Canine Left Ventricular Wall Research Articles

There is no transmural heterogeneity in an index of action potential duration in the canine left ventricle

Transmural heterogeneity in ventricular repolarization demonstrated in vitro has been difficult to confirm in vivo. Whether this discrepancy reflects a physiological phenomenon or a methodological problem remains a vivid matter of debate despite a plethora of experimental work. Therefore, we have measured the relevant electrophysiological parameters first in vivo and repeated these in the same heart and at identical sites in vitro. Methodological issues were tackled by using both unipolar and bipolar recordings. Physiological issues were explored by measuring both local and functional electrophysiological parameters. In 10 healthy dogs, 2 high-resolution needle electrodes were inserted into the left ventricle. Effective refractory periods (ERP) as well as activation recovery intervals (ARI) were determined at each electrode along both needles at basic cycle lengths (BCL) of 850 and 300 ms, respectively. After excision of the heart, ERP and ARI measurements were repeated in the arterially perfused wedge preparations. First, we observed that ERPs and ARIs were significantly shorter in vivo than in vitro. Mean ERPs and ARIs of all muscle layers were relatively uniform throughout the ventricular wall in vivo. The transition from the in vivo to the in vitro preparation was associated with a significant albeit small increase of mean ARIs in the subendocardium, whereas interlayer differences in mean ERPs did not reach statistical significance as in vivo. In the intact canine left ventricular wall, a more or less homogeneous distribution in transmural ERP and ARI is present.

NS5806 Activates the Transient Outward Potassium Current in the Canine Ventricle and Provides a New Model of the Brugada Syndrome

Background: The Brugada syndrome (BrS) is characterized by elevated ST segments in the right precordial leads, ventricular tachycardia and sudden death. The syndrome has been linked to loss-of-function of sodium and calcium channels, however the transient outward potassium current (Ito) is thought to be central in the pathogenesis of BrS. We assessed the effects of Ito augmentation in a mammalian model using a novel activator of Ito, NS5806. Methods and Results: Voltage-clamp experiments were performed on midmyocardial cells isolated from the canine left ventricular wall. At 40mV NS5806 (10 μM) increased peak Ito by 79±4 % and the time-course of inactivation was slowed (from Tau=12.6±3.2 ms to 20.3±2.9 ms). We next assessed the effect of increased Ito in the development of BrS phenotype using canine ventricular wedge preparations. NS5806 increased the epicardial action potential (AP) phase 1 magnitude, whereas the APs of endocardial cells were largely unaffected. The accentuated epicardial notch was associated with an accentuated J-wave on the ECG. Loss of the epicardial AP dome at some sites but not others, led to development of phase-2-reentry and polymorphic ventricular tachycardia. NS5806 was able to induce the BrS phenotype in wedges from both right and left ventricles of the canine heart; at 15 μM NS5806 BrS developed in 4/6 right ventricular preparations compared to 2/10 left ventricular preparations. Conclusion: The Ito agonist NS5806 recapitulates the electrographic and arrhythmic manifestation of BrS, providing evidence in support of its pivotal role in the genesis of the disease. Our findings suggest that a genetic defect leading to a prominent gain of function of Ito could explain variants of BrS in which ST segment elevation are evident in both right and left ECG leads.

Human mesenchymal stem cells as a gene delivery system to create cardiac pacemakers.

We tested the ability of human mesenchymal stem cells (hMSCs) to deliver a biological pacemaker to the heart. hMSCs transfected with a cardiac pacemaker gene, mHCN2, by electroporation expressed high levels of Cs+-sensitive current (31.1+/-3.8 pA/pF at -150 mV) activating in the diastolic potential range with reversal potential of -37.5+/-1.0 mV, confirming the expressed current as I(f)-like. The expressed current responded to isoproterenol with an 11-mV positive shift in activation. Acetylcholine had no direct effect, but in the presence of isoproterenol, shifted activation 15 mV negative. Transfected hMSCs influenced beating rate in vitro when plated onto a localized region of a coverslip and overlaid with neonatal rat ventricular myocytes. The coculture beating rate was 93+/-16 bpm when hMSCs were transfected with control plasmid (expressing only EGFP) and 161+/-4 bpm when hMSCs were expressing both EGFP+mHCN2 (P<0.05). We next injected 10(6) hMSCs transfected with either control plasmid or mHCN2 gene construct subepicardially in the canine left ventricular wall in situ. During sinus arrest, all control (EGFP) hearts had spontaneous rhythms (45+/-1 bpm, 2 of right-sided origin and 2 of left). In the EGFP+mHCN2 group, 5 of 6 animals developed spontaneous rhythms of left-sided origin (rate=61+/-5 bpm; P<0.05). Moreover, immunostaining of the injected regions demonstrated the presence of hMSCs forming gap junctions with adjacent myocytes. These findings demonstrate that genetically modified hMSCs can express functional HCN2 channels in vitro and in vivo, mimicking overexpression of HCN2 genes in cardiac myocytes, and represent a novel delivery system for pacemaker genes into the heart or other electrical syncytia.

