models, and our results may not be representative of dilated cardiomyopathy in human subjects.
In conclusion, these studies of directly measured ventricular interaction isolated to systolic phases and to diastolic phases of the cardiac cycle indicate that pacing-induced dilated cardiomyopathy increases systolic interactions without substantial effect on diastolic interactions and that the pericardium increases diastolic but not systolic ventricular interactions. Volume loading in normal and CHF pigs increases both systolic and diastolic ventricular interactions with and without the pericardium opened.
This work was supported in part by grant R01-HL-45608 from the NHLBI.
- Received January 24, 1995.
- Accepted February 28, 1995.
- Copyright © 1995 by American Heart Association
Elzinga G, van Grondelle R, Westerhof N, van den Bos GC. Ventricular interference. Am J Physiol. 1974;226:941-947.
Glantz SA, Parmley WW. Factors which affect the diastolic pressure-volume relationship curve. Circ Res. 1978;42:171-179.
Glantz SA, Misbach GA, Moores WY, Mathey DG, Lekven J, Stowe DF, Parmley WW, Tyberg JV. The pericardium substantially affects the left ventricular diastolic pressure-volume relationship in the dog. Circ Res. 1978;42:433-441.
Janicki JS, Weber KT. The pericardium and ventricular interaction, distensibility and function. Am J Physiol. 1980;238(Heart Circ Physiol 7):H494-H503.
Spadaro J, Bing OHL, Gaasch WH, Weintraub RM. Pericardial modulation of right and left ventricular diastolic interaction. Circ Res. 1981;48:233-238.
Maruyama Y, Ashikawa K, Isoyama S, Kanatsuka H, Ino-Oka E, Takishima T. Mechanical interactions between four heart chambers with and without the pericardium in canine hearts. Circ Res. 1982;50:86-100.
Olsen CO, Tyson GS, Maier GW, Spratt JA, Davis JW, Rankin JS. Dynamic ventricular interaction in the conscious dog. Circ Res. 1983;52:85-104.
Santamore WP, Li KS, Nakamoto T, Johnston WE. Effects of increased pericardial pressure on the coupling between the ventricles. Cardiovasc Res. 1990;24:768-776.
Santamore WP, Shaffer T, Papa L. Theoretical model of ventricular interdependence: pericardial effects. Am J Physiol. 1990;259(Heart Circ Physiol 28):H181-H189.
Slinker BK, Goto Y, LeWinter MM. Systolic direct ventricular interaction affects left ventricular contraction and relaxation in the intact dog circulation. Circ Res. 1989;65:307-315.
Brown CD, Chow E, Farrar DJ. Left ventricular unloading decreases the rate of isovolumic right ventricular pressure decline. Am J Physiol. 1993;265(Heart Circ Physiol 34):H1663-H1669.
Little WC, Badke FR, O’Rourke RA. Effect of right ventricular pressure on the end-diastolic left ventricular pressure-volume relationship before and after chronic right ventricular pressure overload in dogs without pericardia. Circ Res. 1984;54:719-730.
Slinker BK, Chagas ACP, Glantz SA. Chronic pressure overload hypertrophy decreases direct ventricular interaction. Am J Physiol. 1987;22(Heart Circ Physiol 22):H347-H357.
Santamore WP, Burkhoff D. Hemodynamic consequences of ventricular interaction as assessed by model analysis. Am J Physiol. 1991;260(Heart Circ Physiol 29):H146-H157.
Yamaguchi S, Harasawa H, Li KS, Zhu D, Santamore WP. Comparative significance in systolic interaction. Cardiovasc Res. 1991;25:774-783.
Woodard JC, Chow E, Farrar DJ. Isolated ventricular systolic interaction during transient reductions in left ventricular pressure. Circ Res. 1992;70:944-951.
Farrar DJ, Woodard JC, Chow E. Pacing-induced dilated cardiomyopathy increases left-to-right ventricular systolic interaction. Circulation. 1993;88:720-725.
Maughan WL, Kallman CH, Shoukas A. The effect of right ventricular filling on the pressure-volume relationship of the ejecting canine left ventricle. Circ Res. 1981;49:382-388.
Yaku H, Slinker BK, Bell SP, LeWinter MM. Effects of free wall ischemia and bundle branch block on systolic ventricular interaction in dog hearts. Am J Physiol. 1994;266(Heart Circ Physiol 35):H1087-H1094.
Slinker BK, Glantz SA. End-systolic and end-diastolic ventricular interaction. Am J Physiol. 1986;251(Heart Circ Physiol 20):H1062-H1075.
Kanazawa M, Shirato K, Ishikawa K, Nakajima T, Haneda T, Takishima T. The effect of pericardium on the end-systolic pressure-segment length relationship in canine left ventricle in acute volume overload. Circulation. 1983;68:1290-1298.
Chow E, Woodard JC, Farrar DJ. Rapid ventricular pacing in pigs: an experimental model of congestive heart failure. Am J Physiol. 1990;258(Heart Circ Physiol 27):H1603-H1605.
Maughan WL, Sunagawa K, Sagawa K. Ventricular systolic interdependence: volume elastance model in isolated canine hearts. Am J Physiol. 1987;253(Heart Circ Physiol 22):H1381-H1390.
Slinker BK, Goto Y, LeWinter MM. Direct diastolic ventricular interaction gain measured with sudden hemodynamic transients. Am J Physiol. 1989;256(Heart Circ Physiol 25):H567-H573.
Bemis CE, Serur JR, Borkenhagen D, Sonnenblick EH, Urschel CW. Influence of right ventricular filling pressure on left ventricular pressure and dimension. Circ Res. 1974;34:498-504.