Calcium channel heterogeneity in canine left ventricular myocytes

Regional variations in the electrophysiological properties of myocytes across the left ventricular wall play an important role in both the normal physiology of the heart and the genesis of arrhythmias. To investigate the possible contributions of calcium channels to transmural electrical heterogeneity, whole-cell patch-clamp recordings were made from isolated canine epicardial and endocardial left ventricular myocytes. Two major differences in Ca2+ channel properties were found between epi- and endocardial cells. First, the L-type Ca2+ current was larger in endocardial than in epicardial myocytes. The average peak current density at +10 mV in endocardial myocytes was 3.4 ± 0.2 pA pF−1, and was 45 % higher than that in epicardium (2.3 ± 0.1 pA pF−1). The kinetic properties of the L-type current in epi- and endocardial cells were not significantly different. Second, a low-threshold, rapidly activating and inactivating Ca2+ current that resembled the T-type current was present in all endocardial myocytes but was small or absent in epicardial myocytes. This T-like current had an average peak density of 0.5 pA pF−1 at −40 mV in endocardial cells. In most endocardial cells the T-like Ca2+ current comprised two components: a Ni2+-sensitive T-type current and a tetrodotoxin-sensitive Ca2+ current. We conclude that there are considerable regional variations in the density and properties of Ca2+ channels across the canine left ventricular wall. These variations may contribute to the overall transmural electrical heterogeneity.

Calcium channel heterogeneity in canine left ventricular myocytes.

Regional variations in the electrophysiological properties of myocytes across the left ventricular wall play an important role in both the normal physiology of the heart and the genesis of arrhythmias. To investigate the possible contributions of calcium channels to transmural electrical heterogeneity, whole-cell patch-clamp recordings were made from isolated canine epicardial and endocardial left ventricular myocytes. Two major differences in Ca2+ channel properties were found between epi- and endocardial cells. First, the L-type Ca2+ current was larger in endocardial than in epicardial myocytes. The average peak current density at +10 mV in endocardial myocytes was 3.4 +/- 0.2 pA pF-1, and was 45 % higher than that in epicardium (2.3 +/- 0.1 pA pF-1). The kinetic properties of the L-type current in epi- and endocardial cells were not significantly different. Second, a low-threshold, rapidly activating and inactivating Ca2+ current that resembled the T-type current was present in all endocardial myocytes but was small or absent in epicardial myocytes. This T-like current had an average peak density of 0.5 pA pF-1 at -40 mV in endocardial cells. In most endocardial cells the T-like Ca2+ current comprised two components: a Ni2+-sensitive T-type current and a tetrodotoxin-sensitive Ca2+ current. We conclude that there are considerable regional variations in the density and properties of Ca2+ channels across the canine left ventricular wall. These variations may contribute to the overall transmural electrical heterogeneity.

Relation between local myocardial growth and blood flow during chronic ventricular pacing.

Several studies have shown that, per unit mass, myocardial blood flow (MBF) and oxygen consumption are similar in hypertrophic and non-hypertrophic ventricles. This observation may be explained by the degree of myocardial growth matching the increase in oxygen demand. Such matching may, however, not be perfect at the local level, because substantial heterogeneity of MBF exists within the ventricular wall. We investigated to what extent local growth and MBF are matched after redistribution of workload within the left ventricular (LV) wall. Redistribution of workload was established by ventricular pacing at physiological heart rate, which induces asynchronous activation and contraction. Local wall mass (2D-echocardiography) and MBF (fluorescent microspheres) were determined in the canine LV wall before (t=0) and after 6 months of normal sinus rhythm (SHAM group, n=5) or 6 months of pacing at the LV free wall (PACE group, n=8). During acute pacing MBF (ml/min/g) increased with increasing distance to the pacing site. Local relative MBF (rMBF, local MBF normalized to mean MBF in the LV wall) varied from 0.8 adjacent to the pacing site to 1.2 in remote regions. After 6 months of pacing these regional differences had disappeared, probably due to changes in wall mass, which increased with increasing distance to the pacing site (by up to 39+/-13%). In SHAM animals rMBF at t=0 correlated well with rMBF 6 months later (r=0.71). In PACE animals, however, this correlation was poor (r=0.33), because rMBF increased in regions close to the pacing site with initial rMBF<1 and rMBF decreased in regions remote from the pacing site with initial rMBF>1. After redistribution of workload within the LV wall as induced by ventricular pacing, local load-regulated growth tends to equalize MBF distribution, but local adaptation of MBF also depends on initial MBF.

Effect of Laminar Orthotropic Myofiber Architecture on Regional Stress and Strain in the Canine Left Ventricle

Recent morphological studies have demonstrated a laminar (sheet) organization of ventricular myofibers. Multiaxial measurements of orthotropic myocardial constitutive properties have not been reported, but regional distributions of three-dimensional diastolic and systolic strains relative to fiber and sheet axes have recently been measured in the dog heart by Takayama et al. [30]. A three-dimensional finite-deformation, finite element model was used to investigate the effects of material orthotropy on regional mechanics in the canine left ventricular wall at end-diastole and end-systole. The prolate spheroidal model incorporated measured transmural distributions of fiber and sheet angles at the base and apex. Compared with transverse isotropy, the orthotropic model of passive myocardial properties yielded improved agreement with measured end-diastolic strains when: (1) normal stiffness transverse to the muscle fibers was increased tangent to the sheets and decreased normal to them; (2) shear coefficients were increased within sheet planes and decreased transverse to them. For end-systole, orthotropic passive properties had little effect, but three-dimensional systolic shear strain distributions were more accurately predicted by a model in which significant active systolic stresses were developed in directions transverse to the mean fiber axis as well as axial to them. Thus the ventricular laminar architecture may give rise to anisotropic material properties transverse to the fibers with greater resting stiffness within than between myocardial laminae. There is also evidence that intact ventricular muscle develops significant transverse stress during systole, though it remains to be seen if active stress is also orthotropic with respect to the laminar architecture.

Heterogeneous Loss of Connexin43 Protein in Ischemic Dog Hearts

Ischemia causes cell decoupling in the myocardium. Prolonged ischemia activates proteases and causes degradation of structural proteins as well as gap junctions. There is little information about the degradation of gap junction protein during the early time period after acute ischemia. The purpose of the present study was to investigate connexin43 (Cx43) protein degradation and distribution patterns in the canine left ventricular wall during 1 to 6 hours of ischemia. Ischemia of canine left ventricular myocardium was induced by ligation of the left anterior descending coronary artery. Following a period of in situ ischemia of up to 6 hours, samples were harvested, and standard paraffin slides were prepared for Cx43 and wheat germ agglutinin double labeling. Cx43 distribution was visualized by confocal microscopy. In controls, homogeneous distribution of Cx43 staining was determined. Ischemia caused a loss of Cx43 with a heterogeneous pattern by mixing foci of infarcted cells among normal cardiac myocytes. To determine if the changes were induced by heterogeneous reduction in the blood supply, an in vitro ischemic model was studied to induce more homogeneous ischemia. Western blot analysis of these in vitro ischemic tissue samples revealed a reduction of Cx43 protein concentration with a 50% decay time of 4.8 hours. Cx43 dephosphorylation was detected after 1 hour of in vitro ischemia. Heterogeneous loss of Cx43 was found in the in vitro ischemic tissue. There were no significant changes in Cx43 staining density during the first hour of ischemia at a time when dephosphorylation of the protein was observed. After 1 hour of ischemia, Cx43 was reduced at intercalated disk areas, and, after 6 hours, most Cx43 disappeared at intercalated disk areas, while small amounts of Cx43 remained at side-to-side junctions. Cx43 undergoes both distribution and concentration changes following acute cardiac ischemia. The loss of Cx43 protein is heterogeneous. Cx43 dephosphorylation occurred within 1 hour following ischemia.

Characteristics and distribution of M cells in arterially perfused canine left ventricular wedge preparations.

Much of the characterization of the M cell to date has been accomplished using isolated tissues and cells. This study uses an arterially perfused wedge preparation to examine the characteristics and distribution of M cells within the anterior wall of the canine left ventricle under more physiological conditions. Floating microelectrodes were used to record transmembrane action potentials simultaneously from epicardial, M region, and endocardial or subendocardial Purkinje sites in isolated arterially perfused canine left ventricular wedge preparations. A transmural ECG was recorded concurrently. M cells with the longest action potentials were found in the deep subendocardium in wedge preparations isolated from the anterior wall of the left ventricle. Fairly smooth transitions in action potential duration (APD) were observed except in the region between epicardium and deep subepicardium. Tissue resistivity increased 2.8-fold in this region and much more modestly in the deep subendocardium. Dispersion of APD90 across the left ventricular wall averaged 51+/-19 and 64+/-25 ms at basic cycle lengths of 1000 and 2000 ms, respectively, whereas transmural dispersion of repolarization time was smaller (34+/-18 and 45+/-25 ms), owing to the endocardial to epicardial activation sequence. We conclude that the qualitative differences between the 3 ventricular cell types previously described in isolated tissues and cells are maintained in intact canine left ventricular wall preparations in which the myocardial cells are electrically well coupled. As anticipated, differences in APD are quantitatively smaller because of electrotonic interactions among the 3 cell types. Our data indicate that transmural dispersion of repolarization is the result of intrinsic differences in APD of cells spanning the ventricular wall as well as a heterogeneous distribution of tissue resistivity across the wall.

Numerical Simulator for Estimation of Mechanical Functions of Left Ventricle. 3rd Report. Stress Distribution in Left Ventricular Wall.

Stress distributions in a canine left ventricular wall are simulated during one cardiac cycle by employing 3-dimensional finite element models developed in order to estimate the functions of the left ventricle. The appropriateness of the numerical simulator is examined by comparing the pressure volume relationships of the left ventricle computed by the present system with those obtained by the corresponding canine experiments. The initial geometry of the left ventricle is assumed to be a prolate spheroid in order to facilitate easy estimation of the complex computed numerical results. It is recognized that the distributions of the wall stress, e.g., the circumferential stress, near the equatorial surface are quite different from those of the thick cylinder subjected to the internal pressure and that the magnitudes and the directions of the principal stresses are closely related to both the contractility and the orientation of the myocardial muscle fiber.

Finite element stress analysis of left ventricular mechanics in the beating dog heart

A three-dimensional finite element model was used to explore whether or not transmural distributions of end-diastolic and end-systolic fiber stress are uniform from the apex to the base of the canine left ventricular wall. An elastance model for active fiber stress was incorporated in an axisymmetric model that accurately represented the geometry and fiber angle distribution of the anterior free wall. The nonlinear constitutive equation for the resting myocardium was transversely isotropic with respect to the local fiber axis. Transmural distributions of end-diastolic fiber stress became increasingly nonuniform from midventricle toward the apex or the base. At a typical diastolic left ventricular pressure (1 kPa), the differences between largest and smallest fiber stresses were only 0.5 kPa near midventricle, compared with 4.6 kPa at the apex, and 3.3 kPa at the base. Transmural fiber stress differences at end-systole (14 kPa) were relatively small in regions from the base to the midventricle (13–22 kPa), but were larger between midventricle and the apex (30–43 kPa). All six three-dimensional end-diastolic strain components were within or very close to one standard deviation of published measurements through the midanterior left ventricular free wall of the passive canine heart [Omens et al., Am. J. Physiol. 261, H918–H928 (1991)]. End-systolic in-plane normal and shear strains also agreed closely with published experimental measurements in the beating dog heart [Waldman et al., Circ. Res. 63, 550–562 (1988)]. The results indicate that, unlike in the midventricle region that has been studied most fully, there may be significant regional nonhomogeneity of fiber stress in the normal left ventricle associated with regional variations in shape and fiber angle.

Heat transport in the canine left ventricular wall.

The rate of rise of local myocardial temperature (dT/dt) evoked by left coronary artery main stem occlusion has been proposed in the literature as a measure of local metabolic heat production, assuming heat loss due to diffusion to be negligible. In a previous study (ten Velden, G. H. M., G. Elzinga, and N. Westerhof. Circ. Res. 50: 63-73, 1982), we showed that this assumption was not valid. With this information in mind, in an attempt to study local metabolism, we compared, in anesthetized dogs, the dT/dt with the temperature distribution over the left ventricular wall. We found the value of dT/dt to be reproducible in time and to reproducibly depend on location. Negative values as well as positive values were measured; values even higher than the maximal possible temperature slope, calculated from the energy equivalent of left ventricular oxygen consumption and the specific heat of cardiac tissue, were found. Transmural distribution of the dT/dt showed positive values epicardially and negative values endocardially, while, as previously shown, a parabola-like shape of the transmyocardial temperature distribution existed. Our findings demonstrate that dT/dt by left coronary main stem occlusion cannot be used as a measure of local myocardial heat production.

A force-length-time relationship describes the mechanics of canine left ventricular wall segments during auxotonic contractions.

We examined regional mechanics of the left ventricular free wall in naturally pumping dog hearts during beta-blockade. Local systolic wall force (F) and segment length (L) were obtained with an auxotonic force gauge and an ultrasonic dimension gauge, both inserted at the equatorial level of the wall to measure F and L in the circumferential direction. Shortening velocity (-dL/dt) was obtained by differentiation of L. Preload and afterload were changed by acute caval and/or aortic occlusion so that a wide variation in F, L, -dL/dt, and dF/dt during shortening was obtained. In all experiments, F vs. L at identical time (t) after end-diastole (ED) fell on well-defined lines, irrespective of the -dL/dt line (t = 200 msec) was equivalent to a drop of approximately 50% in F at 10% reduction in L. No defined relationship was observed between F, L, and -dL/dt. However, by superimposing F, L, and -dL/dt curves from contractions of high EDL and high -dL/dt on those from contractions of low EDL and low -dL/dt, and comparing F and -dL/dt at identical L and t, a slightly lower F (difference 2.23 +/- 1.09 g, P approximately 0.05) could be associated with the higher -dL/dt (difference 0.6 +/- 0.1 muscle length/sec, P less than 0.001). These data suggest that the F-L-t relationship is a valid descriptor of auxotonic contractions in the ventricular wall, and that the direct effect of shortening velocity on the wall force is modest.

Effects of coronary dilators on segmental forces in normal and ischemic regions of the canine left ventricular wall.

Changes in segmental forces of contraction (determined by a strain gauge arch) in non-ischemic (normal) and ischemic regions of the same left ventricular wall were studied in dogs anesthetized with morphine and pentobarbital. A branch of the left anterior descending coronary artery was completely occluded, and 10 min after occlusion coronary dilators were injected intravenously. The results can be summarized as follows. 1) Coronary occlusion markedly reduced the segmental force of contraction in the ischemic region, while it did not affect that in the normal region. Heart rate did not change markedly after coronary occlusion. 2) Either nitroglycerin (100 microgram/kg) or sodium nitrite (2.5 mg/kg) increased heart rate, and it also increased segmental force of contraction in the normal region while it decreased that in the ischemic region. 3) Papaverine (1 mg/kg) increased heart rate and segmental forces of contraction in both normal and ischemic regions. 4) Dipyridamole (250 microgram/kg) slightly increased heart rate, and it also increased segmental force of contraction in the normal region but not in the ischemic region.

Comparison of the distribution of intramyocardial pressure across the canine left ventricular wall in the beating heart during diastole and in the arrested heart. Evidence of epicardial muscle tone during diastole.

Computations of compliance of the left ventricle (LV) during diastole assume passive tissue characteristics. To evaluate this assumption, we measured diastolic LV intramyocardial pressure simultaneously in the subepicardium and subendocardium in 18 open-chest dogs, using 1-mm in diameter micromanometers. Subepicardial pressure, 26 +/- 1 mm Hg (mean +/- SEM) exceeded subendocardial pressure, 14 +/- 1 mm Hg (P less than 0.001), and it exceeded left ventricular end-diastolic pressure (LVEDP) (9 +/- 1 mm Hg) (P less than 0.001). After an infusion of dextran-40 (10 dogs), subepicardial diastolic pressure increased to 42 +/- 4 mm Hg which was higher than diastolic subendocardial pressure, 26 +/- 2 mm Hg (P less than 0.001) and LVEDP, 24 +/- 2 mm Hg (P less than 0.001). Following cardiac arrest (12 dogs) with the intramyocardial probes unchanged in position, LV intracavitary pressure, 9 +/- 1 mm Hg, and subendocardial pressure, 13 +/- 3 mm Hg, did not differ significantly from the pressures in the beating heart. Subepicardial pressure, 9 +/- 1 mm Hg, was lower than in the beating heart (P less than 0.001). Following distention of the arrested LV (12 dogs), subepicardial pressure, 31 +/- 7 mm Hg, was lower than both subendocardial pressure, 58 +/- 12 mm Hg (P less than 0.001) and LV intracavitary pressure, 54 +/- 11 mm Hg (P less than 0.001). These observations indicate that tone is maintained by the subepicardium during diastole. Furthermore, the LV wall does not appear to behave as a passive shell during ventricular filling.

Transmural myocardial deformation in the canine left ventricular wall

Page H523: T. R. Fenton, J. M. Cherry, and G. A. Klassen. “Transmural myocardial deformation in the canine left ventricular wall.” Page H527: in Figure 8, orientations of epi and endo axes should be interchanged. Page H528: in discussion at top left of page, results do not contradict results of Ingels et al. (22); they are reasonably comparable. Page H528: in Figure 9, orientations of epi and endo axes should be interchanged.

Transmural myocardial deformation in the canine left ventricular wall.

A biplane cineradiographic technique was used to measure deformation of the myocardium as indicated by small lead spheres implanted into the anterior left ventricular wall of anesthetized dogs. Deformation was resolved into nine separate components for each of the epicardial, middle, and endocardial layers. The data illustrate the mechanical effect of myocardial fiber orientation and the ability of muscle layers to deform differentially. In order to present an overview of all the results, the implications of the deformation components are discussed first separately and then in a coordinated fashion.

Relationship between epicardial S-T segment changes and myocardial metabolism during acute coronary insufficiency.

Canine left ventricular wall metabolites (ATP, creatine phosphate, glycogen, glucose, glucose-6-phosphate, and lactate) were assessed in the ischemic region of the wall after 17 minutes of ligation of the left anterior descending coronary artery at its midlevel. Variations in the degree of lactate accumulation were found between different sites in the same ischemic region and between the outer and the inner left ventricular wall at the same site. High levels of lactate were associated with S-T segment elevation in the epicardial electrocardiogram, but no lactate accumulation or only mild lactate accumulation was found at isoelectric sites. Lactate accumulation at isoelectric sites was higher in the outer wall than it was in the inner wall; the opposite tendency was observed at sites of S-T segment elevation. In addition to high lactate levels, sites of S-T segment elevation demonstrated a more pronounced depletion of ATP and creatine phosphate, indicating not only a marked anaerobic glycolysis but also a more pronounced overall anaerobic stress. No relationship was found, however, between the absolute magnitude of S-T segment elevation and the degree of lactate accumulation or ATP depletion at sites of S-T segment elevation within the ischemic region. The data obtained in this study demonstrate that during myocardial ischemia epicardial sites of S-T segment elevation are sites of pronounced subendocardial and epicardial ischemia and of anaerobic metabolism.

Differential effects of chromonar and nitroglycerin on intramyocardial oxygen tension

Tissue oxygen tension of the subepicardial (Epi Po 2) and subendocardial (Endo Po 2) regions of the canine left ventricular wall was determined polarographically. Chromonar consistently increased Epi Po 2; Endo Po 2 was decreased in half the studies and elevated in the other half. Coronary blood flow was maximally increased by chromonar and reactive hyperemia abolished. Nitroglycerin selectively increased Endo Po 2 before and after chromonar. This suggests that arteriolar dilatation is not a requisite for the rise in Endo Po 2 by nitroglycerin.

Diastolic geometry and sarcomere lengths in the chronically dilated canine left ventricle.

In 10 dogs left ventricular geometry and ultrastructure were examined 3 to 12 weeks after chronic ventricular dilatation was induced by means of a large arteriovenous shunt. Following cardiac catheterization, the hearts were arrested and fixed in diastole at the left ventricular end-diastolic pressures (LVEDP) which existed in the beating heart. Transmural LVEDP were increased in all animals (avg. 27 mm Hg). The diastolic pressure-volume relationship was shifted to the right, the average end-diastolic volume of the chronically dilated ventricles (avg. 103 ml) being larger than that of previously studied acutely dilated ventricles (avg. 72 ml, P &lt;.01), but calculated diastolic wall stress values were not different. Sarcomere lengths in the chronically dilated hearts averaged 2.19±.02 µ ( SE ) (range 2.11 to 2.27), a value near the apex of the sarcomere length-tension curve, but not significantly different from sarcomere lengths in acutely dilated ventricles. Slippage between myofibrils, reflected by a loss of normal alignment of the Z lines, appeared to be one mechanism underlying this adaptation. The findings indicate that sarcomeres in the canine left ventricular wall are remarkably resistant to chronic as well as to acute overstretch. They further imply that little or no Frank-Starling reserve mechanism was available in these ventricles and raise the possibility that a descending limb of the heart's performance as a pump need not reflect a descending limb at the sarcomere level.